Chemomechanical coupling mechanism of F1-ATPase: Catalysis and torque generation

FEBS Letters 587 (2013) 1030–1035

journal homepage: www.FEBSLetters.org

Review

Chemomechanical coupling mechanism of F1-ATPase: Catalysis and torque generation Rikiya Watanabe, Hiroyuki Noji ⇑ Department of Applied Chemistry, School of Engineering, The University of Tokyo, Bunkyo-ku, Tokyo 113-8656, Japan

a r t i c l e

i n f o

Article history: Received 20 December 2012 Revised 29 January 2013 Accepted 29 January 2013 Available online 8 February 2013 Edited by Wilhelm Just Keywords: F1 ATP hydrolysis Catalysis Torque generation Single-molecule technique Chemomechanical coupling

a b s t r a c t F1-ATPase (F1), a rotary motor protein driven by ATP hydrolysis, is unique with respect to its high efficiency and reversibility in converting chemical energy into mechanical work. Single-molecule studies have improved our understanding about the energy-conversion mechanism of F1 and the chemomechanical-coupling scheme under ATP hydrolysis conditions. A novel single-molecule technique was recently established to estimate the free-energy change of F1 during catalysis at elementary-step resolution, advancing our understanding about the energy-conversion mechanism of ATP hydrolysis and synthesis. The energy conversion mechanism of F1 elucidated from singlemolecule studies provides us with important insights into the operating principles underlying molecular motors. Ó 2013 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

1. Introduction The terminal reaction of oxidative phosphorylation, ATP synthesis is catalyzed by FoF1-ATP synthase (FoF1) [1,2] (Fig. 1). FoF1, which is embedded in the mitochondrial inner membrane, thylakoid membrane of chloroplasts, and the plasma membrane of bacteria, catalyzes ATP synthesis from ADP and inorganic phosphate (Pi) driven by a proton motive force (pmf) across the membranes (Fig. 1a). Thus, FoF1 is the molecular energy-converter that mediates heterologous energy conversion between pmf and the phosphoryl-transfer potential of ATP. The unique feature of the molecular mechanism of FoF1 is that this enzyme employs the mechanical rotation of the inner rotor complex for energy interconversion [3–8]. As the term suggests, FoF1 consists of two parts, F1 and Fo, each of which functions as a molecular motor. F1, the segment of FoF1 that protrudes from the membrane, possesses catalytic reaction centres for ATP hydrolysis/synthesis. When isolated from Fo, F1 shows high ATP hydrolytic activity; therefore, it is Abbreviations: DG, change in free energy; KdATP, dissociation constant of ATP; DG2(h), energy difference between transition state and post-reaction state; DG1(h), energy difference between transition state and pre-reaction state; KEhyd, equilibrium constant of ATP hydrolysis; F1, F1-ATPase; FoF1, FoF1-ATP synthase; Pi, inorganic phosphate; PONbind, probability of the post-ATP binding in the ‘ON’ state; PONhyd, probability of the post-hydrolysis in the ‘ON’ state; pmf, proton motive force; TIRF, total internal reflection fluorescence; TS, transition state of catalysis ⇑ Corresponding author. Fax: +81 3 5841 1872. E-mail address: [email protected] (H. Noji).

