Heterogeneity of photosynthetic membranes from Rhodobacter capsulatus: size dispersion and ATP synthase distribution

The density distribution of photosynthetic membrane vesicles (chromatophores) from Rhodobacter capsulatus has been studied by isopicnic centrifugation. The average vesicle diameters, examined by electron microscopy, varied between 61 and 72 nm in different density fractions (70 nm in unfractionated chromatophores). The ATP synthase catalytic activities showed maxima displaced toward the higher density fractions relative to bacteriochlorophyll, resulting in higher specific activities in those fractions (about threefold). The amount of ATP synthase, measured by quantitative Western blotting, paralleled the catalytic activities. The average number of ATP synthases per chromatophore, evaluated on the basis of the Western blotting data and of vesicle density analysis, ranged between 8 and 13 (10 in unfractionated chromatophores). Poisson distribution analysis indicated that the probability of chromatophores devoid of ATP synthase was negligible. The effects of ATP synthase inhibition by efrapeptin on the time course of the transmembrane electric potential (evaluated as carotenoid electrochromic response) and on ATP synthesis were studied comparatively. The ATP produced after a flash and the total charge associated with the proton flow coupled to ATP synthesis were more resistant to efrapeptin than the initial value of the phosphorylating currents, indicating that several ATP synthases are fed by protons from the same vesicle.     

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.     

Physiological ligands ADP and Pi modulate the degree of intrinsic coupling in the ATP synthase of the photosynthetic bacterium Rhodobacter capsulatus

The proton-pumping and the ATP hydrolysis activities of the ATP synthase of Rhodobacter capsulatus have been compared as a function of the ADP and P(i) concentrations. The proton pumping was measured either with the transmembrane pH difference probe, 9-amino-6-chloro-2-methoxyacridine, or with the transmembrane electric potential difference probe, bis(3-propyl-5-oxoisoxazol-4-yl)pentamethine oxonol, obtaining consistent results. The comparison indicates that an intrinsic uncoupling of ATP synthase is induced when the concentration of either ligand is decreased. The half-maximal effect was found in the submicromolar range for ADP and at about 70 microM for P(i). It is proposed that a switch from a partially uncoupled state of ATP synthase to the coupled state is induced by the simultaneous binding of ADP and P(i).     

Rhodobacter capsulatus 1-deoxy-D-xylulose 5-phosphate synthase: steady-state kinetics and substrate binding

1-Deoxy-d-xylulose 5-phosphate synthase (DXP synthase) catalyzes the thiamine diphosphate (TPP)-dependent condensation of pyruvate and d-glyceraldehyde 3-phosphate (GAP) to yield DXP in the first step of the methylerythritol phosphate pathway for isoprenoid biosynthesis. Steady-state kinetic constants for DXP synthase calculated from the initial velocities measured at varying concentrations of substrates were as follows: k(cat) = 1.9 +/- 0.1 s(-1), K(m)(GAP) = 0.068 +/- 0.001 mM, and K(m)(pyruvate) = 0.44 +/- 0.05 mM for pyruvate and GAP; k(cat) = 1.7 +/- 0.1 s(-1), K(m)(d-glyceraldehyde) = 33 +/- 3 mM, and K(m)(pyruvate) = 1.9 +/- 0.5 mM for d-glyceraldehyde and pyruvate. beta-Fluoropyruvate was investigated as a dead-end inhibitor for pyruvate. Double-reciprocal plots showed a competitive inhibition pattern with respect to pyruvate and noncompetitive inhibition with respect to GAP/d-glyceraldehyde. (14)CO(2) trapping experiments demonstrated that the binding of both substrates (pyruvate and GAP/d-glyceraldehyde) is required for the formation of a catalytically competent enzyme-substrate complex. These results are consistent with an ordered mechanism for DXP synthase where pyruvate binds before GAP/d-glyceraldehyde.     

Coupling of proton flow to ATP synthesis in Rhodobacter capsulatus: F(0)F(1)-ATP synthase is absent from about half of chromatophores.

