V-ATPase of Thermus thermophilus Is Inactivated during ATP Hydrolysis but Can Synthesize ATP*

The ATP hydrolysis of the V1-ATPase of Thermus thermophilus have been investigated with an ATP-regenerating system at 25 °C. The ratio of ATPase activity to ATP concentration ranged from 40 to 4000 μm; from this, an apparent K m of 240 ± 24 μm and a V max of 5.2 ± 0.5 units/mg were deduced. An apparent negative cooperativity, which is frequently observed in case of F1-ATPases, was not observed for the V1-ATPase. Interestingly, the rate of hydrolysis decayed rapidly during ATP hydrolysis, and the ATP hydrolysis finally stopped. Furthermore, the inactivation of the V1-ATPase was attained by a prior incubation with ADP-Mg. The inactivated V1-ATPase contained 1.5 mol of ADP/mol of enzyme. Difference absorption spectra generated from addition of ATP-Mg to the isolated subunits revealed that the A subunit can bind ATP-Mg, whereas the B subunit cannot. The inability to bind ATP-Mg is consistent with the absence of Walker motifs in the B subunit. These results indicate that the inactivation of the V1-ATPase during ATP hydrolysis is caused by entrapping inhibitory ADP-Mg in a catalytic site. Light-driven ATP synthesis by bacteriorhodopsin-VoV1-ATPase proteoliposomes was observed, and the rate of ATP synthesis was approximately constant. ATP synthesis occurred in the presence of an ADP-Mg of which concentration was high enough to induce complete inactivation of ATP hydrolysis of VoV1-ATPase. This result indicates that the ADP-Mg-inhibited form is not produced in ATP synthesis reaction.

V o V 1 -ATPases and F o F 1 -ATPases constitute two subclasses of the ATPase/ATP synthase superfamily (1,2). V o V 1 -ATPases are present in the membranes of lysosomes (3), clathrin-coated vesicles (4), chromaffine granules (5), and the central vacuoles of yeast (6). They are responsible for vacuolar acidification, which plays an important role in a number of cellular processes (1). V o V 1 -ATPases are also found in the plasma membranes of most archea (7)(8)(9) and some kinds of eubacteria (10 -12). Several studies indicate that the physiological role of V o V 1 -AT-Pases of some archea and the thermophilic eubacterium Thermus thermophilus is ATP synthesis coupled to proton flux across the plasma membranes (7,9,(13)(14)(15).
V o V 1 -ATPases consist of two functional assemblies, a periph-eral V 1 moiety and a membrane integrated V o moiety, which are counterparts of the F 1 and F o moiety of the F o F 1 -ATPase (1,(15)(16)(17). The peripheral V 1 moiety is composed of two major subunits, A and B, and other minor subunits. Both structural analysis and sequence homology indicate an evolutionary relationship between V o V 1 -ATPases and F o F 1 -ATPase and that the A and B subunit of V o V 1 -ATPase are homologous to the ␤ and ␣ subunit of F o F 1 -ATPases (2). The A subunit of V o V 1 -ATPases contains the Walker motifs (1), which are critical for nucleotide binding (18,19). Labeling of the A subunit by 2-azido-[ 32 P]ATP correlates well with inactivation of ATPase activity, with complete inactivation observed upon modification of a single A subunit per complex (20). These findings indicate that the catalytic site of the V o V 1 -ATPase is located on the A subunit.
