Volume 155, number 1 FEBS LETTERS May 1983

Inorganic pyrophosphate-driven ATP-synthesis in liposomes containing membrane-bound inorganic pyrophosphatase and Fo-Fl complex from Rhodospirillum rubrum

Pal NyrCn and Margareta Baltscheffsky

Department of Biochemistry, Arrhenius Laboratory, University of Stockholm, S-106 91 Stockholm, Sweden

Received 14 March 1983

PPi driven ATP synthesis has been reconstituted in a liposomal system containing the membrane-bound energy-linked PPiase and coupling factor complex, both highly purified from Rhodospirillum rubrum. This energy converting model system was made by mixing both enzyme preparations with an aqueous suspension of sonicated soybean phospholipids and subjecting to a freeze-thaw procedure. In the presence ofADP,Mg ‘+ , Pi and PP, the system catalyzed phosphorylation by up to 25 nmol ATP formed x mg -’ x min-‘, at 2O”C, which was sensitive to uncouplers and inhibitors of phosphorylation such as oligomycin, efrapeptin and N,N’-dicyclohexylcarbodiimide.

Pyrophosphatase Liposome Reconstitution A TPase A TP synthesis pump

1. INTRODUCTION We have developed a method for obtaining a highly purified membrane-bound PPase from R. Inorganic pyrophosphatase is a product of rubrum chromatophore [lo]. This pure membrane- photophosphorylation in Rhodospirillum rubrum bound PPase, incorporated into phospholipid chromatophores [l] and is an energy donor for vesicles in the presence of PPi and Mg’+, can func- several energy-linked reactions in R. rubrum tion as a H+-pump. PPi can drive the synthesis of chromatophores including cytochrome reduction ATP from ADP and Pi in R. rubrum [2,3], transhydrogenation [4,5], NAD+ reduction chromatophores [ 111. [6] and the carotenoid shift [7]. The utilization of Here, we have obtained phosphorylation of PPi apparently involves the formation of an ADP to ATP at reasonable rates, by incorporation energized state, also indicated by the fact: that of the membrane-bound inorganic PPase (func- PPi-energized chromatophores of R. rubrum tioning as a ) with the DCCD- generate a membrane potential (A$) and sensitive ATPase complex from R. rubrum into transmembrane pH difference (ApH) [8]; and that pre-formed liposomes. a partly purified membrane-bound PPase from R. rubrum, incorporated in a phospholipid mem- 2. MATERIALS AND METHODS brane, acts as a PPi-dependent electric generator 191. Triton X-100, DTE, FCCP, IDP, MDP, valinomycin, oligomycin, nigericin, Trizma-base, Abbreviations: DCCD, N,N’-dicyclohexylcarbodiimi- cholic acid and asolectin (crude soybean de; DTE, dithioerythritol; FCCP, carbonyl cyanide phospholipids used after an acetone wash [12]) p-trifluoromethoxyphenylhydrazone; IDP, imidodi- were purchased from Sigma. DCCD from Huka phosphate; MDP, methylenediphosphonate AG (Bucks). Luciferin/luciferase assay kit was Enzymes: Membrane-bound inorganic pyrophosphatase from LKB. Efrapeptin D was a kind gift of Pro- (EC 3.6.1.1) and adenosine 5 ’ -triphosphatase (EC fessor B. Beechey (Shell Research, Sittingbourne, 3.6.1.4) Kent).

Published by Elsevier Science Publishers B. V. 00145793/83/$3.00 0 Federation of European Biochemical Societies 125 Volume 155. number 1 FEBS LETTERS May 1983

