PHOTOSYNTHETIC MECHANISMS OF GREEN PLANTS PHOTOSYNTHETIC MECHANISMS RHOTOSYNTHETIC MECHANISMS OF GREEN PLANTS Papers presented at a symposium sponsored by the Committee on Photobiology of the National Academy of Sciences — National Research Council with the support of the National Science Foundation. O: 3- CO nj □ ; ° i r^ I D I m 14-18 October 1963 Chairman: Bessel Kok Organizer: Andre T. Jagendorf Publication 1145 National Academy of Sciences — National Research Council Washington, D.C. 1963 LIBRARY OF CONGRESS Catalog Card Number 63-65396 CONTENTS Page FOREWORD '^^^^ I. SPECTROSCOPIC AND FLUORESCENCE ANALYSIS OF OXIDATION-REDUCTION CATALYSTS Studies on Primary Reactions and Hydrogen or Electron Transport in Photosynthesis by Means of Absorption and Fluorescence Difference Spectrophotometry of Intact Cells. ... 1 L. M. N. Duysens Correlation Between Absorption Changes and Electron 1 ft Transport in Photosynthesis B. Rumberg, P. Schmidt-Mende , J. Weikard and H. T. Witt Photosynthetic Electron Transport 35 Bess el Kok • Fluorescence Studies 45 Bess el Kok Light-Driven Cytochrome Reactions in Anacystis and Euglena ^^ John M. Olson and Robert M. Smillie The Temperature Insensitive Oxidation of Cytochrome F in Green Leaves - A Primary Biochemical Event of Photosynthesis ' "o Britton Chance and Walter D. Bonner, Jr. Light Induced Optical Changes in Green Leaves 82 Walter Bonner and Robert Hill Action of Two-Pigment System on Fluorescence Yield of Chlorophyll A 91 W. L. Butler and N. I. Bishop Principles of a Theory of Energy Utilization in Photosynthesis ^^^ James Franck and J. L. Rosenberg The Mechanism of Photosynthesis 112 Eugene Rabinowitch Page Fluorescence in Two-Pigment Systems 122 J. L. Rosenberg and Tevfik Bigat Relationship between Light Induced EPR Signal and Pigment P700 131 Helmut Beinert and Bessel Kok EPR and Optical Studies on Scenedesmus Mutants 138 Ellen C. Weaver and Norman I. Bishop A Method for Calculating Quantum Yields for the Formation of Reaction Intermediates 147 Daniel Rubinstein Light-Induced Rapid Absorption Changes During Photosynthesis. II. 430 m^ Absorption Changes in Aged Chloroplasts in the Presence of PMS and Ascorbate .... 153 Bacon Ke II. ELECTRON TRANSPORT PATHS - BIOCHEMICAL INVESTIGATIONS The Electron Transport System of Photosynthesis Deduced from Experiments with Mutants of Chlamydomonas Reinhardi 158 R. P. Levine Effects of Quinones and Oxygen in the Electron Transport System of Chloroplasts 174 Achim Trebst, Herbert Eck and Sieglinde Wagner Photosynthetic Electron Trcinsport and Phosphorylation 195 Daniel I. Arnon Characteristics of Tritium Incorporation into Illuminated Chloroplasts 213 Brian Colman and Wolf Vishniac Indophenol Dyes: Catalysts and Uncouplers of Photophosphorylation 219 Donald L. Keister Photosynthetic Phosphorylation in the Presence of Naturally Occurring Substances 228 C. C. Black and A. San Pietro Characterization of Allagochrome and its Biosynthesis in Leaf Extracts 235 Helen M. Habermann Page Some Effects of Oxygen in Photosynthesis by Chloroplast Preparations 243 F. R. Whatley in. STUDIES WITH ISOLATED ELECTRON CARRIERS Photosynthetic Pyridine Nucleotide Reductase. IV Further Studies on the Chemical Properties of the Protein 252 Keelin T. Fry and Anthony San Pietro Photochemical Reactions of Plastocyanin in Chloroplasts 262 Sakae Katoh and Atusi Takamiya Evidence for the Role of Several Quinones in the Electron Transport System of Chloroplasts 27 3 R. A. Dilley, M. D. Henninger and F. L. Crane The Pathway of Metmyoglobin and NADP Reduction by Illuminated Chloroplasts 278 H. E. Davenport On the Participation of Cytochrome f in Photo- synthetic Electron Transport 284 Giorgio Forti, Maria Luisa Bertole and Bruno Parisi The Photosynthetic and Respiratory Systems in Euglena Gracilis 291 Fulvio Perini IV. ENHANCEMENT STUDIES: GAS EXCHANGES Enhancement 301 Jack Myers Emerson Enhancement Effect and Two Light Reactions in Photosynthesis 318 Govindjee Separation of the Effects of Two Photochemical Reactions by Studies of Oxygen Exchange 335 C . S . French Chromatic Transients and Enhancement Recorded by the Glass Electrode 345 L. R. Blinks Observations on the Function of Chlorophyll a^ and Accessory Pigments in Photosynthesis 352 David C . Fork Page Light -Induced Oxygen Reactions in Isolated Chloroplasts 362 Yaroslav de Kouchkovsky and Jean-Marie Briantais Studies with Flash Illumination on the Enhancement Effect in Chloroplasts 371 C. P. Whittingham and P. M. Bishop Some Effects of Monochromatic Light on Oxygen Evolution and Carbon Dioxide Fixation in Chlorella Pyrenoidosa 381 Max H. Honnmersand Illumination Dependence of Enhancement 391 T. T. Bannister and M.J. Vrooman The Relation Between Pigment Concentration and Photosynthetic Capacity in a Mutant of Chlamydonnonas Reinhardi 400 G. C. McLeod, G. A. Hudock, and R. P. Levine V. RESPIRATION, PHOTOSYNTHESIS, AND HYDROGEN METABOLISM Photoreactions and Respiration 409 George Hoch and Olga v. H. Owens Some Flavin Interactions with Grana (Seen in a Different Light) 42 j Birgit Vennesland Utilization of Far-Red Light by Green Algae and the Problem of Oxygen Evolution 436 H. Gaff r on, W . Wiessner and P. Homann On The Interrelation of the Mechanisms for Oxygen and Hydrogen Evolution in Adapted Algae 441 Norman I. Bishop and H. Gaffron Effect of Light on Respiration 452 G. Krotkov VI. FUNCTION OF PIGMENTS AND PIGMENT COMPLEXES Aggregated Chlorophyll in vivo 455 S . S . Br ody and M . Br ody Properties of Chlorophyll Protein Isolated from Leaves of Chenopodium Album 479 Atusi Takamiya, Hirosi Obata and Eijiro Yakushiji Page Studies of the Constitution and Photochemical Activity of an Isolated Chlorophyll Complex 486 M. B. Allen and J. C. Murchio Soluble Protein-Pigment Complexes from Spinach Chloroplasts 496 Josephs. Kahn Photoreduction of NADP by Ascorbate and Hematoporphyrin . . . 504 Anthony San Pietro, Leo P. Vernon and Dorothy Limbach Photooxidation of Reduced PMS by Chloroplasts and Chlorophyll A under Anaerobic Conditions in the Presence of Quinones 509 Leo P. Vernon, Waldo S. Zaugg and Elwood Shaw Fluorescence, Energy Transfer, and SH-Groups in Photosynthetic Pigments of Red and Blue-Green Algae 519 Eiji Fujimori and Kenneth Quinlan Studies of the Localization, Physicochemical Properties, and Action of Phycocyanin in Anacystis Nidulans 5Z7 John A. Bergeron VU. CHLOROPLAST STRUCTURE AND ORIENTED MOLECULES Experiments made to Elucidate the Molecular Structure of Chloroplasts 537 Wilhelm Menke Oriented Molecules and the Structure of Chloroplasts 545 R. A. Olson with the collaborative assistance of W . H. Jennings Studies with Cyanidium Caldarium. 11. The Fine Structure of Pigment-Deficient Mutants 560 Lawrence Bogorad, Frank V. Mercer, and Rosemary Mullens Chlorophyll's Lipid Environment 571 A. A. Benson The Chloroplast Structure in Photosynthesis 575 Jerome J. Wolken Vin. MECHANISM OF PHOSPHORYLATION, AND STRUCTURAL DEFORMATIONS Control of Chloroplast Structure by Adenosine Triphosphate . . . 587 Lester Packer Page Studies on the Mechanism of Photophosphorylation 599 A. T. Jagendorf and Geoffrey Hind On The Coupling of Photophosphorylation to Electron Transport 611 Mordhay Avron and Noun Shavit The Stoichiometry of Photophosphorylation 619 Thomas Punnett Structure-Function Relationships in Proteins and their Possible Bearing on the Photosynthetic Process 625 Rufus Lumry IX. PATH OF CARBON AND ASSOCIATED METABOLISM Recent Kinetic Studies on the Carbon Reduction Cycle 635 J. A. B as sham Glycolate Pathway 648 N. E. Tolbert An Evaluation of the Carbon Reduction Pathways of Photosynthesis 663 Martin Gibbs The Production of Glycollate During Photosynthesis 675 C. P. Whittingham, R. G. Hiller, and M . Bermingham Chloroplast Nucleotide Coenzynnes 684 William L. Ogren and David W. Krogmann Metabolism of Inorganic Polyphosphates in Growing Chlorella Cells 688 Shigetoh Miyachi X. MISCELLANEOUS TOPICS The Decay of Delayed Light at Short Times 698 William Arnold and J. B. Davidson Effects of Photosynthetic Poisons on Delayed Light in the Millisecond Time Range ^01 Walter F. Bertsch, J. B. Davidson, and J. R. Azzi Light Scattering by Chloroplasts in the UV 711 William F. Prickett, F. Dudley Bryant, and Paul Latimer Vll Page Interference of Emission Changes with Fast Absorption Changes in the Flash Spectroscopy of Algae "717 Edgar Inselberg and J. L. Rosenberg Effects of Hydrostatic Pressure on Induction 7 ? A Transients of Oxygen Evolution "^" William Vidaver Effects of Photodynamic Treatment, Ultraviolet Radiation and Gamma Radiation on the Photosynthesis and Hill Reaction of Chlorella ^^^ John D . Spikes and Dennis C. Hall On The Variability in the Activity of the Photo synthetic Mechanisms ^42 Constantine Sorokin SUMMARY ^^^ LIST OF PARTICIPANTS "^59 FOREWORD This volume contains the papers submitted to a synnposium on "Photo- synthetic Mechanisnns of Green Plants," held at Airlie House, Warrenton, Virginia, Oct. 14-18, 1963. The symposium was proposed and sponsored by the Committee on Photobiology , of the National Academy of Sciences— National Research Council, Carl P. Swanson, Chairman. The funds for the symposium were provided by a generous grant from the National Science Foundation, with ancillary support from the Kettering Foundation, Yellow Springs, Ohio. It was the feeling of the Committee that progress in the field since the last meeting of this kind, Gatlinburg, 1955, warranted a summing up and organization of our newer information. In retrospect, a brief paper by Blinks in that last meeting proved to be the forerunner of the period in photosynthesis research which we attempted to crystallize at Airlie. Blinks' report on chro- matic transients brought the realization that photosynthesis is not "color blind"— and that different pigments might sensitize different photoprocesses . Soon followed the observation by the late Dr. Emerson of the enhancement effect in which lights of two different wavelengths proved to exert a greater effect if given simultaneously than if given individually. This enhancement of net rate was rationalized by the observation of a push-and-puU effect of two different colors upon intermediate catalysts of the process: i.e. , P700 and cytochrome f . The analysis of photosynthesis in terms of two distinct photo- reactions, their features and their coupling, has consequently been the main area of concentration during the last years. This interest is reflected in a large percentage of the papers in this symposium. Enhancement, transients, and respiratory interactions with at least one light reaction have been studied in great detail. Considerable spectroscopic evidence concerning primary and early events has accumulated. Also on the basis of biochemical studies and analysis using mutants, a picture of photosynthetic electron transport is be- ginning to emerge, albeit somewhat hesitantly. There seems to be fair agree- ment at present about the nature of the first photoreaction, producing an as yet unidentified strong reductant, and a weak oxidant (P700). Compared to this the details of the other photoact and the evolution of oxygen associated with it are still quite obscure. Considerable progress is reported in regard to the nature and function of known and newly discovered constituents of the electron transport chain: chloroplast ferredoxin, transhydrogenase, plastoquinones , cytochrome_f , and plastocyanine. Photophosphorylation in chloroplasts , discovered about the time of the previous meeting in Gatlinburg, appears to be far from a finished problem. Some newer aspects are presented here, such as a large pool of high-energy intermediate, capable of inaking ATP in the dark. It is interesting to note that probably no area of photosynthesis can yet be considered a closed chapter. Although the path of carbon in photosynthesis IX has been familiar for a considerable time, the present fluid state of this field is excellently surveyed, with clear indications as to where revisions must be forthcoming in the future. Elegant work on the chemical structure of lamellae points up our continued ignorance as to the precise functional significance of the structures, or even as to the mode of attachment of chlorophyll to protein. Our introduction cannot attempt a detailed survey of the contents of this book. It must further be stated that some areas of photosynthesis were com- pletely omitted in the planning of this nneeting. We hope the reader realizes that the constant shifting of the focus of attention often makes it difficult to discriminate between the intrinsic importance of a given aspect and the number of workers interested in it. Every so often someone manages to remove an- other stone from the wall through which we all want to see, and the crowds tend to flock around the new peep-hole. To a certain extent we must apologize for the inelegant composition of this volume, the lack of editing and the somewhat brutal measures taken to secure early publication. We are grateful to all contributors for their coop- eration in providing concise summaries of their recent important work, and we hope that the exposed cross-section of photosynthesis as of the early sixties will prove of value. It is a pleasure to acknowledge the conscientious efforts of Miss Inger Hermann, Secretary of the Photobiology Committee, and of her staff. Bess el Kok, Chairman Andre Jagendorf , Organizer I. SPECTROSCOPIC AND FLUORESCENCE ANALYSIS OF OXIDATION-REDUCTION CATALYSTS STUDIES OH PRIMARY REACTIOllS AUD HYDROGEN OR ELECTRON TRANSPORT IN PHOTOSYHTHESIS BY MEANS OP ABSORPTION AND FLUORESCENCE DIF- FERENCE SPECTROPHOTOMETRY OP INTACT CELLS L.N.M. Duysens INTRODUCTION AND METHODS As requested by the organizers of this symposium, in this paper investigations mainly carried out at our laboratory will be reviewed and interpreted. In most experiments the time courses of changes in light ab- sorption and emission were recorded at various wavelengths in the visible and adjacent spectral regions for suspensions of in- tact photo synthesizing cells. These changes were brought about by suddenly admitting relatively strong so-called actinic light of constant intensity. This light was switched off after a time varying from one second to about one minute. The changes in ab- sorption were measured by means of a separate weak modulated measuring, beam, which in general did not cause changes in ab- sorption ' J the fluorescence was in general excited by modu- lated light of such a/^Qw intensity so as not to cause changes in fluorescence yield^ . The measuring apparatus was only sen- sitive to the modulated measuring beam or to the fluorescence, excited by the modulated exciting beam. These modulations and the use of suitable filters prevented direct effects of the ac- tinic light on the measuring apparatus. The absorption and fluor- escence changes were recorded by means of a one-fourth or one second recorder. Rapid changes in fluorescence yield were brought about by an electronic flash, and were measured by means of an oscillograph. This apparatus has a response time of the order of one usee, and the flash served both as exciting and actinic beam • A difference spectrum presents a change in the absorption spectrum or/Of the fluorescence spectrum occurring during il- lumination » Difference spectra may be obtained from the ti- me c\irves of absorbancy or fluorescence changes (see ). These spectra sometimes permit the identification of substances cau- sing the spectral changes. If e.g. the difference spectrum is L.H.H* Duysens equal to the difference of the absorption spectra of oxidized and reduced cytochrome, this shows that upon illumination a cy- tochrome becomes oxidized. Similarly, if the fluorescence dif- ference spectrum is equal to the fluorescence spectrum of chlo- rophyll a, this means that the chlorophyll a fluorescence in- creases upon illumination. Difference spectra determined between various time limits make it possible in principle to analyze the time course of photo synthetic intermediates which show ab- sorption or fluorescence changes upon illumination. By measuring the effectivity of the actinic light as a func- tion of the wavelength, it is possible in principle to identi- fy the pigments which are responsible for the photochemical re- action or reactions which cause changes in the absorption spec- trum or the fluorescence spectrum. If these changes are quali- tatively different in light of different actinic wavelengths, then it follows that more than one photochemical reaction dri- ven by pigment systems with different absorption spectra occurs. This is the case in oxygen evolving photo synthesizing organisms. In fluorescence experiments one more parameter is available: the wavelength of the exciting light. The action or effective- ness spectrum for exciting the fluorescence of a certain com- pound is proportional to the sum of the absorption spectra of the fluorescing substance and of the absorption spectra of other pigments multiplied by the efficiences of transfer of ex- citatien energy from these pigments to the fluorescing substan- ce • Since it is possible to bring about oxidation or reduction of a great number of non-physiological substances by means of extracts of bacteria or higher plants, it is in general not possible to conclude from these experiments, whether a certain redox reaction occurs in the living cell, even if the reacting substances are known to occur in the cell. For this reason mainly, we have concentrated on studying reactions in intact cells in order to establish which reactions occur i^ vivo . Most if not all reactions observed by absorption and fluores- cence techniques in living cells were redox reactions. In order to find out whether a reaction is a main pathway in photosyn- thesis or a side reaction, the quantum efficiency of this re- action may be determined. Since the quantum requirement per hydrogen atom or electron transported for COp reduction is 2, an efficiency of the same order of magnitude may be expected for the redox reactions in the photo synthetic chain. L.K.M. Duysens (Qualitative evidence suggests a hydrogen transport scheme with two different photochemical reactions in series. If the two pigment systems have different pigment compositions, in ab- sence of interaction two different action spectra for the va- rious reactions are found under the appropriate conditions. We feel that quantitative data, such as quantum efficiences and ac- tion spectra for the various reactions, are very useful per- haps necessary criteria to establish whether a hydrogen trans- port scheme may be correct. RESULTS AlID IKTERPRETATION For the sake of brevity the results will not be discussed in a historical and inductive sequence, but rather in a logical and deductive one. The mechanism of hydrogen or electron trans- port in algae or higher plants is discussed at the hand of the scheme shown in Fig. 1. The arrows point into the direction of hydrogen transport. Of the two components of the redox couples, only the component pre- ponderant in darkness is represented. Cytochrome C 420 occurs in darkness in the reduced form. The approximate B* values of the redox couples are indicated at the left side: tne more strongly reduciiig redox couples occur at the top of the drawing. The two heavy upward pointing arrows indicate the two primary photochemical reactions, 2 and 1, symbolized by the light quan- ta hv2 and hv-^. Most other arrows point downward* which indica- tes that the reactions occur spontaneously with a loss of free energy. Such reactions may be "coupled" to a reaction in which a gain of free energy occurs: the phosphorylation of ADP. The two primary photochemical reactions, photoreactions 1 and 2, &xe driven by two distinct pigment systems with different ab- sorption and action spectra. Action spectra of these systems are shown for two species as inserts in the scheme. The action spec- trum of pigment system 1 in the red alga Porphyridium cruentun shows that for photoreaction 1 quanta incident at 680, which are mainly absorbed by chlorophyll a, are about equal- ly effective as quanta absorbed at 56O mfi mainly by the "acces- sory pigment" phycoerythrin . The spectrum of system 2 shows that quanta incident at 56O mji, which are absorbed mainly by the phycoerythrin of pigment system 2, are much more effective in photoreaction 2 than quanta incident at 680 or 430 ofi, which are absorbed by chlorophyll a^ . Comparison of the action spectra of the two systems shows that light absorbed by Porphyridium at 680 and 43O m^ is more effective in exciting pigment system 1 than system 2, and that light absorbed at 56O a*i is more ef- L.N.M. Duysens rlydi"o^»n or fle^trori transport, in phptosynthetis ^M ""I* Pig. 1. Scheme for photosynthesis. The direction of hy- drogen or electron transport of oxygen evolving photo syn- thetic organisms is indicated by the direction of the ar- rows. The redox substances surrounded by a rectangle may be partly bypassed. The two light reactions are represen- ted by the large open arrows. Action spectra of the two pigment systems driving these reactions are shown for two algal species. A more detailed description is given in the text. fective in excitixxg system 2. In Chlorella , chlorophyll b is more active in system 2. Both systems contain about equal amounts of chlorophyll a, but a form of chlorophyll a with absorption at lunger wavelengths is preponderant in system 1. Wavelengths shorter than 680 m^x^ are more effective for system 2 than for system 1 in Chlorella . The relative amounts and activities of the various pigments being present in the two systems is not only different for different species, but also for different cultures of the same species. L.N.M, Duyaens Evidence for the occurrence of the redox couples shown in Fig. 1 will be discussed below. We restrict ourselves here to a few general statements. According to the scheme* upon onset of illumination a reduction will occur of all compounds at the right hand side at the top of the scheme, and an oxidation of H2O and ZH. Upon illumination with light of a wavelength, which ia mainly absorbed by system 1 (and which we call light 1. such as 430 and 680 mn in Porphyridium , or 700 mjx in Chlorella ) , redox components in between the two systems, such as ^ and PH, tend to accumulate in the oxidized state; in light 2 (such as light of 560 mM in Porphyridium and of wavelength shorter than 680 mji in Chlorella ) these redox couples may be expected to ac- cumulate in the reduced state. Upon darkening, the initial re- dox state in the dark is usually reestablished due to reactions with redox substances shown in the scheme, or with other cell constituents. The action spectrum for the initial r^te of re- duction of X and all compounds written at the right hand side of Xwill be proportional to the action spectrum of system 1; the same will be true for the initial rate of oxidation of the sub- stances between system 1 and 2, with the exception of ^, if the electron or hydrogen transport between Q and the substances is interrupted by an inhibitor such as DCMU, In the presence of this inhibitor, the action spectrum for the initial rate of re- duction of <^ will be proportional to the action spectrum of system 2. The overall electron or hydrogen transport during steady state photosynthesis will be limited by that system which has the smallest activity at the wavelength of the actinic light. Thus, the action spectrum of photosynthesis will follow the lower of two curves shown in the inserts of Pig. 1. If, on the other hand, the action spectrum of photosynthesis is mea- sured in the presence of a strong constant background of light » mainly absorbed "by system 2, this action spectrum will be pro- portional to the action spectrum of eystoaO. ^ , and vice versa . This was in fact the way French and Myers ineasured action spectra of Chlorella , and we have interpreted these spectra as action spectra of system 1 and 2. The spectra for Porphyridium , also shown in Fig. 1, were measured in various ways (see below). The scheme of Fig. 1 makes it possible/to understand the '• Emerson effect ", Emerson's discovery ^ ' stimulated research that established and identified the two algal photosynthetic systems. This effect may be formulated and "explained" as fol- lows (cf. ), If light beams 1 and 2 (e.g. light of 700 and 650 m^. respectively in Chlorella , or 68O and 56O mjx in Porphy- ridium ) are applied simultaneously, then the rate of photosyn- L.N.M. Duysens thesis in both beams applied together is foiuid to be greater than the sum of the rates in each beam separately. In light 1, system 1 is producing oxidized intermediates at a higher rate than system 2, and in light 2, system 2 is producing reduced intermediates at the higher rate. If both beams are applied simultaneously, the oxidized and reduced intermediates formed in excessin each beam separately react with each other, which results in an enhanced electron transport or an enhanced rate of photosynthesis. Cytochrome reactions Purple bacteria. Illumination of PB^P?)®iV^9$®f4^ causes the oxidation of one or more cytochromes^ * * * * • In purple bacteria presumably only one photochemical system, analogous but not /identical to system 1 in algae and higher plants is present^ * . Oxidized cytochrome may be reduced either by a so-called hydrogen donor, which has a function analogous to H2O in algae, or by a reductant, e.g. by XH, formed in the light. In/ ihe, latter reaction so-called cyclic phosphorylation may occur ' . The sequence of the reactions in purple bac- teria between the various cytochromes present, has not yet been definitely established. Olson and Chance^ 'made an extensive investigation of cytochromes in the purple bacteria Chromatium . They concluded that at least four cytochromes were present, dis- tinguished by different time courses and difference spectra. One of these cytochromes, G 423,5 (with a difference spectrum having a soret maximum ^*/4S^»5 uni) is still oxidized at the temperature of liquid air . Oxidation of other cytochromes was not observed at this temperature, but it was suggested that the oxidation of these cytochromes could not be observed, be- cause these cytochromes were oxidized spontaneously in darkness during cooling. The rate of oxidation of C 423.5 was the same at -I96 as at 20 C. Cytochrome photooxidation ie an extremely efficient reac- tion. Assuming that the/Sgecific absorption difference at 420 m|i is 62 /(mM»cm),01son calculated that the number of quanta required for the oxidation of one cytochrome molecule (the quantum requirement) was about 1 within rather^large limits of error. Making similar assumptions, Vredenberg calculated a minimum requirement of 0.6, which suggests that the true quantum requirement is 1 and that the true specific absorbancy diffe- rence at 423 mji is about 100 /(mM'cm). The equality of the rates (and thus that of the quantum requirements of cytochrome oxidation) at 20 and at -I70 C was confirmed. However, it was found that L.N.li. Duyeens the rate (or the quantum requirement) decreased at -220 C, Also it was established that all cytochromes, which were at 20 C in the reduced state, remained in the reduced state during cooling in the dark; except for C 423.5, no photooxidation occurred at -170°C. The rate of photooxidation of cytochrome C 422 strongly decreased for temperatures below -90 C (see Pig. 2), 5 sec CHROMATIUM Pig, 2. Time courses showingpg^^o'too^i*^^*^*^'^ Qix* of cyto- chrome C 422 in Chromatium "• 'at various temperatures but for the same intensities. The initial rate of oxi- dation decreases below -90 C. The decrease in the rate upon lowering the temperature was re- latively more pronounced at higher intensities of the actinic light, which also indicated that the rate was limited by a tem- perature dependent reaction. The photooxidation of C 422 (above -90^C) was observed under conditions, in which C 423.5 ''as oxi- dized, and was found to proceed with a quantum requirement of 1, if the specific absorbancy difference is assumed to be 100/ (mWcm). These dark redox reaction may differ from other dark reactions which stop at temperatures close to the freezing point in that the participating molecules do not have to diffuse before reacting. These observations suggest that C 423.5 ^^^ C 422 are located on the same protein or cell constituent, and that oxidation of C 423.5 causes a change in confirmation of this protein, which brings the haem moiety of C 422 in the^right position to be photooxidized even at a temperature of -100 C by the probably adjacent primary photooxidant, P 890 (see below), Cy to chrome reactions in algae. The absorption difference spectrum, light minus dark, of the red alga Porphyridium cruen^ L.K.M. Buysens tuBj which has peaks at 420 and 555 Ba|i,is similar but not iden- TTcal to the difference of thi'^ab sorption spectra of oxidized and reduced cytochrome f or £. We call this cytochrome C 420. The maximum of the /difference spectrum of cytochrome £ in vitro occurs at a 424 mn\ 21 123)5 the band shift may have been caused by the extraction procedure* slnv contradistinction to the action spectrum of photosynthesis^-''^ , which shows much stronger photo synthetic activity of phycoerythrin thaja^of chlorophyll a, the action spectrum for cytochrome oxidation (2)' shows strong activity for chlorophyll a but little activity for phycoery- thrin; furthermore, the kinetics of cytochrome oxidation were qualitatively different at 680 and 56O mii, which indicated that at least two photochemical reactions participated in the cyto- chrome reactions. If the cytochrome is brought in the oxidized state by illumination with actinic light of 68O m^, addition of ^strong light of 5^0 m|i caused a reduction of the cytochrome U5»o7^ As discussed in the introduction, oxidation by light 1 and reduction by light 2 can be explained by means of scheme 1, and, in fact, a scheme like this was first proposed on basis of the cytochrome reactions. The light-driven reduction of cyto- chrome was inhibited by DGMU, hydroxylamine or M-ethylure thane. This suggests that these inhibitors inhibit one or more reac- tions between photoreaction 2 and C 420, Evidence will be gi- ven below that DCMU inhibits between ^ and quinone. Prom scheme 1 it follows that in the presence of an inhibitor like DCMU the action spectrum for the initial rate of C 420 oxidation is pro- portional to the action spectrum of system 1. The action spec- trum of system 1 of Porphyridium between 540 and 710 mix shown in the insert of the scheme was determined in this way. The ac- tion spectrum of photosynthesis, measured against a background of relatively strong light of 56O m^ is, according to the dis- cussion in the introduction, also proportional to the action spectrum of system 1, and was found to be similar to the action spectrum for cytochrome oxidation, albeit not identical. So far only ad hoc explanations have been given of the deviation be- tween the two spectra . Similar observations were made for the blue-green alga Anacystis nidulans ^^4;, Phospho pyridine nucleotide reduction . Upon illtimination of photosynthetic organisms an increase in fluorescence in general occurs in the blue region. The fluorescence difference spectrum and that for the excitation of this fluorescence are similar to the corresponding spectra of PNH bound to an enzyme. Since DPN and TPN do not fluoresce, these experiments indicate that upon illumination of photosynthetic organisms reduction of py- L.K.H* Duysens (25) ridine nucleotide occvirs^ . In algae, pyridine nucleotide reduction could be_mo8t readily- studied in the blue-green alga Anacystia nidulansC^A), As pre- dicted by the scheme of Pig, 1, the action spectrum for the ini- tial rate of pyridine nucleotide reduction was found to be roughly proportional to the action spectrum for C 420 oxidation in the presence of an inhibitor of the cytochrome reducing re- action. Measurements of the quantum requirement for this reduc- tion were consistent with the assumption that one hydrogen ab- sorbed by system 1 was sufficient for the transport of one hy- drogen-equivalent to pyridine nucleotide. Since about 8 quanta (absorbed by both pigment systemsl^are probably necessary for the reduction of 1 C0„ molecule (26), this implies that each photochemical system requires one quantum per transported hy- drogen atom or electron. The quantum requirement for C 420 oxi- dation by system 1 is estimated to be higher/th^n 2 for Porphy- ridium ^ ' and still higher for Anacystis '(24), which suggests that part of the hydrogens transported from OH to P 700 bypasses C 420. For this reason we have put C 420 within a rectangleo Vredenberg (unpublished observations) observed a auch higher efficiency for cytochrome oxidation after cooling to 2 C. Quinone reactions Addition of plastoquinone stimulates the/Hill reaction under certain conditions as first shown by Bishop . This has been taken as a proof of participation of plastoquinone as a redox intermediate in the Hill reaction. It is also conceivable, how- ever, that the quinone acts as a structural factor, which is necessary for optimal rate of the Hill reaction. Also the ob- servation that quinones are reduced or oxidized by chloroplasts does not prove, as was argued in a general way in the introduc- tion, /that quinone is a photosynthetic intermediate. Recently Amesz^ 'obtained, upon illumination of the blue-green alga Anacystis , in the ultraviolet region a difference spectrum which was similar to the difference spectrum of oxidized minus reduced plastoquinone (see Fig. 3). Fig. 3 then indicates that light ab- sorbed by system 1 causes oxidation of quinone, and that light absorbed by system 2 favors its reduction. The reduction of qui- none was inhibited by low concentrations of DCMU. The time courses of the oxidation and reduction indicate thatthe quantum efficiency for these reactions is rather high. The tl^ value of quinone, and the fact that DCMU inhibits quinone reduction in- dicates that quinone is located between *4 and C 420. The amoxait of quinone participating in this light-driven redox reaction is only about 1^ of the amount of chlorophyll ^ present. This is 10 L.N.M. Baysens Fig. 3. The difference spectrum with the blaok circles is the abserbancy difference of the steady states in intense light of 680 DM (light l) and in light of 620 mi (light 2), and that with open circles is the absorbancy difference at lower intensities, between the steady states in light of 680 and 620 mji and in light of 620 mji alone. The bro- ken curve is the difference of the spectra of oxidized and reduced plastoquinone. The first spectrum indicates, that besides oxidation of quinone in light of 680 n^ other reactions occur. only a small fraction of the total amount of the quinones, which are present in the chloroplast(at a concentration of about 1/10 of that of chlorophyll). It was not possible to conclude from the difference spectrum, which of the numerous quinones dis- covered in chloroplasts is the photoactive one. 11 L.N.M. Duysens The primary photooxidanta of system _1 : P 89O and P 700 The first observed reversible change in absorbancy upon il- lumination of photo synthesizing organisms was a decrease in ab- sorption close to the baoteriochlorophyll peak at around 89O m^ in purple bacteria^-^* ■^', In various species studied, this de- crease was accompanied by an decrease at about 810 m^i and an increase of absorption at about 790 m^i. These absorption chan- ges were attributed to the oxidation of a small(about 27^) spe- cial fraction of the bacteriochlorophyll, which fraction we now call PH 89O. Since the maximum of the differegge spectrum of PH 890 varies iHpdifferent purple bacteria ^-^' and under various conditions^ ^in a similar way as the maximum of the absorption spectrum of the bacteriochlorophyll-type B 89O (see the next paragraph), PH 89O is probably closely related to bac- teriochlorophyll, but not necessarily chemically identical. The shape of the difference spectrum indicates that PH 89O consists of at least two bacteriochlorophyll-like molecules, one with a maximum at about 8IO, the other with a maximum between 87O and 890 mn, depending upon the species. Upon photoconversion of PH 890 to P 890 the maxima shift from about 810 and 89O m^ to about-7^0 and I25O m|i. The last maximum was observed by Clayton wO). Similar changes in the infrared region can be brought about by adding mixtures of potassium ^errj-r and fer^i- cyanide. The E'n value of P was found to be 0.^1^ ', ^'^^^K^N and about O.44 (Duysens, unpublished observation). Clayton^ provided direct evidence indicating that the absorption peeik at 890 mil was. almost completely bleached upon oxidation. Arnold and Clayton made the important observation that the lig^* induced changes in infra-red absorption occurred even at 1 K in dried extracts of bacteria. In purple bacteria, the bulk of the pigments consists of ca- rotenoids and three bacteriochlorophyll types, called B 800, B 850 and B 890 after the approximate location of the absorp- tion maxima, which vary somewhat from species to species. Only B 890 shows fluorescence. Light energy absorbed by the carote- noids and the other bacteriochlorophyll types besomea only ac- tive in photosynthesis through transfer to B 890^^*-^. The fluorescence yield increases upon increasing the intensity of the actinic light. The increase^occurs at lower intensities in the absence of hydrogen donors^ -^. A quantitative correlation was found under various conditions between the increase i^'i^) fluorescence and the decrease in absorption aroxaid 89O mix • These observations can be quantitatively explained by the hy- pothesis that excitation energy is transferred from B 89O to 12 L.H.M. Duysens FE 890 ; upon receiving a quantum, FH S90 is bleached, yielding F 890. Since energy transfer by induced resonance requires ab- sorption of the energy receiving molecule in the region of the fluorescence spectrum of the energy transferring molecule, the transfer of energy from B 89O to FH 890 stops and the fluores- cence yield of B 89O increases. For a simple energy transfer model, if it is assumed that FH 89O does not fluoresce, a line- ar relation can be derived between the inverse of the fluores- cence yield and the increase in absorption. Experimentally, a linear relationship is observed indeed, which supports the hy- pothesis. fkX^ confirms the earlier suggestion of Wassink and coworkers that bacteriochlorophyll fluorescence increases because the "energy acceptor" of photosynthesis becomes exhaus- ted. / gN In algae and chloroplasts, Kok^ 'observed a decrease in ab- sorption at 700 m|i, which was in several aspects similar to the bleaching at 89O m^ in pxirple bacteria. The bleaching could also be brought about by mixtures of potaBsiumQferri-/and ferrocya- nide, the E'q value being about 0.44^ '. Kok^^ 'also obser- ved that the bleaching was brought about most strongly by light absorbed by system 1, and concluded from indirect evidence that the bleaching was reversed by light absorbed by system 2, in a similar way as the oxidation-reduction shifts in cytochrome C 420, Vredenberg (unpublished observations) confirmed this conclusion directly. Thus F 700 is probably located between systems 2 and 1. The high E'q value and other evidence indi- cates that F 700 is located between C 420 and system 1. Fresu- mably F 7OO is, as Kok suggested, the primary oxidant of system 1. Because of technical difficulties, it was so far not possi- ble in our laboratory to establish a quantitative correlation between the bleaching of FH 70O and the increase in fluores- cence of chlorophyll aj^, i.e. the chlorophyll a of system 1. The main difficulty was, that the fluorescence of chlorophyll a^^ is snail compared to the fluorescence of the chlorophyll a of system 2. There is, however, indirect evidence that chloro- phyll a^ 3^?2| fluorescence and that changes in this fluores- cence occur ^ * i in addition to changes in the fluorescence yield of chlorophyll a^ and &2f there are also fluorescence changes at 720 and in some species at 730 mji. The infrared fluorescence is-primarily excited by blue light^ >-J»5>42; (see also ^^* ^^ . Changes in the fluorescence yield of chlorophyll &2 and their correlation with a primary pho tore due tant of system 2* The action spectrum for the excitation of the chlorophyll a 13 L.N.M* Dujsens ronoxua- fluorescence in Porphyridixim cruentum shows Hnich more pron( ced activity of phycoerythrin than of chlorophyll a^ ' ,(see also^ ). Prom this it follows not only that excitation energy is transferred from phycoerythrin to the fluorescing chloro- phyll a, but also that the fluorescence is mainly caused by the chlorophyll a of system 2. From the observation that the ratio of the value at the phycoerythrin maximum and that at the chlo- rophyll a maximum was larger for the action spectrum of chloro- phyll a fluorescence than for the absorption spectrum, it had been already earlier concluded that two forms of chlorophyll a occurred. One of these is a fluorescent form, receives its ener- gy to a large extent from phycoerythrin and/the\other is non- or weakly fluorescent form of chlorophyll a^^'-'', Thes^gforms were recently called chlorophyll a^ and chlorophyll a^^ • Upon illumination of a suspension of Porphyridium cruentum, which previously had been in the dark, with green light (light 2) the fluorescence yield rapidly increased, by a factor of about two, and then slowly declined to a steady state value, which was only slightly higher than the value in the dark. The excitation spectrum for this fluorescence increase revealed that it was caused by chlorophyll &2' ^^* ^* ^^^ maximum of the increase in fluorescence yield, or shortly thereafter a strong blue beam was admitted in addition to the green actinic beam, the fluorescence yield rapidly dropped to a value not much higher than the steady state value. This phenomenon was obser- ved for all species of algae of different groups investigated: for various red and blue-breen\algae, the green alga Chlorella and for spinach chloroplasts^ « Light 2 was found to strongly increase the chlorophyll a^ fluorescence and light 1 to lower this enhanced fluorescence. If these changes are attributed to chemical changes in a compound Ci present in the neighbourhood of chlorophyll &2, then it can be concluded from scheme 1 that ^ is located between systems 2 and 1. Apparently the oxidized form "^ quenches the fluorescence, and the reduced form GiH does not quench the chlorophyll a^ fluorescence. DGMU does not inhi- bit the increase in fluorescence, but it abolishes the decrease in fluorescence (or the reoxidation of ^iH)by light 1. Apparently ^ ±3 different from C 420, P 700, or quinone, since the reduc- tion of these compounds and not the oxidation is inhibited by DCMU. Furthermore the first two compounds are in the dark in general present in the reduced form, while Q, is present in the oxidized form. In spinach chloroplasts the chlorophyll fluores- cence is strongly enhfinced by the addition of a small amount of dithionite. These experiments show that <^, is located between 14 L.N.M. Buysens system 2 and the other compounds mentioned, and suggest that QB is the primary photoreductant of system 2. It was observed (Duysens and Kamp, unpublished experiments) by means of an oscillographic technique that by illuminating a suspension of Chlorella by means of an electronic flash the yield of the fluorescence at 685 m^i mainly due to chlorophyll a2 could be tripled. 10 '°einstein/cm* I Fluorescence yield ^ at 685 mu as a function of total in- cident energy , for flashes of relative intensities varying from 2.6 to 100. The initial increase depends upon the total energy but not upon the intensity of the flash. For intense flashes for which the risetime of the fluo- rescence yield is of the order of 10 ^l8ec the maximxim level attained decreases. Fig. 4 shows that the initial increase in a rather wide range of intensities studied, is proportional to the absorbed energy, and independent of the intensity of the flash. If the intensi- ty is sufficiently high that the risetime becomes shorter than 10 jisec, then the yield does not increase up to the maximum at- tained at lower flash intensities. The shape of the time course depends on the conditions of the algae, but the appro- 15 L.K.ll. Duyeens ximate proportionality of the initial rate of increase with in- tensity and the progressive depression of the "steady state" at increasing flash intensities are general phenomena. Apparently this depression is caused either directly or indirectly by a light reaction. In some experiments the fluorescence yield goes through a maximum and decreases appreciably below the value achieved within 10 usee after the starting of the flash: the strongest depression appeared to occur after about 100 jisec. Whether the depression is caused by a rapid reoxidation of QH by P 700 »by another component of system 1, by a short cirr cuit of Z to QH, or by another reaction is a subject of fur- ther experimentation. Photo synthetic pho sphoryla tion We do not have experimental evidence on the sites of phos- phorylation. As far as I know most experimental evidence from other laboratories is consistent with the assumption that non- cyclic photophosphorylation(46) occurs between QH and P, and analogy with the respiratory system suggests that plastoquinone and cytochrome may participate in the pho sphoryla ting reactions. The possible sites of phosphorylation are given in the scheme by the arrow with ADP. The fact that cyclic phosphorylation in chloroplasts is, under certain conditions, not inhibited by DCMU suggests then it occurs in dark reactions between XH and P 700. The free energy loss in this redox reaction would be sufficient to permit the production of 2 molecules of ATP per 2 electrons transported, but the measured quantum requirements indicate a two times lower efficiencj(47) , SmOMARY A discussion is given of experimental evidence obtained in our laboratory concerning the mechanism of hydrogen or electron transport in photosynthesis. A scheme is given in which two dif- ferent photochemical reactions operate in series. Unpublished experiments are presented concerning the presumed primaiy pho- toreductant of the system 2 (which system is closely connected to the production of oxygen), and concerning the photooxidation of cytochromes in purple bacteria at low temperatures. It is shown that the initial rate of these photooxidations is de- creased upon lowering the teciperature, and it is argued that the cytochrome oxidation is a temperature dependent dark reac- tion. However, at "normal" intensities the rate of this reaction 16 L.H«H. Duysens becomes only limiting at very low temperatures, for some cyto- chromes at -100 C or even lower. REFERENCES (1) L.N.M. Duysens, in: Research in Photosynthesis. Edited by H. Gaffron, 59-67 • Interscience, New York, (1957). (2) L.N.M. Duysens, in: Progress in Photobiology. Edited by B.C. Christensen and B. Buchmann, 135-142. Elsevier Publ. Co, Amsterdam, (196I). (3) L.N.M. ])uysens and H.E. Sweers, in: Studies on Microal- gae and Photo synthetic Bacteria. Special Issue of Plant and Cell Physiology. Edited by Jap. Soc. Plant Physiolo- gists, 353-372. Univ. of Tokyo Press, (1963). (4) L.N.M. Duysens, Nature 168 , 548 (l95l). (5) L.N.M. Duysens, Thesis Utrecht (1952). (6) L.N.M. Duysens and J. Amesz, Biochim. Biophys. Acta, 6;^, 243-260 (1962). (7) C.S. French, J. Myers and G.C, McLeod, in: Symposia on Comparative Biology, vol. 1: Comparative Biochemistry of Photoreactive Systems. Edited by M.B, Allen, 361-365. Academic Press, New York-London, (I960). (8) R. Emerson, Ann. Rev. Plant Physiol., 9, 1-24 (1958). (9) L.N.M. Duysens, Nature, 173 * 692 (19547. (10 (11 B. Chance and L. Smith, Nature, 175 « 803 (1955). L.N.M. Duysens, in: Research in Photosynthesis. Edited by H. Gaffron, 164-173. Interscience, New York, (1957). (12) J.M. Olson and B. Ghance, Arch. Biochem. Biophys, 88 , 26-53, (I960). (13) L. Smith and B. Chance, Ann. Rev. Plant Physiol., £, 449-482 (1958). , , . X fl4) J. Amesz, Biochim. Biophys. Acta, 66, 22-36 (1963). (15) L.N.M. Duysens, J. Amesz and B.M. Kamp, Nature, 190 , 510-511 (1961). (16) A.W. Frenkel, Brookhaven Symposia in Biology, 11., 276-288 (1958). (17) D.I. Arnon, in: A Symposium on Light and Life. Edited by W.D. McElroy and B. Glass, 489-566. The Johns Hopkins Press, Baltimore (196I). (18) B. Chance and M. Nishimxxra, Proc. Natl. Acad. Sci. U.S., 46, 19 (I960). (19) J.M. Olson, Science, 135 » 101 (1962). (20) W.J. Vredenberg and L.N.M. Duysens, Biochim. Biophys. Acta. In the press. (21) H.E. Davenport and R. Hill, Proc. Royal. Soc, 139 B , 17 L.N.k. Duysens 327 (1952). ^ ^ . (22) S. Katoh, Hature, 186, 138-139 (i960). (23) L.ir.M. Duysens, Science 121 , 210 - 211 (1955). (24) J. Amesz and L.N.M. Duysens, Biochim. Biophys. Acta, 164, 261-278 (1962). L.H.M. Duysens, Proc. 4th Intemat, Congr. Biochem. , held in Vienna, 303 . Pergamon Press, London (1959). B. Kok, in: Handbuch der Pflanzenphysiologie Band V. Edi- ted by: W. Ruhland, 566-633. Springer Verlag Berlin, Gbt- tingen, Heidelberg (i960). N.I. Bishop, Proc. Hatl. Acad. Sci. US, 4^, 1696-1702 (1959). J. Amesz, Biochim. Biophys. Acta. In the Press. L.N.M. Duysens, W.J. Huiskamp, J.J. Vos and J.M. van der Hart, Biochim. Biophys. Acta, 1£, 188-190 (1956). R.K. Clayton, Photochem. Photobiol., 1., 201-210 (1962). J.C. Goedheer. Brookhaven Symposia in Biology, 11, 325-331 (1958). L.N.M. Duysens, Brookhaven Symposia in Biology, 11, 10-25 (1958). , , , R.K. Clayton, Photochem. Photobiol., 1, 305-311 (1962). W. Arnold and R.K. Clayton, Proc. Natl. Acad. Sci. US., 46, 769 (I960). B.C. Wassink, E. Katz and R. Dorrestein, Enzymologia, 10, 285 (1942). W.J. Vredenberg and L.N.M. Duysens, Nature, 197 » 355-357 (1963). , , E.G. Wassink, Adv. in Enzymology, 11, 91-199 (1951 ;• B. Kok, Acta Bot. Neerl., 6, 316-336 (1957). B. Kok, Biochim. Biophys. Acta, ^, 527 (l96l). H.T. Witt, A. Miiller and B. Rumberg, Nature, 122,, 967-969 (1961). B. Kok and G. Hoch, in: A Symposium on Light and Life. Edited by W.D. McElroy and B. Glass, 397-423. The Johns Hopkins Press, Baltimore (l96l). L.N.M. Duysens, Proc. Royal Soc, 1^ B, 301-313 (1963). W.L. Butler, Biochim. Biophys. Acta, 64, 309*317 (1962). J. Lavorel, Plant Physiol., 2>±, 204-209 (1959). C.S. French and V.K. Young, J. Gen. Physiol., ^t 873*890 (1952). A.T. Jagendorf, Survey of Biological Progress, Vol. IV, 181. Academic Press, New York (1962). H.C. Yin, Y.K. Shen, K.M. Shen, S.Y. Yang and K.S. Chin, Scientia Sinica (Peking) 10, 976 (I961). CORRELATION BETl'.rEEN ABSORPTION CHAl^IGES AND ELECTRON TRANSPORT IN PHOTOSYNTHESIS B. Rumberg, P. Schmidt-Mende , J. Weikard and H.T. Witt With the method of sensitive flash photometry we separated and analyzed as yet 7 different types of absorption changes. By this method it was possible to derive a rather detailed reaction scheme on the electron transport system in photosynthesis (1). In the following we will discuss some more details which support the probability of this scheme. I. PHOTOOXIDATION OF CHLOROPHYLL - a ^- Under "normal" conditions (eJiiting photosynthesis with red light between 65o - 7oo rau) mixed absorption changes can be ob- served between ^oo - 800 mu in chloroplasts and chlorella (s. fig.1 top) (this difference spectrum does not include the ab- sorption changes with life times .^ I0 ^ sec). The changes at 7o3 mu were discovered by Kok (2). The com- pound causing these changes he called P 7oo. He reported also changes at ^5o mu in inactivated aceton-extracted chloroplasts (3) and suggested that both changes may be caused by a chloro- phyll-a. But according to our measurements (1) there exist at least 5 different types of absorption changes around 45o mu caused by different compounds. Therefore, it is not established that P 700 is a chlorophyll-a. We tried therefore to separate out of the overall difference spectrum in photoactive prepara- tions the spectrum which belongs to P 7oo only. 1 . Separation of the difference spectrum of chlorophyll-a -j- in 3 way s a) Aged chloroplasts . Aged chloroplasts (spinach) reactivated by addition of reduced PIVIS (ascorbic acid in excess) show a simplified difference spectrum (4) very similar to that shown in fig.1, bottom. b) Plastoquinone extracted chloroplasts . Plastoquinone ex- tracted chloroplasts (by petroleum ether) show a simplified spectrum (s. fig.1, bottom), which is very similar to that men- tioned under a) . c) Addition of CMU to chloroplasts . CMU-poisoned chlorella or chloroplasts show a difference spectrum which is similar with that of a) and b). d) Trapping at -1^o°C . The difference spectrum of trapped products (i) at -150^^0 of fresh chloroplasts is shown in the 18 19 B. Rumberg, P. Schmidt-Mende, J. Weikard aud H.T. Witt top of fig. 2. It shows changes around 7o$ mu (dotted line) and in addition oxidized cytochrome (solid line). The cytochrome is masking changes around 45o mu which may be caused by other sub- stances. Addition of reduced PMS (phenazine methosulf ate) (as- corbic acid in excess), prevents the trapping of oxidized cyto- chrome (bottom in fig. 2). Under these conditions a difference spectrum can be seen at -15o which is similar to that of a), b ) , and c ) . e) A fifth method is described in 7,b). 2. Kinetics During the flash the absorption decreases very fast (-i. 1o sec in chlorella). In the dark a backreaction takes place in~1o~ sec at 2o^C (fig. 5a, 6a) (4)(5). 5. Identification of Ghl-a j The upper results (5 equal spectra under different conditions) suggest that the changes at ^5o mu and 7o3 mu are caused by one substance. This was additionally proved by comparing the kinetik behaviour of both changes in reactivated aged chloroplasts under different conditions. The lifetime and the magnitude are identi- cal at both wave-lengths at different values of pH (4) (fig. 5) and also at different concentrations of added reduced PMS_(^). Decreases of absorption changes just with the two absorption bands of chlorophyll-a indicate that very probably a chlorophyll -a (Chl-aj) is in action (-4-). 4. Oxidation of Chl-a j That the decreases of absorption indicate an oxidation of Chl-a^ in the light, was provided in our experiments by the fact that in aged chloroplasts reduced PMS or reduced DPIP (2,6-Dichlor- phenol-indophenol) can be directly coupled to the light product (4)(1). This is demonstrated by the strong acceleration of the decay time with increasing concentrations of red PMS or red DPIP (fig.-M-) and by the demonstration of a first order reaction (fig. 4). The electron-acceptor of Ghl-aj is called Z. Obviously in aged chloroplasts photooxidized Chl-a-j- is directly reduced by electrons provided from red PMS: Z ^r^ Ghla-j- < red PMS (The arrows indicate the flow of electrons). 5. Ohl-a-r - oxidation as a primary act The production of Ghl-a-. within -i 1o~5 sec and the trapping at -15o G give evidence th4t this oxidation is a primary act (4). 5. The effect of far red background light Far red actinic light between 7oo and 72o mu can only oxidize Ghl-a-,- (2) (in chlorella and fresh chloroplasts with Kill-oxi- dantsj. Actinic light with wave-length -c 7oo mu results in an oxidation of Chl-a-p followed by a reduction of Ghl-at • There- fore light of <. 7oo mu is obviously channeled into two reaction 20 B. Rumberg, P. Schmidt-Mende , J. Weikard and H.T. Witt centres. One part - hv-j- - is channeled to Chl-a-j- (as far red light), a second part - hVjy - is obviously channeled to a se- cond reaction centre w;here it provides electrons from a natural electron-donor for the reduction of Chl-a-j- • hv-r hv-p-r Z < — Chl-aj < ...Ai 9 With hVj-light alone (far red light 7oo - 72o mu) Chl-a-j- should accumulate in its oxidized form. Measuring absorption changes at 7o3 mu the measuring beam should already oxidize Chl-a-j-. The mag^ nitude of this oxidation must depend on the intensity of the measuring beam (fig.5)« a) In soft measuring light (7o5 mu) practically no Chl-aj - oxidation takes place. A supplementary red flash (-s: 7oo mu) at t^ with hv-p- and hv-p-^-light has the following effect: The hv^- llght oxidizes Chl-ai immediately. The hv-j-j- light provides electrons from a second reaction centre for the reduction of Chl-aj. The reduction takes place in»-^1o"^ sec (fig. 5a). b) In medium measuring light (7o5 mu) started at t^ Chl-a-j- accumulates partially in the oxidized form (fig. 5b). A supple- mentary red flash ( ^ 7oo mu) at t, oxidizes the rest of Chl-a-j- immediately. Afterwards all Chl-ai is again reduced in ^ — ^1o~^ sec (fig. 5b and 5b'). Therefore, -fche changes pass the zero line. c) In stronger measuring light (7o5 mu) started at t^ nearly all Chl-ay accumulates in the oxidized form (fig. 5c). Therefore in stronger measuring light practically only positive absorption changes take place. These effects are the basic for the follow- ing chapter. 7. The electron-acceptor Z of Chl-a j Changing from soft to stronger measuring light we should aspect - according to the scheme in fig. 5 - a shift from negative ab- sorption changes to positive ones as indicated by fig.5a,b,c. Such a shift has been observed in chlorella (fig. 5a and 5b). However, in chloroplasts without any Hill-oxidants in stronger measuring light no shift to positive absorption changes takes place (fig. 6 top). This indicates that no accumulation of Chl- a+ has occurred. The reason may be that Chl-a+ can be reduced b^ a backflow of elacirons from the electron acceptor Z: Z < Chl-aj < ? If this is true, trapping of electrons of Z should prevent the backflow and allow an accumulation of Chl-a^ : hvj tivjj ox.S-j- < Z < Ghl-a-j- < ■? Indeed with the addition of a number of oxidized substances (ox.S-p) positive absorption changes occur which are characteri- stic for accumulation of Chl-a^ (fig. 6 bottom). This effect gives two important informations. 21 B. Eumberg, F. Schmidt-Mende , J. l^eikard and H.T. Witt a) The shift from negative to positive values indicates which added oxidized substances are electron-acceptors of Z~ (5). So far as investigated all substances surnamed under ox.Sj in fig. 1o react with Z. S.B. Henriques in our laboratory showed that also TPN-reductase - but not TPN alone - traps electrons from Z~ From the highest redox potential of ox.Sj (methyl viologen) follows for the redox potential of Z~/Z a value of ^-o,^^ Volt (4). Because TPN-reductase is reduced by Z~, TPN-reductase must contain a redox system (1). Arnon (7) et al have shown that this is ferredoxin. TPN-reductase was already described in de- tail (7). b) By shifting the absorption changes of Chl-a-j- from negative to positive values it is possible to separate the difference spectrum of Chl-aj from the overall difference spectrum (fig.1 top) under complete natural conditions (fig. 7) (compare 1,c). 8 . Water as the ultimate electron donor of Chl-a -j- It is long known that CMU blocks especially the oxidation of water (9) • If water is the ultimate electron donor for Chl-aj, any reaction of Chl-aj should vanish in the presence of CIi/IU. hvj ^"^TT oxS-;^ < Z < — Chl-aj < ••• a^O This is true for chlorella but this is not the case in fresh chloroplasts (fig. 8 mid). The reason is again the backflow of electrons from Z~ to Ghl-at , which keeps the cycle in action. Trapping the electrons of Z~ by addition of oxidized substances ox.S-p (i.e. benzyl viologen), results in the disappearance of the changes caused by Chl-aj (fig. 8 bottom). This fact is a fur- ther possibility for the determination of those substances which can accept electrons from Z~ (5). In aged chloroplasts, where Chl-a-p is supplied directly by electrons of red PMS (see 4), the adaition of CMU has no influ- ence on the absorption changes of Chl-a-j- (4) . 