Assessment of total catalytic sites and the nature of bound nucleotide participation in photophosphorylation

Modulation of the mitochondrial permeability transition by cyclophilin D: moving closer to F(0)-F(1) ATP synthase?

Cyclophilin D was recently shown to mask an inhibitory site of the mitochondrial permeability transition pore (PTP) for phosphate, and to constitutively bind F(0)-F(1) ATP synthase resulting in the slowing of ATP synthesis and hydrolysis rates, thus regulating matrix adenine nucleotide levels. Here we review the striking similarities of the factors affecting the threshold for PTP induction, to those affecting binding of phosphate to formerly proposed sides on F(1)-ATPase affecting ATP hydrolytic activity, including critical arginine residues, matrix pH, [Mg(2+)], adenine nucleotides and proton motive force. Based on these similarities, we scrutinize the hypothesis that in depolarized mitochondria exhibiting reversal of F(0)-F(1) ATP synthase operation, the genetic ablation of cyclophilin D or its inhibition by cyclosporin A results in accelerated proton pumping by ATP hydrolysis, opposing a further decrease in membrane potential and promoting high matrix phosphate levels, both negatively affecting the probability of PTP opening.

Regulatory mechanisms of proton-translocating F(O)F (1)-ATP synthase

H(+)-F(O)F(1)-ATP synthase catalyzes synthesis of ATP from ADP and inorganic phosphate using the energy of transmembrane electrochemical potential difference of proton (deltamu(H)(+). The enzyme can also generate this potential difference by working as an ATP-driven proton pump. Several regulatory mechanisms are known to suppress the ATPase activity of F(O)F(1): 1. Non-competitive inhibition by MgADP, a feature shared by F(O)F(1) from bacteria, chloroplasts and mitochondria 2. Inhibition by subunit epsilon in chloroplast and bacterial enzyme 3. Inhibition upon oxidation of two cysteines in subunit gamma in chloroplast F(O)F(1) 4. Inhibition by an additional regulatory protein (IF(1)) in mitochondrial enzyme In this review we summarize the information available on these regulatory mechanisms and discuss possible interplay between them.

ATP synthase--past and future

This paper gives an overview of a lecture scheduled for the opening of the 10th European Bioenergetics Congress. In this lecture I plan to first reflect on the accomplishments of some of the individuals who were involved in research on the ATP synthase during the past 50 years. Then I will give a brief view of the present information about rotational catalysis by the ATP synthase. This will be followed by a discussion of some results from my laboratory that call for additional experimentation. Finally I will direct attention to other questions about the ATP synthase that should be addressed in future studies.

Conformational changes of the mitochondrial F1-ATPase epsilon-subunit induced by nucleotide binding as observed by phosphorescence spectroscopy

Changes in conformation of the epsilon-subunit of the bovine heart mitochondrial F1-ATPase complex as a result of nucleotide binding have been demonstrated from the phosphorescence emission of tryptophan. The triplet state lifetime shows that whereas nucleoside triphosphate binding to the enzyme in the presence of Mg2+ increases the flexibility of the protein structure surrounding the chromophore, nucleoside diphosphate acts in an opposite manner, enhancing the rigidity of this region of the macromolecule. Such changes in dynamic structure of the epsilon-subunit are evident at high ligand concentration added to both the nucleotide-depleted F1 (Nd-F1) and the F1 preparation containing the three tightly bound nucleotides (F1(2,1)). Since the effects observed are similar in both the F1 forms, the binding to the low affinity sites must be responsible for the conformational changes induced in the epsilon-subunit. This is partially supported by the observation that the Trp lifetime is not significantly affected by adding an equimolar concentration of adenine nucleotide to Nd-F1. The effects on protein structure of nucleotide binding to either catalytic or noncatalytic sites have been distinguished by studying the phosphorescence emission of the F1 complex prepared with the three noncatalytic sites filled and the three catalytic sites vacant (F1(3,0)). Phosphorescence lifetime measurements on this F1 form demonstrate that the binding of Mg-NTP to catalytic sites induces a slight enhancement of the rigidity of the epsilon-subunit. This implies that the binding to the vacant noncatalytic site of F1(2,1) must exert the opposite and larger effect of enhancing the flexibility of the protein structure observed in both Nd-F1 and F1(2,1). The observation that enhanced flexibility of the protein occurs upon addition of adenine nucleotides to F1(2,1) in the absence of Mg2+ provides direct support for this suggestion. The connection between changes in structure and the possible functional role of the epsilon-subunit is discussed.