On the other hand, the B subunit of V o V 1 -ATPases lacks Walker motifs. A recent study reported that the B subunit in the V o V 1 -ATPase of clathrin-coated vesicles was modified by 3-O-(4-benzoyl)benzoyladenosine 5Ј-triphosphate (21). However, any direct evidence for the nucleotide binding to the isolated B subunit of the V o V 1 -ATPase has not been reported yet. Structural similarity and sequence homology of the major subunits of V o V 1 -ATPases and F o F 1 -ATPases lead to the hypothesis that the mechanisms of ATP hydrolysis and ATP synthesis by V o V 1 -ATPases are almost identical to those of F o F 1 -ATPases. Nevertheless, the enzymatic properties of V o V 1 -ATPases and F o F 1 -ATPases are different (1). Whereas azide inhibits ATP hydrolysis by F 1 -ATPases by stabilizing the inhibitory ADP-Mg-F 1 -ATPase complex (22)(23)(24), it does not inhibit ATPase activity of V o V 1 -ATPases (1,10).
Precise understanding of V o V 1 -ATPases would allow the comparison to F o F 1 -ATPases and the elucidation of the common essential mechanism for the coupling of proton translocation across a membrane with ATP formation. However, several problems, such as the difficulty of obtaining a large amount of pure enzyme from vacuolar membranes and an unstable V 1 moiety (17), have limited our investigation of enzymatic properties of V o V 1 -ATPases.
T. thermophilus, originally isolated from a hot spring in Japan, is thermophilic, obligatory aerobic, Gram-negative, and chemoheterotrophic eubacterium (25). Its respiratory chain may include energy coupling Site I (26). This bacterium has a large amount of the V o V 1 -ATPase on the plasma membrane, instead of F o F 1 -ATPase (15).
In contrast to eukaryotic equivalents, the V 1 moiety of T. thermophilus is easily detached from the membranes using chloroform treatment and ATPase-active stable complex can be obtained in large amounts (10). Throughout this manuscript, the V 1 moiety from T. thermophilus is refereed to V 1 -ATPase.
The V 1 -ATPase consists of four kinds of subunit with appar-* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ent molecular sizes of 66 (A or ␣), 55 (B or ␤), 30 (␥), and 11 (␦) kDa, which are present in a stoichiometry of A 3 B 3 ␥ 1 ␦ 1 . Similar to its eukaryotic counterparts, the V 1 -ATPase also shows enzymatic properties different from those of F 1 -ATPases, such as low specific activity, high K m values, and resistance to azide inhibition (10). We previously reported a specific activity of the V 1 -ATPase of about 0.1 units/mg of protein at 55°C in the absence of an ATP-regenerating system.
In this report, we demonstrate the particular kinetic behaviors of the V 1 -ATPase of T. thermophilus in the presence of ATP regenerating system, the nucleotide binding properties of the isolated A and B subunits, and ATP synthesis of a V o V 1 -ATPase co-reconstituted with bacteriorhodopsin for the first time.

EXPERIMENTAL PROCEDURES
Materials-V 1 -ATPase and V o V 1 -ATPase were prepared from T. thermophilus plasma membranes using the previously described methods (10,15). Bacteriorhodopsin (bR) 1 was prepared from Halobacterium halobium (27). [␥-32 P]ATP was purchased from NEN Life Science Products. Radioactive ATP (9.25MeBq/ml) was diluted with nonradioactive ATP to the desired specific activity. Egg yolk phosphatidylcholine (type XVI-E) was purchased from Sigma. Other chemicals were purchased from Nacarai Corp.

Construction of Overexpression Systems of the A and B Subunits and
Purification of the Products-The DNA fragments of the A and B subunits were PCR amplified using Ex Taq™ polymerase (TaKaRa) to minimize errors. The sequence of the amplified DNA fragments was confirmed using ABI 373A sequencer. The obtained DNA fragments were ligated into pUC18 to result in vectors pATPA and pATPB, respectively. After transformation in Escherichia coli strain JM 103, the expression of subunit genes was induced by addition of 0.4 mM isopropyl-1-thio-␤-D-galactopyranoside.