termed F1-ATPase. Fueled by ATP hydrolysis, F1 rotates the central rotary shaft against the surrounding stator ring [9]. Fo, the membrane-embedded segment of FoF1, conducts proton-translocation and rotates the cylindrical ring against the stator complex [3,6]. F1 and Fo are connected via central and peripheral stalks, which allow torque transmission between the two motors. Under physiological conditions where pmf is sufficient, Fo generates a larger torque than F1 and reverses the rotary direction of F1, thereby inducing the reverse reaction of ATP hydrolysis, i.e. ATP synthesis [4,5] (Fig. 1a). In contrast, when the pmf diminishes, F1 hydrolyzes ATP and reverses the rotary direction of Fo, thereby inducing Fo to pump protons in order to generate pmf (Fig. 1b). In order to elucidate the working principle of the molecular motors, extensive single-molecule studies on F1 and the complex of FoF1 [2,10–12] have been performed. Due to the stability of the complex and greater ease of handling, F1 has been more intensively investigated compared with Fo [13]. In this review, we focus on the chemomechanical-coupling mechanism of F1. In the first part, we concisely summarize the basic features of rotational dynamics and the proposed reaction scheme of F1 revealed from structural and biochemical studies, as well as single-molecule studies [6,7,9]. In the second part, we introduce more recent single-molecule manipulation measurements. These studies have provided insights on the rotary catalysis mechanisms of F1: the contribution of individual catalytic reaction steps for torque generation, the catalytic power of the conformational subsets that are at a great distance from the stable conformation, and the reaction scheme of ATP synthesis.

0014-5793/$36.00 Ó 2013 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.febslet.2013.01.063

R. Watanabe, H. Noji / FEBS Letters 587 (2013) 1030–1035

1031

Fig. 1. FoF1-ATP synthase. Energy conversion mechanism of FoF1-ATP synthase under ATP synthesis (a) or ATP hydrolysis (b) conditions. Red and green represent the rotor and stator subcomplex, respectively.

2. Part I: general aspects of the F1 motor Bacterial F1 has the simplest subunit composition (a3b3cde). The minimum complex of F1 for function as a rotary motor is the a3b3c subcomplex, in which the a3b3 assembly forms a cylindrical stator ring and the c rotor subunit penetrates the center of the a3b3 ring [9,14]. The first crystal structure of mitochondrial F1 provided many essential insights about the structural basis of the mechanochemical coupling mechanism of F1 [14,15]. There exist three catalytic sites for ATP hydrolysis at the a–b subunit interfaces (primarily on the b subunits), which alternate with three noncatalytic sites at the a–b subunit interfaces [14]. In the crystal structure, the b subunits differ from one another in catalytic state and binding conformation; one binds to the ATP analogue adenylyl–imidodiphosphate (AMP–PNP), the second binds to ADP and azide [16], and the third does not bind to any ligand. These conformations are referred to as bTP, bDP, and bempty, respectively. bTP and bDP assumes a closed conformation, swinging the helical domain (the C-terminal domain) inward towards the nucleotide binding domain, while bempty assumes an open conformation (Fig. 2). The swing motion of the b subunit is thought to generate the rotary torque necessary for the c rotation. 2.1. Single-molecule analysis of the F1 motor To reveal the dynamic features of F1 rotation, several singlemolecule rotational assays [9,12,17,18] have been developed since the method was first developed in 1997 [9]. In most assays, the F1 molecules are immobilized to affix the a3b3 cylinder onto a cover slip, and a rotation marker tag is attached to the protruding part of the c subunit. While a fluorescently labeled actin filament was used as the rotation marker in the pioneering study, micron-sized plastic beads or gold colloids of 40–60 nm are frequently used at present. For single-molecule manipulation experiments, a magnetic bead of 200 nm is attached to the c subunit and controlled by an external magnetic field. In accordance with its pseudothreefold symmetrical structure, F1 hydrolyzes three ATP molecules per rotation. Therefore, the unitary step of rotation is a 120° step [4,19]. The rotary torque of F1 is primarily given as

Fig. 2. Conformational state of the b subunit, depending on the rotary angle. bE, bTP, and bDP represent the b subunit at hc = 40°, 80°, and 200°, respectively. b subunits in open and closed states are colored in red and blue, respectively.

