Pierre Sebban

Heterogeneity of the P+Q−A recombination kinetics in reaction centers from Rhodopseudomonas viridis: the effects of pH and temperature

The decay of the flash-induced absorbance changes related to the primary electron donor, P + , was measured in reaction centers from Rhodopseudomonas viridis . The decay, which occurs by the recombination or back-reaction of P + Q − A , was found to be not asingle exponential and could be decomposed into two components. At pH 9,in the presence of o -phenanthroline, the two phases exhibited rate constants of 350 ± 25 s −1 ( k slow ) and 1400 ± 100 s −1 ( k fast ). In the absence of o -phenanthroline, in Q B -depleted reaction centers, the rate constants were 370 ± 25 and 1700 ± 100 s −1 , respectively. k slow and k fast display temperature-dependences similar to those previously described by Shopes and Wraight ((1987) Biochim. Biophys. Acta 893, 409–425) for the total decay component. Down to 240 K, the temperature-dependences of k slow and k fast indicate thermally activated processes, whichare presumed to proceed by repopulation of the P + I − statefrom P + Q − A . At temperatures below 240 K, an activationless process dominates with rate constants for the two phases whichdepend somewhat on the concentration of glycerol. In aqueous buffer, the limiting values at low temperatures, k T slow and k T fast , were determined to be 155 ± 10 and 460 ± 60 s −1 , respectively. After correction for these limiting rates, the ambient temperature-dependences were linear in an Arrhenius plot. At pH 9, the activation energies (enthalpies, Δ H ) were 0.258 ± 0.009 and 0.278 ± 0.021 eV for the slow and fast phases, respectively. The entropies of activation were also quite large. Consequently the activation free energies (Δ G ) were inverted compared to the activation enthalpies, i.e., Δ G slow = 0.292 ± 0.011 eV and Δ G fast = 0.260 ± 0.022 eV. k slow and k fast were also significantly pH-dependent. As the pH was raised from pH 6, both rate constants showed a shallow minimum at pH 7.5–8, and accelerated significantly as the pH was raised further. The pH-dependence of both components was more marked in the absence than in the presence of o -phenanthroline. The relative proportions of k slow and k fast were also strongly pH-dependent in the range pH 5.5–11.5. The amplitude curves displayed two waves that were dependent on ionic conditions and on the presence of o -phenanthroline. The pH-and temperature-dependences of the rates and amplitudes of the recombination components are interpreted in terms of multiple protonation states of the reaction center affecting the energetics of the thermally activated recombination pathway. The pH-dependence of the rates can be understood as arising from the electrostatic stabilization of the charge-separation states, P + Q − A and P + I − , due to the interaction of at leasttwo protonation sites with the species Q − A and I − . The pH-dependence of the recombination rates, in the range pH 5.5–10, couldbe accounted for in this way, with two distinct values for p K Q A : one with p K Q A about 6, for which protonation resulted in the relative stabilization of P + I − , decreasing Δ G and leading to acceleration of the back-reaction, and one with p K Q A about 9, which relatively stabilized P + Q − A in the protonated state, increasing Δ G and causing slower recombination. The biphasicity of the recombination kinetics in this pH range was ascribed to the comparable rates of the back-reaction and of protonation of the reaction center at most pH values. Thus, the distribution of protonation states established in the dark cannot fully reequilibrate with the new p K values of the charge-separated state, on the timescale of the back-reaction. The behavior described here for the P + QA charge recombination in Rps. viridis is in contrast to that in Rhodobacter sphaeroides , which exhibits monophasic back reaction kinetics at ambient temperatures. It is suggested that in the latter species the back-reaction is sufficiently slow ( k = 10 s −1 ) that protonation states present after the flash can equilibrate prior to the charge recombination.

Electron and proton transfer to the quinones in bacterial photosynthetic reaction centers: Insight from combined approaches of molecular genetics and biophysics

We present here new results together with an overview of the current knowledge on the coupled processes of electron and proton transfer in bacterial reaction centers. The importance of a multidisciplinary approach associating molecular genetics, structural biology, biochemistry and spectroscopy is underlined. We emphasize the electrostatic role of the protein to maintain a negative electrostatic potential near the second quinone electron acceptor in order to: i) accelerate the overall rate of proton transfer from the cytoplasm to this acceptor by increasing the pKs of some groups involved in this process; ii) increase the local proton concentration near this acceptor. We also point out the possibility of long distance propagation of the electrostatic effects through the protein associated with relaxation processes triggered by the formation of the semiquinone anions on the first flash.

SPECTRA OF FLUORESCENCE LIFETIME AND INTENSITY OF Rhodopseudomonas sphaeroides AT ROOM AND LOW TEMPERATURE. COMPARISON BETWEEN THE WILD TYPE, THE C 71 REACTION CENTER‐LESS MUTANT AND THE B800–850 PIGMENT‐PROTEIN COMPLEX

Abstract— Spectra of the fluorescence lifetime and intensity of chromatophores from the wild type Rhodopseudomonas sphaeroides, from the C 71 reaction center‐less mutant and of the B800–850 light harvesting pigment‐protein complex have been studied by phase fluorimetry techniques at different light modulation frequencies at room and low temperature.