More than 10% of the total soluble protein was the recombinant A subunit after induction, about 5% in case of the recombinant B subunit (see Fig. 5). Both subunits were purified by the same method. About 10 g of cells were obtained from an overnight culture, suspended in 50 ml of buffer containing 50 mM Tris-SO 4 (pH 8.0), 50 mM NaCl, and 0.1 mM EDTA, and disrupted by sonication. The membrane fraction and cell debris were removed by centrifugation, and the supernatant was applied onto a DEAE Sephacel column (2 ϫ 10 cm) equilibrated with Buffer A (50 mM Tris-Cl (pH 8.0) and 0.1 mM EDTA). The column was washed with 200 ml of Buffer A, and the proteins were eluted with linear NaCl gradient (0 -0.5 M). Fractions containing the recombinant subunit were chosen after SDS-PAGE analysis (15% of acrylamide gels). The fractions were combined, and solid ammonium sulfate was added to a final concentration of 1.2 M and stirred for 1 h. After precipitation by centrifugation, the supernatant was applied onto a butyl-toyopearl column (Tosoh Corp.; 2 ϫ 10 cm). Proteins were eluted with a reverse ammonium sulfate gradient (1.2-0 M). Fractions containing the subunit were combined, and proteins were precipitated by ammonium sulfate. The precipitate was dissolved with a minimum volume of Buffer A, and it was applied onto a Sephacryl S-300 gel permeation column (1 ϫ 90 cm) equilibrated with Buffer A plus 50 mM Na 2 SO 4 . The column was eluted with the same buffer and fractions containing the subunits were chosen after SDS-PAGE analysis (15% of acrylamide gels). The purified subunits were stored at 4°C until use.
Analytical Methods-The protein concentrations of V 1 -ATPase were determined by measurement of absorbance at 280 nm using a factor of 0.59 for the absorbance of 1 mg/ml protein. The factor of 0.59 was determined by quantitative total amino acid analysis and spectral data. Unless otherwise specified, ATPase activity was measured at 25°C with an enzyme-coupled ATP-regenerating system. The reaction mixture contained 50 mM Tris-Cl (pH 8.0), 100 mM KCl, different concentrations of ATP-Mg, 2.5 mM phosphoenolpyruvate, 50 g/ml pyruvate kinase, 50 g/ml lactate dehydrogenase, and 0.2 mM NADH in a final volume of 1.2 ml. Typically, the reaction was started by addition of the enzyme dissolved in 50 mM Tris-SO 4 (pH 8.0), 50 mM Na 2 SO 4 , 0.2 mM EDTA to 1.2 ml of the assay mixture, and the rate of ATP hydrolysis was monitored as the rate of oxidation of NADH determined by the absorbance decrease at 340 nm. The data were stored in a computer for further analyses. The spectrometer was equipped with a small stirrer for rapid mixing and with a device that enabled us to start the reaction by injecting the enzyme solution without opening the lid. We confirmed that the maximum dead time was less than 5 s after starting the reaction. A phosphate-molybdate assay was used for the measurement of the ATPase activity in the presence of ADP. The standard reaction mixture was prepared in a final volume of 0.5 ml, containing 4.5 mM ATP-Mg, 2 mM MgCl 2 , 100 mM KCl, and 50 mM Tris-Cl (pH 8.0). The reaction mixture was preincubated for 3 min at 25°C before starting the reaction by addition of enzyme. The reaction was terminated with 0.3 ml of 3% perchloric acid, and the amount of P i was assayed as described previously (28).
Enzyme-bound nucleotide was analyzed by anion exchange high performance liquid chromatography. Bound nucleotides were released from the enzyme by addition of 5 l of 60% perchloric acid to 50 l of the enzyme solution. Thereafter, the mixture was incubated on ice for 10 min. Then, 5 l of 5 M K 2 CO 3 solution were added to the mixture and incubated on ice for 10 min. The resulting pellet was removed by centrifugation at 4°C. The supernatant was applied to a Cosmopak-200 column equilibrated with 0.1 M sodium-phosphate buffer (pH 7.0). The column was eluted isocratically with the same buffer at a flow rate of 0.8 ml/min. The nucleotide was monitored at 258 nm. The peak area was determined by automatic integration.