40 pN nm rad1 [19,20], although other values of torque have also been reported [21,22]. The torque values may vary among different

1032

R. Watanabe, H. Noji / FEBS Letters 587 (2013) 1030–1035

species. It has been reported that the free energy released upon ATP hydrolysis is approximately equal to the mechanical work of the rotation against the hydrodynamic friction [19]. Therefore, it is surmised that the energy conversion efficiency of F1 is exceedingly high [23], a hypothesis supported by a recent study [24]. In relation to this observation, F1 displays very high reversibility; F1 catalyzes the synthesis of ATP from ADP and Pi when the rotor is forcibly rotated in the reverse direction [4,5]. 2.2. Reaction scheme of catalysis and rotation The reaction scheme of F1 has been extensively studied by resolving the rotational process into discrete rotational steps. The stepping rotation of F1 was first observed under ATP-limiting conditions, where F1 made distinct pauses at three specific rotary angles that were 120° apart from one another [19]. A mutant F1 and the ATP analogue ATPcS both markedly retarded the ATP hydrolysis step of F1, and were also used to observe clear rotational dwells caused by the hydrolysis-waiting state. The reason for this phenomenon is that the time constant of ATP hydrolysis on the wild-type F1 was only approximately 1 ms long, too fast for analysis, unless a high-speed recording of unloaded rotation was used [17,25]. The angular positions waiting for the ATP binding and hydrolysis differ; the hydrolysis-waiting position is set at +80° from the binding-waiting angle. Therefore, these angles are referred to as binding angle and catalysis angle, respectively. Moreover, subsequent experiments have identified the angular positions of ADP release and Pi release as the binding angle and catalytic angle, respectively [26,27]. Although several uncertainties remain, most sections of the reaction scheme have been revealed. The proposed reaction scheme was summarized in Fig. 3. For each catalytic site on the b subunit, ATP binding occurs when the c subunit is orientated at a specific binding angle. When the c rotates by 200°, bound ATP is cleaved into ADP and Pi. The ADP and Pi generated are released at 240° and 320°, respectively. Thus, after one revolution, the catalytic site completes one turnover of catalysis (ATP ? ADP + Pi). The same reactions occur in the remaining catalytic sites, although the reaction phases differ by 120°.

2.3. Imaging of conformational change of the b subunit While a few catalytic residues are located on the a subunit, the b subunit possesses most of catalytic residues, causing it to be principally responsible for torque generation. The conformational transition between the open and closed states is expected to induce c rotation [14]. To elucidate the correlation of the b conformational state and the c rotation, Masaike et al. developed an elaborate optical system to simultaneously monitor the conformational change of the catalytic b subunit with the angular position of the c subunit [28]. To visualize the conformational state of the b subunit, a fluorescent dye molecule was attached to the C-terminal domain of b and the orientation of the attached dye was determined vis à vis that of the transition dipole moment using total internal reflection fluorescence (TIRF) with polarization modulation. In this study, the b subunit changed its conformation from an open to closed state upon ATP binding (0°–80° in Fig. 3). Subsequently, the b subunit transited the conformationfrom a closed to an open state upon c rotation from 200° (hydrolysis angle) to 320° (Pi-release angle), as proposed based on the crystal structure and proposed reaction scheme. Interestingly, an intermediate conformation between open and closed states was newly discovered at an ADP-releasing angle of 240°; the b subunit assumed a partially closed state at 240°. Although an apparently similar ‘half-closed’ conformation was observed in the crystal structure of mitochondrial F1 with the catalytic sites all occupied with bound nucleotides [29], this would not represent the partially closed state, where the transient conformation of the ADP-bound state after ATP binding is transitioned to b empty, while the partially closed state represents the ADP-bound state prior to ATP binding. The structural details of the partially closed state remain to be refined.

3. Part II: single-molecule manipulation The reversibility of chemomechanical coupling reactions on F1 implies that the rate and equilibrium constants of elementary reaction steps of catalysis are tightly modulated upon c rotation; when the c is reversely rotated, the affinity of the c subunit for ADP or Pi is increased while that for ATP is decreased. The equilibrium constant of the ATP hydrolysis step is also biased towards the direction of synthesis upon the reverse rotation. However, single-molecule analysis does not have the capability to assess the catalytic properties of enzymes in conformational states at a great distance from the stable one. We recently developed a novel single-molecule manipulation protocol in order to elucidate how F1 modulates its catalytic properties upon c rotation [27,30,31]. In the following sections, we introduce the primary concepts and the results of this experiment. We also discuss several implications concerning the energy conversion mechanism of F1 based on the results of this study. 3.1. Stall and release experiment with magnetic tweezers

Fig. 3. Chemomechanical coupling scheme of F1. Elucidated chemomechanical coupling scheme under ATP hydrolysis conditions. The circles and red arrows represent the catalytic state of the b subunits and the angular positions of the c subunit, respectively. While the one featured catalytic site shown in blue undergoes the binding and catalytic events shown, the remaining two catalytic sites undergo the same events simultaneously, but are offset by 120° and 240°.

