Kinetics of the generation of a photo-induced electric potential in chromatophores of photosynthetizing bacteria

Flash-induced formation of an electric potential difference (delta psi) was monitored by a direct method in chromatophores associated with the collodion phospholipid membrane. In Rhodospirillum rubrum and Rhodopseudomonas sphaeriodes chromatophores, the kinetics of delta psi generation exhibit fast (tau less than or equal to 0.3 microseconds) and slow (tau congruent to 200 microseconds) phases, the latter observed in the presence of exogenous quinones. Comparison of the kinetic and potentiometric characteristics of the process with those of electron transport reactions suggests that the fast phase of delta psi rise is due to charge separation between the primary electron donor, P870, and primary electron acceptor QIFe; the slow phase, which is inhibited by o-phenanthroline, is due to electron donation from QIFe to the secondary acceptor, quinone QII. The kinetics of delta psi decay include components arising form the recombination of primary separated charges (tau congruent to 30 ms) and from the passive discharge of the membrane (tau congruent to 400 ms; tau congruent to 1400 ms). From a redox titration of the photo-induced electric signal and the photo-induced absorption changes of P870 at different pH meanings, the value of pK for the primary acceptor FeQI was found to be 7.4 in Rps. sphaeroides chromatophores. In Chromatium minutissimum, a phase ( tau congruent to 20 microseconds) was observed in addition to those seen in Rps. sphaeroids and R. rubrum which was explained by the reduction of P890+ from the high potential cytochrome c555. Possible distribution of the electron transport components in the chromatophore membrane are discussed.     

Kinetic factors limiting the synthesis of ATP by chromatophores exposed to short flash excitation.

ATP synthesis was measured after chromatophores from Rhodopseudomonas capsulata had been subjected to illumination by single turnover flashes fired at variable frequencies. Three processes were examined, which under different conditions can limit the net yield of ATP. (1) A process with an apparent relaxation time of 10-20 ms. This reaction probably limits the rate of ATP synthesis in continuous illumination. It has similar time dependence to the stimulation of the carotenoid shift decay by ADP after a single flash. (2) An active state of the ATPase only persists when the chromatophores are excited more often than once in 10 s. This state decays with similar kinetics to the entire carotenoid shift decay. Full activation is achieved after two flashes. (1) and (2) are not significantly affected by concentrations of antimycin A sufficient to block electron flow through the cytochrome b/c2 oxidoreductase and abolish phase III in the generation of the carotenoid shift. (3) In the presence of antimycin A, after the third, fourth and subsequent flashes ATP synthesis is limited by the quantity of reducing equivalents transported through the reaction centre rather than by the level of the electrochemical proton gradient.     

Flash-induced photophosphorylation in Rhodospirillum rubrum chromatophores. I. The relationship between cytochrome c-420 content and photophosphorylation.

The content of cytochrome c-420 in Rhodospirillum rubrum chromatophores prepared by grinding with alumina is 5--10% of that in whole cells, and 20--40% in chromatophores by 'French' pressing. Flash-induced phosphorylation of various chromatophores which varied in cytochrome content from 7 to 40% is proportional to the cytochrome content. Extrapolating the cytochrome c-420 content to that observed in whole cells, a ratio ATP/P+X- near 1 is calculated. At low flash intensity the phosphorylation per flash is proportional to flash energy. Photophosphorylation in flashes given after a time of several minutes is only slightly dependent on the number of flashes. If the flashes are spaced from 0.1 to 10 s, relative phosphorylation in the first flash is about 70% and in the second 90+ of that observed in the following flashes. Proton binding is not affected by the cytochrome c-420 content and a ratio of H+/P+x- of 2.3 was found. These results can be explained by a working hypothesis in which charge separation occurring at one reaction centre and the resulting electron transport mediated amongst others by c-420, results in the injection of two protons into an ATPase, this in contrast to a chemiosmotic mechanism, where the protons are released in the chromatophore inner space.     

Succinate dehydrogenase in Rhodopseudomonas sphaeroides: subunit composition and immunocross-reactivity with other related bacteria.

