Photosynthetic phosphorylation and molecular oxygen

1314 BIOCHEMISTRY: ARNON ET AL. PROC. N. A. S. thermodynamic forces toward the most probable form, native ribonuclease. Some of the less likely possibilities mentioned above can only be rigorously excluded upon completion of current experiments on the nature of the pairing of half-cystine residues during the lag phase. The authors wish to thank Mrs. Juanita Cooke for her expert assistance in many of these experiments. * Present address: Massachusetts General Hospital, Boston 14, Massachusetts. t On leave of absence from The Weizmann Institute of Science, Rehovoth, Israel. 1 Sela, M., F. H. White, Jr., and C. B. Anfinsen, Biochim. et Biophys. Acta, 31, 417 (1959). 2 White, F. H., Jr., J. Biol. Chem., 236, 1353 (1961). 3Anfinsen, C. B., and E. Haber, ibid., 236, 1361 (1961). 4Aqvist, S. E. G., and C. B. Anfinsen, ibid., 234, 1112 (1959). 6 Peterson, E. A., and H. A. Sober, J. Am. Chem. Soc., 78, 751 (1956)-. 6 Hirs, C. H. W., S. Moore, and W. H. Stein, J. Biol. Chem., 200, 493 (1953). 7 Spackman, D. H., W. H. Stein, and S. Moore, Anal. Chem., 30, 1190 (1958). 8 Anfinsen, C. B., R. R. Redfield, W. L. Choate, J. Page, and W. R. Carroll, J. Biol. Chem., 207, 201 (1954). 9 Crestfield, A. M., Smith, K. C., and F. W. Allen, ibid., 216, 185 (1956). 10 Richards, F. M., Compt. rend. trav. lIb. Carlsberg, Ser. Chim., 29, 315 (1955). 11 Boyer, P. D., J. Am. Chem. Soc., 76, 4331 (1954). 12 Bray, G. A., Anal. Biochem., 1, 279 (1960). 13 Sela, M., and C. B. Anfinsen, Biochim. et Biophys. Acta, 24, 229 (1957). 14 Sela, M., C. B. Anfinsen, and W. F. Harrington, ibid., 26, 502 (1957). - Haber, E., M. Sela, and C. B. Anfinsen, Federation Proc., 20, Part I, 217 (1961). 16 Experimental work is now in progress to determine the nature of the pairing of half-cystine residues at various times during the early stages of reoxidation. Preliminary results indicate that pairing is quite random and that "incorrectly" formed bonds are present. 17 Harrington, W. F., and J. A. Schellman, Compt. rend. trav. lib. Carlsberg, Ser. Chim., 30, 21 (1956). 18 Sluyterman, L. A. A., Biochim. et Biophys. Acta, 48, 429 (1961). 19 Haber, E., and C. B. Anfinsen, unpublished data. PHOTOSYNTHETIC PHOSPHORYLATION AND MOLECULAR OXYGEN* BY DANIEL I. ARNONt M. LOSADA, F. R. WHATLEY, H. Y. TsuJIMOTO, D. 0. HALL, AND A. A. HORTON DEPARTMENT OF CELL PHYSIOLOGY, UNIVERSITY OF CALIFORNIA, BERKELEY Communicated July 3, 1961 Oxygen and photosynthesis were first linked about 200 years ago when both were discovered almost simultaneously. The earliest concept of photosynthesis was that of planetary ventilation in which illuminated plants exchanged CO2 or "bad air" for 02 or "vital air" (see historical review1). A mechanism for this gas exchange was proposed in 1796 by Ingenhousz.2 Green plants, he suggested, absorb from "carbonic acid in the sunshine, the carbon, throwing out at that time the oxygen alone, and keeping the carbon to itself as nourishment." 2 For over a hundred years afterward, the view that CO2 assimilation always involved a liberation of oxygen gas was so firmly entrenched that it was even extended to the dark CO2 assimilation by chemosynthetic bacteria.3-5 The idea VOL. 47, 1961 BIOCHEMISTRY: ARNON ET AL. 1315 of photosynthesis without oxygen evolution seemed a contradiction of terms until the conclusive work of van Niel6' 7with photosynthetic bacteria firmly established that, basically, the assimilation of CO2 at the expense of light energy does not depend on the liberation of oxygen or on the presence of oxygen. Bacterial photosynthesis occurs under strictly anaerobic conditions and includes organisms that are obligate anaerobes (for example, Chlorobium and Chromatium6, 7). Although molecular oxygen was shown to be unnecessary and often even detrimental to bacterial photosynthesis, its role in the energy transformations that occur during photosynthesis in green plants has gained more attention with time. In 1939, Hill postulated8 that the consumption of molecular oxygen is necessary for photosynthesis of green plants in order to supply additional energy for CO2 assimilation. He envisaged "first a reduction of a substance by chloroplasts giving oxygen and then a reoxidation of the reduced substances by part of the oxygen together with a simultaneous reduction of the necessary equivalent of carbon dioxide." 9 In 1951 Burk and Warburg'0 reported the separation of photosynthesis into a light reaction and a dark reaction on the basis of manometric measurements with intact Chlorella cells. In the words of Warburg, "in the light reaction, one molecule of 02 will develop per molecule of chlorophyll ... whereas during the dark period following the end of illumination it can be observed manometrically ... that two-thirds of the oxygen gas developed during the light period undergoes a back reaction." '1 This dark consumption of oxygen is, in Warburg and Burk's view, a photosynthetically induced respiration that supplies two-thirds of the needed energy and is, therefore, indispensable for CO2 assimilation in photosynthesis. As summed up recently by Warburg, "Keine (CO2) Fixierung ohne Atmung." 12 A biochemical model for oxygen consumption in photosynthesis in accordance with the views of Hill and Warburg was proposed by Vishniac and Ochoa.13 They suggested that the ATP used for CO2 assimilation in photosynthesis is formed by a collaboration between chloroplasts and mitochondria. In their model, chloroplasts reduced pyridine nucleotide in light and ATP was formed when mitochondria reoxidized the reduced pyridine nucleotide with molecular oxygen, by the dark process of oxidative phosphorylation. A similar model for ATP formation has been used in several general schemes of photosynthesis, as, for example, in the one proposed in 1951 by Holzerl4 and in 1954 by Calvin's group (ref. 1a, Fig. 7). In all these schemes, oxidative phosphorylation by mitochondria, and hence oxygen consumption, was an essential feature of the mechanism of photosynthesis. A mechanism of photosynthesis that includes a consumption of molecular oxygen cannot apply to the strictly anaerobic bacterial photosynthesis and leads, therefore, to the inference that the mechanism of energy conversion in bacterial photosynthesis is basically different from that in plant photosynthesis. However, the discovery of photosynthetic phosphorylation in isolated spinach chloroplasts16 and, shortly thereafter, in cell-free preparations of Rhodospirillum rubrum, 17 has pointed to mechanisms for ATP formation at the expense of light energy that are common to green plants and photosynthetic bacteria. A unified concept of photosynthesis in plants and bacteria, based on photosynthetic phosphorylation as the common denominator is now possible and has been presented elsewhere. 