Ferredoxin and flavodoxin from the cyanobacterium Synechocystis sp PCC 6803.

The unicellular cyanobacterium Synechocystis sp PCC 6803 is capable of synthesizing two different Photosystem-I electron acceptors, ferredoxin and flavodoxin. Under normal growth conditions a [2Fe-2S] ferredoxin was recovered and purified to homogeneity. The complete amino-acid sequence of this protein was established. The isoelectric point (pI = 3.48), midpoint redox potential (Em = -0.412 V) and stability under denaturing conditions were also determined. This ferredoxin exhibits an unusual electrophoretic behavior, resulting in a very low apparent molecular mass between 2 and 3.5 kDa, even in the presence of high concentrations of urea. However, a molecular mass of 10,232 Da (apo-ferredoxin) is calculated from the sequence. Free thiol assays indicate the presence of a disulfide bridge in this protein. A small amount of ferredoxin was also found in another fraction during the purification procedure. The amino-acid sequence and properties of this minor ferredoxin were similar to those of the major ferredoxin. However, its solubility in ammonium sulfate and its reactivity with antibodies directed against spinach ferredoxin were different. Traces of flavodoxin were also recovered from the same fraction. The amount of flavodoxin was dramatically increased under iron-deficient growth conditions. An acidic isoelectric point was measured (pI = 3.76), close to that of ferredoxin. The midpoint redox potentials of flavodoxin are Em1 = -0.433 V and Em2 = -0.238 V at pH 7.8. Sequence comparison based on the 42 N-terminal amino acids indicates that Synechocystis 6803 flavodoxin most likely belongs to the long-chain class, despite an apparent molecular mass of 15 kDa determined by SDS-PAGE.     

Ferredoxin and flavodoxin in eastern Antarctica pack ice

The presence, concentration and distribution of the iron regulated proteins, ferredoxin and flavodoxin, was investigated in pack ice off eastern Antarctica using SDS-PAGE gels. Bands corresponding to ferredoxin and/or flavodoxin were observed in all but eight of the 102 core sections analysed. Flavodoxin was found in most of the ice samples and was strongly correlated with chlorophyll a standing stock. The widespread distribution of flavodoxin here is not thought to indicate iron-limitation as many of the dominant species, such as Fragilariopsis cylindrus, Cylindrotheca closterium, are known to produce this protein under iron-replete conditions and thus the significant correlation between flavodoxin and biomass is likely to be the result of widespread constitutive flavodoxin expression among the diatoms that commonly inhabit sea ice. High concentrations of ferredoxin were predominantly derived from core sections on the floes closest to the continent and also in the upper portion of these floes. There was a consistent lack of ferredoxin expression in the high biomass bottom communities. The absence of ferredoxin is likely to indicate a reduced supply of iron but the significance of this reduced iron supply cannot be inferred on the basis of protein expression alone. Furthermore, in the present study the observed variability in the flavodoxin:ferredoxin ratio may not simply reflect the iron nutritional status of the community, but probably results from changes in the abundance of species capable of expressing ferredoxin.     

Ferredoxin: the central hub connecting photosystem I to cellular metabolism

Photosynthetica - Ferredoxin (Fd) is a small soluble iron-sulfur protein essential in almost all oxygenic photosynthetic organisms. It contains a single [2Fe-2S] cluster coordinated by four...     

Light-induced alteration of low-temperature interprotein electron transfer between photosystem I and flavodoxin

Electron paramagnetic resonance (EPR) was used to study light-induced electron transfer in Photosystem I-flavodoxin complexes. Deuteration of flavodoxin enables the signals of the reduced flavin acceptor and oxidized primary donor, P(700)(+), to be well-resolved at X- and D-band EPR. In dark-adapted samples, photoinitiated interprotein electron transfer does not occur at 5 K. However, for samples prepared in dim light, significant interprotein electron transfer occurs at 5 K and a concomitant loss of the spin-correlated radical pair P(+)A(1A)(-) signal is observed. These results indicate a light-induced reorientation of flavodoxin in the PSI docking site that allows a high quantum yield efficiency for the interprotein electron transfer reaction.     

