ATP Hydrolysis and Synthesis of a Rotary Motor V-ATPase from Thermus thermophilus*

Vacuolar-type H+-ATPase (V-ATPase) catalyzes ATP synthesis and hydrolysis coupled with proton translocation across membranes via a rotary motor mechanism. Here we report biochemical and biophysical catalytic properties of V-ATPase from Thermus thermophilus. ATP hydrolysis of V-ATPase was severely inhibited by entrapment of Mg-ADP in the catalytic site. In contrast, the enzyme was very active for ATP synthesis (∼70 s–1) with the Km values for ADP and phosphate being 4.7 ± 0.5 and 460 ± 30 μm, respectively. Single molecule observation showed V-ATPase rotated in a 120° stepwise manner, and analysis of dwelling time allowed the binding rate constant kon for ATP to be estimated (∼1.1 × 106 m–1 s–1), which was much lower than the kon (= Vmax/Km) for ADP (∼1.4 × 107 m–1 s–1). The slower \batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \(k_{\mathrm{on}}^{\mathrm{ATP}}\) \end{document} than \batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \(k_{\mathrm{on}}^{\mathrm{ADP}}\) \end{document} and strong Mg-ADP inhibition may contribute to prevent wasteful consumption of ATP under in vivo conditions when the proton motive force collapses.

Vacuolar-type H ϩ -ATPases (V-ATPases) 2 are found in a wide range of organisms. V-ATPase in eukaryotes functions as an ATP hydrolysis-driven proton pump that carries out acidification of cellular compartments, such as lysosomes, and extracellular fluid in the case of renal acidification, bone resorption, and tumor metastasis (1). A family of V-ATPases is also found in archaea and some eubacteria (the prokaryotic V-ATPase family) (2-7). 3 A major role of the V-ATPase in prokaryotes is to produce ATP, a function performed by F 0 F 1 in eukaryotes and most eubacteria. V-ATPase and F 0 F 1 function by a rotary ATP synthase/ATPase mechanism (1). The hydrophilic domain of both V-ATPase and F 0 F 1 (called V 1 and F 1 , respectively) is responsible for ATP synthesis/hydrolysis and is connected via the central rotor and peripheral stator stalks to the transmembrane domain (V 0 and F 0 , respectively), which functions as an ion channel (1,8). Although composition and arrangement of subunits differ considerably between V-ATPase and F 0 F 1 , they seem to share a common rotary catalysis mechanism, catalyzing the interconversion of the energy from proton translocation across membranes and the energy of ATP synthesis/hydrolysis through rotation of the central rotor subunits (8). It is thought that rotary catalysis is basically reversible (8,9). When the transmembrane electrochemical gradient of protons (proton motive force (pmf)) is of sufficient strength, pmf drives rotation of a central rotor shaft to synthesize ATP. In contrast, when pmf is weak, the enzymes become an ATPdriven proton pump that rotates in the opposite direction driven by the energy released by ATP hydrolysis. Indeed, it has been shown that yeast V-ATPase, which functions as a proton pump in vivo, is able to catalyze ATP synthesis when exposed to an electrochemical gradient in vitro (10). In addition the F 1 portion of F 0 F 1 can synthesize ATP when the rotor shaft is forced to rotate in a direction opposite that of ATP hydrolysis (11,12). It is well known that the ATP hydrolysis reaction catalyzed by both V-ATPase and F 0 F 1 is highly regulated by a number of different mechanisms to prevent wasteful ATP consumption (1,13). One such mechanism is Mg-ADP inhibition, whereby Mg-ADP binds into the catalytic site of the V 1 and F 1 domains and thus inhibits ATP hydrolysis (14 -17). Some enzymes appear to be inhibited irreversibly by these mechanisms (18,19).