18-20 A key premise in this concept is that, in green plants as in baqteria, photosynthetic phosphorylation is fundamentally independent of molecular oxygen. Since this 1316 BIOCHEMISTRY: ARNON ET AL. PROC. N. A. S. has recently been questioned12' 21 we have re-examined the relation of oxygen to photosynthetic phosphorylation. The purpose of this article is to bring together and add to the evidence which confirms and extends our earlier conclusions that photosynthetic phosphorylation, both in green plants and photosynthetic bacteria, is basically independent of molecular oxygen and hence of respiration. Cyclic and Noncyclic Photophosphorylation.-Photosynthetic phosphorylation (photophosphorylation) is a term coined in 1954 to describe a light-induced ATP formation by isolated chloroplasts, without the aid of mitochondria and without the consumption of oxygen.16 The over-all reaction of photosynthetic phosphorylation is represented by equation (1): light ADP + P ATP (1) Reaction 1, in which the sole product is ATP, was subsequently designated cyclic photophosphorylation, to distinguish it from a second photophosphorylation reaction by isolated chloroplasts (equation (2)) which was found a few years later22 and named noncyclic photophosphorylation.18 light TPN + 2H+ + ADP + P + 20H- ) TPNH2 + ATP + H20 + '/202 (2) The terms "cyclic" and "noncyclic" refer to the electron flow mechanisms which have been proposed'8' 19 for these two reactions that now jointly constitute the process of photosynthetic phosphorylation. In cyclic photophosphorylation all of the biochemically effective light energy is used for ATP formation. In noncyclic photophosphorylation only a portion of the biochemically effective light energy is used for the formation of ATP; the remainder is used for the formation of a reductant, TPNH2, and the excretion (evolution) of oxygen. A nonphysiological variant of noncyclic photophosphorylation (reaction 2) is reaction 3, in which TPN is replaced by ferricyanide22' 23 (represented here by Fe3+). light 2Fe3+ + ADP + P + 20H- - 2Fe2+ + ATP + H20 + 1/202 (3) The significance of cyclic and noncyclic photophosphorylation to the over-all mechanism of photosynthesis in green plants is that they supply-aside from oxygen as the excreted by-product-the two products, ATP and TPNH2, which alone suffice for converting C02 into carbohydrates in the dark.24 25 Both ATP and TPNH2 (also oxygen) are formed by noncyclic photophosphorylation (equation (2)) but cyclic photophosphorylation (equation (1)), which supplies only ATP, is also needed for C02 assimilation because the ATP formed in noncyclic photophosphorylation is insufficient to convert CO2 to carbohydrate.26 Chloroplast fragments ("grana") do not respire27' 28 and hence, according to Warburg, would not be expected to fix CO2 ("Grana atmen nicht und fixierien nicht" 12). However, by fractionating isolated chloroplasts we have shown that grana fail to assimilate C02 not because they are unable to respire but because they lack the requisite chloroplast enzymes. These are water-soluble and are readily leached out in the preparation of "grana."29 The soluble chloroplast enzymes, either in situ or when extracted from chloroplasts, assimilate CO2 to the level of carbohydrates and do so solely24 at the expense of ATP and TPNH2- VOL. 47, 1961 BIOCHEMISTRY: ARNON ET AL. 1317 the two compounds formed in chloroplasts by cyclic and noncyclic photophosphorylation (equations (1) and (2)) and not by respiration. CO2 assimilation proper in photosynthesis, whether in whole cells30 or in isolated chloroplasts,24' 25 consists of the same exclusively dark, enzymatic reactions3' that are now known to operate in nonphotosynthetic cells.32-34 There is, so far as we know, no valid experimental basis, for postulating in photosynthesis a peculiar photochemically produced "reducing power" that is capable of direct reduction of CO2 to the level of carbohydrate by some special mechanism unknown elsewhere in biochemistry (see, for example, ref. 35 and 36). If we abandon the notion of an unspecified, hypothetical "reducing power" and interpret photosynthesis in terms of known biochemical mechanisms, then the unique feature of photosynthesis as a biological process is not CO2 assimilation-a dark process driven, in both photosynthetic and nonphotosynthetic cells, by ATP and reduced pyridine nucleotide-but the formation of these two compounds by cyclic and noncyclic photophosphorylation. The discovery of photosynthetic phosphorylation was part of a coincident, broader finding'6' 27 that isolated chloroplasts are capable of carrying out a complete extracellular photosynthesis, i.e., a conversion of CO to starch and sugar, at physiological temperatures and with no energy supply except visible light. The view that chloroplasts are the sites of complete photosynthesis was once widely held without the support of critical experimental evidence and was later abandoned because of evidence to the contrary, only to be formulated anew on the basis of improved methods for isolating functional chloroplasts from leaves (see review'). The discovery of cyclic and noncyclic photophosphorylation by isolated chloroplasts (also of CO2 assimilation, see review') was confirmed and extended in other laboratories, notably