Energy-dependent conformational changes in the epsilon subunit of the chloroplast ATP synthase (CF0CF1)

Incubation of spinach chloroplast thylakoids with pyridoxal 5'-phosphate modified the epsilon subunit of ATP synthase (CF0CF1). Illumination of thylakoids stimulated the modification of one specific amino acid residue of the epsilon subunit by a factor of 3. Endoproteinase Glu-C treatment of the isolated epsilon subunit and fractionation of the peptides by high performance liquid chromatography revealed a major fluorescent peptide with the sequence GKRQKIE. Further treatment of this peptide with endoproteinase Arg-C gave a strongly fluorescent tripeptide (GXR). From the primary structure of the epsilon subunit, the specifically modified residue was deduced to be Lys-109. This suggests the energy-dependent conformational changes in the epsilon subunit which change the surroundings of Lys-109 and alter the reactivity of this residue.     

Single channel H+ currents through reconstituted chloroplast ATP synthase CF0-CF1.

The purified chloroplast ATP synthase (CF(0)-CF(1)) was reconstituted into azolectin liposomes from which bilayer membranes on the tip of a glass pipette ('dip stick technique')and planar bilayer membranes were form ed. The CF(0)-CF(1) facilitated ion conductance through the bilayer membranes. Our results clearly indicated that the observed single channel currents were carried by H+ through the isolated and reconstituted chloroplast ATPase. We demonstrated that in proteoliposomes it is the whole enzyme complex CF(0)-CF(1) and not the membrane sector CF(0) alone that constitutes a voltagegated, proton-selective channel with a high conductance of 1-5 pS at pH 5.5-8.0. After removal of CF(1) from the liposomes by NaBr treatment the membrane sector CF(0) displayed various kinds of channels also permeable to monovalent cations. The open probability P(0) of the CF(0)-CF(1) channel increased considerable with increasing membrane voltage [from P(0) less than or equal to 1% (V(m) less than or equal to 120 mV) to P(0) less than or equal to 30% (120 mV less than or equal to Vm 200 mV)]. In the presence of ADP (3 microM) and P(i) (5 microM), which specifically bind to CF(1), the open probability decreased and venturicidin (1 microM), a specific inhibitor of H+ flow through CF(0) in thylakoid membranes, blocked the channel almost completely. Our results, which reveal a high channel unit conductance, and at membrane voltages less than 100 mV low open probability with concomitant mean open times in the micros timescale (less than 100 micros) for the energy coupling in the enzyme complex. At physiological membrane voltages for photophosphorylation (about 30 mV) the enzyme complex would then display a time-averaged conductance of about 1 fS.     

Subunit interactions within the chloroplast ATP synthase (CF0-CF1) as deduced by specific depletion of CF0 polypeptides

The proton-linked ATP synthase (CF1-CF0) of chloroplasts consists of a catalytic component (CF1) and a membrane-embedded part (CF0) that interacts with CF1 and contains a proton channel. The subunits of CF0 which are involved in binding of CF1 were studied by examining the effect of selective depletion of subunits I, II, and IV of CF0 from the chloroplast ATP synthase on the association of the remaining CF0 subunits with CF1. Dissociated CF0 subunits were identified by sucrose density gradient centrifugation. Removal of subunit IV alone from CF0-CF1 did not cause dissociation of the other CF0 subunits from CF1. Upon removal of both subunits I and IV from CF0-CF1, subunit II also dissociated, but subunit III was still bound to CF1. Thus, at least two subunits of CF0, I and III, directly associate with CF1. Subunit II is unlikely to bind CF1 directly and may associate with subunit I. Although depletion of subunit IV does not cause dissociation of CF0 from CF1, its interaction with CF1 subunits is uncertain.     