V-ATPase from the thermophilic bacterium, Thermus thermophilus, has been extensively investigated by biochemical and biophysical methods. Subunit rotation coupled to ATP hydrolysis of the V 1 portion has been visualized using a single mole-cule detection technique (20 -22), and recent work has succeeded in resolving each step in the ATP hydrolysis reaction of the V 1 domain (23). In these previous studies, a mutant enzyme (the S232A/T235S double substitution in subunit A, the TSSA mutation) was used to investigate the suppression of Mg-ADP inhibition. However, rotation analysis of the wild type enzyme has yet to be described. Moreover, the effects of the V 0 domain on rotary catalysis by the V 1 domain, including effects on stepwise rotation, dwell time, torque, etc., remain unclear. Previous work has shown that the V-ATPase is highly active as an ATP synthase with an H ϩ /ATP ratio of 4.0 Ϯ 0.1 (24). However, the kinetic parameters (K m and V max ) of the ATP synthesis reaction by the V-ATPase have not been reported yet. In addition, the effect of the TSSA mutation on ATP synthesis has not been investigated. It is clear that a detailed quantitative analysis of ATP synthesis/hydrolysis by the holo-enzyme (i.e. V-ATPase) is required for a deeper understanding of this remarkable rotary catalysis mechanism.
In this study, we visualized rotational steps, which correspond to each catalytic reaction, of wild type V-ATPase from a thermophilic eubacterium, T. thermophilus, at a single molecule level. We also quantitatively analyzed ATP synthesis using enzyme reconstituted into liposomes. It was possible to determine several kinetic parameters of V-ATPase for both ATP synthesis and hydrolysis reactions. These parameters explain why T. thermophilus V-ATPase has a higher preference for ATP synthesis than ATP hydrolysis.

EXPERIMENTAL PROCEDURES
Preparation of Avi Tag or His Tag Introduced V 1 -The Avi Tag sequence (GLNDIFEAQKIEWHE) was introduced at the N terminus of the A subunit by PCR-based mutagenesis. The Avi-tagged wild type V 1 (A (AviTag)3 B 3 DF) and the Avi-tagged mutant V 1 (A (AviTag/C28S/S232A/T235S/C255A/C508S)3 B (C264S)3 DF) were expressed in Escherichia coli as described previously (23,25). The E. coli cells were suspended in 100 mM sodium phosphate (pH 8.0), 1 mM EDTA (buffer A) and disrupted by sonication, followed by heat treatment at 65°C for 30 min. After removal of denatured E. coli proteins by centrifugation at 19,000 ϫ g for 60 min, ammonium sulfate was added to the supernatant to a final concentration of 0.5 M. The solution was applied to a Butyl-650 M Toyopearl column (TOSOH) equilibrated with buffer A supplemented with 0.5 M ammonium sulfate. After washing the column with the same buffer, the protein was eluted with buffer A. The buffer was exchanged to 20 mM Tris-HCl (pH 8.0) containing 1 mM EDTA by an Amicon Ultra filtration unit (Millipore), and the solution was applied to a UNO-Q column (Bio-Rad). Proteins were eluted with a linear gradient of NaCl, from 0 to 400 mM. The eluted protein was applied to a Superdex HR200 column equilibrated with 20 mM MOPS-NaOH (pH 7.0), containing 150 mM NaCl, concentrated with an Amicon Ultra filter unit. Purified sample was stored at 4°C until use. The His-tagged wild type V 1 (A (His-10/C28S/C508S)3 B (C264S)3 D (E48C/Q55C) F) and the His-tagged mutant V 1 (A (His-10/C28S/S232A/T235S/C255A/C508S)3 -B (C264S)3 D (E48C/Q55C) F) were expressed in E. coli and purified as described previously (23,25).
