Chloroplast Coupling Factor CF1 Research Articles

Conformational Changes in Chloroplast F1-ATPase Caused by Thiol-Dependent Activation and MgADP-Dependent Inactivation

Abstract—The state of tyrosine residues of the chloroplast coupling factor CF1 was studied by spectrophotometric titration. It was shown that some tyrosine residues of CF1 underwent deprotonation at pH values of the medium much lower than the pK of free tyrosine. The number of such residues depends on both the conformational state of the enzyme and the composition of the medium. They are abundant in CF1 with the γ-subunit that contains a disulfide bridge. Bridge reduction leads to a decrease of their number and, accordingly, an increase in the number of residues that undergo deprotonation at a pH higher than the tyrosine pK. The minimum number of residues that dissociated within the 6.0–9.0 pH range was observed in the reaction mixture containing Mg2+ or MgADP. It is assumed that the changes in pK values for tyrosine residues result from the presence or absence of positively charged amino-acid residues in their neighborhood, which is indicative of alterations in the tertiary structure of the enzyme. Deprotonation of a considerable part of tyrosine residues in the presence of Mg2+ or MgADP occurs within an abnormally narrow pH range and demonstrates the cooperative transition to the new conformational state of the enzyme. Comparison of the data obtained with our previous kinetic data indicates that the titration characteristics and the respective structures of CF1-ATPase observed in the presence of Mg2+ or MgADP result from reversible inactivation caused by MgADP binding to one catalytic site and one noncatalytic site.

The effect of medium viscosity on kinetics of ATP hydrolysis by the chloroplast coupling factor CF1.

The coupling factor CF1 is a catalytic part of chloroplast ATP synthase which is exposed to stroma whose viscosity is many-fold higher than that of reaction mixtures commonly used to measure kinetics of CF1-catalyzed ATP hydrolysis. This study is focused on the effect of medium viscosity modulated by sucrose or bovine serum albumin (BSA) on kinetics of Ca(2+)- and Mg(2+)-dependent ATP hydrolysis by CF1. These agents were shown to reduce the maximal rate of Ca(2+)-dependent ATPase without changing the apparent Michaelis constant (К m), thus supporting the hypothesis on viscosity dependence of CF1 activity. For the sulfite- and ethanol-stimulated Mg(2+)-dependent reaction, the presence of sucrose increased К m without changing the maximal rate that is many-fold as high as that of Ca(2+)-dependent hydrolysis. The hydrolysis reaction was shown to be stimulated by low concentrations of BSA and inhibited by its higher concentrations, with the increasing maximal reaction rate estimated by extrapolation. Sucrose- or BSA-induced inhibition of the Mg(2+)-dependent ATPase reaction is believed to result from diffusion-caused deceleration, while its BSA-induced stimulation is probably caused by optimization of the enzyme structure. Molecular mechanisms of the inhibitory effect of viscosity are discussed. Taking into account high protein concentrations in the chloroplast stroma, it was suggested that kinetic parameters of ATP hydrolysis, and probably those of ATP synthesis in vivo as well, must be quite different from measurements taken at a viscosity level close to that of water.

The effect of viscosity-increasing agents on ATP synthesis in chloroplast thylakoids

The effect of an increase in the viscosity of the medium on cyclic photophosphorylation in chloroplast thylakoids and the Ca2+-dependent ATP hydrolysis by the chloroplast coupling factor CF1 was studied. Using 0.1–0.2 mM ADP, it was found that the rate of ATP synthesis decreases after the addition of various agents that increase the viscosity of the medium (sucrose, dextran 40, or polyethylene glycol 6000) provided that these agents cause neither uncoupling nor electron-transport inhibition in the absence of ADP. Dextran and polyethylene glycol inhibited ATP synthesis by 50% when their concentrations were much lower (6–10%) than that of sucrose (30–40%), while 50% inhibition of the Ca2+-dependent ATP hydrolysis by CF1-ATPase was observed at higher concentrations of dextran and polyethylene glycol (9–13%) and lower concentrations of sucrose (about 20%). For ADP, the effective Michaelis constant (KM) was shown to increase by factors of 2–3 with increasing viscosity, while the maximum rate of cyclic photophosphorylation remained virtually unchanged. The dependence of KM on the medium viscosity can serve as a criterion for the operation of diffusion-controlled photophosphorylation. Possible mechanisms of the diffusion of ADP and ATP are discussed.