Antibodies were raised against the succinate dehydrogenase (SDH) present in the chromatophores of phototrophically grown Rhodopseudomonas sphaeroides. Crossed immunoelectrophoresis experiments indicated that the SDH present in the cytoplasmic membranes of heterotrophically grown R. sphaeroides is probably the same enzyme observed in the chromatophores. The enzyme was extracted by Triton X-100 in a form which consisted of only two subunits (molecular weight, 68,000 and 30,000) and was not associated with a cytochrome b. The antibodies directed against SDH from R. sphaeroides showed no immunocross-reactivity with SDH from phylogenetically related bacterial species, including Rhodopseudomonas capsulata, Paracoccus denitrificans, Rhodopseudomonas palustris, Rhodospirillum rubrum, and Rhodospirillum fulvum.     

Nanosecond fluorescence from chromatophores of Rhodopseudomonas sphaeroides and Rhodospirillum rubrum

Single-photon counting techniques were used to measure the fluorescence decay from Rhodopseudomonas sphaeroides and Rhodospirillum rubrum chromatophores after excitation with a 25-ps, 600-nm laser pulse. Electron transfer was blocked beyond the initial radical-pair state (PF) by chemical reduction of the quinone that serves as the next electron acceptor. Under these conditions, the fluorescence decays with multiphasic kinetics and at least three exponential decay components are required to describe the delayed fluorescence. Weak magnetic fields cause a small increase in the decay time of the longest component. The components of the delayed fluorescence are similar to those found previously with isolated reaction centers. We interpret the multi-exponential decay in terms of two small (0.01-0.02 eV) relaxations in the free energy of PF, as suggested previously for reaction centers. From the initial amplitudes of the delayed fluorescence, it is possible to calculate the standard free-energy difference between the earliest resolved form of PF and the excited singlet state of the antenna complexes in R. rubrum strains S1 and G9. The free-energy gap is found to be about 0.10 eV. It also is possible to calculate the standard free-energy difference between PF and the excited singlet state of the reaction center bacteriochlorophyll dimer (P). Values of 0.17 to 0.19 eV were found in both R. rubrum strains and also in Rps. sphaeroides strain 2.4.1. This free-energy gap agrees well with the standard free-energy difference between PF and P determined previously for reaction centers isolated from Rps. sphaeroides strain R26. The temperature dependence of the delayed fluorescence amplitudes between 180 K and 295 K is qualitatively different in isolated reaction centers and chromatophores. However, the temperature dependence of the calculated standard free-energy difference between P* and PF is similar in reaction centers and chromatophores of Rps. sphaeroides. The different temperature dependence of the fluorescence amplitudes in reaction centers and chromatophores arises because the free-energy difference between P* and the excited antenna is dominated by the entropy change associated with delocalization of the excitation in the antenna. We conclude that the state PF is similar in isolated reaction centers and in the intact photosynthetic membrane. Chromatophores from Rps. sphaeroides strain R-26 exhibit an anomalous fluorescence component that could reflect heterogeneity in their antenna.     

Kinetics of ubisemiquinone redox changes in primary reactions of bacterial photosynthesis

Absorbance changes at 450 nm of the semiquinone form of the secondary electron acceptor were studied in chromatophores of Rhodospirillum rubrum. When chromatophores are illuminated by a series of single turnover flashes ubisemiquinone is formed and destroyed on alternate flashes at ambient redox potential from 100 to 250 mV. A simple kinetic model of the binary oscillations is suggested. On the base of the model it is shown that the rate constant of electron transfer from primary to secondary quinone after the first flash is larger that after the second flash. Cooperativity in electron transfer from primary to secondary quinone can be explained by electrostatic interactions of charged carriers.     

ATP-synthase of Rhodobacter capsulatus: coupling of proton flow through F0 to reactions in F1 under the ATP synthesis and slip conditions

A stepwise increasing membrane potential was generated in chromatophores of the phototrophic bacterium Rhodobacter capsulatus by illumination with short flashes of light. Proton transfer through ATP-synthase (measured by electrochromic carotenoid bandshift and by pH-indicators) and ATP release (measured by luminescence of luciferin-luciferase) were monitored. The ratio between the amount of protons translocated by F0F1 and the ATP yield decreased with the flash number from an apparent value of 13 after the first flash to about 5 when averaged over three flashes. In the absence of ADP, protons slipped through F0F1. The proton transfer through F0F1 after the first flash contained two kinetic components, of about 6 ms and 20 ms both under the ATP synthesis conditions and under slip. The slower component of proton transfer was substantially suppressed in the absence of ADP. We attribute our observations to the mechanism of energy storage in the ATP-synthase needed to couple the transfer of four protons with the synthesis of one molecule of ATP. Most probably, the transfer of initial protons of each tetrad creates a strain in the enzyme that slows the translocation of the following protons.     