The 'non-exchangeable' nucleotides of F1-F0ATP synthase. Cofactors in hydrolysis?

The F1-F0 ATP synthase bears 6 nucleotide binding sites, only 3 of which turn over during catalysis. The remaining 3 are occupied by slowly exchanging ATP in vivo, although at least 1 molecule is generally lost on isolation of the enzyme in the absence of nucleotide. It is proposed that the function of the slowly exchanging (NC) nucleotides is to participate in catalysis, the terminal phosphate of the bound ATP acting as an acid catalyst in the cleavage/synthesis of the phosphate anhydride bond in the catalytic sites. Such a role has been demonstrated for the bound pyridoxal phosphate moiety in glycogen phosphorylase. Evidence is presented that (i) the NC nucleotide spans the interface between an alpha subunit and its partner beta, interacting near the catalytic binding site on beta; (ii) the phosphate moieties of the catalyzed and NC nucleotide are close in space; and (iii) occupation of the NC nucleotide sites promotes ATP hydrolysis by F1 or its subfragments. All of these findings are required by the proposed mechanism. Relationships between phosphorylase and F1 structures are discussed.

Nucleotide-dependent and dicyclohexylcarbodiimide-sensitive conformational changes in the epsilon subunit of Escherichia coli ATP synthase

The rate of trypsin cleavage of the epsilon subunit of Escherichia coli F1F0 (ECF1F0) is shown to be ligand-dependent as measured by Western analysis using monoclonal antibodies. The cleavage of the epsilon subunit was rapid in the presence of ADP alone, ATP + EDTA, or AMP-PNP + Mg2+, but slow when Pi was added along with ADP + Mg2+ or when ATP + Mg2+ was added to generate ADP + Pi (+Mg2+) in the catalytic site. Trypsin treatment of ECF1Fo was also shown to increase enzymic activity on a time scale corresponding to that of the cleavage of the epsilon subunit, indicating that the epsilon subunit inhibits ATPase activity in ECF1Fo. The ligand-dependent conformational changes in the epsilon subunit were also examined in cross-linking experiments using the water-soluble carbodiimide 1-ethyl-3-[3-(dimethylamino)propyl]-carbodiimide (EDC). In the presence of ATP + Mg2+ or ADP + Pi + Mg2+, the epsilon subunit cross-linked product was much reduced. Prior reaction of ECF1Fo with dicyclohexylcarbodiimide (DCCD), under conditions in which only the Fo part was modified, blocked the conformational changes induced by ligand binding. When the enzyme complex was reacted with DCCD in ATP + EDTA, the cleavage of the epsilon subunit was rapid and yield of cross-linking of beta to epsilon subunit low, whether trypsin cleavage was conducted in ATP + EDTA or ATP + Mg2+. When enzyme was reacted with DCCD in ATP + Mg2+, cleavage of the epsilon subunit was slow and yield of cross-linking of beta to epsilon high, under all nucleotide conditions for proteolysis.(ABSTRACT TRUNCATED AT 250 WORDS)

Catalytic site nucleotide and inorganic phosphate dependence of the conformation of the epsilon subunit in Escherichia coli adenosinetriphosphatase

The rate of trypsin cleavage of the epsilon subunit of Escherichia coli F1 (ECF1) has been found to be ligand-dependent, as measured indirectly by the activation of the enzyme that occurs on protease digestion, or when followed directly by monitoring the cleavage of this subunit using monoclonal antibodies. The cleavage of the epsilon subunit was fast in the presence of ADP alone, ADP + MG2+, ATP + EDTA, or AMP-PNP, but slow when Pi was added along with ADP + Mg2+ or when ATP + Mg2+ was added to generate ADP + Pi (+Mg2+) in the catalytic site(s). The half-maximal concentration of Pi required in the presence of ADP + Mg2+ to protect the epsilon subunit from cleavage by trypsin was 50 microM, which is in the range measured for the high-affinity binding of Pi to F1. The ligand-dependent conformational changes in the epsilon subunit were also examined in cross-linking experiments using the water-soluble carbodiimide 1-ethyl-3-[3-(dimethylamino)propyl]carbodiimide (EDC). In the presence of ATP + Mg2+ or ADP + Mg2+ + Pi, the epsilon subunit cross-linked to beta in high yield. With ATP + EDTA or ADP + Mg2+ (no Pi), the yield of the beta-epsilon cross-linked product was much reduced. We conclude that the epsilon subunit undergoes a conformational change dependent on the presence of Pi. It has been found previously that binding of the epsilon subunit to ECF1 inhibits ATPase activity by decreasing the off rate of Pi [Dunn, S. D., Zadorozny, V. D., Tozer, R. G., & Orr, L. E. (1987) Biochemistry 26, 4488-4493]. This reciprocal relationship between Pi binding and epsilon-subunit conformation has important implications for energy transduction by the E. coli ATP synthase.