Ferredoxin reduction by polarographic methods]

The reduction of iron-sulphur protein of the higher plant ferrodoxin has been studied by polarographical methods. Ferredoxin initiates a reversible wave with E1/2=--0,61 v (N. C. E.) at pH 7. Protein absorption greatly influences the electrochemical reduction. The protons have been shown to take part in the electrode reaction. The potentiometrically obtained data about the difference between E1/2 and E0=--0.70 v and its causative factors are discussed. As a result of the experiments with modification of ferredoxin active centre it has been concluded that the active centre participates in the polarographical reduction.     

Characterization of a redox-active cross-linked complex between cyanobacterial photosystem I and its physiological acceptor flavodoxin.

A covalent complex between photosystem I and flavodoxin from the cyanobacterium Synechococcus sp. PCC 7002 was generated by chemical cross-linking. Laser flash-absorption spectroscopy indicates that the bound flavodoxin of this complex is stabilized in the semiquinone state and is photoreduced to the quinol form upon light excitation. The kinetics of this photoreduction process, which takes place in approximately 50% of the reaction centres, displays three exponential components with half-lives of 9 microsec, 70 microsec and 1 ms. The fully reduced flavodoxin subsequently recombines with P700+ with a t1/2 of 330 ms. A corresponding flavodoxin semiquinone radical signal is readily observed in the dark by room temperature electron paramagnetic resonance, which reversibly disappears upon illumination. In contrast, the light-induced reduction of oxidized flavodoxin can be observed only by first-flash experiments following excessive dark adaptation. In addition, the docking site of flavodoxin on photosystem I was determined by electron microscopy in combination with image analysis. Flavodoxin binds to the cytoplasmic side of photosystem I at a distance of 7 nm from the centre of the trimer and in close contact to a ridge formed by the subunits PsaC, PsaD and PsaE.     

Kinetic evidence for the PsaE-dependent transient ternary complex photosystem I/Ferredoxin/Ferredoxin:NADP(+) reductase in a cyanobacterium.

A mutant of Synechocystis PCC 6803, deficient in psaE, assembles photosystem I reaction centers without the PsaE subunit. Under conditions of acceptor-side rate-limited photoreduction assays in vitro (with 15 microM plastocyanin included), using 100 nM ferredoxin:NADP(+) reductase (FNR) and either Synechocystis flavodoxin or spinach ferredoxin, lower rates of NADP(+) photoreduction were measured when PsaE-deficient membranes were used, as compared to the wild type. This effect of the psaE mutation proved to be due to a decrease of the apparent affinity of the photoreduction assay system for the reductase. In the psaE mutant, the relative petH (encoding FNR) expression level was found to be significantly increased, providing a possible explanation for the lack of a phenotype (i.e., a decrease in growth rate) that was expected from the lower rate of linear electron transport in the mutant. A kinetic model was constructed in order to simulate the electron transfer from reduced plastocyanin to NADP(+), and test for possible causes for the observed change in affinity for FNR. The numerical simulations predict that the altered reduction kinetics of ferredoxin, determined for the psaE mutant [Barth, P., et al., (1998) Biochemistry 37, 16233-16241], do not significantly influence the rate of linear electron transport to NADP(+). Rather, a change in the dissociation constant of ferredoxin for FNR does affect the saturation profile for FNR. We therefore propose that the PsaE-dependent transient ternary complex PSI/ferredoxin/FNR is formed during linear electron transport. Using the yeast two-hybrid system, however, no direct interaction could be demonstrated in vivo between FNR and PsaE fusion proteins.     