Chromatographic purification of the chloroplast ATP synthase (CF0-CF1) and the role of CF0 subunit IV in proton conduction

Chromatographic procedures were developed to purify chloroplast ATP synthase (CF0-CF1) in large amounts and to resolve subunits from this enzyme. The ATP synthase thus obtained has high ATP-Pi exchange and Mg2(+)-ATPase activities upon incorporation into asolectin liposomes. The purity of this preparation was about 95%. By modifications of this chromatographic procedure, we purified subunit IV-deficient CF0-CF1, subunit IV-deficient CF0, and subunit IV. Both ATP-Pi exchange and Mg2(+)-ATPase activities were impaired by depletion of subunit IV from CF0-CF1. Partial restoration of these activities was obtained by reconstituting subunit IV-deficient CF0-CF1 with subunit IV. The impairment of these activities was likely caused by a loss in proton conductivity of CF0 upon removal of subunit IV. The dicyclohexylcarbodiimide-sensitive Mg2(+)-ATPase of subunit IV-deficient CF0-CF1 was not as sensitive to the depletion of subunit IV as ATP-Pi exchange. Nearly 90% of subunit IV could be removed, but Mg2(+)-ATPase activity was inhibited by only 40-60%. Thus subunit IV of CF0-CF1 may not participate directly in proton transfer but may have a role in organizing and/or stabilizing CF0 structure.     

Activation of the CF0-CF1, ATP synthase from spinach chloroplasts by chloroplast lipids

The interactions of CF₀-CF₁ with different lipids were studied by following the stimulation of Mg-ATPase and of P_i-ATP exchange activities of reconstituted CF₀-CF₁ proteoliposomes. The following results were obtained: (1) Both P_i-ATP exchange and Mg-ATPase activities are stimulated by lipids. Furthermore, the inhibition of Mg-ATPase by N,N′-dicyclohexylcarbodiimide is dependent on the interactions of CF₀-CF₁ with lipids. (2) A polar lipid extract of thylakoid membranes stimulates Mg-ATPase activity of CF₀-CF₁ more efficiently than phospholipids. The relative effectiveness of Mg-ATPase stimulation is: chloroplast lipids > soybean phospholipids > phosphatidylcholine/phosphatidylserine (4: 1) > phosphatidylcholine. The rate of P_i-ATP exchange in chloroplast lipids CF₀-CF₁ proteoliposomes is, however, lower than in soybean lipids CF₀-CF₁ proteoliposomes, due to their higher permeability to protons. Addition of 10% phosphatidylserine to chloroplast lipids reduces their permeability to protons and stimulates P_i-ATP exchange. (3) The kinetic mechanism of ATPase stimulation by chloroplast lipids is by decreasing the K_m (ATP) and by increasing V_max in comparison to soybean lipid proteoliposomes. This may explain the low affinity for ATP and the slow turnover rate of the purified enzyme in artificial lipids in comparison to the native enzyme in chloroplast thylakoids. (4) Chloroplast lipids lacking monogalactosyldiacylglycerols only poorly activate CF₀-CF₁. A large stimulation of P_i-ATP exchange is obtained by a mixture of 60% monogalactosyldiacylglycerol and 40% of the rest of the chloroplast lipids, but not by mixtures of monogalactosyldiacylglycerol with phospholipids. Hydrogenation of the unsaturated fatty acids of monogalactosyldiacylglycerol inhibits the activation of CF₀-CF₁. (5) The results suggest that: (a) interactions of specific chloroplast lipids with CF₀-CF₁ activates the enzyme by increasing its turnover and its affinity for ATP; (b) specific requirements for CF₀-CF₁ activation are the presence of monogalactosyldiacylglycerols together with another chloroplast lipid component and of highly unsaturated fatty acids.     