those of Jagendorf,37-40 Wessels,41 42 Vennesland,43 44 and Hill. Most of the work has been done with spinach chloroplasts but recently Whatley et al.46 have also demonstrated cyclic and noncyclic photophosphorylation (and CO2 assimilation) in chloroplasts isolated from several other species of plants.47 Of the two photophosphorylation reactions (equations (1) and (2)), the cyclic type (equation (1)) appeared the more basic from the standpoint of a general mechanism of energy transformation in photosynthesis since this reaction was found in representatives of all the different groups of photosynthetic organisms. Williamsw found photosynthetic phosphorylation in the obligately anaerobic photosynthetic bacteria, Chromatium and Chlorobium. Cyclic photophosphorylation in algal preparations was found by Thomas and Haans6' and Petrack and Lipmann.52 (For a more complete review of literature see ref. 53.) Special Features of Cyclic and Noncyclic Photophosphorylation.-The unique features of cyclic photophosphorylation are that ATP is formed with no added electron donor and no added electron acceptor. Cyclic photophosphorylation consumes neither respiratory substrate nor molecular oxygen. Both are replaced by light53 which provides the free energy required for the synthesis of pyrophosphate bonds of ATP. This seemed at first not to be the case in photosynthetic bacteria. Frenkel's cell-free preparations of R. rubrum became substrate-dependent after washing; the rate of phosphorylation was doubled on adding a-ketoglutarate.17 But in subsequent experiments Frenkel54 ruled out the dependence on an added chemical substrate and once this fundamental point was clarified, the equivalence 1318 BIOCHEMISTRY: ARNON ET AL. PROC. N. A. S. of cyclic photophosphorylation in chloroplasts and bacterial particles seemed probable.55 The independence of photosynthetic phosphorylation from a respiratory substrate is confirmed by the independence of this process from CO2 assimilation. Photosynthetic phosphorylation occurs when the enzymes of the reductive carbon cycle are removed by washing the chloroplasts, CO2 is not supplied to the reaction mixture, and the reaction vessels contain KOH in the center well. This would not exclude a possible catalytic participation of CO2 in photophosphorylation (cf. Warburg et al.12 and Vennesland et al.56). What can be definitely excluded is that substrate amounts of carbon compound(s) are first synthesized in light from CO2 and are then used as electron donors for the formation of ATP. While cyclic photophosphorylation was recognized as a common denominator of all photosyntheses, noncyclic photophosphorylation as represented by equation (2)-a reaction that liberates oxygen-seemed at first destined to remain a special feature of plant photosynthesis, since oxygen is not evolved in bacterial photosynthesis. However, Losada et al.57 (cf. also Duysens et al.58) have recently separated reaction 2 into two distinct photochemical reactions (a) a photooxidation of hydroxyl ions that yields oxygen (equation (4)) and (b) a photophosphorylation reaction proper in which the photoreduction of triphosphopyridine nucleotide is coupled with the formation of adenosine triphosphate (equation 5). light 2A + 20H- > 2A- + H2O + 1/202 (4) light TPN + 2H+ + ADP + P + 2A- >-TPNH2+ ATP + 2A (5) light Sum: TPN+2H++ADP+P+2OH--- TPNH2 + ATP + H20 + 1/202 (2) Only the photooxidation of hydroxyl ions (equation (4)) is peculiar to green plants since only green plants seem capable of using OH- as an electron donor with the resultant excretion of molecular oxygen. Reaction 5, the noncyclic photophosphorylation reaction proper, occurs also in photosynthetic bacteria.59 Since they use inorganic or organic electron donors, such as thiosulfate or succinate instead of OH-, oxygen evolution does not occur in the reaction. Figure 1 depicts the over-all scheme for noncyclic photophosphorylation in green plants that we now envisage. The intermediate A in Figure 1 is both the electron acceptor for the first light reaction and the electron donor for the second light reaction (cf. equations (4) and (5)). We have used indophenol dyes as an experimental device to separate the two photochemical reactions. The natural intermediates with which the dyes have interacted have not been identified. They may include cytochrome components60 or a quinone.61, 62 The terminal physiological electron acceptor in noncyclic photophosphorylation (B in Fig. 1) is pyridine nucleotide. Our original scheme for noncyclic photophosphorylation (ref. 22, Fig. 3) envisaged the phosphorylation step as occurring between the unknown reductant (H), formed by the photolysis of water and TPN, whereas our present scheme, based on the electron flow theory,18' 53 links the phosphorylation step with the oxidation of cytochrome by photoactivated chlorophyll. VOL. 47, 1961 BIOCHEMISTRY: ARNON ET AL. 1319 (?e,-- B - e v Chi ~+) Cyt - A I et! LIGHT pigment J(4, OH O - H20 [OH] LIGHT \4°2 FIG. 1.-Scheme for noncyclic photophosphorylation of the green plant type. In the first light reaction, the pigment molecule (see text) becomes excited by the absorption of a quantum of light. The excited pigment donates its high-energy electron (e -) to an intermediate electron acceptor (A) and accepts an electron from an hydroxyl ion. The oxidation product (OH) of the hydroxyl ion is the precursor of molecular oxygen. In the second light reaction, the chlorophyll molecule (Chl), excited by the absorption of a quantum of light, donates its high-energy electron (e-) to the terminal electron acceptor (B) and accepts, via the cytochrome system (Cyt), an electron from the reduced intermediate (A-) formed in the first light reaction. The phosphorylation step is linked with the transfer of the electron from cytochrome to chlorophyll. (.e-,. B [g tcz~te- e- A P. LIGHT FIG. 2. -Scheme for noncyclic photophosphorylation of the bacterial type. The chlorophyll molecule (Chl) becomes excited by the absorption of a quantum of light. The excited chlorophyll donates its high