Energy-driven subunit rotation at the interface between subunit a and the c oligomer in the F(O) sector of Escherichia coli ATP synthase.

Subunit rotation within the F(1) catalytic sector of the ATP synthase has been well documented, identifying the synthase as the smallest known rotary motor. In the membrane-embedded F(O) sector, it is thought that proton transport occurs at a rotor/stator interface between the oligomeric ring of c subunits (rotor) and the single-copy a subunit (stator). Here we report evidence for an energy-dependent rotation at this interface. F(O)F(1) was expressed with a pair of substituted cysteines positioned to allow an intersubunit disulfide crosslink between subunit a and a c subunit [aN214C/cM65C; Jiang, W. & Fillingame, R. H. (1998) Proc. Natl. Acad. Sci. USA 95, 6607--6612]. Membranes were treated with N,N'-dicyclohexyl-[(14)C]carbodiimide to radiolabel the D61 residue on less than 20% of the c subunits. After oxidation to form an a--c crosslink, the c subunit properly aligned to crosslink to subunit a was found to contain very little (14)C label relative to other members of the c ring. However, exposure to MgATP before oxidation significantly increased the radiolabel in the a-c crosslink, indicating that a different c subunit was now aligned with subunit a. This increase was not induced by exposure to MgADP/P(i). Furthermore, preincubation with MgADP and azide to inhibit F(1) or with high concentrations of N,N'-dicyclohexylcarbodiimide to label most c subunits prevented the ATP effect. These results provide evidence for an energy-dependent rotation of the c ring relative to subunit a.

The rotary binding change mechanism of ATP synthases.

The F(0)F(1) ATP synthase functions as a rotary motor where subunit rotation driven by a current of protons flowing through F(0) drives the binding changes in F(1) that are required for net ATP synthesis. Recent work that has led to the identification of components of the rotor and stator is reviewed. In addition, a model is proposed to describe the transmission of energy from four proton transport steps to the synthesis of one ATP. Finally, some of the requirements for efficient energy coupling by a rotary binding change mechanism are considered.

Rotation of the epsilon subunit during catalysis by Escherichia coli FOF1-ATP synthase.

We report evidence for catalysis-dependent rotation of the single epsilon subunit relative to the three catalytic beta subunits of functionally coupled, membrane-bound FOF1-ATP synthase. Cysteines substituted at beta380 and epsilon108 allowed rapid formation of a specific beta-epsilon disulfide cross-link upon oxidation. Consistent with a need for epsilon to rotate during catalysis, tethering epsilon to one of the beta subunits resulted in the inhibition of both ATP synthesis and hydrolysis. These activities were fully restored upon reduction of the beta-epsilon cross-link. As a more critical test for rotation, a subunit dissociation/reassociation procedure was used to prepare a beta-epsilon cross-linked hybrid F1 having epitope-tagged betaD380C subunits (betaflag) exclusively in the two noncross-linked positions. This allowed the beta subunit originally aligned with epsilon to form the cross-link to be distinguished from the other two betas. The cross-linked hybrid was reconstituted with FO in F1-depleted membranes. After reduction of the beta-epsilon cross-link and a brief period of catalytic turnover, reoxidation resulted in a significant amount of betaflag in the beta-epsilon cross-linked product. In contrast, exposure to ligands that bind to the catalytic site but do not allow catalysis resulted in the subsequent cross-linking of epsilon to the original untagged beta. Furthermore, catalysis-dependent rotation of epsilon was prevented by prior treatment of membranes with N,N'-dicyclohexylcarbodiimide to block proton translocation through FO. From these results, we conclude that epsilon is part of the rotor that couples proton transport to ATP synthesis.

Subunit rotation in Escherichia coli FoF1-ATP synthase during oxidative phosphorylation.

We report evidence for proton-driven subunit rotation in membrane-bound FoF1-ATP synthase during oxidative phosphorylation. A betaD380C/gammaC87 crosslinked hybrid F1 having epitope-tagged betaD380C subunits (betaflag) exclusively in the two noncrosslinked positions was bound to Fo in F1-depleted membranes. After reduction of the beta-gamma crosslink, a brief exposure to conditions for ATP synthesis followed by reoxidation resulted in a significant amount of betaflag appearing in the beta-gamma crosslinked product. Such a reorientation of gammaC87 relative to the three beta subunits can only occur through subunit rotation. Rotation was inhibited when proton transport through Fo was blocked or when ADP and Pi were omitted. These results establish FoF1 as the second example in nature where proton transport is coupled to subunit rotation.