Interaction of ATP with the isolated subunit was assayed by UV difference spectra at 25°C with a Double Beam Spectrophotometer model U-3200 (Hitachi Corp.) using a pair of matched double cells, as described in Ref. 29. The subunits were dissolved in 10 mM Tricine-NaOH buffer (pH 8.0) at a final concentration of 10 M.
Preparation of Proteoliposomes and Light-induced ATP Synthesis-Proteoliposomes containing bR and V o V 1 -ATPase were reconstituted according to the procedure by Richard et al. (30). The reconstitution was performed at 25°C in 25 mM potassium phosphate buffer (pH 7.3), 50 mM K 2 SO 4 , and 50 mM Na 2 SO 4 . Unilamellar liposomes were prepared by reverse phase evaporation using phosphatidylcholine and resuspended at a lipid concentration of 4 mg/ml. Triton X-100 was added to a final concentration of 8 mg/ml. bR was solubilized from purple membranes with 2 mg/ml Triton X-100. Then, 50 l of solubilized bR solution (5 mg of protein/ml) and 10 l of V o V 1 -ATPase solution (3 mg of protein/ ml) were added to 850 l of liposome solution. n-Octyl-␤-D-glucopyranoside was added to a final concentration of 20 mM, and the mixture was incubated for 5 min. Then, pyranine (excitation, 450 nm; emission, 510 nm) was added to the mixture at a final concentration of 0.2 mM. The detergent was removed by four successive additions of 80 mg/ml washed Bio-beads SM-2 (Bio-Rad). The measurement of the fluorescence of pyranine trapped inside the liposomes confirmed the low leakage of protons from the reconstituted liposomes. The mixture was incubated at 40°C and preilluminated for 15 min before the ATP synthesis reaction was started by addition of 2 mM MgSO 4 . Aliquots of the illuminated sample were taken at different reaction times, and the reactions were quenched with trichloroacetic acid. The ATP content was measured using a luciferin-luciferase assay. Fig. 1a shows the time course of ATP hydrolysis measured with the ATP regenerating system. Hydrolysis of 200 -4000 M ATP by V 1 -ATPase proceeded in three distinct phases. Within 10 s after the reaction was started by addition of enzyme, an apparent short initial lag was observed (Fig. 1a). The initial lag phase rapidly transformed to a second phase of high rate of hydrolysis. Then, the rate of hydrolysis was decelerated slowly. The ATP hydrolysis almost stopped after 20 min (data not shown). Fig. 1b shows the change of absorbance at 340 nm over time (dA 340 /dt), illustrating the three phases of hydrolysis. The decrease in dA 340 /dt corresponds to the activation of the ATPase and the increase to the inactivation. The results indicate that the initially inactive species of the V 1 -ATPase were rapidly activated, and the activated species were re-inactivated during ATP hydrolysis.

Time-dependent Change of the ATPase Activity of V 1 -ATPase-
Interestingly, both the rates of initial activation and subsequent inactivation of the V 1 -ATPase were dependent on the ATP concentration in the assay mixture. As shown in Fig. 1a, the rapid inactivation was only observed at high concentrations of ATP. The inactivation almost disappeared and the rate of ATP hydrolysis was almost constant during turnover at 30 M ATP (Fig. 1, a and b, trace 0.03 mM). The V o V 1 -ATPase also exhibited similar activation and inactivation phases in hydrolyzing 0.01-4 mM ATP-Mg (Fig. 1c). The time-dependent changes were analyzed using a simple sequential model for the activation and inactivation processes, and the apparent first order rate constants for each process were calculated by nonlinear regression fitting. The apparent rate constants were plotted against the ATP concentrations. As shown in Fig. 2, the rate constants of inactivation exhibited a monophasic dependence on ATP concentration, and the half-maximum rate of inactivation was attained at 140 M ATP. The associated maximum inactivation rate constant was 0.0055 s Ϫ1 . We calculated the ATP concentration for the half-maximum rate of activation with the same method. It was 70 M ATP, and the associated maximum activation rate constant was 0.65 s Ϫ1 (Fig. 2b).