Preparation of V-ATPase-A His 3 tag was introduced to the C terminus of the V 0 -c subunits using an integration vector system (26,27). The recombinant T. thermophilus strain was grown at 70°C for ϳ24 h under strong aeration in a medium containing 8 g of polypeptone, 4 g of yeast extract, and 2 g of NaCl/l. The harvested cells (30 g) were suspended in 200 ml of 50 mM Tris-HCl (pH 8.0) and disrupted by sonication. The membranes were precipitated by centrifugation at 96,000 ϫ g for 20 min and washed twice with the same buffer. The washed membranes were suspended in 20 mM sodium phosphate (pH 8.0), 10 mM imidazole-HCl, 100 mM NaCl, and 10% (w/v)Triton X-100, followed by sonication. The debris was removed by centrifugation at 96,000 ϫ g for 30 min, and the supernatant was applied to a nickel-nitrilotriacetic acid (Ni 2ϩ -NTA) Superflow column (Qiagen), which was then washed thoroughly and eluted with 20 mM sodium phosphate (pH 8.0), 200 mM imidazole-HCl, 100 mM NaCl, and 0.05% (w/v) octaethylene glycol monododecyl ether. The fractions containing proteins were collected and dialyzed against 20 mM Tris-HCl (pH 8.0), 1 mM EDTA, and 0.05% (w/v) octaethylene glycol monododecyl ether for 10 h at 4°C. The dialyzed solution was applied to a Resource Q column (Amersham Biosciences). Proteins were eluted with a linear NaCl gradient, from 0 to 500 mM. The V 0 and V-ATPase were eluted in different fractions and applied to a Superdex HR200 column and eluted with 20 mM MOPS-NaOH (pH 7.0), 150 mM NaCl, and 0.05% (w/v) n-dodecyl ␤-D-maltoside (DDM) (buffer B). To obtain the reconstituted V-ATPase, the purified V 1 and V 0 were mixed (V 1 /V 0 molar ratio was Ͼ3) for 1 h at room temperature, and unreacted V 1 was removed by a Superdex HR200 column equilibrated with buffer B. The reconstituted V-ATPase was concentrated with an Amicon Ultra filtration unit and stored at 4°C until use.
Protein Concentration, Bound Nucleotide, and ATP Hydrolysis Assays-The protein concentration of V 1 was determined from UV absorbance calibrated by quantitative amino acid analysis; 1 M gives 0.36 A at 280 nm. The protein concentrations of V 0 and V-ATPase were determined by the BCA assay kit (Pierce) using known concentrations of V 1 as standards.
Nucleotides tightly bound to wild type V 1 were removed as described previously (25). The enzyme solution was dialyzed against 100 mM sodium phosphate (pH 8.0) and 10 mM EDTA (buffer C) and heated at 65°C for 10 min, followed by cooling on ice for 30 min; this process was repeated five times. The enzyme solution was then applied to a PD-10 column (Amersham Biosciences) equilibrated with buffer C. After repeating this application procedure several times, the enzyme solution was applied to a Superdex 200HR column equilibrated with 20 mM MOPS-NaOH (pH 7.0) and 150 mM NaCl and concentrated with an Amicon Ultra filter unit. The enzyme was stored at 4°C before use. The amount of nucleotide bound to the enzymes was measured using an ODS-80TS column (TOSOH) eluted with 100 mM sodium phosphate (pH 6.9), containing 4 mM EDTA, monitoring absorbance at 260 nm.
The ATP hydrolysis activities of the Avi-tagged reconstituted V-ATPase and the Avi-tagged V 1 were measured as follows; the reactions were initiated by the addition of V 1 or V-ATPase solutions into 1.6 ml of assay mixture consisting of 50 mM Tris-HCl (pH 8.0), 100 mM KCl, 2 mM MgCl 2 , 0.05% (w/v) DDM, 1 mM phosphoenolpyruvate, 0.2 mM NADH, 50 g/ml pyruvate kinase, 50 g/ml lactate dehydrogenase, and a range of concentrations of Mg-ATP. The rate of ATP hydrolysis was monitored continuously as the rate of oxidation of the NADH, determined by the absorbance decrease at 340 nm. N, NЈ-Dicyclohexylcarbodiimide (DCCD) sensitivity of ATP hydrolysis activity was measured after a 1-h preincubation of 100 M DCCD with V 1 or V-ATPase. These experiments were carried out at 25°C.