Identification of amino acid residues at nucleotide-binding sites of chaperonin GroEL/GroES and cpn10 by photoaffinity labeling with 2-azido-adenosine 5'-triphosphate.

Although the chaperonin GroEL/GroES complex binds and hydrolyzes ATP, its structure is unlike other known ATPases. In order to better characterize its nucleotide binding sites, we have photolabeled the complex with the affinity analog 2-azido-ATP. Three residues of GroEL, Pro137, Cys138 and Thr468, are labeled by the probe. The location of these residues in the GroEL crystal structure [Braig, K., Otwinowski, Z., Hedge, R., Boisvert, D., Joachimiak, A., Horwich, A. & Sigler, P. (1994) Nature 371, 578-586: Boisvert, D. C., Wang, J., Otwinowski, Z., Horwich, A. L. & Sigler, P. B. (1996) Nat. Struct. Biol. 3, 170-177] suggests that 2-azido-ATP binds to an alternative conformer of GroEL in the presence of GroES. The labeled site appears to be located at the GroEL/GroEL subunit interface since modification of Pro137 and Cys138 is most readily explained by attack of a probe molecule bound to the adjacent GroEL subunit. Labeling of the co-chaperonin, GroES, is clearly demonstrated on gels and the covalent tethering of nucleotide allows detection of a GroES dimer in the presence of SDS. However, no stable peptide derivative of GroES could be purified for sequencing. In contrast, the GroES homolog, yeast cpn10, does give a stable derivative. The modified amino acid is identified as the conserved Pro13, which corresponds to Pro5 in Escherichia coli GroES.

Nucleotide-depleted beef heart F1-ATPase exhibits strong positive catalytic cooperativity.

Catalytic cooperativity is a central feature of the binding change mechanism for F0F1-ATP synthases. However, in a recent publication (Reynafarje, B. D., and Pedersen, P. L. (1996) J. Biol. Chem. 271, 32546-32550), Reynafarje and Pedersen claim that cooperative effects are an artifact caused by endogenous nucleotides and that when such nucleotides are removed, the multiple catalytic sites on MF1 behave independently during ATP hydrolysis. In contrast to this conclusion, we show here that when ATP is loaded at a single catalytic site on nucleotide-depleted MF1, the rate of product release is accelerated by up to 5 x 10(4)-fold by the binding of ATP at adjacent catalytic sites. Hence, nucleotide-depleted MF1 is not an exception but does in fact show strong cooperative interactions. In addition, evidence is presented supporting a random order for product release during ATP hydrolysis.

Subunit rotation in F0F1-ATP synthases as a means of coupling proton transport through F0 to the binding changes in F1.

The rotation of an asymmetric core of subunits in F0F1-ATP synthases has been proposed as a means of coupling the exergonic transport of protons through F0 to the endergonic conformational changes in F1 required for substrate binding and produce release. Here we review earlier evidence both for and against subunit rotation and then discuss our most recent studies using reversible intersubunit disulfide cross-links to test for rotation. We conclude that the gamma subunit of F1 rotates relative to the surrounding catalytic subunits during catalytic turnover by both soluble F1 and membrane-bound F0F1. Furthermore, the inhibition of this rotation by the modification of F0 with DCCD suggests that rotation in F1 is obligatorily coupled to rotation in F0 as an integral part of the coupling mechanism.

ATP hydrolysis by membrane-bound Escherichia coli F0F1 causes rotation of the gamma subunit relative to the beta subunits.