By analogy to the inactivation of F 1 -ATPases (24,31,32), we suspected that the inactivation was caused by entrapment of inhibitory ADP at a catalytic site. To investigate the effect of ADP for the inactivation, we assayed the ATPase activity of V 1 -ATPase by the measurement of inorganic phosphate with or without the ATP regenerating system. When 50 g/ml pyru-vate kinase and 2 mM phosphoenolpyruvate were present in the assay mixture, the hydrolysis of ATP proceeded linearly up to 4 min after addition of enzyme. Then, the rate of hydrolysis gradually decelerated 5 min after the reaction was started. On the other hand, in the absence of the ATP-regenerating system, the deceleration of the rate of hydrolysis occurred more rapidly, and the hydrolysis completely stopped at 6 min (Fig. 3a). The apparent rate constant of inactivation of the V 1 -ATPase under these conditions was about 3 ϫ 10 Ϫ3 s Ϫ1 in the presence of an ATP-regenerating system, and about 7 ϫ 10 Ϫ3 s Ϫ1 in the absence of an ATP regenerating system. These values are in the same order of magnitude as the value deduced in Fig. 2a. Fig.  3b shows the effect of ADP in the assay mixture on the rate of hydrolysis of ATP. Various amounts of ADP were added to the assay mixtures in the absence of a regenerating system, and the hydrolysis of 4 mM ATP was assayed. Increasing ADP concentrations in the assay mixtures led to a decrease of the rate of ATP hydrolysis.
Furthermore, 1 M V 1 -ATPase was preincubated with various concentrations of ADP-Mg for 120 min at 25°C, and the residual ATPase activities were measured in the presence of 4 mM ATP-Mg. As shown in Fig. 3c, the extent of inactivation of The rates of inactivation and activation were estimated by nonlinear curve fitting assuming sequential activation and inactivation during the ATPase assay. Inset, corresponding Eadie-Hofstee plots. a, the rate of inactivation was plotted as a function of the ATP concentration. The range of ATP concentrations was 40 -4000 M. The solid line represents the best fit. b, the rate of activation was plotted as a function of the ATP concentrations. The solid line represents the best fit. the V 1 -ATPase was dependent on the concentration of added ADP-Mg.
These results strongly suggest that the mechanism of inactivation is similar to that of the ADP-Mg inhibition observed for F 1 -ATPase.

Kinetics of ATP Hydrolysis by V 1 -ATPase-Because the ATP
hydrolysis by V 1 -ATPase did not proceed linearly over time, it is difficult to define the rate at a given ATP concentration. In the present study, we plotted the maximum rate of ATP hydrolysis at each ATP concentration (Fig. 4). The ATP concentration ranged from 40 to 4000 M. The kinetic data were also plotted in V/S versus V form (inset, Eadie-Hofstee plot). From this plot, an apparent K m of 240 Ϯ 24 M and an apparent V max of 5.2 Ϯ 0.5 units/mg were deduced. An apparent negative cooperativity that is frequently observed for F o F 1 -or F 1 -ATPase (31,(33)(34)(35)(36)(37)(38)(39)(40)(41)43) was not observed for the V 1 -ATPase here.