9. Some properties of the Chl-aj-reaction by the depe the light-induced changes at 7o5 mu on the ratio of ferro/ferri- cyanide . From such measurements Kok (2) estimated a value of + 0,43 Volt (pH 5 - 9>5)« Our measurements gave nearly the same potential + o,^ Volt (pH 6-8) (fig. 9) (-4-). The pH-independen- ce indicates a pure electron transfer. b) Temperature of inactivation . The Chl-a-p-reaction is stable up to 65''C (1). ^ c) pH-Range . The Chl-a-^ reaction is stable between pH 4 and pH 11 (1). ^~ by means of hVj-background light (7oo - 72o mu) 22 B. Rumberg, P. Schmidt-Mende , J. 7/eikard and H.T. Witt 1 o . Reaction scheme The results of this chapter are summarized in fig. 1o (left). In chlorella and fresh chloroplasts Chl-aj is reduced by electrons originating from water with the help or hVjj-light . After sup- pressing water-oxidation (by aging, extraction of plastoquinone , addition of CMU) , Chl-at can be reduced by backflow of electrons from Z~ or by added reduced FMS or reduced DPIP. In these circum- stances one light reaction cycle (1) of the overall electron transport system of photosynthesis has been completely isolated (1o)(1'1 )(1) . This reaction cycle must be responsible for the sy- stem that operated when Vernon, Zaugg and Kamen (12) used aged chloroplasts with reduced substances which are capable of reduc- ing TFN in the light. II. FHOTQREDUCTION OF X AND PLASTOQUINONE RESPECTIVELY 1 . Analysis of X Out of the overall difference spectrum (under "normal" conditions) we isolated one part: the Chl-aj-spectrum with peaks at 43o and 7o3 mu. The other part with peaks around ^75 mu and 5^5 mu has been investigated already in detail (15)(1o). The results are summa- rized in fig.lo (right). Instead of X we introduced already Q in fig.lo (see below). Part of the changes are caused by a photore- duction of a substance X (redox potential — Volt) . The elec- tron donor is Y (redox potential:?' +o,8 Volt). Y oxidizes water. The natural electron acceptor must be according to the result of the last chapter - finally Chl-at . Added oxidized substances as surnamed under oSjj act as artificial electron acceptors. That the cycle is sensitized by a chlorophyll-a (Chl-ajj) follows from the action spectra (11). In the following we will give fur- ther evidence for the probability of this scheme. 2. Separation of the difference spectrum of X To separate the reaction cycle of X from the natural electron acceptor (Chl-a-j-), one has firstly to trap the electrons of X by addition of ox.Sjj (for instance ox. DPIP). To suppress any reac- tion of Ghl-a-p (backflow of electrons from Z~ to Chl-a-j-), one has secondly •fco oxidize permanently Chl-aj by further addition of ferricyanide . Ferricyanide keeps also DPIP permanently oxidiz- ed. Fig. 11 shows that with addition of DPIP (1o"^ M/1) and ferri- cyanide (1o~5 M/1) the difference spectrum of Chl-aj vanishes and that of X is separated (1o)(2o)(1). In these circumstances a second light reaction cycle (II) of the overall electron-trans- port system has been completely isolated. This reaction cycle must be responsible for the system which was described by Losada et al (11) (oxygen-producing chloroplasts with ferricyanide and DPIP) . 3 . Relation of X and Plastoquinone-content a) Bishop (lA-) has shown that plastoquinone is somewhere in- volved in the electron transport system of photosynthesis. The 23 B. fiumberg, F. Schmidt-Mende , J. Weikard and H.T. Witt redox potential of plastoquinone ( — Volt) is very similar to that of X (--^0 Volt). This encouraged us to look for the rela- tion of X and plastoquinone. b) Petroleum ether extraction of chloroplasts (fig. 12) re- sults in a decrease of the absorption changes of X. Following condensation of synthetic plastoquinone results in a complete reappearance of the difference spectrum or even more (15% Both results indicate that X is closely related or identical with plastoquinone (15) • But plastoquinone in vitro is not ac- companied by absorption changes at 475 S-nd 515 niu. If X is iden- tical with plastoquinone, we have to assume that in vivo the re- action of plastoquinone influences the absorption of its sur- rounding pigments, resulting in changes at 475 and 515 niu. 4. The difference spectrum of plastoquinone Absorption changes which are directly related to the reduction of pure plastoquinone in vitro show changes with a maximum at 254 mu (15). Corresponding changes should occur in vivo. V/e have observed light induced changes in active chloroplasts (17) which correspond to those in vitro. 5. Separation of the difference spectrum of plastoquinone We reported that with addition of DPIP (ferricyanide in excess), the difference spectrum of Chl-a-j- vanishes while that of X is furthermore in action. Because or the demonstrated relation of X and plastoquinone we supposed that under these conditions also the difference spectrum of plastoquinone should be still in ac- tion. Measurements at 254 mu show indeed that with addition of DPIP and ferricyanide the magnitude of the changes at 254 mu are influenced not at all while those of Chl-aj at 7o5 mu are com- pletely quenched. 6. Relation between absorption changes of X and plastoquinone The absorption changes of plastoquinone (254 mu) and X (475 and 515 mu) were measured in chlorella and chloroplasts (with DPIP + Ferricyanide) in relation to light intensities, duration of the flash, temperature and pH. We added also different concentratiom of CMU and changed the content of plastoquinone by different ex- tractions. Under all these conditions the magnitude as well as the time course at 254 mu and 515 niu correspond to each other. In fig. 15 the changes at 515 and 254 mu are compared as function of extraction and recondensation of plastoquinone. Fig. 14 shows the decrease of both changes with increasing extractions of pla- stoquinone. The changes of Chl-aj (upper curve) at 7o5 mu are influenced not at all. This indicates in a different way (com- pare 5) that plastoquinone seems not to be involved in the re- action cycle of Chl-aj. 7 . X and plastoquinone respectively as electron acceptor of water oxidation That X is the electron acceptor of lastly water in reaction cyc- le II, we gave in the last years several evidences. A further 24 B. Rumberg, P. 3chmidt-Mende , J. Weikard and H. T. Witt proof that X and plastoquinone resp. is the electron acceptor of water is given in the following experiment: We measured simulta- neously with the extraction and recondensation of plastoquinone and the corresponding disappearance and reappearance of the ab- sorption changes at 515 mu the oxidation of water (by Op-produc- tion measurements). Comparing such results (fig. 15) it can be seen that the magnitude of the changes at 515 mu parallels the rate of Op-production. On the other hand the changes at 515 parallels the changes at 25^ mu (fig. 13, 14). 8. Reaction scheme The results show that plastoquinone reacts at the position of Z. X is identical with plastoquinone or closely related to it. Therefore we introduce in the already shown scheme instead of X plastoquinone Q (18)(15). III. THE ACTION SPECTRA OF THE TWO SEPARATED REACTION CYCLES I AND II The action of cycle I is optically represented by the absorp- tion changes of Chl-aj at 45o mu and chemically by the rate of reduction of oxSj, (i.e. benzylviologen) . The action of cycle II is optically representated by the magnitude of the absorption changes at 515 mu and chemically by the rate of oxidation of wa- ter (Op-production). Pig. 15 (top) show the optically measured action spectra (11) of the two separated cycles and on the bottom the chemical ac- tion spectra are shown. (The optical action spectra could be measured with a much higher precision than the chemical ones. The chemical action spectra are preliminary ones.) The results are in agreement with the Emerson-effect (2o): The peaks indi- cate those types of chlorophylls which provide the two reaction centres Chl-aj and Chl-a-j-j per energy migration with light ener- gy (11). These different -feypes are noted in fig.lo. The long wave-length limit for reaction cycle I is at '^ 73o mu, the long wave-length limit for reaction cycle II is already at ,r^ 7oo mu . The action spectrum for the reduction of plastoquinone was fur- thermore investigated directly at 254 mu . It has a long wave- length limit at 7oo mu in accordance with the measurements at 515 mu and of oxygen production. IV. THE COUPLING OF REACTION CYCLE I AND II 1 . F irst demonstration Optically demonstrations of the existence of two coupled light reactions were done in the following way: Kok (2) showed that far red light causes oxidation of P 7oo at 7o3 mu and shorter wave-length its reduction. Duysens (21) showed in red algae that red light causes oxidation of cytochrome and shorter wave-length its reduction. We showed in green plants (18) that far red light causes oxidation of cytochrome while shorter wave-length causes the reduction of X which reduce cytochrome in the dark. 25 B. Rumberg, F. Schmidt-Mende , J. Weikard and H.T. Witt The above reported results shorn that when under natural con- ditions (chlorella) there exists a coupling between the reaction cycle I and II, it should be possible to demonstrate the coupl- ing of the cycles I and II through the absorption changes of Chl-a-. directly at ^3o mu . (Because of the strong fluorescence in the red region a coupling at 7o3 mu can directly not be de- monstrated) . On the other hand we selected with plastoquinone a first che- mical representative of cycle II and it should be possible to de- monstrate the coupling of cycle I and II also at 25^ mu. J^ccording to the action spectra Ghl-a-j- can be oxidized with light a^ 7oo mu, i.e. 72o mu (decrease at 43o mu) and Chl-a-j- should be reduced with -^7oo mu light, i.e. 55^ mu (increase at 430 mu). 63^ mu light produces namely reduced plastoquinone (Q,~) and when there exists a coupling between reaction cycle I an.d II, Q~ should provide Chl-at with electrons. Vice versa plastoquinone can be reduced with -=• 7do mu light, i.e. 53'^- mu (decrease at 254 mu) and Q~ should be oxidized with >'7oo mu light, i.e. 72o mu (increase at 254 mu). 72o mu light produces namely Chl-at (see above) which accepts the electron of Q," when there existi a coupl- ing between reaction I and II. The results in fig. 17 confirm these predictions. 2. Second demonstration It should be pointed out that as yet in all cases coupling of two light reactions in photosynthesis was demonstrated only by two different light colours. The following experiments show a different and completely independent demonstration of coupling. We separated chemically the reaction cycle 1 from cycle II and investigated the different properties of these separated reac- tions (chloroplasts of spinach) (1o)(2o)(1). Some typical properties of reaction cycle I are for instance: a) pH-range between 4-11; b) independence from the presence of CMU. (CMU blocks water oxidation); c) practically independent from aging. Corresponding properties of reaction cycle II are: a) pH-ran- ge between 5 - 8; b) strong dependence from the presence of CMU; c) strong dependence from aging. '/iThen however both cycles are completely coupled (fresh chloro- plasts with Hill-oxidants ox.Sj or whole chlorella cells), the behaviour of reaction cycle I ^demonstrated for instance by ab- sorption at 7o3 mu) and the behaviour of reaction cycle II (de- monstrated for instance by absorption changes at 515 mu)should show one and the same dependence as function of pH, CMU age, etc. Figs.18-2o show the result. The influence is the same at both wave lengths (5). This proves in a completely different way the quantitative cooperation between both reaction cycles (5). 26 B. Rumberg, P. Schmidt-Mende , J. Weikard and H.T. V/itt V. A POOL OF PHOTOACTIVE FLASTOQUIHONE From the magnitude of absorption changes of Chl-aj at 7o3 mu and of Q at 254 mu it follows (together with the extinction co- efficients of pure chlorophyll and pure plastoquinone) the ratio of Chl-a-^: Q « 1 : 4. Q is the photochemical engaged plastoqui- none; the total amount of plastoquinone is ten times higher (22)- This means that there exists a pool (reservoir) of photoactive plastoquinone between both light reactions (s. fig. 21) shuttling electrons from reaction centre I to II. This result is also in accordance with the fact that after extraction of plastoquinone this substance can be recondensed into the same position. This seems to be possible only when there exists a gap between reac- tiion centre I and II which is filled up with larger amounts of plastoquinone. The redox potential of this pool is probably con- trolled by oxidants (Op) as well as by reductants. This pool seems to be the reason for most of the so-called "induction pho- nomena" in photosynthesis. VI. REAC TION MECHANISM OF THE OVER ALL PROCESS The coupling of reaction cycle I and II leads to a reaction scheme of the over all process of the electron transport in pho- tosynthesis which is mapped out in (1o)(1). An intermediate bet- ween both cycles is probably cytochrome. This was demonstrated in (ia(4)(1o)(1). 27 B. Rumberg, P. Schmidt-Mende , J. Weikard and H.T. Witt Literature ( ( ( ( ( 1) 2) 3) ^) 5) ( ( 6) 7) ( ( 8) 9) do) (11) (12) (15) (1^) (15) (16) (17) (18) (19) (2o) (21) (22) Witt,H.T.,Kuller,A.,and Rumberg, B. ;Nature ,192.98? (1965) Kok,B.,and Hoch,G., Light and Life, 597 (1961) Kok,B., Biochim.Biophys.Acta, ^, 52? (1961) Witt,H.T.,Muller,A. ,and Rumberg B. ; Nature ,192,967 (1961) Witt,H.T. ,Miiller,A. ,and Rumberg B. ; Colloque internatio- nal sur "La Photosynthese" a Gif s/Yvette (1962) Rumberg, B.,Muller, A., and Witt , H.T. jNature ,191,85^ (1962) Davenport, H.E. ,Hill,R. , and Whatley,F.R. jProc .Roy .Soc .B, 159,5^6 (1952) ^^^^^ San Pietro,A., and Lang, H.M. ; J.Biol.Chem. , 251, 211 (195^ Tagawa,K.,and Arnon,D.I.; Nature ,1^, 557 (1962) Bishop, N. I.; Biochim.Biophys.Acta, 22, 2o5 (1958) Nakomoto,T. ,Krogmann,D.W. ,and Vennesland ,B. ; J.Biol. Chem., 25ii, 2785 (1959) Jagendorf , A.T. , and Margulies,M. ;Arch.Biochem. Biophys . , 9o, 184 (i960) Miiller, A. , Rumberg, B. , and Witt , H.T. ;Proc .Roy .Soc .B, 157 , 515 (1963) Miiller, A., Fork, D.C., and Witt,H.T.; Z.Naturf. ,18b, 142 (-^65) Vernon , L . F . , and 122 (1954) Vernon,! .P. ,and Kamen,M.D.; Arch. Biochem. Biophys . , ^, Zaugg,W.S.-, Biochem. Witt, H. T. , and Moraw,R.,Z. Phys.-Chem. 255, 285 (1959) Witt, H. T. , and Miiller, A. ; 1 (1959) 235 , 2728 (i960) Neue Folge, 2o, Z.Phys.-Chem. ,Neue Folge, 21_, Bishop, I. N., Biochem., 4^, 1696 (1959) T.; Z.Naturf .,18b. 139 Weiksrd, J. ,Miiller,A. , and Witt,H. (1963) Crane, F.L.; PlantPhysiol . , 24, 128 (1959) Klingenberg,M. , Miiller, A. , Schmidt-Mende, P. ,and Witt, H.T. ; Nature, 194, 579 (1962) Witt, H.T., Miiller, A. , and Rumberg, B.; Nature ,191, 194 (1961) V;itt,H.T. , Miiller, A. , and Rumberg, B.; Angew.Chem., 75,5o7 (1961) Losada,M. ,Virhatley,F.R. ,and Arnon,D.I.; Nature , 19o, 6o6 (1961) Govindjee,R. ,and Rabinowitch,E. ; Science , 152 , 555 (196o); Biophys. J., 1,73 (196o); 377 (1961). Govindjee ,R. , Thomas, J. B. , and Rabinowitch,F. ; Science ,152 ,421 (i960) Duysens,N.M.; Nature, 175 , 69? (1954) !fi£ , S"! (,i2; ferricyanide was added as Hill oxidant. Fip;.16. -faction spectra of reaction cycles I and II. Top: Measur- ed by the magnitude of the absorption changes at ^35 mu in aged spinach chloroplast fragments in the presence of 1o~^ M/1 PMS plus I0-2 M/1 ascorbate (2o"C, pH = 7,2) and at 515 mu in normal spinach chloroplast fragments in the presence of 1,5.1o ^ M/1 DFIP plus I0-2 M/1 ferricyanide (2o C, pH = 7,2). Bottom: Measure ed by the rate of Op-production of illuminated spinach chloro- plast fragments in the presence of 2.1o"5 M/1 DFIP plus 6.10"-* M/1 ferricyanide (2o 0, pH = 7,2) and measured by the rate of 0^-consumption of illuminated aged spinach chloroplast fragments ife the presence of lo"^ M/1 DFIP plus 5-1o ^ M/1 ascorbate plus 5.1o"5 \,/\ benzylviologene (2o G, pH = 8). (Measured by U.Siggel in our laboratory) . Fig. 17. Top: Time course of the absorption changes_at ^3o mu in spinach chloroplast fragments in the presence of I0 ^ M/1 benzyl- viologene illuminated with 72o mu-light (1/2 sec), followed by 658 mu-light (10""^ sec) (2o°C, pH = 7,2). Bottom: Time course of the absorption changes at 25-^ mu in chlorella cells illuminated with 658 mu-light (1/5 sec), followed by 72o mu-light (1 sec) (2o^G, pH = 7,2). Fig. 18. Top: Absorption changes at 515 mu and 7o5 mu in whole spinach chloroplasts in the presence of 1o~^ M/1 ferricyanide as function of aging. (2o C, pH = 7,2). Bottom: Absorption changes at 515 mu and 7o5 mu in whole spinach chloroplasts in the pre- sence of 1o~^ M/1 ferricyanide as function of pH (2o C). Fig. 19. Half-life of the absorption changes at 515 mu and 7o5 mu (strong measuring light) in chlorella cells in dependency of temperature (pH = 8). The chlorella culture differs from that used in the measurements of fig. 5* (In chloroplasts identical kinetics at 515 mu and 7o5 mu could as yet not be established). Fig.2o. Absorption changes at 515 mu and 7o5 mu in whole spi- nach chloroplasts in the presence of I0 o M/1 indigocarmine as function of concentration of added ClllU. (2o G, pH = 7,2). Fig. 21 . See text. 30 c .0 I o o Ol c o ^ .,0' IP A ,0' ^w v normil \\ .,0' ■w-' V ^ilh ptirol -tthtr \j fig-l itc-lO- wave-length *10'5mI PM5 31CPMII Asc • 705 m/j i 435 my ? "V^ without additions - liO'C \ f 700 mn fig. 3 fig. 2 (AD, fig. 4 ♦ PMS(c),Asc(3tO M/l) 3-ro-^ lime (sec) 31 T mttniiljf of 100 X Chlij 0%^ ^S ^100% time fig. 5 705my no additions benzyl - viologene WW'^SK fig 6 time 705 m^J 430 mfj fig. 7 WtO'^sec time ferncyanide/ferroc/onidt I ^ -v- fig.9 705 mfj no additions 6 ta^MIl CMU *6 icr^Mii CMU tO-^Mfl bemytnologtne sec time fig. 8 32 Phfouint mrlhciul'ile ChlO-695 Indite"^'"' ^.. c»-} Brntfl rtotogtfw LnlO - OO^ ,4,th,i ..ciogf^ (Chlb-eSO) IPt rrA/cMM \ fig. 10 Of- Sg ___ f5_ 2-6-D>cMoeph9nol--naofitmnel rhfimlpfttnel - inttofihtftei Telufl»n» blur Ttuomn* 7^ HjO Chlo-67* Chi b- 650 hiji O , lO'J _ «> O) c /\ o •C / V V ♦DP/P ♦ FtCy ■I ID"-' 1 1 1 1 1 wave-length fig. J J c o e- o to -Q D C o -c o ® normal fflrocled plailo- quinone ^ recondensed 1 / \ X I *^^ V y ' V/ ® o. /■•■ ■••-Ss. \-> f V^ 0) fS. C\ h:::^ 450 500 550 mfj wave -length fig.l2 33 o •c "i" 5»5m(/ '0001 BH J^S • aooi i^_ ^^S •0.001 BH 9^ -aooi -aooi -QDOI normal portly extracted plaslo- quinone recondensed fig.l3 c "V/ *\ ^"^A o ^^^ Q \s^ V- o \.,_^ i5 \^i D A 705 mil O a 2Umti "N. ■ • srs/i^ • ^s. ■ r n ■c • ^ increasing extraction with petrol-ether — fig.U BOmin mg Chlorofitiyll 3 o o'^ 1 "I 1 III c 9 /0(\ change of //\V absorption \ 'C^^atS}5mp •5 ^ y \y\ change of yA \ \y absorption 4. 1 1 M\, 600 650 700 750 mfi ¥ •Q E O 2 ^. to O _ «, O o> c o fig.15 1 Z Jl J. □ Z^ptastoauinone extracted \^ recondensed 600 650 700 wove -length mfj figlS 34 TPN- Z^ Chi a J Q Q Q Chi a J Y^H20 Q J fig. 21 35 PHOTOSYNTHETIC ELECTRON TRANSPORT Bessel Kok I, The first photoreaction of photosynthesis mediated by P700 It is well established that in aerobic photosynthesis two photoreactions occur. Photosystem I is sensitized by "long wave" chlorophyll which feeds absorbed quanta into a special long wave pigment "P700" (13). Upon excitation P700 loses an electron and remains in a bleached form which exerts the prop- erties of a weak oxidant (normal potential P/P"'"= 0.^3 volt). The electron acceptor ("X~") is a strong reductant of a po- tential lower than -0.42 volt (one can show that it reduces methyl viologen) . Figure 1 and Table 1 bring quantitative evidence that photoact I is mediated by P700: Fig. 1 shows an experiment made by Hoch and Martin (l) with fresh chloro- plasts. The reduction of TPN was studied as a function of Equivalence Between Rates of P7Q0 Turnover and TPN Reductio Reaction mixture (pH 7.8) contained per ml(in u. moles) : Chlorophyll ; 0.0 17 ; TPN : 0. 5 ; ADP : 1 ; PO4 : 10 ; Mg** : 2.5 ; PPNR .- saturating Illumination 7 10 ml/ /leq. /min P700 = AO. P. 700 (A-B signal) x 1800 (RPM) 80 000 (E mol. ) x mg Chlorophyll fL eq. /min TPNH2 AO D 340(B-W signal) 3100 (E mol/2) x I{min) x mg Chlorophyll P 700 TPN Ratio 2. 1 Z. 1 1 660 670 6B0 690 700 7)0 WAVELENGTH (m/i.) Figure 1 Table 1 wavelength both in the absence and the presence of the poison DCMU. Reduced indophenol was added in the latter case to re- lieve the DCMU inhibition (2). In the unpoisoned chloro- plasts, two quanta of 65O-68O mu light, are required to transfer one electron to TPN, the quantum yield drops severly in wavelengths beyond 69O mu. In the presence of DCMU and reduced dye, however, the quantum yield is relatively low at 36 Bessel Kok short wavelengths but rises with increasing wavelength until it approaches unity (1 eq./hV). Obviously, in long wave light we observe only photosystem I, a process which occurs with a quantum yield of one and can retain a considerable fraction C^ ^0%) of the absorbed light as chemical energy. The experiment described in Table 1 was made with our differ- ence spectro-photometer (3) in which a sample is exposed to a series of light flashes (l800 per minute). The dark periods are long enough to allow dark conversion of the photoproducts made in each flash. The apparatus measures cyclic absorbency changes of intermediates as well as the net result of many cycles, provided a color change accompanies the events. In Table 1 we observed the repetitive bleaching (in each flash) and re-reduction (in the dark) of P700. If for every P700 which is bleached, one electron is transferred to the primary reductant (X) and from there to TPN, then the total amount of TPNH2 accumulated should correspond to the sum of all P700 molecules which have turned over during the time the flashing light was given. This indeed proved to be the case if we assumed a (change of) molar extinction AE=8xlO^ L/mole. cm. for P7OO -a value typical for the red absorption band of a chlorophyll. This good agreement, together with the high quantum yield observed in Fig. 1 yields quantitative support to the proposal that F7OO is the photoconverter of photosys- tem I in photosynthesis. II, Photoreduction and -oxidation of indophenol dye, site of phosphorylation. In the long wave light or in the presence of DCMU the sec- ond photoreaction of photosynthesis is inoperative. Earlier (4) we have shown that DCMU does not really inhibit the photo- reduction of DCPIP by chloroplasts, but only causes the reduced dye to react back with photo-oxidized P700 (whose Reduction ot DCPIP by weak flashing light of two wavelengths. The atriount reduced per flash is compared to the net rate during one light-dark period. 725 mil values in parenthesis are corrected for absorbed intensity equal to the 674 light. Wavelength actinic light mji Fractional absorption % Rate per flash 10-5O.D. Rate per 1/30 sec. 10-^ O.D. Ratio flash yield net rate 674 ^80 49 3i 1. 5 725 = 6 9 1= 100) 0.28(= i) 32 Table 2 37 Bessel Kok normal reduction by the second photoact is impeded by the poison) . Table 2 shows that essentially the same happens in long wave light: In this experiment, performed like the one shown in Table 1, we simultaneously measured the amount of dye re- duced in each flash — whether reacting back or remaining in the reduced form — and the net amount of DCPIPH accumulated by all flashes. If N flashes of 674 rau light are given, an amount of DCPIPH accumulates which nearly equals N times the amount reduced per flash. However, if 725 mu flashes are given instead, only 1/30 of this amount accumulates, 97% of the reduced dye is re-oxidized in the dark periods by P700 . The long wave drop of the quantum yield of dye reduction (5) therefore is due to a lack of the normal reductant generated by photosystem II which reacts with P700+ much faster than DCPIPH. Actually, if one corrects for the low fractional absorption of 725 mu light, the data of Table 2 show that in this wavelength area dye is reduced with twice the quantum yield observed in 67^ light (analogous to the reduction of TPN in expt. Fig, 1). We may conclude that in these discussed experiments DCPIP is reduced at the same locus as TPN: by the primary reductant X~ made in the first photoact. It is clear then that in the absence of the second photoact (DCMU or long wave light), indophenol dye mediates a vigorous cyclic electron transport. Trebst and Eck (6) showed that this cycle is coupled to ATP formation and therefore must include the site of photophos- phorylation. We agree with Vi/itt, et. al. (7) that at least in fresh chloroplasts DCPIPH does not reduce P700+ directly but via another intermediate. It remains to be proven that this intermediate is cytochrome b^ or plastoquinone such as would be required by hypotheses which correlate one of these com- ponents with photophosphorylation. The hi^h normal potential of the dye and the considerable concentration of the oxidized form which can result from its photo-oxidation (even in the presence of ascorbate cf. k Table 1) argue against this pos- sibility. A more likely site for reduced dye to re-enter the cycle is cyt. f. From an energetic viewpoint, this would exclude ATP formation at this locus (cf. k) , 38 Bessel Kok III . Photo-oxidation of cytochromes and plastocyanin by- detergent treated chloroplasts . Photosystem II is quite sensitive to ageing, heating or detergent, Photosystem I, hov/ever, survives such treatments to a considerable extent. Possibly by opening up the chloro- plast structure, detergent makes P700 accessible for large molecules such as cytochrome c, f, or plastocyanin so that they can be photo-oxidized (like reduced indophenol dye). The photo-oxidation of cytochrome c by this material was first observed by Niemann and Vennesland (8), We may briefly summarize the results of our own recent studies of such photo- oxidations (9)« The process is sensitized quite efficiently by long wave light (the yield approaches 1 eq./hv) involving photosystem I — P700* being the photo-oxidant. Two types of catalysts accelerate the photo-oxidation of f errocytochrome cl Viologen, flavin and other auto-oxidizable, single electron transfer agents stimulate by mediating between primary re- ductant X~ and oxygen. Cytochrome f (Cyt. 552 from Euglena) and plastocyanin (10) stimulate by mediating between P?©©"*" and ferrocyt. c. Figure 2 illustrates the effect of various concentrations of plasto- cyanin upon the rate of cyt. c photo-oxidation in the presence and absence of viologen. Rates approaching 5000 eqs./chl. hour have been observed, severalfold higher than found in 10 12 14 16 18 20 22 24 CONCENTRATION OF PLASTOCYANIN COPPER 28«10"*M Figure 2 photosynthesis or chloroplast reductions. The rates proved remarkably independent of temperature. (Q 10-^1. 3i+ 30"- -5*0) Cyt. f acts very similar to plastocyanin but never 39 Bessel Kok yielded rates higher than ~800 eq/chl. hour. A peculiarity of these catalysts is their effectiveness in weak as well as in strong light — they increase the quantum yield of the reaction. Plastocyanin and cyt. f, if added in substrate amounts in- stead of cyt. c are photo-oxidized themselves with high rates. An interesting feature of these reactions is that the rate is dependent upon the redox state of the substrate. For instance, if cyt. f is added in half-oxidized, half-reduced form, the rate is only half maximal. (In the presence of excess cyt. c, cyt. f is kept in the reduced form and the overall reaction proceeds with optimal rate until the depletion of cyt. c.) 20 40 60 80 7o REDUCED CYTOCHROME f 100 Figure 3 Figure 3 illustrates that in weak as well as in strong light, the rate of cyt. f photo-oxidation is proportional to the ratio cyt, f red. /cyt. f total. To explain this, we assume that external cyt. f equilibrates with a cyt. f mole- cule fixed in the chloroplast matrix in close proximity to P700. A photoact can only be successful if not only P700 but also its associated cyt. f is in the reduced state before the quantum hits. The cytochrome transfers an electron to P700''" immediately after the latter has lost its electron to X, and thus prevents a back reaction between the photoproducts. Such a charge transfer complex between cyt. f and P700 em- bedded in the chloroplast matrix was already shown by Witt et al. (7). In scheme Fig. k we indicated that plasto- cyanin could react in two ways with P700 , directly or via 40 Bessel Kok cyt. c hi/ 1 chlorophyll Figure k cyt, f. In many chloropiast preparations plastocyanin sus- tained much higher photo-oxidation rates than cyt. f, indica- ting, that it bypasses cyt. f and reacts directly with P700. Our preliminary data indicate that the kinetics of plasto- cyanin photo-oxidation do not essentially differ from those of cyt. f oxidation (cf. Fig. 3). This suggests that not only cyt. f, but also plastocyanin has a fixed locus in the chloropiast matrix and can operate in a complex with P700. This intimate cooperation between the photoreceptor, a haem and a copper enzyme, all of high potential, present in equal amounts in the chloroplasts, has prompted some speculations which will be described in the following section. IV. Discussion The first photoact yields besides X~ (re-oxidized by sub- strate) a weak oxidant P700'''. The second (short wave sen- sitized) photosystem is left with the tasks to evolve oxygen (at least to assist this process) and at the same time re- reduce P700"*". One of the products of the second photoact, therefore, must be a reductant "Y~" of a potential lower than +0.^3 volt. Just how much lower, is presently a point of discussion since the chemical nature of Y is unknown. For instance, as mentioned by Dr. Hoch in this symposium, the possibility is not excluded that Y~ is as strong a reductant as X~, This would mean a "parallel" operation of the two photoacts and require an energetic coupling between them by dismutation reactions in the dark. The other extreme is the assumption that the potential of 4L Bessel Kok Y/Y" is only slightly lower than that of P700, i.e. the sec- ond photoact is strictly in "series" with the first one. We have mentioned this possibility on an earlier occasion {k) and illustrate it again in scheme Fig. 5* These two extremes share one feature, namely, that the quantum requirement of the overall process could be less than 2 per electron (<8 per O2): V\/e feel the data of Fig. 1 and Table 1 definitely show a requirement of one quantum to generate X" and P"*". In the parallel formulation the extra energy required to evolve O2 starting from P does not neces- sarily have to come in one quantum "packages" and in the series formulation the energy of 2 (short wave) quanta might suffice to evolve oxygen from a preformed complex. A minimum requirement of 2 hv/el. is more rigidly demanded by the part-parallel part-series scheme [cf. Hill and Bendall (11)3 which is presently accepted by most workers in the field. In this formulation the reductant of the second photoact hvl hvn TPNH, Figure 5 should be able to reduce cytochrome bg or plastoquinone (po- tential of Y/Y~£0.0 volt). The reaction chain connecting the two photoacts: provides an attractive site for photophosphorylation--coupled to the main stream of electron transport. If one does not 42 Bessel Kok accept this intermediate chain Y — ^ P (of. section II) the only reasonable alternative to conceive a generation of ATP coupled to substrate reduction is the locus indicated in scheme Fig. 5 as X~ — ^ X^ (4), This hypothesis of "substrate level" photophosphorylation requires, however, that the pri- mary reductant of the first photoact has a quite low poten- tial. Scheme Fig. 5 contains additional speculations which must appear objectionable to many of you and might not survive future evidence. However, it is not presented as the ulti- mate truth but rather to illlustrate some possibly useful thoughts. One feature is a "sharing" of traps by the two photoreactions visualized by the identification of P700 with the light collector of photoreaction II, i.e. ^uysens' "Q" (12) and »'itt's "Q II" (7). The asymmetric shape of the 700 mu difference band (cf. 13, Fig. 2 and 13 ) suggests a shift of the 700 mu absorption to a {-^^0%) weaker band at a ('-'13 mu) shorter wavelength. Photo-oxidized P700 (P690'*") thus could conceivably again function as a light trap — compare e.g. the photochromic back and forth shift of the plant pigment phytochrorae (l4). The second photoact then would be sensitized by all wavelengths except those beyond 69O mu (of course, one must assume that P69O undergoes alternate dark steps as well). Such a competitive sharing of trapping pigment by the two photoacts allows a self-regulating sensitization mechanism ("spill over"), in line with earlier thoughts of Franck (15) and Myers (I6). Our fluorescence data, although supporting this spill over from one photosystem to the other, do not seem to favor an identification of the conversion centers and indicate a more complex type of mechanism (this volume). A main argument for closely connecting the two traps i.e. for a short path between the intermediate products of the two photoacts — regardless of their nature--is the following: The spectroscopic evidence such as obtained by Witt, ^uysens, and ourselves for P700 operating in a charge transfer complex with cyt. f and probably plastocyanin, subject to a push-pull operation by two photoacts, is impressive. However, most of the information originates from measurements under extrem-e or abnormal circumstances. Under favorable conditions of ef- ficient, steady state photosynthesis in whole algae, neither P7OO nor cyt. f can be observed by our flashing light method which is geared to the "time constant" of photosynthesis and has a time resolution of a millisecond. With their much 43 Bessel Kok faster method Witt et al. (7) still fail to observe the reduc- tion of cyt. f, presumably carried out by a dark step follow- ing the second photoact. (Spectroscopic measurements which use continuous actinic illumination cannot discriminate be- tween compounds directly involved in the electron transport chain and those which are indirectly affected.) Unless one considers a complete re-interpretation of available data, one must assume that the early events in photosynthesis occur with such extreme rapidity that they escape observations with present methods. Also, this feature seems to leave_little room for an intermediate electron transport chain Y — > P . Finally, one cannot help drawing an anology between the (copper and haem containing) cytochrome oxidase in respira- tion and the peculiar combination of plastocyanin and cyto- chrome f in photosynthesis (cf. section III) — reason why we have suggested in scheme Fig, 5 that photosynthetic oxygen evolution might be a light driven reversal of the terminal respiration step (4). Presently, the evidence concerning the photosynthetic light reactions is so much in flux and of qualitative nature that the speculations forwarded do not appear too much out of line. V. Acknowledgements The author gratefully acknowledges the collaboration of Mrs. Pat VVoolf and Mr. Hans Rurainski. The work reported was supported in part by the National Institutes of Health, the Air Force Office of Scientific Research, the Air Force School of Aerospace Medicine and the National Aeronautics and Space Administration. 44 Bessel Kok. REFEKENCES 1. Hoch, G. and Martin, I., Arch. Biochem. Biophys., 102 , 430-438 (1963). 2. Vernon, L. P. and Zaugg, W. S., J. Biol. Chera., 232» 2728-2733 (i960). 3. Kok, B., Plant Physiol., Jit, 184-192 (1932). 4. Kok, B.; Cooper, B., and Yang, L., In: Microalgae and Photosynthetic Bacteria, Plant and Cell rhysiol., 373- 396 (1963). 3. Chen, S. L., Plant Physiol., 22, 33-^1 (1932). 6. Trebst, A. and Eck, H., 2,.f. Naturf. l6f_, 453-^61 (I96I). 7. Witt, H. T.; Muller, A., and Humberg, B. , Nature, 192 , 967-969 (1961). Nature, 197, 987-991 (1963). 8. Nieman, R. H.,and Vennesland, B., Plant Physiol., 3iti 2^^-26Z (19^9). 9. Kok, B.; Rurainski, H., and Harmon, A., Plant Physiol., In press. 10. Katoh, S., Nature, I86, 333-3^ (I96O). 11. Hill, R. and Bendall, F. , Nature, I86, 136-137 (I960). 12. Duysens, L.K.M. and Sv;eers, H. E., In: Microalgae and Photosynthetic BtiCteria, Plant and Cell Physiol, 333- 372 (1963). 13. Kok, B., Biochim. Biophys. Acta, _22, 399-^01 (1956). Biochim. Biophys. Acta, _48, 327-333 (I96I). 14. Hendricks, S. 3.; Butler, W. L., and Siegelman, J. Phys. Chem., 66, 2330-2535 (1962). 15. Franck, J., Proc. Nat. Acad. Sci., 44, 9^1-9^8 (1958). 16. Myers, J. and Graham, J. R., Plant Physiol., 38_, 105- 116 (1963). 45 FLUORESCENCE STUDIES Bessel Kok I, A photoinitiated emission at 698 mu ♦ Brody and Brody (1) mentioned that upon cooling to 77*'K the 685 mu fluorescence maximum shifts to 69O mu. We ob- served, hov;ever, that the location and intensity of the 685 mu band ("FSS^") are practically independent of temperature. Below -150°C a distinctly separate band develops at 696-698 mu (denoted "F700") which keeps increasing with decreasing tem- perature until in many cases it is the most pronounced emis- sion (cf. Figs. 2, k, and 5). This emission occurs in all organisms we have investigated, in acetone extracted or deter- gent treated chloroplasts, but not in solutions of chlorophyll. In fresh chloroplasts, leaves or algae, a striking feature of this emission band is a requirement of light in order to fully develop. Figure 1 (top curve) shows an experiment made in the following way: a sample of chloroplasts was kept in i E 3 Efnan "" NORMAL ^^^ — " / PHOTOINHlBITEDs. /•^Eo 1 1 1 2 3 4 5 6 « ON SECONDS 7 8 9 Figure 1 the dark for a few minutes and then cooled to liquid nitrogen temperature — also in darkness. Upon addition of the exciting light, the fluorescence intensity rises instantaneously to a certain level (Eq) and from thereon much more slowly until it reaches a final value (Ejj^ax^* An immediate rise of fluorescence is typical for pigment in solution, for most of the low temperature emission at 730 mu ("F730") and also for the 685-697 bands in aged or heated chloroplasts. It indicates that already, before light is 46 Bessel Kok given, the responsible pigment molecules are in the fluores- cent state. The exponential build-up of part of the 698 emission in fresh leaves or chloroplasts implies that the responsible pigment is not in the fluorescent state to begin with, but converted into it by light. We therefore, must assume the presence of trapping centers which cease to be traps as soon as they have received a quantum. One can conceive two possible mechanisms: (a) the trapping pigment bleaches upon excitation and ceases to absorb light from surrounding pigment so that the latter is free to fluo- resce, until the trap is again restored in a consecutive pro- cess, (b) the trapping pigment does not bleach but converts an associated molecule. Unless the latter conversion is re- stored in a consecutive reaction the next quantum cannot be used in photochemistry and will be re-emitted by the trapping pigment. In either case, one can determine the number of traps by measuring the number of quanta required to raise the fluo- rescence from Eo to Emax* Assuming a quantum requirement of one per trapping molecule and absence of restoration reactions at 77*K, our measurements indicate a concentration of 1 trap per about —50 chlorophyll molecules. Though still preliminary, our data probably show the correct order of magnitude; Kaut- sky et al. (2), observed at room temperature a similar but much faster initial rise of fluorescence and computed the presence of 1 quencher per --^OO chlorophylls. Using Porphy- ridium, Duysens and Sweers (3) arrived at a ratio 1:150. V.'e indeed observed at 77°K for F7OO (and probably F685 as well) a 10 times slower rise than at 300*'K for F685. This detrapping at low temperature indicates a photochemi- cal phenomenon probably correlated with a primary act of photosynthesis. Since as far as we know no corresponding color change accompanies this detrapping, the second of the two mechanisms discussed above might be involved. The number of traps in this system is much (^^lOx) greater than expected on the basis of the classical photosynthetic unit. Under appropriate conditions the rise time of F685 at room temperature reveals the same small photosynthetic unit. A trapping pigment present in so high a concentration ( — '2-5% of Chi.) might be detectable by rather direct methods. In the following we will discuss some further observations concerning long wave absorption and emission bands which could 47 Bessel Kok be related to it. II. Correlation between P700, C7OO and the low temperature emission bands. With leaves at yy^K Butler (.k) observed a band at 705 mu in the absorption spectrum as well as in the excitation spec- trum of fluorescence. He assumed (a) that this pigment "C705" was the emitter of the strong 730 mu fluorescence band found earlier by Brody (l) at 77''K and (b) that 0705 was identical to the long wave chlorophyll "P7OO" functioning as the trap- ping center in the long wave photoact (system I) of photosyn- thesis. Identity of P7OO and C705 appeared unlikely because the data indicated a concentration of C705 as high as 2-5% of total chlorophyll, whereas we never observed P700 in a concen- tration higher than 1 per 3OO or 400 chlorophylls (5). An identification of P7OO with F73O did not appear likely either: The photochemical bleaching of P700 is irreversible at 77°K and a fluorescence emission would have to come from its oxi- dized form. The possibility is not excluded, however, that P700 ox still does absorb, viz . , at 69O mu (6). The follow- ing experiments bear on these questions: Expt. J^ig. 2 shows emission spectra (77°K) of chloroplasts briefly treated v/ith increasing concentrations of acetone in water (7). The data show that a low concentration of acetone (20%) decreases the 730 mu fluorescence relatively less than the short wave bands. Higher concentrations produce band CHLOROPLASTS TREATED WITH INDICATED PERCENTAGES ACETONE TT-K 700 720 WAVELENGTH mfi Figure 2 shifts and additional emissions in the long wave region, and also lower the yield severely. (The numbers in brackets indi- cate the factors applied to match the curves.) 75% Acetone 48 Bessel Kok removes practically all chlorophyll and P700 and the extracted material shows an emission spectrum identical to that of a dilute chlorophyll solution. Treatment with 65-70% acetone, which barely yields a loss of P700, results in material which shows relatively less fluorescence at 730 mu and relatively more at 685 and 698 mu. As was to be expected, the emission of such preparations is not affected by the redox state of P700. The data of Fig, 2 indicate a closer correlation of P700 with the two short wave bands than with the long wave emissionCs). SCENEOESMUS A rwrmol mulont • 8 1^ 77»K / J Difference .,,^^_ 77°K // \\ SCENEOESMUS 6 300* K / / \\ MUTANT #8 'A \ / / A WILD TYPE 5 - x6— n' / / / \ // \ " 4 - I / 1 / ( / 1 / / / ''xT\ \ 3 / / ; / / / 1 1 1 1 ' ^J ^\ A 2 I 1 1 ' / -''' ^^^ v\ I 1 / / / / / '/ / j^ ' ^^^^^"^::F -^t^ . .^'' 600 550 700 WAVELENGTH mp. 700 750 WAVELENGTH m^ Figure 3 Figure h Figures 3 and k show data obtained with normal Scenedesmus cells and mutant #8 of Dr. N. Bishop. This mutant is capable of performing the quinone Hill reaction but not photosynthe- sis and photoreduction. Vi/eaver and Bishop (8) noticed that it lacked the light induced fast EPR signal, according to Beinert, et al., (9) due to the oxidized form of P700. lie indeed, did not find a light induced turnover of P700 in this mutant or extracts prepared from it. The difference between the two absorption spectra in Fig. 3 shows a distinct band at 700 mu amounting to ^10% of the absorbance at 678 mu. This confirms Butler's (k) and Brown and French's (lO) earlier ob- servations of long wave absorption bands. One is tempted to identify "C7OO" with the small amount (-5%) of long wave "oriented" chlorophyll which will be discussed in this sym- posium by Dr. R. Olson (cf. also 11). The high concentration of C7OO and its failure to undergo reversible photobleaching, argue against its identity with photocatalyst P700. On the other hand, the ability of the mutant to evolve oxygen |be it at lower than normal rate (8)] despite the virtual absence of C7OO suggests that both pigments are part of photosystem I. 49 Bessel Kok One could speculate that (some 20 molecules of) C700 func- tion as energy collectors for (one molecule of) P700. Ab- sence of CyOO in the mutant then would explain the lack of a light induced P700 bleaching and EPR signal — energy transfer being impeded. ( This hypothesis still allows the presence of pyOO in the mutant — which remains to be proven). An alternate hypothesis identifies C70O with P700: One could conceive that one of the oriented C700 molecules upon excitation loses an electron to primary acceptor (X) and re- gains it from cytochrome f or plastocyanin (6). If X, cyt. f and P.C. were only present in 1/20 the concentration of C700 (1/400 chl.) one would observe at any time only one "P7OO" per 400 chlorophylls. This hypothesis would fit sensibly with a rigid orientation of C70O around the reaction loci (11). Ill, Sensitization of the various emission bands. Figure k shows the fluorescence emission spectra of the two types of Scenedesmus. At room temperature the 685 fluo- rescence of the mutant is as high or higher than that of the normal cells (even if the latter are poisoned with DGMU to stop energy flow in order to obtain comparable conditions). We are not certain whether the anomalous emission between 700 and 760 mu in Fig. 4 is typical. At 77®K the mutant fluo- resces stronger than the v/ild type at 730 mu, relatively weaic at 685 mu, and the 698 band is practically absent. Figures 3 and k show a distinct correlation between C7OO and the fluorescence at 698 mu; both are practicaJLly absent in the mutant. The simplest explanation is that C7OO is the emitter of F7OO. The data, furthermore, indicate that the (77*'K) fluores- cence at 730 mu does not directly originate from either C7OO or P700; F73O is high in the mutant lacking C70O. It actu- ally appears as if in the normal alga, C7OO functions as a quencher for F685 at room temperature and possibly for F730 at 77*'K. Quenching at room temperature could at least partly be an indirect effect: if C70O and/or P7OO operated in photo- system I they would provide substrate for photosystem II, e.g. "Qox" in (3). However, quenching of F730 at low temperature must be due to a competition for absorbed quanta between C7OO and "C730" (the chloro:chyll responsible for F730). We now meet a difficulty: Butler (4) observed, and we con- firmed, that in leaves, chloroplasts , and green algae (where F73O is found at 7l6 mu) the sensitization of F730 shows a 50 Bessel Kok maximum at 700-705 mu. This 700 maximum is quite weak in mutant #8 and subtraction of the excitation spectra measured with the two algae (as done in Fig. 3 for the absorption spectra) yields a distinct band at 698 mu, present in the normal alga only. Obviously C700 sensitizes the emission at 730 mu. This might still be compatible with a quenching of F73O if one assumes that C70O transfers energy from sensitiz- ing pigment (Chl. a) to P730, but either can be bypassed or is needed only in a low concentration such as still might be present in the mutant: F700 F730 hV — » Chl a — ^ C700^— ^ 0730^ (l) I r In the previous section we have tentatively located C7OO in photosystem I, if scheme (l) were correct, F730 should be sensitized by the same pigment system. To check this, we have measured the excitation spectra for F7OO and F730 in chloroplasts and various types of algae. In each species these spectra proved to be quite similar for the two emis- sions — which is an argument for scheme (l). Hovi/ever, the data failed to indicate a correlation with"photosystem I": In Anacystis the excitation of F7OO and F73O revealed the typical ineffectiveness of 68O mu light (1/2 - 1/3 of 63O light) and also lacked the 700 band which is so evident in chloroplasts and green algae. (The red alga TX 27, however, showed a distinct sensitization band at 710 mu.) Figure 3 shows the fluorescence emission of the alga TX 27 excited either by green light (3^6 + 378 mu) which sensitizes photosystem II (and I) via phycoerythrin or by blue light (^36 mu) which mainly sensitizes photosystem I. The multiplication factors required to bring the spectra of Fig. 5 to equal hei^-ht (indicated on the curves) show that green light is much more effective than blue, both at room temperature (4x) and at 77°K (l6x). Although, at 77*K the 730 band is predominant in the blue excited emission, it stiLl is induced (9x) more effectively by green light. For the 683 and 698 emissions this ratio is ^50 fold. The data of Fig. 3 [xn accord with those of Brody and Brody (l)J do not support the simple thesis that "F730" originates from photosystem I. They further show that F7OO as well as F683, even more exclu- sively than F730,are sensitized by green light i.e. by photo- system II. Actually, also the excitation spectra measured with green cells--although more difficult to analyze — favor this conclusion. 51 Bessel Kok 700 720 740 WAVELENGTH m/j. Figure 5 10 20301O5O607OBO9OIOO RELATIVE LIGHT INTENSITY Figure 6 IV, Fluorescence yield and electron transport. Figure 6 shows measurements (made at room temperature) of fluorescence as a function of intensity, run concurrently with measurements of the rate of dye reduction by fresh chloroplasts. In confirmation with the earlier analysis of Lumry et al. (12), the data show that efficient electron transport corresponds to a low yield of fluorescence. In the absence of a Hill oxidant the yield is high at all intensi- ties. In the presence of ferricyanide the yield is low in weak light. Addition of NH4CI — which by uncoupling phosphor- ylation accelerates the rate by about a factor two — lowers the fluorescence yield even further (to 25% of the control value). In higher intensities electron transport approaches its saturation rate and the fluorescence yield rises again. Effect of Subsequent Additions upon the Fluorescence Yield of Fresh Chloroplasts in Weak Light Addition Rel. Yield None PPNR (Saturating) 10"% TPN io-5m j)cm 10""^M DCPIP + Ascorbate 100 98 52 86 82 The above table shows that addition of PPNR (which does not induce significant electron transport), has no effect, whereas successive addition of TPN lowers the yield. Addi- tion of DCMU again brings the fluorescence close to the con- trol value — indicating that electron transport stops. A 52 Bessel Kok subsequent addition of indophenol dye and ascorbate, which restores the photoreduction of TPN does not restore the suppression of fluorescence. We may assume that the photo- reduction of TPN in the presence of DCMU and reduced dye requires only the long wave photoreaction (6). Although this conversion can occur with a high quantum yield, (13) it is not reflected in a suppression of fluorescence. DISCUSSION The following scheme seems to satisfy most of the presented data. It assumes that the bulk of the light absorption is carried out by one pigment assembly only. Some fluorescence escapes at all times from the partners of this collector sys- tem. Absorbed quanta drain preferentially in trapping mole- cules II which occur in a concentration of 1 per 50 Chi. (cf. Fig. 1). As long as these traps are unexcited (and kept in the receptive state by a dark reaction) only the accidental (fixed) fluorescence occurs. FIXED F650 F685 J Collector Pigment TRAP3E - DCMU [H] ^ VARIABLE F685 + I / / BYPASS 1 SPILLOVER QUENCHED AT 300° K F700 F730 iChLai]"' ^C700 / C730 P700 02 After a trap II is photoconverted, however, the next excitation will flow on to the long wave pigment C7OO ["spill over" (16)] . Since the coupling with C700 is rather weak, this transfer competes with an additional ("variable") fluo- rescence emission of the collector system [cf . Franck (17)J . If all absorbed quanta flow towards C7OO (as in the absence 53 Bessel Kok. of substrate or the presence of DCMU which blocks alternate path II), the variable fluorescence will be maximal, although system I might still operate at full efficiency (cf. Fig. 6, Table 1). Secondary collector C700 transfers its excitations to the final trap of "photosystem I": P700. Upon excitation P700 produces a strong reductant ("H") and is left behind as a weak oxidant. Trap II, after its photoconversion, is capable of reducing PyOO"*" and producing O2. Except for the fact that it should be (associated with) a chlorophyll type pigment, the chemical nature of this trap and its mode of dark conversion are immaterial for the present discussion. The data indicate that in green plants C7OC and trap II occur in about equal concentration — but they seem to exclude the identity of the two compounds. [System II can operate in the absence of system I (cf. Section II)]. A peculiar feature of this scheme is that the "primary collector unit" amounts to only ^0 pigment molecules--a sit- uation also found in bacterial photosynthesis (I8). For op- timal operation, each unit should, on the average, transfer quanta to T II and C7OO in the proper ratio (e.g. 1:1), but since the mechanism is self-regulating it does not require a definite ratio between T II and C7OO or an association of these with a distinct group of collector pigment. This self- regulation is effected by the connecting electron transport system mediated by P700--fed at both ends by some 10 small units. The result of this "double focussing" is the classicsil unit of 400 chlorophylls. The scheme provides for a second "switch" in the transfer chain: quanta which fail to find P7OO can either be degraded in C7OO itself, [at 77'*K re-emitted as F7OO (cf. Fig. k)] or escape to C730 — v/hich also emits only at low temperature. (Long wave light does not yield fluorescence at 300°K.) The fact that at 77*K a considerable fraction of the absorbed light ( 50%) can be re-emitted at 730 mu (cf. Fig. 5) suggests that this is a funci^ionally significant process. One can as- cribe an important task to pigment C730: the harmless degrad- ation of excess quanta which might otherv;ise lead to photo- inhibition. In green plants (C700) amounts to only 3-10% of (Chi.) and causes a decline of the quantum yield of O2 evolution only beyond 69O mu : Due to the impossibility of reversed quantum flow traps II are not excited by light absorbed by G7OO 54 Bessel Kok itself. In blue-green algae, however, even 680 mu light is used ineffectively. Vi/e therefore, assumed that in these algae C700 is replaced (or amplified) by a significant fraction of the total chlorophyll [indicated (Chi. a 1)] . ACKNOWLEDGEMENTS The author gratefully acknov/ledges the collaboration of Miss Louisa Yang, Mrs. Pat VVoolf, Mr. Robert Trimble, and Mr. Hans Rurainski. The work reported v/as supported in part by the National Institutes of Health, the ^ir Force Office of Scientific Research, the Air Force School of Aerospace Medicine and the National Aeronautics and Space Administration. 55 Bessel Kok REFERENCES 1. Brody, S. S. and Brody, M. , Arch. Biochem. Biophys., 93 > 521-525 (1961). 2. Kautsky, H. ; Appel, W. , and Amann, H., Biochem, Zeitschrift, 352 , 277-292 (I96O). 3« Duysens, L.N.M, and Sweers, H. E., In: Microalgae and Photosynthetic Bacteria, pp. 553-572. Plant and Cell Phys. (1965). 4. Butler, W. L., Arch. Biochem. Biophys., 93, kl3- k22 (196I). 5. Kok, B. and Hoch, G., In: Light and Life, vV, D. McElroy and B. Glass eds., Johns Hopkins Press (I96I). 6. Kok, B, (this volume). 7. Kok, B., Biochim. Biophys. Acta, 48, 527-533 (I96I). 8. Weaver, E. C. and Bishop, N. I., Science, lAO, 1095-1097 (1963). 9. Beinert, H, and Kok, B. (this volume). 10. Brown, J. S. and French, C. S., Biophys. J. 1_, 539-550 (I90I). 11. Sauer, K. and Calvin, M. , J. Mol. Biol., _4, 451-466 (I962). 12. Lumry, R; Mayne, B. and Spikes, J. D., Faraday Soc. Disc. 27, 149-160 (1959). 13. Hoch, G. and Martin, I., Arch. Biochem. Biophys., 102 , 430-438 (1963). 14. Butler, v;. L., Biochim. Biophys. Acta, 64, 309-317 (1962). 15. 01son» R. A. (this volume). 16. Myers, J. and Graham, J. R., Plant Physiol., 3_8, IO5-II6 (1963). 17. Franck, J., Proc. Nat. Acad. Sci., 44, 941-948 (1958). 18. Clayton, R. K., Ann. Rev. Plant Physiol., l4, 159-l80 (1963). LIGHT-DRIVEN CYTOCHROME REACTIONS IN ANACYSTIS AND EUGLENA John M. Olson and Robert M. Smillie The basic similarity in cytochrome physiology between Anacys - tis nidulans and Eiiglena gracilis , strain Z, is impressive in view of their gross dissimilarities in size, structure, and pig- ment content. We have investigated the cytochrome reactions by sensitive spectroj^otometric methods in order to gain some in- sight into the patterns of energy transfer from the various light receptors to the reaction centers involved in the two photochemi- cal reactions of green plant photosynthesis and also to elucidate the pathways of photosynthetic electron transfer. The major thrust of this presentation will be the implications of experi- ments on whole cells in which both wavelength and intensity of monochrcmatic actinic light have been systematically varied. Some preliminary observations of the effect of carbonyl cyanide-m chlorophenylhydrazone (CCCP) are presented, and the light-driven reduction of cytochrome b,- in Euglena chloroplast fragments is described. INTACT ALGAE Two Light Reactions ; The evidence for two essential light re- actions which most clearly laid the precedents for the present work was that obtained by Kokd), Wittv^), and Duysens^^). The light-induced oxidation of c- or f-type cytochromes in green plants was clearly established, and the light-driven reduction by a second photoreaction was demonstrated. We have confirmed the observations by Amesz and Duysens(^) of light-driven cytochrome reactions in Anacystis and have identified the major ccmponent to be cytochrome f-555 on the basis of the alpha trough at 556 mii in light-minus -dark difference spectra. In Euglena the high poten- tial cytochrome -5 52 reacts to light. The effects of the two photochemical reactions on these f-type cytochrcanes are illustra- ted in Figure 1. In both algae far-red light causes a rapid oxi- dation to the steady-state level; but light below a certain crit- ical wavelength causes an initial rapid oxidation followed by a slower reduction to the final steady-state when the proper inten- sity is used. The diphasic kinetics disappear as the light in- tensity is lowered as shown in Figure 2. The curves ("initial peak" and "steady-state") for 0.62 ^ light in Figure 2 are 56 57 John M. Olson and Robert M. Smillie O 03 'H ■P (0 OJ -P 4) CO -P -d X o tiO ^ -d OH ^ a 0} o ^ H-H a ft OJ -P -H ^ f >-H <>H 1* -H 0) 0) tlOiH CVJ 3 (U O (U ^ >;CJ • ,a ^ +3iH o O -P nix(«z*av-°«*av) .002 1 , u. UL V o < ::» , s^ z ? ° ^ u. O - -I 3 II -< liJ °x r» ^ V. u> ~ or in V li. II ^ V «i °'\ '0-^ ^ . •a '§ s > g-*' O CO -H O-H 1 -PH O O (U (0 •H (U CO (U S -H ta fi U iO -P d +3 f-l CO OJ CD OJ V) €^ -p >H d 1 d >0 0) •H -P dP • -P O -H JE50N 4) :3 -p-d o 1 d -d o fd • -H >-p CJ o <1> -p ■p •H X CO 2 ■H lr^ OJ w pi 1 tn I !-> ( 0) OJ ■P cd -p -a c o •H -p CO •xi •H X o CM tJ ::« o ,0 CO CO •H a en ■P •H +3 • I •H ^ I u CD O -P CO fl CO o LTN J-P H 1 ^ ' ^ ^ E 0) 0) g X § ^ OJ- o U Xi f-i UJ -^^ .'^ ^tt° 1 p 9) o o o j^ d Pu O -H fi 61 John M. Olson and Robert M. Smlllie maximum net rate of oxidation during the "light -on" phase is much less than the initial rate of reduction observed in the "light- off" phase. A simple explanation of the "sluggish" cytochrome ox- idation is that the rate of cyclic electron flow is rapid enough at room temperature to compete vith the oxidation reaction at low light intensity. The maximimi net rate of oxidation upon illumin- ation is not proportional to intensity at these very low intensi- ties; the rate vs. intensity curve is sigmoid as shown in Figure 5, As the temperature is dropped, however, the rate cxurve more nearly approaches a straight line. Quantum Requirements ; Preliminary estimates of the quantum requirement for cytochrome -5 52 oxidation in Euglena range from 2 to 8 quanta per electron based on rates of absorbancy change upon illxomination. In Figure 5 the slope of the rate curve at 2* indi- cates a quantum requirement of 2 if ^6552 " ^®5U0 ^^ assumed to be 2 X 10^ M-1 cm"^. The lowest observed values are substantial- ly lower than the estimate of 7-IO for Anacystis (^) . Cytochrome b Reactions ; Light-induced reactions of cytochrome b in whole cells are ctoservable only under special circxjmstances. Under physiological conditions cytochrome f-555 is the main pig- ment to respond in Anacystis ; sometimes a slight response of the low potential cytochrome C-550 is also obseirved. In Euglena , only the high potential cytochrome -5 52 is observed. Cytochrome b oxidation in Anacystis caused by far-red light (O.7O n) can be observed in addition to cytochrome f oxidation when cells are cooled to 2" C or \^en cells are permitted to become anaerobic at room temperature. In both Anacystis and Euglena , the addition of 5 X 10-5 M CCCP permits the light-induced oxidation of cytochrome b with far-red light (R^ only) either with or without the oxida- tion of f-type cytochrome (Fig. 6). In Anacystis the cytochrome b oxidation is superimposed on the usual cytochrome f + c oxida- tion. In Euglena , the cytochrome b oxidation appears to replace cytochrOTie-552 oxidation initially, but the cytochrome -5 52 light reaction reappears almost completely after ^4-0 min without appre- ciable change in the cytochrome b reaction. The mechanism of CCCP action on photosynthetic electron transfer is not known. An attempt to demonstrate the light-induced reduction of cyto- chrome b by R2 in whole cells ( Euglena ) indicated a possible transient small increase in reduced cytochrome upon illimiination with high intensity red light (O.65 u) . The steady-state change was, however, either zero or a slight net oxidation. 