Kinetic factors limiting the synthesis of ATP by chromatophores exposed to short flash excitation

ATP synthesis was measured after chromatophores from Rhodopseudomonas capsulata had been subjected to illumination by single turnover flashes fired at variable frequencies. Three processes were examined, which under different conditions can limit the net yield of ATP.(1) A process with an apparent relaxation time of 10–20 ms. This reaction probably limits the rate of ATP synthesis in continuous illumination. It has a similar time dependence to the stimulation of the carotenoid shift decay by ADP after a single flash.(2) An active state of the ATPase only persists when the chromatophores are excited more often than once in 10 s. This state decays with similar kinetics to the entire carotenoid shift decay. Full activation is achieved after two flashes.(1) and (2) are not significantly affected by concentrations of antimycin A sufficient to block electron flow through the cytochrome $\text{b}\text{c}{}_2$ oxidoreductase and abolish phase III in the generation of the carotenoid shift.(3) In the presence of antimycin A, after the third, fourth and subsequent flashes ATP synthesis is limited by the quantity of reducing equivalents transported through the reaction centre rather than by the level of the electrochemical proton gradient.     

Delayed fluorescence from Rhodopseudomonas sphaeroides following single flashes.

Delayed fluorescence from Rhodopseudomonas sphaeroides chromatophores was studied with the use of short flashes for excitation. Although the delayed fluorescence probably arises from a back-reaction between the oxidized reaction center bacteriochlorophyll complex (P+) and the reduced electron acceptor (X-), the decay of delayed fluorescence after a flash is much faster (tau1/2 approximately 120 mus) than the decay of P+X-. The rapid decay of delayed fluorescence is not due to the uptake of a proton from the solution, nor to a change in membrane potential. It correlates with small optical absorbance changes at 450 and 770 nm which could reflect a change in the state of X-. The intensity of the delayed fluorescence is 11-18-fold greater if the excitation flashes are spaced 2 s apart than it is if they are 30 s apart. The enhancement of delayed fluorescence at high flash repetition rates occurs only at redox potentials which are low enough (less than +240 mV) so that electron donors are available to reduce P+X- to PX- in part of the reaction center population. The enhancement decays between flashes as PX- is reoxidized to PX, as measured by the recovery of photochemical activity. Evidently, the reduction of P+X- to PX- leads to the storage of free energy that can be used on a subsequent flash to promote delayed fluorescence. The reduction of P+X- also is associated with a carotenoid spectral shift which decays as PX- is reoxidized to PX. Although this suggests that the free energy which supports the delayed fluorescence might be stored as a membrane potential, the ionophore gramicidin D only partially inhibits the enhancement of delayed fluorescence. With widely separated flashes, gramicidin has no effect on delayed fluorescence. At redox potentials low enough to keep X fully reduced, delayed fluorescence of the type described above does not occur, but one can detect weak luminescence which probably is due to phosphorescence of a protoporphyrin.     

Dichroism of bacteriochlorophyll in chromatophores of photosynthetic bacteria.

The dichroism was measured in films of air-dried and, consequently, flattened chromatophores of Chromatium vinosum, Rhodopseudomonas sphaeroides and Rhodospirillum rubrum. The values (deltaA/A) of dichroism in C. vinosum were found to be -1.05 at 590 nm and 0.75 in the near infrared region. The values of dichroism in R. sphaeroides were -0.70 at 590 nm and 0.80 at 870 nm. The values of dichroism in R. rubrum were -1.45 at 590 nm and 0.97 at 870 nm.     

The extent of the stimulated electrical potential decay under phosphorylating conditions and the H+/ATP ratio in Rhodopseudomonas sphaeroides chromatophores following short flash excitation.