Binding of ATP to the Isolated Subunits-The kinetics features of inactivation of V 1 -ATPase during the ATPase reaction described above are similar to those recently reported for the mutant F 1 -ATPase, which lacks nucleotide binding at the noncatalytic sites (32). In that case, turnover-dependent inactivation was explained as the failure to recover from the ADP inhibited state due to the inability of nucleotide binding at noncatalytic sites. If the V 1 -ATPase lacks nucleotide binding to the noncatalytic subunit (subunit B), the similarity of V 1 -ATPase to mutant F 1 -ATPase is well understood. The lack of Walker motifs of B subunit of V 1 -ATPase further underlined this idea. Thus, in an attempt to characterize nucleotide binding to the isolated subunits of V 1 -ATPase, we constructed over expression systems for the A and B subunits and purified the products. We could successfully obtain large amounts of these subunits (Fig. 5). The isolated recombinant A and B subunits migrated as single protein bands with little contamination.
Difference absorption spectra have been used to probe the binding of adenine nucleotides to the F 1 -ATPase and its isolated subunits (29). When these proteins bind ADP or ATP, difference spectra are induced by a red shift of the absorption maximum accompanied by a slight decrease of the magnitude. The difference absorption spectra induced by the interaction of ATP with the isolated A or B subunits showed significantly different profiles (Fig. 6). Upon addition of ATP-Mg to the isolated A subunit, a trough at 260 nm and a peak at 280 nm were observed, and the magnitudes of the peak and the trough depended on the amount of the added ATP. Saturation was In the presence of the ATP regenerating system, the assay mixtures contained 50 g/ml pyruvate kinase and 2 mM phosphoenolpyruvate. The reactions were stopped at the indicated times, and the generated P i was measured by the P i -molybdate assay. b, time course of P i generation in the presence of various ADP-Mg concentrations in the assay mixture. The reactions were started by the addition of enzyme to the assay mixtures. The reactions were stopped at indicated times, and the P i in the assay mixture was measured. observed when the concentration of ATP reached about an equimolar concentration to the A subunit (Fig. 6a, inset). In contrast, neither a trough nor a peak was observed for the B subunit (Fig. 6b).
2Ј,3Ј-O-(2,4,6-Trinitrophenyl)-ATP is known to have a significantly higher affinity for the nucleotide binding sites of F 1 -ATPase and to give a clear difference spectrum in the visible wavelength region upon binding (29,42,43). The difference spectrum was induced by addition of 2Ј,3Ј-O-(2,4,6-trinitrophenyl)-ATP to the isolated A subunit, but was not induced in case of the B subunit (data not shown). These results suggest that the isolated A subunit can bind 1 mol of ATP-Mg per enzyme, but the isolated B subunit cannot bind ATP-Mg. The ability and inability of ATP binding of the A and B subunit are consistent with the presence and absence of the Walker motifs in these subunits.
Analysis of Bound Nucleotide of V 1 -ATPase-To analyze bound nucleotide, the V 1 -ATPase was preincubated with or without 1 mM ADP-Mg for 60 min, and free nucleotides were removed with a Sephadex G-50 centrifuge column. The enzyme was denatured with perchloric acid, and the amount of released adenine nucleotide was quantified by anion exchange high performance liquid chromatography. The analysis revealed that endogenous ADP on the purified V 1 -ATPase preparation was less than 0.1 mol per mol of enzyme, whereas ADP-Mgpreincubated V 1 -ATPase contained 1.5 mol of ADP per mol of the enzyme.
T. thermophilus V o V 1 -ATPase Synthesizes ATP Coupled to Proton Flux-ATP synthesis is the physiological role of T. thermophilus V o V 1 -ATPase (15). To measure ATP synthesis, we co-reconstituted V o V 1 -ATPase with bR into proteoliposomes. A light-induced pH gradient was generated, and the proton permeability was monitored using the fluorescence of the pHsensitive probe pyranine trapped inside of the co-reconstituted V o V 1 -ATPase-bR proteoliposomes. Light-induced proton translocation was observed and the accumulation of protons saturated after illumination for 10 min. Because the level of accumulated proton remained relatively constant after the illumination, the V o V 1 -ATPase-bR proteoliposomes have a low proton permeability (Fig. 7, inset).  After preillumination for 15 min to generate a stable transmembrane pH gradient, the ATP synthesis reactions were started by addition of MgSO 4 (indicated as time 0 in Fig. 7). The synthesized ATP was measured with the luciferin-luciferase assay. ATP synthesis proceeded with a constant rate for 40 min after starting the reaction. The constant rate of ATP synthesis was found to be 0.67 mol of ATP mg Ϫ1 min Ϫ1 , which is 3-4 times larger than the one of F o F 1 -ATPase from thermophilic bacterium, PS3 (30). The time-dependent inhibition by ADP observed in case of ATP hydrolysis was not observed in ATP synthesis.