Correlation between the ATP synthetic active state and the ATP hydrolytic active state in chloroplast ATP synthase-ATPase complex CF0 . CF1

Illumination of chloroplast thylakoids activates ATP synthase-ATPase complex CF0 . CF1. The time course of ATP synthesis is linear if ADP and Pi are added before or simultaneously with illumination. ATP synthesis initiated by adding the substrates in the light exhibits a curvilinear time course with a low initial rate (Vi). Vi, but not the rate at a steady state, decreases with increasing preillumination time with a half-time of 2 s. Coincident with this decrease in Vi, activation of ATP hydrolysis takes place. In the postillumination dark, restoration of Vi is observed: Vi increases with increasing time intervals between the end of illumination and the addition of the substrates with simultaneous reillumination (half-time of 3 s). Coincident with this restoration of Vi, inactivation of ATP hydrolysis takes place. Such an increase in Vi in the postillumination dark is not observed in thylakoids pretreated with N-ethylmaleimide. These results suggest the following: in the light, the ATP synthetically active, but ATP hydrolytically inactive state (Es) converts to the ATP hydrolytically active, but ATP synthetically inactive (or less active) state (Eh) in the absence of ADP and Pi. The N-ethylmaleimide pretreatment inhibits this process. In the postillumination dark, the reverse conversion takes place.     

CF0, the proton channel of chloroplast ATP synthase. After removal of CF1 it appears in two forms with highly different proton conductance

The discharge of the flash-induced transmembrane voltage through the exposed proton channel, CF0, of the chloroplast ATP synthase, CF0CF1 was investigated. EDTA treatment of thylakoid membranes exposed approximately 50% of total CF0 by removal of the CF1 counterparts. This greatly accelerated the decay of the transmembrane voltage, as was apparent from electrochromic-absorption changes of intrinsic pigments and by pH-indicating-absorption changes of added dyes. Two decay processes were discernible, one rapid with a typical half-decay time of 2 ms, and a slower one with a half-decay time variable between 20-100 ms. Both were sensitive to CF0 inhibitors, but only the rapid decay process was also inhibited by added CF1. CF1 was effective in surprisingly small amounts, which were significantly lower than those previously removed by EDTA treatment. This finding corroborated our previous conclusion that the rapid decay of the transmembrane voltage was attributable to only a few high-conductance channels among many CF0 molecules, typically in the order of one channel/CF1-depleted EDTA vesicle. Inhibition of photophosphorylation in control thylakoids was measured as function of the concentration of CF0 inhibitors. It was compared with the inhibition of proton conduction through exposed CF0 in EDTA vesicles. Photophosphorylation and proton conduction by the high-conductance form of CF0 were inhibited by the same low inhibitor concentrations. This suggested that the high-conducting form of CF0 with a time-averaged single-channel conductance of 1 pS [Lill, H., Althoff, G. & Junge, W. (1987) J. Membrane Biol. 98, 69-78] represented the proton channel in the integral enzyme, which acted as a low-impedance access from the thylakoid lumen to the coupling site in CF0CF1. The slow decay process was attributed to a majority of low-conductance CF0 channels, i.e. about 50 molecules/vesicle. The conductance of these channels was more than 100-fold lower and they did not compete with the very few highly conducting channels for rebinding of added CF1. The low proton conduction of the majority of exposed CF0 molecules, possibly due to a structural rearrangement, may be protecting the thylakoid membrane against rapid energy dissipation caused by accidental loss of CF1. It may also explain the low single-channel conductance of bacterial F0 reported in the literature.     

Proton flux through the chloroplast ATP synthase is altered by cleavage of its gamma subunit

Electron transport, the proton gradient and ATP synthesis were determined in thylakoids that had been briefly exposed to a low concentration of trypsin during illumination. This treatment cleaves the gamma subunit of the ATP synthase into two large fragments that remain associated with the enzyme. Higher rates of electron transport are required to generate a given value of the proton gradient in the trypsin-treated membranes than in control membranes, indicating that the treated membranes are proton leaky. Since venturicidin restores electron transport and the proton gradient to control levels, the proton leak is through the ATP synthase. Remarkably, the synthesis of ATP by the trypsin-treated membranes at saturating light intensities is only slightly inhibited even though the proton gradient is significantly lower in the treated thylakoids. ATP synthesis and the proton gradient were determined as a function of light intensity in control and trypsin-treated thylakoids. The trypsin-treated membranes synthesized ATP at lower values of the proton gradient than the control membranes. Cleavage of the gamma subunit abrogates inhibition of the activity of the chloroplast ATP synthase by the epsilon subunit. Our results suggest that overcoming inhibition by the epsilon subunit costs energy.     