F(0)F(1)-ATP synthase (H(+)-ATP synthase, F(0)F(1)) utilizes the transmembrane protonmotive force to catalyze the formation of ATP from ADP and inorganic phosphate (P(i)). Structurally the enzyme consists of a membrane-embedded proton-translocating F(0) portion and a protruding hydrophilic F(1) part that catalyzes the synthesis of ATP. In photosynthetic purple bacteria a single turnover of the photosynthetic reaction centers (driven by a short saturating flash of light) generates protonmotive force that is sufficiently large to drive ATP synthesis. Using isolated chromatophore vesicles of Rhodobacter capsulatus, we monitored the flash induced ATP synthesis (by chemoluminescence of luciferin/luciferase) in parallel to the transmembrane charge transfer through F(0)F(1) (by following the decay of electrochromic bandshifts of intrinsic carotenoids). With the help of specific inhibitors of F(1) (efrapeptin) and of F(0) (venturicidin), we decomposed the kinetics of the total proton flow through F(0)F(1) into (i) those coupled to the ATP synthesis and (ii) the de-coupled proton escape through F(0). Taking the coupled proton flow, we calculated the H(+)/ATP ratio; it was found to be 3.3+/-0.6 at a large driving force (after one saturating flash of light) but to increase up to 5.1+/-0.9 at a smaller driving force (after a half-saturating flash). From the results obtained, we conclude that our routine chromatophore preparations contained three subsets of chromatophore vesicles. Chromatophores with coupled F(0)F(1) dominated in fresh material. Freezing/thawing or pre-illumination in the absence of ADP and P(i) led to an increase in the fraction of chromatophores with at least one de-coupled F(0)(F(1)). The disclosed fraction of chromatophores that lacked proton-conducting F(0)(F(1)) (approx. 40% of the total amount) remained constant upon these treatments.     

Monte Carlo simulation from proton slip to "coupled" proton flow in ATP synthase based on the bi-site mechanism

ATP synthase couples proton flow to ATP synthesis, but is leaky to protons at very low nucleotide concentration. Based on the bi-site mechanism, we simulated the proton conduction from proton slip to “coupled” proton flow in ATP synthase using the Monte Carlo method. Good agreement is obtained between the simulated and available experimental results. Our model provides deeper insight into the nucleotide dependence of ATP catalysis, and the kinetic cooperativity in three catalysis subunits. The results of simulation support the bi-site mechanism in ATP synthesis.     

Activation of the H(+)-ATP synthase in the photosynthetic bacterium Rhodobacter capsulatus.

The regulation of the membrane-bound H(+)-ATPase from the photosynthetic bacterium Rhodobacter capsulatus was investigated. In the presence of uncouplers the rate of ATP hydrolysis was about 40 mM ATP/M bacteriochlorophyll (Bchl)/s. Without uncouplers this rate increased and if, additionally, the chromatophores were illuminated, it was almost doubled. If uncouplers were added shortly after illumination, the rate increased to 300-350 mM ATP/M Bchl/s. Obviously, energization of the membrane leads to the formation of a metastable, active state of the H(+)-ATPase. The maximal rate of ATP hydrolysis can be measured only when first all H(+)-ATPases are activated by delta mu H+ and when the delta mu H+ is abolished in order to release its back pressure on the hydrolysis rate. The half-life time of the metastable state in the absence of delta mu H+ is about 30 s. It is increased by 3 mM Pi to about 80 s and it is decreased by 1 mM ADP to about 15 s. Quantitatively, the fraction of active H(+)-ATPases shows a sigmoidal dependence on pHin (at constant pHout) and the magnitude of delta psi determines the maximal fraction of enzymes which can be activated: delta pH and delta psi are not equivalent for the activation process.     

Purification and properties of the H(+)-nicotinamide nucleotide transhydrogenase from Rhodobacter capsulatus.

1. H(+)-transhydrogenase from Rhodobacter capsulatus is an integral membrane protein which, unlike the enzyme from Rhodospirillum rubrum, does not require the presence of a water-soluble component for activity. 2. The enzyme from Rb. capsulatus was solubilised in Triton X-100 and subjected to ion-exchange, hydroxyapatite and then gel-exclusion column chromatography. SDS/PAGE of the purified enzyme revealed the presence of two polypeptides with apparent Mr 53,000 and 48,000. Other minor components which were stained on the electrophoresis gels or which were revealed on Western blots exposed to antibodies raised to total membrane proteins, were probably contaminants. 3. Antibodies raised to the 53-kDa and 48-kDa polypeptides cross-reacted with equivalent polypeptides in Western blots of solubilised membranes from Rb. capsulatus, Rhodobacter sphaeroides and Rhs. rubrum. The significance of this finding is discussed in the context of the hypothesis [Fisher, R.R. & Earle, S.R. (1982) The pyridine nucleotide coenzymes, pp. 279-324, Academic Press, New York] that the soluble component associated with H(+)-transhydrogenase from Rhs. rubrum is an integral part of the catalytic machinery. Antibodies against the 48-kDa and 53-kDa polypeptides of the Rb. capsulatus enzyme cross-reacted with equivalent polypeptides in solubilised membranes of Escherichia coli. 4. The dependence of the rate of H- transfer by purified H(+)-transhydrogenase on the nucleotide substrate concentrations under steady-state conditions, the effects of inhibition by nucleotide products and the inhibition by 2'-AMP and by 5'-AMP suggest that the reaction proceeds by the random addition of substrates to the enzyme with the formation of a ternary complex. 5. In conflict with this conclusion, the reduction of acetylpyridine adenine dinucleotide (AcPdAD+) by NADH in the absence of NADP+ by bacterial membranes was earlier taken as evidence for the existence of a reduced enzyme intermediate [Fisher, R.R. & Earle, S.R. (1982) The pyridine nucleotide coenzymes, pp. 279-324, Academic Press, New York]. However, it is shown here that although chromatophore membranes of Rb. capsulatus catalysed the reduction of AcPdAD+ by NADH, the reaction was not associated with the purified H(+)-transhydrogenase. Moreover, in contrast with the true transhydrogenase reaction, the reconstitution of AcPdAD+ reduction by NADH (in the absence of NADP+) in washed membranes of Rhs. rubrum with partially purified transhydrogenase factor, was only additive.     