Interaction of ADP and ATP with noncatalytic sites of isolated and membrane-bound chloroplast coupling factor CF1.

This study of ATP and ADP binding to noncatalytic sites of membrane-bound CF1 (ATP synthase) revealed two noncatalytic sites with different specificities and affinities for nucleotides. One of these is characterized by a high affinity and specificity to ADP (Kd=2.6+/-0.3 microM). However, a certain increase in ADP apparent dissociation constant at high ATP/ADP ratio in the medium allows a possibility that ATP binds to this site as well. The other site displays high specificity to ATP. When the ADP-binding site is vacant, it shows a comparatively low affinity for ATP, which greatly increases with increasing ADP concentration accompanied by filling of the ADP-binding site. The reported specificities of these two sites are independent of thylakoid membrane energization, since both in the dark and in the light the ratios of ATP/ADP tightly bound to the noncatalytic sites were very close. The difference in noncatalytic site affinity for ATP and ADP is shown to depend on the amount of delta subunit in a particular sample. Thylakoid membrane ATP synthase, with stoichiometric content of delta-subunit (one delta-subunit per CF1 molecule), showed the maximal difference in ADP and ATP affinities for the noncatalytic sites. For CF1, with substoichiometric delta subunit values, this difference was less, and after delta subunit removal it decreased still more.

Influence of growth temperature on structure, thermostability and kinetic properties of the ATPase coupling factor AF1 of a thermophilic blue-green alga (cyanobacterium).

The coupling factor AF1 isolated from cells of the thermophilic blue-green alga Mastigocladus laminosus grown at 40 degrees C, 50 degrees C and 60 degrees C is investigated and compared with the chloroplast coupling factor CF1. It is demonstrated that the structure of AF1 is affected by the growth temperature. The AF1 from 60 degrees C shows a split alpha-band in dodecylsulfate/polyacrylamide gel electrophoresis which is not observed in AF1 from 40 degrees C or from 50 degrees C. Only the AF1 from 60 degrees C aggregates with allophycocyanin. The AF1 from 60 degrees C is stable up to 85 degrees C for 60 min whereas the AF1 from 40 degrees C and 50 degrees C denature between 65 degrees C and 70 degrees C. CF1 is much less stable. In contrast to the structure, the kinetic properties of the coupling factor are not influenced by the growth temperature. Furthermore, the reaction rates of all three AF1 preparations have the same temperature optimum which jis also identical to that of CF1. The kinetic properties of the AF1 from 50 degrees C were determined in more detail. In the absence of ADP the saturation curve for ATP shows Michaelis-Menten kinetics with Km = 480 micromol and V = 30 micromol ATP hydrolyzed X mg protein-1 X min-1. With the addition of ADP this curve becomes sigmoidal and V decreases. This behaviour and the Hill values suggest an allosteric enzyme with two active sites acting cooperatively. ADP as a product of the ATPase reaction inhibits the enzyme. The inhibition is not simply competitive, it also influences the cooperativity of the active sites.

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 gamma-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 kd(ATP) approximately/= 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.

Interaction of sulfite with the noncatalytic and catalytic sites of chloroplast coupling factor cf1.