Single Molecular Analysis for ATP Hydrolysis-Streptavidincoated beads and nickel-nitrilotriacetic acid (Ni 2ϩ -NTA)coated coverslips were prepared as described before (28,29). A flow cell (5-10 l) was made of two coverslips (bottom, 24 ϫ 36 mm 2 , and top, 24 ϫ 24 mm 2 ) separated by two spacers of 50-m thickness. The glass surface of the bottom coverslip was coated with Ni 2ϩ -NTA. Buffer D (50 mM Tris-HCl (pH 8.0), 100 mM KCl, 0.05% (w/v) DDM) containing 1 mg/ml bovine serum albumin was first applied to the flow cell and incubated for 5 min to block nonspecific binding of the enzyme. The biotinylated V 1 or V-ATPase (1-10 nM) in buffer D containing 1 mg/ml bovine serum albumin was then applied to the flow cell and incubated for 5 min. Unbound V 1 or V-ATPase was washed out with 20 l of buffer D and 40 l of 20 mM potassium phosphate (pH 8.0) containing 0.05% (w/v) DDM (buffer E). Then 0.1% (w/v) streptavidin-coated beads in buffer E were applied to the flow cell and incubated for 10 min. Unbound beads were removed by washing with 40 l of buffer E and 40 l of buffer D. Finally, observation of rotation was initiated after infusion of 80 l of buffer D supplemented with a range of concentrations of Mg-ATP, 2 mM MgCl 2 , 1 mM phosphoenolpyruvate, and 50 g/ml pyruvate kinase. Rotation of the bead was recorded with a charge-coupled device camera (300-RCX, Dage-MTI, Michigan City, IN) at 30 frames/s (fps) using a phase-constant microscope (IX70, Olympus) with ϫ100 objective lens (N.A., 1.30, Olympus). For rapid recording, we acquired images of the rotating bead with a dark field microscope (IX70, Olympus) equipped with a mercury lamp and with a complementary metal oxide semiconductor camera (Hi-Dcam, NAC Image Technology, Tokyo) at 1,000 fps. Custom software (created by Ryohei Yasuda, Kengo Adachi, and Library) was used for analyses of the bead movements and dwelling times of steps. Timeaveraged rotation speed was calculated over five consecutive revolutions. All experiments were carried out at 23-25°C. For single molecular experiments, the Avi-tagged reconstituted V-ATPase and the His-tagged V 1 were used.
ATP Synthesis Assay-Phosphatidylcholine (type II-S, Sigma) was suspended to a final concentration of 40 mg/ml in 10 mM HEPES, 10 mM MgSO 4 , 1 mM KCl adjusted to pH 7.5 with NaOH (buffer F). The suspension was sonicated in a bath sonicator. For reconstitution, the liposomes were diluted to a final concentration of 32 mg/ml in buffer F and mixed with purified V-ATPase (final concentration of 50 -150 g/ml). The proteoliposomes were generated using a freeze-thaw sonication method (24) and stored at room temperature for 1 day, which did not affect ATP synthesis activity.
The amount of ATP catalyzed by V-ATPase was monitored continuously with luminescence by luciferin/luciferase (ATP bioluminescence assay kit CLS-II; Roche Applied Science) in a spectrofluorometer (FP-6500, Jasco), monitoring at 550 nm in the absence of exciting light as described previously (24). First, 30 l of proteoliposomes were mixed with 15 l of acidic medium (300 mM maleic acid/NaOH (pH 4.9), 20 mM MgSO 4 , 1 mM KCl) and incubated for 3 min. The ATP synthesis reaction was initiated by addition of 30 l of the acidified suspension into 970 l of basic medium (100 mM Tricine-NaOH (pH 8.5), 2.5 mM MgSO 4 , 100 mM KCl, a range of concentrations of sodium phosphate and ADP, 40 ng/ml valinomycin, 2.16 mg/ml luciferin/luciferase). The ATP synthesis rate was determined from the slope of the luminescence intensity curve for the period of 0 -5 s after the start of the reaction. The luminescence intensity was calibrated by addition of 200 pmol of ATP. All experiments were carried out at 25°C. For the ATP synthesis assay, the V-ATPase purified from T. thermophilus membranes (wild type V-ATPase) and the Avi-tagged reconstituted mutant V-ATPase (the TSSA mutant V-ATPase) were used.