F1 ATPase from the thermophilic bacterium PS3 (TF1) shows ATP modulation of oxygen exchange

The ATPase from the ATP synthase of the thermophilic bacterium PS3 (TF1), unlike F1 ATPase from other sources, does not retain bound ATP, ADP, and Pi at a catalytic site under conditions for single-site catalysis [Yohda, M., & Yoshida, M. (1987) J. Biochem. 102, 875-883]. This raised a question as to whether catalysis by TF1 involved alternating participation of catalytic sites. The possibility remained, however, that there might be transient but catalytically significant retention of bound reactants at catalytic sites when the medium ATP concentration was relatively low. To test for this, the extent of water oxygen incorporation into Pi formed by ATP hydrolysis was measured at various ATP concentrations. During ATP hydrolysis at both 45 and 60 degrees C, the extent of water oxygen incorporation into the Pi formed increased markedly as the ATP concentration was lowered to the micromolar range, with greater modulation observed at 60 degrees C. Most of the product Pi formed arose by a single catalytic pathway, but measurable amounts of Pi were formed by a pathway with high oxygen exchange. This may result from the presence of some poorly active enzyme. The results are consistent with sequential participation of three catalytic sites on the TF1 as predicted by the binding change mechanism.

Catalytic and regulatory effects of light intensity on chloroplast ATP synthase

The incorporation of water oxygens into ATP made by photophosphorylation is known to be increased markedly when either Pi or ADP concentration is lowered. The present studies show a similar increase in oxygen exchange when light intensity is lowered even with ample ADP and Pi present. The number of reversals of bound ATP formation prior to release increases about 1 to about 27 in the presence of dithiothreitol and to 5 in its absence. The equilibrium of the bound reactants still favors ATP at low light intensity, as shown by measurement of the amount of bound ATP rapidly labeled from [32P]Pi during steady-state photophosphorylation. Changes observed in the interconversion rate in the absence of added thiol are likely involved in the regulation of the dark ATPase activity in the chloroplast. The interconversion rate of bound ATP to bound ADP and Pi in the presence of thiol is about the same at low and high light intensities. This rate of bound ATP formation is not sufficient, however, to account for the maximum rate of photophosphorylation. Thus, when adequate protonmotive force is present, the rate of conversion of bound ADP and Pi to bound ATP, and possibly that of bound ATP to bound ADP and Pi, must be increased, with proton translocation being completed only when bound ATP is present to be released. These observations are consistent with the predictions of the binding change mechanism with sequential participation of catalytic sites and are accommodated by a simplified general scheme for the binding change mechanism that is presented here.(ABSTRACT TRUNCATED AT 250 WORDS)

Steady-state rate of F1-ATPase turnover during ATP hydrolysis by the single catalytic site

Under the conditions of ATP regeneration and molar excess of nucleotide-depleted F1-ATPase the enzyme catalyses steady-state ATP hydrolysis by the single catalytic site. Values of Km = 10(-8) M and Vm = 0.05 s-1 for the single-site catalysis have been determined. ADP release limits single-site ATP hydrolysis under steady-state conditions. The equilibrium constant for ATP hydrolysis at the F1-ATPase catalytic site is less than or equal to 0.7.