Anabaena flavodoxin as an electron carrier from photosystem I to ferredoxin-NADP+ reductase. Role of flavodoxin residues in protein-protein interaction and electron transfer

Biochemical and structural studies indicate that electrostatic and hydrophobic interactions are critical in the formation of optimal complexes for efficient electron transfer (ET) between ferredoxin-NADP(+) reductase (FNR) and ferredoxin (Fd). Moreover, it has been shown that several charged and hydrophobic residues on the FNR surface are also critical for the interaction with flavodoxin (Fld), although, so far, no key residue on the Fld surface has been found to be the counterpart of such FNR side chains. In this study, negatively charged side chains on the Fld surface have been individually modified, either by the introduction of positive charges or by their neutralization. Our results indicate that although Glu16, Glu20, Glu61, Asp65, and Asp96 contribute to the orientation and optimization of the Fld interaction, either with FNR or with photosystem I (PSI) (presumably through the formation of salt bridges), for efficient ET, none of these side chains is involved in the formation of crucial salt bridges for optimal interaction with FNR. These data support the idea that the FNR-Fld interaction is less specific than the FNR-Fd interaction. However, analysis of the reactivity of these mutated Flds toward the membrane-anchored PSI complex indicated that all mutants, except Glu16Gln, lack the ability to form a stable complex with PSI. Thr12, Thr56, Asn58, and Asn97 are present in the close environment of the isoalloxazine ring of FMN in Anabaena Fld. Their roles in the interaction with and ET to FNR and PSI have also been studied. Mutants at these Fld positions indicate that residues in the close environment of the isoalloxazine ring modulate the ability of Fld to bind to and to exchange electrons with its physiological counterparts.     

O2 reduction by photosystem I involves phylloquinone under steady-state illumination

O₂ reduction was investigated in photosystem I (PS I) complexes isolated from cyanobacteria Synechocystis sp. PCC 6803 wild type (WT) and menB mutant strain, which is unable to synthesize phylloquinone and contains plastoquinone at the quinone-binding site A₁. PS I complexes from WT and menB mutant exhibited different dependencies of O₂ reduction on light intensity, namely, the values of O₂ reduction rate in WT did not reach saturation at high intensities, in contrast to the values in menB mutant. The obtained results suggest the immediate phylloquinone involvement in the light-induced O₂ reduction by PS I.     

One-electron reduction of flavodoxin. A fast kinetic study.

The reaction kinetics of fully oxidized flavodoxin from Clostridium MP with the hydrated electron have been investigated by the pulse radiolysis method. Four spectrally distinct processes have been observed with the ultimate formation of the singly reduced flavin form of the protein. The last two species obtained in the reaction sequence are spectrally similar, and are connected through a reaction which is first order. It is proposed that this reaction involves a protein conformational alteration.     

Evidence for the involvement of PSI-E subunit in the reduction of ferredoxin by photosystem I.

Of the stroma-accessible proteins of photosystem I (PSI) from Synechocystis sp. PCC 6803, the PSI-C, PSI-D and PSI-E subunits have already been characterized, and the corresponding genes isolated. PCR amplification and cassette mutagenesis were used in this work to delete the psaE gene. PSI particles were isolated from this mutant, which lacks subunit PSI-E, and the direct photoreduction of ferredoxin was investigated by flash absorption spectroscopy. The second order rate constant for reduction of ferredoxin by wild type PSI was estimated to be approximately 10(9) M-1s-1. Relative to the wild type, PSI lacking PSI-E exhibited a rate of ferredoxin reduction decreased by a factor of at least 25. After reassociation of the purified PSI-E polypeptide, the original rate of electron transfer was recovered. When a similar reconstitution was performed with a PSI-E polypeptide from spinach, an intermediate rate of reduction was observed. Membrane labeling of the native PSI with fluorescein isothiocyanate allowed the isolation of a fluorescent PSI-E subunit. Peptide analysis showed that some residues following the N-terminal sequence were labeled and thus probably accessible to the stroma, whereas both N- and C-terminal ends were probably buried in the photosystem I complex. Site-directed mutagenesis based on these observations confirmed that important changes in either of the two terminal sequences of the polypeptide impaired its correct integration in PSI, leading to phenotypes identical to the deleted mutant. Less drastic modifications in the predicted stroma exposed sequences did not impair PSI-E integration, and the ferredoxin photoreduction was not significantly affected. All these results lead us to propose a structural role for PSI-E in the correct organization of the site involved in ferredoxin photoreduction.     

Structure of photosystem I.