62 John M. Olson and Robert M. Smillie EFFECT OF CI-CCP ON CYTOCHROME OXIDATION IN EUGLENA [X = 0.70/1, 1 = 11x10-* EINSTEIN cm-2 sec"'] X .01 ^ ^ .02 -.03 DIFFERENCE SPECTRUM LIGHT-DARK 5x10"' M CI-CCP _L J_ _L 520 540 560 WAVELENGTH IN m/i 580 Fig. 6 EUGLENA CHLOROPIAST FRAGMENTS Cytochrome b^ Reduction ; When washed chloroplast fragments are prepared from Euglena cells in late log phase, most of the cytochrome -5 52 is lost. When such fragments are suspended in .025 M Tris, pH 7.8 with 10-3 M MgCl2, strong light causes a gradual reduction of cytochrcsne bg which remains bound to the fragments (Fig. 7). If the light is turned off after the reduc- tion is complete, the reoxidation in the dark is extremely slow. If, however, the light is kept on, and the intensity dropped a factor of 10, a light-induced oxidation of the cytochrome occurs after an initial lag of about 20 sec (see upper left insert in Fig. 7). The observation of a light-induced reduction or a light-induced oxidation of cytochrome bg depending upon light in- tensity and the redox state of the cytochrome suggested that both R2 and R, were still functioning. The absence of significant ox- idation tor reduction) of cytochrome bg in the absence of light indicated that this cytochrome cannot be merely a component of an electron transfer system which "short-circuits" a single photo- chemical reaction. 63 John M. Olson and Robert M, Smillie LU CO O h- <: <: LU -J e) ID LU 1 ■ I 1 1 1 1 O - o- -V ^ ^or •< ^<; o ^- ^•L o LU to in ' 'O tiO a (u o ts C ij^ O XJ fH > fl CO rv Jh (U 4J CO •H ^ G ^ u O O ^ :! -P o w XI B v_' +> -H 0) Eh >, O S UA -O O P< o f- q; o >H '^ -p D • • to ^ w -p r^ CO H CJ -P 4-> O (h ,H O C CO iH 4-> -PS 0) • •H p (0 K ft *•— ^ -P c S to O CO S* O -H O rH C > < Q P^ (U f-. C O C o ;3 O SI Jh 0) ^.0 • ^d O O dJ «M +3 bO-rt H > Cm -P ^ X C CO •H O (i< o X -H x; t:) t3 £Oi''(°^^av-^av) i t- o e in - M I •H (- (^ C3 2 lU _l LU > o 1 <\J > If) 64 John M. Olson and Rotert M, Smillie The action spectrum of the light-induced reduction of cyto- chrone b^ in chloroplast fragments is compared to the action spectrtmi of cytochrome -5 52 oxidation in whole cells poisoned with DCMU in Figure 8. The former spectrum corresponds to the action spectrum of R^* and is further evidence that chlorophyll a is effective in Rp. The difference between the normalized spectra for Rt and Rg indicates again that a far-red pigment analogous to P-700 is active in R^, but not in R2. Effect of 1,10 Phenanthroline ; This inhibitor (10"3 m) de- creased the steady-state level of cytochrome b^ reduction during illumination to about 10-20 per cent of the total change observed in the absence of inhibitor. In addition, a rapid reoxidation of the cytochrome b^ took place immediately when the light was turn- ed off. Therefore, no light-induced oxidation of cytochrome bg at low intensity could be observed in the presence of 1,10 phen- anthroline. It appeared that the cytochrome bg was now part of a "short circuit" of R2. Interaction of Cytochrcme-^^^ : When purified cytochrCTie-552 (ca. 10"^ M) frcm Euglena is added to washed chloroplast frag- ments (ca. 10"^ M chlorophyll), illumination causes the oxidation (or reduction) of the added cytochrome -5 52. High intensity illum- ination causes the oxidation of cytochrome -5 52. VThen the light is turned off, a rapid reduction occurs in the dark with a half time of ca. 6 sec. (The light-induced reduction of cytochrome bg which is concurrent with the oxidation of cytochrome-552 is about 1/3 the change observed in the absence of added cytochrome-552.) Very low intensity illumination causes a relatively slow oxida- tion of cytochrome-552, but the final extent of oxidation is con- siderably greater than that observed with high intensity. Fur- thermore, no reduction is observed upon cessation of the light. A light-induced reduction of cytochrome-552 can be demonstrated with high intensity illumination after partial oxidation with low intensity light. -k Addition of 5 x 10 M 1,10 phenanthroline to the reaction mix- ture slows down the reduction of cytochrome-552. The light-in- duced oxidation at high intensity is at least doubled and the sub- sequent reduction in the dark is slowed to about l/lOth the rate in the absence of inhibitor. SUMMARY AND CONCLUSIONS 1) There is not necessarily a one-to-one correspondence be- tween a given light-absorbing pigment (e.g. chlorophyll a) and a 65 John M. Olson and Robert M. Smillie given light reaction (e.g. Ri ) . In either Anacystis or Euglena, a photon absorbed by chlorophyll a has a significant probability of driving R^ instead of R, . Analysis of energy transfer path- ways should Be made in terms of relative probabilities of energy transfer to R^^ and R2 from a given light absorber. 2) The cytochrome oxidation vs. intensity curve (Fig. 2) for actinic wavelengths -vdiich activate both R, and R^ suggest that the transfer ratio R-i/R2 ^^Y increase as the light inten- sity is raised to very high levels. This implies that R2 ap- proaches light saturation at lower intensities than R, . 3) Cytochrome f (or -552) is oxidized by R-j^ via a mechanism relatively insensitive to temperature, and is reduced by R2 and/ or a short-circuit pathway around R, . Cytochrcane-552 oxidation apparently requires 2 quanta per electron. Cytochrome reduction has a temperature coefficient of about 2 per 10 degrees. k) Light-driven reduction of cytochrome b^ in Euglena chloro- plast fragments and the interaction of added cytochrome -5 52 sup- ports the proposal of Hill and B€ndall(9) that electron flow frcxn R2 to R^ is mediated by cytochrome br and cytochrome f (or -552). 5) The mechanism of 1,10 phenanthroline inhibition of the Hill reaction and photosynthesis may be linked to the apparent short-circuiting of R2 via cytochrome b^ in chloroplast frag- ments. ^^*^^ REFERENCES AND NOTES B. Kok & G. Hoch, in Light and Life (W.D. McElroy & B. Glass, eds.), Johns Hopkins Press, Baltimore, I96I, p. 39T« 2 H.T. Witt, A. Muller, & B. Rumberg, Nature 192, 967, I96I. 3 L.N.M. Duysens & J. Amesz, Biochim. Bio^Aiys. Acta 64, 2U3, I962. ^ J. Amesz & L.N.M, Duysens, Biochim. BiojAiys. Acta ^, 26l, I962. 5 B. Kok & H. Beinert, Biochem. Bioi^ys. Res. Ccmmun . 9, 3^9,1962. ° W.L. Butler, Arch. Biochem. Biophys . 93, 4l3, I96I. T J. Myers & C.S. French, J. Gen. Physiol . k3, 723, I96O. ° M.B. Allen, L.R, Piette, & J.C. Murchio, Biochim. Biophys. Acta Q 60, 539, 1962. f R. Hill & F. Bendall, Nature I86, I36, I96O. Research carried out at Brookhaven National Laboratory under the auspices of the U. S. Atonic Energy Commission. The coopera- tion of Dr. John A. Bergeron in culturing the algae under con- ditions of rapid autotrophic growth is appreciated. ID THE TEMPERATUEE INSENSITIVE OXIDATION OF CYTOCHROME F IN GREEN LEAVES - A PRIMARY BIOCHMICAL EVENT OF PHOTOSYNTHESIS Britton Chance and Walter D. Bonner, Jr. The temperature insensitive oxidation of a cytochrome compon- ent adjacent to chlorophyll has proved to be an incisive tool in the study of the basic mechanisms of primary photo-reactions at the biochemical level. In the context of this paper the "primary" light-induced reaction is that which involves a chemical step, and therefore we exclude the physical processes by which chloro- phyll may become activated. A number of aspects of the temperatiire insensitive oxidation of cytochrome c in the purple sulfur bacteria Chromatium have been siommarized elsewhere (l), and special attention has recent- ly been paid to the fact that the reaction is not only temp- erature insensitive but is also viscosity insensitive (2). The application of this method to green plants has not previously been very fruitful. While H. T. Witt has recorded the low temp- erature oxidation of cytochrome f, the observation has been re- stricted to chloroplast suspensions and, in fact, to chloroplast suspensions which are specifically treated with sucrose in order that the effects be observed (3). As yet no observations have been made of the low temperature response of cytochrome f in the intact leaves of green plants, although numerous observations have been made at room temperature in leaves and in suspensions of algae; for example, Chlorella {k) , Anacystis (5) and Porphy- ridiiom (6)and in chloroplast s (T). This paper describes the application of the double-beam spec- trophotometer to the detailed quantitative study of cytochrome f oxidation at 77 K in a variety of leaves. The experimental res\alt allows an evaluation of the role of cytochrome f oxidation in relation to the"PYoo" absorption band. In addition, we pro- vide an accvirate representation of the relationship between light absorption in chlorophyll a and the rate of cytochrome f oxida- tion in leaves. Materials . The materials used were market spinach leaves, leaves plucked 66 67 Spinach Leaf (59-1) Spinach Chloroplasts Off (57-"57") Off Seal* On(2X) Swiss Chard -Leaf (59-6) Off Scale Mung Bean Leof Off I '^^"^ O" Oft Scale Figure 1. Light- induced oxidation of cytochrome f in three types of leaves and one preparation of chloroplasts. 680 Ta\i actinic light; a downward deflection corresponds to a decrease at 555 or 55^ niM- measured with respect to the reference wavelength (5^0 or ^'•i-^ ni^). Light intensities employed relative to each other are indicated by the symbols "IX" or "2X". Experiment 59, 1 - ^. Material ConparlBOn of Cytochrome t kinetics In different leaves and in chloroplasts at 77^ K ^D* 10 ^D%ec Intensity* lo"* ii>D/sec/ Btpt. unit Intensity Spinach leaf (market) 555-5^ 0.0069 0.18 1.5 0.12 57a-l Spinach leaf (fresh) 555-:5ltO 0.00l»5 0.55 3.3 0.17 59 -1 Spinach Chloroplasts 555-5UO 0.0050 0.90 0.25 3.3 1.0 0.27 0.25 57b 58 57a 59 Swiss Chard leaf fresh 555-5'>0 COOS'* 0.59 1.5 0.39 59 6 Mung bean leaves ^^.^ O.OOW (etiolated) 0.35 3.3 0.11 57a-55 per leaf thickness except for chloroplasts (Im path, 0.279 mg/nl) Arbitrary uslts TABLE I 68 Britton Chance and Walter D. Bonner, Jr. directly from yoxing spinach and Swiss chard plants, and chloro- plast preparations (kindness of Mr. Stephen Sikes). Etiolated Mung bean leaves were obtained directly from the plant and were greened for several hours with the red light prior to use. The properties of these leaves have been reported on in more detail by Dr. Bonner (8). The leaves are held lightly between two lucite plates. These preparations were rapidly frozen by plung- ing them into liquid nitrogen in an aluminxom holder of small specific heat. Occasionally, the leaves crack at low temperat\ire^ especially if they are held too tightly by the lucite plates. The leaf absorbancies are very high, not only due to their chlorophyll content (0.3-0.^4- mmoles/kg wet weight) but also due to the non-specific light absorption of the leaf. The effective transmission in this apparatus is 1. 6 per cent at TOO mii and even less at shorter wavelengths. Thus, nearly complete absorption of the incident light occurs. An example of the responses of these fovir types of biological materials is given in Fig. 1 where we display on a compressed time scale the optical density decreases at 555 ^ measured with respect to a nearby reference wavelength (5^0 or 5^^ ^V-) > ^^ each one of the four records, it is seen that the optical density change of the order of approximately 0.005 is obtained. It is seen that a rapid downward deflection of the traces (corresponding to a decreased absorbancy at 555 niii) is obtained when the actinic light (680 rD\i) is tTxrned on. In each case, the leaf thicknesses were between ,k and .6 mm. Two levels of actinic intensity are employed (IX and 2X). In the case of the chloroplast suspension and the Mung bean leaves the actinic light is interrupted period- ically and the oxidation is seen to come to a halt. Thus, the effects of the measviring light are not objectionable under the experimental conditions. However, small effects can be observed in the chloroplast suspension. Table I summarizes a number of properties of the cytochrcme f responses in the four types of leaves and the one type of chloro- plast preparation studied. As indicated in the Table, the opti- cal density chants for the leaf thickness are in the range of 0,Oi4-i4- (Mung bean leaves) to O.OO69 for market spinach leaf. The rates of cytochrome f oxidation, which will be discussed in more detail below, when put on a basis of an arbitrary unit of light intensity, are of a similar order of magnitude although the fresh Swiss chard leaf seems to excel the other biological ma- terials, even the chloroplast preparation. 69 Britton Chance and Walter D, Bonner, Jr. EXPEEUMENTAL METHODS A Dewar flask is attached onto the emergent light fitting of the doulJle-heara spectrophotoaeter (9, 10). Tiie one inch photo -rnul- ti-olier is attached directlir to the housing of the Dewar flask. Since this Dewar has right-angle windows, side illumination is readily obtained from either a tvingsten lemp (680 va\i illumination) with appropriate interference filter or a mercury arc (^3^ "Ua illxomination) with a multielement filter (Eppendorf). The cuvette is held in the pictxire and contains an aluminiom block with a pair of lucite plates between which the leaf is gently pressed. A smsill mirror serves to reflect the actinic light upon the leaf, out of the way of the measuring light. All other aspects of the double-beam spectrophotometer are as in previous communications (11). The levels of actinic illumination are low; since cyto- chrome f responds as a "quantum counter", high levels are not required to cause maximal oxidation as they are at room tempera- ture. We are not reporting quantum efficiencies for cytochrome f oxidation in this paper and therefore light intensities are reported in arbitrary \inits. The values are about 10"^^ to 10"^/ Einsteins/cm^/sec . \^ea k36 m\x excitation is employed, the rate of cytochrome f oxidation is slower than with 680 m^i excitation, as indicated in the Figures below. However, the extent of the reaction is the same, since the reaction is essentially irreversible. Absorbancy enhancement . With cytochrome c the absorbancy increment in the oxidized minus reduced spectra at '^9 rap is if. 5 times greater. Presumably, such values apply to leaves, but detailed controls are necessary to ensure this. This factor is not known for P700. For the purposes of this paper, we are emphasizing the ratio of the rates at 555 and 705 m\x. This comparison may be made by assuming the same enhancement for the two wavelengths without the need for its absolute value. Effect of the measuring light . At low temperatures, cytochrome f becomes effectively a "quan- tum counter" and therefore, prior illTomination with undue inten- sities of excitation light will vitiate the desired response. We have, however, illustrated the effect of a high level of measuring light intensity in Figure 2 . In this illustration, the measuring light is that obtainable from a 3 mn slit of the Bausch 70 Actinic Light Off Spinach L«of j»- 0.7x10' XMc"'x0.5mm"' 0.013 xtec xcm t t Meos. Light On , „. 555-540m/i '-'9''' (3m,. Interval) (gaomM) 50 tac Figu re 2. Illustrations of the kinetics of the light-induced cytochrome f oxidation vmder conditions where the measuring light is of sufficiently high intensity to cause rapid oxidation by itself. Time proceeds from left to right. Tlie deflection of both traces with tirae indicates a decrease of ab- sorbaiicy at 555 rjn with respect to 5^0 mu . The sensitivities employed in the tv70 traces are different and are indicated on the diagram. Response tine of the tv/o traces is also different; that of the lower sensitivity has a lower response time corresponding to less than l/2 second, while the trace for the higher sensitivity has a response tirae of approximately 2 seconds. The moment of illtuniiiation with actinic light is indicated, and its wave- lengtli is 680ajJ, A spinach leaf employed in this experiment \7as a mature plant (market spinach) . The rate of absorbancy change on illumination with actinic light is indicated per leaf thickness (0.5 mm) or per centimeter. Experiment 53 A 1. Switt Chord Ltof 5S5-540m^ log I./I'OOOI log I./I-0004 1 77"> K Swiss Chord Leaves Spinach Leaves X(m/i) X(m/i) 680m/i Illumination 540 550 560 570 0-1— • 540 550 560 ogI./I'0004 -0.005- -SOttc- 0.005 FIGURE 3 FIGURE k Figure 3. A ccirpririson of the light response of Swiss chard leaf at 300 77^'kV ^Tlie convention in recording is similar ix> that of Figure 2; i.e., t\ro traces of different sensitivities and response speeds. The actinic il- lumination is 680 ran. Time proceeds from left to right; initial rate of ab- sorbancy change is calculated per leaf thickness (O.k mm). Experiment 59-6. Figure h. Lot; temperature difference spectra for Sv/iss chard and spinach leaves obtained with the double -beam spectrophotometer, according to the method illustrated by the preceding Figure. A number of similar leaves are selected and illuminated consecutively at Im temperatures. The absorb- ancy changes are indicated per leaf thickness, 680 mm actinic illumination. Experiment 62, 58* and 71 Britton Chance and Walter D, Bonner, Jr. and Lonib monochromators. Upon the moment of illumination of the sample, a steady state deflection of the two traces is observed which increases when the actinic light (680 mu) is turned on. It is seen that the measviring light intensity is adequate to cause an appreciable rate of oxidation of cytochrome f and roughly half of the cytochrome f has been oxidized by the time the actinic light is t\irned on. A criterion of satisfactory operation may be taken as the ra- tio of the response time of the spectrophotometer (0.3 sec for 10 to 90 per cent in this case) to the time for the measuring light to cause an arbitrary oxidation of cytochrome f (lO per cent is used here). This ratio, M, is over kO here and may be useful in comparing different spectrophotometers. With the actinic illumination employed here, a response time of 1 sec (10 to 90 per cent) is adequate in order to diminish the oxidation of cytochrome f prior to illumination. VJe have reduced the spectral interval from 3 to about 1 mji. In most of the traces which we report here, the effect of the meas\iring light prior to illumination with actinic light is negligible; the ratio M is over 25, even when using 7OO mij. as the measuring light. The recording is usually made with double traces at different gains and at different response speeds. Thus, the rapidly re- sponding trace ( showing an upvrard deflection) is at a lower gain and at a higher response speed (l sec), while the downward de- flecting trace is at a higher gain an a lower response speed (2- 3 sec). However, the respective upward and downward deflections of both traces correspond to a decrease of absorbancy at 555 with reference to 5^0 m^. In these studies in which the absorbancy changes due to P70O are measured, certain controls and precautions are observed to insure that fluorescence changes are not causing artifac- tual responses. First, it should be pointed out that the observations of Butler (12) on the fluorescence of chloro- phyll show that fluorescence enhancement of approximately 20 per cent caused by light in the region of 620 myi is decreased considerably by light of wavelengths of TOO mn. Therefore, a fluorescence artifact would be in the opposite direction from the absorbancy decrease observed; i.e., 705 ^P- illumination decreases the long wave fluorescence of chlorophyll relative to that of 635 niP-. Thus, with the double-beam spectrophoto- metric technique, the absorbancy change woi^ld not be confused with the fluorescence change, as they are in opposite direc- tions. Secondly, the abil5.ty to record the kinetics on a fast 72 Britton chance and Walter D. Bonner, Jr. as well as a slow time scale permits the observation of the rela- tively instantaneous fluorescence change as separate from the kinetics of the P70O oxidation. Steady state illumination allows the possibility of the time discrimination between the fluorescence and absorbancy changes as the fluorescence change woiold occur in a short time, while absorbancy changes, as clearly indicated in the charts, requires about half a minute to reach the steady state level. We have, therefore, reasonable assurance that the double-beam method guards against the very annoying fluorescence due to chlorophyll in emission, in the region of TOO miji. When the double-beam spectrophotometer is used with intense ac- tinic illumination of fluorescent materials, where the average photocurrent may increase greatly during actinic illumination, it is necessary to connect the photomultiplier output directly to the ac amplifier, bypassing the chopper contacts used for calibraUon or for single- ended operation. However, a doubling of the aver- age photocurrent causes no difficulties. EXPERD^[SM!AL RESULTS A comparison of the light induced kinetics at room and low temp - eratures. Figure 3 illustrates room and low temperature kinetics of a Swiss chard leaf upon illumination with 68O m\i actinic light. This comparison is facilitated by the possibility of measuring the room temperatvire kinetics before filling the De>ra,r flask with liquid nitrogen. On the left hand portion are the room tempera- ture kinetics which are seen first to be rather small in amplitu* in comparison with the low temperature kinetics. (The rate of reaction is apparently more rapid at the low temperature in spite of the decrease of sensitivity to absorbancy changes at the low temperatures.) Two effects are undoubtedly involved. First, an enhancement of the absorbancy change due to increased light scat- tering and a sharpening of the cytochrome band occvirs. Second, the dark reduction of cytochrome f which, at room temperature is almost identical in rate to the oxidation, does not occur, and hence, the correct velocity constant for cytochrome f oxidation can be obtained at the lovr temperatui'e. This diagram suggests that measurements of quantum requirements for cyi:ochrome reac- tions may be more acctirately measured at the low temperatures (ll) Difference spectrum for illumination at low temperatvires . By choosing a n-umber of leaves of similar size and hence thickness, it is possible to repeat the experiment of the previous Figure 3 at various wavelengths of measuring light and to obtain S«M Chord LMf 77*K 555-540m^ 680 m^ T °l690m,. log I./I. 0.001 On / ^ 73 ^'\ On ,/ \' / 670m^ 4.IO"*Mc' V' 0"660m^ t -^^ ,°" MaoMring Light On t ^ 1 555— 540 m^ ^bm^ ^. o T w U «> «;-. « <« (— y * •v u O \ c V o K ^ d iJ.b- \ O (A e \ .O o V < 1 If) 0- 6' ^ O *0 660 680 700 a> If) o X. u> Act inic L ght X (m^) Measuring Light FIGURS 5. FIGUEE 6. F igure ^ . Illustrating the effect of light of various vravelen^ths in the red region upon the rate of cytoclirome f oxidation in the S\7iss chard leaf at 77° K. Tine proceeds from left to right. Measuring light slits set at 1 mu cause little tneasurahle oxidation, Hov/ever, illumination with 680 n+i givt-o rise to a rate of ij- x lO"'^ per second per leaf thickness {O.h mm). The tines at which the illumination is changed to various other wavelengths is indicated on the diagr-aa. Eiwerinent 63-I5. Figur e 6. A plot of the effect of red illumination upon the rate of cyto- chroriT oxidation for a a^iss chard leaf at 77° K. The points are taken from the initial velocity" of a nuraber of similar Swiss chard leaves. The rates of absorhancy change are per leaf thickness (0.4 mm). The diagram also in- dicates the rate obtained lith measuring light only. The data are corrected for the small change of energy distriT^ution of the monochromatic light over the spectral interval. Ecperiment 63. Swiss Chard Leaf TT'K "'^ 555-540m/i 1.0x10 -sec' V'^ ! t' Actinic Light On t 436m/i ;s: Measuring Light On t Off log I./I= 0.001 T ).00 1 -50sec- 7O -,r Figure 7. Activation of cj'tochxome f oxidation in &7iss chard leaf at 77 K with J4-36 mti actinic light obtained from a medium pressure mercury arc (see Fig. 6) . Tlie convention used in recording is similar to that in previous ex- periments. Experiment 6I-IO. 74 Britton Chance and Walter D. Bonner, Jr. thereby, a "difference spectrum" for the actinic effect. This is illustrated in Fig. 1|. for Swiss chard leaves and for spinach leaves . In both cases there is a large dimin[ution of absorption which has a maxinrum very near 555 m^^- In spinach leaves, there is a possibility that a satellite band characteristic of cyto- chrome f is observed, although further experimentation is desir- able to ensure this. This satellite band is not observed in Swiss chard leaves, although insufficient data are available at present to substantiate this difference. It is significant, however, that the band at low temperature Isat 555 m\i, whereas the peak in acetone extracted spinach quantasomes is clearly at 552 mn (13). It is probable that this difference is a real one, that cj^ochrome exists in a different state in the leaf than in the acetone treated chloroplasts or in the extracted pigment. Relative quantxmi efficiency for cytochrome f oxidation . As the preceding Figures clearly indicated, the velocity of cytochrome f is under control of the intensity of the sctinic light. It has occurred to us that it would be of considerable interest to determine the effect of the wavelength of light upon cytochrome f oxidation since we have here for the first time, the isolated prim.ary chemical event of the leaf. Instead of the fixed wavelength actinic beam a Bausch and Lomb 200 mm focus grating monochromator (1200 line grating) is employed, and is ilUuninated with a tungsten lamp. The energy distribution of this canbination is found to be prac- tically flat in the region of interest (620-710 m|i). ?fenual ro- tation of the wavelength knob controls the rate of this chloro- phyll at low temperatures. The monochromator was set with a 2 imn slit width (6 m[i spectral interval) which proved to be adequate to give rates of oxidation of c\rtochrome f (see Fig. 3) f lar::e compared with the rate caused by the measuring light. As illustrated by Fig. 5., a Swiss chard leaf cooled at 77 K illuminated first with a measiiring light, it is seen that the rate caused by the measuring light is insignificant. Vfhen the 680 mil. light is turned on an abrupt deflection of the traces is observed which proceeds considerably more rapidly for 680 m|jL than it does for 69O mia. In fact, when 700 mil light is employed, the trace is nearly horizontal. The deceleration is, however, reversible, and when 67O and then 68O rau radiation is employed there is rin abrupt acceleration. Since the course of cytochrome f oxidation follows an approximately exponential function, measurements of the rate were restricted approximate- ly to the first third of the course of the reaction. 75 Britton Chance and Walter D. Bonner, Jr. Furthermore, relative rates on ad.^acent segments of the cvirve were employed. A summary of the resxolts of the nujtiber of experiments is plot- ted in Fig. 7 . On the ordinates are represented the rates of ahsorbancy decrease at 55O-5UO mn (as measiired fiom the slope of the curve similar to that of the preceding Figure), the abcissa are values of vavelenijth of the actinic light. The dashed line at the bottom of the Figure is the rate obtained by the measure- ment light only, and this is seen to be negligible for all wave- lengths employed. The graph shows a plateau in the region of 66O-68O mn with an abrupt decrease at 69O and TOO mu. The curve may be interpreted as a decrease in quantiom efficiency of electron transfer between chlorophyll and cytochrome f; the efficiency falls to half its value at approximately 695 m^. A similar fall-off is fotuid with frozen chloroplast suspensions. One measiirement of a decrease of efficiency at 6k0 m\x report- ed in the oral presentation was foiind to be in error. Absorbancy changes at 700-70^ m^i In order to observe the response of the 700 m\i pigment under conditions where the response of cytochrome f can also be mea- sured, we have employed actinic excitation at k'iS m\i from a medium pressure mercury arc . With this ill\xmination, a rate of cytochrome f oxidation corresponding to 1.0 X 10"^ OD units/sec was obtained as is indicated in Fig. 7^ where illumination occurs for a period of 20 seconds . A second period of 20 second illumination carries the reaction nearly to completion. If the measuring wavelengths are now changed to those appro- priate to the 700 mp. pigment (705-635 mp) Fig. 8, ve find a more rapid rate of change of absorbancy when the meastiring light is on, and note a considerable increase of rate during the 20 sec- ond period of actinic illumination at 14-36 m|i. Since the sensi- tivity is lower than in the recordiiig of cytochrome f, the rate corresponds to 2.7 x 10"^ OD units/sec. A second interval of illumination causes completion of the reaction. A number of features of the reaction are of importeince. First, both the reactions proceed when the measuring light is turned on, and are accelerated simultaneously when illuminated with actinic light. This point will be taken up in the Discussion. By repeating the experiment of Fig. ,3 with different leaves 76 Swiss Chord Leaf yyK 705-635nn/i 27xl0'*sec' T t Measuring I Off On Light On , '^'^^'"'^ Light On 436 m/i h 50sec Off Figure 8. Measurements of thekinetics of absorbancy change in the region of "P '""in a S\7iss chard leaf frozen at the temperature of 77°IC. In this case, the measuring light (slits 1 mh,) causes appreciable activation of the absorb - ancy change. However, the increase on illumination at k36 xrii is marked. The rate of absorbancy changes is calcxaated per leaf thickness {.k ram). Experi- ment 61-10. Splnoch Chloroplotti Off 555-940m/t 77*K T loo I./I* 0.001 ! Actinic Light On \ Y eSOmii log I./I*0.004 (I) J. Mtaa. Light On I" SOiec A Figure 9 . A comparison of room and low temperature response of cytochrome f in a suspension of spinach chloroplasts (279 MG chlorophyll a/ml). The ac- tinic light employed in both cases is of the same intensity and a wavelength of 680 np. Tlie appropriate absorbancy scales for the two traces are indi- cated, tlpon illumination, the absorbancy at 555 mp. decreases with respect to that at ^kO riyi. Experiment 57B 56, 57. pt Ho. X ^°^mx ODTl X 10 Assumed d Ac a: AT Chlorophyll (cm^^xM"^) (ran) MM pM/aec unoles/kg leaf wet weight 1 SJS-Sto 0.0060 1.0 20 0.1. .76 .12 IKX) 8 705-635 0.0125 2.7 80 0.4 .•38 .085 WlO TABLE II 77 Britton Chance and Walter D. Bonner, Jr. and with a variety of measuring wavelengths, the difference spec- trum for the TOO mp. pigment can be determined, and it is fo\md to lie between TOO and T05 mu in accordance with the results of Witt (T) and Kok (ik). At low temperatures, the half width of the band is roughly 10 mjj.. Data calculated from Fig. S are summarized in Table II. The maximum absorbancy change in the long wave region is about twice that at the shorter i/ave region and the rate in the longer wave region three times that in the shorter wave region. We maJce the usual assumption that the extinction coefficient of the TOO mu compound is the same as that of chlorophyll a at 680 m|i, and further assume that the extinction coefficients of cytochroane f and Pyoo change in proportion at low temperatures. On this basis, the ratio of the concentrations of tte two substances is about 2 to 1. The ratio of the rates is of more interest; the molar rates of absorbancy change are about equal (act-ually, the cytochrome f rate is 1.^4- times the Ptoo rate). It should be emphasized, how- ever, that these comparisons are only approximate and may be re- vised when more acciorate data are obtained. DISCUSSION The observation of cytochrome f oxidation isolated from other reactions of the complex matrix of photochemistry and biochem- istry allows a detailed study of the mechanism of the electron transfer reaction between cytochrome f and chlorophyll. While it is not the purpose of this preliminary note to discuss this in detail, it is apparent that the accurate recordings of the kinetics in the illuminated frozen leaf will be of great advan- tage in further experimentation. The present discussion will be limited to a consideration of the low temperature oxidation of cytochrome f in relation to the oxidation of cytochrome cg in photo synthetic bacteria and to the absorbancy change at TOO m^i. Properties of the lovr tonperature oxidations of cytochrome f and cytochrome c^' Our previous observation of the kinetics of oxidation of cyto- chrome C2 in Chromatium at liquid nitrogen temperatures (U) is supported by the observation of the cytochrome f kinetics in leaves. These results greatly extend those obtained on aged chlo- roplast suspensions by Witt (T). It is clear from these results 78 Britton Chance axid Walter D. Bonner, Jr. that the electron transfer reaction "betveen cytochrome and chlor- ophyll is very unusual; its temperature insensltivity underlines the jxrxtaposition of these tvo metalloporphyrins in a vay so that collision reactions are not required for electron transfer. In this paper, ve have not yet determined the quantum requirement for cytochrome f oxidation at the lov temperature, but it can be inferred from the kinetic data of Fig. k vhere the low tempera- ture rate is considerably faster than the room temperature rate, that the quantum efficiency is probably high. It is of interest that the oxidation of cytochrome f at low temperatures is rapid in aged spinach chloroplasts as foiind by Witt and extended by us to fresh spinach chloroplasts (Fig. 9). This is not true of Chromatiijm chromatophores which have been found by us (l5), to be more temperatxire sensitive than the reac- tions in the intact cells; the reaction comes to a halt at 77 K. It has further been noted by Duysens (l6) that some algae show a rate limitation in cytochrome oxidation at low temperatures. Whether or not this is due to a basic difference in the mechanism or whether the essential orientation of cytochrome and chlorophyll is deranged during the freezing of some materials and not others cannot be stated at the present time. In Dreliminary experiments we have examined the effectiveness of red-light in cytochrome f oxidation and find thot the rate of oxidation'falls to half the maximum value at spproximaLoly 695mn. In other words, the quantum efficiency falls to half maximal val- at 695 mn. This result gives evidence for the intimate identifi- cation of cytochrome f with system I in the leaf. Kinetics of cytochrome f and Pyoo The possibility of measuring absorbancy changes corresponding to these two components in the frozen leaf would appear to pre- sent optimal conditions for a critical evaluation of the possi- bilities of their interaction (l^). Under these conditions, thermal reactions in which the two might be involved would be negligible and any possibilities for their direct interaction might be observable. In the intact leaves, we routinely observe absorbancy changes at 555 and 705 m|i typified by the data of Table II, approximately double the absorbancy change at 705 as compared with 555 m^. In fresh chloroplasts the absorbancy change at 555 m|j. is observed in roughly the same relationship to the chlorophyll content as in the intact leaves (see Fig. 9). At 700 m^, less than one fifth the absorbancy change is observed and this is partly reversible on cessation of actinic illumination. This result at 700 m^ may 79 Britton Chance and Walter D. Bonner, Jr. be compared vlth that of Witt (3) who states that he was unable to observe any changes at 700 m[i in fresh chloroplasts . It is possible that the reversible light response at TOO m\x is associ- ated vith some damage to the leaves. One possible explanation is based upon the idea that tvo light-induced oxidations occur at low temperatures . The disap- pearance of absorption at TOO m\x is concluded to represent an oxidation state of chlorophyll but its chemical configuration is quite unknown (l^). The evidence for the oxidation of cytochrome f is fiiroly based upon the disappearance of the characteristic absorption band of ferrocytochrome f . It becomes, therefore, of considerable interest to determine vhich is oxidized more rapidly. As Table II indicates, the absorbancy decrease is relatively more rapid at T05 mn than at 555 m|i. Hovever, a comparison of Figs. 8 to 9 indicates that the reactions come to completion at about the same time. In Table II, ve have attempted to make the com- parison more meaningful by converting the rates of absorbancy change to molar rates, making assumptions vhich need, hovever, a detailed study and critical evaluation. Hovever, the simple assumption that the extinction coefficient of cytochrome and chlorophyll bear the same relation to each other at lov tempera- t\ires as they do at room temperatures brings the rates of the light-induced reactions of cytochrome f and '^^qq closely to the same range, cytochrome f being slightly faster than PyoO* "^^ ^^® basis of any of a number of simple mechanisms, it appears that neither cytochrome f nor Pyoo ^^ ^ ^^'^® limiting intermediate in the oxidation of the other. Even qualitative aspects are useful in this respect. First, there is no induction period in the light -induced oxidation of either of these pigments vhich vould suggest a sequential reac- tion, i.e., a delay in the oxidation of cytochrome f prior to the oxidation of 'P'jqq and vice versa. This lack of induction period is also observed at the measuring light and actinic light intensities. One mechanism, vhich appears to meet the needs of the kinetic data is that cytochrome f and P^^qq are interacting vith different chlorophyll molecules, vith cytochrome f being at the active center of the photosynthetic unit and Pyoo ^^i^S at a chlorophyll molecule vhich is on the energy transfer pathvay from the initial receptor to the active center. It is apparent that a very detai led examination of the quantim requirements for these tvo light -induced oxidations vould be of great importance in this respect. Since both oxidation reactions occur simultaneously, their quantum requirements should be additive . 80 Britton Chance and Walter D. Bonner, Jr. /Actually there -tvcwld appear to he no need for a spectroscopi- cally distinct forra of chlorophyll to be formed simultaneously \i±th the oxidation of cytochrome f , presumably the metastahle state of chlorophyll -which acts as the energy trap for the ini- tial light reaction could accept an electron from cytochrome and transfer it to the electron donor in a temperature insensitive re- action, \7ithout the need for the accumulation of a measurable amount of a chlorophyll intermediate. A second mechanism that fits the needs of the experimental data and which tak;es into account the certainty vith vhlch the disappearance of the 555 niia band of cytochrome f indicates an oxidation of ferro-cytochrome f is that the Pyoo absorption is actually that of a reduced chlorophyll intermediate . Under these conditions, an exact correspondence of the molar rates of change •would be expected. Before this hypothesis can be considered seriously, the apparently sound basis upon which it has been con- cluded that P700 i^ ^^ oxidized forra of chlorophyll must be crit- ically reexamined (l^). SUMMARY 1. The demonstration of the temperature insensitive oxidation of cytochrome f in leaves of three types of plants is reported; difference spectra are provided and the kinetics of the change are measured. 2. The g^u8Ji"tum efficiency for cytochrome f oxidation falls rapidly in the red region, a half maximal efficiency is obtained at approximately 695 it^I^* 3. The rates of light-induced absorbancy changes due to cyto- chrome f and to Pvoo have been compared at temperatures of liquid nitrogen. Wiile ihe times for completion of the two reactions are approximately the same, the TOO mia change corresponds to a larger absorbancy and hence, has a larger optical rate. However, conversion to a molar basis (assuming that the effect of tempera- ture upon the extinction coefficient of chlorophyll is the same as that upon cytochrorae f ) brings the two rates approximately in- to agreement at the two values of light intensity employed here. k. Two reaction mechanisms fit the experimental data: a) that cytochrome f and P700 ^^^^"^ ^^ different chlorophyll molecules, cytochrome f presumably reacting at the photosynthetically reac- tive center, and P700 acting at a chlorophyll molecule involved in energy transfer; and b) that cytochrome f and P^qq react at the same active center and that cytochrome f is an electron donor and P7QQ an electron acceptor. The first mechanism is consistent with che data of other workers; the second is not in agreement with studies of the oxidation-reduction reactions of PyqO' 81 Brltton Chance and Walter D. Bonner, Jr. FOOTNOTE In the verbal presentation of this paper, it was noted that the ahsorbancy decrease at 705 m^ reverted tovards the baseline after cessation of actinic illumination. The experiments have been re- peated, but the results are irregular. Since the main conclusions on the relationship of cytochrome f and T^qq do not depend upon this observation, this paper reports only observations \inder conditions in which the absorbancy change due to illumination at low temperatures is apparently stable. ACKNOWLEDGMENTS The technical assistance of Brigitte Schoener is gratefully acknowledged. This research is supported by the National Science Foundation and the national Institute of Health. REFERENCES 1. Chance, B., Nishiraura, M., Roy, S. B., and Schleyer, H., Warbvirg Festschrift , June I963, in press. 2. Chance, B., et ^. Charles . F. Kettering Foundation Sympos- ium , April I9S3, in press. 3. Witt, H. T., personal communication. k. Lundegardh, H., Natvire, 192, 21^3 (I961). 5. Olson, J. M., Smillie, R. M., and Bergeron, J. A., Proc . Plant Physiol . Meetings , XXIX, February I963. 6. Duysens, L. N. M. , et al.. Nature, I90, 510 (1961). 7. Witt, H. T.,et al.,-TJa-^ure, 197. 98TI1963). 8. Bonner, W. D., Jr., this volume. 9. Chance, B., Science 120, 767 (195^). 10. Chance, B., Rev. Sci. Instr., 22, 619,627,23^^ (1951). 11. Chance, B., and Nishiraura, M., Proc. Nat. Acad. Sci., h6, 19 (i960). 12. Butler, W. L., this volvune. 13. Chance, B., and San Pietro, A., Proc. Nat. Acad. Sci., k9, 633 (1963). Ik, Kok, B., Cooper, B. , and Yang, L., Microalgae and Photo- synthetic Bacteria, Special Issue,^ 373 (I963). 15. Chance, B., and Nishiraura, M., Proc. Vth International Con- gress of Biochemistry, Vol. VI, Oxford, Pergamon Press, 19^3, p. 73. 16. Duysens, L. N. M. , this volume. LIGHT INDUCED OPTICAL CHANGES IN GREEN LEAVES Walter Bonner and Robert Hill INTRODUCTION In a previous account of some work by the present authors (1) accumulated evidence was presented suggesting that light init- iated oxidation-reduction reactions in higher plants involved cytochromes f^ and b^. As yet, there are no unequivocal data relating to rapid changes in the oxidation states of cytochromes f and b^ on illumination of green leaves or isolated chloroplasts. Light induced optical changes, which have been said to correspond to cytochromes, have been described in isolated chloroplasts of higher plants by Lundegardh (2), James and Leach (3), and Muller, et. al_. (4), The present paper describes the rapid oxidation of cytochrome f which occurs on illumination of green leaves and, in addition7 some observations pertaining to cytochrome b(,. An abstract of some of this work has been published (5). METHODS Etiolated leaves of mung bean ( Phaseolus areus ) were obtained from seedlings grown in the dark at 25°. Spinach leaves were obtained from locally grown plants; spinach chloroplasts were pre- pared by standard procedures. Optical measurements were per- formed either with a double-beam differential spectrophotometer or with a rapid scanning split-beam spectrophotometer; both of these instruments were similar to ones already described (6), Leaves were mounted on a specially constructed rack which fitted into a moist chamber and suspensions of chloroplasts or algae were placed directly into the moist chamber. The basic design of the moist chamber has been described previously (7). Actinic light was provided with a Unitron Koehler illuminator equipped with suitable inCerference filters. In all experiments the photomult ipler of the spectrophotometer was shielded from the actinic light with suitable filters. The visual optical obser- vations were performed with a low-dispersion microspectroscope; the optical path of the comparison prism was equipped with two wedged troughs. RESULTS Cytochromes f^ and b^ are easily observed in acetone extracted chloroplasts and in etiolated leaves (1). It is also apparent 82 83 Walter Bonner and Robert Hill that in certain selected algal mutants it is possible to make direct observations on the chloroplast cytochromes (8) and Indeed in one of them, to show light induced oxidations of cytochrome f^ and of a b-type cytochrome (8,9). The small etiolated leaves of mung bean seedlings show intense bands of both plastid cyto- chromes when the leaves are observed through a direct vision micro- spectroscope. Furthermore, it is possible to record a spectrum of the two cytochromes in a suspension of plastids prepared from etiolated mung bean leaves. Figure 1 shows such a spectrum. — I 1 1 1 r 510 530 550 570 590 Fig, 1 Low temperature (77°K) difference spectrum, reduced- oxidized, of a suspension of plastids isolated from etiolated mung bean leaves. The small etiolated mung bean leaves, minus their mid-ribs could be relatively easily layered on a special rack which fitted into the moist chamber and which in turn, fitted into the optical path of the double-beam differential spectrophotometer. Both of the authors are deeply indebted to Dr. Hiroshi Ikuma who possesses great skill in the delicate operation of mung bean leaf mounting. No light induced optical changes could be observed in the etiolated leaves, in the region between 500 and 580 mn, until the leaves had been allowed to produce some chlorophyll. Greening of the leaves was accomplished by illuminating them, in the moist chamber, with lov? intensity red light. The filter used for green- 84 Walter Bonner and Rober Hill ing had a wide band with 707. transmission between 680 and 1000 ti^i Figure 2 shows the development of a light-activated response following chlorophyll formation in the mung bean leaves, a res- ponse that corresponds to the oxidation of cytochrome f. Initial Response I Hour 554-544 Fig. 2 Development of the light activated cytochrome f oxidation during greening of etiolated mung bean leaves. The times indi- cated refer to the number of hours the leaves were exposed to broad-band, low intensity red light. Light induced cytochrome f oxidation could be observed in a relatively short time following initiation of chlorophyll form- ation. Maximal response was obtained after three hours of greening, at which time the leaves were very slightly tinged with green. Further greeningof the leaves did not increase the extent of the light induced cytochrome f oxidation, but did markedly in- crease the rate of the response." A relatively high light inten- sity was required for cytochrome f oxidation in these partially greened leaves (Figure 3). "" 85 Walter Bonner and Robert Hill Effect Of Light Intensity On Cytochrome f Oxidation 50- ^ -^ :: — • • >t/^ / 40- / * in t •» oO c g • • 1 K 20- 5 o SL '° I 0^ 2 4 6 8 Relative Light Intensity 10 Fig. 3 The relation between extent of cytochrome Jf oxidation and light intensity in partially greened mung bean leaves. Light induced cytochrome f^ oxidation was activated with 700-m^i actinic light only; 640, 660, and 680 mp. were ineffective, 700 m^i actinic light did not cause noticable greening of the leaves, a fact that clearly separates the events leading to the res- ponse and the activation of the response, A spectrum of the 700 m^ light activated responses in part- ially greened mung bean leaves is shown in Figure 4, a spectrum that corresponds remarkably well to that of cytochrome f^. Light induced Rttpontn Partially Gratned Mung B«an Leaves T A0.D.»0.005 S._^ I 1 — SOO 510 — I — 520 530 540 550 560 570 580 m^ Woveiength Fig, 4 The light activated optical res- ponses of partially greened mung bean leave plotted as a function of wave length. 86 Walter Bonner and Robert Hill The spectrum of Figure 4 shows the ei -band of cytochrome JE and some semblence of the 3-band; the complete absence of the 518 response is worthy of note. In this spectrum 559 m^i appears to be isosbestic and at longer wave-lengths there is a region of decreased transmission, a decrease in transmission that could be interpreted as light induced cytochrome b^ reduction. In these experiments there were suggestions that as the mung bean leaves formed chlorophyll the ©(-band of cytochrome b^ became broader and shifted toward the red. However, if one measured, during greening, both the amount of chlorophyll and the amount of cyto- chrome b^ using visual optical methods (a direct-vision micro- spectroscope, the optical path of the comparison prism being equipped with two wedged troughs containing respectively a standard chlorophyll a solution and a standard hemochromogen solution) it was observed that cytochrome b^ gradually disap- peared as the chlorophyll concentration increased but theo^-band of cytochrome b^ remained remarkedly sharp at 560 mp, as long as it could be observed. The results from experiments using both the differential spectrophotometer and the direct vision micro- spectroscope can be interpreted in three ways: 1) There is formation of a chlorophyll-cytochrome complex, a complex which could be similar to that of Takamiya et. al^, (this symposium). In the case of the mung bean leaf the complex would be between cytochrome b^ and chlorophyll a since only chlorophyll a is formed during these early stages of greening; 2) One is observ- ing the light activated oxidation of cytochrome bg an oxidation that requires a higher chlorophyll concentration than the corres- ponding reaction with cytochrome f_; 3) The fact that in the partially greened leaf (and in the fully greened leaf also) there is no light induced optical response in the region of cytochrome b5 oC-band absorption (560-570 mn) while mder the same condi- tions there is a rapid light induced oxidation of cytochrome f might point to the conclusion that cytochrome b^ does not par~ ticipate in light activated electron transport. Oxidized cytochrome f^, like oxidized cytochrome c, possesses a region of steadily increasing opacity in the region 560-570 mn; the transmission decrease between 560 and 570 mn shown in the spectrum of Figure 4 can be interpreted as being caused by the formation of oxidized cytochrome f^. For this reason and because the spectrum of cytochrome b^ remains sharp up to the point of its disappearance, the gradual disappearance of cytochrome bg with greening appears to be caused by a light activated oxidation of this component. In these experiments a light induced cyto- chrome b^ reduction was observed, but only in ruptured chloro- 87 Walter Bonner and Robert Hill plasts. No light induced oxidations or reductions of cytochrome b^ were found In whole chloroplasts or in green leaves. Having learned the technique of observing light induced optical changes in partially greened leaves, it was relatively simple to record light induced cytochrome f^ oxidation in a large variety of fully greened leaves, in green algae ( Chlorella , Chlamydomonas ) and in blue-green algae. The light activated spectral responses in a fully green spinach leaf are shown in Figure 5 where again the characteristic assymetric a-band of cytochrome f^ is strikingly apparent, but unlike the partially greened mung bean leaf, the fully greened leaf shows the response at 518 m^.. There is no 518 m|j. response in Euglena graciles or in Anacystis nidulans , an observation in confirmation of Olson and Smillie (this symposium) and of Amesz and Duysens (10), Light Induced Responitt Spinach Ltaf T AOD -0005 _L 1 1 1 1 510 520 530 540 550 560 m^ Fig, 5 The light activated optical responses of a fully green mature spinach leaf plotted as a function of wave length. It is shown in another paper (Chance and Bonner, this symposium) that the light induced oxidation of cytochrome f^ proceeds at 77° K and the oxidation rate is more rapid at this temperature than at 25° C. Even so, the "on" and "off" responses of the partially greened be^n leaves are remarkably fast at 25oC. The partially greened leaf has the terrific virtue that in the kinds of experiments described here there is a negligible effect of the measuring light on the cytochrome f_ response, a situation that depends on the low chlorophyll concentration and hence the high light intensity requirement and one that does not exist in fully greened tissues or cells. The "on" and "off" responses for 88 Walter Bonner and Robert Hill cytochrome f^ oxidation in partially greened mung bean leaves are shown in Figure 6. It may be seen in this figure that the dark reduction of cytochrome f^ is rapid, so rapid in fact that the techniques that have been used previously to observe light induced optical changes in chloroplasts (2,3) and in Chlorella (11) would fail in any attempt to obsei*ve specific light induced changes. AQD.= 0.005 Fig. 6 An illustration of the time relations involved in light activated cytochrome f^ oxidation and the subsequent dark reduction of cytochrome f^ in partially greened mung bean leaves. DISCUSSION The paper documents, for the first time, specific light- induced optical changes in the partially greened leaf as well as in fully greened leaves and in green algae. This papsr also con- firms the observations of Amesz and Duysens (10) and of Olson and Smillie (this symposium) relating to the light-induced cytochrome f^ oxidation in Anacystis and in Euglena . The accumulated exper- ience gained through comparison of the light-activated cytochrome £ oxidation in partially greened leaves and in various fully greened leaves has emphasized the need for considerable care in investigations on fully greened leaves. Extra precautions are required because of a low light requirement that can be met, partially, by the measuring light. The partially greened leaf has three distinct advantages, compared to the green leaf: (a) better optical properties; (b) the high light intensity require- 89 Walter Bonner and Robert Hill ment for cytochrome f_ oxidation, a requirement that negates possible influences of the measuring light, and (c) the control of the development of light activated responses through control of the leaf chlorophyll content. The observation that light-induced cytochrome £ oxidation can be observed after an etiolated leaf has been allowed to forr> some chlorophyll is simply a prelude to the observations that now must follow. What is the role of cytochrome b^, does it form a complex with chlorophyll, is it oxidized or reduced by light, or does it have any role in photosynthetic electron transport? Very precise data are now needed on rates of chlorophyll forma- tion, development of the f^ response, development of oxygen evolu- tion and carbon dioxide assimilation, plastoquinone formation, the time relations involved in the 518 response as well as correla- tions of the above lore with the light-induced completion of the chloroplast structure itself. Preliminary experiments have shown that many hours of greening are required before plastoquinone is formed. Because of the close association, in the minds of various investigators, between plastosquinone and the 518 response, it is very important to follow the development of this response during greening. It is always a pleasure, however, to find that the solution of a specific problem points immediately to other specific and more challenging problems. SUMMARY 1. Cytochromes f_ and b^ can be observed directly, by means of a microspectroscope, in intact leaves of dark grown mung bean seedlings. Spectra of these same cytochromes can be obtained from plastid suspensions prepared from such etiolated leaf tissue, 2. Following the development of a small amount of chloro- phyll in the mung bean leaves, and while cytochrome b^ is still visible in the microspectroscope, the absorption spectrum of cyto- chrome f_ can be plotted from the light activated optical responses as observed in a differential spectrophotometer, 3. Following more prolonged greening the a-band of cyto- chrome b5, as observed visually, remains sharp but gradually disappears without noticeable change in the position of its maximum, 4. The absorption spectrum of cytochrome f^ can be plotted from the light activated optical responses, as observed in the differential spectrophotometer, in a variety of fully greened leaves and in green algae, 5. No light-induced optical responses corresponding to cyto- 90 Walter Bonner and Robert Hill chrome h(, have been observed in either green leaves, whole chloro- plasts derived from these leaves, or in algae. Light activated cytochrome b^ reduction is found in ruptured chloroplasts. ACKNOWLEDGEMENTS This work was supported by a grant from the National Science Foundation, REFERENCES 1, Hill, R. and Bonner, W,D., Light and Life, p, 424-A35, Ed McElroy, W.D. and Glass, B, Johns Hopkins University Press, Baltimore, 1961. 2, Lundegardh, H., Nature 192, 243-248 (1961), 3, James, W,0. and Leech, R.M, , Nature 182, 1584 (1958). 4, Muller, A,, Rumberg, G. and Witt, H.T, , Proc. Roy, Soc. B. 157,313-332 (1963). 5, Bonner, W.D, and Hill, R., Plant Physiol. 38, xxviii, 1963. 6, Chance, B. , Methods in Enzymology, Vol. 4, p. 273, Ed, Colo- wick, S,P. and Kaplan, N,0. Academic Press, Inc,, New York, 1957, 7, Chance, B. and Strehler, B., Plant Physiol, 32, 536-548 (1957X 8, Chance, B. and Sager, R., Plant Physiol. 32, 548-561 (1957). 9, Chance, B. , Schleyer, H. and Legallais, V,, Studies on Micro- algae and Photosynthetic Bacteria, p. 337-346, Japanese Society of Plant Physiologists, The University of Tokyo Press, 1963, 10. Amesz, J, and Duysens, L,N,M. , Biochimica et Biophysica Acta 64, 261-278, 1962. 