energy electron (e-) to the electron acceptor (B) and accepts, via the cytochrome system (Cyt), an electron from an external electron donor (A -). The phosphorylation step is linked with the transfer of the electron from cytochrome to chlorophyll. TPN reduction in noncyclic photophosphorylation (reaction 2), although normally coupled with, may also be uncoupled from, ATP formation. Trebst et al. (ref. 63, Table 1) have shown that ammonia64 suppresses almost completely the formation of ATP in reaction 2 in light without inhibiting the formation of TPNH2. These results speak against the suggestion of Chance and Olson65 that the photoreduction of TPN may be driven by ATP, formed perhaps by cyclic photophosphorylation (cf. also Marr6 and Forti66). The contrary is, in fact, the case. Elec- 1320 BIOCHEMISTRY: ARNON ET AL. PROC. N. A. S. /00 02 produced 80 ~~~~~~~~/00 02 produced % 60 80 Xi 40 - \ /- 60 _a \ t TPN//2 ~~~formed I 'a,~~~~~~~~~~~~~~~~~~q rz, ,00 t 40 t 0_A_ oxygenevolutonand TPNH feductiormed for 20 4~~~~~~0 Oxygen was measured manometriATP formed 0630644 66/ 677 699 FIG. 3.-Effectiveness of monochromatic 66 6 6 6 light, in the red region of the spectrum, on oxygen evolution and TPN reduction, occur- (m ring as separate chloroplast reactions.57 Oxygen was measured manometrically when FIg. evolutionwss of dscribed the photochemical reaction was limited to Fig. 4ig e rfectraegin of mo ncthromato the photooxidation of OH- (equation (4)) lightn mth red regio othespetrum do in a reaction mixture which included, in a foxgn oution andlume cylc oophosp horlafinal volume of 3 ml, chloroplast fragments ioyen, urigasseprnts mclsrop ast reaciens. (Cae) containing 1 mg. of chlorophyll and Oxgn evoluti fon meason as dylesc pribed the following in Mmoles: tris/acetate bufih Fuffre,3. AP formation by; cyclic mphoto fer, pH 8, 40; potassium ferricyanide, 15; phosphorylat (equereantin () was measured and 2, 6 - dichlorophenol indophenol 0i aprllel e imnt inal reatio mi (DCPIP), 0.2. TPN reduction (measured wi inclue ina fial voum ae o m spectrophotometrically at 340 mM) was chloroplast fragments (C18) containing 1 mg carried out in a parallel experiment in which chlorophyll and the following in Mmoles: tris! the photooxidation of water was blocked by HCl buffer, pH 8.3, 80; MgSO4, 5; ADP, 10; the omission of chloride and the addition K5)). 10;i anrdvian K0 (mnaedione) ofe 2ecin X er run Mo CMU The eletro donor 0.3.s qiaen o010Ml Pfre system f reuc cons dof 100 on the ordinate scale is equivalent to moles ascorbate and 0.05 moles of 2,3',6- atoms oxygen evolved per trichlorophenol indophenol (equation (5)). quanta of light absorbed. 60 on the ordinate The reactions were run for 15 mm at 15'C. scale is equivalent to 0.10 ,umole ATP formed Gas phase, nitrogen. per /mole quanta of light absorbed. 100 on the ordinate scale is equivalent to 0.15 MAatoms oxygen evolved per M~mole quanta of light absorbed. 58 on the ordinate scale is equivalent to 0.09 M~moles TPNH2 formed per /Amole quanta of light absorbed. tron transport, resulting in TPN reduction, may proceed without ATP formation, but no photophosphorylation can occur unless it is coupled with electron transport. Experimentally, it is possible to convert noncyclic photophosphorylation in chloroplasts (equation (2) and Fig. 1) to a "bacterial" type (equation (5) and Fig. VOL. 47, 1961 BIOCHEMISTRY: ARNON ET AL. 1321 2) by replacing OH- with an exogenous electron donor at the level of A-. Reduction of pyridine nucleotide is then coupled with the oxidation of the exogenous electron donor and a simultaneous ATP formation but without oxygen evolution.57 We have not definitely identified the "pigment" shown in Figure 1 that is involved in the photooxidation of hydroxyl ions. Recent experiments on the effectiveness of monochromatic light on the photochemical reactions of chloroplasts suggest that this may be a special photosynthetic pigment such as chlorophyll b or another accessory pigment that is found only in oxygen-evolving organisms, i.e. higher plants and algae, rather than chlorophyll a (or the closely related bacteriochlorophyll) found in all photosynthetic cells. (See review of related literature in ref. 57.) In the red region of the spectrum the photooxidation of OH- (equation (4)) was most effective around 644 miu, which corresponds to the maximum absorption peak67 in the red for chlorophyll b (Figs. 3 and 4). The effectiveness of monochromatic light in the red region on photoreduction of TPN (equation (5)) and cyclic photophosphorylation (equation (1)) was distinctly different from that on oxygen evolution (equation (4) and Figs. 3 and 4), thus supporting the view that these are different light reactions. The experiments with monochromatic light favor the conclusion that the photochemical reaction which produces oxygen is different from the other photochemical reactions which produce ATP and reduced pyridine nucleotide. The relation of oxygen produced to ATP formation in noncyclic photophosphorylation will now be examined in more detail. Oxygen as an Excreted By-Product of Noncyclic Photophosphorylation.-The essence of noncyclic photophosphorylation, both in plants and in photosynthetic bacteria, appears to be the formation of ATP which is coupled with an electron transport from an external electron donor to pyridine nucleotide acting as the electron acceptor. As already mentioned, in plants, as in photosynthetic bacteria, oxygen is not consumed in these essential aspects of noncyclic photophosphorylation. Oxygen may be liberated in this process but only in the specific instance, that is characteristic of plant but not of bacterial photosynthesis, when OH- is the external electron donor. This concept is supported by the experimental findings that ATP formation greatly increases the rate of oxygen evolution and the concomitant reduction of the electron acceptor, whether it is ferricyanide or TPN22 23, 39, 68, 69 (equations (2) and (3)). These results run counter to Hill's original hypothesis that in chloroplasts "the reduced product formed in the light might be reoxidized by molecular oxygen to give a phosphorylating system similar to that in chemosynthetic bacteria." 