Proton flux through the chloroplast ATP synthase is altered by cleavage of its gamma subunit

Electron transport, the proton gradient and ATP synthesis were determined in thylakoids that had been briefly exposed to a low concentration of trypsin during illumination. This treatment cleaves the γ subunit of the ATP synthase into two large fragments that remain associated with the enzyme. Higher rates of electron transport are required to generate a given value of the proton gradient in the trypsin-treated membranes than in control membranes, indicating that the treated membranes are proton leaky. Since venturicidin restores electron transport and the proton gradient to control levels, the proton leak is through the ATP synthase. Remarkably, the synthesis of ATP by the trypsin-treated membranes at saturating light intensities is only slightly inhibited even though the proton gradient is significantly lower in the treated thylakoids. ATP synthesis and the proton gradient were determined as a function of light intensity in control and trypsin-treated thylakoids. The trypsin-treated membranes synthesized ATP at lower values of the proton gradient than the control membranes. Cleavage of the γ subunit abrogates inhibition of the activity of the chloroplast ATP synthase by the ε subunit. Our results suggest that overcoming inhibition by the ε subunit costs energy.     

ATP hydrolysis catalyzed by a beta subunit preparation purified from the chloroplast energy transducing complex CF1.CF0

Washing thylakoid membranes with 1 M LiCl causes the release of the beta subunit from the chloroplast energy transducing complex (CF1.CF0) in spinach chloroplasts. This protein purifies by size exclusion chromatography as a 180-kDa aggregate and, thus, is probably composed of a trimer of beta polypeptides. The purified aggregate binds ADP to a high and a low affinity site with dissociation constants of 15 and 202 microM, respectively. Mg2+ is required for ADP to bind to both sites. Manganese binds to the protein in a cooperative manner to at least two sites with high affinity. The beta subunit preparation catalyzes Mg2+-dependent ATP hydrolysis at rates which are comparable to other subunit-deficient CF1 preparations and is increased by treatments known to activate the Mg2+-ATPase activity of CF1. However, Ca2+ is not an effective cofactor for this reaction and treatments which activate the Ca2+-ATPase of CF1 are either ineffective or inhibitory.     

Proton channel of the chloroplast ATP synthase,CF0: Its time-averaged single-channel conductance as function of pH,temperature, isotopic and ionic medium composition

Summary The proton-driven ATP synthase of chloroplasts is composed of two elements, CF₀ and CF₁. The membrane bound CF₀ conducts protons and the peripheral CF₁ interacts with nucleotides. By flash spectrophotometric techniques applied to thylakoid membranes from which about 50% of total CF₁ was removed, we have previously determined the protonic (timeaveraged) single-channel conductance of CF₀. Being in the order of 1 pS, it was sufficiently large to support the proposed role of CF₀ as a low-impedance access for protons to the coupling site in CF₀CF₁. On the other hand, it was too large to be readily reconciled with current concepts of proton supply to and proton conduction through the channel.We studied the time-averaged single-channel conductance of CF₀ under variation of pH, pD, ionic composition, temperature, and water/membrane structure with the following results: (i) CF₀ was proton-specific even against a background of 300mm monovalent or 30mm divalent catins. (ii) While the conductance of CF₀ was pH/pD-independent in the range from 5.6–8.0, in D₂O it was lower by a constant factor of 1.7 than in H₂O (iii) Addition of glycerol diminished the conductance and abolished the isotope effect. (iv) The Arrhenius activation energy was 42 kJ/mol and thus intermediate between the ones found for the water-filled pore, gramicidin (30 kJ/mol), and the mobile carrier, valinomycin (65 kJ/mol).The results implied that CF₀ is endowed with an extremely proton-specific (10⁷-fold) selectivity filter. Its conductance is very high, and its conduction cycle is not necessarily rate limited by a protolytic reaction. The mechanisms of rapid proton supply to the channel mouth and of proton conduction remained enigmatic.     