Owing to the tightly coupled features of catalysis and rotation on F1, we were able to observe the waiting state of an individual reaction step as a distinct rotational pause; the ATP waiting state as the pausing state at ATP-binding angle and the hydrolysis waiting state as the pausing state at the hydrolysis angle. The principle experimental procedures were as follows. First, when F1 performed a rotational pause to wait for the reaction step to be investigated, we stalled the rotor at the target rotational position using magnetic tweezers. In this protocol, a magnetic bead is attached to the c subunit as a probe to visualize its rotational motion as well as to function as a handle for manipulation by the magnetic tweezers. After a specified time period elapsed, the magnetic tweezers were turned off to release F1 in order to observe the response of the motor. F1 displays two principal behaviours:

R. Watanabe, H. Noji / FEBS Letters 587 (2013) 1030–1035

(i) forward rotation to the next pausing angle (denoted as ‘ON’); and (ii) returning back to the previous pausing angle (denoted as ‘OFF’). The ‘ON’ behaviour indicates that the reaction under investigation has been completed when released, whereas the ‘OFF’ behaviour indicates that the reaction has not yet been completed. This behaviour occurs because F1 does not generate rotational torque until all defined reactions occur. In this protocol, the target reaction is the rate-determining step causing the longest rotational pause, while the rotational pause by additional reaction steps is considerably shorter in comparison to the target reaction. We determine the probability of the reaction as the probability of an ‘ON’ event (PON) against all trials. 3.2. Angle dependence of ATP binding and ATP hydrolysis We determined the probabilities of ATP binding and hydrolysis at several stall angles. For the measurement of ATP binding probability (PONbind), the time constant of ATP binding was specifically prolonged by lowering ATP concentration to less than 200 nM. Under this condition, ATP binding takes more than 0.3 s to occur, while other reactions are completed within a 1.5-ms timeframe. For the measurement of ATP-hydrolysis probability (PONhyd), the ATP hydrolysis step was prolonged by using the aforementioned mutant F1 and/or ATPcS, both of which retard the hydrolysis step.

1033

Fig. 4a and b show the probability data of PONbind and PONhyd. Here, we define the angle shown in Fig. 4a and b according to the reaction scheme displayed in Fig. 3; 0° and 200° represent the pausing angle for ATP waiting and for hydrolysis waiting, respectively. Interestingly, even when the c subunit was stalled at an angle that is a great distance from the original pausing angle, ATP binding and hydrolysis still occurs (see the data points of 50° and 150° in Fig. 4a and b), with probabilities that were largely dependent on the stall angle. Additionally, both PONbind and PONhyd achieved plateau levels at rates lower than 100% (Fig. 4a, b). This observation indicates that ATP binding and hydrolysis steps are reversible; during long stalling steps, the system is at equilibrium between ATP binding and release (or ATP hydrolysis and synthesis). Therefore, the plateau levels represent an equilibrium point under this condition. The data points of individual stall angles were fitted using the reversible kinetic model to determine the rate constants of ATP binding and ATP release (or hydrolysis and synthesis). The results of this analysis demonstrated that the rate constants in the forward reactions (i.e. ATP binding, hydrolysis) were exponentially accelerated as F1 rotated in a counterclockwise direction (Fig. 4c), while those in the reverse direction (i.e. ATP release, synthesis) were decelerated or remained constant (Fig. 4d) [31]. As a result, the dissociation constant of ATP (KdATP) and the equilibrium constant of ATP hydrolysis (KEHyd) were exponentially decreased and increased, respectively (Fig. 4e). Fig. 4c–e also displayed rate constants of Pi release/binding that were determined by different experiments [26,27,32]. Therefore, this simple protocol allowed us to determine the kinetic properties of F1, whose conformational state is at a great distance from the stable pausing angle. It was revealed that F1 smoothly modulates the reaction rate constants of each of the elementary steps of catalysis, depending on the rotary angle. This occurs because the rotation is tightly coupled with the bending motion of the b subunit, as well as the closure/opening of the ab interface. 3.3. Reaction energy diagram