45 As Hill and Bonner have recently pointed out, "In such a hypothesis one would expect that addition of inorganic phosphate and ADP to the system would result in decreased rates of (H) acceptor reduction and of oxygen production" 60 (italics ours). Hill has accepted the new experimental findings and concluded that "the above ... hypothesis has to be abandoned." 60 Warburg,'2 however, has retained the hypothesis of oxygen consumption in energy transformations in photosynthesis and questioned the new experimental facts. Specifically, Warburg et al.12 found that with naphthoquinone sulfonate as the hydrogen acceptor, the stoichiometry of oxygen evolution by chloroplasts is not affected by phosphorylation. Warburg et al.12 suggested that the increase in photoproduction of oxygen as a corsequence of 1322 BIOCHEMISTRY: ARNON ET AL. PROC. N. A. S. photophosphorylation-first observed by other investigators22' 68 with ferricyanide (equation (3))-was the result of an uncontrolled change in acidity. Since the increased production of oxygen that accompanied photophosphorylation in the ferricyanide system was found in solutions that were strongly buffered with tris (hydroxymethyl) aminomethane buffer,23 it seemed unlikely that the observed effect was the result of uncontrolled pH changes as suggested by Warburg et al.'2 Moreover, Davenport69 has recently found that the rate of photoreduction of TPN, the physiological hydrogen acceptor in noncyclic photophosphorylation (equation (2)), was also greatly increased at two different pH's when a phosphateaccepting system was present. We have now reinvestigated the effect of photophosphorylation on oxygen evolution by isolated chloroplasts at different hydrogen ion concentrations, using either TPN or ferricyanide as electron acceptors. The pH value of each reaction mixture was checked at the end of the experiment. The results are shown in Figures 5 and 6. With either TPN or ferricyanide as the electron acceptor, we observed a signifi- TPN # (ADP+ P) qj TPN I 7.6 7.8 80 8.2 84 8.6 pH FIG. 5.-Effect of photophosphorylation at various pH values on oxygen evolution accompanying TPN reduction. The reaction was carried out in Warburg manometer vessels at 150C. Illumination, 25,000 Lux for 6 min.; gas phase, nitrogen. In all vessels the reaction mixture contained, in a final volume of 3 ml, 200 Mimoles tris-HCl buffer, 4 ,moles TPN, purified TPN-reductase from spinach, and broken chloroplasts (Pi,) containing 0.2 mg chlorophyll. To the phosphorylating series [TPN plus (ADP plus P)] were added 10 Mmoles MgSO4, 10 Mmoles potassium phosphate and 10 /Amoles ADP. Tris buffer, potassium phosphate, and ADP were each adjusted to the selected pH; the TPN was adjusted to pH 7.5. The pH shift, as determined at the end of the reaction, was found not to exceed 0.1 unit. VOL. 47, 1961 BIOCHEMISTRY: ARNON ET AL. 1329 for a relatively high concentration of chloroplast material suggests that the anaerobic FMN and vitamin K systems require more chloroplast factor(s) than the pseudocyclic oxygen-dependent system. No similar requirements were observed for the anaerobic phenazine methosulfate system. Despite the similarity of the over-all reactions (equations (1) and (8)), the anaerobic, cyclic photophosphorylations catalyzed by vitamin K and FMN can be distinguished experimentally from their oxygen-dependent, pseudocyclic analogues. The latter depend on the photoproduction of oxygen (equations (6), (7), and 8)) which requires chloride", 80 and are sensitive to two well-known inhibitors of oxygen evolution in photosynthesis, CMU8' (p-chlorophenyldimethyl urea) and o-phenanthroline.80 Table 1 shows that, under experimental conditions favorable for each system, TABLE 1 EFFECT OF CHLORIDE ON PHOTOPHOSPHORYLATION IN NITROGEN OR AIR pmoles ATP Formed- Treatment Minus chloride Plus chloride Nitrogen, FMN 5.1 5.7 Nitrogen, vitamin K, 9.7 9.9 Air, FMN 0.5 6.1 Air, vitamin K3 0.4 5.5 In the nitrogen series the illumination was 2,000 Lux for 30 min, and the reaction mixture included, in a final volume of 3 ml, chloroplast fragments (Ci.) containing 2.5 mg chlorophyll and 0.3 moles of FMN or vitamin Ks. In the air series, the illumination was 50,000 Lux for 30 min and the reaction mixture included, in a final volume of 3 ml, chloroplast fragments (Ci,) containing 0.2 mg chlorophyll and 0.003 pumoles of FMN or vitamin Ka. Other components of the reaction mixture were, in pmoles: tris buffer, pH 8.3, 80; KtH32PO4, 15; ADP, 15; and MgSO4, 5. Chloroplasts were prepared in 0.5 M sucrose and chloride-free reagents were used. In the plus chloride series 10 jsmoles NaCl were added. Nitrogen gas was purified by passing through a zinc-methylene blue solution (A. T. Jagendorf, personal communication). the omission of chloride had scarcely an effect on the anaerobic cyclic photophosphorylation with FMN or vitamin K, but severely inhibited their oxygen-de- pendent, pseudocyclic counterparts. Likewise, CMU (also o-phenanthroline53) inhibited the oxygen-dependent, pseudocyclic photophosphorylation but had relatively little effect on their anaerobic, cyclic counterparts (Table 2). TABLE 2 EFFECT OF CMU ON PHOTOPHOSPHORYLATION IN NITROGEN AND AIR WITH HIGH CHLOROPHYLL AT Low LIGHT INTENSITY pmoles ATP Formed Treamn - Nitrogen Air- Treatment Control CMU Control CMU FMN 7.0 6.6 4.2 1.7 Vitamin K3 8.1 7.6 3.3 1.3 PMS 2.7 2.4 2.9 2.8 The reaction mixture included, in a final volume of 3.0 ml, chloroplast fragments (Ci.) containing 2.5 mg chlorophyll and, in moles: tris buffer, pH 8.3, 80; MgSO4, 10; K2H32PO4, 15; ADP, 15; and, where indicated, FMN, 0.3; vitamin Ks, 0.3, or, phenazine methosulfate (PMS), 0.1. The final concentration of CQMU was 2 X 10-6 M. The reaction was run for 30 min at an illumination of 2,000 Lux. The nitrogen gas was purified as described in Table 1. When some experimental conditions are unfavorable for anaerobic, cyclic