In plants and cyanobacteria, the primary step in oxygenic photosynthesis, the light induced charge separation, is driven by two large membrane intrinsic protein complexes, the photosystems I and II. Photosystem I catalyses the light driven electron transfer from plastocyanin/cytochrome c(6) on the lumenal side of the membrane to ferredoxin/flavodoxin at the stromal side by a chain of electron carriers. Photosystem I of Synechococcus elongatus consists of 12 protein subunits, 96 chlorophyll a molecules, 22 carotenoids, three [4Fe4S] clusters and two phylloquinones. Furthermore, it has been discovered that four lipids are intrinsic components of photosystem I. Photosystem I exists as a trimer in the native membrane with a molecular mass of 1068 kDa for the whole complex. The X-ray structure of photosystem I at a resolution of 2.5 A shows the location of the individual subunits and cofactors and provides new information on the protein-cofactor interactions. [P. Jordan, P. Fromme, H.T. Witt, O. Klukas, W. Saenger, N. Krauss, Nature 411 (2001) 909-917]. In this review, biochemical data and results of biophysical investigations are discussed with respect to the X-ray crystallographic structure in order to give an overview of the structure and function of this large membrane protein.     

Energy transfer between photosystem II and photosystem I in chloroplasts.

A model for the photochemical apparatus of photosynthesis is presented which accounts for the fluorescence properties of Photosystem II and Photosystem I as well as energy transfer between the two photosystems. The model was tested by measuring at - 196 degrees C fluorescence induction curves at 690 and 730 nm in the absence and presence of 5mMMgCl2 which presumably changes the distrubution of excitation energy between the two photosystems. The equations describing the fluorescence properties involve terms for the distribution of absorbed quanta, alpha, being the fraction distributed to Photosystem I, and beta, the fraction to Photosystem II to Photosystem I, KT(II yields I). The data, analyzed within the context of the model, permit a direct comparison of alpha and kt(II yields I) in the absence (minus) and presence (+) of Mg-2+ :alpha minus/alpha-+ equals 1.2 and k-minus t)II yields I)/K-+T(II yields I) equal to 1.9. If the criterion that alpha + beta equal to 1 is applied absolute values can be calculated: in the presence of Mg-2+, alpha-+ equal to 0.27 and the yield of energy transfer, phi-+ t(II yields I) varied the presence of Mg-2+, alpha-+ equal to 0.27 and the yield of energy transfer, phi-+ t(II yields I) varied from 0.065 when the Photosystem II reaction centers were all open to 0.23 when they were closed. In the absence of Mg-2+, alpha-minus equal to 0.32 and phi t(II yields I) varied from 0.12 to 0.28. The data were also analyzed assuming that two types of energy transfer could be distinguished; a transfer from the light-harvesting chlorophyll of Photosystem II to Photosystem I, kt(II yields I), and a transfer from the reaction centers of Photosystem II to Photosystem I, kt(II yields I). In that case alpha-minus/alpha+ equal to 1.3, k-minus t(II yields I)/k+ t(II yields I)equal to 1.3 and k-minus t(II yields I) equal to 3.0. It was concluded, however, that both of these types of energy transfer are different manifestations of a single energy transfer process.     

Effect of artificial redox mediators on the photoinduced oxygen reduction by photosystem I complexes

The peculiarities of interaction of cyanobacterial photosystem I with redox mediators 2,6-dichlorophenolindophenol (DCPIP) and N,N,N′,N′-tetramethyl-p-phenylenediamine (TMPD) were investigated. The higher donor efficiency of the reduced DCPIP form was demonstrated. The oxidized form of DCPIP was shown to be an efficient electron acceptor for terminal iron–sulfur cluster of photosystem I. Likewise methyl viologen, after one-electron reduction, DCPIP transfers an electron to the molecular oxygen. These results were discussed in terms of influence of these interactions on photosystem I reactions with the molecular oxygen and natural electron acceptors.     