A point mutation in the ATP synthase of Rhodobacter capsulatus results in differential contributions of Delta(pH) and Delta(phi) in driving the ATP synthesis reaction.

The interface between the c-subunit oligomer and the a subunit in the F0 sector of the ATP synthase is believed to form the core of the rotating motor powered by the protonic flow. Besides the essential cAsp61 and aArg210 residues (Escherichia coli numbering), a few other residues at this interface, although nonessential, show a high degree of conservation, among these aGlu219. The homologous residue aGlu210 in the ATP synthase of the photosynthetic bacterium Rhodobacter capsulatus has been substituted by a lysine. Inner membranes prepared from the mutant strain showed approximately half of the ATP synthesis activity when driven both by light and by acid-base transitions. As estimated with the ACMA assay, proton pumping rates in the inner membranes were also reduced to a similar extent in the mutant. The most striking impairment of ATP synthesis in the mutant, a decrease as low as 12 times as compared to the wild-type, was observed in the absence of a transmembrane electrical membrane potential (Delta(phi)) at low transmembrane pH difference (Delta(pH)). Therefore, the mutation seems to affect both the mechanism responsible for coupling F1 with proton translocation by F0, and the mechanism determining the relative contribution of Delta(pH) and Delta(phi) in driving ATP synthesis.     

Sequence of the indoleglycerol phosphate synthase (trpC) gene from Rhodobacter capsulatus.

We have isolated, cloned, and sequenced the indoleglycerol phosphate synthase gene (trpC) from Rhodobacter capsulatus. Normalized alignment scores comparing the trpC gene of R. capsulatus with the trpC genes of other bacterial species are reported. An unexpected degree of similarity to the trpC gene of Bacillus subtilis was found.     

Assay of ATP in intact cells of the facultative phototroph Rhodobacter capsulatus expressing recombinant firefly luciferase

This study reports on the construction, calibration and use of recombinant cells of Rhodobacter capsulatus expressing the luciferase gene of the North American firefly Photinus pyralis to detect, by bioluminescence, variations of endogenous ATP levels under various physiological conditions. We show that the antibiotic polymyxin B allows luciferin to rapidly move into cell cytosol, but does not make external ATP freely accessible to intracellular luciferase. Notably, in toluene:ethanol-permeabilized cells, the apparent K(mATP) for luciferase (50 microM) is similar to that measured in soluble cell fractions. This finding limits the applicability of the firefly luciferase for monitoring intracellular maximal ATP concentration because dark/aerobic-grown recombinant cells of Rba. capsulatus contain approximately 1.3-2.6+/-0.5 mM ATP. Therefore, the effects of chemical and physical factors such as oxygen, light, carbonyl cyanide m-chlorophenyl hydrazone and antimycin A on ATP synthesis were examined in cells subjected to different starvation periods to reduce the endogenous ATP pool below the luciferase ATP saturation level (< or =0.2 mM). We conclude that the amount of endogenous ATP generated by light is maximal in the presence of oxygen, which is required to optimize the membrane redox poise.     

Kinetic model of ATP synthase: pH dependence of the rate of ATP synthesis

Recently, a novel molecular mechanism of torque generation in the F(0) portion of ATP synthase was proposed [Rohatgi, Saha and Nath (1998) Curr. Sci. 75, 716-718]. In this mechanism, rotation of the c-subunit was conceived to take place in 12 discrete steps of 30 degrees each due to the binding and unbinding of protons to/from the leading and trailing Asp-61 residues of the c-subunit, respectively. Based on this molecular mechanism, a kinetic scheme has been developed in this work. The scheme considers proton transport driven by a concentration gradient of protons across the proton half-channels, and the rotation of the c-subunit by changes in the electrical potential only. This kinetic scheme has been analyzed mathematically and an expression has been obtained to explain the pH dependence of the rate of ATP synthesis by ATP synthase under steady state operating conditions. For a single set of three enzymological kinetic parameters, this expression predicts the rates of ATP synthesis which agree well with the experimental data over a wide range of pH(in) and pH(out). A logical consequence of our analysis is that DeltapH and Deltapsi are kinetically inequivalent driving forces for ATP synthesis.     