Fig. 4. Experimental results for angle dependence of ATP binding and hydrolysis. The time course of the probability of completion of (a) ATP binding/PONbind or (b) hydrolysis/PONhyd for stalling at 50° (red), 0° (black),+50° (cyan) from each of the authentic reaction angles, i.e., 0° for ATP binding and 200° for hydrolysis. Angular dependence of kinetic parameters as determined by single-molecule manipulation; reaction rate constants of forward (c) or reverse (d) reactions and equilibrium constants (e). Solid lines represent the linear regression for ATP binding/release (red), hydrolysis/synthesis (blue), or Pi release/binding (green).

A two-dimensional diagram is conventionally used to describe the reaction coordinate of enzymes in relation to the energy level and the conformational freedom of an enzyme. In such a 2D diagram, an intersection of the two potential wells for the prereaction and post-reaction states represent the transition state (TS; Fig. 5a). Therefore, the position of TS on the axis of the conformational state represents the critical conformational state; the reaction will not occur unless the enzyme changes its conformation over the critical point, in other words, reaction will always occur when the enzyme changes over the critical point. A 2D diagram is inadequate to explain the angle dependence of ATP binding and hydrolysis because reactions can occur at multiple rotational angles. Next, we display a 3D diagram (Fig. 5b), where the x- and y-axis represent the rotational angle of the c subunit and the chemical state, respectively, while the contour lines display free energy. The cross-section along the y-axis at a particular x value represents the 2D energy diagram of the reaction when the c subunit is stalled along the x value. This 3D energy diagram for F1 reaction is essentially consistent with the Sumi–Marcus model that treated intramolecular electron transfer in relation to the solvent or conformational dynamics of the molecule [33], which was recently used to account for the chemomechanical rotary mechanism of F1 [34]. This 3D diagram can account for the angular dependence of ATP binding and hydrolysis when we assume harmonic potentials for the pre-reaction and post-reaction states, as well as a relatively constant level of TS that is represented as the ridge between two potential basins. When the c subunit is stalled at an angle forward from the original pausing position, the activation energy for the forward reaction is reduced while that for the

1034

R. Watanabe, H. Noji / FEBS Letters 587 (2013) 1030–1035

domain of c, streptavidin, and His-tags), and the fact that the actual angle dependence should be steeper [35]. Although we experimentally estimated the stiffness of the c subunit and the entire molecular system from thermally agitated rotary fluctuation [35], the stiffness remains unclear when the c subunit is forcibly rotated far from the stable angle since it is experimentally challenging. 3.5. Reaction scheme for ATP synthesis

Fig. 5. Freeenergy diagram of catalysis on F1. (a) Schematic model of a 2D freeenergy diagram of catalysis of F1. Light green and dark green curves represent the energy potentials of reactant and product. The cross point of the reactant and product potentials represents the transition state of catalysis (TS). The catalysis occurs passing through TS from reactant to product potential (blue arrow). (b) Schematic model of a 3D free-energy diagram of F1 catalysis (left). The free-energy surfaces of pre- or post-catalysis intersect on the line TS, which represents the transition state of catalysis. Points O and O0 represent the energy minima on the reactant and product, respectively. The cross-section at a specific rotary angle is drawn as the right figure. DG represents the free-energy difference of reactant and product. DG1 and DG2 represent the activation energies of forward and reverse reactions, respectively.