2.1. Preparation of chromatophores Tris-HCl (pH 7.5) and Hz0 in 2 ml total vol. The Rhodospirillum rubrum strain Sl was grown assay mixture was incubated at 30°C and the reac- anaerobically in the medium described in [13]. tion was terminated after 10 min by addition of After 40 h of growth, cells were harvested, washed 1 ml 10% trichloroacetic acid. Blanks were run and chromatophores were prepared as in [14]. where trichloroacetic acid was added before the sample. Supernatant (1.6 ml, after centrifugation) 2.2. Preparation of the DCCD-sensitive ATPase was used for Pi determination. Pi was assayed col- complex orimetrically as in [ 171. The PPase activity was The ATPase complex was prepared from R. assayed as the ATPase activity with the exception rubrum chromatophores as in [25]. Isolated that 0.5 mM PPi was used instead of ATP. chromatophores were solubilized with final con- centrations of 0.25% (w/v) sodium cholate plus 2.6. ATP-synthesis 15 mM octylglucoside at -8 mg protein/ml for The amount of ATP synthesized was measured 20 min at 4°C. The purification on the sucrose gra- with the luciferin/luciferase technique, 15-50 ~1 of dient was performed in the presence of 0.1% soy- reconstituted ATPase, PPase liposomes were add- bean phospholipids and 0.2% Triton X-100 [15]. ed directly to a test solution containing 200~1 The most active fraction for ATPase activity was luciferin/luciferase assay (LKB kit), 1 ml 0.2 M quickly frozen and was stored at -70°C. Protein glycylglycine (pH 7.8), 20 ~1 100 mM sodium was 0.5 mg/ml. phosphate, 10~1 10 mM ADP and 50~1 10 mM sodium pyrophosphate. The final concentration of 2.3. Preparation of the membrane-bound PPase MgCl2 was 10 mM as the luciferin/luciferase assay The preparation of the membrane-bound PPase contains a high [MgClz]. Addition of inhibitors is from R. rubrum chromatophores was performed as indicated in the text and figures. The reaction as in [lo]. The purified PPase was concentrated by was carried out at room temperature. The resulting ultrafiltration through an Amincon XMlOOA luminescence was measured in an LKB lumino- membrane and was kept frozen at - 70°C. Protein meter 1250. The light output was calibrated by ad- was 0.1-0.2 mg/ml. dition of a known amount of ATP. Protein was determined according to the 2.4. Preparation of ATPase and PPase liposomes modified Lowry method [18]. The freeze-thaw technique of [16] was used to reconstitute the PPase and ATPase into liposomes; 3. RESULTS AND DISCUSSION 40 mg soybean phospholipids were supplemented with 1 ml medium, containing 10 mM Tris-HCl 3.1. PP,-driven A TP formation (pH 7.5), 0.5 mM DTE, 0.5 mM EDTA and The time course of ATP formation in liposomes 0.05% Na-cholate. The suspension was flushed is illustrated in fig.1. With 0.1 pmol PPi the reac- with nitrogen and sonicated in a covered test tube tion was complete in 15 min. At least 50 nmol PPi for 20-25 min at 20°C in a bath-type sonicator must be added to detect phosphorylation. This model G 112 SP lT, Lab. Supplies (Hicksville amount of PPi might give the approx. K,,, for NY). The PPase (0.1 ml desalted preparation [lo], PPase and, thereby, a low rate of PPi hydrolysis. containing lo-20 pg protein) and ATPase (0.1 ml, No reaction was observed if only PPi without containing 50 pg protein) preparations were then liposomes or liposomes without PPi was added to added to 0.5 ml liposomes, sonicated for a few the ATP synthesis assay. ATP hydrolysis was seconds and frozen at -70°C. After thawing at observed with no added PPi (fig.2). The hydrolyz- room temperature, the preparation was stored on ed ATP originated from the ATP contamination ice with only minor losses in activity for at least of the ADP in the sample (0.5-l%). ATP 3 h. hydrolysis could be abolished with oligomycin (fig.2). With ADP, PPi and liposomes present, 2.5. A TPase activity ATP was synthesized at 25 nmol . mg This was assayed in a reaction mixture contain- protein-’ . min-‘. This value can be compared with ing 10 mM MgC12, 1 mM ATP, 1 ml 0.1 M 23 pmol . mg Bchl-’ . h-’ for the chromatophore

126 Volume 155, number 1 FEBS LETTERS May 1983

14Opmol ATP T80pmoL

Fig.1. Time course of ATP synthesis in a suspension of PPase,ATPase liposomes; 25~1 liposomes (-0.07 mg Fig.3. Time course of ATP synthesis as a function of the protein/ml) were suspended in a medium containing amount of liposomes added; (---+ ) 25 ,uI liposomes 1 ml 0.2 M glycylglycine (pH 7.8), 0.2 ml luciferin/luci- added to the same medium as in fig.1 except that 50 ~1 ferase assay, 20~1 100 mM sodium phosphate, 10 pl 10 mM sodium pyrophosphate was present. 10 mM ADP and 504 10 mM sodium pyrophosphate; final [MgClz] was 10 mM; ( -) 25 pl liposomes -30,~M PPi and a corrected value of added. The resulting luminescence was measured in an LKB luminometer 1250. The light output was calibrated 130 PPi/ATP is obtained. by addition of a known amount of ATP. (2) Determining PPi hydrolysis calorimetrically, and ATP synthesis by the luciferase technique:

lmbn This was done under similar conditions, ex- H cept that with the calorimetric assay no phosphate was added; the two rates were com- pared. This method gave 100 PPi/ATP. About 10 PPi/ATP was obtained for the 4Opmol ATP f chromatophore reaction in [lo]. The dif- ference can be explained by assuming that not all liposomes contain both enzymes, so some Fig.2. Time course of ATP hydrolysis in a suspension of of the energy will be wasted. PPase,ATPase liposomes; 50 ,uI liposomes were suspended in the same medium as in fig. 1 except that no 3.2. Effects of energy-transfer inhibitors sodium pyrophosphate was added. At the first arrow The energy-transfer inhibitor DCCD inhibits the 50 ~1 liposomes was added. At the second arrow 50 ~1 membrane-bound PPase of chromatophores to oligomycin (2 mg/ml) was added. -75% [ 191. The titration curves with added DCCP were similar for both PPase and ATPase activities reaction [ll]. If we assume that 1 mg Bchl cor- of chromatophores. However, after purification responds to -40 mg protein, the rate becomes and reconstitution the PPase, but not the ATPase, 9.6 nmol.mg protein-‘.min-‘. As yet, no at- lost the DCCD-sensitivity [14]. So the 50% inhibi- tempts have been made to optimize the conditions tion of phosphorylation in fig.4 might be due to for the phosphorylation reaction; the reaction with the inhibition of the ATPase complex. chromatophores was performed at a higher Oligomycin inhibits energy transfer in temperature. Over the range of concentrations phosphorylation at a site closely connected with tested the phosphorylation rate was proportional the final ATP-forming reaction. Oligomycin is an to the amount of liposomes added (fig.3). The inhibitor of photophosphorylation to ATP [20] stoichiometry of ATP formation to PPi hydrolysis and of the membrane-bound ATPase in chromato- was determined by two different methods. phores 121. However, it does not inhibit photosyn- (1) Adding 0.1 mol PPi to the reaction mixture thetic PPi formation [l], nor does oligomycin in- and detecting the amount of ATP synthesis: hibit membrane-bound PPase activity [2]. Fig.4 This method gave a value of 260 PPi/ATP. shows that oligomycin inhibits phosphorylation in However, the reaction rate leveled off at the liposome system to 80%.

127 Volume 155, number 1 FEBS LETTERS May 1983 $ We infer from the above results that inhibition by DCCD, oligomycin and efrapeptin of liposomal A / phosphorylation is due to the effect on the ATPase complex and not on membrane-bound PPase. _/ 1 3.3. Effects of PPase inhibitors Fluoride is a selective inhibitor of the inorganic pyrophosphatase that has no effect on ATPase or photophosphorylation in R. rubrum chromato- phores [26]. NaF at 10 mM inhibited phosphoryla- tion in liposomes to 1OOVo (fig.5). Methylene diphosphonate is an effective inhibitor of in- organic PPase while not affecting the ATP-linked reactions [ll]. MDP at 0.5 mM inhibited phos- 80pmol ATP _T phorylation to 50% and 0.5 mM IDP. a com-

A C _/1 .-----l

Fig.4. Time course of ATP synthesis in a suspension of PPase,ATPase liposomes; 50 ~1 liposomes were suspended in a medium described in fig.3. At the first J------arrow in A, the liposomes were added; at the second B arrow 10 ~1 10 mM DCCD was added. At the first arrow in B, liposomes were added, at the second arrow 50~1 1 oligomycin (2 mg/ml) was added. At the first arrow in -/ FlUpmol ATP C, liposomes were added; at the second arrow 4,ul 1 efrapeptin (1 mg/ml) was added.