11. Lundegardh, H., Physiol, Plantarum ]_, 375-382 (1954), ACTION OF TWO-PIGMENT SYSTEM ON FLUORESCENCE YIELD OF CHLOROPHYLL A W. L. Butler and N. I. Bishop Light-induced fluorescence yield changes of chlorophyll a in vivo have been related to the two-pigment system of photosynthe- sisT Govindjee, et al./^-' reported that the intensity of flu- orescence emitted by chlorella, when illuminated with 67O- and 700-nm light simultaneously, was less than the sum of the inten- sities obtained with the 67O- and 700-nm beams separately. Butler (2) showed that the yield of chlorophyll a in vivo was re- versibly increased by irradiation with red light and decreased by irradiation with far red. The action spectrum for the effect of far-red light in decreasing the fluorescence yield had a maximum at 705 nm w ) which was similar to the action spectra for the effects of long-wavelength light in enhancement phenomena of the second Emerson effect. Teale ^^^ reported that with green algae and chloroplasts the action spectrum for the effect of light in increasing the fluorescence yield had maxima at ij.70 and 6^0 nm and a shoulder at 67O nm, a typical action spectrum for the ,. shorter wavelength pigment system in green plants. Duysens^-'' ' also showed with red, blue-green and green algae that the fluor- escence yield of chlorophyll a was increased by light absorbfd by the shorter wavelength pigment system (which he called system 2) and decreased by the longer wavelength pigment system (system 1). We have also measured the action spectrum for the light- induced fluorescence yield increase in green leaves and algae and will report the results in the present paper. The symbolism and theoretical framework introduced by Duysens v5jO) ^11 be adopted in this paper. It will be assumed that pigment systems 1 and 2 both contained chlorophyll a and the accessory pigments (chlorophyll b in green plants and phycobil- ins in red and blue-green algae) but that most of the accessory pigment is associated with system 2. Also, according to Duysens, the chlorophyll a in system 1 is weakly or nonf luorescent, while the chlorophyll a in system 2 fluoresces with a variable yield that depends on the redox state of a quenching substance Q. A similar quenching substance was proposed by Kautsky, et al. ^ '•' on the basis of a detailed kinetic analysis of fluorescence 91 92 W. L. Butler and N. I. Bishop transients following the onset of illumination. In the sympli- fied electron transport chain proposed by Duysens and Sweers^ ^> H20 -> (system 2) -^ Q -> cyt -^ P -> (system l) -^ PN Q will quench the fluorescence of the chlorophyll a in system 2 but the reduced state, QH, will not quench. Light absorbed by system 2 will reduce Q to QH, thus increasing fluorescence while light absorbed by system 1 will oxidize QH to Q, thereby quench- ing the fluorescence. P is a small amount of a chlorophyll absorbing near 700 nm. Kok (^^ showed that this component, which he called P 700, is oxidized and thereby bleached by light absorbed by system ^/ai?d that it is reformed by light absorbed by system 2. Butler ^^ ^ studied this component (denoted G-705 in his work) by low-temper- ature absorption and fluorescence excitation spectroscopy and showed that energy absorbed by a large bulk of chlorophyll a is transferred by inductive resonance to P. Presumably, P is the energy trap for system 1. Duysens and Sweers ^ ^ have also pro- posed that Q is the energy trap for system 2. The chlorophyll transfers excitation energy to Q but not to QH. So far, there is no direct spectroscopic evidence for this energy trap. In the scheme proposed by Duysens, system 1 and system 2 should be re- moved from the path of electron flow. They drive the electron transport chain by energy transfer to P and Q. Butler (3) attempted to account for the fluorescence yield changes of chlorophyll a on the basis of energy transfer from chlorophyll a to P but not to Pqx- This explanation, however, did not account for the action of far-red light in decreasing the fluorescence yield. Duysens' scheme, in which Q is the quencher of system-2 chlorophyll, accounts satisfactorily for the experimental observations on fluorescence yield changes and relates these to changes in the electron flow along the electron transport chain. The fluorescence yield measurements thus be- come a convenient assay to determine if the electron flow is functioning. Such measurements will be reported in the present paper on mutants of Scendesmus which have specific blocks in the photosynthetic electron transport chain. METHODS AND MATERIALS The instrument, previously (3) used to measure the relative fluorescence yield immediately following a brief actinic irradi- ation, was modified so that the fluorescence excited at low in- tensity could be measured during the actinic irradiation. The 93 ¥. L. Butler and N. I. Bishop modified, instriinient was similar in principle to that used by Duvsens ^^^ - The fluorescence was excited by a weak (^0 ergs/ cmVsec), 650- nm monochromatic beam which was chopped at 36O cycles/sec. The sample could be simultaneously illuminated with a high intensity (1000 to 2000 ergs/cmVsec), monochromatic acti- nic beam. A cut-off filter, placed between the sample and photo- tube, blocked the low-intensity, measuring beam and the high- intensity, actinic beam but transmitted the fluorescence of wave- lengths longer than 710 nm. It was previously shown with a green leaf (3) that the light-induced fluorescence changes, even when limited to wavelengths longer than 730 nm, were due to changes in the yield of chlorophyll a fluorescence. The photometer incorpo- rated a tuned amplifier which was tuned to the chopping frequency of the measTiring beam so that the alternating fluorescence excit- ed by the chopped, low-intensity beam was measured but the con- stant fluorescence excited by the steady, high-intensity, actinic beam was not. The actinic beam changed the fluorescence yield which resulted in a change in the intensity of fluorescence excited by the chopped beam. The intensity of fluorescence excited by the actinic beam was monitored with a D.C. voltmeter which measured the drop across the anode resistor. The intensity of the actinic light was adjusted at each wavelength such that the intensity of fluorescence excited by this beam was about 20- fold greater than that excited by the measuring beam. Absorption spectra were measured at -196°C with a single-beam recording spectrophotometer similar to one described previously do). The spectrophotometer could also record derivative spectra by differentiating the signal from the photometer electrically. Measurements were made on green bean leaves, on spinach chlo- roplasts and on suspensions of Scenedesmus. The Scenedesmus were wild-type cells and two classes of mutants which have been des- cribed previously by Bishop (^^^2). One class of mutants ("GO2" mutants) will not fix CO2 in the light but will evolve O2 in a quinone Hill reaction. The other class of mutants ("O2" mutants) will not evolve O2 but will photoreduce CO2 in a hydrogen atmosphere. RESULTS MP DISCUSSION Action spectra for the effects of light on the fluorescence yield of a green leaf in air and in nitrogen are shown in Fig. 1. The fluorescence yield of the leaf is somewhat greater in nitro- gen presumably because Q is more reduced. The horizontal lines show the relative intensity of fluorescence in the absence of 94 W. L. Butler and N. I. Bishop actinic light. In the presence of monochromatic actinic light, the intensity of fluorescence excited by the weak, chopped, 650-nm beam is given by the curve through the points. The in- tensity of actinic light was adjusted at each wavelength to give the same level of fluorescence as measured by the B.C. voltmeter. This procedure was adopted to make sure that approximately the same energy was absorbed at each wavelength. The intensity of actinic light at 5^0 nm was $0 percent greater than that at 6^0 nm because less light was absorbed at 5^0 nm. This method does not insure that precisely equal energy will be absorbed at all actinic wavelengths because the fluorescence yield appears to vary approximately 20 percent with wavelength across the visible spectrum. Adjusting to a constant level of fluorescence is more valid, however, than using a constant intensity of actinic light which would make the red and blue maxima in Fig. 1 more pro- nounced simply because those wavelengths were absorbed more 140 RELATIVE FLUORESCENCE 120 YIELD 100 80- 400 500 600 WAVELENGTH -nm 700 Fig. 1. Relative fluorescence yield of green bean leaf in the presence of actinic light vs. wavelength of actinic light. Betails of measurement in text. 95 W. L. Butler and N. I. Bishop effectively. The action spectra of Fig. 1 are strickingly simi- lar to the enhancement spectrum for chlorella shown by Myers and Graham CU). The action spectra show that the wavelengths absorbed prefer- entially by system 2 are the most effective in increasing the fluorescence yield. Green plants do not show a clear-cut re- sponse to system-1 absorption because of the spectral overlap be- tween systems 1 and 2. Beyond 680 nm, however, the absorption spectrum of system 2 falls markedly and the direct absorption by P becomes more important so that Q can be driven largely to the oxidized state by far-red light. It was previously shown that the most effective wavelengths for decreasing the fluorescence yield are those absorbed directly by P \^J. Far-red light does not depress the fluorescence yield of the leaf in air much below the value for a dark leaf. Thus, Q is largely oxidized in the dark in air. The fluorescence yield which obtains when the Q is largely re- duced can be observed by adding DCMU. This herbicide apparently blocks the electron transport chain between Q and cytochrome so that P cannot oxidize QH to). The addition of DCMU causes the_ fluorescence yield to increase approximately 3-fold. This maxi- mal fluorescence yield also obtains momentarily during the tran- sient fluorescence burst which occurs when a darkened leaf is first placed in bright light. The effect of DCMU on spinach chloroplasts is shown in Fig. 2. The chloroplast preparation without DCMU shows a typical two-pigment response. Light at 650 nm increases the fluorescence yield somewhat more than 600-nm light and much more than 690-nm light. After the addition of DCMU (10-^M) the fluorescence yield increased slowly under the influence of the weak, meas-uring light and abruptly with the 650-nm actinic beam. The fluorescence yield drops somewhat when the actinic light is turned off, indicating that QH can be oxi- dized to some extent either by a back reaction or by dark metab- olism. The light-induced fluorescence yield in the DCMU-treated chloroplasts is independent of wavelength, showing that system 1 or P have no influence on Q. The small drop at 690 nm is an artifact due to the small amount of the 690-nm actinic light which leaks through the 710-nm cut-off filter. In the case of the chloroplasts without DCMU, the marked depression of the fluo- rescence yield at 690 nm is due largely to the oxidation of QH by system 1 and P. The light-induced fluorescence yield changes have been used to determine if both pigment systems were operating in mutants 96 W. L. Butler and N. I. Bishop of Scenedesmus. The fluorescence measurements on the wild-type Scenedesmus cells in Fig. 3 show the typical two-pigment con- trol: The fluorescence yield is greater in 650-nm light than in 600-nm light and is markedly depressed by 690-nm light. One set of mutants, called "COg" mutants, fails to fix CO2 in the light but will evolve O2 in a Hill reaction. The fluores- cence measurements in Fig. 3 show that the fluorescence yield of the "CO2" mutant is about 3 -fold greater than that of the 8 6 RELATIVE FLUORESCENCE INTENSITY o in o in 000^ o in au- ID to (DO I MM in (Dlt; ID CD CC o 1 i M 3 o K-lmin.H TIME Fig. 2. Relative intensity of fluorescence {\p < 710 nm) from spinach chloroplasts excited by low-intensity, chopped, 650-nm beam in presence and absence of constant, monochromatic, actinic beam of wavelength indicated. Effect of lO'^M DCMU shown. 97 W. L. Butler and N. I. Bishop wild-type cells. The fluorescence yield is increased further in the presence of actinic light, but all wavelengths are equally effective. The same results were obtained by adding DCMU to the chloroplasts in Fig. 2 and the same results would have been ob- tained by adding DCMU to the wild-type Scenedesmus cells. The fluorescence measurements indicate that system 2 reduces Q in the light but that the electron transport chain is blocked some- place between P and Q because system 1 and P cannot oxidize QH. The block in the electron transport chain prevents the reduction of TPN which is needed for CO2 fixation but does not prevent system-2 activated transport of electrons from H2O through Q to benzoquinone. RELATIVE "^ FLUORESCENCE INTENSITY 3 _ o o o 2 <«- O U) U3 «5 tf> ^ o wild type (3Xsens.) CO2 mutont Oj mutant min- TIME Fig. 3. Relative fluorescence intensity (same measurement as Fig. 2) from Scenedesmus wild-type, "CO2" mu- tant No. 8 and "O2" mutant No. 11 cells. Sensi- tivity of fluorescence measurement was increased 3-fold for wild-type cells so that the fluores- cence yield of the wild-type cells is approximate- ly 1/3 that of the mutant. 98 ¥. L. Butler and N. I. Bishop The other set of mutants, called "O2" mutants, fails to evolve O2 in the light but will photoreduce CO2 in a hydrogen atmos- phere. These mutants show a high fluorescence yield (Fig. 3) which is not affected by actinic light. These measurements are consistent with the photochemical activities of the "O2" mutants which indicate that system 1 is operating, since CO2 can be photoreduced with H2j tiut that system 2 or some reaction closely allied with O2 evolution is not functioning. Low-temperature absorption and derivative spectra of the wild- type and mutant cells are shown in Fig. U. The derivative spectra were used primarily to distinguish the presence of P, the small amount of chlorophyll absorbing near 700 nm. The 1 — . — . — , — 1 — r absorption derivative wild type 1— Oo mutant COj mutant 400 500 600 700 WAVELENGTH -nm 600 700 Fig. U. Absorption and derivative spectra of Scenedesmus wild-type, "O2" mutant No. 11 and "CO2" mutant No. 8 cells at -196°C. 99 ¥. L. Butler and N. I. Bishop absorption spectra show that essentially the same amounts of chlorophyll b, chlorophyll a-670 and chlorophyll a-680 are pres- ent in all samples. Thus, the mutants carry a normal compliment of pigments in systems 1 and 2. P is present in the xd.ld-type cells and in the "O2" mutant, as shovm by the slight shoulder in the absorption spectra on the long-wavelength side of the main chlorophyll-absorption band and by the maximum in the derivative spectra near 700 nm. The spectra for the "CO2" mutant, however, do not show the presence of P. Thus, in these mutants, system-1 pigments cannot affect the redox state of Q because of the block in the electron transport chain at P. System-2 pigments, how- ever, can cause Q to be more reduced in the light. The high fluorescence yield of these mutants in the absence of actinic light suggests that the Q is largely reduced in the dark. It would be of interest to determine if the cytochrome in these mu- tants could be reduced in the light by system 2 and oxidized in the dark as is Q. The block in the "O2" mutants has not been localized. The absence of a light affect on the fluorescence yield shows that Q is not being affected by light even though the absorption spec- trum indicates that system-2 pigments are present. The high fluorescence yield suggests either that Q is fully reduced in the dark or that it is not present. The absence of Q in the "O2" mutants would be analogous to the absence of P in the "CO2" mutants . The fluorescence yield changes are consistent with the photo- synthetic electron transport chain and with the photochemical activities of the wild-type and mutant cells. These measure- ments provide a rapid and convenient method to study metabolic inhibitors and mutations which affect photosynthesis. 100 ¥. L. Butler and N. I. Bishop REFERENCES 1. Govindjee, S. Schimuri, C Cederstrand and E. Rabinowitch, Arch. Biochem. Biophys., 89, 322 (i960) 2. Butler, ¥. L., Plant Physiol. Suppl., 36, IV (l96l) 3. Butler, W. L., Biochim. Biophys. Acta, 6k, 309 (1962) h. Teale, F. ¥. J., Biochem. J., 85, IUP (1962) $. Duysens, L. M. N., Proc. Royal Soc, B, l57, 301 (1963) 6. Duysens, L. M. N. and H. E. Sweers, in Microalgae and Photosynthetic Bacteria, Univ. Tokyo Press, Tokyo, 1963, p 353 7. Kautsky, H., ¥. Appel and H. Amann, Biochem. Z., 332 , 277 (I960) 8. Kok, B. and G. Hock, in A Symp. on Light and Life, Eds., ¥. D. McElroy and B. Glass, Johns Hopkins Univ. Press, Baltimore, Md., 1961, p 373 9. Butler, ¥. L., Arch. Biochem. Biophys., 93_, Ul3 (l96l) 10. Norris, K. H. and ¥. L. Butler, IRE Trans. Biomed. Electronics, BME-8 , 153 (I96l) 11. Bishop, N. I., Nature, 195, S^ (1962) 12. ¥eaver, E. G. and N. I. Bishop, Science, lUO, 1095 (1963) 13. Myers, J. and J. Graham, Plant Physiol., 38, 105 (1963) PRINCIPLES OF A THEORY OF ENERGY UTILIZATION IN PHOTOSYOTHESIS James Franck and J. L. Rosenberg The present paper is an abstract insofar as it leaves out many details for the sake of brevity. An extended discussion will be published elsewhere. A short introduction which will be neither historical nor comprehensive in literatiore citations, will be presented first. Our theory developed slowly from early attempts to find a general point of view for the meagre evidence available at that time. They were mostly based on van Niel's principle of water splitting, certain kinetic phenomena, and fluorescence observa- tions . Along with proposals for which a sound experimental basis existed were a number of guesses which, londer the influence of subsequent experimentation, became obsolete. One of the deduc- tions which we regarded, and still regard, as well founded is that in green plants the excitation energy collected by the bulk of the plant pigments is transferred always to one center of photochemical activity, namely a chlorophyll a molecule in a special position which enables it to use the energy for the photochemical reactions of photosynthesis. Although this state- ment referred originally only to photosynthetic ixnits in green plants, we extended it later to those in all plant cells. The energy- collecting part of the non-photochemically active regions of the unit contains the "protected" pigments, so-called because of the lack of contact between these dye molecules with water and its solutes. Their natxore of the pigments may vary in different classes of plants or even from unit to unit within a given plant cell. In green plants, where all units contain chlorophyll a chlorophyll b, and p-carotene, the energy transfer to the lowest available excitation level is so quick, in accordance with well- known principles applied to dense and partly ordered systems, that the energy absorbed by all dyes reaches the exposed site as chlorophyll a excitation. The excitation arriving at the reac- tion center is usually that of the first excited singlet state, with the consequence that fluorescence visible during photo- synthesis at non-excessive light intensities is predominantly the red fluorescence of chlorophyll a. 101 102 James Franck and J, L. Rosenberg Actually chlorophyll exists in the xonits in two different modifications^ as the occurrence of a minor red-shifted set of absorption "bands indicates. One of these modifications is usually present to a small extent and is contained in only a few units, so that it can exert only a minor influence on photo- synthesis . (At this point we must state that in our previous publications we ascribed the weak spectrum to a so-called n-jt transition which woiild be present in all units. We used special assumptions to explain why the energy would not always flow to this level. Our reasons for abandoning this interpretation will be presented in the forthcoming extended discussion). Under these conditions intensity measurements of the red fluorescence d\iring and in the absence of photosynthesis are an uneq-uivocal criterion of the utilization of the arriving singlet excitation energy for photochemistry. If only singlet energy would be used with optimal quantum yield there should be no fluorescence during photosynthesis. If^ on the other hand, photochemistry always occurred by the action of metastable states formed from the excited singlet state in competition with fluorescence emission, the intensity of fluorescence would be independent of photo- chemical utilization. Observations showed that under the conditions mentioned above the ratio of the intensity of fluo- rescence in the presence or absence of photochemical utilization of excitation turned out to be very nearly one-half. This result was the same for three different methods of suppressing and permitting optimal photosynthetic activity. The obvious conclu- sion is that singlet and metastable excitation energy are \ised equally often for photosynthesis . That both singlet and triplet states are used for photosynthesis equally often is an indication that two photochemical reactions, one at the cost of each of these two states, must be coupled in such a way that they auto- matically occur at the same rate. That this coupling can be explained without contradicting any of the new important obser- vations of the second Emerson effect and of the occurrence, spectroscopy, and photochemistry of the chlorophyll modification absorbing at 7OO mjj., together with earlier and more recent studies of light emission of chlorophyll a is the theme of this paper. No attempt has been published so far to explain the emission and absorption phenomena of chlorophyll a together with the Emerson effect and reversible bleaching on the basis of a two-center model. We have attempted to adapt niomerous published proposals to the above criteria but have not succeeded. 103 James Franck and J. L. Rosenberg NATURE OF THE TWO PHOTOCHEMICAL ACTS For the application of a pictiire in which singlet and meta- stable state energy of chlorophyll a are used for the two photo- chemical steps of photosynthesis, we have to be more specific about these steps. From the work of Hillv^) and Kamenv^j it is known that cytochrome f (designated as Cyt-^ed ^^^ ^y^ox ^°^ ^^^ ferro- and ferri-forms, respectively) plays a role in the photo- chemistry. If the light reactions consist of water-splitting, the cytochrome is regarded in our scheme as the primary OH acceptor while the H acceptor designated as A, is the primary photosynthetic oxidant. From the determination of the standard electrode potential of Cyt in the chlorophyll -cytochrome complex (Chl-Cyt) by Kok(5)^ we know that the Cytox cannot evolve oxygen by dark reactions alone. Therefore a second photochemical step is needed on the Oh side by which CytQ^ is reduced and another enzyme Y is oxidized to a state capable of yielding oxygen in subsequent dark reactions. Both Cyt and Y must be complexed with the exposed chlorophyll. The scheme is summarized in the fol- lowing set of essential chemical reactions, where (l) and (2) are reactions of photo-excited Chi molecules, the first in the metastable state and the second in the excited singlet state, (la) and (2a) represent sequences of enzymatic dark reactions. A + (Chi* .-Cyt J -> AH + (Chl-Cyt ) (l ) ^ met red ox 2 AH -> AHg + A (la) Y , + (Chl* . -Cyt ) -> (Chl-Cyt ^) + Y ^ (2) red ^ ^^"- sing "^ ox^ ^ "^ red' ox Y ^ Y . + ^ (2a) ox red 4 2 GENERAL CONSEQUENCES OF THE SCHEME The phenomena described in this section will be first applied to green plants, with omission of the long wavelength absorption (above 680 m\i) . The consequences of the far-red absorption will be considered later. Since the exposed chlorophyll comes in contact with many kinds of diffusible substances, a possible competition may exist for the 104 James Franck and J. L. Rosenberg use of metastable excitation of Chi between A and other oxidants, some as harmful as moleciolar oxygen. Although in vitro experi- ments show that photo-oxidation of chlorophyll is much less likely thaa simple oxygen quenching of excitation, the former process does occur with a small yield. In any event, mere impacts of oxygen with CM.* may divert the use of excitation from photosynthesis . Since it is known that oxygen has a negli- gible influence on photosynthetic rates in regions of low irradiation intensity, we condLude that A must be present in at least a high enough concentration (lO~^ molar or more) to win out over Oo for the excitation. Among substances which do not meet this criterion for A are most of the materials present in catalytic amoiints, including TPN. Phosphoglyceric acid, PGA, does meet the requirement and has been tentatively identified by us with substance A. In support of this view is the fact that other respiratory intermediates, normally present in smaller concentrations than PGA, become the substitute oxidants for Reaction (l ) during periods of unavailability of PGA, such as cyanide inhibition, induction periods, and periods of greening of etiolated leaves' ^^5 J. At intensities approaching saturation the photosynthetic oxidant becomes limiting, and oxygen becomes a more effective competitor. This is the explanation for the strong depressing influence of oxygen on the saturation rate of photosynthesis. Although photo-oxidation of the exposed chlorophyll still proceeds with only a small quantum yield under these conditions, in the absence of a recovery process there would be a progres- sive accumulation of units whose exposed chlorophylls, oxida- tively bleached, are unable to allow excitation to move to the reaction centers. Another competing process occurring with a probability rising with irradiation intensities is the collection by Chl-^jj^g^ of a second quantiom of excitation before Reaction (l ) has been sensitized. This double excitation, leading to photo- ionization and the associated afterglow, will not be discussed further in this paper, except for the remark that the connected bleaching would res\ilt in a disruption of the flow and utiliza- tion of excitation energy collected by the protected chlorophylls. The disruption of energy flow may be avoided by a built-in recovery process, (5), analogous to the reaction described by Kok(3) and by Wittv°) for the special case of the 7OO m|j. absorbing form of chlorophyll. This recovery of normal exposed Chi would be rapid if the Cyt in its reaction center complex is Chi + Cyt ^ ^ Chi + Cyt (3) red ox ^ -^ ' 105 James Franck and J. L. Rosenberg in the reduced form. The rate woilLd then he so great that the reversible bleaching becomes non- observable in difference spectra experiments. Since photo-oxidation might attack any exposed Chi at random, efficient statistical recovery requires that practically all the Cyt molecules at the reaction centers be in the reduced form during steady- state illumination. We nov face the problem: how can most of the Cyt be in the reduced form? Simple kinetic analysis of our scheme shows that if each photosynthetic unit were functionally independent, half of all the Cyt should be in the reduced form and half oxidized during steady light-limiting illumination. In fact in the intro- duction we implied an independence of the imits. We now qualify this statement by introducing an interaction at the level of the exposed chlorophylls . We postiolate an energy migration of sing- let excitation among the exposed chlorophylls in a larger super- unit of the chloroplast. This proposal does not invalidate our conclusion that the far-red absorbing modification in green plants does not act as a trap for normal chlorophyll a singlet excitation, as will be dlsc\issed in the section on the Emerson effect. With this limited type of super-organization we gain the advantage that the steady- state fraction of Cyt in the oxidized condition is lowered to l/2 n, where n is the n\imber of participating lonits in the exchange. We visualize n to be of the order of magnitude of 100 or more. On this basis we can •understand why good photosynthesis does not change the level of oxidized cytochrome, as shown by difference spectra. That our scheme provides an automatic adjustment of equal opportunity for photochemical use of singlet and metastable excitations is obvious. Next we must mention several consequences of our scheme for the fluorescence yield of exposed chlorophylls . a) Butler has shown that freshly formed chlorophyll in a greening leaf has a fluorescence yield practically as high as that of chlorophyll in organic solvents. W/ Absolute measure- ments show that the fluorescence yield in vitro is about 25^ while that of fully developed chloroplasts, in which only the exposed chlorophyll a is responsible for fluorescence, is G°Jo in the absence of photosynthesis. This difference may be ascribed to the influence of the heavy atom, the iron of the coraplexed Cyt, on the transition to the metastable state. 106 James Franck and J. L. Rosenberg b) The fluorescence of the protected chlorophyll, if the absorbed energy is forced to remain inside the unit, should thus also be higher than of the exposed chlorophyll. If;, for instance by photo-oxidation of the exposed chlorophylls during strong illumination or by other methods, the energy is prevented from migrating outvard in the unit, fluorescence would be observed from the protected pigment molecules with up to a four-fold increase in yield. Indeed such observations have been made with a variety of external treatments (°^ 9 ) some of them destructive for the chloroplastsv^'-'). c) Our scheme forces us to the conclusion that half of the excited exposed chlorophyll molecules emit their filLl share of fluorescence, independent of the level or even absence of photo- synthesis, while the other half, involved in Step (2), use singlet excitation immediately dioring good photosynthesis without any fluorescence. The consequence, that the lifetime of excita- tion of those emitting the fluorescence, is independent of photo- synthesis, has been proved by several types of observation ' '. EMERSON EFFECT AUD RE VERSIBLE BLEACH ING AT TOO my. Green Plants We first discuss our interpretation of the far-red absorp- tions. After being forced by a number of new observations (to be described in the forthcoming extensive discussion) to abandon oirr previous idea that nit transitions were responsible for the long wavelength absorption. We based oiur picture on a proposal originally made by Brody that chlorophyll exists within the photosynthetic apparatus partly in monomeric form and partly in aggregates(l2) , This idea received strong support from the discovery by Olson et al. that the absorption at 700 to.\s. and the fluorescence above TOO mjj. are highly polarized in the planes of the lamellae, while the major components observed in both absorption and luminescence are not polarized, ^'l^^l^j Specif- ically, we visualize most of the pigment molecules as being in a partly disordered, amorphous state within their \inits, while in some small fraction of the units there are regions of two- dimensional crystallinity bordering on the reaction center. As are all absorption peaks of the crystalline patches the red absorption is shifted slightly toward the inra-red and is split into two components . This is indeed the type of behavior observed by chlorophyll microcrystallites in vitro by Rabinowitch et al.^ 5/ Fig, 1 shows the red shift and splitting 107 James Franck and J. L. Rosenberg 700 ove length in m>j Fig. 1. Absorption spectriom of chlorophyll a in ether 10-5 M^ cijrve 2; of colloidal chlorophyll siispension in JOfo aqueous methanol, crosses of curve 1. Curves from reference (l5)- of the long wave-length peak in their "colloidal" samples in comparison with monomeric material. Of the three chlorophyll a components fo\and by French and Brown(l6)^ ve ascribe the main peak at about 67O to the amorphous chlorophyll, and the two minor peaks at about 68O and 7OO m\x to the two differently polar- ized Davydov components of the crystalline chlorophyll. We note that the relative proportions of crystalline to non- crystalline chlorophyll may vary with ciiltixring conditions and need have no particular stoichiometric value. Also, the extent of the absorp- tion shift depends on the size of the microcrystallites'lT j and a distribution of crystal size woiiLd be expected to cause broad- ening of the bands. If OTor scheme makes possible a self-consistent explanation of the Emerson effect, it must provide the answer to five questions. a) Why does the photosynthetic quantum yield fall in the far- red with increasing wave-length of absorption in the crystalline regions? 108 James Franck and J. L. Rosenberg "b) Why does the stronger Davydov absorption at 680 ni|i give a good quantum yield while at the TOO m^i component the quantum yield is reduced to one-fourth or less? c ) Why can simultaneous or almost simultaneous absorption by the amorphous chlorophyll enhance the effectiveness of absorp- tion by the crystals? d) What is the role of the reversible photo- or oxidative bleaching and why is this bleaching observed mainly in the 700 m^l absorption? e) In view of the energy migration through the super-unit which we introduced in the previous section, what prevents the degradation of excitation in the amorphous layers to the lowest excitation level of the crystalline layers, 7OO m\x, by migration throiigh the exposed chlorophylls? To answer the first three questions we use the idea previously advanced by us that excitation provided by 7OO mn absorption is not sufficient to excite the singlet level of the exposed cMLorophyll, which is an amorphous type molecule having its absorption at about 675 m[x whether it is attached to an all- amorphous -unit or to a unit containing a crystalline patch. As a result excitation moving quickly in the crystal to the contact point with the exposed chlorophyll will suffer there a trans- ition into a metastable state. A re-interpretation of Becker's observations (18) shows that the order of the metastable levels of crystalline and amorphous chlorophyll is the reverse of that of the singlet levels, and that a transfer of metastable energy to the amorphous exposed molecule will occur. The higher energy Davydov crystal component, 68O m|j. is energetically so close to the singlet level of the exposed chlorophyll that by thermal fluctuation the 68O excitation may easily excite the singlet level of the exposed chlorophyll. In competition with this process, degradation of the 68O to 7OO or even further to the metastable level is an inherently slower event. If a reason can be found why the metastable state of an exposed chlorophyll is inefficient in sensitizing Reaction (2), we could understand the long wave-length decline and the enhancement. We considered several possible mechanisms, as follows : a) Back-reaction between AH and CytQx 109 James Franck and J. L. Rosenberg As opposed to singlet excitation, metastatle excitation cannot participate in the migration at the super--unit level. As a result, each Cyto^ formed during a period of irradiation with 700 m[x only must wait a much longer period to receive a quantim of excitation for sensitizing Step (2). The Cyto^ "^H "thus have a greater chance to he destroyed by a competing dark reaction, such as one with AH. Our objection to this possi- bility is that it has no connection with the bleaching of crystalline chlorophyll, which plays such a great role in observed difference spectra. b) Photo- oxidation of exposed chlorophyll. Althovigh photo-oxidation may occur with small probability with any type of irradiation, there are two important differ- ences when the oxidized chlorophyll is at the reaction center of a crystal -containing unit. The first is that the res\alting Chl+ captures an electron from a neighboring crystalline chloro- phyll because the ionization potential for the latter is lower. This in itself would slow the recovery reaction (5). The second difference is that half of the Cyt is oxidized at the steady state and is unavailable for the recovery reaction (3). This follows from the non-migration of metastable energy among exposed molecules and is in agreement with the observed increase in oxidized cytochrome dirring long wave-length irradiation. A progressive accumiAlation of units containing Chl+ within the layers woiild lead to low efficiency if the trapped hole impedes the flow of excitation energy outward. Our objection to this possibility is that it is in contradiction to the finding by Gorindjee and Rabinowitch of the long wave-length decline and enhancement for the Hill reaction in the absence of oxygen^l9J. c) Alternative electron transfer in competition with Step (2), We propose that triplet excitation can be used to sensitize Reaction (h), which is a reversal of the recovery reaction (3), Cyt + Chi* ,^ Cyt + Chl"^ {h) "^ ox met while singlet excitation cannot be so used. In the first instance, such an event would resiolt in the deto\xr of two quanta from photosynthetic purposes. Secondarily, it would produce much greater losses because the resultant bleached Chl would transfer its oxidation to crystalline chlorophyll. As in (b ) above, the trapped Chl+ within the layer would impede the outward flow of excitation until it has imdergone recovery by a dark recapture of an electron from the Cytred. The basic idea 110 James Franck and J. L, Rosenberg imderlying this possibility is not so much an ad hoc hypothesis as it might seem. It is not unplausible to assTune that in an excited Chl-Cyt complex more charge vould tend to be transferred from the Chi to the Cyt for triplet excitation than for singlet under the influence of the mutual repulsion of two electrons with parallel spins. This explanation provides an interpretation of the Emerson effect, the enhancement by short-wave irradiation^ and the 7OO bleaching. Finally, we have to discuss the possible role of the 7OO m\x levels as a sink for shorter wave-length excitation. In our model, singlet excitation of amorphous chlorophyll molecules within a crystal -containing unit may be degraded to the 7OO level during its passage through the crystal on the way to the reaction center. The converse is not true, however. Singlet excitation of exposed chlorophylls migrating through the super-unit might sensitize a crystalline moleciile bordering on some reaction center, but would do so by preferentially sensitizing the stronger 68O level rather than the 7OO, with which the resonance overlap is poorer. As discussed previously, 68O excitation near the reaction center can be used with high probability for recreating singlet excitation at the exposed site. Thus, an overall degradation of singlet to triplet excitation at the super-unit level can occur with only a small probability. Red and Blue -Green Algae These algae can be incorporated into our general scheme if we postulate the following pict-ure of the photosynthetic apparatus. Most of the chlorophyll is contained in crystal-containing units, so that even 67O m^ absorption has the effect of sending mostly triplet excitation to the reaction centers . The phycobilins are contained in units devoid of chlorophyll, but these units still have chlorophyll a as the exposed pigment at the reaction center. These phycobilinous units can provide interaction with the chlorophyllous units by way of super-unit energy migration in such a manner as to divide the photochemical chores in part between the two kinds of unit. Absorption in the chlorophyll units will be used predominantly for Step (l ) while absorption in the phycobilin units will be used more than half the time in sensitizing Step (2). Ill James Franck and J. L. Rosenberg ACKNOWLEDGMENTS This work was supported "by the United States Air Force Office of Scientific Research of the Air Research and Development Command under contract No. AFi^9( 638)762 and by the Office of Naval Research, Department of the Navy, \inder contract m^0k-kl6 with the University of Pittsburgh, and by the Fels Fund of Philadelphia. REFERENCES 1. Davenport, H. E., and Hill, R., Proc . Roy. Soc . (London) Ser. B, 139, 327 (1952). 2. Kamen, M. D., in "Enzymes: Units of Biological Structure and Function" (D.H. Gaebler, ed. ), p. h-Q^, Academic Press, New York (1956). 3. Kok, B., Biochim. Biophys. Acta, kQ, 527 (1961). k. Franck, J., Handbuch der Pflanzenphysiologie, Vol. V, p. 689, Springer, Berlin (1960). 5. Smith, J.H.C., Proc. Fifth Int. Cong. Biochem., 6, I5I (1963). 6. Witt, H.T., Muller, A., and Rumberg, B., Nature, 192, 967 (1961). 7. Butler, W. L., Arch. Biochem. and Biophys., 92, 287 (1961). 8. Lumry, R., Mayne, B., and Spikes, J. D., Discussions Far. Soc, 27, 1^9 (1959). 9. Weber, G., in "Comparative Biochemistry of Photoreactive Systems" (M.B. Allen, ed. ), p. 395, Academic Press, New York (1960). 10. Tumerman, L. A., Borisora, 0. F., and Rubin, A. B., Biofizika, 6, 6^5 (1961). 11. Brody, S. S., and Rabinowitch, E.I., Science, 125 , 555 (1957). 12. Brody, S. S., and Brody, M., Nature, I89, 5^7 (1961). 13. Olson, R. A., Butler, W. L., and Jennings, W. H., Biochim. Biophys. Acta, 5|+, 615 (I962). ik. Olson, R. A., Butler, W. L., and Jennings, W. H., Biochim. Biophys. Acta, 58, l^U (1962 ). 15. Ichimura, S., and Rabinowitch, E., Science, I3I , 131^ (196O). 16. Brown, J.S., and French, C.S., Plant Physiol., 5^, 305 (1959). 17. Jacobs, E. E., and Holt, A. S., J. Chem. Phys., 20, I526 (1952). 18. Fernandez, J., and Becker, R. S., J. Chem. Phys., ^, ^^-67 (1959). 19. Govindjee, R., and Rabinowitch, E., Biophys. J., 1, 377 (1961). THE MECHANISM OF PIIOTC SYNTHESIS Eugene Rabinowitch I. Photosynthesis is a tripartite process, as indicated schematically in figure 1 — one reaction set dealing with the conversion of CO^ to organic matter, one with the liberation of oxygen from water, and one with the conversion of light into chemical energy. CO2 X (-0.4V) T {CH20) Chl HcO -0; ZH(+0.8V) Fig. 1 The first two sets are "dark," enzymatic processes, occur- ring at an approximately constant level of energy, the third an energy-accumulating photochemical process. This latter must be an oxidation-rcduction --either hydrogen atom transfer, or electron transfer, or a combination of both. This transfer goes from a donor, ZH (an intermediate in the i 112 113 Eugene Pabinowitch lower en/ymatic reaction chain), with a high oxidation po- tential (approximately +0.8 V) , to an acceptor (an inter- mediate in the upper enzymatic reaction chain) with a high reduction potential (approximately -0.4 V). It thus leads to the storage of about 1.2 eV of chemical energy per elec- tron (or hydrogen atom) transferred. Since the reduction of carbon dioxide to the carbohydrate level requires the trans- fer of four hydrogen atoms, the total storage in the reduc- tion of one carbon dioxide molecule is 4.8 eV, or about 110 kcal. per mole. Energy storage in a form other than oxida- tion-reduction energy can play only an auxiliary role in photosynthesis; this applies, in particular, to the forma- tion of high energy phosphate (ATP). 2. Experiments by Emerson (and others) have established that in normal photosynthesis, about 8 light quanta are needed to transfer 4 hydrogen atoms (or electrons) "from ZTT to X" (in figure l). As suggested already in 1947, this could be achieved in two ways: either by eight parallel one-electron transfers, followed by four dismutations ; or by two consec- utive sets of four transfers each. Various recent studies support the second alternative. They suggest that the first transfer leads from ZTI to an intermediate acceptor roughly halfway between XR and Z--in other words, with an oxidation potential of the order of +0.2 Vr-and the second from this to the ultimate acceptor, X. The finding of cytochromes in chl oroplasts , due to T?obin Hill and coworkers, led to speculations on the role of these catalytic proteins in photosynthesis. At first, they have been assigned--by Arnon, among others--to positions on one or the other end of the photochemical sequence. This re- quired, however, ascribing to them an extreme (positive or negative) potential, not characteristic of known cyto- chromes. Much more plausible is Hill's and Bendel's recent suggestion that the chloroplast cytochromes serve as inter- mediates between the two photochemical reactions. Of the two cytochromes found by Hill in chl oroplasts , one belongs to the group of cytochromes bi, ("cytochrome bf,")> and has a n.ormal potential of approximately 0.0 V, while the other, designated as "cytochrome i_" (of cytochrome c_ type), has a potential of +0.37 V. It has been suggested that they serve in series, as indicated in figure 2. 114 Eugene Rabinowitch -0.4 V -0.1 + 0.2 + 0.5 +08 X/XH (O.OV)Cyt bg hi/1 hi/2 Z/HZ Cyt f (+Q37V) System 2 Fig. 2 Interestingly enough, two other oxidation-reduction systems have been identified recently in chloroplasts, one ( plastoquinone) having a potential close to that of cyto- chrome h^f the other ( plastocyanine ) one close to that of cytochrome f. Perhaps, they function as regulating systems, keeping the Fe3+/Fe''+ ratio in the cytochrome systems stable. Another possible function of these quinones could be to pro- vide transition from the "mono-valent" cytochromes to the "di-valent" redox systems at the beginning and end of the reaction sequence. The first oxidation-reduction step, which transfers the hydrogen atom (or electron) from an oxidant, ZIT, with a potential of about +0,8 V, to cytochrome _bg with a potential close to zero, stores about 0,8 eV (or 18,5 kcal, per mole) of the energy of the quantum, (The latter, in the case of chlorophyll a, contains between 35 and 43 kcal, per mole). The photochemically reduced cytochrome t^ then reduces, in a 115 Eugene Rabinowitch nonphotochemical reaction, the cytochrome f^, with a potential of +0.37 V, thus liberating about 8 kcal. of stored energy. This energy does not need to be lost; it seems plausible that, similarly to the electron transfer between cytochromes in the respiration chain, a part of it (about 7 kcal. per electron pair) can be salvaged in a molecule of high-energy phosphate, ATP. The second photochemical reaction transfers the electron from cytochrome f_ to the acceptor, X, again storing about 0.75 eV of potential energy. 3. Schemes similar to figure 2 — presented vertically, hor- izontally, in zig-zags, on circles or curlicues — have been proposed by several authors in recent years. They have arrived at these schemes in different ways; the following appears to me as the logical sequence of findings leading to it : a. Realization that in photosynthesis, four Il-atoms (or electrons) have to be transferred "uphill", over a gap of 1.2 V. b. Finding that quanta are used in this transfer (Emerson) . c. Finding of two cytochromes in chloroplasts, with potentials suggesting their function as "halfway stations" in the transfer of electrons. ( nil 1 ) . d. Observations of the oxidation of cytochrome in chlor- oplasts by light absorbed in one "pigment system," and reduction by light absorbed in another "system" ( Duysens) . e. Observation of an enhancement of the photosynthet ic action of light absorbed in "system 1," by light absorbed in "system 2" (Hmerson). The last-named line of evidence appears to me particular- ly convincing. This matter will be presented in more detail in Govindjee's paper later in this symposium. It is difficult to show that excitation of system 2 is by itself insufficient to bring about complete 116 Eugene Rabinowitch photosynthesis, since photosynthesis is known to proceed, at high rate, in red algae, in 540 in^i-light, absorbed mostly by phycoerythrin. Two possible explanations may be given. In the first place, light absorption in system 2 could lead to resonance transfer of energy to system 1, thus achieving the required co-excitation of the two systems ("spilling over"). Secondly, the presence of some phycoerythrin in system 1 (suggested by Duysens) and the contribution of chlorophyll a to the absorption at 540 m;i, may be sufficient to achieve co-excitation of the two systems even without energy transfer. If the latter is true, the suggestion, pre- sented further below, that the two systems may be contained in separate layers, would become more plausible. Experiments by S. Brody, and by W^. Butler and coworkers suggest that the fluorescence band of chlorophyll a i^ vivo, lies, at ordinary temperatures, at 680 m)i and belongs to chlorophyll a 670. It is thus associated with pigment system 2. (At the liquid air temperature, however, fluor- escence bands appear also beyond 700 nui — and must be attri- buted to chlorophyll a in system l). The primary distinc- tion between the two systems probably consists in their different location in the chloroplasts, implying association with different cellular components. Since chlorophyll mole- cules are known to fluoresce (in solution) only when assoc- iated with polar molecules, and to be non-fluorescent in dry hydrocarbons, one can suggest that pigment system 1 is associated with hydrophobic organic molecules, and system 2 with hydrophilic proteins. It is known that chloroplasts consist of alternating lamellae of more hydrophilic and more hydrophobic nature. It is tempting to associate the fluorescent pigment system 2 with the hydrophilic, and the non-fluorescent pigment system 1 with the hydrophobic layers in the chloroplasts. One could, for example, postulate a bimolecular sheet of pig- ments, with one leaf turned tov/ards the hydrophilic, aqueous layer, and the other towards the hydrophobic, lipoid layer; the two leafs could be separated by an enzymatic layer, containing e. ^. , the two cytochromes. Another alternative is to postulate a more uniform dis- tribution of pigments in the two layers. The first altern- ative is supported, however, by Goedheer's optical findings, which suggest that chlorophyll in the chloroplasts forms separate lamellae less than 10 A thick, (The proteidic and 117 Eugene Rabinowitch the lipoid laj'^ers having a thickness of the order of 30-50 A). Another problem is the relationship between the layered structure, as observed on chloroplast sect ions , and the "cobblestone" appearance of separate lamellae when vicv/ed from above. In the interpretation of the structure of mit- ochondria, the layered structure has been emphasized at some times, and the presence of, more or less spherical, en- zymatic units, at other times. Perhaps, the lamellae in chloroplasts , similarly to those in mitochondria, also bear --or even themselves consist of — spherical units, perhaps identical with the "photosynthet ic units," whose existence was first deduced from the experiments by Emerson and Arnold on photosynthesis in flashing light, and since supported by many other observations. Thomas observed that for chloro- plast particles to be effective in the Kill reaction, they have to contain at least two layers--an observation that fits into the above-derived picture. Pigment System 2 Pigment System 1 Fig. 3 118 Eugene Uabinowitch Figure 3 represents a two-layer photochemical system. V/ithin each layer, resonance transfer must take place both between identical pigment molecules, (leading to energy migration through the layer, at least within the confines of a single photosynthet ic unit), and between pigments ab- sorbing at the shorter wavelengths and those absorbing at the longer wavelengths. The energy quanta absorbed in each unit may be thus conveyed to a single enzymatic site. The chlorophyll molecules immediately associated with this site may be different in the two layers; their absorp- tion band may be located at 700-710 m^ in the hydrophobic system 1, and nearer 670 mfi in the hydrophilic system 2. The question arises: How is it possible for the photoxida- tive process in one layer to be effectively coupled with a photoreduct ive process in the opposing layer, preventing the loss of energy by back reactions within the layer? Is the existence of photosynthet ic units in both layers, with their enzymatic centers in juxtaposition, sufficient to insure effective correlation, evidenced by utilization of almost all light quanta absorbed, (at least, in weak light)? In the light of Menke's results, one is tempted to equate the photosynthet ic units in the hydrophilic layer with his "crystallites," and consider the possibility of chlorophyll molecules in the hydrophobic layer not being associated in units at all. However, bringing quanta absorbed by such scattered molecules into action at the proper reaction site, would be very difficult. The effectiveness of coupling of events in two layers de- pends on the time interval allowed between the two photo- chemical reactions. The observations, by Myers and French, of the Emerson enhancement in flashing light suggest that the intermediates involved may be long-lived enough to per- mit effective coordination of the two processes, even in weak light. (Otherwise, the light curves, showing the rate of photosynthesis as function of light intensity, would be quadratic in weak light and only gradually approach linear- ity)! ,. Another question is that of the mechanism of function of the sensitizing pigments. Do they themselves serve as re- versible oxidation-reduction catalysts, (or more exactly, "photocatalysts" ) in the electron (or hydrogen) transfer chain? The search for evidence of reversible oxidation- reduction reactions of chlorophyll in the sensitization of 119 Eugene Pabinowitch photosynthesis in vivo remains inconclusive. We know from the work of Krasnovsky, Evstigneev and coworkers, that, in vitro , chlorophyll sensitizes energy-storing oxidation- reductions, (such as that of riboflavin by ascorbic acid), by itself undergoing reversible reduction. Measurement of difference spectra by Coleman (in our lab- oratory), and by Kok at RIAS, suggested that in sufficiently intense light, difference bands appear in the region of chlorophyll absorption, in the red as well as in the blue, indicative of reversible bleaching of chlorophyll, beginning when saturation of photosynthesis sets in. However, neg- ative difference bands appearing in the neighborhood of 680-690 mp, could be due to changes in fluorescence. (Fluorescence caused by the modulated measuring light beam could be made more intense by strong "background" of non- modulated actinic light; it is known that the yield of fluorescence in vivo is about doubled when cells pass from the "light-limited" into the "light-saturated" state). Coleman thought that he had disposed of these objections by re-routing the light beam under a right angle to its orig- inal direction, and decided that not more than 10'^ of the difference band he observed at 680 m;i could be due to fluor- escence effects. More recently, however, Mr. Rubinstein in our laboratory found--by measurements in polarized light-- that a much larger proportion, (perhaps, the whole) of the negative difference bands observed at 680 mji, were in fact due to fluorescence changes. Independently, similar con- clusions were reached by Krasnovsky and coworkers. However, certain true changes in absorption spectra in the chlorophyll a^ absorption region, do occur; they include a "negative" band at 648 mp, and a "positive" one at 658 mu. These bands were first observed by Strehler, and also noted by Coleman. Recently, they were studied much more precisely by Rubinstein in our laboratory. Their occurence suggests that some reversible photochemical transformations do occur in the chlorophyll system during photosynthesis; but whether these are in the nature of oxidation-reductions (such as would be expected if chlorophylls were to serve as a wayside station in the transfer of electrons or TT-atoms in photo- synthesis) remains uncertain. Kok found a difference band at about 705 mji, which he attributed to reversible photo- chemical transformation of a long-wave component of the chlorophyll system directly associated with an enzymatic center and serving as a final "sink" in the energy trans- 120 Eugene Rabinowitch fer chain. In this case, too, the possibility of this change being in the nature of oxidation-reduction remains uncertain, Witt considered certain difference bands he ob- served in the short-wave region of the spectrum as indica- tive of reversible oxidation and reduction of chlorophyll, but this interpretation, too, remains hypothetical. The old problem of chemical (photocatalytical) versus purely "physical" sensitizing action of chlorophyll in photosynthesis, thus remains an elusive one. 5. Two types of fluorescence measurements need to be inter- preted for the two-layer pigment picture to be convincing. One is the above-mentioned finding, by Franck and coworkers, (as well as by Wassink and others), that the fluorescence yield of photosynthesis doubles (from about 1.5^ to about 3^) when photosynthesis becomes light-saturated. In our picture, chlorophyll is supposed to fluoresce in the hydrophilic layer, where it sensitizes the oxidation of water, and to be non-fluorescent in the hydrophobic layer, where it sensitizes the reduction of the organic substrate, X. Additional ad_ hoc hypotheses are needed to explain in this picture why the fluorescence yield should double at light saturation of photosynthesis. (Pranck sees in this finding the evidence that of the two photochemical steps, one is brought about by fluorescent, and one by metastable chlorophyll molecules). The other finding is due to S. Brody, who found in our laboratory that the lifetime of chlorophyll fluorescence in living cells is slightly less than one half of that in sol- ution — while the steady fluorescence yield is at least 10 times smaller (3^ vs. 30^), Brody' s value was subsequently confirmed by more precise measurements of Tomita and Murty, They support the presence j_n vivo of two forms of chloro- phyll, but suggest that the fluorescent form accounts for only 20^ of the total, and the non-fluorescent for 80^, How- ever, Brody' s results were obtained by means of very brief flashes of illumination; the average light intensity was very low, and the cells could be presumed to remain in the dark-adapted state. Lifetime measurements should be repeat- ed with cells illuminated with a background light, suffix ciently strong to maintain steady photosynthesis. 