photophosphorylation, the high affinity of reduced FMN or vitamin K, for molecular oxygen may still give rise to a pseudocyclic, oxygen-dependent photophosphorylation. Thus, we have observed in the presence of 10-4 M FMN or vitamin K3 at a high light intensity, and with a relatively low concentration of chloroplast material (0.25 mg chlorophyll/3 ml) that photophosphorylation became oxygendependent, even when the reaction was started in purified nitrogen gas that was 1330 BIOCHEMISTRY: ARNON ET AL. PROC. N. A. S. TABLE 3 EFFECT OF CMU ON PHOTOPHOSPHORYLATION IN NITROGEN AND AIR WITH Low CHLOROPHYLL AT HIGH LIGHT INTENSITY - yjmoles ATP Formed Nitrogen Ai Treatment Control CMU Control C 1[U FMN 5.6 1.5 5.4 1.1 Vitamin K3 8.6 4.7 5.3 1.0 PMS 9.3 9.9 8.9 8.6 Experimental conditions as in Table 2 except that less chloroplast material (Ci.) was used (0.25 mg chlorophyll) and the illumination was increased to 20,000 Lux. Reaction time, 15 min. treated to remove oxygen impurities (Table 3). Phosphorylation became oxygendependent when the catalytic amounts of FMN or vitamin K? were photochemically reduced by chloroplasts. An equivalent amount of oxygen was then liberated which was sufficient to sustain an oxygen-dependent photophosphorylation (cf. Trebst and Eck75). The oxygen dependence of the system was determined by blocking the oxygen evolution reaction with CMU. ATP formation then became strongly inhibited. (It seems likely that a similar dependence on photochemically generated oxygen accounts for the results reported by Vennesland et al. in ref. 56, Table 2.) Photophosphorylations catalyzed by phenazine methosulfate were resistant to inhibition by CMU and o-phenanthroline,53 regardless whether the reaction was carried out under conditions favoring the anaerobic cyclic, or the oxygen-dependent pseudocyclic pathway (Tables 2 and 3). Concluding Remarks.-From the standpoint of cellular physiology, the role of oxygen in ATP formation in photosynthesis provides an interesting contrast to the role of oxygen in ATP formation in respiration. The high thermodynamic efficiency of respiration, and its superiority over fermentation in using the free energy of the degradation of foodstuffs for synthesis of the pyrophosphate bonds of ATP, is made possible by the key role that molecular oxygen plays as the terminal electron acceptor in oxidative phosphorylation by mitochondria. Only with oxygen can the oxidation of foodstuffs be complete (to CO a ld water) and their free energy released in full for metabolic purposes. In the last seven years biochelnical investigations of photosynthesis in cell-free systems revealed no such role for oxygen in ATP formation by chloroplasts and bacterial particles. Here, ATP formation occurs in cyclic photophosphorylation under strictly anaerobic conditions where the participation of oxygen is excluded. I i noncyclic photophosphorylation, oxygen may replace TPN as a terminal electron acceptor, but, as discussed later, this results in a loss instead of a gain in the yield of physiologically useful energy. The strictly anaerobic character of cyclic photophosphorylation in bacterial preparations needs no elaboration since photosynthesis in. bacteria is firmly established as an anaerobic process.7 In isolated chloroplasts, anaerobic cyclic photophosphorylation has been demonstrated with phenazine methosulfate as a catalyst, and also with vitamin K3 and FMN. With the latter, a far more rigid control of experimental conditions was required than with phenazine methosulfate. Special experimental safeguards are needed for demonstrating in chloroplasts an anaerobic cyclic photophosphorylation catalyzed by vitamin K3 and FMN, because these substances, when reduced' photochemically, readily react with small concentrations of molecular oxygen. Phenazine methosulfate also reacts readily VOL. 47, 1961 BIOCHEMISTRY: ARNON ET AL. 1331 with oxygen, but its affinity for oxidized cytochromes82 seems to be greater than for oxygen; it is thus able to catalyze a cyclic electron flow (Fig. 8) even in the presence of molecular oxygen. Noncyclic photophosphorylation consists of ATP formation that is coupled with the reduction of pyridine nucleotide and the oxidation of an exogenous electron donor. The process, which was originally found in isolated chloroplasts,22 has now also been demonstrated in bacterial particles.59 The difference between noncyclic photophosphorylation in the two cases centers on the electron donor. For isolated chloroplasts the electron donors are hydroxyl ions which on oxidation form oxygen; this accounts for oxygen evolution in plant photosynthesis. Photosynthetic bacteria cannot use hydroxyl ions but only such electron donors as organic acids or reduced sulfur compounds6' 7; this accounts for the absence of oxygen evolution in bacterial photosynthesis. Instead of oxygen, oxidized bacterial electron donors are formed, which may be further metabolized or may accumulate and yield, for example, elemental sulfur.6 The role of oxygen in photosynthesis is, therefore, only incidental and not vital to the energy transformation reactions, as it is in respiration. Oxygen is evolved because it is an excreted product of the oxidation of hydroxyl ions-the electron donor in noncyclic photophosphorylation by chloroplasts. Experimentally, when another electron donor is substituted for hydroxyl ions, noncyclic photophosphorylation in chloroplasts becomes converted to the bacterial type: ATP formation is coupled with pyridine nucleotide reduction but no oxygen is evolved.57 In a cell-free system, oxygen evolution by chloroplasts will be observed manometrically only under certain conditions. First, the acceptor for the electrons "expelled" from chlorophyll in the primary photochemical reaction must not react preferentially with the components of the photosynthetic electron transport chain (cf. Trebst and Eck83). If it does react, as in the case of phenazine methosulfate, then cyclic photophosphorylation, in which oxygen is not evolved, will result. Second, the electron