Efrapeptin, a small peptide (16 amino acids long) isolated from the fungus Folypocladium in- flatum [21] inhibits phosphorylation and ATP hydrolysis by the coupling factor ATPases from mitochondria [22], chloroplasts [23] and Escher- ichia coli [21]. Efrapeptin also inhibits ATP hydrolysis and phosphorylation in R. rubrum [24], but has no effect on the adenylate kinase or energy-linked PPase [25]. Efrapeptin is a time- Fig.5. Time course of ATP synthesis in a suspension of PPase,ATPase liposomes; 50 ~1 liposomes were dependent inhibitor. Efrapeptin at 4 pg/ml in- suspended in a medium described in fig.3. At the first hibits the phosphorylation in the liposomal system arrow in A, the liposomes were added; at the second to nearly 100% after 5 min (fig.4). Unlike oligo- arrow 30 ~10.4 NaF was added. At the first arrow in B, mycin, efrapeptin does not inhibit luciferase at liposomes were added; at the second arrow 30 ~120 mM concentrations that inhibit the R. rubrum ATPase MDP was added. At the first arrow in C, liposomes were ]241. added; at the second arrow 30 ~120 mM IDP was added.

128 Volume 155, number 1 FEBS LETTERS May 1983 petitive inhibitor of PPase, decreased phosphory- liposomal PPi-driven ATP synthesis to 50%, lation by 5-fold (fig.5). whereas nigericin, an ionophore of the type that forms an uncharged complex with K+ and 3.4. Effects of uncouplers and ionophores transports this ion in exchange for H+, inhibits Valinomycin, which is a cyclic depsipeptide an- completely (fig.6). The uncoupler FCCP inhibits tibiotic, creates a specific K+ permeability in mem- phosphorylation to 100% (fig.6). This effect is ex- branes. The PPi-driven ATP synthesis in pected if the ATP synthesis is driven by the proton chromatophores was inhibited to nearly 100% by gradient created by the PPase, since FCCP would a combination of valinomycin and NHXl [l 11. A prevent the formation of a pH gradient across the combination of valinomycin and KC1 inhibits the liposomal membrane. Table 1 gives a summary of the effects of dif- ferent compounds tested on the liposomal phosphorylation. The ATPase and PPase activity in the liposomes, measured by the calorimetric A method, under similar conditions as the ex- periments with the luciferase technique, are shown in table 2. Taken in total, these inhibitor-studies illustrate the dependence of the PPase, the coupling factor ATPase and coupled liposomes for the phosphory- lation reaction in the model liposomal system. Liposomes with only the PPase and the ATPase 1 were also examined (not shown) and the result 6 clearly showed the necessity of both enzymes for the phosphorylation reaction. J Purified energy-generating membrane com- _/ ponents like bacteriorhodopsin [27] and photosystem I [28] have been reconstituted f80pmol ATP

Table 1 Effect of different compounds tested on the liposomal phosphorylation

Additions Concen- Phosphorylation tration rate in liposomes __!/ Imin (O70control) H DCCD 80 PM 50 Fig.6. Time course of ATP synthesis in a suspension of Oligomycin 75 pg/ml 20 ATPase,PPase liposomes; 50 ~1 liposomes (-0.07 mg Efrapeptin 3 rg/ml 0 protein/ml) were suspended in a medium containing NaF 10 mM 0 1 ml 0.2 M glycylglycine (pH 7.8), 0.2 ml MDP 0.5 mM 50 luciferin/luciferase assay, 20 ~1 100 mM sodium IDP 0.5 mM 20 phosphate, lOpI 10 mM ADP, 50~1 10 mM sodium Valinomycin pyrophosphate and 204 1 M KCI; final [MgClz] was (15 mM KCI) 4 rM 50 10 mM. At the first arrow in A, liposomes were added; Nigericin at the second arrow 5 pl 1 mM valinomycin was added. (15 mM KCI) 15 gM 0 At the first arrow in B, liposomes were added; at the FCCP 3.0 pM 0 second arrow 24 10 mM nigericin was added. At the first arrow in C, liposomes were added; at the second 504 liposomes were assayed for ATP-synthesis as in arrow 3 pl 1 mM FCCP was added. section 2, with different inhibitors present