121 Eugene Rabinowitch Cbviously, more studies are needed before it could be said that the fluorescence data confirm (or contradict) the two-layer scheme, suggested in this paper, 6. The lamellar structure of the photosynthet ic apparatus in general — and the two-leaf structure discussed here in part- icular--may favor effective storage of light energy in two ways: (l) by permitting resonance energy migration within the layers, and thus facilitating the utilization of quanta absorbed in a large number of chlorophyll molecules, for chemical transformations in a — by geometrical necessity, much smaller--number of enzymatic sites; and (2) by causing spatial separation of the oxidation and reduction products in two different layers, thus making immediate back reac- tions unlikely. (For example, the organic reduction pro- ducts may be given off into a hydrophobic, lipoid layer, while the photoperoxides, and ultimately, oxygen, are evolved into the hydrophilic, proteidic layer). In this connection, 1 would like to mention one experiment. On my suggestion. Dr. Mathai carried out the oxidation-reduction reaction between ferrous iron and thionine dye in a two- phase system, (ether + water). This reaction is in some respect analogous to photosynthesis, or even more, to Kras- novsky's reaction between chlorophyll and ascorbic acid; it stores energy equivalent to a potential difference of about 0.3 V, by using visible light energy, absorbed in a dye molecule. When illumination stops, the reaction is reversed and the color of the dye returns. Vhen this reaction is carried out in a suspension of ether droplets in water, the leucothionine is extracted into the ether, while the ferric ions stay in the aqueous phase; the two phases could be sep- arated in a funnel, and both remain colorless. But v/hen alcohol is added, permitting the two phases to mix, the de- layed back reaction takes place, and the mixture immediately becomes dark purple. This simple experiment demonstrates the effectiveness of phase separation of products for chem- ical storage of light energy in an oxidation-reduction system. FLUORESCENCE IN TWO-PIGMENT SYSTEMS J. L. Rosenberg and Tevfik Bigat We report here on some of our recent experimental observations of fluorescence in algae and leaves, with particiilar reference to the interactions between the two pigment systems and between the two photochemical reactions of photosynthesis. A ixnified model for energy collection and for the photochemical sequence of photosynthesis will be assiomed, applicable to green algae, phycobilin- containing algae, and green leaves. RED ALGAE Experimental Fluorescence observations of Porphyridium cruentum were carried out under conditions of chromatic transients occiirring when the wave-length of illumination is changed. In addition steady-state intensities of fluorescence were observed under differential excitation of the chlorophyll a and of the phycobilins. Days ens ' publication of a similar study appeared while our own manuscript was in preparatlon(l ) . We therefore need not give great experimental detail, but mention partic- ularly those points of difference between oixr work and his. We will use his notation wherever possible: The density of the cell suspension was such as to allow 25- 50fo transmission at 680 m\i in the 1 mm optical path of the fluorescence cell. The sample could be illuminated with two beams entering the cell along the same path, a blue beam at k-k^ m|j, absorbed principally by chlorophyll a and a green beam peaking at 535 m^i absorbed principally by phycoerythrin. These beams served both as actinic and fluorescence excitation beams . Fluorescence was observed at 4-5° to the actinic beam through a 679 mp. interference filter appropriately supplemented to eliminate all scattering and phycobilin fluorescence. For approximately equal incident intensities of the two beams, the steady-state fluorescence d-uring simultaneous irradiation was 122 123 J. L. Rosenberg and Tevfik Bigat not only less than the sum of the fluorescences with separate excitations but was less than the fluorescence with the green beam only. In a typical experiment, for which the green beam intensity was 97 aJ^d. the blue 90, in units of 10"11 einstein/cmS sec, the relative steady-state fluorescence intensities for green, blue, and green plus blue excitations were I30, 25, and 100 respectively. The attainment of these steady values in changing from one type of illumination to another was achieved by way of interesting transients. Fig. 1 summarizes the trans- ients accompanying addition or removal of blue excitation to a sample under continued green irradiation. The transient SO- TO - t 60 50 U w "2 40- 30 - 20- ro- -BLUEi-GREEN- -GREEN- gStc. Time Fig. 1. Chlorophyll a fluorescence of Forphyridium showing transients due to shifting from green to green plus blue illumination and the reverse . The regime of illumination was not changed \antil the steady-state of fluorescence was reached for each type of illumination. Incident intensity for the green beam was 97 a^d. for the blue, 90 (in 10~11 einstein/cm^-sec ) . accompanying the shift from blue to green excitation is shown in Fig. 2. Here the initial instantaneous rise in fluorescence is followed by a slower sigmoid rise to a new steady-state at a rate increasing with Increasing green light intensity. ¥e failed to observe the maximum found by Duysens for Forphyridium but not found by him to a large extent in other species . We 124 J. L. Rosenberg and Tevfik Bigat believe that Ms raaxim-um might be a normal induction outb\arst, and that its occurrence might require a longer elapsed period at low photosynthetic activity (dark or blue light) before switching to green than was allowed in ovor experiments . TIME (seconds) Fig. 2. Chlorophyll a fluorescence of Porphyridium showing the chromatic transient accompanying the shift from blue excitation (l = 128) to green. Intensity of green^ in imits of 10"11 einstein/cm^-sec: A, I = l^Sj B, I = 79; C, I = ko. Discussion (1) The above results, like those of Duysens with red algae and of Govindjee et al. with Chlorella( 2 ) , may be interpreted on the basis of two pigment systems. System 1, containing the biolk of the chlorophyll a in the red algae and the long-wave absorbing component in the gredn algae, gives very little chlorophyll a fluorescence; while System 2, containing the phycobilins in the red algae and the bulk of the chlorophyll a in the green algae, is capable of a strong chlorophyll a fluorescence. To accommodate the experimental results we start with a model different from that of Duysens in one very important respect. He postulates that System 1 is req.uired for the sensitization of photochemical Step (l) and that System 2 can sensitize only photochemical Step (2). The substrate for Step (2), Q is 125 J. L. Rosenberg and Tevfik Bigat an oxidized direct or indirect product of Step (l) and is reduced photo chemically "by the fluorescent state of System 2. We base our interpretation on the model of Franck and Rosenberg, ^ 5j in ■which the non- fluorescent System 1 efficiently sensitizes only Step (l), as in Duysens ' scheme^ but in which System 2 may sensi- tize either Step (l) or Step (2), Step (l) by way of a meta- stable state and accompanying fluorescence and Step (2) by way of the excited singlet state without fluorescence. With this dif- ference we propose the following correlations with experiment: (a) The action spectrxim determined by Duysens for photo- synthesis against a background of strong green light, \1) similar to the action spectriom for cytochrome oxidation in the presence of DCMQ, (^/ shows equal effectiveness of chlorophyll a and phycoerythrin . We assign this as the action spectrum for Step (l), not System 1, and expect it to be a combined action spectrxom for System 1 plus System 2. (b) System 2 is capable of efficient photosynthesis by itself., This correlates with the observed proportionality between the action spectra for photosynthesis and for chlorophyll a fluor- escence. (5 j Although good absolute quantxam yield data may not exist showing a minimum quantum requirement of 8 for red algal photosynthesis, there certainly is good evidence with Chi or ell a and with brown and blue algae that the minimum quantum require- ment may be achieved with monochromatic light over a wide wave- length region. (d) If the same distribution of function between two pigment systems exists in all the algae, this fact could be explained only if one of the systems may sensitize both photo- chemical steps. (c) A long-wave absorbing pigment, like P7OO, cannot act as a sink for singlet excitation energy during normal photosynthesis lander illumination absorbed by System 2, even when System 2 is being used for Step (l). If we ass-ume that P7OO removes excita- tion in connection with Step (l), then the observed chlorophyll a fluorescence would have to arise by competition with Step (2). This is in contradiction to the finding in our experiments and in those of others that the fluorescence from System 2 is depressed particularly when this system is working with high efficiency in sensitizing Step (2), namely^ when simultaneous blue irradiation provides adequate Q substrate. An attempt to retain the assump- tion of trapping by P7OO by allowing a small fluorescence leak during the flow of excitation from System 2 to P7OO would not work, because a small fluorescence leak of chlorophyll a is always accompanied by a many- fold greater leak into the 126 J. L. RosenlDerg azid Tevfik Bigat metastable state of chlorophyll a, and the combined energy losses for fluorescence and metastable state formation would be too excessive to allow for good quantum yields. (d) Existing fluorescence data need not he interpreted to exclude the possibility that Q^ the substrate for Step (2) is. . oxidized cytochrome f. This exclusion has been made by Buys ens '^ and in somewhat different language by Kautsky et al.(7) on the ground that the fluorescence is low after a dark period when cytochrome f is known to be reduced. It must be emphasized that oxidized Q is not the only quencher of chlorophyll a fluorescence. It is well known that after a dark period metabolic oxidants substitute for the normal primary photosynthetic oxidants during the beginning of the induction period,^ ■' Even on steady irradia- tion at intensities below compensation this type of substitution occurs and is accompanied by lower than normal fluorescence yields. (9) The lowering of fluorescence under such conditions can be explained in terms of a postulated adsorption of these substitute oxidants at the chlorophyll and their consequent ability to use chlorophyll a singlet excitation directly in a variant of Step (l ), thus quenching fluorescence . ^-'-^'' (e) Since the above model allows System 2 to sensitize both photochemical steps, it is necessary for reasons of geometrical economy about the bifunctional reaction center that there not be too many different photochemical substrates that are part of the built-in photosynthetic apparatus. For this reason, as well as for others discussed elsewhere in this Symposium, ^ ^'' we propose that Q, the normal quenching substrate for Step (2), is oxidized cytochrome f and that the same cytochrome in the reduced form is a substrate for Step (l). (f) In the absence of substitute reactions discussed in (d) the chlorophyll a fluorescence yield in the light-limiting region depends mainly on the ratio, r, of the rate of use of System 2 for Step (l) to its use for Step (2). The shape of the transients shown in Figs. 1 and 2 can be accounted for very well in terms of the adjustment of the value of r to the illumination regime, with the detailed model of Franck and Rosenberg predic- ting that r may range from a value close to zero, immediately after a period of System 1 light absorption and oxidized cyto- chrome accumulation, to a value of unity during steady-state absorption by System 2 alone. ^^j 127 J. L. Rosenberg and Tevfik Bigat GREEN PLANTS We report here two experiments with green plants which bear on the problem of interaction between the two photochemical systems. By a procedxare described in detail elsewhere^ we determined the fluorescence spectriim at the one-second peak of the induction outb"urst following a dark period relative to the spectrum at steady-state irradiation. Fig. 3 summarizes some typical results, o I s > o « a: 6- 660 680 700 720 740 Wove - length in mp. 760 Fig. 3- Relative fluorescence spectrum of a fresh Forsyi:hia leaf at the induction maximum. The leaf was illuminated in air with blue-green light, I = 800 x 10--'--'- einstein/cm^-sec, following a 10-minute dark period. R^ is intensity of emission at wave-length \ at the steady-state, reached after about 20 seconds of irradiation. The abscissa, measuring the ratio of R^y^ to R^YQ (^si'^g 679 W- a.s a reference wave-length), would follow the dashed horizontal line If the induction outburst had the same spectrum as the steady- state fluorescence. The two curves' are for different samples . v-'-'^'' The c\arves of Fig. 3 show that there is relatively more increase in 679 n^M' fluorescence (the main chlorophyll a band) d\aring the outbxorst than in the longer wave-length component. The same was found for many kinds of samples, both leaves and 128 J, L. Rosenterg and Tevfik Bigat algae. These results are similar to those reported ^y, v Lavorel for Chi or el l a hy a completely different method/11-' and are consistent with the finding of Butler that there vas practically no increase in the ability of light at wave-lengtJiB. above TOO m^ to excite fluorescence at the induction outburstr: '' The simplest interpretation of these facts is that the far- red fluorescing pigment^ corresponding to P7OO in absorption, can participate in photochemistry only indirectly or only by way of its metastable state. If this pigment belongs to System 1 and acts as an energy sink for other pigments within System 1, there would be a unique assignment of the photo- sensitizing role for Step (l) to a metastable state, whether System (l) or System (2) absorption acts lead to this photo- chemical act. Fig. k shows the results of some typical fluorescence transients in leaves accompanying the reduction of incident Intensity from a value above saturation to a lower value. These experiments were a confirmation, and extension of a phenomenon reported by Franck et al.^^^). The different courses of Curves A and B in Fig.~5 suggests a difference in mechanism of satioration for the aerated and C02-d.epleted samples. In the aerated leaf, curve A of Fig. k, the major cause of saturation is probably located in the oxygen- liberating reactions following Step (2). Under such condi- tions oxidized cytochrome accumulates but is not able to undergo Step (2) as rapidly as quanta are supplied to the reaction centers. When the incident intensity is lowered, a region of greater quanta! utilization is reached. Efficient photochemistry will start with Step (2) to reduce the backlog of oxidized cytochrome. The fluorescence yield will thus start low at the reduced Intensity and will slowly rise as the excess oxidized cytochrome is reduced and the two photochemical steps again come into phase with each other . In the COg- deprived sample. Curve B of Fig. h, the major cause of saturation is the limitation in supply of the nat-ural photosynthetic oxidant for Step (l). In such a case, prolonged irradiation at high light leaves the cytochrome principally in the reduced state. On a sudden reduction in light intensity a more efficient quantal utilization will be felt primarily in Step (l), ajid the fluorescence will remain above the final steady- state value until the two photochemical 129 J. L. Rosenterg and Tevfik Bigat c 35 B e — — e o* 1 .d^ 25 15 - 4 6 Time (Steondt) Fig. k. Fluorescence at 680 m^i of a bean leaf following the transition from high light to low light irradiation. Incident intensity in "blue-green prior to zero time^ in units of 10"^ einstein/cm^-sec: A, I = 336, B, I = 250. Intensity was reduced to one-eighth initial intensity at zero time. A: leaf in moist air; B, leaf in COo-free air , (107 steps come into phase with each other. The above explanation may not include all factors responsible for the transients in Fig. h but does provide a basis for understanding the dif- ference between the two cases studied. ACKNOWLEDGMENTS This work was supported by the Office of Naval Research, Department of the Navy, under contract W.30k-kl6 with the University of Pittsb\argh. We thank Dr. Mary B. Allen of the Kaiser Foundation Research Institute for supplying the Poirphyridium strain -used in this work. 130 J. L. Rosenberg and Tevfik Bigat REFERENCES 1. Duysens, L. N. M., Proc. Roy. Soc, BI57, 3OI (1965). 2. Govindjee, Ichimura, S., Ceder strand, C, and Rabinovitch, E., Arch. Biochem. Biophys., 89, 522 (196O). 3. Franck, J., and Rosenberg, J. L., This volume. k. Duysens, L. N. M., and Amesz, J., Biochim. Biophys. Acta, 6k, 2k^ (1962). 5. Duysens, L. N. M., Thesis, Univ. of Utrecht (l952). 6. Emerson, R., and Lewis, C, Am. J. Botany, _50, 165 (19^3); other references in Chapter 30 of Rabinowitch, E., "Photosynthesis," Vol. II, Part 1, Interscience, New York (1951)- 7. Kautsky, H., Appel, W., and Amann, H., Biochem. Z., 332, 277 (i960). 8. Franck, J,, in "Handbuch der Pflanzenphysiologie, " Vol. V, p. 689, Springer, Berlin (196O). 9. Latimer, P., Bannister, T. T., and Rabinowitch, E., Science, 12|+, 585 (1956). 10. Rosenberg, J. L., Bigat, T., and DeJaegere, S., Biochim. Biophys . Acta, in press . 11. Lavorel, J., Biochim. Biophys. Acta, 60, 5IO (1962). 12. Butler, W. L., Biochim. Biophys. Acta, 9^, 309 (1962). 13. Franck, J., French, C. S., and Puck, T. T., J. Phys . Chem., i+5, 1268 (19^1). RELATIONSHIP BETWEEN LIGHT INDUCED EPR SIGNAL AND PIGMENT P700 Helmut Beinert and Bessel Kok In two previous communications (1,2) we have reported ob- servations on the narrow (--^8 gauss) EPR signal induced by light in photosynthetic materials. (This signal is also known as signal I (Ref. 3) or signal II (Ref. 4) and rapidly decaying or "R" signal (Ref. $)• Ti^e first paper (l) de- scribed properties of the signal observed in a preparation of the red alga strain TX 27 that was largely deprived of phyco- bilin and enriched in "P700" by partial extraction of the chlorophyll using 72% acetone (6). The signal appeared on illumination at room or liquid nitrogen temperature or after chemical oxidation by means of ferricyanide . Double integra- tion of the derivative signal indicated a concentration of spins nearly identical to the concentration of P700 as deter- mined by difference spectroscopy assuming a molar extinction of 10~5 M~ xcm~ and complete bleaching in the photoact. A second extraction of the algal preparation, now with S0% in- stead of 72% acetone, removed the EPR signal as well as the difference spectroscopic signal at 700 mu. We concluded that these results could be most readily explained if the photo- oxidized form of P700 was responsible for the narrow, fast decaying EPR signal in photosynthesis. The second paper (2) reported observations with whole cells of the blue-green alga Anacystis at room temperature. The relative strength of the narrow EPR signal was measured as a function of intensity of either one or both of two wave- lengths: 630 mu absorbed by phycocyanin and quite effective in provoking photosynthesis and 710 mu absorbed by (long wave) chlorophyll and rather ineffective in provoking photo- synthesis. The EPR signal showed the same behavior as the oxidized form of P700 which was observed earlier spectro- scopically {?): Long wave light proved much more effective in provoking the EPR signal than short wavelengths, or a combination of the two lights. These data therefore also indicated the possible identity of the EPR signal with oxi- dized P7OO, presumably the oxidized moiety generated in the long wave photoreaction of photosynthesis. Since its reduc- tion requires the reduced moiety generated in the short wave photoact, P700 tends to accumulate in the absence of the short wave light. 131 132 Helmut Beinert and Bessel Kok Weaver and Bishop (8) recently failed to observe the nar- row EPR signal in a Scenedesmus mutant (#8). Spectroscopic examination of this alga indeed did not reveal a light in- duced bleaching of P700 (9). Spectroscopic determination of P700 revealed a maximum ab- sorbancy change at 700 mu of 1 unit per 300-400 units of total chlorophyll absorbancy (at^ — ^675 mu). Its soret band (at hj>2 mu) as well as its nearly identical solubility in organic solvents indicate P700 is a chlorophyll _a molecule. We assume that a special binding site causes a long wave change of its absorption band and underlies its function as a photoconverter . This assumption assigns to P700 a molar extinction coefficient similar to that of chlorophyll a ( -^10^ in vivo ) . A second assumption: that complete bleaching occurs in the photoact yields a ratio of one trapping center per 300- 400 sensitizing chlorophyll molecules — in good agreement with measurements of the "photosynthetic unit" (10,11). If the oxidized form of P700, P700*, which is formed in the long wavelength photoact, were identical with the free radical species observed by EPR, the quantitative relation- ship, between the number of spins represented by the EPR signal and the amount of P700 detected optically, which we observed in a preliminary experiment (1), should hold rather generally for photosynthetic materials and furthermore, the kinetics of appearance and disappearance of the EPR signal should match that of the typical absorption band of P700, In the present work we undertook a direct approach at quantitation of the number of spins represented by the light induced EPR signal and the amount of chlorophyll and P700 present in a variety of materials. Knowledge of any consist- ent quantitative relationships would obviously be of interest, even if the EPR signal were not due to P700'^ itself or a closely associated electron carrier or trap. Vife were aware during the course of this work that quanti- tation of EPR signals is beset with many difficulties, that we had to make certain assumptions, which are not readily amenable to experimental verification, and that a definitive identification of the component responsible for the EPR sig- nal could not be expected from our approach. The EPR data were evaluated on the assumption that the light induced free 133 Helmut Beinert and Bessel Kok radical species are fully detectable by EPR and that no interactions interfere with this detectability . The concentration of unpaired electrons was measured by EPR spectroscopy at ambient and low (-50" to -70") tempera- ture as will be described in detail elsewhere (12). For the experiments at room temperature a double cavity was used, which held both sample and standard in matched flat cells. A benzene solution of diphenylpicrylhydrazyl (DPPH) was used as a steindard at room temperature and a pitch sample in KCl and nitrosyldisulfonate in KOH at low temperature. The standards as well as the integration procedure used were cross checked. Conditions were chosen that saturation of the EPR signals with microwave power did not occur or was suf- ficiently small that it could be corrected. In order to ensure saturation of the pigment suspension witn light, i.e. maximal signal development, the signal amplitude of serial dilutions of these suspensions was measured until a linear relationship between signal amplitude and concentration was observed. The double integration of the derivative signals, which resulted in the quantitative estimate of unpaired spins, was based on signals obtained in this linear region. Values were also observed in the dark and after addition of ferri- cyanide. The value obtained with ferricyanide can in many cases serve as a control for light saturation and maximal signal development, as the concentration of unpaired elec- trons produced by an excess of this oxidant is either very similar to that obtained on illumination under saturating conditions or larger. Only materials were selected for study in which overlapping signals (broad signal, 20 gauss, slow decaying or "S" signal, and Mn (II) signal) were absent or small. Overlap, when occurring, was corrected. Lumn Results obtained under satisfactory experimental condi- tions are summarized in the table. The last vertical col\ gives the calculated ratio of unpaired spins per P700, Most values cluster around a ratio of 2, although the low tempera- ture experiments on the TX 27 preparations and the experi- ments on Anacystis yielded higher ratios. We have no explanation for the high values obtained at low temperature. Although a different standard (pitch) was used at low temper- ature than at room temperature, a comparison of both stand- ards gave excellent agreement. It is likely that the high values obtained for whole Anacystis and the same preparation after sound treatment are due to overlap of the narrow EPR signal with the broad light induced EPR signal in such 134 Helmut Beinert and Bessel Kok complex preparations. Although overlap was corrected on the basis of the dark signals observed after illumination, this correction is not entirely satisfactory as the broad signal is more intense during illumination. In the last two lines determinations on chromatophores from Rhodospirillum rubrum are reported. Since P700 is not a constituent of these organisms only the ratio of bacterio chlorophyll [determined according to (13)3 to unpaired spins is given. Approximately 3% of this chlorophyll are thought to represent a photocon- verter, P89O, similar to P7OO. On this basis a spin per P89O ratio of 0.4 to 0.5 would be obtained. We are aware that our experiments cannot provide a final decision or an identification of the narrow light induced EPR signal; they could at best rule out or make appear plausible certain possible interpretations. In assessing the signifi- cance of the values we obtained, two principal considerations are pertinent: The first is concerned with the accuracy of our quantitation procedures and the second with the question of whether all radicals and radical species formed in the illuminated samples were, in fact, detected by EPR. To the first point we can say that use of the double sam- ple cavity and carefully matched cells, the use of independ- ently standardized standards, attention to the conditions of saturation with light and microwave power and the consistency of the results obtained in the determinations at both room and low temperature, make it very unlikely that gross errors were committed. Nevertheless, in view of the uncertainties in the absolute values of EPR standards and in comparison of different materials, we think that accumulation of errors could have led to values which are in error by a factor of 2 or 3- The consistancy of the results indicates that these errors, if incurred, are not random but systematic and due to certain incorrect assumptions. The second point of concern is related to the question as to what type of paramagnetic species is in fact responsible for the observed signal. The simplest assumption, on which our experiments here are based, is that a single free radical species arises, which has a structure and environment such that it can be quantitatively detected by the EPR technique. However, since a one electron oxidation produces the radical, it appears possible that a second radical, formed by the corresponding one electron reduction (in the extreme case a free electron) is simultaneously generated. There are no 135 Helmut Beinert and Bessel Kok indications we know of from the behavior of the observed EPR signal that it may represent two different species. This would still not exclude such a possibility, for which one could see some support in the high ratios of free spins produced per molecule of P700 (cf. Table l). A serious objection to the interpretation of our experi- mental results could arise from the possibility that we might not be detecting more than a fraction of the radicals actu- ally produced. Several explanations could be given for this. We may be dealing with "lifetime" broadening, i.e., short relaxation time of the unpaired electron; an exchange inter- action may broaden the line; or, in case a free electron were generated, it could be trapped at non-equivalent sites and therefore, experience varying local magnetic influences, which would lead to line broadening. Such arguments cannot at present, be refuted on experimental grounds. If they were valid our values would set the lower limit of radical concen- tration. We may then conclude from our data in the light of these considerations that the number of unpaired electrons induced by light in the photosynthetic material studied is either closely similar or bigger, certainly not smaller than the amount of P700 present. It is of interest to note that other components of the photosynthetic system in plants, have been reported to occur at a concentration of the same range as P700 and the light induced EPR signal studied here. However, the metal constituents of two of these: cytochrome f and plastocyanin are certainly not responsible for this signal. ACKNOWLEDGEMENTS We are indebted* to Dr. J. Heise and Mr. R. W. Treharne of the Kettering Laboratories, Yellow Springs, Ohio, for their kind collaboration in the experiments with the double sample cavity and to Mrs. I. Harris for technical assistance. This work was supported by the USPHS through research grants (GM-05073 and GM-O6762) and a research career program award (GM-K6-I8, kk2) from the Division of General Medical Sciences to H. Beinert. 136 Helmut Beinert and Bessel Kok Table 1 Concentrationa of Chlorophyll, P700 and Detectable Light- Induced Free Radicals in Various Photosynthetic Materials Material 1 Temp. °C Chlorophyll 10 "^M P 700 10 ""^M Spins 10 ""^M Ratio: Chlorophyll per spins Ratio: Spins per ^70n TX 27 broken, washed 25 -5? 72 2.5 38 1.3 67 h.3 108 58 1.8 3.3 TX 27, washed acetone extracted 25 -53 50 2.0 65 2.6 110 9.0 ^5 22 1.7 3.5 Anacystls whole cells 25 -53 120 6 ko 2 107 11 55 2.7 5.5 Anacystls broken, washed 25 -53 52 2.7 I'' 0.9 65 3.8 80 71 3.8 k.2 Chloroplasts acetone extracted -72 2k 35 79 30 2.3 Chloroplasts fresh aged acetone extracted -70 -70 -70 100 72 16 25 18 23 51 38 58 196 190 28 2.0 2.1 2.6 R.rubrum 25 -53 210 5.3 330 6.3 6k 137 Helmut Beinert and Bessel Kok REFERENCES 1. Beinert, H. ; Kok, B, and Hoch, G., BBRC, 2' ^09 (1962). 2. Kok, B., and Beinert, H., BBRC, 9, 3^9 (1962). 3. Commoner, B., in "Light and Life", W. D. McElroy and B, Glass, eds., p. 356. Johns Hopkins Press, Baltimore (1961). k. Allen, M. B. ; Piette, L. R., and Murdiro, J. C, BBA, 60, 539 (19o2). 5. Weaver, E. C, Arch. Bioch. Bioph., 92, 19^ (I962). 6. Kok, B., Bioch. Bioph. Acta, _48, 32? (1961). 7. Kok, B., and Gott, W. , Plant Physiol. 33, 802 (I96O). 8. iVeaver, E. C, and Bishop, N. I., Science, 1^0 , 1093 (1963). 9. Kok, B., This symposium. 10. Emerson, R., and Arnold, Vi/., J. Gen. Physiol., 13., 391 (1932). 11. Kok, B., Bioch. Bioph. Acta, _22, 399 (1956). 12. Beinert, H. and Kok, B., In preparation. 13. Cohen-Bazire, G. ; Sistrora, W. R. and Stanier, R. Y., J. Cellular Comp. Physiol., 49, Z3 (1957). EPR AJTO OPTICAL STUDIES ON SCENEDESMUS MUTMTS Ellen C. Weaver and Norman I. Bishop There is more than one way to obtain information on single e- lectron transfers as they are taking place in an illuminated pho- tosynthetic system. One which we have been using for some time is electron paramagnetic resonance (EPR) spectroscopy(l) . There are two light induced resonances which are typical for the chloro- plasts of higher plants and several species of algae. They can "be differentiated on the basis of g- value, line shape, and the ki- netics of their formation and decay. The most prominent one forms and decays in less than a second; hence the designation R, for "rapidly decaying". It has a g-value of 2.0025, is about eight gauss wide, is unstructured and Gaussian in shape. It has been shown to be dependent on the presence of chlorophyllU;2) . The other resonance forms rather quickly, but persists in the absence of illumination for periods up to an hour or more; hence the des- ignation S, for "slowly decaying". It has a somewhat higher g- value, 2.0046, is twenty gauss wide, and displays partially re- solved hyperfine structure. It has been tentatively identified with the semiquinone of plastoquinonev 3) . There are two classes of mutants which have provided further- evidence on the role of the two EPR signals in photosynthesis^.^) . Thesepossess all the readily identifiable wild type pigments in normal quantities, and yet are unable to photosynthesize. Those in one class are termed "oxygen" mutants, because, although they are able to photoreduce carbon dioxide with hydrogen, have a greatly reduced quinone-Hill reaction. All of these have a typi- cal R signal, but display only a trace of the persistant S sig- nal. This observation, together with that on manganese-deficient cultures, which also lack both Hill reactivity and the S signal'^ 3 j and photosynthetic bacteria which evolve no oxygen and also lack the S signal, enables one to identify the broad, structured sig- nal with the ability of the system to evolve oxygen in photosyn- thesis, although other interpretations have been proposed^ -' . They all possess the normal distribution of plastoquinones; some other essential link in the electron transport chain, not direct- ly observable with EPR spectroscopy, is missing. 138 139 Ellen C. Weaver eind Norman I. Bishop The second class are termed "CO2" mutants because they have a greatly reduced ability to photoreduce carbon dioxide with hydro- gen, although the quinone-Hill reaction is relatively intact. One of these, Mutant 8, shows no R signal at all, even though it has abundant chlorophyll (^). Dr. Warren Butler's investigations on this mutant (5) indicate that it lacks P7OO, but information on this point for the other "COg" mutants is lacking. Fig. 1 illus- trates the behavior of the cultures under consideration. LIGHT OFF S-SIGNflL LIGHT ON "COJ' MUTANT ^^'V^ ,AkU. S- SIGNAL ONLY "Oj' MUTANT tW^V*"**'^**''*' R- SIGNAL ONLY ,.» M^'^V, '^*'^ WILD TYPE 'H- f S-SIGNAL VV \JV ■A**^ g- VALUE H — ► T V I s a R SIGNALS ^V*4MVr'^ Fig. 1 Comparison of the EPR spectra in mutant and wild type Scenedesmus . Due to its long decay time, the S signal can be seen in the absence of illumination, whereas the R signal decays within seconds, g-value may be thought of as the EPR ana- logue of wavelength as used in optical spectroscopy. All spectra are made with suspensions containing approximately 5 x 10" cells per milliliter and with identical instrumental parameters . 140 Ellen C. Weaver and Norman I. Bishop Another "CO2" mutant^ No. l8^ has a trace of R signal when ex- posed to bright light, hut this may reflect the somewhat higher overall rate of photosynthesis of which it is capable. However, several other "CO2" mutants have an R signal of approximately wild type proportions. Although a system capable of carrying out phot- oreduction can produce an R signal, the converse is evidently not true; and R signal is not a reliable indicator of photoreduction. We were fortunate, therefore ^ to be able to compare the behavi- or of the mutants by optical means with the results already on hand. The difference spectrophotometer in Prof. Melvin Calvin's laboratory is somewhat similar to the one described by Dr. Bessel Kok(6) in that it makes use of repetitive flashes of actinic light. A spinning disc with a sector removed provides alternating light and dark periods. The absorption is measured at a given wave- length by a photomultiplier immediately after the flash, and again just before the flash; the difference in the absorption (light minus dark) is plotted against wavelength as the monochromator is slowly driven. This particular machine is described in detail in a forthcoming paperVn. The spectra displayed here utilize a flash of 3 msec followed by a dark period of 30 msec; thus, only reversible changes with time constants which fall within these limits are detectable. WILD TYPE SCENEDESMUS 3 msec FLASH 30 msec DARK -'^Jj«Jv^*■)!^*W/«^ — ''■■'^'■./^fMJ^- _1 I L. 500 540 Fig. 2 Changes in absorption induced by 3 msec flashes in a sus- pension of Scenedesmus . The absorption was measured k msec after the onset of the flash, and again at the end of the 30 msec dark period. A tungsten lamp was used with a Corning 2030 filter passing only the wavelengths between GhQ m^ and 75O mfj,. The curve seen here is the first minus the second measurement, (light minus dark). The base line is the absorption just before the flash. 141 Ellen C. Weaver and Norman I. Bishop The difference spectrum of wild type Scenedesmus , agrees in its major features with other published difference spectra for this organism^ -' . In contrast is the spectrum of Mutant 8, a "CO2" niu- tant, which displays virtually no spectral shifts; yet Mutant l8^ in all other respects quite similar to Mutant 8, has a difference spectrum which resembles wild type (Fig. 3). Among the "O2" mu- tants the same situation holds true: some (e.g. No. 11) have no spectral shifts, while others (e.g. No. ko) look very much like wild type. There is no obvious correlation between either of the EPR signals and any of the prominent positive or negative shifts in the region we were able to scan. +0.5 10 o " o o -0.5- — 1 1 1 NO. 8 "CO2" MUTANT 1 r -I 1 1 r ■H*^HH^f»^M^^ Fig. 3 Changes in absorption determined as in Fig. 2. The upper trace is that of No. 8, a "CO2" mutant, and displays al- most no changes. No. 11, an "O2" mutant, is similar in its lack of signal. The lower trace is of another "O2" mutant, No. I8; the "O2" mutant No. kO resembles it. Fortunately, it was also possible to examine the time course of an absorption change at any one wavelength by using the machine in the following way: a neon flash bulb was substituted for the chop- per, allowing the time sequence of light and dark to be completely controlled. With the very short time constants employed, sensi- tivity had to be sacrificed. This difficulty was circumvented by use of a coirputer of average transients (CAT); this device 're- 142 Ellen C. Weaver and Norman I. Bishop members' the signal from each of a large number of repetitions of each light and dark sequence. Noise, being random, is averaged out, while any consistent signal, however small, is huilt up to recognizable proportions. We devoted most of our attention to the shift at 520 m^a; the ^+78 m|j. shift seems to be related in a direct way to it in that it behaves in a similar way with varying conditions. Fig. Ua is a trace of the rise and decay of this shift in wild type cells; both curves are approximately exponential with the 0.1 second flash employed here. The whole cycle, light plus dark, is two seconds in duration. Fig. 4b at once tells us why we were seeing no shifts with the 3 msec flash in Mutant 11; although the posi- tive shift is rapid, the decay is very much slowed down. If the absorption change is not reversible, the repetitive flash device will detect no change. For Number 8, (Fig. kc) both rise and de- cay of the absorp)tion shift at 525 W- are slow. However, both LIGHT ON Fig, k Time course of the absorption change at 525 W- produced by a neon flash in wild type and mutant Scenedesmus . Each curve is the summation of 100 repetitions of the light-dark cycle, made with the aid of a CAT (see text). 143 Ellen C. Weaver and Norman I. Bishop this "CO2" mutant and the one most like it, Mutant 18, are capable of large shifts in absorption at 525 mjj, if the light period is long enough. Figure 5 demonstrates that the absorption in No. 8 is still increasing; even at the end of a flash 0.4 second long. LIGHT OFF "COj MUTANT NO. 8 -0.4 sec FLASH 50 RUNS 525 LIGHT OFF, ^O.isec FLASH ^\^ 100 RUNS ttlGHT ON Fig. 5 Comparison of the effect of a 0.1 second flash in a total period of 2 seconds, which produces very little change in the 525 mp. absorption, with a O.k second flash (period ^i- seconds) which produces a large change. The picture is quite different when wild type is subjected to flashes of 0,5 second in length. There is a two phase rise at 525 mji; first a very steep one, lasting just over 0.1 second, followed by a slower increase in absorption. Evidently at least two absorption changes, one considerably slower than the other, are taking place. The time course of the ^4-25 vcnx shift, however, remains exponential. If one observes the behavior of a represen- tative "O2" mutant. No. 11, with a 0.5 second flash, the time course is more conplicated (Fig. 6). There is a steep rise, followed by a decay which sets in while the light is still onj further decay takes place when the light is turned off. It is al- most as though some substance present in this mutant was decreas- ing in optical density at 525 ^^^> "but with a slower time course than the positive shift. Possibly this substance does not decay 144 Ellen C. Weaver and Norman I. Bishop o o in < UJ o LIGHT OFF NO.II- NO. II 525 m^ {50 RUNS) TOTAL PERIOD 4 SECONDS 0.5 sec LIGHT ON LIGHT OFF Fig. 6 Time course of alDsorption changes with a 0.5 second flash at 525 m|j, in wild type and an "O2" mutant (No. 11). The two phase rise in wild type is contrasted with the negative shift in the same preparation at 425 i^M- which remains approximately ex- ponential. Note the decay during the 0.5 second flash in No. 11. entirely in the dark period, thus accounting for the greater initial increase in absorption observed with the shorter flash (Fig. 7). The time course behavior of No. 8, wild type, and No. 11 com- pose a series of increasing complexity. These were suspended in water without added substrate. However, addition of benzoquinone to No. 8, while allowing it to evolve oxygen, does not perceptib- ly change its optical behavior. Nor does saturating No. 11 with hydrogen gas, which it can use for the photoreduction of carbon- ate, change the characteristics displayed here. However, it has been shown that adapting wild type Scenedesmus to hydrogen over a number of hours enables it to carry out anaerobic photoreduction. 145 Ellen C. Weaver and Norman I. Bishop HT OFF OXYGEN MUTANT NO. 1 1 T OFF 0.5 sec FLASH =^^ 525 m/i TOTAL PERIOD 4 SECONDS I I I I I 1 1 0.5 sec - Fig. 7 Absorption changes in Mutant 11 with flashes of different duration. Each curve is the s'uramation of 50 repetitions of the k second light-dark cycle. De-adaptation at high light Intensities is prevented by the ad- dition of 3(3,4-dichlorophenyl)-l-l-dimethylurea (DCMJ) to prevent the evolution of oxygen. Scenedesmus thus treated "behaves very much like untreated Mutant 11 in its capacity for photoreduc- tion(9); it was therefore not unexpected that the adapted wild type should display the 525 ni|a ahsorption shift with kinetics re- sembling those of Mutant 11, as can be seen in Fig. 8. We are still not able to equate any of the spectral shifts in the region we have been examining, i.e. that between 3^0 mji and 580 m|j,, with either of the typical free radical signals. The pre- sence or absence of characteristic EPR signals may in general be correlated with the photochemical behavior of the mutant; however, the rapid kinetics of the formation and decay of these signals in the mutants has not yet been studied. It is notable that all the major absorption changes appeared in each strain, whether or not metabolic activity was occurring. The kinetics of these changes were strikingly altered in the mutant cultures, and it is hoped that further study will provide a basis for explanation of these time course differences, and perhaps yet reveal underlying rela- tionships between observations by these two differing experimen- tal techniques. 146 Ellen C. Weaver and Norman I. Bishop tn < 111 (£ O WILD TYPE LIGHT ADAPTED TO H2 rN_ OFF + HCOj" / ^-^^--v-^ + I0"5 M DCMU / V 25 RUNS ^ixV 525 m/i 1 1 1 1 TOTAL PERIOD 4 SECONDS 1 1 1 1 1 ' Fig. 8 Absorption change in hydrogen adapted wild type Scenedesitius with 0,1 mg NaHCO-3 in 10 ml of cell suspension and 10-5 M DCMU. The time course is similar to that of the "O2" mutant^ No. 11. The physiological "behavior of these two prepara- tions has been shown to he similar^). Acknowledgments ; ¥e thank Prof. Melvin Calvin, in whose labora- tory a part of this work was done, and Mr. Irwin Kuntz, Jr., who designed and "built the difference spectrophotometer. He was an active participant in the experiments. Parts of this investigation were supported "by National Science Foundation Grant GB 97^, contract No. AT( ^0-1 ) -2687 between the United States Atomic Energy Commission and the Florida State University, and the Fels Fund. LITERATURE CITED (1) E. C. Weaver and H. E, Weaver, Photochem. and Photoblol . 2, 325 (1963) (2) G. M. Androes, M. F. Singleton and M. Calvin, Proc . Natl . Acad. Scl., U. S . ^8, 1022 (1962) (3) E. C. Weaver, Arch. Biochem. Biophys . 99, 193 (1962) {\) E. C. Weaver and N. I. Bishop, Science ll+O, 1095 (1963) (5) Warren Butler, Personal Communication (6) B. Kok, Plant Physiology 3^1-, l84 (1959) (7) I. D. Kuntz, Jr., P. Loach, and M. Calvin. Submitted to Biochem J . (8) Kok, B., Nature 179, 583 (l957) (9) N. I. Bishop, Nature 195, 55 (1962) A METHOD FOR CALCULATING QUANTUM YIELDS FOR THE FORMATION OF REACTION INTERMEDIATES Daniel Rubinstein The appearance or disappearance of absorption bands during illumination of photo synthetic organisms has been observed by many investigators beginning with Duysens (1). Whether or not these absorp- tions bands are due to compounds in the main path of photosynthesis could be best determined by quantum yield measurements. Photo- synthesis in monochromatic red light has been observed to have a quantum requirement of eight (±2) quanta per molecule of oxygen evolved ^^\ A much larger quantum requirement for the appearance of an absorption band would suggest that this band is not due to a compound in the mainstream of photosynthesis. Quantimi yields have usually been estimated by examining the rate of increase in absorbance, and dividing this rate by that of the light absorption. The increase in absorbance is then converted into increase in concentration (moles per liter) by assuming a plausible value for the extinction coefficient. In some cases, such as the appearance of a band at 420 mn in Porphvridium (3) or Chromatium (4) attributable to a cytochrome, the extinction coefficient is known from measurements in vitro; but for nearly all other difference bands, whose molecular origin is unknown, a guess must be made. In such cases, it is usually assumed that the extinction coefficient of the unknown compound is approximately the same as that of chlorophyll (or a similar organic pigment). Many plots of absorption changes as a function of exciting light intensity have been published (4-10). in general, these have been plotted to determine whether several absorption bands belong to the same pigment, or whether an absorption change at a certain wave- length should be attributed to more than one pigment(6) . If the plots for changes at different wavelengths are identical, they may be due to the same pigment. If a "light curve" (as these plots are designated) appears to have inflections, more than one pigment may be involved. Usually, no further information is sought from these plots. A more detailed analysis shows, however that additional information maybe deduced from them. Specifically, the shape of the light curve may 147 148 Daniel Riobinstein permit evaluation of the quantum yield without having to assume an arbitrary value for the extinction coefficient, or without having to know exactly the amount of energy absorbed by the sensitizing pigment. This evaluation is based on the following postulated reaction Chi ^ Chi* (1) scheme: t^t ^^1 2 1 Chl*+A > A + Chi kg (3) A^+ B > B''^+ A The first equation describes the absorption of light and of the pigment, Chi (which may be chlorophyll) to Chi*. The rate of formation of Chi* is equal to the number of absorbed quanta, K I. The rate constant, k , for the decay of Chi* includes all Chi* > Chi transitions, except the one leading to the formation of the absorption band belonging to the molecule A (eq. 2). In the second equation, it is postulated that the excited molecule, Chi*, reacts with A to form A^. In the third equation, A-'- decays, during both light and dark periods, by reacting with a substrate, B, assumed to be present in excess. The rate constant for the decay of A-*- (k3, eq. 3) can be taken directly from the time-trace after cessation of illumination. The steady concentration of Chi* is easily seen to be (4) KI o ,[Chl*] = ■ k^ + k^ [A] et rate of formation of A during i' dfA^l dt = k^ [Chi*] [A] -kgLAY (5) It is further postulated that the saturation of the absorption change observed at high light intensities, is due to a limited quantity of reactant A, available for reaction with Chi*. 149 Daniel Rubinstein It follows that 1, r A 1, [A] + [A^] = [A^]^^ =[AJ (6) By combining equations (5) and (6): ^' ki+k2([A^];;;,-[A^]) (7) The q\iantuin reqmrement is the number of quanta needed for the ^ appearance of an absorption band corresponding to a jingle molecule A . Eq. (7) predicts that the initial rate of formation of A ^^^ll be propor- tional to the absorbed light intensity. The ratio d [A ] / KI^ when [A"*"] > has been defined as the quantimi yield y by Olson and Chance (4) and others (3) (11). In the above equation AJAi \^l^\^, (8) 7 = dt = KI^ k-Tk^fAl] max A^ > The steady state expression following from (7) is I k k k (9) 0=13 + o Only absorption changes, not concentration changes are actually observed; concentrations can be converted to absorbances (optical densities by dividing them by e (since OD = e [A1] ). Rearranging the terms, we obtain: I k_ k^ k„ (10) o 3 13 OD - .K k2K(OD^^^-OD) 150 Daniel Rubinstein This is the equation for a straight line when Iq/OD is plotted against 1/ (OD ^ - OD) ' ^ max ' I, 0. D. K3 ry^^ K2 K£ jO^ ^ 1 ' — ^ O.D. max -O.D. Figure 1 The intercept on the abscissa and the ordinate are respectively: k^. and (11) (12) eK The intercept on the abscissa is related to the quantum requirement l/y by the following expression: 1 , . ^^ k, e (13) ( y - 1 ) OD max \ ' The knowledge of the absolute number of absorbed quanta, K I^ is not needed, to calculate the quantum requirement, since the constant, K, is absent from equation (13). The only terms needed for this determination are the change in OD reached at high light intensity. Calculation of the extinction coefficient from the intercept on the ordinate (fig. 1) requires, on the other hand, the knowledge of the number of absorbed light quanta, since the constant, K, remains in the expression (12). 151 Daniel Rubinstein Discussion In utilizing eq. (10) for the determination of quantum yields and extinction coefficients by extrapolation of the experimental curves to the two axes, accurate knowledge of several variables is required. A "close fit" of the individual points to the straight line is essential for the extrapolation to be meaningfxil. It is essential that the relative value of I be measured by means of a linear detector, such as a photomultiplier, or by utilizing a constant source of light with calibrated neutral density filters. In this determination of quantum yield, we assume a particular set of reactions in which ki, k2 and ks were assumed constant and independ- ent of I (eq. l-3y. Also, the conservation law was assumed to apply to A; that is A + A-*- was assumed to be constant. This idealized case may not always be realized, especially at high light intensities. This will be indicated by deviations from the straight line in our plot. For example, a positive deviation might mean that k^ increases with light intensity. Other deviations in the same direction may be caused by increases in the rate of eq. (3) due to a photochemical reaction supplying substrate for the back reaction. These and other deviations can often be revealed by other plots such as those of absorbance vs. time or temperature. Any intermediate A which is present in limiting concentration could be detected not only by the absorption of light, but by other methods (eg . electron spin resonance) and the same equation will apply, provided the effect reaches saturation by the same reaction mechanisms. It is to be noted that the plot in fig. 1 does not give the "total" number of absorbed quanta needed to produce one molecule of A^, but that of the quanta absorbed by the "active" pigment participating in eq. (1). The quantum yields calculated from the above plot may be used in the study of light reaction mechanisms in mixtures of "active" and "inactive" pigments, particularly when the "total" quantum yield can be obtained by conventional methods. A more detailed study of these relationships will be published later. Thanks are due to Professor Eugene Rabinowitch, Professor Gregorio Weber, Professor Bernard Abbott and Dr. Ashish Ghosh for valuable discussions. 152 Daniel Rubinstein References (1) Duysens, L. M. N. , Science, 120, 353 (1954). (2) Emerson, R. and Chalmers, R. Plant Physiol . 30, 504 (1955). (3) Duysens, L. M. N. , Proc. Third Intern. Congress on Photo- biology , Elsevier, 1960, pp. 135-142. (4) Olson, J. M. and Chance, B. , Arch. Biochem. Biophys. , 88 , 40 (1960). (5) Strehler, B. L. and Lynch, V. , Arch. Biochem., Biophys. , 70, 527 (1957). (6) Rabinowitch, E., J. Phys. Chem. , 66, 2557 (1962). (7) Clayton, R. K. , Photochem. Photobiol. , 1, 313 (1962). (8) Witt, H. T. and Moraw, R. , Z. Phys. Chem. , Neue Folge, 20, 253 (1959). (9) Chance, B. and Strehler, B., Plant Physiol . , 32, 536 (1957). (10) Kok, B. , Acta Botan. Neerl . , 6, 316 (1957). (11) Clayton, R. K. , Photochem. Photobiol. , 1, 305 (1962) LIGHT- INDUCED RAPID ABSORPTION CHANGES DURING PHOTOSYNTHESIS. II. 430 mp ABSORPTION CHANGES IN AGED CHLOROPLASTS IN THE PRESENCE OF PMS AND ASCORBATE Bacon Ke Chemical separation of two light reactions in photosynthesis and the re- constitution of the reaction system with artificial reagents have been amply demonstrated. Vernon and Zaugg ^^^ have shown that aged chloroplasts which had lost the capacity for oxygen evolution were able to photoreduce TPN when reduced DPIP was added. Other evidence for the chemical separation of two light reactions has been obtained from experiments on the relief of CMU inhibition of TPN reduction by ascorbate and redox dyes ^'^' and from experi- ments with mutants of Chlamydomonas reinhardi ^^i and Scenedesmus ^^K Some preliminary observations on the 430 m^ absorption changes in aged chloroplasts and the response of these absorption changes to ascorbate and PMS will be reported in this note. In aged chloroplasts the 515 nn^ absorption increase was either negligible or completely absent, indicating that the reac- tion associated with oxygen evolution was inactivated. Rapid absorption changes were studied with an instrument basically simi- lar to that reported by Witt (^). Construction details of the instrument used in the present work will be reported elsewhere (6). Polychromatic red (620 - 720 n^j) flashes with a duration of 2 x 10-5 gee were used for excitation. Aging of spinach chloroplast was done at 4° C for one week. The chloroplast suspension usually contained about 30 ;ig chlorophyll in a total volume of 3. ml 0. 1 M pH 7 phosphate buffer. Whereas fresh chloroplasts do not show any absorption change at 430 m;j, at least not under the experimental conditions and sensitivity used here, aged chloroplasts show a light- induced absorption decrease (Fig. 1, curve a). At the time resolution available, the rise time of the signal was estimated to be less than 10''* sec, and the half life of decay was 4 - 5 x 10' sec. The 430 m^ absorption change in aged chloroplasts can be completely abolished by 2 x 10"^ M ferricyanide, but is unaffected by ferrocyanide at a concentration as high as 10" '^ M. In a reducing medium of 2 x 10" ^ M ascor- bate, the absorption change was enhanced and the half life was shortened to 2 X 10" '^ sec (Fig. 1, curve b). The loss of 430 m>i absorption change caused by ferricyanide can be fully restored by adding ascorbate. Upon addition of PMS to a concentration of 3 x 10' ^ M to the aged 153 154 Bacon Ke FLASH .1 .2 .3 .4 TIME, SECONDS Fig. 1. Light- induced absorption change at 430 mp in: a. week- old chloroplasts. b. a + 2 X 10'^ M ascorbate. c. b + 3 X 10" 6 M PMS. chloroplasts already containing ascorbate, the absorption change was im- mediately converted to one composed of a rapidly decaying portion (i/> 10"^ sec) superimposed on a portion having the same decay time as before PMS addition (Fig. 1, curve c). By varying the measuring- beam wavelength between 400 and 450 mp, a difference spectrum was obtained as shown in Fig. 2. The spectrum shows a broad negative peak covering the 420-430 mja region and a smaller positive peak in the 405 mp region. Examination at 703 mp revealed an absorption decrease with kinetic be- havior very similar to that of the slow portion of the 430 m;j change. No rapidly decaying portion was observed in the 703 m;j absorption change. The composite absorption change shows an interesting dependency on the intensity of the excitation flashes (Fig. 3). Starting from a given saturating intensity (Imax^ ^^^ gradually decreasing it, the signal remained practically unchanged down to 50% intensity. Decreasing the intensity to below 50%, the rapidly decaying portion started to disappear. Between 30 and 15% intensity. 155 Bacon Ke only the slowly decaying portion remained Below 15% intensity, the slowly decaying portion also decreased. 400 410 420 430 440 450 WAVELENGTH. m>i Fig. 2. Difference spectrum of aged chloroplasts in the presence of 2 X 10-3 M ascorbate and 3 x lO'^ M PMS. ~ Fig. 3. Dependency of the composite peak on illumination intensity, o total peak height. • height of the slowly decaying portion. 156 Bacon Ke Because the 430 mp absorption change in aged chloroplasts was enhanced and its decay was accelerated by ascorbate, and because the absorption change was abolished by ferricyanide, it may be inferred that an oxidation reaction is responsible for the absorption decrease. Since the absorption de- crease occurred at both 430 and 703 m^j, and the slowly decaying portion of the 430 nryj absorption change and the 703 m^ absorption change have the same kinetics, it is reasonable to assume that these absorption changes are caused by the oxidation of chlorophyll, probably the far- red absorbing pigment P-700 ^^^ The breadth and peak position of the difference spectrum in Fig. 2 indi- cate that cytochrome oxidation, probably that of cytochrome f, may also be partly responsible for the 430 m>i absorption change. Thus, "the initial ab- sorption decrease may be interpreted as being due to photooxidation of chloro- phyll, part of which rapidly extracts electrons from cytochrome f in the pres- ence of ascorbate and PMS at the stated concentrations. The light intensity dependency of the composite curve suggests that the latter reaction occurs only when the light intensity exceeds a certain level. Cytochrome f then re- reduces rapidly in the presence of PMS and chlorophyll re-reduces~more slowly (by reduced cytochrome, PMS, ascorbate, or endogenous reductants). A similar reaction route has recently been proposed by Witt and co-workers for PMS concentrations less than 10"^ M, but no experimental details were given ^^'. Absorption decreases at 430 m^ with rapid decay times of ^ 10"* sec have previously been reported by Moraw and Witt ^^^ The so-called "type O" change has been attributed to the tt -tr* triplet state. The so-called "type 1" absorption decrease has been observed in many types of algae containing chlorophyll a and consequently described as due to the formation of a chloro- phyll a derivative. The "type 1" signal was observed only at high illumination intensities and no saturation could be reached even at extremely high inten- sity. Judging from these characteristics, it can be concluded that the 430 mp signal observed in the present work is not identical with either the "type O" or "type 1" 430 m^ signals reported. Furthermore, the difference spectra for the various types of absorption changes are entirely different. In CMU- treated chloroplasts with ascorbate and PMS at concentrations similar to those used here, Jagendorf and Margulies concluded from a high ATP/TPNH ratio that a cyclic electron flow must also have occurred in addition to photoreduction of TPN. It is not known whether a similar situation exists here, and if so, what effect it might have on the transient absorption change. However, experiments on the light-induced 430 mji absorption changes in the presence of ascorbate and PMS were conducted under both aerobic and anaerobic conditions, and practically identical results were ob- tained. Under anaerobic conditions PMS should exist exclusively in the re- duced state in the presence of excess ascorbate. More extended and detailed experiments on these absorption changes will be reported elsewhere. 157 Bacon Ke This is Contribution No. 135 from the Charles F, Kettering Research Laboratory. The author wishes to thank Miss Elena Ngo for technical assistance. REFERENCES 1. L. P. Vernon and W. S. Zaugg, J. Biol. Chem. , 235, 2728(1960). 2. A. T. Jagendorf and M. Margulies, Arch. Biochem. Biophys. , 90, 184 (1960). 3. R. P. Levine and R. M. Smillie, Proc. Nat. Acad. Sci. , 48, 417 (1962). 4. N. Bishop, Nature, 195, 55 (1962). 5. H. T. Witt, R. Moraw, and A. Miiller, Z. Phys. Chem., 20, 193(1959). 6. B. Ke, R. W. Treharne, and C. McKibben, Rev. Sci. Instruments (submitted). 7. B. Kok, Biochim. Biophys. Acta, 48, 527 (1961). 8. H. T. Witt, A. Miiller, and R. Rumberg, Nature, 197, 987 (1963). 