Inhibition of photosystem I and photosystem II in chloroplasts by UV radiation

The effects of UV radiation on the low temperature fluorescenceand primary photochemistry of PSII and PSI of spinach chloroplastswere studied. Fluorescence induction curves at –196°Cwere measured at 695 nm for PSII fluorescence and at 730 nmfor PSI fluorescence to determine both the initial F_o and finalF_M levels. The primary photochemistry of PSII was measured asthe rate of photoreduction of C-550 at – 196°C, thatof PSI as the rate of photooxidation of P₇₀₀ at –196°C.The results were analyzed in terms of a model for the photosyntheticapparatus which accounts for the yields of fluorescence andprimary photochemistry. According to this analysis UV radiationincreases nonradiative decay processes at the reaction centerchlorophyll of PSII. However, the effect of UV radiation isnot uniform throughout the sample during irradiation so thataccount must be taken of the fraction of PSII reaction centerswhich have been irradiated at any given time. UV radiation alsoinactivates P700 and causes a slight increase in nonradiativedecay in the antenna chlorophyll of PSI. All fluorescence ofvariable yield, F_V = F_M–F_o, at 730 nm is due to energytransfer from PSII to PSI so that the sensitivity of Fv to UVradiation is the same at 730 and 695 nm. ¹Present address: Department of Biology, Faculty of Science,Toho University, Narashino, Chiba 275, Japan. ²Present address: Central Research Laboratories, Fuji PhotoFilm Co., Ltd., 105 Mizonuma, Asaka-Shi, Saitama 351, Japan. (Received September 10, 1975; )     

Structural alignment of ferredoxin and flavodoxin based on electrostatic potentials: implications for their interactions with photosystem I and ferredoxin-NADP reductase

The two proteins ferredoxin and flavodoxin can replace each other in the photosynthetic electron transfer chain of cyanobacteria and algae. However, structure, size, and composition of ferredoxin and flavodoxin are completely different. Ferredoxin is a small iron-sulfur protein (approximately 100 amino acids), whereas flavodoxin is a flavin-containing protein (approximately 170 amino acids). The crystal structure of both proteins from the cyanobacteria Anabeana PCC 7120 is known. We used these two protein structures to investigate the structural basis of their functional equivalence. We apply the Hodgkin index to quantify the similarity of their electrostatic potentials. The technique has been applied successfully in indirect drug design for the alignment of small molecule and bioisosterism elucidation. It requires no predefined atom-atom correspondences. As is known from experiments, electrostatic interactions are most important for the association of ferredoxin and flavodoxin with their reaction partners photosystem I and ferredoxin-NADP reductase. Therefore, use of electrostatic potentials for the structural alignment is well justified. Our extensive search of the alignment space reveals two alignments with a high degree of similarity in the electrostatic potential. In both alignments, ferredoxin overlaps completely with flavodoxin. The active sites of ferredoxin and flavodoxin rather than their centers of mass coincide in both alignments. This is in agreement with electron microscopy investigations on photosystem I cross-linked to ferredoxin or flavodoxin. We identify residues that may have the same function in both proteins and relate our results to previous experimental data.     

Electrogenic reduction of the primary electron donor P700 by plastocyanin in photosystem I complexes

An electrometric technique was used to investigate electron transfer between spinach plastocyanin (Pc) and photooxidized primary electron donor P700 in photosystem I (PS I) complexes from the cyanobacterium Synechocystis sp. PCC 6803. In the presence of Pc, the fast unresolvable kinetic phase of membrane potential generation related to electron transfer between P700 and the terminal iron–sulfur acceptor F_B was followed by additional electrogenic phases in the microsecond and millisecond time scales, which contribute approximately 20% to the overall electrogenicity. These phases are attributed to the vectorial electron transfer from Pc to the protein-embedded chlorophyll dimer P700⁺ within the PsaA/PsaB heterodimer. The observed rate constant of the millisecond kinetic phase exhibited a saturation profile at increasing Pc concentration, suggesting the formation of a transient complex between Pc and PS I with the dissociation constant K_d of about 80 μM. A small but detectable fast electrogenic phase was observed at high Pc concentration. The rate constant of this phase was independent of Pc concentration, indicating that it is related to a first-order process.     

Electron transfer in photosystem I.

This mini-review focuses on recent experimental results and questions, which came up since the last more comprehensive reviews on the subject. We include a brief discussion of the different techniques used for time-resolved studies of electron transfer in photosystem I (PS I) and relate the kinetic results to new structural data of the PS I reaction centre.