Purification, characterization and nucleotide sequence of the periplasmic C4-dicarboxylate-binding protein (DctP) from Rhodobacter capsulatus

A periplasmic binding protein essential for high-affinity transport of the C4-dicarboxylates malate, succinate and fumarate across the cytoplasmic membrane of the purple photosynthetic bacterium Rhodobacter capsulatus has been purified to homogeneity and some of its ligand-binding properties characterized. The protein was not produced in a Tn5 insertion mutant unable to transport C4-dicarboxylates under aerobic conditions in the dark. Wild-type DNA corresponding to the location of the transposon insertion site was subcloned and a 1.5 kb section sequenced. A complete open reading frame of 999 bp was identified that encoded a 333-residue protein (DctP) with a molecular weight of 36,128 with a 26-residue amino-terminal signal peptide. The identify of this protein with the purified dicarboxylate-binding protein and the position of the predicted signal peptide cleavage site was confirmed by N-terminal sequencing. No significant homology with other proteins was detected in database searches. A GC-rich region of dyad symmetry was located 7 bp downstream of the dctP translational stop codon. This structure may be of significance in regulating the relative abundance of DctP and other dct gene products which comprise the high-affinity dicarboxylate transport system in this bacterium.     

Cloning and characterization of the ddsA gene encoding decaprenyl diphosphate synthase from Rhodobacter capsulatus B10

A decaprenyl diphosphate synthase gene (ddsA, GenBank accession No. DQ191802) was cloned from Rhodobacter capsulatus B10 by constructing and screening the genome library. An open reading frame of 1002 bp was revealed from sequence analysis. The deduced polypeptide consisted of 333 amino acids residues with an molecular mass of about 37 kDa. The DdsA protein contained the conserved amino acid sequence (DDXXD) of E-type polyprenyl diphosphate synthase and showed high similarity to others. In contrast, DdsA showed only 39% identity to a solanesyl diphosphate synthase cloned from R. capsulatus SB1003. DdsA was expressed successfully in Escherichia coli. Assaying the enzyme in vivo found it made E.coli synthesize UQ-10 in addition to the endogenous production UQ-8.     

Two distinct proton binding sites in the ATP synthase family

The F1F0 ATP synthase utilizes energy stored in an electrochemical gradient of protons (or Na+ ions) across the membrane to synthesize ATP from ADP and phosphate. Current models predict that the protonation/deprotonation of specific acidic c ring residues is at the core of the proton translocation mechanism by this enzyme. To probe the mode of proton binding, we measured the covalent modification of the acidic c ring residues with the inhibitor dicyclohexylcarbodiimide (DCCD) over the pH range from 5 to 11. With the H+-translocating ATP synthase from the archaeum Halobacterium salinarium or the Na+-translocating ATP synthase from Ilyobacter tartaricus, the pH profile of DCCD labeling followed a titration curve with a pKa around neutral, reflecting protonation of the acidic c ring residues. However, with the ATP synthases from Escherichia coli, mitochondria, or chloroplasts, a clearly different, bell-shaped pH profile for DCCD labeling was observed which is not compatible with carboxylate protonation but might be explained by the coordination of a hydronium ion as proposed earlier [Boyer, P. D. (1988) Trends Biochem. Sci. 13, 5-7]. Upon site-directed mutagenesis of single binding site residues of the structurally resolved c ring, the sigmoidal pH profile for DCCD labeling could be converted to a more bell-shaped one, demonstrating that the different ion binding modes are based on subtle changes in the amino acid sequence of the protein. The concept of two different binding sites in the ATP synthase family is supported by the ATP hydrolysis pH profiles of the investigated enzymes.     

Intracytoplasmic membrane vesiculation in light-harvesting mutants of Rhodobacter sphaeroides and Rhodobacter capsulatus

Abstract In Chlamydomonas reinhardtii there are three glutamate dehydrogenase isozymes which can use both NADH and NADPH as cofactors and respond differently to different nitrogen sources and several stress conditions. From data of induction of isozymes in different metabolic situations, we propose a possible physiological role for each of them in algal carbon and nitrogen metabolism.