Inactivation of V 1 -ATPase May Be Caused by Entrapment of Inhibitory ADP-Mg in a Catalytic
Site-The results presented clearly show three distinct phases of ATP hydrolysis by the V 1 -ATPase in the presence of an ATP regenerating system (Fig.  1, a and b). The initial lag apparently shows the presence of initial inhibited species of V 1 -ATPase. The inhibited enzyme changed rapidly to an active form. The cause of this initial inhibition is not clear at present. The activated enzyme was re-inactivated during turnover. The apparent half-maximum rate of the inactivation was attained at 140 M ATP, which reflects the affinity of the ATP binding site for the inactivation. The presence of ADP-Mg in the assay mixture increased the rate of turnover-dependent inactivation.
The specific activity of V 1 -ATPase is 5.2 unit/mg protein at 25°C in the presence of an ATP-regenerating system, but we previously reported that the specific activity of V 1 -ATPase was about 0.1 unit/mg protein at 55°C in the absence of ATP regenerating system (10). Probably, the low specific activity was due to the turnover-dependent inactivation of the V 1 -ATPase. In addition, contaminant ADP in the ATP solution could increase the rate of inactivation, so that ATP hydrolysis almost stopped 1-2 min after starting the reaction.
Previously, a similar slow inactivation of yeast V-ATPase was reported by Kibak et al. (44). They showed sulfite eliminates the inactivation, but in our case, 33 mM Na 2 SO 3 apparently inhibited the initial rate of the ATP hydrolysis of T. thermophilus V 1 -ATPase (data not shown).
In the case of mitochondorial F 1 -ATPase, Jault and Allison (34) indicated that three kinetics phases were present at low ATP concentration. An initial burst phase decelerated rapidly to a slow intermediate phase, which, in turn, gradually accelerated to a final steady-state rate. They postulated that the transition of the initial burst phase to the slow intermediate phase was caused by accumulation of inhibitory ADP-Mg at a catalytic site and that the transition of the intermediate phase to the final steady state was caused by binding of ATP to noncatalytic sites, which promoted the dissociation of inhibitory ADP-Mg from the affected catalytic site (31,34). Recently, Matsui et al. (32) observed a rapid and nearly complete turnoverdependent inactivation of a mutant F 1 -ATPase, which lacks the ability of nucleotide binding at noncatalytic sites. The mutant F 1 -ATPase was also completely inactivated by prior incubation with stoichiometric ADP-Mg; thus, they concluded that the entrapment of ADP-Mg in a catalytic site caused the turnoverdependent inactivation. Interestingly, they also observed the ATP concentration dependence of the rate of inactivation and found that the half-maximum rate of inactivation was attained at 5 M ATP, which coincides with one of the two K m values, about 4 M, obtained from initial rate analysis (32). In this study, we observed a similar inactivation of V 1 -ATPase during turnover of ATP hydrolysis. The K m of 240 M determined for the hydrolysis of 40 -4000 M ATP may be comparable to the apparent K d of 140 M for inactivation. Furthermore, nearly complete inactivation of V 1 -ATPase was attained by prior incubation of V 1 -ATPase with ADP.