acceptor must not be easily oxidized by molecular oxygen. This is often, but not always, related to redox potential. Thus, TPN with a redox potential, at pH 7, of E'o = -324 mV shares this property with ferricyanide (E'o = 360 mV) and p-benzoquinone (E'o = 293 mV). In the intact plant, oxygen is normally evolved during photosynthesis because the TPNH2 formed in noncyclic photophosphorylation is not reoxidized by the oxygen formed, but is used up in carbon assimilation. A special case of noncyclic photophosphorylation by isolated chloroplasts occurs when oxygen acts as the terminal electron acceptor and becomes reduced to hydrogen peroxide. The photoreduction of oxygen by isolated chloroplasts, discovered by Mehler,72 has been investigated by Brown and Good84 and Good and Hill,76 who suggested that it is "an artifact" 84 and that "the direct reduction of oxygen does not ordinarily occur to a significant extent in vivo." 76 In a cell-free system containing microcatalytic amounts of FMN or vitamin K3, oxygen may replace TPN as the electron acceptor in noncyclic photophosphorylation by chloroplasts and give rise to an oxygen-dependent, pseudocyclic photophosphorylation. The substitution of oxygen for TPN is analogous to the substitution of ferricyanide for TPN. ATP is still formed in each case, but the reductant produced is no longer TPNH2 but either ferrocyanide, or, when oxygen 1332 BIOCHEMISTRY: ARNON ET AL. PROC. N. A. S. is the electron acceptor, H202 or water. Thus, the intervention of oxygen in noncyclic photophosphorylation results in a loss of biochemically useful energy which would have normally been used for forming a strong reductant, i.e. TPNH2. When oxygen is the electron acceptor and water is the final product of oxygen reduction (equations (6) and (7)) the energy loss is equivalent to a redox potential of 1139 mV (difference between. Eo' -815 mV for the water/oxygen system an(d Eo' = -324 mV for the TPN/TPNH2 system). If hydrogen peroxide is allowed to accumulate as the final product of the oxygen reduction, the computed energy loss would be equivalent to a redox potential of 1006 mV. * The following abbreviations are used: ATP, adenosine triphosphate; ADP, adenosine diphosphate; P, orthophosphate; TPN, TPNH2, oxidized and reduced forms of tiiphosphopyridine nucleotide; FMN, riboflavin phosphate (flavin mononucleotide); PMS, phenazine methosulfate. t Aided by grants from the National Institutes of Health and the Office of Naval Research. 1 Arnon, D. I., Bull. Torrey Botan. Club, 88 (1961). 2 Ingenhousz, J., Essay on the food of plants and the renovation of soils, London (1796). 3Lebedev, A., Biochem. Z., 7, 1 (1907). 4Lebedev, A., Ber. deut. botan. Ges., 27, 598 (1909). 6 Lebedev, A. F., Izvest. Donskovo Gosudarst. Univ., 3, 25 (1921), translated in Am. Rev. Soviet Med., 5, 15 (1948). 6Van Niel, C. B., Arch. Mikrobiol., 3, 1 (1931). 7 Van Niel, C. B., in Photosynthesis in Plants, ed. J. Franck and W. E. Loomis (Ames, Iowa: Iowa State College Press, 1949), p. 437. 8 Hill, R., Proc. Roy. Soc. (London), B 127, 209 (1939). " Hill, R., Symposia Soc. Exptl. Biol., 5, 221 (1951). 10 Burk, D., and 0. Warburg, Z. Naturforsch., 6b, 12 (1951). "1 Warburg, O., Science, 128, 68 (1958). 12 Warburg, O., G. Krippahl, H. S. Gewitz, and W. Z. Volker, Z. Naturforsch., 14b, 712 (1959). 13 Vishniac, W., and S. Ochoa, J. Biol. Chem., 198, 501 (1952). 14 Holzer, H., Z. Naturforsch., 6b, 424 (1951). 15 Bassham, J. A., A. A. Benson, L. D. Kay, A. Z. Harris, A. T. Wilson, and M. Calvin, J. Am. Chem. Soc., 76, 1760 (1954). 16 Arnon, D. I., M. B. Allen, and F. R. Whatley, Nature, 174, 394 (1954). 17 Frenkel, A. W., J. Am. Chem. Soc., 76, 5568 (1954). 18 Arnon, D. I., Nature, 184, 10 (1959). 19 Arnon, D. I., M. Losada, M. Nozaki, and K. Tagawa, ibid., 190, 601 (1961). 20 Stanier, R. Y., Bacteriol. Rev., 25, 1 (1961). 21 Krogmann, D. W., and B. Vennesland, J. Biol. Chem., 234, 2205 (1959). 22 Arnon, D. I., F. R. Whatley, and M. B. Allen, Science, 127, 1026 (1958). 23 Arnon, D. I., F. R. Whatley, and M. B. Allen, Biochim. et Biophys. Acta, 32, 47 (1959). 24 Trebst, A. V., H. Y. Tsujimoto, and D. I. Arnon, Nature, 182, 351 (1958). 25 Losada, M., A. V. Trebst, and D. I. Arnon, J. Biol. Chem., 235, 832 (1960). 28 Trebst, A. V., M. Losada, and D. I. Arnon, ibid., 234, 3055 (1959). 27 Arnon, D. I., in Enzymes: Units of Biological Structure and Function, ed. 0. H. Gaebler (New York: Academic Press, Inc., 1956), p. 279. 28 Arnon, D. I., M. B. Allen, and F. R. Whatley, Biochim. et Biophys. Acta, 20, 449 (1956). 29 Whatley, F. R., M. B. Allen, L. L. Rosenberg, J. B. Capindale, and D. I. Arnon, ibid., 20, 462 (1956). 30 Calvin, M., Proc. Intern. Congr. Biochem., 3rd Congr., Brussels, 1955 (New York: Academic Press, Inc., 1956), p. 211. 31 Vishniac, W., B. L. Horecker, and S. Ochoa, Advances in Enzymol., 19, 1 (1957). 32 Trudinger, P. A., Biochem. J., 64, 274 (1956). 33 Aubert, J. P., G. Milhaud, and J. Millet, Ann. inst. Pasteur, 92, 515 (1957). 34 McFadden, B. A., and D. E. Atkinson, Arch. Biochem. Biophys., 66, 16 (1957); McFadden, B. A., J. Bacteriol., 77, 339 (1959). VOL. 47, 1961 BIOCHEMISTRY: ARNON ET AL. 1333 '5 Strehler, B. L., Arch. Biochem. Biophys., 43, 67 (1953). 36 Gaffron, H., in Autotrophic Microorganisms, ed. B. A. Fry and J. L. Peel (Cambridge, England: Cambridge University Press, 1954), p. 163. 37 Avron, M., and A. T. Jagendorf, Nature, 179, 428 (1957). 38 Avron, M., A. T. Jagendorf, and M. Evans, Biochim. et Biophys. Acta, 26, 262 (1957). 39 Jagendorf, A. T., in The Photochemical Apparatus, Brookhaven Symposia in Biology, No. 11 (1958). 40 Jagendorf, A. T., and M. Avron, J. Biol. Chem., 231, 277 (1958). 41 Wessels, J. S. C., Biochim. et Biophys. Acta, 25, 97 (1957). 42 Ibid., 29, 113 (1958). 43 Chow, C. T., and B. Vennesland, Plant Physiol., 32 (Supp.), iv (1957). 44 Nakamoto, T., D. W. Krogmann, and B. Vennesland, J. Biol. Chem., 234, 2783 (1959). 46 Hill, R., and D. A. Walker, Plant Physiol., 34, 240 (1959). 46 Whatley, F. R., M. B. Allen, A. V. Trebst, and D. I. Arnon, ibid., 35, 188 (1960). 