The interaction between sulfite, an efficient Mg2+-dependent F1-ATPase activator, and chloroplast CF1-ATPase was studied. The sulfite anion was shown to inhibit ADP and ATP binding to the noncatalytic sites of CF1. The stimulating activity of sulfite persists when all noncatalytic sites are nucleotide-occupied. Phosphate, a competing candidate for binding to CF1 catalytic sites, suppresses this activity. These results support the suggestion that the stimulation of Mg2+-dependent ATPase activity of CF1 is caused by sulfite binding to its catalytic sites.

Mg2+ -dependent inactivation / H+ -dependent activation equilibrium of chloroplast F1-ATPase

The catalytic part of chloroplast thylakoid ATPase, the chloroplast coupling factor CF1, is reversibly inactivated during incubation in the presence of Mg2+. The inactivation has two phases. Its fast phase occurs at basic pH of the incubation medium (k = 6 min-1), while the slow phase ( k = 0.1-0.2 min-1) depends on pH only slightly throughout the studied range (5.5-9.0). As followed from changes in the inactivation effect of magnesium ions, Mg2+ affinity for the enzyme decreases dramatically with decreasing medium pH. The pH-dependence of Mg2+ dissociation apparent constant suggests that the binding/dissociation equilibrium is determined by protonation/deprotonation of specific acid-base groups of the enzyme. The analysis of pH-dependence plots gives the equilibrium constant of magnesium dissociation (3-9 μM) and the dissociation constant of the protonated groups pK 5.8-6.7). Sodium azide is known to stabilize the inactive CF1-MgADP complex; when added to the incubation medium it diminishes the Mg2+ dissociation constant and has no effect on the dissociation constant of the acid-base groups. At lower pH, Mg2+-inactivated CF1-ATPase reactivates. Octyl glucoside accelerates the reactivation, while Triton-100 affects it only slightly. The reactivation rate of membrane-bound CF1 (thylakoid ATPase) inactivated by preincubation with Mg2+ in the presence of gramicidin is a few times higher than that of isolated CF1. These results suggest that the reactivation of isolated and membrane-bound CF1-ATPase is determined by protonation of a limited number of acid-base groups buried in the enzyme molecule.

Spinach chloroplast coupling factor CF1-alpha 3 beta 3 core complex: structure, stability, and catalytic properties.

A minimal chloroplast coupling factor CF1 core complex, containing only alpha and beta subunits, has been isolated from spinach thylakoids [Avital, S., & Gromet-Elhanan, Z. (1991) J. Biol. Chem. 266, 7067-7072]. This CF1(alpha beta) exhibited a low MgATPase activity, which was stimulated but not inhibited by low concentrations of the species-specific CF1 effector tentoxin. As is reported here, the structure of CF1(alpha beta) could not be determined due to its instability. However, its pretreatment with high tentoxin concentrations resulted in a remarkable 50-fold stimulation of the MgATPase activity as well as stabilization of its hexameric structure, thus enabling the isolation of a more active CF1-alpha 3 beta 3 complex by size-exclusion chromatography. A detailed characterization of the MgATPase activity of this tentoxin-stabilized CF1-alpha 3 beta 3 hexamer, as compared to the activity of a CF1 complex lacking the epsilon subunit, revealed similar apparent Km values and a similar stimulation by the presence of 100 microM tentoxin in the assay medium, but drastic differences in all other tested assays. Most pronounced were their different temperature profiles and different responses to all added inhibitors and stimulators of the CF1 MgATPase activity and to excess free Mg2+ ions. The specific properties of the stable CF1-alpha 3 beta 3 hexamer are identical to those earlier reported for its parent-unstable CF1(alpha beta). These results indicate that, although the CF1 gamma subunit is not required for the low CF1(alpha beta) ATPase activity nor for the higher activity of the tentoxin-stabilized CF1-alpha 3 beta 3, it plays a central role in obtaining the typical functional properties of the CF1-ATPase. Kinetic cooperativity could not be critically tested as yet with any F1-alpha 3 beta 3. However, tentoxin, as azide, has been shown to inhibit multisite but not unisite catalysis. Therefore, the observation that CF1-alpha 3 beta 3 is only stimulated by tentoxin suggests that the required presence of CF1-gamma for obtaining inhibition by tentoxin reflects the role of this subunit in cooperative interactions between the catalytic sites.