Synthesis of ATP by soluble mitochondrial F1 ATPase and F1-inhibitor-protein complex in the presence of organic solvents

The F1 and F1-inhibitor-protein complex synthesized tightly bound ATP from ADP and Pi when the organic solvents dimethylsulfoxide (20-50% v/v), ethylene glycol (20-60% v/v) or poly(ethylene glycol) 4000 and 8000 (30-50% w/v) were included in the assay media. There was no synthesis of tightly bound ATP in the absence of organic solvents. In the presence of 50% dimethylsulfoxide, maximal synthesis of ATP was obtained at pH values between 6.5 and 7.7. In both F1 and F1-inhibitor-protein there was no synthesis of ATP in the absence of MgCl2. The rate of ATP synthesis became faster as the MgCl2 concentration in the medium was raised from 0.1-10 mM. The Km for Pi of F1 was in the range of 0.8-1.5 mM. The Km for Pi of the F1-inhibitor-protein was much higher than that of F1 and could not be measured. In the presence of 10 mM MgCl2 and 2 mM Pi, the rate constants of ATP synthesis by F1 and F1-inhibitor-protein were 5.2-10.4 h-1 and 3.5-5.9 h-1 respectively. For both enzymes the rate constant of ATP hydrolysis was 0.69 h-1. The tightly bound ATP of F1 and F1-inhibitor-protein were hydrolyzed at a much slower rate when either the Pi concentration or the MgCl2 concentration was suddenly decreased. Both in presence and absence of Mg2+, 40-60% of the radioactive tightly bound ATP synthesized by F1 was hydrolyzed when non-radioactive ATP was added to the assay medium. This was not observed when F1-inhibitor-protein was used.

Orthophosphate-pyrophosphate exchange catalyzed by soluble and membrane-bound inorganic pyrophosphatases. Role of H+ gradient

A comparative study of the orthophosphate-pyrophosphate exchange reaction catalyzed by the soluble pyrophosphatase from baker's yeast and by the membrane-bound pyrophosphatase of Rhodospirillum rubrum chromatophores was performed. In both systems the rate of exchange increased when the pH of the medium was raised from 6.0 to 7.8 and when the MgCl2 concentration was raised from 0.1 mM to 20 mM. For the yeast pyrophosphatase the exchange rates measured at different pH values and in the presence of 6.7 to 8.8 mM free Mg2+ superimposed as a single curve when plotted as a function of the concentrations of either HPO4(2-) or MgHPO4. This was not observed with the use of R. rubrum chromatophores. With yeast pyrophosphatase, the Km for Pi was higher than 10 mM and could not be measured when the free Mg2+ concentration in the medium was lower than 0.5 mM. There was a decrease in the Km for Pi when the free Mg2+ concentration was raised to 6.7-8.8 mM or when, in the presence of low free Mg2+, the organic solvents dimethylsulfoxide (20% v/v) or ethyleneglycol (40% v/v) were included in the assay medium. In the presence of 6.7-8.8 mM free Mg2+ the Km for total Pi was 7 mM at pH 7.0 and 12 mM at pH 7.8. For the ionic species HPO4(2-) and MgHPO4, the Km values were 5.8 mM and 4.2 mM respectively. In the presence of 0.24-0.42 mM free Mg2+ and either 20% (v/v) dimethylsulfoxide or 40% (v/v) ethyleneglycol the Km values for total Pi, HPO4(2-) and MgHPO4 were 7.6, 3.5 and 0.5 mM respectively. With R. rubrum chromatophores, the Km for Pi in the presence of 5.5-7.5 mM free Mg2+ was very high and could not be measured. In the presence of 0.24-0.45 mM free Mg2+ the ratio between the velocities of hydrolysis and synthesis of pyrophosphate measured at pH 7.8 with yeast pyrophosphatase and chromatophores of R. rubrum were practically the same. When the free Mg2+ concentration was raised to 5.5-8.8 mM this ratio decreased from 1028 to 540 when the yeast pyrophosphatase was used and from 754 to 46 when chromatophores were used.

Integrated functioning of the chloroplast coupling factor

This review is focused on some functional characteristics of the chloroplast coupling factor. The structure of the enzyme and the putative role of its subunits are recalled. An attempt is made to discriminate the driving force and the activator effects of the electrochemical proton gradient. Respective roles of delta pH, delta phi, external and internal pH are discussed with regard to mechanistic implications. The hypothesis of a functional switch of the enzyme between two states with better efficiency either in ATP synthesis or in ATP hydrolysis is also examined. A brief survey is made on some problems complicating quantitative studies of energy coupling, such as localized chemiosmosis, delta pH and delta phi computations, and scalar ATPases. The main data on the enzyme activation and the energy-dependent release of tightly bound nucleotides are summarized. The arguments for and against the catalytic competence of theses nucleotides are reviewed. Lastly, some prevailing models of the catalytic mechanism are presented. The relevance of nucleotides binding change events in this process is discussed.