47 Other accounts of the discovery of photosynthetic phosphorylation and CO2 assimilation by isolated chloroplasts are sometimes found in the literature. Thus in 1956, J. A. Bassham and M. Calvin, in Currents in Biochemical Research, ed. D. B. Green (New York: Interscience, 1956), ascribed the discovery of CO2 assimilation by isolated chloroplasts to Boychenko and Baranov.48 In 1959, Calvin (ref. 49, p. 152) ascribed the discovery of both CO2 assimilation and ATP synthesis by isolated chloroplasts to his own laboratory. 48 Boychenko, E. A., and V. I. Baranov, Doklady Akad. Nauk S.S.S.R., 95, 1025 (1954); Chem. Abstr., 48, 8881 (1954). 49 Calvin, M., Rev. Modern Phys., 31, 147 (1959); also in Biophysical Science-A Study Program, ed. J. C. Oncley (New York: John Wiley and Sons, 1959), p. 152. 50 Williams, A. M., Biochim. et Biophys. Acta, 19, 570 (1956). 51 Thomas, J. B., and A. M. Haans, ibid., 18, 286 (1955). 52 Petrack, B., and F. Lipmann, in Light and Life, ed. W. D. McElroy and B. Glass (Baltimore: Johns Hopkins Press, 1961), p. 621. 6 Arnon, D. I., ibid., p. 489. 64"Frenkel, A. W., J. Biol. Chem., 222, 823 (1956). 56 Geller, D. M., "Photophosphorylation by Rhodospirillum rubrum preparations," Doctoral Dissertation, Div. Med. Sci., Harvard University (1957). 66 Vennesland, B., T. Nakamoto, and B. Stern, in Light and Life, ed. W. D. McElroy and B. Glass (Baltimore: Johns Hopkins University Press, 1961), p. 609. 67 Losada, M., F. R. Whatley, and D. I. Arnon, Nature, 190, 606 (1961.) 58 Duysens, L. N. M., J. Amesz, and B. M. Kamp, ibid., 190, 510 (1961). 69 Nozaki, M., K. Tagawa, and D. I. Arnon, these PROCEEDINGS, 47, 1334 (1961). 60 Hill, R., and W. D. Bonner, in Light and Life, ed. W. D. McElroy and B. Glass (Baltimore: Johns Hopkins University Press, 1961), p. 424. 61 Crane, F. L., Plant Physiol., 34, 128 (1959); Bishop, N. I., these PROCEEDINGS, 45, 1696 (1959). 62 Krogmann, D.. W., Biochem. Biophys., Research Commun., 4, 275 (1961). 63Trebst, A. V., M. Losada, and D. I. Arnon, J. Biol. Chem., 235, 840 (1960). 64 Krogmann, D. W., A. T. Jagendorf, and M. Avron, Plant Physiol., 34, 272 (1959). 65 Chance, B., and J. M. Olson, Arch. Biochem. Biophys., 88, 54 (1960). " Marrb, E., and E. Forti, Science, 126, 976 (1957). 67 French, C. S., in Handbuch der Pftanzenphysiolqgie, ed. W. Ruhland (Heidelberg: Springer, 1960), vol. 5, part 1, p. 252. 68 Avron, M., D. W. Krogmann, and A. T. Jagendorf, Biochim. et Biophys. Acta, 30, 144 (1958). 69 Davenport, H. E., Biochem. J., 77, 471 (1960). 70 Arnon, D. I., paper presented at the Cell Symposium, Amer. Assn. for the Advancement of Sci., Berkeley Meeting (1954); Science, 122, 9 (1955). 71 Arnon, D. I., M. B. Allen, and F. R. Whatley, Biochim. et Biophys. Acta, 20, 449 (1956). 72 Mehler, A. H., Arch. Biochem. Biophys., 33, 65 (1951). 73 Ibid., 34, 339 (1951). 74 Mehler, A. H., and A. H. Brown, ibid., 38, 365 (1952). 75 Trebst, A., and H. Eck, Z. Naturforsch. (in press). 1334 BIOCHEMISTRY: NOZAKI ET AL. PROC. N. A. S. 76 Good, N., and R. Hill, Arch. Biochem. Biophys., 57, 355 (1955). 77 Krogmann, D. W., J. Biol. Chem., 235, 3630 (1960). 78 Arnon, D. I., in Handbuch der Pflanzenphysiologie, ed. W. Ruhland (Heidelberg: Springer, 1960), vol. 5, part 1, p. 773. 79 Baltscheffsky, H., Svensk Kemisk Tidskrift, 72, 4 (1960). 'I Warburg, O., in Heavy Metal Prosthetic Groups and Enzyme Action (Oxford, England: Clarendon Press, 1949), p. 213. 81 Wessels, J. S. C., Biochim. et Biophys. Acta, 19, 548 (1956). 82 Massey, V., ibid., 34, 255 (1959). 83Trebst, A., and H. Eck, Z. Naturforsch., 16b, 44 (1961). 84 Brown, A. H., and N. Good, Arch. Biochem. Biophys., 57, 340 (1955). NONCYCLIC PHOTOPHOSPHORYLATION IN PHOTOSYNTHETIC BACTERIA BY M. NOZAKI, K. TAGAWA, AND DANIEL I. ARNON* LABORATORY OF CELL PHYSIOLOGY, UNIVERSITY OF CALIFORNIA, BERKELEY Communicated August 2, 1961 Noncyclic photophosphorylation is the formation of adenosine triphosphate in a light-induced electron flow from an external electron donor to an electron acceptor such as pyridine nucleotide. In photosynthesis of green plants the electron donors are the hydroxyl ions of water and hence the over-all reaction also produces, aside from ATP and PNH2,l molecular oxygen, which is liberated when OH- is photooxidized (see Fig. 1 in preceding article2). Photosynthetic bacteria cannot use OH- as an electron donor for photosynthesis but use a variety of reductants stronger than OH-, among them hydrogen gas, succinate, or thiosulfate. With hydrogen gas no input of light energy is required for the reduction of pyridine nucleotide, since hydrogen gas, with the aid of bacterial hydrogenase, can reduce pyridine nucleotide in the dark.3 The contribution of light to bacterial photosynthesis is then limited to the formation of ATP by cyclic photophosphorylation.4 I With other electron donors such as, for example, succinate, the additional input of energy that is necessary to reduce pyridine nucleotide is provided by light. Thus, in bacterial photosynthesis with succinate as the electron donor, light energy serves a dual purpose: it supplies ATP by cyclic photophosphorylation and, by photooxidizing the electron donor, it also provides electrons for reducing pyridine nucleotides by a noncyclic electron flow mechanism. Photoreduction of pyridine nucleotide (DPN) by cell-free preparations of Rhodospirillum rubrum6 I and Chromatium3 has been observed but attempts to find a simultaneous ATP formation, analogous to the noncyclic photophosphorylation in chloroplasts, led to negative results.7' 8 As pointed out by Vernon and Ash,8 a demonstration of ATP formation by a noncyclic electron flow mechanism is more difficult in chromatophores than in chloroplasts because in chromatophores it cannot be experimentally distinguished from a simultaneous ATP formation by cyclic photophosphorylation. In chromatophores, a vigorous cyclic photophosphorylation occurs under anaerobic conditions and without added cofactors,3 whereas in chloroplasts cyclic photophosphorylation is always under experimental