9. R. Moraw and H. T. Witt, Z. Phys. Chem., 29, 17^(1961). II. ELECTRON TRANSPORT PATHS - BIOCHEMICAL INVESTIGATIONS THE ELECTRON TRANSPORT SYSTEM OF PHOTOSYNTHESIS DEDUCED FROM EXPERIMENTS WITH MUTANTS OF CHLAMYDOMONAS REINHARDI R. P. Levine INTRODUCTION Six years have elapsed since Emerson and his coworkers^ ' described two different types of light effects in green plant photosynthesis, and it has been only three years since Hill and Bendall suggested that the electron transport system of photosynthesis could be interpreted in terms of two distinct light-dependent reactions coupled by at least one light-independent reaction^^). Subsequently, there has been almost a surfeit of publications on the two light effects in photo- synthesis; merely a brief reference to the papers presented in this and other recent symposia is sufficient to emphasize the impact that Emerson's original contribution has had on contemporary research into the mechanism of photo- synthesis^"^"^). It is indeed noteworthy that at present there is a fair extent of uniformity among the several schemes proposed for the electron transport system of photo- synthesis. However, the general nature of the schemes, and the lack of suffi- cient data for their support, makes it virtually impossible to accept or deny any one of them. The uniformity of most of the popular schemes is a mixed blessing, for it suggests on the one hand that diverse experimental approaches are leading to a set of final and general conclusions. On the other hand, the specific and significant details by which these schemes differ suggest that there is much to learn before we have a complete understanding of the mechanism of photo- synthesis. The purpose of this contribution to the symposium is to present data con- cerning reactions in the electron transport system of photosynthesis of the unicellular green alga, Chlamydomonas reinhardi , as studied with the wild type strain and mutant strains that are unable to carry out normal photosynthesis. Furthermore, these data will be used as the basis of a model for a sequence of reactions in the electron transport system. Three years ago we became interested in the genetic control of photo- synthesis. The investigation, however, soon turned from one having an orienta- tion primarily of a genetic sort to one more directly concerned with the mecha- nism of photosynthesis per se . The work was begun with the aid of a mutant strain of C . reinhardi that was unable to fix carbon dioxide by photosynthesis^^). It was assumed that this, and other mutant strains isolated subsequently^ '' ^>, 158 159 R. P. Levine were unable to carry out normal photosynthesis because of gene mutation that either; 1) affected the synthesis or activity of enzymes responsible for the formation of components in the electron transport system of photo- synthesis; or, 2) affected the synthesis or activity of enzymes whose function lies in synthetic processes attending normal chloroplast development; or, 3) affected the synthesis or activity of enzymes responsible for the formation of components of the reductive pentose cycle. Clearly, the actual nature of the genetic block could be any one of the afore- mentioned and result in a cell's inability to carry out normal photosynthesis. Furthermore, it was realized that it might be difficult to distinguish, at least on a phenotypic basis, between certain classes of mutants in view of the intimate relationship between chloroplast structure and function. The results and conclusions presented here derive principally from bio-- chemical experiments utilizing chloroplast fragments and, as such, they may not necessarily reveal what happens in the intact cell or chloroplast. However, the model presented here is in substantial agreement with some of the models that have been proposed from studies with whole cells or whole chloroplasts''^"^^. METHODS The methods we have used for investigating the electron transport system of photosynthesis in C. reinhardi stem largely from the well established tech- niques used with higher plants. A description of the methods for assaying the activity of enzymes related to photosynthesis, as well as the techniques for in- vestigating photosynthetic reactions by whole cells and isolated chloroplast fragments of C. reinhardi have been published (see principally references 9-11). Details for the procedures used in isolating and measuring the quantity of dif- ferent components of the electron transport system of photosynthesis have also been described^ ^2, 13)_ ^^i^ methods for investigating the electron spin reso- nance signals in C. reinhardi have been described elsewhere' ^'^'. Photo- reductionUS) by whole cells following adaptation to hydrogen gas was measured manometrically as consumption of hydrogen in the light or as carbon dioxide fixation in the light from C^^.^abeled bicarbonate. The procedures, details of which will be published elsewhere^ "', are similar to those used for Scenedesmus( 1 "7). GENERAL DESCRIPTION OF THE MUTANT STRAINS The mutant strains of C. reinhardi used for the experiments reported here were derived from the wild type strain (No. 137c) by induction with 160 R. P. Levine ultra-violet light, followed by screening for their inability to fix carbon dioxide in the light^'^\ Unlike the wild type strain, each of the mutant strains will not grow in minimal medium unless it is supplemented with sodium acetate^^^). Four different mutant strains will be considered here; namely, ac-21, ac-115, ac-141, and ac-208 (the symbol ac refers to acetate dependence). With the exception of ac- 14r and ac- 208 , the mutants lie in different linkage groups. Although ac-14l"and ac- 208 "are linked they lie about 10 map units apart on opposite sides of the "centromere in linkage group III (see references 19 and 20 for details of the techniques of genetic analysis). In terms of current genetic theory, each mutation should be located in a different cistron or functional unit. That is, each mutation most likely represents a genetic alteration that affects the synthesis or activity of a different enzyme. These mutant strains, while unable to fix carbon dioxide by photosynthesis at the wild type rate, resemble wild type in two important characteristics. First, electron microscopy has revealed that ac-21, ac- 115 and ac- 141 have normal chloroplast structure^^l); ac- 208 has not been examined. Second, there are no major differences in chlorophyll content^^^'. There are, however, minor differences in carotenoid content, but these may be trivial in so far as their being causally related to the inability of the mutant strains to carry out normal photosynthesis^ 13). REACTIONS OF THE ELECTRON TRANSPORT SYSTEM OF PHOTOSYNTHESIS The rationale of the experiments with the mutant strains of C. reinhardi is simply one of inquiring into which known reactions of photosynthesis whole cells or isolated chloroplast fragments can or cannot perform, and then attempt- ing to reconstruct the sequence of reactions which best fits the observed results. Thus, we are following, by analogy, one of the classical procedures of bio- chemical genetics to determine the sequence of events as they occur in a partic- ular biosynthetic pathway. Except for recently obtained data pertaining to the mutant strain ac-208, and to both photoreduction and photophosphorylation as they relate to wild type and the four mutant strains, the results summarized here are discussed in a series of publications'"' °" ^ 1' '■^K Carbon dioxide fixation in whole cells by photosynthesis and photoreduction It has been shown that, in comparison to the wild type strain, whole cells of the mutant strains are highly deficient in their ability to fix carbon dioxide by photosynthesis^^). The maximum rate of fixation by a mutant strain (ac-21) was about two per cent of the wild type rate. 161 R. P. Levine Wild type cells fix carbon dioxide by photoreduction after being adapted to hydrogen in the dark for a period of 15 minutes. The maximuna rate of photoreduction was obtained at a light intensity of about 1500 lux. At intensities greater than 2000 lux the cells reverted to photosynthesis. However, photo- reduction was obtained at a light intensity of 10, 000 lux in the presence of 1 x 10"^M DCMU. At this concentration DCMU causes a 99 per cent inhibition of photosynthetic oxygen evolution and carbon dioxide fixation. At 1500 lux the mutant strains ac-115 and ac-141 gave rates of carbon dioxide fixation and hydrogen consumption comparable to the wild type rate, whereas the rates of ac-21 and ac- 208 were negligible. The addition of DCMU had little effect upon photoreduction by ac-115 and ac-141, and photoreduction was obtained at 10, 000 lux in the absence of DCMU. Both ac- 1 15 and ac- 141 resemble a mutant strain of Scenedesmus de- scribed by Bishop^^"^) that can carry out photoreduction but does not evolve oxygen by photosynthesis and has no Hill reaction with p-benzoquinone. The Hill reaction Further differences between the mutant strains were revealed in a study of the Hill reaction both by whole cells and chloroplast fragments^ \ As seen in Table I, ac-115 and ac-141 have no Hill reaction with a variety of Hill oxidants; ac-21 and ac-208 show Hill reaction activity, except that the latter does not have activity when the oxidant is potassium ferricyanide. Table I Hill reaction, TPN photoreduction, and photophosphorylation by wild type and mutant strains of C. reinhardi Hill reaction* TPN reduction* * Photophosphorylation***" Vitamin DPIP & Strain DPIP Cyt c FeCy a+ b PMS K3&FMN Ascorbate wild type 55.2 28.8 324 35.4 7.8 124 50 8 ac-21 54.6 6.6 114 9.0 ac-115 13. 8 112 30 ac-141 22.2 209 43 ac-208 53.2 14. 7 * fimoles Hill oxidant reduced/hr/mg chlorophyll ** ^imoles TPNH/hr/mg chlorophyll *** /Ltmoles Pi esterified/hr/mg chlorophyll + a signifies TPN reduction when the source of electrons is from water and b signified TPN reduction when the source of electrons is from DPIP and ascorbate 162 R. P. Levine TPM photoreduction An investigation of TPN photoreduction showed that all of the mutant strains possessed an active PPNR and pyridine nucleotide transhydrogenase' ^0^. However, as shown in column five of Table I, chloroplast fragments from each of the mutant strains were ineffective in the photoreduction of TPN in the presence of an excess of purified PPNR when the source of electrons was water(9' 10). On the other hand, all strains except ac- 208 were capable of TPN photoreduction when the electron donor was a catalytic amount of DPIP in the presence of a substrate concentration of ascorbate'*^' '^^K Photophosphorylation Recent investigations have revealed that wild type, ac- 115 , and ac- 141 are capable of carrying out cyclic photophosphorylation with PMS, or the combination of vitamin K3 and FMN, as the added electron carriers whereas ac-21 and ac-208 are not (22), Since ac-21 can photoreduce TPN from DPIP and ascorbate, it was considered important to determine whether or not TPN photoreduction could be accompanied by photophosphorylation in this strain. The wild type strain can carry out non-cyclic photophosphorylation linked to TPN reduction in this manner (Table 1). Numerous attempts with ac-21, how- ever, have failed to give non-cyclic photophosphorylation, though in each instance there was the expected rate of TPN reduction. Electron spin resonance Results of an investigation of electron spin resonance^ 1'*' have shown that wild type, ac-21 and ac-208 have the two ESR signals (I, the fast, narrow signal and II, the slow broad signal) that are characteristic of Chlorella, Chlamydomonas, and Scenedesmus'23-25), However, signal II is missing in ac-115 and ac-141. Signal I in C. reinhardi is light-dependent, and has an action spectrum that resembles the absorption spectrum of chlorophyll a. Signal II occurs in the absence of light, at least in cells that have been cultured in the light^^). COMPONENTS OF THE ELECTRON TRANSPORT SYSTEM OF PHOTOSYNTHESIS The inability of any one of the mutant strains to carry out normal photo- synthesis might be accounted for on the basis of some change in either the quantity or chemical nature of one of the possible components of the electron transport system. We have analyzed, in part, the cytochrome b5, cytochrome f, plastocyanin, plastoquinone, and carotenoid content of wild type, ac - 2 1 , ac- 1 15 , and ac-14l(9' ^2, 13, 26), Since we are still engaged in these analyses, the data presented here are incomplete. 163 R. P. Levine Table 11 summarizes data regarding cytochrome bg, cytochrome f, and plastocyanin^^^\ It can be seen that there is approximately three to four times as much cytochrome £in ac- 115 and ac- 141 as compared to wild type and ac-21. The cytochrome bg as" well as the plastocyanin content of the strains is about equal. Table II Cytochrome b , cytochrome f, and plastocyanin content of wild type and mutant strains of C. reinhardi Cytochrome content Plastocyanin content Strain Cytochrome bg Cytochrome f_ moles chlorophyll /mole cytochrome moles chlorophyll/g atom Cu wild type 90 363 500 ac-2j. 113 489 530 "^-TTS 154 174 429 ■^-141 100 167 560 An analysis of the carotenoid pigments in wild type, ac-2j_, ac- 115 . and ac-141 has revealed similarities among the mutant strains which distinguish Them from the wild type strain when they are cultured in the light^l'"^'. Each mutant strain has both a lower carotenoid content and a lower beta -carotene/ alpha-carotene ratio than light-grown wild type. Interestingly, both the lower total carotenoid content and the lower beta - carotene / alpha - carotene ratio are characteristic of wild type when it is cultured in the dark. However, in spite of this similarity between the mutant strains and dark-grown wild type, each mutant strain has a pattern of types and amounts of carotenoids that distin- guishes it from the other mutant strains and from both light- and dark-grown wild type. The initial investigation of plastoquinone in wild type, ac-2 1, ac- 1 15 , and ac-141 revealed that both ac- 1 15 and ac-_14J_ have five-fold less plastoquinone than"wild type and ac-2J_^^' ^^K This investigation preceded the important dis- covery of Crane and~his~ coworkers in which it was demonstrated that several different plasto- and tocopherylquinones could be extracted from spinach chloroplasts^27, 29)_ Accordingly, our original data for wild type and the mutant strains (Table III) were representative of a combination of quinones, and most likely some of the quinones were missing because of the extraction procedure used. We have recently undertaken a more extensive analysis of the quinones of C. reinhardi, and it has revealed a variety of both plasto- and tocopherylquinones similar to those found in spinach. 164 R. P. Levine Table III Plastoquinone content of wild type and mutant strains of C. reinhardi Strain Moles chlorophyll /mole plastoquinone wild type 15 ac-2j_ 30 ac-115 99 ac-141 "74 DISCUSSION Each of the four mutant strains of C. reinhardi under consideration is unable to carry out normal photosynthesis because some portion of the electron transport system does not function. This loss of function results from the mu- tation of either a structural or regulatory gene. The expression of this muta- tion is seen experimentally as a block in photosynthesis, and the term block will be used to describe the point or points at which the electron transport system is stopped. It is assumed that in each mutant strain the genetic change affects an enzyme that is at least indirectly concerned with the electron trans- port system. The term block, however, is not meant to imply a knowledge of the specific enzymes involved. Furthermore, a single mutation could result in a block at more than one point. If, for example, two components of the sys- tem are formed as part of the same biosynthetic pathway, a mutation that af- fects some common, early step in their biosynthesis could result in the loss of both components and, consequently, the system would be blocked at two different points. In addition, the loss of some single component might result not only in a block within the electron transport system but in the coupling of photophosphorylation to the system as well. The initial model proposed for the electron transport system of photo- synthesis in C. reinhardi<9/ was based upon the hypothesis of Hill and Bendall that there are" two different, light-dependent reactions coupled by at least one light-independent, exergonic reaction. According to this model, and using the terminology of Duysens, light absorbed by system II results in the oxidation of water coupled to the formation of a photoreductant; light absorbed by system I results in the formation of a photo-oxidant and the reduction of TPN. Overall, the photoreductant produced in system II is oxidized by the photo-oxidant pro- duced in system I. The model to be discussed here (Fig. 1) retains the general features of the one presented earlier^^)^ It has gained additional support from recent studies of the mutant strains of C. reinhardi . As further experiments are performed, and other mutant strains studied, the details of the model may change. This is, therefore, only a working model, but it most nearly 165 R. P. Levine PM5- TPN.^ PPNR.^ -4 Chli-^ Cyf-F-^ y^ — x-^ Q-4 ^ Ch,a,-< HoO (' ac-115 ac-141 \ Ascorbate Figure 1. A Model of the Electron Transport System of Photosynthesis in C. reinhardi as Deduced from an Investigation of Four Mutant Strains. (See text for explanation) accommodates the facts we have obtained so far. and we believe it contains the least number of assumptions. Quite clearly, the model is not unique, for it draws heavily upon findings of other investigators whose approach has been somewhat different from ours. Two light-dependent reactions and at least one light -independent reaction Evidence for the two different light-dependent reactions of C. reinhardi shown in the model comes from results of experiments with all four mutant strains. All of the data obtained for ac-115 and ac- Ul ^^' ^^' are consistent with the hypothesis that these two strains are blocked in a reaction associated with system IL in which a photoreductant is produced coupled with the oxidation of water. Chloroplast fragments obtained from these strains showed no Hill reaction activity and were unable to photoreduce TPN. It was predicted that if the block were only in system IL chloroplast fragments could carry out the photoreduction of TPN if a reductant were supplied implying that system 1 was able to function in these two strains. This prediction was borne out when it was shown that chloroplast fragments of ac- 115 and ac- 141 could photoreduce TPN from DPIP and. ascorbate. In this respect both strains resemble wild type which has been inhibited with DCMU or o-phenanthroline, for under these conditions wild type chloroplast fragments can photoreduce TPN in the presence of DPIP and ascorbate. Further confirmation of the ability of these two mutant strains to carry out part of the photosynthetic electron transport was obtained when it was es- tablished that both were able to do cyclic photophosphorylation with PMS and to fix carbon dioxide by photoreduction. 166 R. P. Levine It was also demonstrated that both ac- 115 and ac- 141 had retained the sharp, fast ESR signal but that the slow, broad signal was missing. The former signal has been attributed to a long wave length form of chlorophyll a such as P- 700^25, 30)^ whereas the latter signal may be associated with chlorophyll b, for this signal is absent in a chlorophyll b-less mutant of Chlorella^23)_ j^ the case of ac-115 and ac- 141 it is tempting to correlate the absence of this signal with the inability of these mutant strains to carry out a reaction in system II. In contrast to both ac- 115 and ac- 141 , ac- 208 has Hill reaction activity with all of the Hill oxidants tested except ferricyanide, and yet it cannot photo- reduce TPN from DPIP and ascorbate. These results suggest that the block in ac-208 lies at a side in the system subsequent to the point of entry of electrons TFom DPIP and ascorbate. The block could lie in system I. However, there is no direct evidence for this, and ac- 208 has both ESR signals. The two light-dependent reactions occur in ac-21. However, they must be coupled by at least one light-independent reaction which is blocked in this mutant strain. A photoreductant is produced by system II in ac-21 as evidenced by the fact that there is Hill reaction activity. However, this photoreductant apparently cannot be utilized, for DPIP and ascorbate must be provided in order to obtain TPN photoreduction. These results can be best explained by assuming that a block lies in a light-independent reaction between systems I and II. This explanation is supported by the observation that both ESR signals are generated in cells of ac-21. Thus, inasmuch as the two different ESR signals may reflect systems I and IL the mutant strain is identical to wild type. Both ac-21 and ac-208 pose some interesting questions regarding the site of action of ferricyanide into the electron transport system of C^. reinhardi . Witt, Miiller, and Rumberg^"^^) suggest that ferricyanide belongs to a group of oxidants, termed ox Sj, whose reduction is associated with the oxidation of Chlj. Accordingly, any block in the electron transport system of C. reinhardi lying at a site after system II should result in the absence of a Hill reaction with ferricyanide. Thus, since ac-21 and ac- 208 are blocked after system II, neither of them should have a Hill reaction with ferricyanide. This is contrary to our observations, for ac-21 does have Hill reaction activity with this oxidant, albeit at a rate that is about three times lower than that of wild type. It is con- ceivable that the Hill reaction with ferricyanide could proceed via system II alone in ac-21. If this assumption is made, however, it would be expected that ac-208 would also give a Hill reaction with ferricyanide. Of course, the dilemma with ac-208 could be avoided by making the second assumption that there are two blocks in ac-208; namely, that there is a block in electron transport be- tween the poirvTof entry of electrons from DPIP and the reduction of TPN, and that ac-208 lacks a component of the electron transport system unique to the ferricyanide Hill reaction. These ad hoc assumptions, however, should not be considered significant in the absence of experimental evidence. If the Hill reaction with ferricyanide in ac-21 proceeds from system II alone then its action spectrum might be dif- ferent from that obtained with wild type. We have recently begun, in collaboration 167 R. P. Levine with G. Gingras, measurements of the action spectrum of the ferricyanide Hill reaction by chloroplast fragments of both wild type and ££-^- These measure- ments, obtained with the aid of an oxygen electrode, have revealed so far that the action spectrum for wild type is very similar to the action spectrum for photosynthetic oxygen evolution by whole cells. Photophosphorylation Additional important information for the design of the model was obtained from an investigation of photophosphorylation by wild type and the four mutant strains. Photophosphorylation has been studied in wild type in several different ways(ll). Cyclic photophosphorylation was obtained with either PMS or the combination of vitamin K3 and FMN as electron carriers. Cyclic photophos- phorylation with PPNR^-^^) has not been tested. Non-cyclic photophosphoryla- tion coupled to the photoreduction of TPN was found to occur with either water or DPIP and ascorbate as the reductant. However, non-cyclic photophos- phorylation coupled to ferricyanide reduction could not be demonstrated. The ratio of one ATP produced per two electrons transferred during non- cyclic photophosphorylation in C. reinhardi suggests that there is only one site for non-cyclic photophosphorylation. Furthermore, in agreement with Losada, Whatley, and Arnon^"^'^), non-cyclic phosphorylation lies at a point after the entry of electrons from DPIP and ascorbate into the electron transport system. In the model for electron transport in C. reinhardi under consideration here, the site for photophosphorylation has been placed tentatively between X (a component of the electron transport system and the assumed point of entry of electrons from DPIP and ascorbate) and cytochrome f. A photophosphoryla- tion at this site would, accordingly, be coupled to the oxidation of a reduced X and the reduction of an oxidized cytochrome f. Recently, Forti, Bertole, and Parisi^'^'*) have shown that photophosphorylation in spinach chloroplasts can be coupled to the reduction of cytochrome f and that the stoichiometry is one ATP produced per two electrons transferred. It has also been shown that cyto- chrome f can be oxidized by system l(35-37)_ Further, in Anacystis nidulans (35) cytochrome f can be reduced by system II. Cyclic photophosphorylation with either PMS, or the combination of vitamin K3 and FMN, was obtained only with the mutant strains ac- 1 15 and ac- 14I . This indicates that the cyclic pathway enters the electron transport system after the block in these two strains and, accordingly, does not depend upon a photoreductant produced in system II. The fact that neither ac-2_l^ nor ac-208 gave cyclic photophosphorylation supports the contention that these strains are blocked at sites after the point of entry of the cyclic pathway into the electron transport system. 168 R. P. Levine Since ac-21 can photoreduce TPK fronn DPIP and ascorbate it was assumed thaTT like wild type, it would also carry out photophosphorylation with DPIP and ascorbate; thus, it would be possible to localize the block in this mutant strain to a position before the site of entry of electrons from DPIP and ascorbate. As mentioned earlier it has thus far been impossible to obtain photophosphorylation in this manner with ac-2J. even though TPN is reduced. There are several ways one may interpret the results obtained with ac - 2 1 . First, it might be assumed that there is a block between X and cytochrome f, and in order to obtain TPN photoreduction from DPIP and ascorbate, electrons from these donors must enter the system at a point after X and the site of phosphorylation. This might be at cytochrome £or at Chlj. Second, the block might be in the formation of X itself; it may be either lacking, deficient, or inactive. Once again, in order to explain TPN photoreduction from DPIP and ascorbate, the electrons would have to enter at some alternate site. Third, X might be essential for both electron transport and for the coupling of phos- phorylation to the electron transport system. Consequently, if X were lacking, deficient, or inactive in ac-22^ both electron transport and phosphorylation would be blocked. P hotoreduction and cyclic photophosphorylation A mechanism exists in C. reinhardi for the production of ATP independ- ently from the oxygen evolving" pathway of photosynthesis. Wild type cells can fix carbon dioxide by photoreduction in the presence of a concentration of DCMU that almost completely inhibits both photosynthetic oxygen evolution and carbon dioxide fixation. Photoreduction also occurs in cells of ac- 115 and ac- 141 where the oxygen evolving pathway is blocked as a consequence of mutation. Both of these mutant strains were also found to carry out cyclic photophosphorylation with PMS, thus confirming the finding of several investigators that photophos- phorylation with PMS is independent of the oxygen evolving pathway. These results suggest that ATP is generated during photoreduction by a process of cyclic photophosphorylation, and photoreduction in C. reinhardi , therefore, may be similar to bacterial photosynthesis. That is, a hydrogenase acts to reduce TPN by a light-independent reaction and ATP is produced via cyclic phosphorylation. However, the possibility that the hydrogenase may provide electrons for the production of ATP by a non-cyclic photophosphoryla- tion cannot be excluded. Cyt ochrome f and quinones Among the possible components of the electron transport system in C. reinhardi, the quinones and cytochrome £are the only ones for which marked quantitative differences have been foundn2). xhe low level of plastoquinone in both ac-115 and ac- 141 is of particular interest. The plasto- and tocopheryl- quinolies have be'eii implicated in photosynthesis because they are localized in 169 R. P. Levine chloroplasts(38) and undergo reduction in chloroplasts in the light' •^^"^^'. Fur- thermore, it has been shown that they are essential for the Hill reaction^ 2'' 41-45)^ the photoreduction of TPlsK^^), and for cyclic photophosphorylation(47, 48), Thus, it appears that different plasto- and tocopherylquinones may function at different sites in the electron transport system. The low plastoquinone content of both ac- 1 15 and ac- 141 is not sufficient to account for the complete absence of a Hill reaction in these strains, at least on the basis of the results obtained by Bishop^^^' for sugar beet chloroplasts. However, it is possible that these two mutant strains lack or are deficient in a specific plasto- or tocopherylquinone that functions at a site associated with system II and the oxygen evolving portion of the system, whereas the quinones that remain act at some different site or sites. The cytochrome f content of both ac- 1 15 and ac- 14n ^2) ^g about three to four times greater than that of wild type and ac-21. Coupled with this increase there is at least a doubling of the rate of TPN reduction from DPIP and ascorbate (Table I), and also in cytochrome photo-oxidase activity( 10). These observa- tions are consistent with the idea of Hill and Bendall'^) that cytochrome £is in- volved with system I, and with the observations of the light -dependent oxidation of cytochrome f in algae by system l(35-37). CONCLUSIONS Though our model for the electron transport system of photosynthesis in C. reinhardi lacks complete documentation, it most easily accommodates the observations that have been made with the wild type and four mutant strains. All of the data presented are consistent with a model for electron transport in which there are two light-dependent reactions separated by at least one light- independent reaction. The data suggest that there is a single site for non- cyclic photophosphorylation, aAd that cyclic photophosphorylation can occur independently of the oxygen evolving portion of the system. In addition, one or more plasto- or tocopherylquinones may play an integral role in the electron transport system. The model presented here has the advantage of providing several predic- tions that can be tested experimentally. For example, if ac- 2J_ lacks compo- nent X (Fig. 1), then it should be possible to detect the oxidation of cytochrome £ by system I but not its reduction by system II. Further, both ac- 115 and ac- 141 may show the oxidation of cytochrome f in the light followed by its reduction in the dark(35). in contrast, if the block in ac-208 lies in system I the mutant strain would be expected to show the reduction of cytochrome £by system II but not its oxidation by system I. In addition to testing predictions such as these, the use of the mutant strains provides an opportunity to search for the function of different possible components of the electron transport system, for as mentioned at the outset of this discussion, mutations may have occurred at gene loci that affect the 170 R. P. Levine biosynthesis of some of these components. The establishment of the loss or alteration of any one component in a given strain, taken in conjunction with what we have learned about the strain's loss of photosynthetic reactions, can provide strong evidence for the function of such a component. Clearly, we have merely commenced an extensive series of investigations, for we have by no meams exhausted all of the opportunities for different kinds of experiments with the four mutant strains, and we have yet to investigate in any detail some 15 additional mutant strains. ACKNOWLEDGMENTS The preparation of this paper was greatly facilitated by discussions with Dr. George K. Russell, and the research reported here with C. reinhardi has been the collaborative effort of several people. I wish to acknowledge particu- larly that of Dr. Robert Smillie, Dr. George K. Russell, Dr. Shirley Raps, Mr. Gerald J. Roth, Mr. Gabriel Gingras, and Mrs. Elizabeth E. Levine. I also wish to acknowledge the technical assistance of Miss Joanne Brungard. The research reported here has been supported in part by a research grant from the National Science Foundation, and by research grants from the National Institute of Allergy and Infectious Diseases and the Division of General Medical Sciences, United States Public Health Service. LITERATURE CITED 1. Emerson, R. , Sci. , 125 , 746(1957), Emerson, R. , Chalmers, R. , and Cedarstrand, C. , ProcT Nat. Acad. Sci. (U.S.), 43, 133(1957), Emerson, R. and Chalmers, R. , Phycol. Soc. Amer. News Bull., 11, 51 (1958), and Emerson, R. and Rabinowitch, E. , Plant Physiol., ^, 477, 1960. 2. Hill, R., and Bendall, F. , Nature, 186 , 136(1960). 3. "The Photochemical Apparatus, Its Structure and Function, " Brookhaven Symposia in Biology, 11 (1958). 4. "Light and Life, " W. D. McElroy and H. B. Glass (Editors), Johns Hopkins Press, Baltimore ( 1961). 5. Colloq. Intern. Photosynthese, Centre National de la Recherche Scientifique, Paris, No. 119(1963), in press. Levine, R. P., Proc. Nat. Acad. Sci. (U.S.), 46, 972(1960). /^^i^-~^'^ |uu( LIBRARY j^ 171 R. P. Levine 7. Levine, R. P., Nature, J_88, 339(1960). 8. Levine, R. P., and Volkmann, D. , Biochem. Biophys. Res. Commun. , 6, 263 (1961). 9. Levine, R. P., and Smillie, R. M. , Proc. Nat. Acad. Sci. (U.S.), 48, 417 (1962). 10. Levine, R. P., and Smillie, R. M. , J. Biol. Chem. , 863 , (1963), in press. 11. Russell, G. K., and Levine, R. P., Plant Physiol. , 38, 542, (1963). 12. Smillie, R. M. , and Levine, R. P., J. Biol. Chem., 863 , (1963), in press. 13. Krinsky, N. I., and Levine, R. P., in preparation. 14. Levine, R. P., and Piette, L. H. , Biophys. J., 2, 369(1962). 15. Gaffron, H. , Am. J. Bot. , 27, 273(1940). 16. Roth, G. , Russell, G. K. , and Levine, R. P., in preparation. 17. Bishop, N. I., Nature, 195 , 55(1962). 18. Levine, R. P., and Ebersold, W. T. , Zeit. Vererbuugs., 89, 63, (1958). 19. Levine, R. P., and Ebersold, W. T. , Ann. Rev. Microbiol., _14, 197 (1960). 20. Ebersold, W. T. , Levine, R. P., Levine, E. E. , and Olmsted, M. A., Genetics, 47, 531 (1962). 21. Morovkova, J., and Levine, R. P., unpublished. 22. Levine, R. P., and Russell, G. K. , in preparation. 23. Allen, M. B. , Piette, L. H. , and Murchio, J. C., Biochim. Biophys. Acta, 60, 539 (1962). 24. Weaver, E. , and Weaver, H. E. , Abstracts, Meeting of Biophysical Society, Washington, D.C., February, 19 62. 25. Weaver, E. , and Bishop, N. I., Science, 140 , 1095(1963). 26. Levine, R. P., and Raps, S. , unpublished. 172 R. P. Levine 27. Kegel, C. P., Henninger, M. D. , and Crane, F. L. , Biochem. Biophys. Res. Commun. , 8, 294(1962). 28. Henninger, M. D. , Dilley, R. A., and Crane, F. L. , Biochem. Biophys. Res. Commun., _10, 237(1963). 29. Dilley, R. A., and Crane, F. L. , Plant Physiol. , 38, 452(1963). 30. Beinert, H. , Kok, B. , and Hoch, G. , Biochem. Biophys. Res. Commun., 7, 209 (1962). 31. Witt, H. T., Miiller, A., and Rumberg, B. , Nature, 197 , 987(1963). 32. Tagawa, K. , Tsujimoto, H. Y. , and Arnon, D. I., Proc. Nat. Acad. Sci. (U.S.), 49, 567 (1963). 33. Losada, M. , Whatley, F. R. , and Arnon, D. I., Nature, 190 , 606(1961). 34. Forti, G. , Bertole, M. L. , and Parisi, B. , Biochem. Biophys. Res. Commun., J_0, 384(1963). 35. Duysens, L. N. M. , and Ames z, J., Biochim. Biophys. Acta, 64, 243 (1962). 36. Witt, H. T. , Miiller, A., and Rumberg, B. , Nature, 191 , 194(1961). 37. Olson, J. M., Smillie, R. M. , and Bergeron, J., Plant Physiol. , 38, supplement, p. XXIX (1963). 38. Crane, F. L. , Plant Physiol. , 34, 128(1959). 39. Crane, F. L. , Ehrlich, B. , and Kegel, L. B. , Biochem. Biophys. Res. Commun., 3, 37(1960). 40. Redfearn, E. R. and Friend, J., Phytochemistry, J_, 147(1962). 41. Dilley, R. A. and Crane, F. L. , Biochim. Biophys. Acta, 75, 142(1963). 42. Bishop, N. I., Proc. Nat. Acad. Sci. (U.S.), 45, 1696(1959). 43. Lynch, V., and French, C. S. , Arc. Biochem. Biophys., 70, 382(1957). 44. Henninger, M. D. , and Crane, F. L. , Biochim. Biophys. Acta, 75, 144 (1963). 45. Kegel, L. P., and Henninger, M. D. , Plant Physiol. , 38, supplement, p. XI (1963). 173 R. P. Levine 46. Henninger, M. D. , and Crane, F. L. , Plant Physiol. , ^, supplement, p. XI (1963). 47. Krogman, D. W. , Biochem. Biophys. Res. Commun. , 4, 275(1961). 48. Arnon, D. I., Whatley, F. R. , and Horton, A. A., Federation Proc. , 21, 91 (1961). EFFECTS OF QUINONES AND OXYGEN IN THE ELECTRON TRANSPORT SYSTEM OF CHLOROPLASTS Achim Trebst, Herbert Eck and Sieglinde Wagner Quinones always played a major role in new developments in extracellular photosynthesis with isolated chloroplasts . After Hill had demonstrated oxygen evolution accompanying the photosynthetic reduction of ferric salts by chloro- plasts(l), Warburg introduced p-benzoquinone and naphthoquinone-sulfonate as Hill reagents(^). In the discovery of photosynthetic phosphorylation by Arnon it became soon apparent that vitamin K3 = methylnaphthoquinone is one of the most effective cofactors(-^). When the role of a new class of natural benzoqui- nones in oxidative phosphorylation of mitochondria (the ubiquinones) was investi- gated. Crane also found a similar, but somewhat different endogenous quinone in chloroplasts (plastoquinone)(4) and Bishop showed its importance in the pho- tosynthetic reduction of ferricyanide by chloroplasts^^'. Quinones are there- fore involved in the two principal photosynthetic reactions of chloroplast frag- ments—Hill reaction and photosynthetic phosphorylation— in two ways: as added substrate or cofactor and as an endogenous component of the electron transport chain. 1 . Quinones as substrates for the Hill reaction Numerous compounds have been tested and found suitable as Hill reagents, It does not seem surprising, that organic compounds, functioning as Hill rea- gents, have a quinoid structure, since this is the one most easily reduced. Aronoff and Wessels in 1952 investigated the photosynthetic reduction of a num- ber of substituted benzo- and naphthoquinones by chloroplasts in relation to their redoxpotentials'"" ' ' . Wessels observed no reduction of compounds with a redoxpotential more negative than -loo mV. When Arnon had shown that the Hill reaction with ferricyanide or TPN was coupled to a stoichiometric ATP formation according to the equation^"' X + H^O + ADP + P. XH^ + 1/2 O^ + ATP 2 1 2 ^ . (9) it seemed desirable to investigate the behavior of benzoqumones again Table 1 shows the reduction of a number of substituted p-benzo-, naphtho- and anthraquinones in nitrogen by illumination with broken chloroplasts. Oxy- gen evolution as well as coupled ATP formation was measured. 174 175 Achim Trebst, Herbert Eck and Sieglinde Wagner The table indicates, that in the presence of all quinones ATP is formed, which in most cases is accompanied by oxygen evolution. It seems safe to con- clude, that substituted benzoquinones can be used as Hill reagents and that their reduction is coupled to ATP formation. The non-cyclic type of photophosphoryla- tion(8) can be generalized to probably all Hill reagents. However, as seen in table 1, stoichiometry between oxygen evolution and ATP formation is usually not obtained. With falling redoxpotential of the qui- none, oxygen evolution is lagging behind ATP formation; when the redoxpotential is approaching zero V, no oxygen is evolved at all. This is in agreement with the results of Wessels^^^ The reason for this seems easy to understand. The more negative the redoxpotential, the more autoxidizable is the hydroquinone. Therefore the oxygen evolved in the reduction of the quinone is reacting back with the hydroquinone formed. This was already suggested by Wessels^ as possible explanation for his failure to observe reduction of compounds with a redoxpotential below -loo mV. That such backreactions occur, can now be seen by the formation of ATP without any measurable (by manometric techniques) oxygen evolution. The experiments in table 1 have been performed in nitrogen; apparently the small concentration of oxygen formed is sufficient to react with the hydroquinone formed. Therefore oxygen as well as the quinone/hydroquinone couple are then cycling. substrate ^atoms O evolved ^moles ATP formed ferricyanide p -b enz oquinone toluquinone 2 , 3 -dimethyl-p-benz oquinone 2 , 5-dimethyl-p-benz oquinone trimethylbenz oquinone 2 , 6-dimethoxy-p-benz oquinone 2 , 6 -dimethoxy-methylbenz oquinone 2-hydroxy-benzoquinone-5-propionate 2 -methyl-naphthoquinone phthiokol anthraquinone-2- sulfonic acid o.l 5.0 5,8 4.4 5.1 3,9 4.1 3,3 2,6 o,l o o o o,3 4,8 2,8 2 93 3,o 237 3,1 177 3,9 176 5,1 lo2 5,4 53 5,1 51 2,5 27 7,4 -lo 6,1 -18o 7,1 -25o Table 1: Quinones as Hill reagents Illumination for 15 min at 15° with 3 5ooo Lux in nitrogen. Each vessel contained in /imoles: trisbuffer p 8,o 8o; MgCl2 5; ADP lo; Pj^ lo; quinone 5 or ferricyanide lo; and broken chloroplasts (Pj^^j^) with o,3 chlorophyll in a total volume of 3 ml. 176 Achim Trebst, Herbert Eck and Sieglinde Wagner 2 . Quinones as cofactors of cyclic photophosphorylation If the added components of the system are cycling, no substrate amounts of a quinone should be required. Indeed, all tested substituted o- and p-benzo- quinones are cofactors of a cyclic photophosphorylation^'': by illuminating chloroplasts with catalytic amounts of a quinone (or hydroquinone) ATP is formed (table 2 and 3). Since ATP formation with benzoquinones in catalytic amounts occurs only in the presence of oxygen'^', this has to be called an aerobic or pseudocyclic photophosphorylation. o, 1 /imol cofactor added /imoles ATP formed chlorogenic acid 8,2 caffeic acid 8,1 dihydroxyphenylalanin (DOPA) 7,8 dihydroxyphenylethylamin (Dopamin) 8,6 catechol 6,3 vitamin K_ 6,4 Table 2: o-Hydroquinones as cofactors of aerobic photophosphorylation (conditions as in table 1, 15 min light in air). The term pseudocyclic was introduced by Arnon' '-'' to distinguish an oxygen dependent photophosphorylation from true oxygen independent cyclic photo- phosphorylation. In aerobic (pseudocyclic) photophosphorylation the cofactor is first reduced in a coupled Hill reaction. The reduced cofactor is then re- oxidized by oxygen. Q + H^O + ADP + P. -QH_, + ATP + 1/2 0_ 2 1 2 i. QH^ + 1/2 O^ ►Q + H^O sum: ADP + P. ►ATP 1 The higher efficiency of certain cofactors of photophosphorylation in an atmosphere of air or oxygen has been observed^ ~ ' . All oxidizable com- pounds, whose oxidized form can act as Hill reagent have to be included in the list of cofactors of aerobic photophosphorylation, among them numerous o- and p- hydroquinones , sonne of which are mentioned in tables 2 and 3. Substituted p- and o-benzoquinones , in catalytic amounts are not active as cofactors of true cyclic photophosphorylation in an atmosphere of nitrogen'"'. But a number of naphtho- and anthraquinones are effective in nitrogen^^ ' "' . Such ATP formation independent of oxygen was discovered by Arnon and later 177 Achim Trebst, Herbert Eck and Sieglinde Wagner o 5 1.7 o 1 1.2 o 4 1.3 o 1 4.7 o 8 3.5 o 2 4.3 o 4 5.7 o 6 6,7 o 1 4.9 2 o 4.7 4 3 6,8 6 7 lo, o 5 o 5.0 redoxpotential fxmol ATP o, 1 fxmol cofactor added (E in mV) formed in ^________ N2 air p-benzoquinone 293 2,3-dimethoxy-p-benzoquinone 198 2 , 3-diniethyl-p-benzoquinone 177 2 , 5-benzoquinone-diacetic acid 167 2 , 3 -dinnethoxy-methylbenzoquinone 151 1 . 4-naphthoquinone- sulfonic acid 118 2 , 6-diniethoxy-benzoquinone 53 phenanthrenquinone 28 2 -hydroxy-benzoquinone -propionate 27 vitamin K3 - 1 o 2 -hydroxy -naphthoquinone -154 phthiokol -I80 anthraquinone-2 -sulfonic acid -25o Table 3: Quinones as cofactors of photophosphorylation in nitrogen and in air (conditions as in table 1), termed cyclic photophosphorylation^ ' , to distinguish it from phosphorylation accompanying non-cyclic electron flow in the Hill reaction, which was discussed above. However, Arnon has surmised that the phosphorylating step is identical in the two systems' ^°'. In the truly cyclic electron flow, the reduced cofactor cannot be reoxidized by oxygen (being absent) , but is presumably reoxidized by a component of the endogenous electron transport chain of chloroplasts^ ' . Since the switching from aerobic to cyclic photophosphorylation occurs with co- factors with a redoxpotential around and below zero'9) (table 3 see also (^6 and '), one might conclude, that the endogenous oxidizing component has a redox- potential of about zero volt. Two endogenous compounds of chloroplasts , as far as discovered, have such a redoxpotential: plastoquinone and cytochrome b^ . It is interesting, that Kamen also found an optimum in bacterial photophosphory- lation at a redoxpotential of zero volt(19) The conclusion, that a true cyclic photophosphorylation is possible in broken chloroplasts, has been questioned because stimulation of photophos- phorylation (with suboptimal cofactor concentrations) by oxygen has been ob- served. Also, isotope experiments showed fast O2 -exchange between air and water in cyclic photophosphorylation with Sonne of the original cofactors'^o). However, all this shows is that the reduced cofactor of cyclic photophosphoryla- tion is preferentially reacting with oxygen, if present, rather than with the en- dogenous oxidizing component in the chloroplasts. Whereas oxygen is able to react with very small amounts of a hydroquinone, a certain concentration of the cofactor of true cyclic photophosphorylation has to be used to saturate the re- action with the endogenous oxidizing component of the chloroplasts. Also, a certain specifity in the constitution of a cofactor of cyclic photophosphorylation 178 Achim Trebst, Herbert Eck and Sieglinde Wagner should be expected (whereas almost any redox-catalyst will be active in the presence of oxygen). Inhibition studies with low concentrations of DCMU and KCN(^^'^^' ^^) and the separation of a cyclic photophosphorylation system from the oxygen evolution system^^-^), seems to support strongly the view, that true cyclic photophosphorylation exists, at least when vitamin K3 or PMS as cofactors are used. Of course, this does not mean that cyclic photophos- phorylation, as observed in isolated chloroplasts, is also a physiological reaction. Very recently Arnon showed, that ferredoxin can act, under certain con- ditions, as a cofactor of cyclic photophosphorylation' '. The stimulation of this cyclic photophosphorylation by DCMU^^^' is reminiscent of the cyclic sys- tem with DC PIP as catalyst{22). As table 4 shows, DC PIP is effective as a co- factor of a photosynthetic ATP formation in the absence of oxygen. This ATP formation is not only insensitive to high concentrations of DCMU, but is actu- ally stimulated by it. Photophosphorylation in the presence of oxygen is, how- ever, inihibited. (Since the DCPIP must first be reduced in a DCMU-sensitive Hill reaction, DCMU was added after 1 min pre -illumination). Cyclic photophos- phorylation with DCPIP as catalyst has recently been confirmed in several laboratories (2 5"^^). /imoles ATP formed in additions to o,3 fxmol DCPIP nitrogen air .-- 1.5 6,3 + lo" m DCMU 5,3 o, 1 Table 4: DCPIP as cofactor of cyclic photophosphorylation (conditions as in table 1, 15 min light. DCMU was added after 1 min pre-illumination). 3. Quinones in HO formation Warburg showed that in aerobic photophosphorylation the hydroquinone, acting as a catalyst, is reoxidized by O2 under formation of H2O2' '. Since endogenous catalase of the chloroplasts would decompose most of this H2O2, a catalase inhibitor has to be added in order to observe H2O2 accumulation. KCN, aminotriazole or diethyldithiocarbamate can be used, since these compounds do not interfere with the photosynthetic reactions^^^) . lo"-^m KCN is the most convenient inhibitor. Instead of inhibiting the endogenous catalase, an ethanol/ catalase trap for H2O2 may be added(^°'^^) . The reaction sequence of aerobic photophosphorylation in the presence of KCN is then: Q + HO + ADP + P. -QH^ + 1/2 O, + ATP 2 \ I c. QH^ , O^ .Q + H^O^ 179 Achim Trebst, Herbert Eck and Sieglinde Wagner sum: H^O + 1/2 O^ + ADP + P.- 2 2 1 ►H^O^ + ATP The ratio of ATP and HO formation to O uptake is 1 1 : o,5 (14,22) The Hill reaction with a quinone may therefore be followed not by oxygen evolution, but by oxygen uptake, if the experiment is done in air and lo~-^m KCN is added. Only catalytic amounts of a quinone are required, which is often important, when substrate amounts of a quinone are inhibitory or insoluble. H2O2 formation by illuminating chloroplasts and its stimulation by quinones is long known as Mehler reaction''^°'. It seems, however, somewhat misleading, to speak of oxygen as a Hill reagent, when it is only a variant of a Hill reaction with a quinone. 4. The photooxidation of hydroquinones It is proper to assume, that the reaction between a hydroquinone and oxy- gen under formation of H2O2 in aerobic photophosphorylation is an autoxyda- tion, particularly if the hydroquinone has a low redoxpotential. This is not correct, however, in the case of the oxidation of hydroquinones with rather pos- itive redoxpotentials, since these are not, at ppj 8, readily autoxidizable. Still, p-hydroquinone and even better, chlorogenic acid or dopamin are, as shown in table 2, excellent cofactors of aerobic photophosphorylation and these hydro- quinones are therefore rapidly oxidized by chloroplasts. The suggestion, that a phenoloxidase might be responsible' I'*', can be ruled out, since the experi- ment can be done in the presence of KCN''^'^', which would inhibit the phenol- oxydase. Table 5 shows, that chlorogenic acid, p-hydroquinone and dopamin (with redoxpotentials above +29o mV) are not substantially oxidized by chloroplasts in the dark, whether KCN is absent or not (there is some phenoloxidase activity, as seen in the dopamin experiment). In the light, however, oxygen is taken up and H2O2 accumulates, even and particularly in the presence of KCN (which again inhibits endogenous catalase). in the dark in the light /Liatoms O /imoles H2O2 /iatoms O jumoles H2O2 taken up chlorogenic acid o + lo-^m KCN o p-hydroquinone o " + lo-^m KCN dopamin 1 , 3 + lo"^m KCN o form ed taken up formed 1,6 1,0 6,4 5,8 0,8 5,2 0,8 4,8 2,1 0,8 7,2 7,0 Table 5: Photooxidation of hydroquinones (5 ^mol) by broken chloroplasts (conditions as in table 1; 15 min). 180 Achim Trebst, Herbert Eck and Sieglinde Wagner These hydroqainones therefore seem to be photo oxidized. This photooxi- dation is not a chlorophyll sensitized photooxidation for two reasons. 1. The photooxidation is inhibited by lo""m DCMU and lo'^^m o-phenanthroline^ ' (see also table 11). 2. Treatment of chloroplasts with a detergent and solu- bilizing the chlorophyll destroys the ability to photooxidize hydroquinones^^*^) . The inhibition of this photooxidation by DCMU seems to indicate that the oxygen evolution system of photosynthesis somehow participates. It is difficult to give an explanation. It seems conceivable that a hypothetical peroxyde, which gives off oxygen in usual photosynthesis, is oxidizing the hydroquinone, perhaps via a quinolperoxyde. A mechanism like this has been proposed also by Jagendorf as an explanation for the photooxidation of ascorbate by otherwise unsupplemented chloroplasts^-^^). This theory implicates that oxygen is required in order to evolve oxygen, even in normal photosynthesis. Such a hypothesis has been ad- vanced in particular by Schenck on the basis of the behavior of chlorophyll in chemical photoreactions'-^^'. Certainly more experiments are needed, to sup- port such a view. 5 . Quinone s in the photooxidation of ascorbic acid As already naentioned, the photooxidation of o-hydroquinones shows sim- ilarities to the photooxidation of ascorbic acid. There are several possible ways, in which ascorbic acid may be oxidized by chloroplasts, which cannot be dis- cussed here in detail (see Jagendorf^^-^' for a review of the pertaining literature). Stimulation of ascorbic acid oxidation by quinones was studied in several labo- ratories(^4-4o, 31) Wessels concluded that the quinone stimulated photooxida- • • • • (34) tion of ascorbic acid is a chemical, chlorphyll sensitized, photooxidation^ '. Others, however, concluded that ascorbic acid photooxidation proceeds via all or part of the electron transport chain of chloroplasts'35 , 37)^ Quinone stim- ulated ascorbic acid photooxidation was inhibited by o-phenanthroline in Ikeda's experiments' ''. Substituted p-benzoquinones with a redoxpotential in the range from o till + Zoo mV cannot be used for the stimulation of ascorbic acid photo- oxidation by chloroplasts, since they catalyze already a rapid dark oxidation. Also vitamin K3 at a concentration of lo~m catalyzes a chemical dark oxidation of ascorbic acid. More interesting is the stimulation of ascorbic acid oxidation by low concentrations of vitamin K3 (lo'rn) and by anthraquinonesulfonic acid, which occurs only by illumination with a chloroplast system. This stimulation is inhibited by DCMU (table 6), which suggests that the electron transport chain of photosynthesis in chloroplasts or part of it is participating, lo" m KCN was not inhibiting and was added to prevent H2O2 decomposition by the endogenous catalase. Coupled ATP formation in a ratio 1 to 1 to H2O2 also supports the view that we are not just dealing with a chlorophyll sensitized photooxidation. Two explanations for this stimulation of ascorbate oxidation can be offered (see also Jagendorf^-^'^'): 1. The quinone is reduced in a Hill reaction and rapidly autoxidized under for- mation of H2O2. The H2O2 (perhaps with the help of a peroxidase) is quickly 181 Achim Trebst, Herbert Eck and Sieglinde Wagner oxidizing ascorbic acid. The DCMU sensitivity of the Hill reaction would ac- count for the DCMU inhibition of the overall reaction. However, at the short exposure time in the experiments of table 6, ascorbic acid is stable in the presence of HO. additions to lo ^mol ascorbate o, 1 ixmol anthraquinonesulfonate o, 1 ^mol anthraquinonesulfonate + lo-^m DCMU o,o3 /imol vitamin Ko o,o3 jimol vitamin K^ + lo" m DCMU oxygen uptake H2O2- ATP- fiatoms formation (|Umol) o, 6 0,5 13,5 7,0 7,0 o.l 0,1 0.1 12, o 6.2 6, o,8 0.3 0.1 Table 6: Inhibition of quinone stimulated ascorbate photooxidation by DCMU (conditions as in table 1; 15 min light in air, lo"-^m KCN per vessel). Z. Ascorbic acid is donating electrons into the electron transport chain at a point before the DCMU inhibition site. The electron is transported to the qui- none, which is then autoxidized. This is a mechanism, also proposed by Ikeda^ 'and very recently by Chiba'-^"' and which is in agreement with the ex- periments of Habermann(-^^)and Marre^-^^^). In this view, ascorbate (Aa) would substitute for water as electron donor at the same site. ADP + P. + H^Aa + oiiliiQH + Aa + ATP i 2 ~ ^2 \ -Q + ^2^2 QIL + O^ ►Q + H^O^ sum: ADP + P. + H_,Aa + O^ ►Aa + H^O. + ATP 1 Z Z d c The ratio of H2O2 and ATP formation to O^ uptake in this reaction type is 1 : 1 : 1 as against the ratio of 1 : 1 : o, 5 in aerobic photophosphorylation. This quinone stimulated ascorbic acid photooxidation differs in many re- spects from the DCPIP stimulated ascorbate oxidation. The latter is not inhib- ited by DCMU('^°). One major change introduced by the addition of DCPIP is the point of entry of the electrons donated. As the work of Vernon^ and Witt^'^' clearly indicated, ascorbate donates electrons via DCPIP into the cytochrome chain. Explanation 2 would also account for the ascorbate stimulation of cyclic photophosphorylation with certain cofactors. Vennesland^ showed, that for FMN catalyzed cyclic photophosphorylation catalytic amounts of ascorbate are required and that the ascorbate stimulation is abolished by o-phenanthroline. 182 Achim Trebst, Herbert Eck and Sieglinde Wagner Table 7 indicates, that photophosphorylation catalyzed by indigo- sulfonic acids is like the FMN system stimulated by ascorbate. This stimulation is again DCMU sensitive. 0,2 ymol indigo-sulfonic acid " " +2,5 ymol ascorbate 1' " " " + lo""^ DCMU 0,2 jimol indigo-disulfonic acid " " +2,5 ;imol ascorbate " + lo"'* DCMU o, 1 ^mol FMN " " + 2,5 ^mol ascorbate 1! <• " " + lo"^ DCMU ^moles ; ATP formed 1, I 4, 4 0, 8 1. 2 6, ,1 1, , 1, ,5 7, ,6 1, ,7 Table 7: Stimulation of cyclic photophosphorylation by ascorbate (conditions as in table 1; 15 min light in N^). This might best be explained as a donation of electrons by ascorbate into the electron transport chain before the DCMU block. In the absence of ascorbate, the oxidation of the reduced cofactor by an endogenous oxidizing compound of the chloroplasts is limiting. Ascorbate alone is not a good cofactor, since then the reduction of dehydro- (or monodehydro-) ascorbate is limiting^^ '. Against the explanation that ascorbate substitutes for water is, that a stoichiometry of oxygen evolution and TPNH formation is observed also in the presence of as- corbate. o-Quinones (or rather o-hydroquinones) behave quite different from qui- nones with negative redoxpotentials in the ascorbate oxidation by chloroplasts. o-Hydroquinones do not stimulate ascorbic acid photooxidation by chloroplasts. On the contrary, ascorbate inhibits the photooxidation of these hydroquinones (table 8). o, 1 jLtniol chlorogenic acid 0,1 /xmol chlorogenic acid + lo |imol ascorbate o, 1 jxmol catechol o, 1 ^mol catechol + lo ^imol ascorbate o, 1 ^mol anthraquinonesulfonate o, 1 ^mol anthraquinonesulfonate + lo ^mol ascorbate 14,1 6,7 O2 uptake Hz , O^ formed ^atoms jLXmoles 3,7 3,1 0,8 0,6 6,5 5,5 0,8 0.4 5.6 6, Table 8: Inhibition of o-hydroquinone photooxidation by ascorbate (in compar- ison with anthraquinone) (15 min light in air; lo'^m KCN per vessel). 183 Achim Trebst, Herbert Eck and Sieglinde Wagner This might best be explained by a competition of o-hydroquinone and ascorbate for the same site of oxidation. Since no autoxidizable hydroquinone is formed (as it is from anthraquinone) the system is blocked. 6. Th e action of salicylaldoxime on photosynthetic reactions Salicylaldoxime is a copper chelating agent, which inhibits copper con- taining enzymes, like phenoloxydase^'^^) and cytochromeoxydasev'*'*). It also inhibits photosynthesis in intact Chlorella^'*^' . Table 9 shows the influence of salicylaldoxime on various photosynthetic activities in broken chloroplasts . All reactions involving oxygen evolution (ferricyanide and TPN reduction) as well as cyclic photophosphorylation (vitamin Kj as cofactor) are inhibited by lo" m salicylaldoxime. The only photosynthetic reaction possible in the presence of salicylaldoxime is the reduction of TPN at the expense of DCPIP/ascorbate. For comparison the behavior of DCMU, as worked out by Vernon and others^^^''^^''*^) is included in table 9. The difference between DCMU and sali- cylaldoxime is, that DCMU does not inhibit cyclic photophosphorylation, whereas salicylaldoxime does. In the DCMU experiments, the addition of DCPIP/ascor- bate restores TPNH and ATP formation, whereas in the salicylaldoxime exper- iments coupled ATP formation does not reappear, when TPN is reduced by DCPIP/ascorbate. According to Vernon and Witt^ reduced DCPIP reacts with cyto- chrome f . Since salicylaldoxime does not influence photosynthetic TPN reduc- tion at the expense of DCPIP/ascorbate, the site of inhibition of salicylaldoxime must be before cytochrome f. But since salicylaldoxime inhibits cyclic photo- phosphorylation, its site of inhibition would be after plastoquinone and the sec- ond light reaction (see scheme). The phosphorylation site (either in cyclic or non-cyclic photophosphorylation) cannot be between cytochrome f and TPN, since the reaction sequence: ascorbate - DCPIP - cytochrome f - light - TPN is not coupled in the presence of salicylaldoxime. Witt already argued for rea- sons of redoxpotential that the phosphorylation site has to be between plastoqui- none and cytochrome f^^^). In the DCMU experiments, where the reduction of TPN by DCPIP/ascorbate is coupled to ATP formation^^") , one has to assume, that DCPIP does not react with cytochrome f but with a compound, located in the electron transport chain before the phosphorylation site, possibly plastoqui- none or even with a compound (Y) before the second light reaction, as suggested by Witt(42). The inhibition of photosynthetic reactions in chloroplasts by salicylald- oxime seems interesting; for if it is accepted that salicylaldoxime as copper chelating agent inhibits a copper enzyme in these experiments with chloroplasts, then the site of salicylaldoxime inhibition would indicate the location of this copper enzyme in the electron transport chain of photosynthesis. Such a copper enzyme has already been isolated from chloroplasts. Katoh named it plastocyanine and drew attention to its possible significance in 184 Achim Trebst, Herbert Eck and Sieglinde Wagner 1-^ o T! tn 0) £ > . 1 o o 4-) > rt 1> a. xO o o O -1 r- o in o" CO 00 CX3 o" o" rfi -H IT) — 1 -H —I (N] O (vj O O O O O (U a •1-1 X o Tl .— < (Tl D n-1 2 o •rH 2; U I — 1 (ti u Q tn w s a £ in (M (VJ O 4- + (U XI •iH C o • I-t u CO o (M w c o £ a. O (M 00 (M _ O O XI XI o u u u to o O o rt o o u i-H . — ^ .-H ^H £ •iH 4) S u Z + + + £ U Q £ .—1 tn £ s o Tl »— 1 ■iH X o f— 1 ^ (\j (M D D 1-H o 1 :^ S o •tH o ^^ — ' — ' U O .