Fluorescence lifetime spectra of in vivo bacteriochlorophyll at room temperature

Fluorescence lifetime spectra of Rhodopseudomonas sphaeroides chromatophores have been measured at room temperature by phase fluorimetry at 82 MHz in order to investigate the heterogeneity of the emission. The total fluorescence was decomposed into two main components. A constant component, Fc, centered at 865 nm, represents about 50% of the total emission from dark-adapted chromatophores (Fo) and has a lifetime of 0.55 ns. A variable component is centered at 890 nm. Upon closing the reaction centers, 5-fold increases take place in both emission yield and lifetime of this component. In the dark-adapted state, its lifetime is about 50 ps and its contribution to the total fluorescence is 70% at 890 nm. In the presence of sodium dithionite, a long-lifetime component (τD ⋍ 4 ns) is observed. This probably arises from radical pair recombination between P+ and I− (P, the primary electron donor, is a dimer of bacteriochlorophyll; I, the primary electron acceptor, is a molecule of bacteriopheophytin). Its spectrum is nearly identical to that of the variable component. This emission seems to be present also under nonreducing conditions, although with a much weaker intensity than when the electron acceptor quinone is prereduced.

Relating the Diffusion of Small Ligands in Human Neuroglobin to Its Structural and Mechanical Properties

Neuroglobin (Ngb), a recently discovered member of the globin family, is overexpressed in the brain tissues over oxygen deprivation. Unlike more classical globins, such as myoglobin and hemoglobin, it is characterized by a hexacoordinated heme, and its physiological role is still unknown, despite the numerous investigations made on the protein in recent years. Another important specific feature of human Ngb is the presence of two cysteine residues (Cys46 and Cys55), which are known to form an intramolecular disulfide bridge. Since previous work on human Ngb reported that its ligand binding properties could be controlled by the coordination state of the Fe(2+) atom (in the heme moiety) and the redox state of the thiol groups, we choose to develop a simulation approach combining coarse-grain Brownian dynamics and all-atom molecular dynamics and metadynamics. We have studied the diffusion of small ligands (CO, NO, and O(2)) in the globin internal cavity network for various states of human Ngb. Our results show how the structural and mechanical properties of the protein can be related to the ligand migration pathway, which can be extensively modified when changing the thiols redox state and the irons coordination state. We suggest that ligand binding is favored in the pentacoordinated species bearing an internal disulfide bridge.

pH effect on the biphasicity of the P+QA− charge recombination kinetics in the reaction centers from Rhodobacter sphaeroides, reconstituted with anthraquinones

Abstract The rate constant of decay of the flash-induced absorbance changes related to the primary electron donor, P + , was measured in anthraquinone-reconstituted reaction centers from wild type Rhodobacter sphaeroides (strain Y). The decay was found to be biphasic. At pH 9, the two rate constants are equal to 166 ± 20 s −1 ( k slow ) and 350 ± 30 s −1 ( k fast ), and their amplitudes are 55% and 45%, respectively. This apparent biphasicity is strongly pH-dependent. At pH 11.2, both components are accelerated ( k slow = 370 ± 40 s −1 and k fast = 1440 ± 100 s −1 ) but their relative amplitudes are inverted to 25% and 75%, respectively. The pH dependence curves of both the rate constants and relative amplitudes of the two phases are very similar to what was recently observed in the native reaction centers from Rhodopseudomonas viridis (Sebban, P. and Wraight, C.A., unpublished data). The increase in the rate constants above pH 9 reflects a diminution of the free energy difference between the P + Q A − state and a thermally excited state (possibly P + I − ) via which P + and Q A − recombine. The pH dependence curves of k slow , k fast or the average rate constant, display a p K value (p K QA ) of about 10.3, indicating that the replacement of the native ubiquinone by an anthraquinone shifts the p K of protonations of Q A − compared to the native ubiquinone (p K = 9.8). The replacement of the native Q A by the 1-amino-5-chloroanthraquinone or the 1-chloroanthraquinone confirmed this pK Q A shift. The obtained p K QA value is independent of the presence of terbutryn. In addition to these similarities, the activation parameters of k slow and k fast also behave as in Rps. viridis . From the Arrhenius plots of the two components, we determined that ΔH slow ΔH fast , but because of the quite large entropic contributions, ΔG slow = 0.295 ± 0.01 eV> ΔG fast = 0.276 ± 0.01 eV. It is suggested that the observed biphasicity of the charge recombination is due to the fast recombination rate in the anthraquinone-reconstituted Rb. sphaeroides reaction centers (as in native reaction centers from Rps. viridis ), which prevents the different protonation states reached after the flash from equilibrating. This is contrast to what is observed in native ubiquinone-containing reaction centers where the recombination rate is much slower.