We recently demonstrated that the gamma subunit in soluble F1-ATPase from Escherichia coli rotates relative to surrounding beta subunits during catalytic turnover (Duncan et al. (1995) Proc. Natl. Acad. Sci. USA 92, 10964-10968). Here, we extend our studies to the more physiologically relevant membrane-bound F0F1 complex. It is shown that beta D380C-F1, containing a beta-gamma intersubunit disulfide bond, can bind to F1-depleted membranes and can restore coupled membrane activities upon reduction of the disulfide. Using a dissociation/reconstitution approach with crosslinked beta D380C-F1, beta subunits containing an N-terminal Flag epitope (beta flag) were incorporated into the two non-crosslinked beta positions and the hybrid F1 was reconstituted with membrane-bound F0. Following reduction and ATP hydrolysis, reoxidation resulted in a significant amount of crosslinking of beta flag to the gamma subunit. This demonstrates that gamma rotates within F1 during catalytic turnover by membrane-bound F0-F1. Furthermore, the rotation of gamma is functionally coupled to F0, since preincubation with DCCD to modify F0 blocked rotation.

Rotation of subunits during catalysis by Escherichia coli F1-ATPase.

During oxidative and photo-phosphorylation, F0F1-ATP synthases couple the movement of protons down an electrochemical gradient to the synthesis of ATP. One proposed mechanistic feature that has remained speculative is that this coupling process requires the rotation of subunits within F0F1. Guided by a recent, high-resolution structure for bovine F1 [Abrahams, J. P., Leslie, A. G., Lutter, R. & Walker, J. E. (1994) Nature (London) 370, 621-628], we have developed a critical test for rotation of the central gamma subunit relative to the three catalytic beta subunits in soluble F1 from Escherichia coli. In the bovine F1 structure, a specific point of contact between the gamma subunit and one of the three catalytic beta subunits includes positioning of the homolog of E. coli gamma-subunit C87 (gamma C87) close to the beta-subunit 380DELSEED386 sequence. A beta D380C mutation allowed us to induce formation of a specific disulfide bond between beta and gamma C87 in soluble E. coli F1. Formation of the crosslink inactivated beta D380C-F1, and reduction restored full activity. Using a dissociation/reassembly approach with crosslinked beta D380C-F1, we incorporated radiolabeled beta subunits into the two noncrosslinked beta-subunit positions of F1. After reduction of the initial nonradioactive beta-gamma crosslink, only exposure to conditions for catalytic turnover results in similar reactivities of unlabeled and radiolabeled beta subunits with gamma C87 upon reoxidation. The results demonstrate that gamma subunit rotates relative to the beta subunits during catalysis.

Nucleotide-binding sites on Escherichia coli F1-ATPase. Specificity of noncatalytic sites and inhibition at catalytic sites by MgADP.

Nucleotide-depleted EcF1 binds a maximum of two GTP, ATP, or ADP at noncatalytic sites, whereas all three sites can only be filled by a combination of nucleoside di- and triphosphates. MgPPi prevents binding of GTP and significantly slows ATP binding, suggesting that non-catalytic sites also bind PPi. No binding of GDP at non-catalytic sites could be detected. The slow rate of GTP dissociation from noncatalytic sites (t1/2 = 175 min) is increased 2-8-fold by EDTA, MgPPi, MgADP, or EDTA/ATP, but 23-fold by conditions for ATP hydrolysis. ATP hydrolysis by EcF1, depleted of both its inhibitory epsilon-subunit and endogenous nucleotides, shows a burst of activity. However, it shows a lag if preincubated with MgADP but not when preincubated with Mg2+ alone. For epsilon-depleted EcF1 containing endogenous inhibitory ADP, preincubation with an ATP-regenerating system results in a burst of activity, whereas the control shows a lag. This same enzyme form shows significant inhibition with increasing concentrations of Mg2+ during ATP hydrolysis but lesser levels of inhibition when other NTP substrates are used. With the five-subunit enzyme, increasing amounts of azide cause an increase in the level of inhibition with a corresponding increase in amount of bound nucleotide resistant to rapid chase. Azide-trappable nucleotide is bound at catalytic sites as shown by covalent incorporation of 2-azido-ADP. The results suggest that ligand specificity may not be a reliable means of distinguishing between catalytic and noncatalytic sites and that MgADP inhibition should be taken into account in the kinetic analysis of EcF1 mutants.

Nucleotide binding sites on beef heart mitochondrial F1-ATPase. Cooperative interactions between sites and specificity of noncatalytic sites.