—4 (ti •rH + + + P Q in tn CO o Oh I— I u Q a. o" ni XI u o o tn (Ti I — I o a a. o X o Tl nJ r— 1 u I-H tn I (NJ I a £ rsj (M I ^ ^ ^ O H I— I o £ a. I u a, a o o O X o O in ^^ >^ 1) XJ 00 rn o -4-1 u rn *j rt f .-H a r: o • iH ^ o rC 3 tJ) u .— 1 (U -rH -^ £ o u in XI -I C - to (D E! ,-1 O X! O !-i ^^ ' °°', DCPIP^ ', benzoquinones^^^) and TPN^^^ ' 59))_are inactivated and become plastoquinone dependent, when the chloroplasts were exhaustively extracted with petrolether or acetone. The only exception is the reduction of TPN by DCPIP/ascorbate, which seems to be independent of plastoquinone'^" ' ^^' (see however Crane^ °'). This provides some indication as to the location of endogenous plastoquinone in the electron transport chain of chloroplasts. In agreemient with Witt's interpre- tation of experiments, which gave more direct spectroscopic evidence' ', plas- toquinone is probably the acceptor of the second light reaction and is situated before the cytochrome chain. We have pointed out, however, that incomplete extraction of plastoquinone leads to a somewhat different picture'^"^ By extraction of 7o% of the endoge- nous plastoquinone (which can easily be accomplished by an only short treatment of chloroplasts with petrolether), only the reduction of ferricyanide and oi o-quinones is impaired, but not the reduction of TPN and p-benzoquinones^ '. We have concluded from this, that there is a second site of plastoquinone in the electron transport chain^^"'. This second site has to be in a sidepath to the main chain (leading to TPN) after the first light reaction (see scheme) and is participating only in ferricyanide reduction. This second plastoquinone site has recently been confirmed by Witt by direct spectroscopic observations at 26o mJ^^K Crane recently suggested also several plastoquinone sites in photosyn- thesis(^°). The higher sensitivity of the Hill reaction towards UV-light as com- pared to cyclic photophosphorylation, noticed by Avron'" ', and the stimulation of photosynthetic reactions of chloroplasts by the further addition of plastoqui- none(°^'°^) (surprising in view of the high plastoquinone content of chloroplasts) might also be explained by two plastoquinone sites, only one of which is of phys- iological importance. The two plastoquinone sites are different also in the specifity by which they can be reactivated after petrolether extraction of the endogenous plasto- quinone^^^). Table 13 shows that after incomplete extraction of plastoquinone (removal of plastoquinone at site 2), the reduction of ferricyanide, vitamin K3 and anthraquinone-sulfonic acid is impaired and shows a stimulation by the ad- dition of plastoquinone-45, whereas this is not the case with the reduction of TPN and the two substituted benzoquinones. 189 Achim Trebst, Herbert Eck and Sieglinde Wagner ^mol electrons transferred without with the addition of 0,2 ^mol plastoquinone 5 fxmol TPN lo /xmol ferricyanide 0,2 /xmol 2,3-dimethyl-benzoquinonebutyrate 0,2 fxmol 2,3-dimethoxy-methylbenzoquinone 0,2 ^mol vitamin K^ 0,2 jLimol anthraquinone- sulfonate 2 , o 0,1 1.8 2.1 0,1 0.1 2.1 2,2 2.2 2.2 2,8 2,5 Table 13: Plastoquinone dependence of the reduction of TPN. ferricyanide and quinones in petrolether extracted chloroplasts (only 7o% of endoge- nous plastoquinone removed). o,3 ml dialyzed. watersoluble chloro- plast extract was added in the TPN experiment; quinone reduction was measured by H7O2 formation in the presence of lo'^m KCN. 15 min light in air l^°) . The ferricyanide system in such incomplete extracted chloroplasts can be reactivated by numerous p-benzoquinones . some of which are shown in table 14. There seems to be no structural requirements in the reactivation of the second plastoquinone site, except that it has to be a substituted p-benzoquinone. addition of o,2 j^mol p -b enz oquinone 2 , 3 -dimethoxy-p-benz oquinone 2 , 3 -dimethoxy - methylb enz oquinone 2 . 3 -dimethyl-p-benz oquinone 2 . 3 -dimethyl -b enz oquinone -butyric acid tr imethyl -b enz oquinone p-benzoquinone-2 , 5-diacetic acid vitamin K_ jLimol ferricyanide reduced 0,8 1,0 3.8 3,7 3,8 4,1 4,2 4.4 0,9 Table 14: Reactivation of ferricyanide reduction in petrolether extracted chloroplasts (only 7o% of endogenous plastoquinone removed). 1 5 min light in N . Table 15 shows that after exhaustive extraction of plastoquinone the TPN system is now also plastoquinone dependent (at site one). The reactivation of this plastoquinone site is very specific. Only dimethyl-substituted benzoqui- nones with an isoprenoid side chain of at least 5 C-atoms are active. Allyl- as well as butyrate substitution in position 5 show no activity. 190 Achim Trebst, Herbert Eck and Sieglinde Wagner addition of o,2 /imol of M^T'oI TPNH formed 0,5 plastoquinone-45 ^ > ° plastoquinone-15 ''» ' plastoquinone-5 ^'" 2 , 3 -dimethyl -phytylbenzoquinone 3 > ^ 2 , 3 -dimethyl - 5 - allylbenz oquinone o > o 2 , 3-dimethyl-benzoquinone o> ° 2,3-dimethoxy-methylbenzoquinone o. ^ 2,3-dimethyl-benzoquinone-butyric acid o. 5 ubiquinone-5o °»^ Table 15: Reactivation of TPN reduction by quinones in petrolether extracted chloroplasts (exhaustively extracted, o,3 ml dialyzed, watersoluble chloroplast extract added, 15 min light in N^). II J II II H R = -CH2-GH=C.Qj^^ -CH2-CH=C.g -CH2-CH2-CH2-COOH 5 -5-dimethyl-allyl -5-allyl -5-butyric acid (plastoquinone-5) plastoquinone-45 no no activity "^ Besides two methyl-groups in position 2 and 3, a substitution at position 5 by -CH2-CH=C-(CH3)2 seems to be essential for the activity at the first plasto- quinone site in the electron transport chain of photosynthesis. Krogmann and Crane ° also showed that different photosynthetic re- actions of chloroplasts have different structural requirements in their reacti- vation, after endogenous plastoquinone has been extracted. The following scheme is to indicate the proposed two sites of plastoqui- none function and the site of salicylaldoxime inhibition as discussed in the pre- ceding chapters. 191 Achim Trebst, Herbert Eck and Sieglinde Wagner salicyl- aldoxime t plastoquinone ( 1 ) ADP^ ATP chlorophyll l^"*^ light reac Cu (?) chlorophyll 2^ lightreaction T , Y <- " H2O ■*"■ ascgrbate cytochrome f T _ DCPIP T ascorbate TPN plastoquinone (2). p-benzo- ^ quinones ferricyanide Scheme of photosynthetic electron transport in broken chloroplasts The research reported here has been sponsored in whole or in part by: Deutsche Forschungsgemeinschaft and Air Force Cambridge Research Laboratories, OAR through the European Office of Aerospace Research, United States Air Force. REFERENCES 1. R. Hill a. R. Scarisbrick, Nature (London) 146 , 61 (194o), 2. O. Warburg a. W. Liittgens, Naturwissenschaften 38^, Sol (1944). 3. D. I. Arnon, F. R. Whatley a. M. B. Allen, Biochim. biophys. Acta lb_, 6o7 (1955). 4. F. L. Crane a. R. Lester, Plant Physiol. 33_, VII (1958); F. L. Crane, Plant Physiol. 34, 128 (1959). 5. N. I. Bishop, Proc. Nat. Acad. Sci. USA 45 , 1696 (1959). 6. S. Aronoff, Plant Physiol. 2J_, 393 (1946). 7. J. S. C. Wessels a. E. Havinga, Rec. Trav. chim. Pays-Bas 71, 8o9 (1952); 72, lo76 (1953). 192 Achim Trebst, Herbert Eck and Sieglinde Wagner 8. D. I. Arnon, F. R. Whatley a. M. B. Allen, Science 127 . lo26 (1958). 9. A. Trebst a. H, Eck, Z. Naturforschg. 16b , 44 (1961). 10. D. I. Arnon, M. Losada, F. R. Whatley, H. Y. Tsujimoto, D. O. Hall a. A. A. Horton, Proc. Nat. Acad. Sci. USA 47 , 1314 (1961). 11. J. S. C. Wessels, Biochim. biophys. Acta 29^, 113 (1958). 12. T. Nakamoto, D. W. Krogmann a. B. Vennesland, J. biol, Chem. 234, 2783 (1959). 13. A. T. Jagendorf a. M. Avron, Arch. Biochenn. Biophys. 8o, 246 (1959). 14. O. Warburg, G. Krippahl, H. S. Gewitz a. W. Volker, Z. Naturforschg. 14b , 712 (1959). 15. R. HUl a. D. A. Walker, Plant Physiol. 34, 24o (1959). 16. L. P. Vernon a. M. D. Hobbs, Arch. Biochenn. Biophys. 72_, 25 (1957); A. T. Jagendorf a. M. Avron, J. biol. Chem. 231 , 277 (1958). 17. O. Kandler, Ann. Rev. Plant Physiol. n_, 37 (i960). 18. D. I. Arnon, Nature (London) 184. lo (1959). 19. T. Horio a. M. D. Kamen, Biochem. J_. 144 (1962). 20. A. R. Krall, N. E. Good a. B. C. Mayne, Plant Physiol. 36^, 44 (1961). 21. D. I. Arnon, in "Light and Life" Ed. W. D. McElroy a. B. Glass, The Johns Hopkins Press, Baltimore 1961. 22. A. Trebst a. H. Eck, Z. Naturforschg. I6b , 455 (1961). 23. J. S. C. Wessels, Biochim. biophys. Acta 65 . 561 (1962). 24. K. Tagawa, H. Y. Tsujimoto a. D. I. Arnon, Proc. Nat. Acad. Sci. USA 49, 567 (1963). 25. Z. Gromet-Elhanan a. M. Avron, Biochem. Biophys. Res. Commun. J_o, 215 (1963). 26. D. Keister, J. Biol. Chem. 238, 259o (1963), 27. B. Kok, B. Cooper a. L. Yang, in "Studies on Microalgae and Photosyn- thetic Bacteria", Special Issue of Plant a. Cell Physiol. 1963. 28. B. Vennesland, T. Nakamoto a. B. Stern, in "Light and Life" Ed. W. D. McElroy a. B. Glass, The Johns Hopkins Press, Baltimore 1961. 193 Achim Trebst, Herbert Eck and Sieglinde Wagner 29. A. M. Mehler, Arch. Biochem. Biophys, 33_, 65 (1951); 34, 339 (1951); N. Good, a. R. Hill, Arch. Biochem. Biophys. SJ. 355 (1955). 30. A. Trebst a. S. Wagner, Z. Naturforschg. 17b , 396 (1962). 31. G. Forti a. A. T. Jagendorf, Biochim. biophys. Acta 54, 322 (1961). 32. G. O. Schenck, Naturwissenschaften 4o, 2o5 (1953); 4o, 229 (1953). 33. A. T. Jagendorf, in "Survey of Biological Progress" Ed. B. Glass Aca- demic Press, New York 1962. 34. J. S. C. Wessels, Rec. Trav. chim. Pays-Bas 74, 832 (1955). 35. L. P. Vernon a. E. D. Ihnen, Biochim. biophys. Acta M, 115 (1957). 36. J. W. Hinkson a. L. P. Vernon, Plant Physiol. 34, 268 (1959). 37. S. Ikeda, Mem. Res. Inst. Food Sci. Kyoto Univ. r7_, 1 (1959); j_8, 57 (1959). 38. H. M. Habermann a. A. H. Brown, in "Research in Photosynthesis" Interscience New York 1957. 38a. E. Marre, O. Arrigoni a. G. Rossi, Biochim. biophys. Acta 3£, 56 (1959). 39. Y. Chiba a. S. Okayama, Plant a. Cell Physiol. 3, 379 (1962). 40. A. T. Jagendorf a. M. Margulies, Arch. Biochem. Biophys. 9o, 184 (i960). 41. L. P. Vernon a. M. D. Kamen, Arch. Biochem. Biophys. 5J_, 122 (1954) L. P. Vernon a, W. S. Zaugg, J. biol. Chem. 235 , 2728 (i960). 42. H. T. Witt, A. Muller a. B. Rumberg, Nature (London) 192, 967 (1961). 43. F. Kubowitz, Biochem. Z. 292^, 221 (1937). 44. K. Okunuki, in "Oxygenases" , Ed. O. Hayaishi, Academic Press New York 1962. 45. L. F. Green, J. F. McCarthy a. C. G. King, J. biol. Chem. 128 , 447 (1939). 46. M. Losada, F. R, Whatley a. D. I. Arnon, Nature (London) 19o , 606 (1961). 47. H. T. Witt, Proc. V Int. Congr. of Biochem. Moscow 1961; A. Muller, B. Rumberg, H. T. Witt, Proc. Roy. Soc. London Ser. B 157 , 313 (1963). 194 Achim Trebst, Herbert Eck and Sieglinde Wagner 48. S. Katoh, Nature (London) ^86, 533 {i960); S, Katoh a. A. Takamiya, Nature (London) j^, 665 (1961); S. Katoh, I. Suga, I. Shiratori a. A. Takamiya, Arch. Biochem. Biophys. 94, 136 (I96I). 49. O. Warburg, in " Weiterentwicklung der zellphysiol. Methoden" Georg Thieme Verlag Stuttgart 1962. 50. M. Schwartz, Biochim. biophys. Acta 66_, 292 (1963). 51. B. Vennesland, H. W. Gattung a. E. Birkicht, Biochim. biophys. Acta 66, 285 (1963). 52. A. Trebst, Z. Naturforschg. in press. 53. A. Trebst a. H. Eck, Z. Naturforschg. 18b , lo5 (1963). 54. J. S. C. Weasels, Biochim. biophys. Acta 3£, 195 (i960). 55. D. W. Krogmann, Biochem. Biophys. Res. Commun. 4, 275 (I96I); D. W. Krogmann a. E. Olivero, J. biol. Chem. 237 , 3292 (1962). 56. A. Trebst a. H. Eck, Angew. Chem. 73_, 769 (1961); A. Trebst, Proc. Roy. Soc. London Ser. B. ^l?, 355 (1963); A. Trebst a. H. Eck, Z. Naturforschg. in press. 57. H. T. Witt, A. Muller a. B. Rumberg, Nature (London) 191 , 194 (1961); Angew. Chem. 73_, 5o7 (1961). 58. M. Klingenberg, A. Muller, P. Schmidt-Mende a. H. T. Witt, Nature (London) j^, 379 (1962); J. Weikard, A. Miller a. H. T. Witt, Z. Naturforschg. j^, 139 (1963). 59. D. I. Arnon, F. R. Whatley a. A. A. Horton, Fed. Proc. 2_1_, 91 (1962) D. I. Arnon, Photosynthesis meeting Gif-sur-Yvette 1962. 59a. E. R. Redfearn a. J. Friend, Phytochem. J_, 147 (1962); Proc. Roy. Soc. London Ser. B J^, 364 (1963). 60. M. D. Henninger, R. A. Dilley a. F. L. Crane, Biochem. Biophys. Res. Commun. j^, 237 (1963); M. D. Henninger a. F. L. Crane, Biochem. in press. 61. N. Shavit a. M. Avron, Biochim. biophys. Acta 66^, 187 (1963). 62. S. Ikeda, Memm. Res. Inst. Food Sci. Kyoto Univ. 2j^, 17 (i960); J. Friend a. E. R. Redfearn, Biochem. J. 82, 13p (1961). PHOTOSYNTHETIC ELECTRON TRANSPORT AND PHOSPHORYLATION IN CHLOROPLASTS Daniel I. Arnon The type of photosynthetic phosphorylation first found in chloroplasts--the type now called cyclic photophosphorylation-- in which the sole product of the reaction is ATP, and in which no hydrogen (electron) donor or acceptor is consumed, yielded no ex- perimental evidence for a light-induced electron transport coup- led with phosphorylation (1,2). Direct experimental evidence, as distinguished from supposition, for a coupling between photosyn- thetic phosphorylation and photosynthetic (light-driven) electron transport in chloroplasts came in 1957 with the finding of what we now call noncyclic photophosphorylation (3) • Here the forma- tion of ATP was linked with a thermodynamically "uphill" hydrogen (electron) transfer from water to TPN (or ferricyanide)--a trans- fer that was accompanied by a stoichiometric oxygen evolution. The early hypotheses linking photophosphorylation with photo- synthetic electron transport centered on the photolysis of water as a common primary photochemical event in cyclic and noncyclic photophosphorylation (4). But since 1959 our work has been guided by an "electron flow" hypothesis which limits the photo- oxidation of water and the resulting evolution of oxygen to non- cyclic photophosphorylation (5). The common primary photochemi- cal event (coupled with ATP formation) in both cyclic and non- cyclic photophosphorylation is now envisaged as an electron transfer from excited chlorophyll to a primary electron acceptor molecule and thence, with the aid of appropriate enzyme systems, either to TPN (noncyclic) or back to chlorophyll via the cyto- chrome chain (cyclic electron flow) (5,6). In 1961 this hypothesis was further elaborated (7) to accom- modate the experimental separation of noncyclic photophosphory- lation in chloroplasts into two partial reactions: (a) ATP for- mation without oxygen evolution but coupled with the photoreduc- tion of TPN by the ascorbate-DPIP couple, and (b) photooxidation of water to molecular oxygen (7) . With the separation of these two reactions there was also preliminary evidence that photopro- duction of oxygen is catalyzed by a pigment system different from 195 196 Daniel I. Arnon that required for the photoreduction of TPN (8) . The aim of this article is to summarize the work in our labor- atory since 1961 which extends and supports the electron flow concept of photophosphorylation in chlorop lasts . The discussion will include experiments which led to: (a) The identification of ferredoxin as the most electronegative electron carrier isolated so far from chloroplasts and the elucidation of its role in photosynthetic electron transport. This followed a series of ex- periments on photoreduction of methyl viologen and photoproductLon of hydrogen gas by chloroplasts. (b) Correlations between pig- ment function and photochemical activity of spinach chloroplasts and algal chromatophores, and (c) identification of the position of plastoquinone in the noncyclic electron transport chain of chloroplasts . Some of the material given here was published in more detail elsewhere (9-1^); other reports are in preparation. Extensive reviews of earlier work from this and other laboratories are available (6,l4). Position of plastoquinone in the noncyclic electron transport chain. Crane (15) found that plastoquinone is localized in chloroplasts and Bishop (l6) and Krogmann (1?) have shown that it is required for the photoreduction of ferricyanide or 2,6- dichlorophenol indophenol by isolated chloroplasts. Fig. 1 shows that, after extraction of plastoquinone, chloroplasts lost the ability to photoreduce TPN when water (0H~) was the electron donor but not when the ascorbate-DPIP couple replaced water as the electron donor system (l8). Plastoquinone thus appears to be required in that portion of the photosynthetic electron transport chain in chloroplasts which is concerned with the photooxidation of water to molecular oxygen and which has been identified by Losada et al. (7) as the first of the two light reactions that jointly bring about the transfer of electrons from water to TPN. Plastoquinone may thus be the endogenous chloroplast factor which occupies the position marked as "A" in the noncyclic electron flow scheme of 1961 (Fig. 3 in ref. 7). Similar conclusions about the position of plasto- quinone in the chloroplast electron transport chain were reached by Witt et al. (19). Photoreduction of methyl viologen . According to the electron flow hypothesis, the primary electron carrier, common to the cyclic and noncyclic electron flow, must be able to reduce not 197 Daniel I. Arnon only TPN (E^ = -320 mV, pH 7) but also strongly electronegative redox dyes such as methyl viologen (Eq = -^55 mV, pH 7) , a dye which was shown by Jagendorf and Avron (20) and Hill and Walker (21) to catalyze cyclic photophosphorylation. It would follow from these considerations that an accumulation of photoreduced methyl viologen, although never before demonstra- ted, should be possible if its reoxidation by oxygen or by the cyclic electron flow mechanism of chloroplasts is prevented. This was experimentally shown by Mitsui et al. (22). Table 1 shows that illuminated chloroplasts reduced methyl viologen in the presence of cysteine and dichlorophenolindophenol (DPIP) . No reduction of methyl viologen was observed in the dark or after boiling the chloroplasts for 5 min. Table 1 Photoreduction of Methyl Viologen by Chloroplasts (22) Methyl viologen reduced (umoles/hr/mg chl) Complete system 5^*2 DPIP omitted 5*8 Cysteine omitted 0.2 Water was not the electron donor since the system was incap- able of evolving oxygen, with or without added CMU. The rate of photoreduction of methyl viologen was greatly decreased when DPIP was omitted from the reaction mixture. Thus, DPIP rather than cysteine appeared to be the effective electron donor. However, the function of cysteine was not limited to donating electrons via DPIP since ascorbate, which can also reduce DPIP, did not re- place cysteine in the same system. Only after a mild heat treat- ment (50° C for 10 min.)> which greatly reduced the total photo- activity of chloroplasts, were chloroplasts able to use the ascorbate-DPIP couple as an electron donor for the photoreduction of methyl viologen (at a low rate). The effectiveness of cysteine is explained by its inhibition of the cyclic electron flow by which reduced methyl viologen can be reoxidized by chloroplasts. As shown in Table 2, cysteine strongly inhibited cyclic photophosphorylation catalyzed by methyl viologen, menadione or FMN but not the "shortcut" cyclic photophosphorylation pathway catalyzed by phenazine methosulfate (6). The last observation suggests that the inhibitory effect of cysteine on cyclic photophosphorylation was not related to an activation of ATPase (23) since in that case all photophosphory- 198 Daniel I. Arnon lations would be expected to be affected. Table 2 Inhibition of Cyclic Photophosphorylation with Cysteine (22) ^lmoles ATP Additions formed Methyl viologen 5*2 Methyl viologen, cysteine 0.3 Vit. K3 6.6 Vit. K3, cysteine 0.2 FMN ^.8 FMN, cysteine 0.4 PMS 3-6 PMS, cysteine 3*6 Photoproduction of hydrogen gas by chlorop lasts . The ability of spinach chloroplasts to photoreduce methyl viologen--a dye used as an electron carrier for hydrogenase (24) --suggested that spinach chloroplasts would also be capable of photoproducing hydrogen gas, if they were supplied with a hydrogenase (which they lack) and if oxygen production, usually deleterious to hy- drogenase activity, were suppressed. Photoproduction of hydrogen gas by spinach chloroplasts supple- mented with bacterial hydrogenases was demonstrated by Mitsui and Paneque (25-2?) . When photoproduction of oxygen gas was suppres- sed and cysteine-DPIP (Fig. 2) was used instead of water as the electron donor system, spinach chloroplasts evolved hydrogen gas in the light. The photoproduction of hydrogen gas was accompan- ied by formation of ATP (Fig. 3). Ferredoxin and the equivalence of light and Hg for TPN reduc - tion . At first, photoproduction of H2 by chloroplasts was car- ried out with the aid of a hydrogenase isolated from Chroma tium (25,26) or Desulfovibrio desulfuricans (26). In both these cases photoproduction of H2 required the addition of methyl or benzyl viologen (24). However, with a crude hydrogenase from Clostrid - ium pasteurianum no addition of viologen dye was required for the photoproduction of hydrogen gas by chloroplasts (27). The cell- free extract of C. pasteurianum contained an electron carrier which brought about a photoproduction of hydrogen gas by chloro- plasts. This was consistent with the isolation by Mortenson et al. (28) of a natural electron- transferring factor in C. pasteur - ianum . named by them ferredoxin, which, in that organism, couples 199 Daniel I. Arnon pyruvic dehydrogenase with hydrogenase in the production of Hg from pyruvate. Aside from photoproduction of H2, Tagawa and Arnon (9) found that clostridial ferredoxin catalyzed the photoreduction of TPN by spinach chloroplasts without the participation of the "photo- synthetic pyridine nucleotide reductase" (PPNR) of San Pietro and Lang (29), which has become widely accepted as the specific chloroplast enzyme required for the photoreduction of TPN. More- over, in the presence of added bacterial ferredoxin and hydrogen- ase, ' isolated chloroplasts reduced TPN in the dark with hydrogen gas but without PPNR (see Fig. 5 in ref. 9). Thus, the enzymic apparatus of chloroplasts was found to be able to reduce TPN in- dependently of PPNR and light. These results indicated that the true pyridine nucleotide reductase enzyme of chloroplasts was not a component of the PPNR preparation but a component of the remaining chloroplast fraction. A re-examination of the TPN-reducing system of chloroplasts, des- cribed in more detail elsewhere (9,30), proved this interpreta- tion to be correct. The TPN reductase proper, which had also diaphorase (3I) and transhydrogenase (32,33) activities, was found to be localized in the flavoprotein fraction of chloro- plasts (9). The flavoprotein reductase, which was also isolated by Gewitz and Voelker (3^) and Davenport (35), was recently crys- tallized (Fig. h) by Shin et al. (36,30)- The enzyme reduces TPN either in the light or in the dark with H2 (plus hydrogenase) and requires in either case the collaboration of an electron carrier: bacterial or chloroplast ferredoxin. Chloroplast ferredoxin was isolated and crystallized by meth- ods similar to those used for bacterial ferredoxin (9). Chloro- plast ferredoxin proved to be a non-heme iron protein, localized in chloroplasts, and was similar to Clostridium ferredoxin in having a redox potential (E'q = -^32 mV, pH 7.55) close to that of the hydrogen electrode and in undergoing reversible oxidation- reduction that was measured by spectral changes (9). As shown in Table 3, the protein from spinach chloroplasts had an iron con- tent of 0.89 per cent, which, on the basis of a molecular weight of 13,000, indicates two atoms of iron per mole. Spinach ferredoxin proved to be the same substance as PPNR, the methaemoglobin reducing factor of Davenport, Hill and Whatley (37), the TPN-reducing factor of Arnon, Whatley and Allen (38) and the red enzyme of Gewitz and Voelker (3^) [see review (30)]. The presence of iron in PPNR or the red enzyme was recently re- 200 Daniel I. Arnon ported by Horio and Yamashita (39), Katoh and Takamiya (^0), Fry and San Pietro (4l) and Gewitz and Voelker (3^)- Fry and San Pietro (^1) found that the iron in this protein is associated with "labile" sulfide groups. The presence of "labile" sulfide groups in the protein was also independently found by Gewitz and Voelker (3^+) . Table 3 Iron Analysis of Spinach Ferredoxin (Tagawa, Chain and Arnon, 1963) Ferredoxin Fe found Per cent Minimum Used (mg) (UK) 8.6 Fe M.W. 1.0 0.86 6,490 2.0 17.9 0.90 6,200 3.7 33.3 0.89 6,260 7.^ 6k.k 0.87 6,i+20 The chemical similarities and the functional interchangeabil- ity of bacterial and chloroplast ferredoxin in the photoreduction of TPN (9) suggest that these two substances, although not iden- tical, belong to a family of ferredoxins. Ferredoxins appear to function as electron carriers that transfer to appropriate en- zyme systems the roost "reducing" electrons in cellular metabolism, that is, electrons at a potential of about -420 mV. These come from two sources: hydrogen gas (or substrates producing H^) or illuminated chloroplasts . The role of ferredoxins in photosynthetic electron transport and in utilization and production of hydrogen gas is diagramma- tically represented in Fig. 5' In this scheme, crystalline spinach ferredoxin (Fig. 6) was found to be replaceable by one of several crystalline ferredoxins: that from Clostridium pas - teurianum (9) and those from the photosynthetic bacterium Chroma tium (Fig. 7) and the blue-green alga, Nostoc muscorum (Fig. 8). Separation of the light and dark reactions in photoreduction of TPN . The recognition of the role of ferredoxin in the TPN- reducing system was followed by the physical separation of the photoreduction of TPN by illuminated chloroplasts into two steps: (1) a photochemical reduction of ferredoxin by chloroplasts in the absence of TPN, followed by (2) the dark reoxidation of re- duced ferredoxin by chloroplasts to which TPN was added (42). The results, summarized in Table 4, show that two moles of 201 Daniel I. Arnon reduced ferredoxin are required to reduce one mole of TPN. Thus, the oxidation-reduction of a ferredoxin molecule involves a trans- fer of a single electron. In the absence of evidence to the con- trary, it is attractive to assume that the photor eduction of ferredoxin is the terminal photochemical act of chloroplasts following photon capture--an act that involves a transfer of an electron from excited chlorophyll to the electron acceptor mole- cule in chloroplasts. Table h Stoichiometry of Photor eduction of Spinach Ferredoxin and its Subsequent Reoxidation by TPN in the Dark (10) umoles Ferredoxin (Fd) photoreduced 0.102 Fd reoxidized by TPN in the dark 0.106 TPN reduced 0.0^7 Ferredoxin and photophosphorvlation . If the photoreduction of ferredoxin by chloroplasts is the terminal photochemical act of chloroplasts, then it follows that photoreduction of ferredoxin should be a common feature of the electron flow pathways of either cyclic or noncyclic photophosphorylation. The two path- ways would differ in the electron acceptors beyond ferredoxin. In the case of noncyclic photophosphorylation, the electrons from ferredoxin would be transferred by the flavoprotein reductase to TPN, whereas in the case of cyclic photophosphorylation they would "cycle" back to "electron-deficient" chlorophyll molecules via a chain of endogenous electron carriers (6). A requirement for ferredoxin [then called "TPN-reducing factor (58)] for noncyclic photophosphorylation coupled with TPN reduc- tion was indeed already observed when this process was first dis- covered (3) and this requirement has since been further documen- ted (43-47). Evidence was also obtained several years ago (38) for a requirement of a "TPN-reducing factor" (i.e., ferredoxin) in what is now called pseudocyclic photophosphorylation (8) by chloroplasts. Moreover, Forti and Jagendorf (48) and Black et al. (49) found that under aerobic conditions ferredoxin ("PPNR") stimulates an endogenous photophosphorylation which proceeds in the absence of added cof actors, but is dependent on and consumes, molecular oxygen as the terminal electron acceptor. There was no evidence, however, that ferredoxin catalyzes an anaerobic, cyclic photophosphorylation, when oxygen evolution is effectively inhibited by the presence of CMU. Such evidence was 202 Daniel I. Arnon recently obtained by Tagawa et al. (11) who found that ferredoxin catalyzes an anaerobic cyclic photophosphorylation in chloroplasts which proceeds in the presence of CMU and without the addition of other cofactors. A notable feature of this endogenous, ferre- doxin-catalyzed cyclic photophosphorylation, which distinguishes it from other types of cyclic photophosphorylation in chloro- plasts, is its sensitivity to antimycin A and to low concentra- tions of dinitrophenol (11,12). Since, in mitochondria antimycin A inhibition is considered to be indicative of the participation of cytochrome b in electron transport (50>51)> the sensitivity to antimycin A suggests a possible participation of the cyto- chrome bg component of chloroplasts (52,53) in the ferredoxin- catalyzed cyclic photophosphorylation. A previously puzzling feature of cyclic photophosphorylation in isolated chloroplasts, a feature which distinguished it from cyclic photophosphorylation in bacterial chroma tophores, was a dependence on an added electron carrier such as vitamin K or phenazine methosulfate. A possible, though heretofore experi- mentally unsupported, explanation of this difference was that chloroplasts, but not chroma tophores, lost a soluble constituent in the process of isolation. The recent findings point to chloro- plast ferredoxin as being the water-soluble constituent of cyclic photophosphorylation which is, at least in part, lost from chlcto- p lasts when they are removed from the cell. However, it is still premature to say what role bacterial ferredoxins play in the mechanism of bacterial photophosphorylation. As previously mentioned, ferredoxin appears to be a junction in chloroplasts for the electron transport systems that lead to either cyclic or noncyclic photophosphorylation. Since we now know that TPN is required for noncyclic photophosphorylation but not for cyclic photophosphorylation, it follows that when the photoreduced ferredoxin is reoxidized by TPN ( via the flavopro- tein reductase), noncyclic photophosphorylation would result. When oxidized TPN is unavailable as an electron acceptor, the photoreduced ferredoxin would be reoxidized (directly or indirect- ly) by a cytochrome component of the grana, and cyclic photophos- phorylation would result. It is thus possible to envisage that the availability of TPN as an electron acceptor might serve as a physiological regulator between cyclic and noncyclic photophos- phorylation. Evidence for this view has recently been reported (11). Analysis of chloroplast reactions with monochromatic light . The photoreduction of ferredoxin and the resultant cyclic or 203 Daniel I. Arnon noncyclic photophosphorylation were found to be basically inde- pendent of photoproduction of oxygen by chlorop lasts (12). To demonstrate this independence it was necessary to use special ex- perimental devices such as inhibitors of oxygen evolution, anaer- obic conditions and, most recently, monochromatic light (12) at a wavelength (663 m^) which is absorbed by both chlorophylls a and b and at a wavelength (7O8 m^) which is absorbed by the chlorophyll a pigment system but not by chlorophyll b (5^). The main results may be summarized as follows. At 7O8 mu* i.e., at a wavelength at which light absorption by chlorophyll b was excluded, isolated chloroplasts were unable to use water as a hydrogen donor but retained the photoactivity which did not depend on water as a hydrogen (electron) donor. Thus, little oxygen evolution (Table 1 in ref . 12) was observed at 708 myx (see also ref. 55), but at this wavelength chloroplasts were able to photoreduce ferredoxin and sustain a ferredoxin-catalyzed cyclic photophosphorylation. (Contrary to other reports (56) we found no cyclic photophosphorylation at 708 m\i with phenazine methosulfate.) At 7O8 m^, chloroplasts were also able to sustain a noncyclic photophosphorylation coupled with TPN reduction but only when ascorbate-DPIP couple was supplied to replace water as the hydrogen donor system. The presence or absence of air had no special effect on either cyclic or noncyclic photophosphory- lation at 708 m\x. The photoactivity of isolated chloroplasts at 663 miji differed from that at 7O8 mn but was essentially the same as in white light. Both chlorophyll a and b were able to absorb light at this wavelength; water served as the electron donor for the photoproduction of ferredoxin, and the resulting reduction of TPN and noncyclic photophosphorylation was accompanied by oxygen evolution. At 663 mn, the presence or absence of oxygen had little effect on noncyclic photophosphorylation but had a marked effect on cyclic photophosphorylation catalyzed by ferredoxin. In the absence of oxygen, ferredoxin-catalyzed cyclic photophosphoryla- tion occurred only when electron transport from water was blocked by the addition of CMU. No addition of CMU was required for a ferredoxin-catalyzed cyclic photophosphorylation at 663 m^i in air or, it will be recalled, under either aerobic or anaerobic conditions at 708 m\x. Thus the presence of oxygen was necessary for ferredoxin-catalyzed cyclic photophosphorylation only when the flow of electrons from water remained open; no oxygen was necessary when the electron flow from water was blocked, either 204 Daniel I. Arnon by the use of an inhibitor (CMU) at 66'^ m\x or by the use of far- red monochromatic light (708 m^) . A possible explanation of these effects of oxygen will be given later. Roles of chlorophylls a and b in chloroplast electron trans - port . The results obtained with monochromatic light support and extend our earlier tentative conclusion (8) that the participa- tion of chlorophyll b is essential for oxygen evolution but not for TPN reduction and ATP formation. The diagram shown in Fig. 9 incorporates the new data into our 1961 scheme of electron transport in chlorop lasts (Fig. 3 in ref. 7). We envisage that noncyclic photophosphorylation by chloroplasts is coupled with an "uphill" flow of electrons from water (0H~) to TPN--an electron flow which is driven by two photochemical reactions working in series (cf. 57). The first photoreaction lifts electrons from the redox potential (at pH 7) of water-oxygen (E'q = 0.82 V) to that of plastoquinone (E^'^'O V) and requires the participation of chlorophyll b (henceforth re- ferred to as photoreaction B) . The second photoreaction lifts electrons from the level of cytochrome f (E'q = O.365 V) to that of ferredoxin (E'^ = -0.^3 V) and is driven by the chlorophyll a pigment system (henceforth referred to as photoreaction A). A "primary" phosphorylation site, common to both the cyclic and noncyclic photophosphorylation pathways, is considered to be coupled with a "downhill," dark electron transfer--one favored by the thermodynamic gradient--which joins the two photochemical reactions and probably involves a transfer of electrons from plastoquinone (Q) to the chloroplast cytochromes and thence to chlorophyll a (long black arrow in Fig. 9). In addition, the ferredoxin-catalyzed cyclic photophosphorylation is envisaged as having a least one more phosphorylation site, coupled with the electron transport sector that is marked in Fig. 9 by a broken line. Role of oxygen in cyclic photophosphorylation . The differ- ential oxygen effect on ferredoxin-catalyzed cyclic photophos- phorylation at 663 and 7O8 mn, mentioned previously, is explained by the following hypothesis. Noncyclic electron flow in chloro- plasts is a unidirectional electron transfer from water to TPN and is driven by both photoreactions B and A. The problem of "overreduction" of intermediate electron carriers does not arise as long as the terminal electron acceptor, TPN, is available. A different situation, however, arises in the case of cyclic photo- phosphorylation. To maintain a cyclic electron flow from reduced ferredoxin back to the electron transport chain, the intermediates 205 Daniel I. Arnon in the chain must be at least partly oxidized. If they are kept in a reduced form they cannot accept electrons from reduced fer- redoxin. Our hypothesis proposes that molecular oxygen acts as a redox "buffer" for the electron transport chain involved in cyclic photophosphorylation by chloroplasts . In the presence of oxygen, the electrons from water cannot overreduce the electron carriers in the electron transport chain. Without this "buffering" effect of oxygen, the flow of electrons from water (through photoreac- tion B) overreduces the components of the electron transport chain in chloroplasts and the endogenous cyclic photophosphory- lation via ferredoxin cannot proceed. The hypothesis just presented is supported by experiments with two beams of light (13), as illustrated in Fig. 10. Fig. 10 shows that, under anaerobic conditions, cyclic photophosphoryla- tion at 708 mn is inhibited by the addition of a second monochro- matic beam of light at 663 m^. This chromatic inhibition, which we attribute to overreduction by the 663 mn beam, occurred imme- diately when illumination by the combined 7O8 and 663 mpi beams was preceded by preillumination (under Ng) at 663 m[x (bottom curve. Fig. 10). Without an anaerobic preillumination treatment, the inhibitory effect of the 663 mji beam, added to the 708 m^ beam, was observed only after k min. Evidently an interval of time was needed to bring about a sufficient state of overreduc- tion by the 663 m^ beam. Photochemical activity of subcellular preparations of blue - green algae . In blue-green algae, phycobilins and not chlorophyll b constitute the "accessory" pigment system for chlorophyll a. The phycobilin pigments are water soluble and can be readily separated from chlorophyll a. Thus, cell-free preparations of blue-green algae offer attractive possibilities for testing the view that the photoproduction of oxygen depends on the accessory pigment system and can be experimentally separated from photo- phosphorylation and TPN reduction. Thomas and DeRover (58) have already reported that a loss of phycocyanin is associated with a loss of oxygen evolution by cell macerates of blue-green algae. Petrack and Lipmann (23) found that fragments of Anabaena cells which lost phycocyanin and the capacity for oxygen evolution still retained a capacity for cyclic photophosphorylation. Black et al. (59) showed that cell- free preparations of blue-green algae contain ferredoxin and are able to photoreduce TPN with either water or ascorbate-DPIP as 206 Eleetron Blue chroma tophores donor O2 evolved TPN reduced system lamoles Water ^.0 7.6 Ascorbate- DPIP 0.9 7.2 Daniel I. Arnon the electron donor system. Mitsui and Arnon (60) have compared the photochemical activity of two kinds of particles, "blue" and "green", from Nostoc . The blue particles (prepared with carbowax (6I) and dextrin) con- tained both chlorophyll a and phycocyanin whereas the green particles (prepared with carbowax only) contained little of the phycocyanin pigment. Table 5 Noncyclic Electron Flow in Blue and Green Nostoc Chroma tophores (60) Green chroma tophores O2 evolved TPN reduced (jmoles O.k 0.9 0.2 6.2 As shown in Tables 5 ar^d 6, both blue and green Nostoc parti- cles were able to carry out cyclic photophosphorylation with phenazine methosulfate and photoreduce TPN with the ascorbate- DPIP couple as the electron donor system. However, the green particles had only a feeble capacity for TPN reduction (and oxygen evolution) when water was the electron donor. By contrast^ the blue particles were able to use water effectively as the electron donor for a reduction of TPN and a coupled oxygen evolu- tion. Table 6 Cyclic Photophosphorylation in Blue and Green Nostoc Chromatophores (60) Blue chroma tophores Green chromatophores nmoles ATP formed Light 3.8 3.8 Dark O.5 O.k These results are consistent with the view that in particles of blue-green algae, as in chloroplasts, the accessory photo- synthetic pigment is required for photoproduction of oxygen, but not for photophosphorylation and TPN reduction per se . Concluding remarks . Extensive work from several laboratories, mainly with intact cells, has led to a now widely held view that photosynthesis in green plants involves the cooperation of at 207 Daniel I. Arnon least two pigment systems, each of which carries out separate partial reactions essential to the over-all process (see review, 62). Recent work with isolated chloroplasts, in which the par- tial reactions of photosynthesis have been experimentally separ- ated, has contributed biochemical evidence in support of the idea of at least two collaborative photoreactions. Photoproduc- tion of oxygen requires the participation of chlorophyll b but light absorption by this pigment is not essential for the photo- reduction of ferredoxin and the ensuing TPN reduction and photo- phosphorylation. These seem to be associated with the chloro- phyll a pigment system. Ferredoxin is assigned a key role in the energy conversion process in chloroplasts as an electron carrier in both cyclic and noncyclic photophosphorylation. A hypothesis is presented which attributes to molecular oxygen gas the func- tion of a redox "buffer" in cyclic photophosphorylation by chloroplasts. REFERENCES (1 (2 (3 (4 (5 (6 (7 (8 (9 (10 (11 (12 (13 Arnon, D. I., Allen, M. B. and Whatley, F. R., Nature Jj4, 394 (1954)- Arnon, D. I., Whatley, F. R. and Allen, M. B., J. Am. Chem. Soc. J6, 632^1 (195^) . Paper presented by D. I. Arnon before the Am. Chem. Soc. in New York, Sept. 11, 1957, and published by Arnon, D. I., Whatley, F. R. and Allen, M. B., Science _12I, 33O5 (i9!?8X Arnon, D. I., paper presented at the Cell Symposium, Amer. Assoc. Adv. Sci., Berkeley Meeting (195^); Science 122 . 9 (1955). Arnon, D. I., Nature J^, 10 (1959). Arnon, D. I., in Light and Life , ed. by McElroy, W. D. and Glass, B., 489 (Johns Hopkins Press, Baltimore, 1961). Losada, M., Whatley, F. R. and Arnon, D. I., Nature 190 , 606 (1961). Arnon, D. I., Losada, M. , Whatley, F. R. , Tsujimoto, H. Y. , Hall, D. 0. and Horton, A. A., Proc. Nat. Acad. Sci. (U.S.) 4l, 1314 (1961). Tagawa, K. and Arnon, D. I., Nature 19^, 537 (1962). Whatley, F. R. Acad. Sci Tagawa, K. , Acad. Sci Tagawa, K. , Acad. Sci Tagawa, K, Tagawa, K. , and Arnon, D. I., Proc. Nat, 1247 (1963) (U.S.) 49, 266 (1963) Tsujimoto, H. Y. and Arnon, D. (U.S.) 49, 567 (1963). Tsujimoto, H. Y. and Arnon, D. (U.S.) ^, 544 (1963). Tsujimoto, H. Y. and Arnon, D. , Proc . Na t . , Proc. Nat. , Nature 199, 208 Daniel I. Arnon (14) Jagendorf, A. T., Survey Biol. Progress k, l8l (1962). (15) Crane, F. L. , Plant Physiol. ^, 128 (1959). (16) Bishop, N. I., Proc. Nat. Acad. Sci. (U.S.) h^, 1696 (1959). (17) Krogmann, D. W., Biochem. Biophys. Res. Coratn. h, 275 (1961X (18) Arnon, D. I., and Horton, A. A., Acta Chemica Scand. JJ, S135 (1963). (19) Witt, H. T., Miiller, A. and Rumberg, B., Nature ISJ, 987 (1963). (20) Jagendorf, A. T. and Avron, M, J. Biol. Chem. 211, 277 (1958). (21) Hill, R. and Walker, D. A., Plant Physiol. ^, 240 (1959). (22) Mitsui, A., Paneque, A., and Arnon, D. I. Manuscript in preparation. (23) Petrack, B., and Lipmann, F. , in Light and Life , ed. by McElroy, W. D. and Glass, B., 621 (Johns Hopkins Press, Baltimore, 1961). (24) Peck, H. D., and Gest, H. , J. Bact. 21, 569 (1957). (25) Arnon, D. I., Mitsui, A. and Paneque, A., Science 134 . 1425 (1961). (26) Mitsui, A. and Arnon, D. I., Plant Physiol. 12 (Suppl.) iv (1962). (27) Paneque, A. and Arnon, D. I., Plant Physiol. 12 (Suppl.) iv (1962). (28) Mortenson, L. E., Valentine, R. C. and Carnahan, J. E., Biochem. Biophys. ReSo Coram. 2» ^^8 (1962). (29) San Pietro, A. and Lang, H. M. , J. Biol. Chem. 211, 211 (1958). (30) Shin, M. , Tagawa, K. and Arnon, D. I., Bioch. Z. HS, 84 (1963). (31) Avron, M. and Jagendorf, A. T., Arch. Biochem. 22, 17 (1957). (32) Keister, D. L., San Pietro, A. and Stolzenbach, F. E. , J. Biol. Chem. 211, 2989 (I960). (33) Keister, D. L. , San Pietro, A. and Stolzenbach, F. E., Arch. Biochem. 98, 235 (1962). (34) Gewitz, H. S. and Vtilker, W., Hoppe-Seyler Z. physiol. Chem. 110, 124 (1962). (35) Davenport, H. E. , Nature J^, I5I (1963). (36) Shin, M. , Tagawa, K. and Arnon, D. I., Fed. Proc. 22, 589 (1963). (37) Davenport, H. E., Hill, R. and What ley, F. R. , Proc. Roy. Soc. BI39 . 346 (1952); Davenport, H. E., Biochem. J. 21. 45P (1959). (38) Arnon, D. I., Whatley, F. R. and Allen, M. B., Nature I80. 182 (1957). (39) Horio, T. and Yamashita, T., Biochem. Biophys. Res. Comra. 9, 142 (1962). 209 Daniel I. Arnon (40) Katoh, S. and Takamiya, A., Biochem. Biophys. Res. Conm. 8, 310 (1962). (41) Fry, K. T., and San Pietro, A., Biochem. Biophys. Res. Coram. 9, 2l8 (1962). (42) Whatley, F. R. , Tagawa, K. and Arnon, D. I., Proc. Nat. Acad. Sci. (U.S.) 49, 266 (1963). (43) Jagendorf, A. T., Brookhaven Symposia in Biology, vol. 11, 236 (1958). (44) San Pietro, A., Brookhaven Symposia in Biology, vol. 11, 262 (1958). (45) Arnon, D. I., Whatley, F. R. and Allen, M. B., Biochim. Biophys. Acta ^, 4? (1959). (46) Davenport, H. E., Biochem. J. 21, 471 (I960). (47) Turner, J. F. , Black, C. C. and Gibbs, M. , J. Biol. Chem. 237 . 577 (1962). (48) Forti, G. and Jagendorf, A. T., Biochim. Biophys. Acta ^, 322 (1961). (49) Black, C. C., Fewson, C. A., Gibbs, M. , Keister, D. L. and San Pietro, A., Fed. Proc. 21, 398 (1962). (50) Chance, B. and Williams, C. R. , Advances in Enzym. _12» ^5 (1956). (51) Racker, E. , Advances in Enzym. 2^, 323 (1961). (52) Hill, R., Nature JJ4, 501 (1954). (53) Davenport, H. E. , Nature 1_£0, 1112 (1952). (54) French, C. S., in Handbuch der Pf lanzenphysiologie , ed. by Pirson, A., V. 5. p. 252 (Springer Verlag, Heidelberg, I960) . (55) Fork, D. C, Plant Physiol. ^, 323 (1963). (56) Kok, B. and Hoch, G. , in Light and Life , ed. by McElroy, W. D. and Glass, B., 397 (Johns Hopkins Press, Baltimore, 1961). (57) Wessels, J. S. C, Proc. Roy. Soc. B157 . 345 (1963). (58) Thomas, J. B., and DeRover , W., Biochim. Biophys. Acta j£, 391 (1955) (59) Black, C. C, Fewson, C. A. and Gibbs, M. , Nature 198, 88 (1963). (60) Mitsui, A. and Arnon, D. I. Manuscript in preparation. (61) McClendon, J. H. and Blinks, L. R. , Nature i20. 577 (1962). (62) Smith, J. H. C. and French, C. S., Ann Rev. Plant Physiol. 14, 181 (1963). 210 Daniel I. Arnon Hydrogenose from Desu^fovibrb Fig. 1. Effect of extracting and restoring plastoquinone to spinach chlorop lasts on photo- production of TPN with either water (OH') or ascorbate as electron donor (l8). 30 40 50 minutes Fig. 2. Photoproduction of hydrogen gas by spinach chlor- oplasts supplemented by a hydrogenase from Desulf qvibrio desulfuricans. H2 was identi- fied by adsorption on palla- dium asbestos (Pd-asb) (22,25> "T I r ' 1 ' 1 T- Hydrogenic photophosphorylotion 20 30 minutes Fig. 3* Photoproduction of hydrogen coupled with photo- phosphorylation by spinach chloroplasts supplemented with a hydrogenase from Desulfovibrio desulfuricans hydrogenase flovoprotein Fig. 5. Diagrammatic repre- sentation of the role of ferredoxin as an electron carrier (9) . (26,27) 211 Daniel I. Arnon \-^- • s .) 1 ' f it ^ k ^, <> / ^- Fig. ^. Microphotograph of crystalline f erredoxin-TPN reductase (30) • Fig. 6. Microphotograph of crystalline spinach ferre- doxin (9) . "T.\ \> Igep Fig. 7. Microphotograph of crystalline Chroma tium ferredoxin (Bachofen, Oda and Arnon, 1963). 212 Daniel I. Arnon 150 I Fig. 8. Microphotograph of crystalline ferredoxin from Nostoc (60) . TPN absent gos phose : Ng 708 m/t only 708 -I- 663 66 3-*' 708-^-663 12 Fig. 10. Effect of additional illumination at 663 ni|i on ferredoxin-dependent cyclic photophosphorylation at 7O8 n>n under anaerobic conditions (13). Eo,PH7 TPN LIGHT OH-(IO"^M) Fig. 9. Scheme for two photoreactions and their relation to cyclic and noncyclic photophosphorylation in chloroplasts (13) • CHARACTERISTICS OF TRITIUM INCORPORATION INTO ILLUMINATED CHLOROPLASTS Brlaji Colmaii and Wolf Vishniac It has been shown in previous experiments (1,2), that chloroplasts, incubated with tritiim labeled water in the light, incorporate tritium into chlorophyll a. More tritium is incorp- orated in light than in the dark and little incorporation takes place with boiled chloroplasts. The tritium label of the chloro -phyll was found to be stable to acid but unstable to alkali, which suggested that the chlorophyll was labeled at the C-10 position. Quantitative evaluation of the data was rendered difficult by the unexplained loss of radioactivity from samples of purif- ied chlorophyll a stored in organic solvents at -15°. This loss has been found to be much more rapid when samples are dried prior to counting in a windowless gas flow counter, and makes it impos- sible to follow the kinetics of photo synthetic incorporation. In order to minimize the loss of activity, samples were placed in the counter while still in solution and allowed to dry in the stream of counting gas. By this procedure the rate at which radio -activity dissappeared from labeled chlorophyll could be followed. Labeled chlorophyll a was obtained from chloroplasts which had been illuminated in suspension together with tritiated water. The chloroplasts were extracted with acetone, the acetone fraction transferred to heptane, and chlorophyll a rapidly separated from the other pigments by chromatography on thin layers of powdered sugar. (Sucrose is applied to glass plates as a suspension in methanol; the small amount of sucrose that dissolves in the meth -anol serves to bind the remainder to the plate). Figure 1, curve 2, shows the loss of radioactivity from labeled chlorophyll a isolated from chloroplasts. Curve 1. illustrates a similar decay, but here the chlorophyll was labeled chemically. Purified chloro -phyll a in ether was treated with tritiated water and 2% pyridine at laboratory temperature. The enol-keto tautomerization resulting from this treatment with base, presimjably leads to the replacement of the protium on C-10 with tritivma. 213 214 Brian Colman and Wolf Vishniac Chlorophyll a treated with tritiated water under neutral conditions, that is in boiling ether, incorporates only a fraction of the activity of that treated under basic conditions and the resultant label is relatively stable (Fig. 2, curve 2, cf. curve 1.)- Phaeophytin a labeled by each of the two procedures shows the same types of decay as chlorophyll a. During the loss of radioactivity there is no chemical change in the chlorophyll or phaeophytin as judged by spectroscopic observation. Allomerized chlorophyll a, prepared by the oxidation of chlorophyll a in methanol with gaseous oxygen, was treated with tritiated water in 2fo pyridine in ether. The small amount of radioactivity incorporated and its stability (Fig. 2, curve 3) are consistent with the notion that the tritium label of the intact chlorophyll a is located on C-10. Heating allomerized chlorophyll a in ether in the presence of TOH did not lead to significant labeling either. The loss of radioactivity from chlorophyll does not reflect evaporation of tritiated water originally introduced into the reaction mixture. The procedure for the isolation of the chloro- phyll precludes the retention of any such water. Instead the disappearance of tritium suggests that it is either given off as water vapor or as hydrogen gas resulting from an unknown exchange reaction. The light-dependant incorporation of tritiim into chloroplast components is affected by CMU (p-chlorophenyl-l,l-dimethyl urea). CMJ has been reported (3,i+) to inhibit the Hill reaction, the reduction of TPN and photo synthetic phosphorylation in chloroplast preparations. Pea chloroplasts were incubated with tritiated water under the conditions described in Table 1. At the end of the experiment, the chloroplasts were extracted with acetone and the acetone fraction transferred to heptane. The residue of the chloro -plasts was then succesively extracted with isooctane, methanol and water. The data in Table 1. confirm the earlier results that more activity is taken up by the component pigments of the heptajie fraction in the light than in the dark, and also show that this light stimulated uptake is inhibited by CMU. The accximulation of tritium in the isooctane fraction in the presence of CMU is interesting. This fraction contains at least two quinones which incorporate tritium from water in the light. While tritium incorporated into a quinone would be expected to be readily exchangeable with water, it had been suggested to us by Dr. Calvin that the retained activity could result from the rearrangement of a suitable reduced quinone, for example, reduced Co Q, to form a chroman ring with the concomitant migration of 215 Brian Colman and Wolf Vishnlac the hydroxyl tritium to the side chain. Upon reoxidation the label would therefore be located on the phytol side chain. In a model experiment, purified Co Q from Chromatiiom was reduced with NaBHj^ in the presence of TOH, reoxidised with silver oxide, and reiso- lated. The retention of tritium by the Co Q to the extent of half the specific activity of the TOH, confirmed the possibility of labeling the q.uinone by the mechanism which Dr. Calvin had proposed. The methanol fraction contains six, and the water fraction two chromatographically separable, but unidentified compounds, all of which incorporate more tritium in the light than in the dark. Although we cannot explain the accumulation of tritium in the isooctane fraction, the inhibition by CMU of tritium incorporation into photo synthetic pigments is a further indication that tritium can serve as a useful tracer in the investigation of photo synthetic events. Table 1. Incorporation of tritium by chloroplasts in the presence of CMU. Fraction Dark T.ight Heptane No CMU +CMU No CMU +CMU 13,8^^0 16, 400 8U,U00 10,800 Isooctane 9,1+50 18,350 17,700 31+9,000 Methanol 165,000 150,500 358,000 i+17,000 Water 9,350 15,550 16, 300 16,1+00 Figures in c.p.m. The reaction mixture contained chloroplasts (5 mg chlorophyll) in sucrose-potassivma phosphate, pH 7, 100 mC H^OH, and CMU (lO"%) where indicated. Total volume 1.0 ml. Temperature 10°C. 216 Brian Colman and Wolf Vishniac 1000 -, JOO- locr 50 - -T" I Klnutas Figure 1. Loss of radioactivity from dry chlorophyll samples. Cxarve 1: Chlorophyll a treated with H-^OH in ether plus 2io pyridine. Curve 2: Chlorophyll a isolated from chloroplasts illuminated in presence of H OH. Ordinate in c.p.m. per sample counted. 217 Brian Colinan and Wolf Vishniac 1000 500- 100- 50 - Minutes 50 100 Figure 2. Characteristics of tritiiom labeling of chlorophyll a and allomerized chlorophyll a. ^ Curve 1: Purified chlorophyll a treated with H OH in ether + 2^ pyridine. Curve 2: Purified chlorophyll a boiled in ether with H-^OH Curve 3: Allomerized chlorophyll a treated with H^OH in ether + 2fo pyridine. Ordinate in c.p.m. per sample counted. 218 Brian CoJjnaxi and Wolf Vlshniac References 1. Vishniac,W. and I.A.Rose. Nature, 182 , IO89-9O, (1958). 2. Vlshniac, W. Proc. 5th. Intern. Cong. Bioch em. Vol. VI. 'Mechanism of Photosynthesis' (Ed. H.Tamiya) pp 251-252. Pergamon Press. New York.(l963) 3. Jagendorf,A. and M.Margulies. Arch.Biochem.Biophys., 90, 184-195/(1960). k. Bamberger,E.S., C.C. Black, C.A.Fewson, and M.Gibbs. Plant Physiol. 38, 1+83-^87, (1963). INDOPHENOL DYES: CATALYSTS AND UNCOUPLERS OF PHOTOPHOSPHORYLATION Donald L. Kelster The indophenol dyes, especially 2 ,6-dlchloroindophenol (DCI) and 2,3' ,6-trlchloroindophenol (TCI) have been known to be .^. excellent oxidants for measuring the Hill reaction since ig'^S^ '> In 195'iC2), the demonstration was made that isolated chloroplasts under the influence of light could phosphorylate ADP and in 1958'35 it was discovered that the reduction of TPN was accom- panied by ATP formation. Subsequently it has been shown that ATP formation was coupled to the reduction of other Hill oxidants including ferricyanide , FMN and cytochrome c. However, it has been reported in the literature that the reduction of the iodo- phenol dyes was not accompanied by photophosphorylation^ »^ K This view has led to the postulation that these compounds were reduced by some component of the electron transport chain prior to the site of phosphorylation. Earlier this year three widely separated groups, Shen et gj..^°^ in China, Gromet-Elhanan and Avron^'^ in Israel, and our own^^^ , independently discovered that phosphorylation was coupled to the reduction of indophenols. These compounds are potent uncouplers of photophosphorylation in their oxidized form and therein lies the reason that the phosphorylation coupled to their reduction was previously overlooked, METHODS Unless otherwise specified each reaction mixture per 3 ml contained the following in umoles: Tris, 150; ADP, 2; MgClp, 5; ^^Pi, 2-10; and chlorophyll as noted. After illumination the reactions were terminated by the addition of 0.3 ml of 50% trichloroacetic acid (TCA) and uptake of 32pi determined as previously noted^°\ Anaerobic experiments were performed in Warburg flasks or in 50 ml Erlenmeyer flasks fitted for flushing with gas. The flasks were flushed with a strong stream of argon and shaken throughout the experiment. Ten minutes of flushing was found sufficient to remove essentially all the oxygen. Where reduced DCI or TCI was added, it was reduced with hydrogen using platinum asbestos as catalyst thus eliminating any other reducing agents. One drop of bromine water was added to samples for J'^Pi uptake determinations after the addition of TCA to oxidize the reduced indophenols and prevent their interference with the determination. Cytochrome c was the Type III preparation of the Sigma Chemical Company. 219 220 Donald L. Keister PHOSPHORYLATION COUPLED TO THE REDUCTION OF INDOPHENOLS The phosphorylation coupled to the reduction of the indophenol dyes was demonstrated by all three of the above mentioned groups by using very low concentrations of dye and measuring the small amount of ATP formed by the incorporation of high specific activity ^^?i into ATP during short illumination periods. Gromet-Elhanan and Avron^7) using 5 x lO-^M DCI and short illumi- nation periods measured ATP formation at rates up to I63 ymoles/mg chlorophyll/hr. They further demonstrated that the reduction of DCI was stimulated by the inclusion of a phosphate acceptor system (ADP, Mg"*"*" , and Pi) or by ammonium chloride, a well-known uncoupler of photophosphorylation. These criteria place this compound in the same category as ferrlcyanide^y ^ and tpn <>■'■"'' as Hill oxidants. Shen et_ al. ^^^ using essentially the same techniques but with higher light intensities demonstrated that rates of ATP formation up to 6^40 ymoles/mg chlorophyll/hr could be obtained with 1.28 X 10-5M DCI with an ATP:2e- ratio of 1.0. This is the highest rate yet reported for a non-cyclic phosphorylation. Keister^ ^^ first noted that TCI would catalyze ATP formation while studying the photoreduction of cytochrome £ by chloro- plasts^ll''. This is illustrated in Fig. 1. o 500 UJ o o llJ (c 400 z o i 300 o o 200 UJ -J o 2 100 4 8 12 16 20 24 TCIP CONCENTRATION M x 10^ Fig. 1. The effect of TCI concentration on phosphorylation coupled to cytochrome c reduction. The reaction mixture, pH 7.5, contained 3 ymoles KCN, 3.8 mg cytochrome c, and 10 wg chlorophyll in addition to those described in methods. Illumination was for 2 min with 5000 ft. candles of light. In this system TCI was reduced photochemically and oxidized by cytochrome £. Thus, the TCI remains primarily in the oxidized 221 Donald L. Kelster form. Under these conditions ATP formation was observed to accompany cytochrome c reduction with an ATP:2e- ratio that approached 1.0 at low TCI concentrations (3 x 10"°M). INDOPHENOLS AS UNCOUPLERS OF PHOTOPHOSPHORYLATION Fig. 1 also illustrates the uncoupling effect of TCI. As the TCI concentration was increased the rate of cytochrome £ reduction was stimulated while ATP formation was Inhibited until at 2.5 X 10"5m TCI, very little ATP synthesis occurred. This clearly demonstrates the uncoupling effect of the oxidized indophenols. A further demonstration of this uncoupling effect is illustrated in Fig. 2 which shows the inhibition of cyclic photophosphorylation by TCI. This experiment was performed under argon in the presence of 10~5m £_-chlorophenyl-l ,1-dimethylurea (CMU) to inhibit dye reduction and thus maintain the TCI in its oxidized form. Under these conditions, TCI increasingly inhibits cyclic phosphorylation with almost complete inhibition by 10 M TCI. The curve labelled "reduced TCI" (Fig. 2) demonstrates that the reduced form of the dye has no inhibiting effect. Avron and Jagendorf^^^) first observed the inhibition of cyclic phosphoryla- tion by TCI but at that time concluded that the dye was interfer- ing with electron transport reactions rather than uncoupling or inhibiting phosphate transfers. i 2 4 6 8 TCIP CONCENTRATION MkIO' i Fig. 2. The inhibition of cyclic phosphorylation by TCI. The reaction mixture, pH 8.2, contained 0.06 ymole pyocyanin, 0.03 umole CMU, and 22 ug chlorophyll In addition to those described. The reaction was run in Warburg vessels with pyocyanin, TCI, and CMU being tipped in after flushing for 10 min with argon. The control rate of phosphorylation during a 6 min illumination period with pyocyanin alone was 21i) umoles/mg/hr which was inhibited 20?5 by CMU. i 222 Donald L. Keister Shen et al. ^^^ have also concluded that the indophenols are uncouplers of phosphorylation from two types of experiments. They have previously shown that ATP formation and light-induced electron transport could be separated by pre-illuminating chloro- plasts in the presence of phenazine methosulfate (PMS) but without AD? and Pi. After a short but intense illumination, the addition of ADP and 32pi results in the formation of ATP^-^^-". When increasing concentrations of DCI was added along with the ADP and Pi, the ATP formation was inhibited up to 85% by 10" M DCI while the reduced dye had no effect. This type of inhibition was interpreted as uncoupling. In addition they have measured ATP formation and electron transport by chloroplasts in a system containing both DCI and ferricyanide. With only ferricyanide present an ATP:2e~ ratio of 1.26 was measured. With increasing concentrations of DCI the ratio decreased until complete inhibi- tion of ATP formation was observed with lO'^'M DCI. Ferricyanide reduction was stimulated 25% by this concentration of dye. Fig. 3 illustrates this uncoupling effect of both DCI and TCI on ferricyanide coupled photophosphorylation. Complete uncoupling was observed with 7 x 10-°M TCI while almost a ten-fold higher concentration of DCI was required for complete uncoupling. 