Catalytic properties of the Escherichia coli proton adenosinetriphosphatase: evidence that nucleotide bound at noncatalytic sites is not involved in regulation of oxidative phosphorylation

Nucleotide-depleted F1-ATPase from Escherichia coli was reconstituted with F1-depleted membranes and shown to catalyze high rates of oxidative phosphorylation of ADP and GDP. Adenine nucleotide became bound to the nonexchangeable nucleotide sites on membrane-bound F1 during ATP synthesis, but binding of guanine nucleotides to nonexchangeable sites during GTP synthesis was not detectable. It was possible to reload the nonexchangeable sites on nucleotide-depleted F1 with radioactive adenine nucleotide prior to membrane reconstitution. The radioactive adenine nucleotide did not exchange significantly during oxidative phosphorylation of ADP or GDP. The amount of nonexchangeable adenine nucleotide found in membrane-bound F1 was the same when the nonexchangeable sites were reloaded either prior to membrane reconstitution of the F1 or after membrane reconstitution with nucleotide-free F1 followed by a burst of oxidative phosphorylation of ADP. The results showed that occupation of the nonexchangeable sites on F1 by tightly bound nucleotide is not required for oxidative phosphorylation of GDP (a physiological activity of F1 in the bacterial cell). Also, the results confirm directly that the adenine-specific nonexchangeable sites on F1 are noncatalytic sites. Using this experimental approach, it was possible to look for a regulatory effect of the nonexchangeable nucleotide on oxidative phosphorylation. Nucleotide-depleted F1 was first reloaded with (i) ATP, (ii) ADP, (iii) 5'-adenylyl imidodiphosphate, or (iv) zero nucleotide, and was then reconstituted with F1-depleted membranes. The reconstituted membranes were compared in respect to rates of oxidative phosphorylation of GDP and Km values of GDP and Pi. No regulatory role for the nonexchangeable nucleotide was evident.(ABSTRACT TRUNCATED AT 250 WORDS)

Synthesis of pyrophosphate by chromatophores of Rhodospirillum rubrum in the light and by soluble yeast inorganic pyrophosphatase in water-organic solvent mixtures

Chromatophores of Rhodospirillum rubrum contain a membrane-bound pyrophosphatase that synthesizes pyrophosphate when an electrochemical H+ gradient is formed across the chromatophore membrane upon illumination. In this report it is shown that MgCl2 and Pi have different effects on the synthesis of pyrophosphate in the light depending on whether initial velocities or steady-state levels are examined. When the water activity of the medium is reduced by the addition of organic solvents, soluble yeast inorganic pyrophosphatase (no H+ gradient present) synthesizes pyrophosphate in amounts similar to those synthesized by the chromatophores in totally aqueous medium during illumination, (H+ gradient present). The pH, MgCl2 and Pi dependence for the synthesis of pyrophosphate by the chromatophores at steady-state is similar to that observed at equilibrium with the soluble enzyme in the presence of organic solvents. The possibility is raised that a decrease in water activity may play a role in the mechanism by which the energy derived from the electrochemical H+ gradient is used for the synthesis of pyrophosphate in chromatophores of R. rubrum.

Photoaffinity labeling of the tight ADP binding site of the chloroplast coupling factor one (CF1): the effect on the CF1-ATPase activity

Chloroplast thylakoid membranes contain tightly bound ADP which is intimately involved in the mechanism of photophosphorylation. The photoaffinity analog 2-azido-ADP binds tightly to spinach thylakoid membrane-bound coupling factor one (CF1) and, in a manner similar to ADP, inhibits the light-triggered ATPase activity (Czarnecki, J.J., Abbott, M.S. and Selman, B.R. (1983) Eur. J. Biochem. 136, 19-24). Ultraviolet irradiation of thylakoid membranes containing noncovalently, tightly bound 2-azido[beta-32P]ADP results in the inactivation of both the methanol-stimulated MgATPase activity of the membrane-bound CF1 and the octylglucoside-dependent MgATPase activity of the solubilized enzyme. There is a linear correlation between the loss of enzyme activity and the covalent incorporation of the photoaffinity analog. Full inactivation of catalytic activity is estimated to occur upon incorporation of 1.07 mol analog and 0.65 mol analog per mol enzyme for the methanol- and octylglucoside-stimulated activities, respectively. Since 2-azido-ADP modifies only the beta subunit of the CF1 and since there are probably three beta subunits per CF1, these results indicate strong cooperativity among beta subunits and between the site of tightly bound nucleotides and the catalytic sites.