ATP Hydrolysis by Wild Type Enzymes-In this study, the
Avi-tagged WT-V 1 isolated from E. coli cell cultures was used for a single molecular experiment. To assess the effects of the TSSA mutation (20,21), bulk phase ATP hydrolysis activity of the recombinant WT-V 1 was measured. The recombinant V 1 is referred to as WT-V 1 hereafter. The ATP hydrolysis activity of the isolated WT-V 1 was very low (ϳ1.9 s Ϫ1 ) because the enzyme contains 1.21 Ϯ 0.05 (mean Ϯ S.D.) mol of inhibitory ADP per 1 mol of enzyme on average. To measure the ATP hydrolysis activity of the wild type enzymes, the bound nucleotides were removed by successive phosphate/EDTA treatments as described previously (25). The WT-V 1 used for measurement of ATP hydrolysis activity contained 0.38 Ϯ 0.01 (mean Ϯ S.D.) mol of ADP per 1 mol of enzyme after five times treatments. The ATP hydrolysis activity of WT-V 1 was dependent on the amount of bound ADP (see supplemental Fig. 1), indicating that the WT-V 1 is inactivated by entrapment of one ADP molecule per one enzyme molecule. The WT-V-ATPase was reconstituted from the activated V 1 and V 0 , which was isolated from the membrane fraction of T. thermophilus cells. The ATP hydrolysis activity of the WT-V 1 and WT-V-ATPase was measured at various concentrations of ATP in the presence of 0.05% of DDM. The raw data for ATP hydrolysis by the enzymes are shown in Fig. 1A. After the reaction was initiated by the addition of WT-V 1 , an apparent deceleration of the ATP hydrolysis rate was observed with the ATP hydrolysis activity mainly lost within 15 min. This result is similar to the ATP hydrolysis profiles by the WT-V 1 in the absence of detergent (25). ATP hydrolysis by WT-V-ATPase was also observed, but the calculated inactivation rate of (3.6 Ϯ 0.3) ϫ 10 Ϫ3 s Ϫ1 (mean Ϯ S.E.) was lower than that of WT-V 1 of (7.9 Ϯ 0.1) ϫ 10 Ϫ3 s Ϫ1 (mean Ϯ S.E.). After incubation of the activated enzymes with Mg-ADP, the ATP hydrolysis activities were completely abolished (Fig. 1A). The re-inactivated enzymes contained 1.21 Ϯ 0.08 (mean Ϯ S.D.) mol of ADP per 1 mol of enzyme. Analysis of the TSSA mutants (S232A/T235S double mutation in the A subunit) revealed almost continuous ATP hydrolysis for both TSSA-V 1 and TSSA-V-ATPase in contrast to the wild type enzymes. This indicates that the mutations suppress Mg-ADP inhibition (20) (Fig. 1C). The TSSA mutant of V 1 has been used for previous single molecule experiments because of this characteristic. The bound nucleotide in the TSSA enzymes was almost all removed by a single phosphate/EDTA treatment with 1 mol of the TSSA-V 1 used in these experiments containing Ͻ0.03 mol of ADP. As shown in Fig. 1D, the rate of ATP hydrolysis obeyed simple Michaelis-Menten kinetics. The V max and K m values for the TSSA-V 1 were calculated to be 55.8 Ϯ 0.3 s Ϫ1 and 587 Ϯ 21 M (mean Ϯ S.E.), respectively, consistent with previous results (23). Unlike the wild type enzyme, the V max value for the TSSA-V-ATPase was 2-fold lower than that of TSSA-V 1 , 30.6 Ϯ 0.3 s Ϫ1 (mean Ϯ S.E.). In contrast, the K m value for the TSSA-V-ATPase was almost identical to that of the TSSA-V 1 , 510 Ϯ 11 M (mean Ϯ S.E.). The kinetic parameters for ATP hydrolysis by either V 1 or V-ATPase are summarized in Table 1.
Rotation Assay of Wild Type Enzymes-Single molecule analysis is a powerful technique in the determination of the enzymatic properties of molecular motors as it eliminates artifacts through contamination with denatured or inactivated enzyme molecules. Although our group and others have reported direct observation of subunit rotation of V-ATPase, the rotational steps corresponding to a single ATP hydrolysis reaction could not be observed (21,22). Observation of stepwise rotation of V-ATPase will provide mechanistic insights into the rotary catalysis and allow an assessment of the effect of V 0 on the enzymatic properties of V 1 . In this study, we used the rotation analysis system of V-ATPase as depicted in Fig. 2A. V-ATPase was immobilized onto a Ni 2ϩ -NTA-coating glass surface via His 3 tags introduced into the C termini of the V 0 -c subunits. This method differs from those previously described for immobilizing the protein molecules (21). As shown in Fig. 2B, highly purified V 0 from T. thermophilus membranes and recombinant V 1 expressed in E. coli were used for the reconstitution of V-ATPase. Specific biotinylation of the A subunits in V-ATPase was confirmed by Western blotting, as shown in Fig.  2C. Fig. 2D shows the sensitivity of V-ATPase to inactivation by N, NЈ-dicyclohexylcarbodiimide (DCCD), a specific inhibitor that modifies a critical carboxylate in the proteolipid subunit. DCCD has been generally used as a marker to show that F 0 F 1 is intact (30); if proton translocation and ATP hydrolysis are uncoupled, ATP hydrolysis activity is no longer sensitive to DCCD inhibition. The reconstituted V-ATPase was almost completely inactivated by preincubation with DCCD; the residual activity was nearly 10% after 60 min of incubation at room  temperature. In contrast, the isolated V 1 showed only a slight drop in activity under the same conditions. These results indicate that the engineered V-ATPase used for the rotation assay was fully functional.