Tightly bound sulpholipids in chloroplast CF0-CF1

Highly purified preparations of CF₀-CF₁ from chloroplasts contain a small amount of tightly bound lipids. Extraction and analysis of these lipids show that they are almost exclusively sulpholipids. The calculated amount of bound sulpholipids in spinach and in Dunaliella salina CF₀-CF₁ preparations are 5 and 20 mols/mol enzyme, respectively. Attempts to exchange the bound lipids with other lipids or with detergents have failed, indicating a very strong association with CF₀-CF₁.     

Presteady-state kinetics of ATP hydrolysis by chloroplast CF1-ATPASE

The kinetics of reversible inactivation of chloroplast CF1-ATPase by Mg2+ and ADP was studied. The rate of inactivation obeys the first-order equation and is independent of ADP concentration. An analysis of the dependence of the inactivation rate on Mg2+ concentration demonstrated that the limiting step of inactivation is other than Mg2+ binding, i.e. the subsequent steps which include, in all probability, the conformational changes of the enzyme. The original Mg2+-dependent activity of CF1-ATPase is close to that observed under steady-state conditions in the presence of sulphate and methanol and exceeds the Ca2+-dependent activity approximately 6-fold. Preincubation of CF1-ATPase with Mg2+ results in inhibition of the original activity of the enzyme. This effect is not removed by addition of the ATP-regenerating system (pyruvate kinase + phosphoenol pyruvate) to the preincubation medium but is diminished by sulphite and the non-hydrolyzed analog of ATP--beta, gamma-methyladenosine-5-triphosphate. After addition of AMPPCP to the reaction mixture the initial reaction rate is decreased, while the steady-state rate is increased. It may be concluded that the Mg2+-dependent inactivation of CF1-ATPase is induced by the tightly bound ADP. The latter can be replaced by ATP, which in contrast to ADP does not form an inactive complex with the enzyme. A comparison of experimental results with literature data suggests that the mechanism of "alternating sites" proposed by Boyer et al. for ATP hydrolysis by soluble CF1-ATPase is not realized under the given experimental conditions.     

Molecular mechanisms of rotational catalysis in the F(0)F(1) ATP synthase

Rotation of the F(0)F(1) ATP synthase gamma subunit drives each of the three catalytic sites through their reaction pathways. The enzyme completes three cycles and synthesizes or hydrolyzes three ATP for each 360 degrees rotation of the gamma subunit. Mutagenesis studies have yielded considerable information on the roles of interactions between the rotor gamma subunit and the catalytic beta subunits. Amino acid substitutions, such as replacement of the conserved gammaMet-23 by Lys, cause altered interactions between gamma and beta subunits that have dramatic effects on the transition state of the steady state ATP synthesis and hydrolysis reactions. The mutations also perturb transmission of specific conformational information between subunits which is important for efficient conversion of energy between rotation and catalysis, and render the coupling between catalysis and transport inefficient. Amino acid replacements in the transport domain also affect the steady state catalytic transition state indicating that rotation is involved in coupling to transport.     

Crystallization of coupling factor 1 (CF1) from spinach chloroplast.

Spinach chloroplast coupling factor (CF₁) was crystal-lized at 20°C from 0.05 M TRIS-PO₄, containing 4 mM ATP, 15mM KCl, 1.0 mM EDTA and 1.80 M (NH₄)₂SO₄, at pH 7.8. Some unit cell parameters were determined by electron microscopy and by X-ray diffraction. The cube shaped crystals have a tetragonal lattice, a = b = 135 Å, c = 280 Å with eight molecules per unit cell; possible space group P422 or P4₂2₁2, hence half a molecule in the asymmetric unit. Crystals grown at pH 7.5 in the absence of ATP have an orthorhombic lattice, a = 125 Å, b = 145 Å, c = 169 Å (C222₁), eight molecules per unit cell.     