Purified subunit δ of chloroplast coupling factor CF1 reconstitutes photophosphorylation in partially CF1‐depleted membranes

The ATP synthase of chloroplasts consists of the proton channel, CF0, and the catalytic part, CF1, which carries nucleotide-binding sites on subunits alpha and beta. The still poorly understood interaction between CF0 and the catalytic sites on CF1 is mediated by the smaller subunits gamma, delta and epsilon of CF1. We investigated the ability of purified delta to block proton leakage through CF0 channels after their exposure by removal of the CF1 counterpart. Thylakoids were partially depleted of CF1 by EDTA treatment. This increased their proton permeability and thereby reduced the rate of photophosphorylation. Subunit delta was isolated and purified by FPLC [Engelbrecht, S. and Junge, W. (1987) FEBS Lett. 219, 321-325]. Addition of delta to EDTA-treated thylakoids reconstituted high rates of phenazine-methosulfate-mediated photophosphorylation. Since delta does not interact with nucleotides by itself, the reconstitution was due to a reduction of the proton leakage through open CF0 channels. The molar ratio of purified delta over exposed CF0, which started to elicit this effect, was 3:1. However, if delta was added together with purified CF1 lacking delta, in a 1:1 molar ratio, the relative amount over exposed CF0 was as low as 0.06. This corroborated our previous conclusion [Lill, H., Engelbrecht, S., Schönknecht, G. and Junge, W. (1986) Eur. J. Biochem. 160, 627-634] that only a very small fraction of exposed CF0 was actually proton-conducting but with a very high unit conductance. CF1 including delta was apparently rebound preferentially to open CF0 channels. Although the ability of delta to control proton conduction through CF0 was evident, it remains to be established whether delta acts as a gated proton valve or as a conformational transducer in the integral CF0CF1 ATPase.

The proton channel, CF0, in thylakoid membranes

We investigated the conductance of pea thylakoid membranes and their capacity for photophosphorylation as function of the extraction of chloroplast coupling factor CF1. The degree of extraction was varied via the incubation time in EDTA-containing hypo-osmolar medium and was measured by rocket electroimmunodiffusion. The conductance of thylakoid membranes was measured by flash kinetic spectrophotometry. The time course of extraction followed the time course of thylakoid swelling. Contrary to expectation increasing loss of CF1 did not primarily increase the velocity of proton efflux from each vesicle. Instead proton-tight vesicles were converted to leaky ones, which lost phosphorylating activity. Two subpopulations occurred, although both types of vesicles, leaky and proton-tight ones, were CF1-depleted to a similar degree. This implied that only a small fraction of CF1-lacking CF0 was functional as a proton channel. Tight vesicles had no functional channels while leaky ones had at least one. We determined the proportion of tight vesicles in three independent ways: via the residual phosphorylation activity, via measurements of proton efflux and via measurements of the electric relaxation across the membrane. The results obtained were identical. A statistical evaluation of the data led us to the following conclusions. EDTA treatment produced vesicles containing approximately 10(5) chlorophyll molecules, equivalent to a total of approximately 100 CF0CF1 per vesicle. Even at the highest degree of extraction (75% of total CF1 extracted) only 2.5 out of 75 exposed CF0 per vesicle were proton-conducting. The unit conductance of one open CF0 channel was 169 +/- 18 fS at pH 7.5 and room temperature. At an electrical driving force of 100 mV this was equivalent to the passage of approximately 10(5) protons/s. The most important consequence of this relatively high unit conductance was that a single open CF0 channel was capable of dissipating the protonmotive force of one vesicle, thereby deactivating the whole remaining catalytic capacity of this vesicle.