The wild type enzymes have a strong propensity to lapse into the Mg-ADP inhibited form during catalytic turnover; the rate of inactivation of enzyme is dependent on the ATP concentration in the assay buffer. At 4 mM ATP, almost half the WT-V-ATPase molecules are inhibited within 5 min of addition of ATP, and nearly all molecules cease ATP hydrolysis within 15 min (see Fig. 1A). The inhibition rate is too fast to allow observation of the rotating molecules under ATP-saturated conditions. Below 2 M ATP, however, it was possible to observe the rotating beads. A bead attached to the A subunit of WT-V-ATPase rotated stepwise, pausing every 120° (Fig. 3, A and B), like V 1 . The dwell time between successive 120°steps in WT-V-ATPase at low ATP concentrations corresponds to the time that WT-V-ATPase is waiting for binding of ATP, because binding of ATP is the rate-limiting step under these conditions. Based on histogram analysis (Fig. 3, C and D), k on for ATP of WT-V-ATPase was estimated to be (1.03 Ϯ 0.04) ϫ 10 6 M Ϫ1 s Ϫ1 and (1.11 Ϯ 0.01) ϫ 10 6 M Ϫ1 s Ϫ1 (mean Ϯ S.E.) at both 1 and 2 M ATP, respectively.
Rotation of activated WT-V 1 was also investigated in the presence of 0.05% DDM. To visualize the rotation of the D subunit, the enzymes were immobilized onto a Ni 2ϩ -NTAcoated glass surface by a His 10 tag introduced into the A subunit, and a duplex of streptavidin-coated beads was attached to the biotin-labeled D subunit. To determine the k on value for ATP to WT-V 1 in the presence of detergent, we observed stepwise rotation of WT-V 1 at low ATP concentration. Stepwise rotational beads were found at both 0.5 and 2 M ATP (Fig. 3, E and F). Based on the histogram analysis for dwell times (Fig. 3, G  and H), the k on value of WT-V 1 for ATP was estimated at 0.5 and 2 M ATP to be (1.38 Ϯ 0.02) ϫ 10 6 M Ϫ1 s Ϫ1 and (1.26 Ϯ 0.01) ϫ 10 6 M Ϫ1 s Ϫ1 (mean Ϯ S.E.), respectively. The k on value of WT-V 1 is nearly equal to that of WT-V-ATPase. The V max of WT-V 1 for ATP hydrolysis is roughly estimated to be ϳ270 s Ϫ1 as a product of k on of 1.3 ϫ 10 6 M Ϫ1 s Ϫ1 (single molecular analysis) and a K m of 205 M (bulk-phase analysis). These kinetic parameters estimated by single molecule assay are summarized in Table 1.