Calcium binding to the chloroplast andE. coli (CF0) F0 subunit III (c) of the ATP-synthase

Summary Subunit III and c, the 8 kDa components of the chloroplast CF₀, andE. coli H⁺ channel complexes respectively, were isolated and purified for the purpose of studying their Ca⁺⁺-binding properties. Purified subunit III or c as well as the unfractionated organic-solvent soluble preparation from chloroplasts were used in a⁴⁵Ca⁺⁺-ligand blot assay known to detect high affinity Ca⁺⁺-binding sites in proteins. Both subunit III and c showed strong⁴⁵Ca⁺⁺-binding. None of the other CF₀ subunits bound Ca⁺⁺ and of the CF₁ only a weak binding was detected in the region of the , subunits. The Ca⁺⁺-binding was inhibited after treating the proteins in solution by derivatizing aqueously exposed carboxyl groups with a water soluble carbodiimide plus a nucleophile, after de-formylation of the N-terminal methionine, or with a subsequent treatment with La³⁺. Dicyclohexylcarbodiimide treatment (no nucleophile was added) of thylakoid membranes, which derivatizes the hydrophobically located Glu 61 (Asp 61 inE. coli), did not inhibit the Ca⁺⁺-binding in either protein. The data indicate that for both proteins the carbonyl group of the formylated N-terminal Met-1 and probably the carboxyl group of the subunit III (or c) C-terminal provide some of seven essential oxygen ligands normally required for defining a Ca⁺⁺-binding site in proteins. Based on the accepted models for the hairpin conformation of the subunit III (c), it seems clear that the Ca⁺⁺-binding site can form on the lumenal side of the membrane in the functional CF₀ structures or on the periplasmic side of theE. coli membrane. A working hypothesis we are testing is that Ca⁺⁺-binding to the CF₀ (or F₀) can form an easily reversible gating site such as to enhance the probability for membrane-localized H⁺ gradients being coupled to ATP formation under moderate energization loads, but under excess energization the local H⁺ ion concentration may build up high enough to displace the bound Ca⁺⁺, resulting in delocalization of the H⁺ gradient. The latter situation seems, in chloroplasts at least, to function as a signal for over-energization; i.e., excess light absorption, a potential stress situation for plants. Lumenal acidification appears to be a trigger for initiating stress alleviation responses.On leave from the Institute of Soil Science and Photosynthesis, Russian Academy of Sciences, Puschchino, Russia.     

ATP binding to noncatalytic sites of chloroplast coupling factor CF1

A kinetic analysis of ATP binding to noncatalytic sites of chloroplast coupling factor CF₁ was made. The ATP binding proved to be unaffected by reduction of the disulfide bridge of the CF₁ -subunit. The first-order equation describing nucleotide binding to noncatalytic sites allowed for two vacant nucleotide binding sites different in their kinetics. As suggested by nucleotide concentration dependence of the rate of nucleotide binding, the tight binding was preceded by rapid reversible binding of nucleotides. Preincubation of CF₁ with Mg²⁺ resulted in a decreased rate of ATP binding. ATP dissociation from noncatalytic sites was described by the first order equation for similar sites with a dissociation rate constant k _d (ATP) 10^–3 min^–1. Noncatalytic sites of CF₁ were shown to be not homogeneous. One of them retained the major part of endogenous ADP after precipitation of CF₁ with ammonium sulfate. Its two other sites differed in kinetic parameters and affinity for ATP. Anions of phosphate, sulfite, and especially, pyrophosphate inhibited the interaction between ATP and the noncatalytic sites.