reverse reaction is increased. In this diagram, the rotary torque is represented as the slope of the potential basin of the post-reaction state. 3.4. Torque generation during ATP binding and hydrolysis The angle dependence of ATP binding and hydrolysis is not directly related to the torque generated during the reactions. However, when we consider the rate of the reverse reaction as a barometer of the energy difference between TS and the postreaction state (DG2), torque generation upon reaction can be estimated from the angle dependence of the reverse reaction rate. When we assume a relatively constant energy level for the ridge of TS, the deviation of the activation energy (dDG2(h)/dh responds to the slope of the potential of the post-reaction state (equivalent to torque). Therefore, the slopes of the reverse reaction rates (koffATP and ksynATP) in a semi-log plot (Fig. 4d) represent the torque during the ATP binding or hydrolysis step, respectively. It is evident that ATP release displays a markedly steeper slope than the ATP synthesis reaction, suggesting that the ATP hydrolysis step contributes markedly less to torque generation. Torque generation during the ATP binding step is said to be 11 pN nm, while that of ATP hydrolysis is almost 0. However, it should be noted that these values are underestimated, because the c subunit was twisted upon manipulation with magnetic tweezers, primarily due to the elastic elements of the experimental setup (e.g. the outwardly protruding

The observed angle dependence of ATP binding and hydrolysis also has important implications concerning the reaction scheme for the ATP synthesis of F1. The dissociation constant of ATP (KdATP) at the ATP-waiting angle (0° in Fig. 4e) was determined to be approximately 108 M. Since the ATP concentration in vivo is in the millimolar range (103 M), this result suggests that F1 cannot release bound ATP at 0°. According to the angle dependence of KdATP calculated, the c subunit must be rotated over 40° in a clockwise direction, for KdATP to become >1 mM for efficient ATP release. In addition, the equilibrium constant of ATP hydrolysis (KEHyd) at hydrolysis waiting angle (200° in Fig. 4e) was approximately 1, which is too low to explain the high efficiency of ATP synthesis. According to the angle dependence of KEHyd, the c subunit must be rotated over 40° in a clockwise direction for efficient ATP synthesis. These results suggest that the reaction scheme for ATP synthesis is not a simple reverse version of that for ATP-driven rotation (shown in Fig. 3). These findings suggest that all angular positions for reaction steps for ATP synthesis (Pi binding, ADP binding, synthesis, and ATP release) are essentially different from those of Pi release, ADP release, hydrolysis, and ATP binding, and should be changed depending on the conditions. 3.6. Future prospect The introduced single-molecule manipulation study revealed that the mechanism of ATP synthesis is not a simple reverse process of ATP hydrolysis. This result will contribute to furthering our understanding of the mechanism of ATP synthesis, the physiological role of F1. In addition, this study is fundamentally applicable to studies involving other molecular motors or machines. The most promising experiment is the application of this protocol to V1ATPase. Although angle dependence of the ATP binding of V1-ATPase has already been studied, hydrolysis has yet to be investigated [36]. These studies would reveal the generality and uniqueness of our findings in single-molecule studies on F1. Another important finding is the structural basis of the observed angle dependence. For future studies, a structural analysis of the conformational state of the b subunit in the transient state should be undertaken. References [1] Yoshida, M., Muneyuki, E. and Hisabori, T. (2001) ATP synthase–a marvellous rotary engine of the cell. Nat. Rev. Mol. Cell Biol. 2, 669–677. [2] Junge, W., Sielaff, H. and Engelbrecht, S. (2009) Torque generation and elastic power transmission in the rotary FoF1-ATPase. Nature 459, 364–370. [3] Duser, M.G., Zarrabi, N., Cipriano, D.J., Ernst, S., Glick, G.D., Dunn, S.D. and Borsch, M. (2009) 36 Degrees step size of proton-driven c-ring rotation in FoF1ATP synthase. EMBO J. 28, 2689–2696. [4] Rondelez, Y., Tresset, G., Nakashima, T., Kato-Yamada, Y., Fujita, H., Takeuchi, S. and Noji, H. (2005) Highly coupled ATP synthesis by F1-ATPase single molecules. Nature 433, 773–777. [5] Itoh, H., Takahashi, A., Adachi, K., Noji, H., Yasuda, R., Yoshida, M. and Kinosita, K. (2004) Mechanically driven ATP synthesis by F1-ATPase. Nature 427, 465– 468. [6] Diez, M., Zimmermann, B., Borsch, M., Konig, M., Schweinberger, E., Steigmiller, S., Reuter, R., Felekyan, S., Kudryavtsev, V., Seidel, C.A. and Graber, P. (2004) Proton-powered subunit rotation in single membrane-bound FoF1-ATP synthase. Nat. Struct. Mol. Biol. 11, 135–141. [7] Sambongi, Y., Iko, Y., Tanabe, M., Omote, H., Iwamoto-Kihara, A., Ueda, I., Yanagida, T., Wada, Y. and Futai, M. (1999) Mechanical rotation of the c