We have studied the properties of beef heart mitochondrial F1 having inhibitory MgADP bound at one of the three catalytic sites and various levels of occupancy of the three noncatalytic nucleotide sites including zero, two, or three ADP/ATPs or two ADP/ATP plus one GTP. The properties examined include the rate of MgATP-dependent reactivation and the rate of increase in the fraction of F1 containing transiently bound intermediates. For each form of the enzyme tested, the rate of reactivation closely paralleled the rate of increase in the level of bound intermediates, indicating that when one catalytic site on F1 is blocked by inhibitory MgADP, the remaining two sites are incapable of residual uni- or bi-site activity. It was also found that the stability of the MgADP-inhibited complex decreases with full occupancy of the noncatalytic sites. This demonstrates that the noncatalytic sites modulate the properties of catalytic sites. Finally, it was found that the noncatalytic sites on mitochondrial F1 do not, as has long been believed, bind adenine nucleotides exclusively. Evidence is presented that both GTP and PPi bind tightly at noncatalytic sites.

A model for the catalytic site of F1-ATPase based on analogies to nucleotide-binding domains of known structure.

An updated topological model is constructed for the catalytic nucleotide-binding site of the F1-ATPase. The model is based on analogies to the known structures of the MgATP site on adenylate kinase and the guanine nucleotide sites on elongation factor Tu (Ef-Tu) and the ras p21 protein. Recent studies of these known nucleotide-binding domains have revealed several common functional features and similar alignment of nucleotide in their binding folds, and these are used as a framework for evaluating results of affinity labeling and mutagenesis studies of the beta subunit of F1. Several potentially important residues on beta are noted that have not yet been studied by mutagenesis or affinity labeling.

Nucleotide binding sites on mitochondrial F1-ATPase. Electron spin resonance spectroscopy and photolabeling by azido-spin-labeled adenine nucleotides support an adenylate kinase-like orientation.

A spin-labeled photoaffinity ATP analog, 2-N3-2',3'-SL-ATP (2-N3-SL-ATP) was specifically loaded at catalytic (exchangeable) or noncatalytic (nonexchangeable) nucleotide-binding sites on nucleotide-depleted mitochondrial F1-ATPase. Photolysis of the enzyme complexes resulted in the specific modification of beta-Tyr-345 when the catalytic sites were occupied and beta-Tyr-368 when noncatalytic sites were filled. These are the same amino acid assignments that were made previously using 2-N3ATP. The results demonstrate that the attachment of a spin label moiety to the ribose ring does not prevent proper binding of the analog at both types of nucleotide sites on F1-ATPase and suggest that the probe can be used for investigations of the nucleotide-binding sites using ESR spectroscopy. Enzyme that is in complex with the 2-N3-SL-ATP exhibits an ESR spectrum that is typical for highly immobilized nitroxyl radicals both in the dark or after photolysis. Additional peaks in the high- and low-field regions arise due to dipolar spin interactions most likely involving a pair of catalytic and noncatalytic sites. The two sites are calculated to be approximately 15 A apart. This distance, obtained through ESR spectroscopy, combined with the finding that the 2 labeled amino acids are only 23 residues apart from each other, further supports an adenylate kinase-like arrangement of nucleotide binding sites on F1-ATPase where catalytic and noncatalytic sites are in close proximity (Vogel, P. D., and Cross, R. L. (1991) J. Biol. Chem. 266, 6101-6105).

Adenine nucleotide-binding sites on mitochondrial F1-ATPase: studies of the inactive complex formed upon binding ADP at a catalytic site.

ADP-induced inhibition of mitochondrial F1-ATPase has been studied. It is shown that in the presence of magnesium and the absence of light, the photoaffinity ADP analog, 2-azido-ADP, induces a reversible inhibition of native F1 that is indistinguishable from that obtained with ADP. Photolysis of the inactive complex results in the predominant labeling of a catalytic-site peptide identified previously (Cross et al., 1987, Proc. Natl. Acad. Sci. USA 84, 5715-5719). Dissociation of the inactive complex formed between F1 and ADP is biphasic with a rapid azide-insensitive phase followed by a slow azide-sensitive phase (k approximately 3 x 10(-3) s-1). It is also shown that incubation of the ADP-inhibited enzyme with EDTA or phosphate does not result in release or migration of ADP from the catalytic site. However, it does convert the complex to a form that reactivates in the presence of 100 microM ATP at a rate too rapid to observe using manual mixing.