3'-chloroindophenol was also shown to be an uncoupler at higher concent rat ions . ^©DCI '-a ZxlO-^M 4 1 10"= M X 6.7xI0-5m ^.-^J ©- 2 4 6 8 CONCENTRATION, M X 10' 2 3 4 MINUTES Fig. 3. ( ferricyan mixture , and 100 u in method Fig. n. ( formation 1.23 X 10 those des samples w TCI reduc Left) The uncounline effect of indophenols on ide coupled phosphorylation. The reaction pK B.O, contained ? umoles of ferricyanide g chlorophyll in addition to those described s. Right) The time course of TCI reduction and ATP . The reaction mixture, pH 8.0, contained "'^M TCI and 112 ug chlorophyll in addition to cribed in methods per 3 ml of solution. One ml ere removed for measurement of J'^Pi uptake and t ion . 223 Donald L, Keister Losada e^ a_l. ^-'-^^ first reported that when indophenols were added to a reaction containing ferricyanide the rate of oxygen evolution was accelerated, whereas the coupled phosphorylation was inhibited. Since they also found that there was no inhibi- tion of ATP formation coupled to TPN reduction by the same concentrations of dye (in fact there was a stimulation), they proposed that the indophenols were the direct electron acceptors, not coupled to phosphorylation, which were reduced at a point in the electron transport chain prior to the site of phosphorylation. In the TPN system the indophenols were rapidly reduced and not reoxidized by TPN (the reduced dyes are not uncouplers). Also, oxygen evolution can be blocked by CMU, and ATP formation and TPN reduction restored by reduced dye and ascorbate ^-'•5) , thus eliminating the requirement for the photooxidation of water. On the basis of these observations they proposed a separation of the photooxidation of water and non-cyclic phosphorylation into two distinct photochemical reactions. Although it appears that these are indeed two separate reactions, the use of the indophenol dyes to separate them was not valid since the above results demon- strated that there was ATP formation coupled to their reduction and that they are potent uncouplers. Shen et_ al_. (") have even cast some doubt on whether in a system containing both ferricyanide and indophenol, that the dye is the direct electron acceptor. Taking advantage of the differential inhibition of DCI and ferricyanide reduction by hydroxyquinoline N-oxide^^°) (HOQNO), they demonstrated that in a system containing both ferricyanide and DCI, the degree of inhibition by HOQNO corresponded to the degree of inhibition observed with ferricyanide alone and not to that observed with DCI alone. Thus it appears that in the DCI-ferricyanide system, ferricyanide is the direct electron acceptor and the DCI an inert uncoupler. CYCLIC PHOTOPHOSPHORYLATION The ability of reduced indophenols to reduce a photochemical oxidant and thus restore the reduction processes in chloroplasts , in which the normal process has been blocked by inhibiting oxygen evolution either with CMU or aging' -^^^ has been established. From these results it could have been postulated that the reduced indophenols should catalyze a cyclic electron transport since they can be both oxidized and reduced by chloroplasts. This was first demonstrated by Trebst and Eck'-^^; who reported that reduced DCI and TCI catalyzed a phosphorylation by chloroplasts that was not inhibited by DCMU. They interpreted this as a cyclic phosphorylation. These results were later confirmed by three laboratories (°~°) . Krogmann and Vennesland^^ ) earlier had reported that the indophenol dyes could mediate an oxygen dependent photophosphory- lation and suggested that the phosphorylation occurred upon the photooxidation of the dye and not during the reduction. This suggestion was based upon: 1) reduced dyes catalyzed the ATP formation; 2) the phosphorylation was inhibited by a nitrogen atmosphere; and 3) DCMU inhibition could be reversed by reducing 224 Donald L. Keister the indophenol prior to illumination. The observations can all be explained by cyclic phosphorylation. Fig, H illustrates a time course of the reduction of TCI and ATP formation under aerobic conditions. There was no ATP forma- tion in this experiment until the reduction of dye had almost reached completion. The rate of phosphorylation was the greatest Just after the completion of dye reduction and was proportional to time thereafter. The ATP formation in Krogmann and Vennesland's experiments was measured only at a single time interval, and thus the inhibition of ATP formation before the dye was reduced was overlooked. The concentration dependence of phosphorylation with DCI and TCI is shown in Fig. 5. Approximately a 3-fold higher concentra- tion was required for optimum ATP formation under argon than with an air atmosphere. It is to be noted that concentrations optimal for aerobic phosphorylation catalyze very little anaerobic phosphorylation and thus would appear to be inhibited by removing oxygen. One explanation for the different optimal concentrations is that the reoxidation of the dye by the chloroplast is the rate limiting step in the anaerobic system and the active site for the oxidation has a lower affinity for the dye than does the active site for the reduction. Since the autooxidation of the indophenols by oxygen is rather marked at alkaline pH's, this source of oxidized dye would enhance the turnover of dye in the aerobic system thus shifting the optimal concentration toward the lower optimal concentration of the reducing system. Therefore, a lower concentration would be required for the aerobic phosphorylation. The effect of light intensity on ATP formation with reduced indophenols as compared with pyocyanln and FMN is illustrated in Fig. 6. The FMN and pyocyanin curves are identical for both the aerobic and anaerobic systems, therefore, only one curve was drawn for each. The FMN system became saturated at fairly low light intensities as has been previously reported for fmN^^"^ and TPNH^^9; which catalyze non-cyclic phosphorylation, whereas the pyocyanin which catalyzes cyclic phosphorylation did not saturate even at very high light intensities^^^ ) . Fig. 6 demonstrates that under aerobic conditions with reduced DCI as catalyst of photophosphorylation the reaction became saturated at low light intensities similar to that of FMN, whereas under argon the curve is similar to that of pyocyanin which was not saturated with light. These results are indicative that anaerobically the indophenols catalyze a cyclic phosphorylation similar to that of pyocyanin, whereas with oxygen present the reaction appears to be of the non-cyclic type. However, in view of other results it is more probable that it is a combination of cyclic and non- cyclic phosphorylation. 225 Donald L. Kelster 120 100 I 80 o XT. O < (O o 0.8 PYOCYANIN ,.';J^ ^^^\^r^ ^.'•'V^DCI ,.*'''y/^ ANAEROBIC 0.6 '''*\y^ ,'^ y^ oci ^ •f't/' "^ AEROBIC /v/y 0.4 - © ^ - / /i ^A- FMN ^"-^A 2 r ^ n 1 . . . , , , 2 4 6 8 INDOPHENOL CONC, MxlO 10 12 4 Fig. 5 4000 8000 FOOT CANDLES Fig. 6 12000 Fig, 5. The effect of reduced indophenol concentration on aerobic and anaerobic phosphorylation. The reaction was illuminated in 50 ml Erlenmeyer flasks fitted for flushing with gas. Illumination was provided by 75 watt photoflood lamps which provided 5^00 ft. candles. Curves a and b, gas phase is air. Curves c and d, gas phase is argon. Fig. 6, The effect of light intensity on phosphorylation catalyzed by DCI, pyocyanin, and FMN. The reaction mixture contained either 2.6 x IQ-'^M DCI, IQ-'^M FMN, or 10~5m pyocyanin and 80 ug chlorophyll in addition to those reagents described in methods. The reaction was illuminated from underneath in Warburg vessels. The light intensity was varied by changing the distance of the light source from the flask and by inserting wire screens. Anaerobic flasks were flushed for 10 mln with argon before illumination. Illumination time was varied so that curves were comparable, CONCLUSIONS The indophenol dyes have now been shown to catalyze two types of photophosphorylation; a non-cyclic phosphorylation coupled to their reduction, thus placing them in the same category as ferricyanide as a Hill oxidant; and a cyclic photophosphorylation similar to that catalyzed by PMS and pyocyanin. In addition they act as a mediator to supply electrons from ascorbate and thus restore TPN reduction by a system in which oxygen evolution has been blocked or destroyed^ ■'■5) . It appears probable that this site of action and the site where the dye is reoxidized during cyclic phosphorylation are the same. The indophenols were also demonstrated to be potent uncouplers of photophosphorylation at concentrations normally used to * \ 226 Donald L. Kelster measure the Hill reaction. It is this property of the dyes which led previous investigators to believe that there was no ATP formation coupled to their reduction and thus postulate by necessity that the dyes intercept electrons at some point in the electron transport chain prior to the site of phosphorylation. The uncoupling action of the dyes probably accounts for the observations of Whittingham and Bishop'' ^\ who recently observed that the time between flashes of light was considerably shorter for optimum production of oxygen with DCI than with ferricyanide as the electron acceptor. They suggested, therefore, that production of oxygen by reduction of dye did not proceed through the same thermal reaction (phosphorylation) required for ferri- cyanide reduction. They also found that if the ferricyanide reduction was uncoupled from phosphorylation by ammonium, ions , the optimum time between flashes was of the same magnitude as with DCI. Witt and coworkers ^'^'^ ^ observed that DCI accelerates the decay of "Type 2" absorption increases at 515 my in chloro- plasts. This decay is a thermochemical process, and it is likely that the acceleration of decay can now be explained by the uncoupling effect of the indophenols. It is of interest that Low et al. ^ ^^ have reported that oxidized but not reduced DCI inhibited Pi-ATP exchange reactions in rat liver mitochondria. Hill and V/alker'^^^ several years ago observed that, "The phosphorylation reaction itself now appears as a part of normal photochemically induced H-transfer." They also observed, "It does not follow. . .that all substances capable of being reduced by illuminated chloroplasts prepara- tions would be capable of initiating phosphorylation. The coupling between reduction and phosphorylation can be abolished and an active agent can become inhibitory at higher concentrations so that affinities relating to the chloroplast system have to be in a suitable range." Contribution No. 13^^ of the Charles F. Kettering Research Laboratory. 227 Donald L. Keister PHOTOSYNTHETIC PHOSPHORYLATION IN THE PRESENCE OF NATURALLY OCCURRING SUBSTANCES C. C. Black and A. San Pietro After the initial observation of photosynthetic phosphorylation i^'^i it was demonstrated that many common laboratory redox chemicals support photophosphorylation. A list of these chernicals would include: methyl phenazonium methosulfate, or pyocyanine ^3-5). ferricyanide ^6); viologen dyes (^' ''' ^); anthraquinone ^^'; dimethyl safranin sulphonate '^^>; methylene blue (5); indigo carmine (^); and indophenol dyes (9-i2)_ -phe natural occur- rence, hence physiological role, of these chemicals is doubtful. The following substances occur naturally, may have a role of physiological importance, and have been shown to support photophosphorylation: vitamin K "13). FMN (1^): riboflavin (l^); NADP (^'^^^ plastoquinone ^ 'l. allagochrome (i^h a "flavone- type" compound (^^): cytochrome c ^ ^>; and photosynthetic pyridine nucleotide reductase (PPNR)^^"^ . Although these sub- stances occur in nature and may play a physiological role, it appears that they are involved in the electron transfer pathway rather than the phosphory- lation process. Since scant experimental data was available concerning the components involved in the phosphorylation mechanism, research was undertaken to isolate a naturally occurring substance or substances involved in photophos- phorylation. Evidence will be presented which indicates that we have successfully obtained a new naturally occurring catalyst of photosynthetic phosphorylation. The catalyst(s) has been detected in all types of photo- synthetic organisms, and the catalyst from one organism has been shown to initiate photophosphorylation with both chloroplast and chromatophore frag- ments. We have tentatively assigned the name phosphodoxin to the catalyst. ISOLATION AND DISTRIBUTION Acetone powders of intact spinach chloroplasts were the first source of phosphodoxin ^^^\ We soon learned that phosphodoxin was not destroyed by heating at 100° C for periods up to 30 minutes; therefore, whole leaves or whole cells extracted with boiling water proved to be convenient sources of phosphodoxin. Most of the work in this paper will deal with spinach phos- phodoxin since it has been studied more exhaustively than phosphodoxin from other photosynthetic organisms. More extensive data relating to the activity of phosphodoxin isolated from photosynthetic bacteria can be obtained in references 21 and 22. A partial list of the photosynthetic organisms in which phosphodoxin has been detected is given in Table 1, along with the 228 229 C. C. Black and A. San Pietro pmoles of ATP produced. Phosphodoxin appears to be ubiquitous among photosynthetic organisms. Table 1 Photosynthetic Organisms Containing Phosphodoxin jjmoles of ATP/mg Source chlorophyll/hr. Spinach, chloroplasts 196 Spinach, leaves 220 Phormidium luridum , whole cells 49 Tribonema aequale , whole cells 27 Euglena gracilis , whole cells 25 Chlorella pyrenoidosa , whole cells 210 Rhodospirillum rubrum , chromatophores 54 Rhodospirillum rubrum , whole cells 50 Chromatium , strain D, chromatophores 56 Chromatium , strain D, whole cells 115 All preparations assayed with spinach chloroplasts Spinach phosphodoxin was purified by acetone fractionation and paper chromatography ^^^>. Absorption spectra of spinach phosphodoxin is given in Fig. 1. The absorption spectrum shifts with pH, in contrast to the fluorescence activation spectra, which does not show a pronounced shift with pH (Fig. 2). Since the intensity of the fluorescence spectrum is pH- dependent but the activation spectrum is not (Fig. 2), it appears that the alkaline form of spinach phosphodoxin may be the fluorescent type. Alkaline solutions of spinach phosphodoxin are yellow, while acid solutions are nearly colorless. The fluorescent maximum was at 440 m^ and the activa- tion maximum was at 358 mp. Aqueous solutions of spinach phosphodoxin are stable at 4° C when stored near neutrality. Boiling phosphodoxin for 10 minutes in N HCl does not affect its activity, whereas boiling in N NaOH inactivates the phospho- doxin. Irradiation by ultraviolet light does not alter its activity. Spinach phosphodoxin does not contain a functional metal (23). At the present stage of our research, we do not know the structure of phosphodoxin, nor are we certain that each photosynthetic organism contains the same compound. 230 C. C. Black and A. San Pietro 240 280 320 360 400 WAVELENGTH m/. 440 480 • ACTIVATION ^ r FLUORESCENCE / W \ IM Phosphote \ Buffer pH 7.0 /'xY-'" . \^ / /' A \ \ IN HCI ^^V_ 350 450 WAVELENGTH, m/i Fig. 1. (Left) Lower portion: absorption spectra of spinach phosphodoxin. Solid line, 0. 1 N NaOH; broken line, 0. 1 N HCI. Upper portion: difference spectrum. Fig. 2. (Right) Activation and fluorescence spectra of spinach phosphodoxin. BIOLOGICAL CHARACTERISTICS The endogenous spinach chloroplast fragment photophosphorylation (0. 5 to 3 umoles of ATP/mg chlorophyll/hr) is stimulated over 200-fold by the addition of spinach phosphodoxin (Fig. 3). As indicated in Fig. 4, this aerobic reaction is linear for short periods of illumination. In an earlier publication ^^^^ the reaction was only 60 per cent inhibited under nitrogen. Further work with prepurified nitrogen indicates that the reaction with spinach chloroplast fragments definitely requires aerobic conditions (Fig. 4). It should be noted that this is in contrast to the photo- phosphorylation catalyzed by Rhodospirillum rubrum phosphodoxin with R. rubrum chroma tophore fragments, which is unaffected by anaerobiosis^ ' >. Photophosphorylation with spinach chloroplast fragments plus spinach phosphodoxin has been shown: to be linear with chlorophyll up to 40 ^grams of chlorophyll per ml; to have a pH optimum in Tris-HCl buffer between 7. 4 and 7.8; to require a divalent ion for maximum activity; to respond in a sigmoid fashion to increasing intensity of white light with a distinct lag up to 100 foot-candles, reaching saturation between 1000 and 2000 foot-candles; and to be unaffected by PPNR, pyridine nucleotide transhydrogenase, and the antibody to the transhydrogenase. From Fig. 5 it can be seen that photo- phosphorylation catalyzed by spinach phosphodoxin is sensitive to the usual inhibitors of photophosphorylation. In addition to these inhibitors 50% 231 C. C. Black and A. San Pietro 200 0. ^ 150 < -I E I 100- o _i o 2 .05 I .15 ml of PHOSPHODOXIN 20 . AIR .15 /^•' • * ■ .10 / 05 No .ARGON ^e-0-.--o..--.- ^ I 10 15 MINUTES 20 Fig. 3. (Left) Effect of concentrations of spinach phosphodoxin on photophosphorylation with spinach chloroplast fragments. Fig. 4. (Right) Time course and effect of anaerobic conditions on photophosphorylation with spinach chloroplast fragments plus spinach phosphodoxin. inhibition was obtained with the following compounds at t he indicated concen- trations: atebrin, 10'^ M; antimycin A, 4 x 10"^ M; Cd , 10"^ M; and NH4"'", 6 X 10'"* M. Arsenite at concentrations as high as 10' ^ M did not affect photophosphorylation with spinach phosphodoxin. Since phosphodoxin is a naturally occurring catalyst, it was of interest to study photophosphorylation in the presence of other known catalysts. Total photophosphorylation in the presence of NADP and PPNR plus increas- ing amounts of spinach phosphodoxin did not change from that observed with only NADP and PPNR. In the presence of ferricyanide, a definite in- hibition of ATP production was observed. A distinct stimulation of PMS- catalyzed photophosphorylation was observed, varying between 2- and 4-fold at low levels of phosphodoxin. Spinach phosphodoxin is active with chloroplasts from higher plants and chromatophores from photos ynthetic bacteria. Conversely, phosphodoxin from photos ynthetic bacteria is active with spinach chloroplast fragments. This crossing of activity irrespective of the photos ynthetic organisms from which the phosphodoxin was isolated is illustrated in Table 1 and in Fig. 7. Further demonstrations of this crossing of activity are given in references 21 and 22. 232 C, C. Black and A. San Pietro zoo too, c^. E I a. o 50 L— + PMS ./ -0- +NADP ■*^ _^'*' Phosphodoxin only + Ferncyanide .05 .1 15 ,2 il of Phosphodoxin Fig. 5. (Left) Effect of inhibitors on photophosphorylation with spinach chloroplast fragments plus spinach phosphodoxin. Fig. 6. (Right) Effect of spinach phosphodoxin on photophos- phorylation with spinach chloroplast fragments in the presence of other electron acceptors. ELECTRON PARAMAGNETIC RESONANCE SIGNAL The EPR signals observed with spinach phosphodoxin have been studied in cooperation with Dr. John Heise ^^4) Aqueous solutions of spinach phos- phodoxin exhibit a light- induced, pH- dependent EPR signal. The intensity of the EPR signal is increased with alkaline conditions in a fashion similar to the relative fluorescence intensity (Fig. 2). No dark EPR signal is observed with spinach phosphodoxin alone and the light- induced signal decays in the dark. Figure 8 demonstrates the effects of spinach phosphodoxin on the EPR signal of spinach chloroplast fragments in 4. 2 x 10" 2 M Tris-HCl buffer, pH 7. 8. The small characteristic light- induced EPR signal of spinach chloroplast fragments alone (25) can be observed by comparing curves 1 and 4 of Fig. 8. A sharp decrease in the dark EPR signal of spinach chloroplast fragments upon the addition of spinach phosphodoxin can be observed by com- paring curves 1 and 2 of Fig. 8. Upon illumination of spinach chloroplast fragments plus spinach phosphodoxin, an increased EPR signal was observed (compare curves 3 and 4 of Fig. 8). This increased EPR signal was observed immediately upon illumination and decayed with continuous illumination. Examination of the data indicates that spinach phosphodoxin contributes primarily to signal 2 ^25) of spinach chloroplast fragments '24)^ 233 C. C. Black and A. San Pietro .05 .10 .15 .20 ml of CHROMATIUM Foctor UJ -40 a: Fig. 7. (Left) Effect of Chromatium phosphodoxin on photophos- phorylation with spinach chloroplast fragments. Fig. 8. (Right) Effect of spinach phosphodoxin on EPR signals of spinach chloroplast fragments. Cuive 1 - Dark; chloroplasts alone. Curve 2 - Dark; chloroplasts plus phosphodoxin. Curve 3 - Light; chloroplasts plus phosphodoxin. Curve 4 - Light; chloroplasts alone. In summary, we have isolated a new naturally occurring, thermostable, water-soluble catalyst (phosphodoxin) of photophosphorylation which appears to be ubiquitous among photos yn the tic organisms. Phosphodoxin from each photos ynthetic organism appears to be active with both chloroplast and chromatophore fragments. Spinach phosphodoxin alone has a light- induced EPR signal and in combination with spinach chloroplast fragments decreases the dark EPR signal and increases the light EPR signal, and contributes primarily to signal 2. This is Contribution No. Laboratory. 129 from the Charles F. Kettering Research REFERENCES 1. Frenkel, A., J. Am. Chem. Soc. , 76, 5568(1954). 2. Arnon, D. I., M. B. Allen, F. R. Whatley, Nature, n4, 394(1954). 3. Geller, D. M. , and J. D. Gregory, Federation Proc. , J^, 260(1956). 4. Jagendorf, A. T. , and M. Avron, J. Biol. Chem., 231, 277(1958). 5. Hill, R. , and D. A. Walker, Plant Physiol. , 34, 2415T1959). 6. Arnon, D. I. , F. R. Whatley, and M. B. Allen, Science, 127, 1026 (1958). 7. Jagendorf, A. T. , Federation Proc. , 18, 974(1959). 234 C. C. Black and A. San Pietro 8. Davenport, H. E. , Proc. Royal Soc. 157B , 332(1963). 9. Gormet-Elhanan, Z. , andM. Avron, Blochem, Biophys. Res. Comm. , ^, 215 (1963). 10. Keister, D. , J. Biol. Chem. , 238, PC2590 (1963). 11. Shen, G. M. , S. Y. Yang, Y. K. Shen, H. C. Yin, Scientia Sinica, 12, 685 (1963). — 12. Krogmann, D. W. , and B. Vennesland, J. Biol. Chem., 234, 2205(1959). 13. Arnon, D. I. , F. R. Whatley, and M. B. Allen, Biochim. Biophys. Acta, J^, 607 (1955). 14. Whatley, F. R. , M. B. Allen and D. I. Arnon, Biochim, Biophys. Acta, J^, 605 (1955). 15. San Pietro, A., Brookhaven Symp. Biol. , J_l, 262(1958). 16. Krogmann, D. W. , Biochem. Biophys. Res. Comm., 4, 275(1961). 17. Habermann, H. , and A. R. Krall, Biochem. Biophys. "Res. Comm. , 4, 109 (1961). 18. Krogmann, D. W. , and M. Stiller, Biochem. Biophys. Res. Comm. , 7, 46 (1962). 19. Keister, D. L, , A. San Pietro, Arch. Biochem. Biophys. , in press. 20. Black, C. C. , C. A. Fewson, M. Gibbs, D. L. Keister, and A. San Pietro, Federation Proc. , 21, 398 (1962). 21. Black, C. C. , A. San Pietro, D. Limbach, and G, Norris, Proc. Natl. Acad. Sci. U. S. , 50, 37 (1963). 22. Black, C. C. , and A. San Pietro, p. 223-231 in Bacterial Photosynthesis (H. Gest, A. San Pietro and L. P. Vernon, eds. ). Antioch Press, Yellow Springs, Ohio, 1963. 23. Black, C. C. , A. San Pietro, G. Norris and D. Limbach, Plant Physiol. , in press. 24. Black, C. C. , J. J. Heise, and A. San Pietro, manuscript in prepara- tion. 25. Commoner, B. , J. J. Heise, and J. Townsend, Proc. Natl. Acad. Sci. U.S., 42, 710 (1956). CHARACTERIZATION OF ALLAGOCHROME AND ITS BIOSYNTHESIS IN LEAF EXTRACTS Helen M. Habermann Allagochrotne is a blue-green pigment present in alkaline homo- genates of a wide variety of plant species. Other colored sub- stances (including at least two yellow fractions) are normally pre- sent in crude allagochrome preparations. A survey of the distribu- tion of allagochrome in higher plant groups and a description of techniques which have been developed for the purification and ass^ of allagochrome have been published (2,5) _ There is at least cir- cumstantial evidence that allagochrome is involved in photophos- phorylation and respiration (2,^,6)^ The purpose of this paper is to summarize the recent progress to- ward understanding the structure and biosynthesis of allagochrome. Until recently there were few clues about the chemical nature of this pigment and at times the available information was misleading. ALLAGOCHROME BIOSYNTHESIS For the past several months we have proceeded on the hypothesis that allagochrome is a chlorogenic acid derivative. Briefly, the evidence for involvement of chlorogenic acid in allagochrome syn- thesis is as follows: 1) It is well known that a green derivative forms spontaneously from chlorogenic acid in the presence of ammonia. The ammonia de- rivative is spectrophotometrically different from allagochrome. Another chlorogenic acid derivative which forms slowly in the pres- ence of glycine in alkaline medium resembles allagochrome more closely. The latter derivative is reduced to a yellow form by hy- drosulfite, is autoxidizable and turns red in acid. In other words it has properties which are descriptively the same as those of allagochrome. 2) The addition of chlorogenic acid during the grinding of plant materials testing positively for allagochrome increases yields up to several fold. Absorption spectra of such enhanced crude prep- arations match those of control preparations rather well except for slight shifts of peak position in the red and major increases in absorbancy in the blue with a shift to shorter wavelengths for 235 236 Helen M. Habermann the minimum in difference spectra. Chlorogenic acid extracted from leaves with boiling water is just as effective in increasing yields as the commercially available chemical. 3) The participation of polyphenol oxidase in formation of colored derivatives of chlorogenic acid (especially during aging or after injury of leaves) has been proposed by a number of investigators (8,9)^ The rapid secondary synthesis of excess allagochrome from chlorogenic acid added during grinding suggested that an enzyme of this type is present and could account for part or all of the pig- ment found in leaf extracts. The extent of enzymatic formation of allagochrome during grinding was estimated in two ways: by adding cyanide to the extracting medium or by heating leaves in boiling water before extraction. Cyanide added to the extracting buffer reduces allagochrome values (see fig. 1). Heating leaves before extraction similarly reduces allagochrome values in controls and inhibits the secondary synthesis of pigment from added chlorogenic acid. Immersing leaves in boiling water for as short a time as 5 sec. reduces allagochrome values to about half, but longer heating results in proportionately less additional reduction of pigment values of the extracts (see fig. 2). Reduction in allagochrome values is not proportional to removal of chlorogenic acid. Less than 17o is removed by 5 sec. in boiling water and maximum extrac- tion is achieved by this means in 1 1/2 to 2 min . In experiments testing the effects of heating on the secondary synthesis of al- lagochrome from added chlorogenic acid, only 30 sec. heating be- fore grinding reduced the secondary synthesis to zero; shorter heating resulted in reduction of the secondary synthesis in pro- portion to the time of treatment over the entire range of added amounts of chlorogenic acid (see fig. 3). 4) Plants testing negatively for allagochrome respond in two pos- sible ways when chlorogenic acid is added during grinding: they either remain negative or become positive. This may be inter- preted as an allagochrome negative, enzyme negative vs. allago- chrome negative, enzyme positive situation; or it may indicate the presence of a natural antioxidant which interferes with quinone formation. We have not yet resolved this question. It seems quite probable then that the synthesis of allagochrome from chlorogenic acid and glycine is mediated by an enzyme of the polyphenol oxidase type. This has been substantiated by experi- ments in which sunflowers were grown on copper deficient nutrient solutions (7) . Such plants showed progressively lower copper con- tent of leaves formed at successive nodes and markedly decreased allagochrome levels. At the higher nodes, allagochrome values dropped to between 10 and 207, of controls. There was an increased susceptibility to cyanide poisoning of respiratory oxygen uptake and photosynthetic oxygen production in leaves deficient in copper 237 Helen M. Habermann 10 o 2 • — 1 1 1 1 — 1 1 — ( 1 < . 1 . • 1 1 • _1 — // — 1 1 1 ^.: 10"* I0-' 10" 10" I /' Chlorogenic ocid extracted 500 O ■ 400 300' 200 100 o Concentration of cyanide (Molar) 0(Ui51 20 40 60 80 100 120 Seconds in boiling water before extraction -I 1 1 r 1 1 T" conlrols 10 20 30 40 50 60 70 ;jMoles chlorogenic ocid added per gm leaf Fig. L (upper Left). Effects of cyanide in the extracting buffer on allagochrome values of node 3 sunflower leaves. Vertical bars through points indicate standard deviations of means of six determinations. Fig. 2 (upper right). Effects of immersion in boiling water prior to extraction on allagochrome values of node 8 sunflower leaves. Chlorogenic acid removed was estimated from optical density of aqueous extracts. Fig. 3 (bottom). Effects of heating on the secondary synthesis of allagochrome from chlorogenic acid added to the extracting buffer. Leaves were immersed in boiling water for 10, 20, or 30 sec. before extraction. 238 Helen M. Habermann and allagochrome . Although we now have some working hypotheses concerning the synthesis of allagochrome and evidence concerning its precursor, the reality of the pigment as a component of the living cell re- mains in doubt. Extraction at high pH , under oxidizing conditions and in a buffer containing glycine provides all the necessary con- ditions for production of an artifact. Attempts to find another suitable buffer for extraction have been unsuccessful except for borate-NaOH buffer. In this case, allagochrome values of sunflow- er leaves were only 1/3 to 1/4 what they were with glycine-NaOH buffer. It was not possible to increase yields in borate buffer by adding glycine to the grinding medium. Although there are sev- eral indications that at least part of the allagochrome found in extracts of leaves exists in vivo, unequivocal proof of its natu- ral occurrence is not yet available. CHARACTERIZATION OF ALLAGOCHROME Whether present in vivo or not, the pigment allagochrome re- mains a chemically interesting molecule. The following paragraphs summarize the kinds of information now available on which some speculations concerning the configuration of allagochrome and of the associated yellow pigments can be based. The latter pigments are of increased interest at the present time because of their re- semblance to phosphodoxin, a catalyst of photophosphorylation re- cently reported by Black et al (^^ . Molecular weight A preliminary ultracentrifugal analysis was made with a 0.15% solution of allagochrome in a Tris-HCl buffer (pH 8.2). Even pro- longed centrifugation at maximum speed (59,780 rpm) did not result in the formation of a boundary and the observed schlieren patterns were characteristic of the transient states observed in an ap- proach to equilibrium. Calculations of the molecular weight were made by a method suggested by Ginsberg e_t aj. '-^^ and a value of 720 was obtained. The marked optical density of the solution lim- ited the accuracy attainable by this method. The value should therefore be considered a lower limit and the true molecular weight probably lies between 720 and 1400. This method of analy- sis gave no information about the homogeneity of the sample but did indicate that no protein was present because of the absence of high molecular weight components. Electron spin resonance The first ESR spectra of solid allagochrome and a frozen water solution (containing 2 mg/ml) gave a 12 gaus peak to peak signal at g=2.005. This was the only signal found at that time and on this basis it was concluded that the common paramagnetic metal 239 Helen M. Habermann ions are not present in purified preparations of the pigment. However, direct chemical analyses of allagochrome had indicated a low but persistent Cu content. These results excluded the possi- bility that allagochrome might contain Cu++ . They also proved that the color changes of allagochrome, especially its reversible reduction, cannot be accounted for on the basis of copper but must rather be accounted for by a probably quinonoid chromophoric group, ESR signals of the type obtained previously only in the solid have been obtained recently in solution - probably because of in- creased allagochrome concentration (50 mg/ml vs. 2 mg/ml) . This is a free radical signal with some fine structure and present in very low concentrations. The molecule is not all in a free radi- cal form but seems in a state of constant presence of a small a- mount of free radical. With more concentrated preparations there is a small but recognizable +3 iron signal. No studies have yet been made of the possible effects of light on the ESR signals of allagochrome . Nuclear magnetic resonance Table I summarizes information from NMR spectra of a series of pigments and derivatives prepared from Helianthus leaves. Table I NMR Spectra of Helianthu s Pigment Samp les in D2O Sample Weight(mg) D20(tr il) NMR peaks from TMSP) (in cycles 1. 2. 3. 4. Allagochrome Allagochrome Yellow I Yellow I 207.5 54.6 135.3 166.4 2 1 1.5 1.5 214.9 214.9 214.7 214.8 210.5 209.2 211(broad) 5. Ppt. from Yellow I in 12N HCl (probably Na acetate) 153.4 2 6. Supernatant from 7. Yellow I in 12N HCl Yellow II 224.8 185.1 2 2 233.4 216.2 211.7 It seems probable that all of these compounds except 5 are closely related and may be derivatives of breakdown products of chlorogenic acid plus glycine. These spectra suggest that the quinic acid moiety of chlorogenic acid is not present in the col- ored products. 240 Helen M. Habermann Infrared spectra Infrared absorption spectra of allagochrome and yellow pigment preparations from Helianthus and Chrysanthemum strongly supported the indications of NMR spectra that these are closely related com- pounds. Spectra of allagochrome preparations from these two spe- cies were nearly identical. A R-CH=CHC-OR' configuration is defi- nitely not present in allagochrome (no band between 1715 and 1800 cm"-'-), although such an ester linkage joins together the caffeic and quinic acid portions of the chlorogenic acid molecule. There is no free NH2 group but there are several indications of bonded H n 1 N-H of amino or amide, probably of an amide of the type R-C-N-R . Another phenomenon associated with the measurement of infrared spectra is the observation that there is a change in spectrum of allagochrome during the scanning period. Such changes were ob- served with pigments prepared from both Helianthus and Chrysanthe - mum leaves. It is not known whether such changes are a conse- quence of the rather high intensity incandescent illumination re- ceived by the samples during measurement of spectra or are merely due to water in the samples. Elemental analyses The results of elemental analyses run on a lyophilized sample of Helianthus allagochrome are summarized in Table II. Table II Elemental analysis of Helianthus allagochrome Element Per cent 24.36% 24.39% H 4.23% 4.52% N 12.31% 12.19% P 0.1 % Residue 32.25% 32.19% Loss at 100° 2.99% 2 . 38% The large residue probably indicates a high salt content. We have recently added passage through a mixed bed of ion exchange sepha- dex as a final step in purification. The resulting samples are much darker green after lyophilization (almost black) . 2A1 Helen M. Habermann SUMMARY The increase in allagochrome values of leaves when chlorogenic acid is added to the extracting medium strongly indicates that this pigment is a chlorogenic acid derivative. An inhibition of the secondary synthesis of allagochrome from chlorogenic acid by cyanide or heat treatment of leaves is consis- tent with the hypothesis that this synthesis is mediated by an en- zyme of the polyphenol oxidase type. The reduction of allago- chrome values in plants deficient in copper adds further support to this hypothesis. Estimated molecular weight of allagochrome based on analytical ultracentrifuge studies is 720. NMR and infrared spectra indicate that allagochrome and the yellow pigments fractionated from it during purification are closely related chemically. ESR spectra showed a free radical signal at g=2.005. It remains to be determined whether allagochrome is present in vivo or is formed by the enzymatic oxidation of chlorogenic acid during grinding. Present evidence supports the hypothesis that at least part of the assayed amounts of this pigment could be present before extraction. ACKNOWLEDGEMENTS I am indebted to Drs. Ulrich Weiss, Marc S. Lewis and Chester DeLong of the Laboratory of Physical Biology, Institute of Arthri- tis and Metabolic Diseases, National Institutes of Health, for analytical ultracentrifuge, electron spin resonance and nuclear magnetic resonance data. Dr. H. Knorr of the Charles F. Ketter- ing Research Laboratory, Yellow Springs, Ohio, kindly did infra- red absorption measurements. These studies were supported by grants from the National Sci- ence Foundation (G-12757) and the U.S. Public Health Service (GM 07659-03). LITERATURE CITED 1. Black, C.C., A. San Pietro, D. Limbach and G. Norris. Proc . Nat. Acad. Sci. U.S. 50:37-43 (1963). 2. Garrick, L.S. and H.M. Habermann. Am. J. Botany. 49:1078- 1088 (1962) . 3. Ginsberg, A., P. Appel and H.K. Schachman . Arch. Biochem. & Biophys. 65:545-566 (1956). 242 Helen M. Habermann 4. Habermann, H.M. In: Progress in Photobiology . Elsevier Pub. Co., Amsterdam, pp. 576-580 (1961). 5. Habermann, H.M. Plant Physiol. 38:381-389 (1963). 6. Habermann, H.M. and A.R. Krall. Biochem. & Biophys . Res. Connn. 4:109-113 (1961) . 7. Habermann, H.M. Plant Physiol. 38(suppl . ) : liii (1963). 8. Shiroya, M. and S. Hattori. Physiol. Plantarum. 8:358-369 (1955). 9. Sondheimer, E. In: Plant Phenolics and their industrial sig- nificance, Proc . Plant Phenolics Group of N. Amer . , Corvallis, Oregon, pp. 15-38 (1962). SOME EFFECTS OF OXYGEN IN PHOTOSYNTHESIS BY CHLOROPLAST PREPARATIONS F. R. Whatley It has been suggested that the presence of oxygen is neces- sary for photosynthetic phosphorylation. Thus, Nakamoto et al. (1) showed that the cyclic photophosphorylation catalyzed by flavin mononucleotide (FMN) was greatly stimulated by oxygen when very low concentrations of the cofactor were supplied. The formation of ATP in this system was found to be accompanied by an oxygen exchange (2,3), suggesting that, under the experi- mental conditions employed (compare also ref. h) , the mechanism of the electron transport involved a reaction of reduced FMN with oxygen. But this mechanism cannot be a general one for cyclic photophosphorylation since, as is generally accepted, phenazine methosulfate (PMS) catalyzes a true cyclic photophos- phorylation not involving an oxygen exchange under either aero- bic or anaerobic conditions (2,3>^)' However, even in such a truly cyclic system oxygen has been found to play a role, apparently adjusting the redox balance of the system under some conditions . Another system in which a part of the photosynthetic appara- tus of chloroplasts has been linked with oxygen consumption is the "cytochrome £ photooxidase" activity observed by Nieman and Vennesland (5) in digitonin extracts of chloroplasts. This article discusses two aspects of the participation of oxygen in photosynthetic reactions of chloroplasts: (1) in the photooxidation of ferrocy tochrome c by digitonin extracts of chloroplasts and (2) as a "poising agent" to regulate electron flow in the true cyclic photophosphorylation catalyzed by PMS in broken chloroplasts. ( 1 ) Photooxidation of reduced cytochrome c . Confirming the results reported by Nieman and Vennesland (5), it was found that digitonin extracts of chloroplasts (prepared by continued extraction of spinach chloroplasts by L'fo digitonin, followed by the centrifugation procedure described by Nieman 243 244 F. R. Whatley and Vennesland) oxidized f errocytochrome c at a slow rate in the dark in air but not in argon, and that this dark oxidation, which proceeded in accordance with the thermochemical gradient, was completely suppressed by 10"'* M KCN. Ferrocytochrome c was, however, oxidized when the system was illuminated in the pre- sence of KCN-treated digitonin extracts and oxygen was supplied. No photooxidation occurred under an atmosphere of argon or in the dark, as shown in Fig. 1. Nieman and Vennesland concluded from similar evidence that a "cytochrome c photooxidase" had been unmasked in their preparation. However, as will now be reported, triphosphopyridine nucleotide (TPN) can be substituted for oxygen as a terminal electron acceptor in an atmosphere of argon. To accomplish this it was necessary to add both the electron carrier ferredoxin and the enzyme ferredoxin-TPN reduc- tase to transfer the electrons produced by the photoreaction to TPN. An experiment showing the photooxidation of ferrocyto- chrome c by TPN under argon is shown in Fig. 2 and represents another experimental manifestation of Nieman and Vennes land's apparent cytochrome c photooxidase activity. Photooxidation of ferrocytochrome c by TPN proceeded against a thermochemical gradient, unlike the reaction with oxygen, and obviously required an input of light energy. It was not imme- diately apparent why the oxidation of ferrocytochrome c by oxygen should also need an input of light. However, our results suggest that the photooxidation of ferrocytochrome c is a mani- festation of the terminal portion of the electron transport chain of noncyclic photophosphorylation in chloroplasts (see ref. 6), in which TPN acts as the physiological electron accept- or, but can be replaced unspecif ically by molecular oxygen. Additional support for the view that the digitonin extracts re- tained the terminal portion of the electron transport chain comes from an experiment, shown in Fig. 3> in which ferrocyto- chrome c was replaced by the ascorbate/dichlorophenol indophenol dye couple. In the light, but not in the dark, TPN became photoreduced at the expense of the oxidation of ascorbate. A preliminary report of these results has appeared elsewhere (7) • (2) Cyclic photophosphorylation catalyzed by phenazine metho - sulfate . It is generally accepted that PMS catalyzes a truly cyclic electron flow in chloroplasts, accompanied by ATP formation (2-4). The principal evidence to support this conclusion may be summarized: (i) No oxygen exchange was observed to accompany the ATP formation (2,3), (ii) the rates of phosphorylation in 245 F. R. Whatley I S ^ I 0.3 05 0.7 1 1 1 1— Photooxidotion of reduced cytochrome c by Q^ 09- l.l $=* light 4 6 minutes 8 10 Fig. 1 . Experimental details as described by Nieman and Vennesland (5). A decrease in O.D. at 550 \Wi corresponds to the oxidation of ferro- cytochrome c. 06 I 09[- 10 II Photooxidatlon of cytochrome c in argon by TPN ligttt -TPN 6 8 minutes 10 12 Fig. 2 . Experimental details as in Fig. 1, except that a partially purified preparation of "photosynthetic pyridine nucleotide reductase" (11) was added to each vessel, and 1 i,imole TPN was added as indicated. Fig. 5 . Experimental details as in Fig. 2, except that f errocytochrome £ was re- placed by 20 lamoles ascorbate + 0.2 umoles dichlorophenol indophenol (both omitted in "-ascorbate" treatment). Increase in O.D. at 3^0 mn corresponds to reduction of TPN. 246 F. R. Whatley air and under nitrogen were similar at saturating concentrations of PMS, and (iii) there was little effect of CMU on the rate of ATP formation anaerobically with saturating PMS (4). However, some unexpected effects of oxygen on the PMS-catalyzed cyclic photophosphorylation have been found. These effects will be discussed in the context of an interpretation proposed by Tagawa et al. (8) of the effect of oxygen on the cyclic photo- phosphorylation catalyzed by ferredoxin. In Table 1 is shown the effect of increasing the concentra- tion of the cof actor, PMS, on the formation of ATP in argon or PMS added (usrams) Argon 0.3 0.3 0.2 1 O.k 3 1.1 10 7.6 30 8.7 ^^^^ The reaction mixture Effect of air on cyclic contained, in a final photophosphorylation catalyzed by PMS volume of 3 ml, broken ATP formed in 15 min. chloroplasts (Pig) (nmoles) containing 0.1 mg /^ij- chlorophyll and the ' following in micro- ^•^ moles: tris buffer, "•^ pH 8.3, 80; MgClg, 10; 1.2 ADP, 10; K2HP3204, 10; 3.8 and the amounts of 8.8 phenazine methosulfate 9.2 indicated. The exper- iments were carried out in Warburg manometer vessels at 15° C at 20,000 lux. Prior to turning on the light, the "argon" series was flushed with purified argon gas for 5 minutes. ATP formation was measured as described previously (11). in air. Although at the saturating concentration of PMS com- monly employed (3O ^Lgrams per 3 ml reaction mixture) there was no difference between the rate in air or argon, a large stimu- latory effect of air became apparent at lower concentrations of PMS. When 3 jigrams PMS were used the reaction was not satura- ted by PMS (under our experimental conditions), but there was a threefold stimulation by air of the ATP formation. A similar stimulation has also been observed by Jagendorf and Avron (9) . The addition of small amounts of oxygen was found to produce the same effect as air. In Table 2 are reported the results obtained when 3 ngrams PMS were used as the cofactor for cyclic photophosphorylation and small amounts of oxygen gas were added to a reaction vessel of approximately 20 ml capacity. When 10 lomoles oxygen had been added the rate of ATP formation was max- 247 F. R. Whatley imal, but smaller amounts of oxygen produced a large stimulatiou 10 pinoles oxygen added to a vessel of 20 ml volume corresponds to about 1^ oxygen in the gas phase. Table 2 Experimental condi- Effect of small amounts o f oxvaen on tions were as given in cyclic photophosphorylation catalyzed legend to Table 1. 5 b y a limiting amount of PMS under argon ^g,-anis PMS were added in each vessel. The reaction vessels were capped with a serum bottle cap and flushed with purified argon. Small amounts of oxy- gen gas were then in- jected into the gas phase through the cap. ATP formed in Oxygen added ^0 minutes (piiiol es) _ (l imoles) 1.7 2 h.l k 5.6 10 7.8 These results appear similar to those obtained with the FMN- catalyzed pseudo-cyclic photophosphorylation, which has been shown to require oxygen as a terminal electron acceptor. As already stated, however, the oxygen exchange data do not support a similar interpretation for the PMS system. Consequently, an alternative explanation was sought. If the effect of oxygen in the PMS system were due to the oxidation of some component of the system, it seemed probable that the addition of another oxi- dant would produce a similar stimulation of the ATP formation. Since ferricyanide is not toxic to photophosphorylation by chloropiasts (e.g., it supports ATP formation in noncyclic photophosphorylation) it was chosen as a suitable substitute for oxygen. When tested in a system provided with a limiting amount (;> digrams) of PMS, ferricyanide was indeed capable of stimulating cyclic photophosphorylation with PMS under argon, as is shown in Table 3. The table indicates very clearly that the ferricyanide acts in a "catalytic" fashion, and not as a substrate; the addition of 1 pinole ferricyanide (which in non- cyclic photophosphorylation might have given 0.5 lamoles ATP) caused an increment of 5 lomoles ATP. The effect of ferricyanide is not due to the oxygen which it will produce via a Hill re- action when the light is turned on. This will be seen by com- paring the results in Tables 2 and 3 (both experiments carried out at the same time) ; 1 (.unole ferricyanide, which gave a large stimulation, would yield 0.25 iimoles oxygen, an amount much less than is needed to give a measurable effect when injected into the gas phase. 248 F. R. Whatley Table 3 Effect of small amounts of ferricyanide on cyclic photophosphorylation catalyzed by a limiting amount of PMS under argon Ferricyanide added (umoles) 0.25 0.5 1.0 1.5 2.0 2.5 ATP formed in 30 minutes (umoles) 1.7 3.5 6.2 6.1 7.9 8.0 8.4 Experimental condi- tions as described in legend to Table 1. 3 Jig PMS were added to each vessel. Ferri- cyanide was added as indicated and remain- ed in contact with the chloroplast frag- ments and the PMS for 5 minutes while the vessels were flushed with argon, and be- fore the light was turned on. The inhibitor p-chlorophenyl-dimethyl-urea (CMU) also affec- ted cyclic photophosphorylation with PMS, as can be seen by ex- amining the data presented in Table h. The addition of CMU (2.10"^ M final concentration) to the PMS system under argon Table 4 Effect of CMU on c yclic photophosphory lation catalyzed by PMS in air or argon ATP formed in 15 min (|.imoles) PMS 1 2 3 k added argon + air + (usrams) argon CMU air CMU 0.3 0.3 O.k 0.2 0.3 0.2 0.6 0.6 0.2 1 0.i+ 1.7 1.2 0.6 3 1.1 5.8 3.7 1.1 10 7.6 8.9 8.8 1.9 30 8.r 9.2 9.2 3.7 Experimental conditions as described in legend to Table 1. CMU to a final concentration of 2.10"^ M was added where indicated. produced a large stimulation of ATP formation at lower concen- trations of PMS (Columns 1,2), this effect resembling very closely the effect of air (Column 3). When 3O ^grams PMS were added no effect was produced by CMU. On the other hand, the 249 F. R. What ley addition of CMU to the PMS system in air caused a profound in- hibition of ATP formation (Column k) , which was only partly over- come by ^0 iagrams of PMS. Jagendorf and Avron (9) have made similar observations with both CMU and o-phenanthroline. These results do not at first appear to support the conclusion that PMS catalyzes a truly cyclic photophosphorylation, but are com- patible with this conclusion when the data are interpreted in terms of a "poising" action of oxygen (redox regulation). In a recent paper Tagawa et al. (8) have advanced such an hypothesis to explain their results on cyclic photophosphoryla- tion catalyzed by ferredoxin. They point out that noncyclic electron flow is a unidirectional electron transfer from water to TPN, driven by two photoreactions, B and A (systems 2 and 1, respectively, in Duysens ' terminology (10)). The intermediates in the electron transport chain will be kept in a partly re- duced, partly oxidized state (i.e., "poised") as long as TPN (or ultimately COg) is available, and no "overreduction" or "overoxidation" can occur. However, to maintain a cyclic elec- tron flow from a reduced cofactor back to the electron transport chain, as is required for cyclic photophosphorylation, the inter mediates of the electron chain must be kept in a suitable redox balance. If they are kept fully reduced, as by photoreaction B, they cannot accept electrons from the reduced cofactors. Tagawa et al. proposed the hypothesis "that molecular oxygen normally regulates the redox balance for the electron transport chain involved in cyclic photophosphorylation by chlorop lasts . In the presence of oxygen, the electron transport system does not become overreduced by the flow of electrons from water. But under anaerobic conditions the flow of electrons from photo- reaction B overreduces the components of the electron transport chain in chloroplasts and this overreduction cannot be counter- balanced by the regulatory oxidizing action of oxygen." As shown in Table 1, the rate of ATP formation at lower con- centrations of PMS was much less under argon than in air. This is interpreted to mean that, under argon, electrons flowing from photoreaction B overreduce the intermediates of the electron transport chain; and this slows down or prevents the return of electrons from photoreaction A via PMS to the chain. Oxygen can counterbalance the overreduction by oxidizing some portion of the intermediates and so bring about a regulation of the electron flow. Smaller amounts of oxygen produce a proportion- ately smaller response, although the action of oxygen is clearly catalytic (Table 2). The fact that very small amounts of ferri- cyanide (Table 3) stimulate the cyclic photophosphorylation with 250 F. R. Whatley PMS in a similar fashion emphasizes the idea that redox reagents other than oxygen can regulate ("poise") the system by partly oxidizing the intermediates. It appears likely from the results in Table 1 that even PMS itself in larger amounts (IO-3O fagrams) can bring about the redox regulation of the electron transport chain. When photoreaction B was prevented by the addition of the specific inhibitor, CMU, the flow of electrons from water to the electron transport chain was stopped, and no overreduc- tion could occur. There was thus no need for oxygen as a regu- lator under argon in the presence of CMU (Table 4). On the other hand, in the presence of air some contribution of elec- trons from water by photoreaction B appears to be necessary in order to maintain a suitable redox balance. The addition of CMU caused a large inhibition of cyclic photophosphorylation in air (Table 4), probably because oxygen overoxidized the inter- mediates of the electron transport chain, thus preventing a supply of electrons to photoreaction A. A possible explanation of the result shown in Table 1 might have been that in air the PMS is converted to the oxidation product pyocyanine (cf. Hill and Walker, 13) ^^<^ that pyocyanine functions more effectively as a catalyst at lower concentrations than PMS. However, as is shown in Table 5> pyocyanine-catalyzed cyclic photophosphorylation behaves like the PMS catalyzed sys- tem in its reaction toward aerobic and anaerobic conditions and towards CMU. Effect of CMU and ^ Table 5 of air on cyclic photophosphorylation catalyzed by limiting amounts of pyocyanine Pyocyanine added (micrograms) ATP formed in I5 minutes argon + air argon CMU ((jmoles) air + CMU 1 2 3 1.2 0.5 1.1 2.9 0.7 2.1 4.3 0.7 3.6 0.4 0.5 0.4 Experimental conditions as described in legend to Table 4, except that pyocyanine was substituted for PMS. Thus with PMS or pyocyanine as the cofactor, just as with ferredoxin, oxygen is able to play a regulatory role to combat the apparent tendency of photoreaction B to overreduce the intermediates of the electron transport chain. When photo- reaction B is suppressed the system is already in a suitable 251 F. R. Whatley redox balance without oxygen, and in the case of both PMS and pyocyanine can be shown to become easily overoxidized by oxygen. The hypothesis of a regulatory role for oxygen is consistent with the results of Jagendorf and Avron on the effect of air, CMU and o-phenanthroline on the PMS system, for which they were earlier unable to offer an explanation (9) . The hypothesis also suggests an explanation for the stimulatory effect of DCMU on the cyclic photophosphorylation catalyzed by the indophenol dyes, which has been described by Trebst and Eck (12). REFERENCES (1) Nakamoto, T., D. W. Krogmann and B. Vennesland, J. Biol. Chem. 2^, 2783 (1959). (2) Nakamoto, T., D. W. Krogmann and B. E. Mayne, J. Biol. Chem. 2^, l8ii3 (I960). (3) Krall, A. R. , N. E. Good and B. E. Mayne, Plant Physiol. ^, kk (1961). (h) Arnon, D. I., M. Losada, F. R. Whatley, H. Y. Tsujimoto, D. 0. Hall and A. A. Horton, Proc. Natl. Acad. Sci. (U.S.) 42, l^lh (1961). (5) Nieman, R. H. and B. Vennesland, Plant Physiol. ^, 2^5 (1959). (6) Losada, M. , F. R. Whatley and D. I. Arnon, Nature jj^, 606 (1961). (7) Horton, A. A. and F. R. Whatley, Plant Physiol. ^, Suppl. V. (1962). (8) Tagawa, K. , H. Y. Tsujimoto and D. I. Arnon, Nature 199 . 1247 (196 3). (9) Jagendorf, A. T. and M. Avron, Arch. Biochem. Biophys. 80, 246 (1959). (10) Duysens, L. N. M. , J. Amesz and B. M. Kamp, Nature 190 , 510 (1961). (11) Whatley, F. R. and D. I. Arnon in Methods in Enzymology VI . in press. 3. P. Colo^^7ick and N. 0. Kaplan, eds . Academic Press, New York. (12) Trebst, A., and H. Eck, Z. fUr Naturforsch. ^6, 455 (1961). (13) Hill, R. and D. A. Walker, Plant Physiol. ^, 2^0 (1959). III. STUDIES WITH ISOLATED ELECTRON CARRIERS PHOTOSYNTHETIC PYRIDINE NUCLEOTIDE REDUCTASE. IV FURTHER STUDIES ON THE CHEMICAL PROPERTIES OF THE PROTEIN Keelin T. Fry and Anthony San Pietro The first demonstration that a soluble factor can be added back to chloro- plasts to reconstitute their over-all electron- transport reaction was reported by Davenport, Hill and Whatley ' ^ Mn 1952. They showed that washed chloroplasts lost their capacity for reducing muscle methaemoglobin in the light. The addition of the soluble fraction restored the activity. They fur- ther demonstrated that the soluble methaemoglobin reducing factor was thermolabile. Lang and San Pietro (2, 3) were unaware of the prior work of Davenport et al. (i^when they reported that chloroplasts contain a soluble protein which catalyzed the photochemical reduction of pyridine nucleotides. At that time there wa s no evidence to suggest that the proteins isolated independently by these two groups might be identical. The possible identity of these two pro- teins, namely, photos ynthetic pyridine nucleotide reductase (PPNR) and the methaemoglobin reducing factor, only became apparent when they were available in purified form (^' ^). During the course of their studies on the methaemoglobin reducing factor, Davenport and Hill ^^^ observed that the preparation catalyzed the reduction of a number of haem- proteins, including cytochrome c, by illuminated chloroplasts. Moreover, this factor was shown to catalyze the reduction of NADP by illuminated chloroplasts ' ^ At the suggestion of Dr. H. E. Davenport, the ability