Enzymatic activities in thylakoid membranes, which form medium [32P]NDP and [32P]ATP from 32Pi. Polynucleotide phosphorylase and adenylate kinase

Soluble chloroplast coupling factor 1 (CF1) and the ATP synthase complex, under uncoupled conditions, can form bound ATP from tightly bound ADP and medium Pi. This partial reaction is a powerful probe of the mechanism of ATP synthesis. During our study of the synthesis of bound ATP by CF1 other enzyme activities, which generate [32P]nucleotides from 32Pi, were characterized and controlled. Two enzymes present at significant levels in the preparations are polynucleotide phosphorylase and adenylate kinase. Polynucleotide phosphorylase (PNPase) was found both in thylakoid and CF1 preparations and catalyzed the formation of [beta-32P]ADP via its Pi----ADP exchange activity. The formation of [beta-32P]ADP during net photophosphorylation is attributable to adenylate kinase action on the [32P]ATP formed since hexokinase and glucose effectively block its production. In addition, PNPase also degraded RNA present in thylakoid preparations yielding all four [32P]nucleoside diphosphates. PNPase was also shown to catalyze a Pi----ATP exchange that is dependent on RNA primers and other cofactors.

ATP synthesis catalyzed by the purified erythrocyte Ca-ATPase in the absence of calcium gradients

The Ca2+-transporting ATPase of erythrocytes was isolated by calmodulin affinity chromatography. The backward reaction of the ATPase was investigated. The phosphorylation of the solubilized enzyme by Pi required Mg and was inhibited by Ca and vanadate in the micromolar concentration range. Significant amounts of phosphoenzyme could be obtained only in a medium containing high dimethyl sulfoxide concentrations (greater than 25%) in order to diminish water activity at the phosphorylation site. The phosphoenzyme formed in this way could not phosphorylate ADP. However, upon addition of Ca2+ ions and dilution of dimethyl sulfoxide in the phosphorylated preparation (water activity jump), a highly reactive phosphoenzyme species was obtained which could transfer phosphate in nearly stoichiometric amounts to ADP to form ATP.

Effects of deuterium on the kinetics of beef heart mitochondrial ATPase

A study was done examining the steady-state kinetics of F1-catalyzed ATP and ITP hydrolyses in the presence or absence of D2O as a function of temperature. The steady-state kinetic parameters kcat and kcat/Km were obtained. For ATP hydrolysis, kcat/Km was independent of temperature in the presence or absence of D2O, while kcat/Km for ITP hydrolysis increased in both cases. The relative magnitudes of change of kcat and kcat/Km in the presence and absence of D2O over the temperature range studied were much different for the cases of ATP and ITP hydrolysis. A normal isotope effect was observed in plots of kcat H2O/kcat D2O versus temperature for ATP hydrolysis, which increased then leveled off as temperature increased. An inverse isotope effect at low temperatures changed to a normal isotope effect and increased dramatically as temperature increased during ITP hydrolysis. The results are discussed in terms of the nature and location of the rate-limiting steps in the reaction mechanisms.

Localization of the tight ADP-binding site on the membrane-bound chloroplast coupling factor one

The photoaffinity analog 2-azido-ADP (2-azidoadenosine 5'-diphosphate) was used as a probe of the spinach chloroplast ATP synthase. The analog acted as a substrate for photophosphorylation. Several observations suggested that 2-azido-ADP and ADP bound to the same class of tight nucleotide binding sites: (a) 2-azido-ADP competitively inhibited ADP tight binding (Ki = 1.4 microM); (b) the concentration giving 50% maximum binding, K0.5 for analog tight binding (1 microM) was similar to that observed for ADP (2 microM); (c) nucleotide tight binding required prior membrane energization and was completely reversed by re-energization; (d) the tight binding of 2-azido-[beta-32P]ADP was completely prevented by ADP; (e) the analog inhibited the light-triggered ATPase activity at micromolar concentrations. Ultraviolet irradiation of washed thylakoid membranes containing tightly bound 2-azido-[beta-32P]ADP resulted in the covalent incorporation of the label into the membranes. Denaturing polyacrylamide gel electrophoresis of the labeled membranes demonstrated that the beta subunit of the coupling factor one complex was the only polypeptide in the thylakoid membranes which was labeled. These results identify the beta subunit of the coupling factor as the location of the tightly bound ADP on the thylakoid membranes.