Single Molecule Analysis of TSSA Mutants-Rotation of single molecules of the TSSA-V-ATPase was also investigated. The rotational speed was basically proportional to the ATP concentration between 10 and 100 M (Fig. 4A). At 4 mM ATP, a rotational rate of 9 revolutions/s was observed. One-third of the bulk-phase ATP hydrolysis rate of the TSSA-V-ATPase was roughly equal to the rotation rate below 500 M ATP, suggesting that three ATPs were consumed per rotation. Below 50 M ATP, the bead attached to the A subunit rotated stepwise, pausing every 120° (Fig. 4, B and C), like WT-V-ATPase. The dwell time between successive 120°steps in the TSSA-V-ATPase at low ATP concentrations corresponds to the time that the TSSA-V-ATPase is waiting for binding of ATP. Fig. 4, D and E, show histograms of the dwell times at 10 and 20 M ATP, respectively. By fitting the dwell times with a single exponential equation, k on for ATP to the TSSA-V-ATPase was estimated at 10 and 20 M ATP to be (6.21 Ϯ 0.23) ϫ 10 4 M Ϫ1 s Ϫ1 and (5.62 Ϯ 0.11) ϫ 10 4 M Ϫ1 s Ϫ1 (mean Ϯ S.E.), respectively. These values for the TSSA-V-ATPase are about 20-fold lower than that of WT-V-ATPase, suggesting that the TSSA mutant V-ATPase has reduced affinity for ATP.
The rotation of the TSSA-V 1 was also investigated in the presence of detergent. Similar to the TSSA-V-ATPase, the mutant V 1 exhibited an ATP hydrolysis rate comparable with the rotation rate of the enzyme concentrations of ATP of less than 100 M (see supplemental Fig. 2A). The TSSA-V 1 showed stepwise rotation below 10 M of ATP concentration (see supplemental Fig. 2, B and C). The k on value for ATP of the TSSA-V 1 was estimated at 4 and 10 M ATP to be (1.93 Ϯ 0.03) ϫ 10 5 M Ϫ1 s Ϫ1 and (1.51 Ϯ 0.02) ϫ 10 5 M Ϫ1 s Ϫ1 (mean Ϯ S.E.), respectively (see supplemental Fig. 2, D and E). These values are comparable with previous results in the absence of detergent (23).
Kinetic Analysis of ATP Synthesis by V-ATPase-The physiological role of the V-ATPase of T. thermophilus is ATP synthesis (14). To determine kinetic parameters for the ATP synthesis reaction, the isolated WT-and TSSA-V-ATPase were reconstituted into proteoliposomes using a freeze-thaw and sonication method. The proteoliposomes were energized by a  transmembrane pH difference (acid-base transition: pH out ϭ 8.4, pH in ϭ 4.9) and a K ϩ /valinomycin diffusion potential ([K ϩ ] out ϭ 100 mM, [K ϩ ] in ϭ 1 mM). The generated pmf is estimated to be 295 mV from the Nernst equation. The ATP synthesized by the enzymes was measured by a luciferin/luciferase system as described under "Experimental Procedures." After addition of acidic proteoliposome solution into the base buffer containing luciferin/luciferase and ADP or P i at indicated concentrations, the rapid increase of luminescence intensity was observed linearly within 5 s. The rate of ATP synthesis was calculated from the slope at t ϭ 0. The raw data for ATP synthesis at various ADP or P i concentrations are presented in supplemental Fig. 3, A and B. Fig. 5A shows  Table  2. These results clearly indicate that the TSSA mutation affects the K m for both ADP and Pi. The V max values for ATP synthesis are also decreased in the TSSA mutant enzymes.

DISCUSSION
In this study, we investigated the detailed enzymatic properties of T. thermophilus V-ATPase using biophysical and bio-chemical methods. A comparison of ATP-driven rotation of wild type V-ATPase with that of the V 1 domain indicates that both rotate stepwise, pausing roughly every 120°. The histograms of dwell time between steps, which corresponds to time waiting for ATP, revealed that the binding rate constant k on of wild type V-ATPase is nearly identical to that of the wild type V 1 . In addition kinetic parameters of V-ATPase measured by bulk phase analysis were similar to those of V 1 ( Table 1). These results indicate that the enzymatic properties of V 1 are not affected by association with V 0 .
The ATP hydrolysis activity of WT-V-ATPase from T. thermophilus is easily abolished by Mg-ADP, whereas the enzyme is capable of effective ATP synthesis from ADP and phosphate using pmf (Fig. 5). The F 0 F 1 from a thermophilic eubacterium Bacillus sp. PS3 also entraps an inhibitory Mg-ADP in a catalytic site, resulting in enzyme inhibited for ATP hydrolysis (31). However, this entrapped Mg-ADP in the F 0 F 1 is capable of being released from the catalytic site (32,33). Thus, the F 0 F 1 from PS3 can catalyze both the ATP synthesis/hydrolysis directions. In this respect, the V-ATPase of T. thermophilus seems to be more similar to the F 0 F 1 from Paracoccus denitrificans, which catalyzes rapid ATP synthesis coupled to the pmf, whereas ATP hydrolysis occurs but at a very low rate (19).