1. In chromatophores from Rps. sphaeroides, the stimulation by ADP and Pi of the electric potential decay indicated by the carotenoid shift is greater than the stimulation of the decay of pH change indicated by the colour change of added cresol red under similar conditions. This difference is attributed to H+ consumption during the synthesis of ATP. The ratio of H+ translocated across the membrane to ATP synthesized was estimated to be approximately 1.7 H+/ATP. 2. The stimulation of the electrical potential decay by ADP and Pi was found to be a constant fraction (10%) of the total decay when the flash intensity was varied. No 'critical' or 'threshold' potential was observed. 3. The stimulated electrical potential decay after a second flash, given within a few seconds of the first, was related to the amplitude of the electrical potential produced by the second flash (10%) but neither to the dark time between the flashes, nor to the total extent of the electrical potential above the dark level. These results are consistent with two hypotheses (a) the chromatophores are a mixed population of vesicles, only a small fraction (10%) of which possess an active ATP synthesizing system (b) the activity of the ATP synthesizing system, though driven by a proton motive force, is controlled by electron transport processess. If alternative (a) is correct then the overall single turnover flash yield of 1 ATP per 1470 bacteriochlorophyll measured in (1) would mean that the yield of the active vesicles is approximately 10 ATP per 1470 bacteriochlorophyll or 30 ATP per vesicle. 4. The stimulation of the electrical potential decay by ADP and Pi is approximately 40% less in antimycin-treated chromatophores. It is shown that this is probably a consequence of antimycin-inhibited H+-release on the inside of the chromatophore vesicles following a flash.     

The incorporation of reaction centres into membranes from a bacteriochlorophyll-less mutant of Rhodopseudomonas sphaeroides.

Reaction centres purified from a blue-green mutant R-26 of Rhodopseudomonas sphaeroides can be incorporated into bacteriochlorophyll-less membranes purified from an aerobically-grown bacteriochlorophyll-less mutant 01 of R. sphaeroides. This can be accomplished by raising the temperature of the mixture or by addition of the detergent sodium cholate and its subsequent removal by dilution or dialysis. Optimum conditions for the reconstitution are at 4 degrees C in the presence of 1% cholate and soybean phospholipid (2 : 1, w/w, with membrane protein). Isopycnic sucrose density gradient centrifugation of such preparations shows that reaction centres and light-harvesting pigment-protein complex bind to the membranes. Reconstituted membranes exhibit light-induced steady-state cytochrome absorbance changes resembling those observed in chromatophores prepared from the photosynthetically-grown mutant R-26. The effect on these absorbance changes of varying reaction centre content in the membrane has been studied, and the time course of the interaction between 01 membrane cytochrome c2 and added reaction centre examined. Cytochrome b photoreduction and cytochrome c2 photo-oxidation were observed in the reconstituted preparation; each increased following the addition of antimycin A, suggesting that a cyclic light-driven system had been reconstituted.     

Reactions between primary and secondary acceptors of photosystem II in Chlorella pyrenoidosa under anaerobic conditions as studied by chlorophyll fluorescence.

The kinetics of the fluorescence yield phi of chlorophyll a in Chlorella pyrenoidosa were studied under anaerobic conditions in the time range from 50 mus to several minutes after short (t 1/2 = 30 ns or 5 mus) saturating flashes. The fluorescence yield "in the dark" increased from phi = 1 at the beginning to phi approximately 5 in about 3 h when single flashes separated by dark intervals of about 3 min were given. After one saturating flash, phi increased to a maximum value (4-5) at 50 mus, then phi decreased to about 3 with a half time of about 10 ms and to the initial value with a half time of about 2 s. When two flashes separated by 0.2 s were given, the first phase of the decrease after the second flash occurred within 2 ms. After one flash given at high initial fluorescence yield, the 10-ms decay was followed by a 10 s increase to the initial value. After the two flashes 0.2 s apart, the rapid decay was not followed by a slow increase. These and other experiments provided additional evidence for and extend an earlier hypothesis concerning the acceptor complex of Photosystem II (Bouges-Bocquet, B. (1973) Biochim. Biophys. Acta 314, 250-256; Velthuys, B. R. and Amesz. J. (1974) Biochim. Biophys. Acta 333, 85-94): reaction center 2 contains an acceptor complex QR consisting of an electron-transferring primary acceptor molecule Q, and a secondary electron acceptor R, which can accept two electrons in succession, but transfers two electrons simultaneously to a molecule of the tertiary acceptor pool, containing plastoquinone (A). Furthermore, the kinetics indicate that 2 reactions centers of System I, excited by a short flash, cooperate directly or indirectly in oxidizing a plastohydroquinone molecule (A2-). If initially all components between photoreaction 1 and 2 are in the reduced state the following sequence of reactions occurs after a flash has oxidised A2- via System I: Q-R2- + A leads to Q-R + A2- leads to QR- + A2-. During anaerobiosis two slow reactions manifest themselves: the reduction of R (and A) within 1 s, presumably by an endogenous electron donor D1, and the reduction of Q in about 10 s when R is in the state R- and A in the state A2-. An endogenous electron donor, D2, and Q- complete in reducing the photooxidized donor complex of System II in reactions with half times of the order of 1 s.     