The analogy of the inactivation of F 1 -ATPases leads us to postulate that the inactivation of V 1 -ATPase is caused by the entrapment of inhibitory ADP-Mg on the catalytic sites. The role of the noncatalytic nucleotide binding sites of F 1 -ATPase during ATP hydrolysis is the release of inhibitory ADP bound at the catalytic sites. It was shown that a mutant of F 1 -ATPase that lacked the ability of nucleotide binding on the noncatalytic sites exhibited strong inhibition by ADP (32). The strong inactivation of the V 1 -ATPase seems similar to that of the mutant of F 1 -ATPase. Actually, the B subunit of the V 1 -ATPase does not have a region homologous to the Walker motifs A and B. As expected, the interaction of ATP and the isolated B subunit was not observed in measurements of difference spectra. Taken together, we prefer the view that the noncatalytic B subunit in the T. thermophilus V 1 -ATPase does not bind nucleotide or has a only very weak affinity for nucleotides. The results obtained suggest that the inactivation of V 1 -ATPase was due to the failure of binding of ATP to noncatalytic sites.
Endogenous ADP on V 1 -ATPase was less than 0.1 mol per mol of enzyme, whereas 1.5 mol of ADP per mol of the enzyme was bound to V 1 -ATPase after the preincubation with 1 mM ADP-Mg for 1 h. Because the loaded nucleotides could not be removed by centrifugation elutions, this nucleotide is thought to bind the catalytic site with high affinity.
The mechanism of initial activation of V 1 -ATP is unknown. Unlike the irreversibly turnover-dependent inactivated enzyme, the initial inhibited enzyme is rapidly activated by the binding of ATP. Furthermore, the initial activation occurred at a lower ATP concentration range than the turnover dependent inactivation. These results clearly show that the initial inhibited form is not identical to the irreversible inactivated form. Further studies will be necessary to clarify the characteristics of the bound adenine nucleotides and the initial inhibited form.
V o V 1 -ATPase Can Synthesize ATP in Co-reconstituted Liposomes-Several findings indicate that the physiological role of V o V 1 -ATPases in some archea and T. thermophilus is the synthesis of ATP coupled to a proton flux (7-10, 13, 15, 25, 26). The results in this study give direct evidence for the ability of T. thermophilus V o V 1 -ATPase to synthesize ATP coupled to proton flux. This is the first report of ATP synthesis with a reconstituted proteoliposome of a V o V 1 -ATPase. We used the V o V 1 -ATPase-bR co-reconstituted proteoliposomes because a steady pH potential is attained by light-induced proton pumping. The reaction mixture for ATP synthesis contained 2 mM ADP-Mg, which is sufficient to induce complete inactivation of the V o V 1 -ATPase for ATP hydrolysis. However, ATP synthesis continued for up to 40 min indicates that ADP-Mg-induced inactivation does not occur under ATP synthesis conditions. It is possible that the membrane potential and/or the pH gradient protects V o V 1 -ATPase from ADP-Mg inactivation.
The particular characteristics of T. thermophilus V o V 1 -ATPase, where ATP hydrolysis-turnover-dependent inactivation occurs under ATP hydrolysis condition but ATP synthesis is not inhibited in the presence of a proton motive force, are thought to be favorable for physiological ATP synthesis. When the proton motive force is close to zero for the cell, the hydrolysis of intracellular ATP may be inhibited by the generation of the inactivated species of the V o V 1 -ATPase, so that a rapid decrease of intracellular ATP is avoided.
During the preparation of this report, an interesting paper by Bald et al. was published (45). They reconstituted the mutant F o F 1 -ATPase, which lacks nucleotide binding to the noncatalytic site, into liposomes with bR and examined lightdriven ATP synthesis. Contrarily to the quickly inhibition of ATP hydrolysis activity, the mutant F o F 1 -ATPase synthesized ATP at nearly constant rate up to 60 min. This results further reinforces the similarity of the V o V 1 -ATPase to the mutant F o F 1 -ATPase, which lacks nucleotide binding to the noncatalytic subunit.