ISOLATION and SPECTROSCOPIC CHARACTERIZATION OF THE B875 ANTENNA COMPLEX OF A MUTANT OF Rhodopseudomonas sphaeroides

Abstract— We describe a procedure of purification of the B875 antenna complex isolated from the 3P17 mutant strain of Rhodopseudomonas sphaeroides, enriched in B875. The integrity of this isolated complex, as well as a very low content of residual B800‐850 antenna, was suggested from low temperature absorption and resonance Raman spectra. Time resolved experiments were also carried out. The important result is the identity of the fluorescence lifetime of the B875 isolated complex (0.64 ± 0.03 ns) with that of the B875 antenna in vivo (0.63 ns), in the membrane of the C71 reaction center‐less mutant of Rhodopseudomonas sphaeroides, measured in our previous study.

Study of wild type and genetically modified reaction centers from Rhodobacter capsulatus: structural comparison with Rhodopseudomonas viridis and Rhodobacter sphaeroides

Reaction centers from the purple bacterium Rhodobacter (Rb.) capsulatus and from two mutants ThrL226-->Ala and IleL229-->Ser, modified in the binding protein pocket of the secondary quinone acceptor (QB), have been studied by flash-induced absorbance spectroscopy. In ThrL226-->Ala, the binding affinities for endogenous QB (ubiquinone 10) and UQ6 are found to be two to three times as high as the wild type. In contrast, in IleL229-->Ser, the binding affinity for UQ6 is decreased about three times compared to the wild type. In ThrL226-->Ala, a markedly increased sensitivity (approximately 30 times) to o-phenanthroline is observed. In Rhodopseudomonas viridis, where Ala is naturally in position L226, the sensitivity to o-phenanthroline is close to that observed in ThrL226-->Ala. We propose that the presence of Ala in position L226 is responsible for the high sensitivity to that inhibitor. The pH dependencies of the rate constants of P+QB- (kBP) charge recombination kinetics (P is a dimer of bacteriochlorophyll, and QB is the secondary quinone electron acceptor) show destabilization of QB- in ThrL226-->Ala and IleL229-->Ser, compared to the wild type. At low pH, similar apparent pK values of protonation of amino acids around QB- are measured in the wild type and the mutants. In contrast to Rb. sphaeroides, in the wild type Rb. capsulatus, kBP substantially increases in the pH range 7-10. This may reflect some differences in the respective structures of both strains or, alternatively, may be due to deprotonation of TyrL 215 in Rb. capsulatus. At pH 7, measurements of the rate constant of QA to QB electron transfer reveal a threefold greater rate in the reaction centers from wild type Rb. capsulatus (65 +/- 1 0 ps)-1 compared to Rb. sphaeroides.We suggest that this may arise from a 0.7-A smaller distance between the quinones in the former strain. Our spectroscopic data on the wild type Rb. capsulatus reaction center suggest the existence of notable differences with the Rb. sphaeroides reaction center structure.

Intermediate states between P* and Pf in bacterial reaction centers, as detected by the fluorescence kinetics

The fluorescence lifetimes of the reaction centers isolated from the wild‐type Rhodopseudomonas sphaeroides purple bacterium and those from the R26 mutant strain, lacking the carotenoid, were measured at low redox potential. In addition to the prompt fluorescence occurring directly from P* and the long delayed emission related to the radical pair state Pf, two other components are present. We suggest that they may come from intermediate states between P* and Pf, or reflect the stabilization of Pf itself.

The position of QB in the photosynthetic reaction center depends on pH: A theoretical analysis of the proton uptake upon QB reduction

Electrostatics-based calculations have been performed to examine the proton uptake upon reduction of the terminal electron acceptor Q(B) in the photosynthetic reaction center of Rhodobacter sphaeroides as a function of pH and the associated conformational equilibrium. Two crystal structures of the reaction center were considered: one structure was determined in the dark and the other under illumination. In the two structures, the Q(B) was found in two different positions, proximal or distal to the nonheme iron. Because Q(B) was found mainly in the distal position in the dark and only in the proximal position under illumination, the two positions have been attributed mostly to the oxidized and the reduced forms of Q(B), respectively. We calculated the proton uptake upon Q(B) reduction by four different models. In the first model, Q(B) is allowed to equilibrate between the two positions with either oxidation state. This equilibrium was allowed to vary with pH. In the other three models the distribution of Q(B) between the proximal position and the distal position was pH-independent, with Q(B) occupying only the distal position or only the proximal position or populating the two positions with a fixed ratio. Only the first model, which includes the pH-dependent conformational equilibrium, reproduces both the experimentally measured pH dependence of the proton uptake and the crystallographically observed conformational equilibrium at pH 8. From this model, we find that Q(B) occupies only the distal position below pH 6.5 and only the proximal position above pH 9.0 in both oxidation states. Between these pH values both positions are partially occupied. The reduced Q(B) has a higher occupancy in the proximal position than the oxidized Q(B). In summary, the present results indicate that the conformational equilibrium of Q(B) depends not only on the redox state of Q(B), but also on the pH value of the solution.