The kinetics of carotenoid absorption changes in intact cells of photosynthetic bacteria

The kinetics of carotenoid absorption changes have been measured in intact cells of Rhodopseudomonas capsulata after short flash excitation. The observed changes were consistent with the thesis that they indicate the development and dissipation of membrane potential. In the generation of the absorption changes in anaerobic cells, fast (complete in 0.5 ms) and slow (half-time 3 ms) components can be distinguished. The slow component corresponds kinetically to the rate of cytochrome c re-reduction and is similarly antimycin sensitive. These data are similar to those observed in isolated chromatophores which have been artifically poised with redox mediators. In aerobic intact cells the kinetic profile is altered, mainly because the decay of the carotenoid change is much faster. Inhibition of respiration with KCN leads to flash-induced changes similar to those in anaerobic cells. At least two components can be distinguished in the decay of the carotenoid absorption changes in anaerobic intact cells. Only the faster decay component was inhibited by venturicidin which suggests that it corresponds to H⁺ flux through the F₀F₁-ATPase during ATP synthesis. The contribution of the venturicidin-sensitive decay to the total decay was dependent upon the initial amplitude of the carotenoid absorption change produced by the flash group. This suggests that there is an apparent threshold of membrane potential for ATP synthesis. Supporting evidence was provided by the finding that venturicidin stimulated the steady-state light-induced carotenoid absorption change at high but not at low light intensities. The entire decay of the carotenoid absorption changes was stimulated by carbonyl cyanide p-trifluoromethoxyphenylhydrazone in a manner that can be interpreted as an ionophore catalysing the dissipation of membrane potential.     

Oxidation of cytochrome c2 and of cytochrome c by reaction centers of Rhodospirillum rubrum and Rhodobacter sphaeroides. The effect of ionic strength and of lysine modification on oxidation rates

The oxidation of cytochrome c2 by the photooxidized reaction center bacteriochlorophyll, P+-870, in chromatophores of Rhodospirillum rubrum can be described using second-order kinetics at all ionic strengths. In a system consisting of isolated R. rubrum reaction centers and purified R. rubrum cytochrome c2, the oxidation of cytochrome c2 also follows second-order kinetics. In both cases, the reaction rates at low ionic strength are weakly dependent on the ionic strength. The data suggest that the cytochrome remains mobile at very low ionic strength, since the observed kinetics can be easily explained assuming no significant tight binding of cytochrome c2 to the reaction center. In a system consisting of equine cytochrome c and reaction centers of either R. rubrum or Rhodobacter sphaeroides, the cytochrome c oxidation rate depends more strongly on the ionic strength. The high reaction rates at low ionic strength suggest that a significant portion of the cytochrome is bound. Using equine cytochrome c derivatives modified at specific lysine residues, it was shown that both R. rubrum and Rb. sphaeroides reaction centers react with equine cytochrome c through its exposed heme edge.     

The localized coupling of bacterial photophosphorylation: Effect of antimycin a and N,N-dicyclohexylcarbodiimide) in chromatophores from Rhodopseudomonas sphaeroides,Ga, studied by single turnover event analysis

1. The inhibition by antimycin A of the cyclic electron transfer has been studied in chromatophores from Rhodopseudomonas sphaeroides Ga following an approach based on the analysis of the relaxation kinetics of the reaction center optical changes in pulsed light. The recovery kinetics of the bacteriochlorophyll redox state have been found to be clearly biphasic. The half-times of the fast phase (13 ms) and slow phase (about 400 ms) were not modified by antimycin in a range of concentrations from 0.1 to 9 μM. On the other hand the percentage extent of the fast phase, which reflects the rate of the cyclic electron transfer, was monotonically decreased by increasing concentrations of the inhibitor. This indicates that antimycin decreases progressively the fraction of the photosynthetic units, active in cyclic electron transfer. 2. The ATP yield per flash observed under conditions of controlled inhibition of electron flow was strongly dependent upon the amount of active redox cycles. On the other hand, the amplitude of the carotenoid band shift, which has been demonstrated unequivocally to be correlated to the ATP yield per flash in uninhibited chromatophores, was not affected by antimycin up to a 40% inhibition of electron flow. 3. The effect of a progressive limitation by DCCD in the number of active ATP synthetase complexes on flash-induced phosphorylation has been examined. The decrease in ATP yield observed over a wide range of flash frequencies is related simply to the ATPase activity and to phosphorylation in continuous light, irrespective of the value of the membrane potential, which appears to be stabilized by this inhibitor. 4. As a whole, the results obtained at low concentrations of antimycin and under conditions of partial inhibition by DCCD evidence a localized coupling between the redox reactions and phosphorylation.     