R. Watanabe, H. Noji / FEBS Letters 587 (2013) 1030–1035

[8] [9] [10] [11]

[12]

[13]

[14]

[15] [16]

[17]

[18]

[19]

[20] [21]

[22]

subunit oligomer in ATP synthase (FoF1): direct observation. Science 286, 1722–1724. Ueno, H., Suzuki, T., Kinosita Jr., K. and Yoshida, M. (2005) ATP-driven stepwise rotation of FoF1-ATP synthase. Proc. Natl. Acad. Sci. USA 102, 1333–1338. Noji, H., Yasuda, R., Yoshida, M. and Kinosita Jr., K. (1997) Direct observation of the rotation of F1-ATPase. Nature 386, 299–302. Okuno, D., Iino, R. and Noji, H. (2011) Rotation and structure of FoF1-ATP synthase. J. Biochem. 149, 655–664. Zimmermann, B., Diez, M., Borsch, M. and Graber, P. (2006) Subunit movements in membrane-integrated E FoF1 during ATP synthesis detected by single-molecule spectroscopy. Biochim. Biophys. Acta 1757, 311–319. Ishmukhametov, R., Hornung, T., Spetzler, D. and Frasch, W.D. (2010) Direct observation of stepped proteolipid ring rotation in E. coli FoF1-ATP synthase. EMBO J. 29, 3911–3923. Zimmermann, B., Diez, M., Borsch, M. and Graber, P. (2006) Subunit movements in membrane-integrated E FoF1 during ATP synthesis detected by single-molecule spectroscopy. Biochim. Biophys. Acta 1757, 311–319. Abrahams, J.P., Leslie, A.G., Lutter, R. and Walker, J.E. (1994) Structure at 2.8 Å resolution of F1-ATPase from bovine heart mitochondria. Nature 370, 621– 628. Stock, D., Gibbons, C., Arechaga, I., Leslie, A.G. and Walker, J.E. (2000) The rotary mechanism of ATP synthase. Curr. Opin. Struct. Biol. 10, 672–679. Bowler, M.W., Montgomery, M.G., Leslie, A.G. and Walker, J.E. (2006) How azide inhibits ATP hydrolysis by the F-ATPases. Proc. Natl. Acad. Sci. USA 103, 8646–8649. Yasuda, R., Noji, H., Yoshida, M., Kinosita Jr., K. and Itoh, H. (2001) Resolution of distinct rotational substeps by submillisecond kinetic analysis of F1-ATPase. Nature 410, 898–904. Adachi, K., Yasuda, R., Noji, H., Itoh, H., Harada, Y., Yoshida, M. and Kinosita, K. (2000) Stepping rotation of F1-ATPase visualized through angle-resolved single-fluorophore imaging. Proc. Natl. Acad. Sci. USA 97, 7243–7247. Yasuda, R., Noji, H., Kinosita Jr., K. and Yoshida, M. (1998) F1-ATPase is a highly efficient molecular motor that rotates with discrete 120 degree steps. Cell 93, 1117–1124. Hayashi, K., Ueno, H., Iino, R. and Noji, H. (2010) Fluctuation theorem applied to F1-ATPase. Phys. Rev. Lett. 104, 218103. Panke, O., Cherepanov, D.A., Gumbiowski, K., Engelbrecht, S. and Junge, W. (2001) Viscoelastic dynamics of actin filaments coupled to rotary F-ATPase: angular torque profile of the enzyme. Biophys. J. 81, 1220–1233. Hornung, T., Ishmukhametov, R., Spetzler, D., Martin, J. and Frasch, W.D. (2008) Determination of torque generation from the power stroke of Escherichia coli F1-ATPase. Biochim. Biophys. Acta 1777, 579–582.