Single molecule analysis for ATP hydrolysis by V-ATPase, indicated a binding rate constant k on of 1.1 ϫ 10 6 M Ϫ1 s Ϫ1 for ATP. This value is ϳ30-fold lower than that of the F 0 F 1 from PS3 (3.6 ϫ 10 7 M Ϫ1 s Ϫ1 ) (30), suggesting a much slower rate of ATP binding by V-ATPase from T. thermophilus during ATP hydrolysis.
We also determined kinetic parameters for the ATP synthesis reaction. If V-ATPase is evenly oriented within the liposomes, the V max of ATP synthesis is estimated to be ϳ140 s Ϫ1 . This value is nearly equal to the V max of 147 s Ϫ1 , a product of k on (1.1 ϫ 10 6 M Ϫ1 s Ϫ1 , from single molecule analysis) and K m (134 M, from bulk phase analysis) obtained from this study. The estimated k on (V max /K m ) for ADP is 1.4 ϫ 10 7 M Ϫ1 s Ϫ1 , ϳ13fold higher than the k on for ATP during ATP hydrolysis. These two enzymatic properties of the V-ATPase from T. thermophilus, 1) lower k on for ATP and 2) irreversible inactivation because of Mg-ADP inhibition, might prevent wasteful consumption of ATP when the pmf has collapsed because of inhibition of respiration, for example, at mesophilic temperatures.
We have also investigated the enzymatic properties of the TSSA mutant enzymes, which have low propensity for Mg-ADP inhibition during ATP hydrolysis. Single molecule analysis for ATP hydrolysis revealed that the TSSA mutants of V 1 and V-ATPase have apparently lower k on values for ATP than the wild type enzymes. In addition, TSSA-V-ATPase has an

Rotary Catalysis of V-ATPase in Both Directions
ϳ20-fold lower V max /K m value for ADP than wild type in the ATP synthesis reaction ( Table 2). The bound-ADP in TSSA-V 1 was mostly removed by one phosphate/EDTA treatment even though the wild type enzyme still contains ϳ0.3 mol of ADP per 1 mol of enzyme after five successive phosphate/EDTA treatments (data not shown). These results indicate that the binding affinity to ADP is decreased in the TSSA mutants. Noticeably, the V max rate of TSSA V-ATPase for ATP synthesis was about five times lower than of that of wild type V-ATPase. This suggests that one or more of the elementary step(s) involved in ATP synthesis are prolonged by the TSSA mutation.
In TSSA-V-ATPase, the ATP hydrolysis rate of ϳ35 s Ϫ1 calculated from the K m from the bulk phase analysis and k on of the single molecule analysis is nearly equal to the V max value (ϳ30 s Ϫ1 ) of bulk phase analysis. Indeed, the ATP hydrolysis rate of TSSA-V-ATPase estimated from single molecule analysis is consistent with that obtained from bulk phase analysis at a range of from 4 to 1000 M ATP (Fig. 4A). These results indicate that most of the TSSA-V-ATPase molecules are active for ATP hydrolysis. In contrast, ATP hydrolysis by wild type V-ATPase has a V max value (ϳ41 s Ϫ1 ) from bulk phase analysis much lower than 143 s Ϫ1 , a product of K m (bulk phase) and k on (single molecule). The V max value of wild type V 1 (ϳ40 s Ϫ1 ) of bulk phase analysis was also much lower than ϳ270 s Ϫ1 , a product of K m (bulk phase) and k on (single molecule). This inconsistency is likely caused by contamination with Mg-ADP-inhibited wild type V-ATPase.
In this study, we investigated the ATP synthesis reaction of the V-ATPase only by bulk phase analysis, and the kinetic parameters obtained indicate that the enzyme avoids wasteful consumption of ATP when the pmf has collapsed in vivo. Future work will aim to observe the ATP synthesis reaction using single molecule techniques which will further clarify the kinetic parameters for ATP synthesis.