Equilibrium and kinetic parameters for the binding of inhibitors to the QB pocket in bacterial chromatophores: dependence on the state of QA

The equilibrium and kinetic parameters for the binding of various inhibitors to the Q(B) pocket of the bacterial reaction center were investigated in chromatophores from Rhodobacter capsulatus and Rhodobacter sphaeroides. By monitoring the near-IR absorption changes specific to Q(A)(-) and Q(B)(-), we measured the fraction of inhibited centers in the dark and the kinetics and extent of inhibitor displacement after one flash due to the formation of the Q(A)Q(B)(-) state. The inhibitor release rate was much faster for triazines and o-phenanthroline (t(1/2) in the 50 ms to 1 s range) than for stigmatellin (t(1/2) approximately 20 s). For inhibitors with a rapid release rate, the fast phase of P(+) decay observed in the absence of secondary donor reflects the competition between P(+)Q(A)(-) recombination and inhibitor release: it is thus faster than the P(+)Q(A)(-) recombination, and its relative extent is smaller than the fraction of initially inhibited centers. At appropriate inhibitor concentrations, one can have almost total binding in the dark and almost total inhibitor displacement after one flash. Under such conditions, a pair of closely spaced flashes resets the two-electron gate in a single state (Q(A)Q(B)(-)), irrespective of the initial state. The apparent dissociation constant of terbutryn was significantly increased (by a factor of 4-7) in the presence of Q(A)(-), in agreement with the conclusion of Wraight and co-workers [Stein, R. R., et al. (1984) J. Cell. Biochem. 24, 243-259]. We suggest that this effect is essentially due to a tighter binding of ubiquinone in the Q(A)(-) state.     

Secondary electron transfer in reaction centers of Rhodopseudomonas sphaeroides. Out-of-phase periodicity of two for the formation of ubisemiquinone and fully reduced ubiquinone.

Electron transfer between purified reaction centers from Rhodopseudomonas sphaeroides and exogenous ubiquinone has been studied in the presence of electron donors by measurements of light-induced absorbance changes following a sequence of short actinic light flashes. Each odd flash promotes the formation of a molecule of ubisemiquinone; after each even flash the semiquinone disappears and a molecule of the fully reduced quinone appears. We interpret these results by means of a model where a specialized molecule of ubiquinone is reduced by the primary electron acceptor in a one-electron transfer reaction after each flash, and is reoxidized by a molecule of the ubiquinone pool in a two-electron transfer reaction every two flashes.     

Photooxidation and light-induced transport of phenazine methosulfate in chromatophores of purple bacteria

The light-induced interaction of phenazine methosulfate (PMS) with chromatophores of the purple bacteria Rhodospirillum rubrum and Rhodopseudomonas sphaeroides was studied, using an ion-specific electrode. Illumination caused an initial rapid increase in the concentration of methylphenazinium cation (MP+) and a subsequent slow (1-3 min) decrease of the MP+ concentration to a low steady level. The rapid phase of the light-induced MP+ concentration change is specifically enhanced by ascorbate. The slow phase (uptake of MP+ from the medium) is stimulated on addition of valinomycin, which is known to collapse the membrane potential of energized chromatophores, and is partly inhibited by NH4Cl, which enhances the membrane potential in chromatophores. The light-induced uptake of MP+ is sharply stimulated by dibromothymoquinone. It is concluded that the initial rapid increase of the MP+ concentration in the outer medium results from the oxidation of the reduced PMS by photooxidized reaction centers. The slow decrease of the external MP+ concentration is due to active transport of MP+ into the internal space of the chromatophores via a mechanism of a chemiosmotic type. The accumulation of MP+ is directly mediated by the redox reactions of PMS at the outer and inner surfaces of the photosynthetic membrane, which are involved in cyclic electron transport.