1035

[23] Wang, H.Y. and Oster, G. (2002) The Stokes efficiency for molecular motors and its applications. Europhys. Lett. 57, 134–140. [24] Toyabe, S., Watanabe-Nakayama, T., Okamoto, T., Kudo, S. and Muneyuki, E. (2011) Thermodynamic efficiency and mechanochemical coupling of F1ATPase. Proc. Natl. Acad. Sci. USA 108, 17951–17956. [25] Spetzler, D., Ishmukhametov, R., Hornung, T., Day, L.J., Martin, J. and Frasch, W.D. (2009) Single molecule measurements of F1-ATPase reveal an interdependence between the power stroke and the dwell duration. Biochemistry 48, 7979–7985. [26] Adachi, K., Oiwa, K., Nishizaka, T., Furuike, S., Noji, H., Itoh, H., Yoshida, M. and Kinosita Jr., K. (2007) Coupling of rotation and catalysis in F1-ATPase revealed by single-molecule imaging and manipulation. Cell 130, 309–321. [27] Watanabe, R., Iino, R. and Noji, H. (2010) Phosphate release in F1-ATPase catalytic cycle follows ADP release. Nat. Chem. Biol. 6, 814–820. [28] Masaike, T., Koyama-Horibe, F., Oiwa, K., Yoshida, M. and Nishizaka, T. (2008) Cooperative three-step motions in catalytic subunits of F1-ATPase correlate with 80 degrees and 40 degrees substep rotations. Nat. Struct. Mol. Biol. 15, 1326–1333. [29] Menz, R.I., Walker, J.E. and Leslie, A.G. (2001) Structure of bovine mitochondrial F1-ATPase with nucleotide bound to all three catalytic sites: implications for the mechanism of rotary catalysis. Cell 106, 331–341. [30] Hirono-Hara, Y., Ishizuka, K., Kinosita Jr., K., Yoshida, M. and Noji, H. (2005) Activation of pausing F1 motor by external force. Proc. Natl. Acad. Sci. USA 102, 4288–4293. [31] Watanabe, R., Okuno, D., Sakakihara, S., Shimabukuro, K., Iino, R., Yoshida, M. and Noji, H. (2012) Mechanical modulation of catalytic power on F1-ATPase. Nat. Chem. Biol. 8, 86–92. [32] Adachi, K., Oiwa, K., Yoshida, M., Nishizaka, T. and Kinosita Jr., K. (2012) Controlled rotation of the F1-ATPase reveals differential and continuous binding changes for ATP synthesis. Nat. Commun. 3, 1022. [33] Sumi, H. and Marcus, R.A. (1986) Dynamic effects in electron-transfer reactions. J. Chem. Phys. 84, 4894–4914. [34] Mukherjee, S. and Warshel, A. (2011) Electrostatic origin of the mechanochemical rotary mechanism and the catalytic dwell of F1-ATPase. Proc. Natl. Acad. Sci. USA 108, 20550–20555. [35] Okuno, D., Iino, R. and Noji, H. (2010) Stiffness of gamma subunit of F1-ATPase. Eur. Biophys. J. 39, 1589–1596. [36] Tirtom, N.E., Okuno, D., Nakano, M., Yokoyama, K. and Noji, H. (2013) Mechanical modulation of ATP-binding affinity of V1-ATPase. J. Biol. Chem. 288, 619–623.