The adenine nucleotide-binding site on yeast 3-phosphoglycerate kinase. Affinity labeling of Lys-131 by pyridoxal 5'-diphospho-5'-adenosine.

The adenine nucleotide analog [3H]pyridoxal 5'-diphospho-5'-adenosine (PLP-AMP) is a potent and highly specific inactivator of yeast 3-phosphoglycerate kinase. Supportive evidence includes the finding that 1) during a 10-min incubation, half-maximal inactivation is given by 10 microM PLP-AMP, 2) covalent incorporation of 1.2 mol of PLP-AMP/mol of enzyme is sufficient to give complete inactivation, and 3) MgATP gives near complete protection against modification and inactivation by PLP-AMP. Following reaction with PLP-AMP and reduction with NaBH4 to form a stable adduct, the enzyme was digested with endoproteinase Lys-C and peptides were separated by reversed-phase high-performance liquid chromatography. The single major labeled peptide was purified and sequenced, and the modified residue was identified as Lys-131. The crystal structure of enzyme in the open conformation shows Lys-131 to reside within a loop of flexible random coil positioned at the outer edge of the central binding cleft, approximately 2 nm from the surface of the cleft that comprises part of the MgATP-binding site (Watson, H. C., Walker, N. P. C., Shaw, P. J., Bryant, T. N., Wendell, P. L., Fothergill, L. A., Perkins, R. E., Conroy, S. C., Dobson, M. J., Tuite, M. F., Kingsman, A. J., and Kingsman, S. M. (1982) EMBO J. 1, 1635-1640). We conclude that the structural element containing Lys-131 undergoes substantial movement during the ligand-induced conformational change known to occur during formation of the ternary complex, resulting in the positioning of a basic residue near a negatively charged substrate. Since similar affinity-labeling results have been presented for hexokinase (Tamura, J. K., LaDine, J. R., and Cross, R. L. (1988) J. Biol. Chem. 263, 7907-7912), we further suggest that movement of positive charge into the central cleft may be a common step in the tight binding of nucleotides by bilobal kinases.

Adenine nucleotide-binding sites on mitochondrial F1-ATPase. Evidence for an adenylate kinase-like orientation of catalytic and noncatalytic sites.

Nucleotide-depleted mitochondrial F1-ATPase (F1[0,0]) is inhibited by the diadenosine oligophosphate compounds, AP4A, AP5A, and AP6A (where APxA stands for 5',5'-diadenosine oligophosphates having a chain of x phosphoryl groups linking the two adenosine moieties). When F1[0,0] is preincubated with these compounds and then assayed for ATP hydrolysis activity under conditions that normally allow turnover at all three catalytic sites, the maximal level of inhibition observed is 80%. However, when assayed at lower ATP concentrations under conditions that allow simultaneous turnover at only two of the three sites, no inhibition is observed. A decrease in the number of phosphoryl groups that links the adenosine moieties to less than 4 (AP3A, AP2A) converts the compound to an activator of ATP hydrolysis, similar in effect to that obtained when one mol of ADP or 2-azido-ADP binds at a catalytic site on F1[0,0]. Inhibition by the compounds requires the presence of at least one vacant noncatalytic site. Evidence is provided that the probes also interact with a catalytic site. The stoichiometry for maximal inhibition by AP4A is 0.94 mol/mol of F1. The data presented support a model for the structure of nucleotide-binding sites on F1 that places catalytic and noncatalytic sites in close proximity in an orientation analogous to the ATP and AMP binding sites on adenylate kinase. Inhibition of the enzyme by the dinucleotide compounds can be explained by the cross-bridging of one of the catalytic sites to a noncatalytic site in analogy to the inhibition of adenylate kinase by AP5A. The residual capacity for bi-site catalysis indicates that the second and third catalytic sites remain catalytically active.