129 Volume 155, number 1 FEBS LETTERS May 1983

Table 2 I51 Keister, D.L. and Yike, N.J. (1967) Biochemistry 6, 3847-3857. The PPase and ATPase activity in liposomes 161 Keister, D.L. and Yike, N.J. (1967) Arch. Bio- them. Biophys. 121, 415-422. ATPase and PPase activity I71 Baltscheffsky, M. (1969) Arch. Biochem. Biophys. (umol PPi or ATP 130, 646-652. hydrolyzed. mg-’ . min-‘) PI Barsky, E.L., Bench-Osmolovskaya, E.A., Ostroamov, S.A., Samuilov, V.D. and Skulachev, PPase activity 0.590 V.P. (1975) Biochim. Biophys. Acta 387, 388-395. + 3 pM FCCP 0.800 I91 Kondroskin, A.A., Remennikov, V.G., Samuilov, ATPase activity 0.150 V.D. and Skulachev, V.P. (1980) Eur. J. Biochem. + 3 PM FCCP 0.690 113, 219-222. 1101 Nyren, P., Hajnal, K. and Baltscheffsky, M. (1983) 20~1 and 140~1 liposomes were used for PPase and submitted. ATPase activity measurements, respectively. The illI Keister, D.L. and Minton, N.J. (1971) Arch. Bio- measurements were done under the same conditions as them. Biophys. 147, 330-338. ATP-synthesis measurements and the amount of Pi WI Kagawa, Y. and Racker, E. (1971) J. Biol. Chem. realized was measured calorimetrically (see section 2 for 246, 5477. details) u31 Bose, S.K., Gest, H. and Ormerod, J.G. (1961) J. Biol. Chem. 236, 13-14. [I41 Shakhov, Y., Nyren, P. and Baltscheffsky, M. (1982) FEBS Lett. 146, 177-180. together with purified CFe-Ft into liposomes. In 1151 Pick, U. and Racker, E. (1979) J. Biol. Chem. 254, these cases the membrane was energized by light- 2793-2799. driven proton transport, and coupled ATP syn- I161 Kasakaro, M. and Hinkle, R.C. (1977) J. Biol. Chem. 252, 7384-7390. thesis was observed. In a model system we have us- Rathbun, W.B. and Betlach, W.M. (1969) Analyt. ed a chemically-driven (PPi) proton-pump for v71 Biochem. 28, 436-445. energization of the membrane and coupled the ml Peterson, G.L. (1977) Anal. Biochem. 83, energy liberated by PPi hydrolysis to ATP 346-356. synthesis. [I91 Batlscheffsky, M., Baltscheffsky, H. and Boork, J. (1982) in: Electron Transport and Photophosphorylation (Barber, J. ed) Top. ACKNOWLEDGEMENTS Photosynth., vol.4, pp.249-272, Elsevier, This work was supported by a grant from the Amsterdam, New York. Swedish Nature Science Research Council to M.B. WI Baltscheffsky, H. and Baltscheffsky, M. (1960) Acta Chem. Stand. 14, 257. The authors thank I. PontCn for preparation of the WI Linnett, P.E. and Beechey, R.B. (1979) Methods DCCD-sensitive ATPase. Enzymol. 55, 493-495. WI Susa, J.B. and Lardy, H.A. (1975) Mol. Pharmacol. 11, 166-173. REFERENCES t231 Lucero, H.A., Rovizzini, R.A. and Vallejos, R.H. (1976) FEBS Lett. 68, 141-146. PI Baltscheffsky, H., Von Stedingk, L.-V., Heldt, H.- t241 Lucero, H., Lescano, W.I.M. and Vallejos, R.H. W. and Klingenberg, M. (1966) Science 153, (1978) Arch. Biochem. Biophys. 186, 9-14. 1120-l 124. [251 Lundin, A., Lundin, U.K. and Baltscheffsky, M. PI Baltscheffsky, M., Baltscheffsky, H. and Von (1979) Acta Chem. Stand. B33, 608-609. Stedingk, L.-V. (1967) Brookhaven Symp. Biol. 19, WI Keister, D.L. and Minton, N.J. (1971) Biochem. 246. Biophys. Res. Commun. 42, 932-939. I31 Baltscheffsky, M. (1967) Biochem. Biophys. Res. v71 Winget, G.D., Ranner, N. and Racker, F. (1977) Commun. 28, 270-276. Biochim. Biophys. Acta 460, 490-499. [41 Keister, D.L. (1967) Brookhaven Symp. Biol. 19, WI Hauska, G., Samoray, D., Orlich, G. and Nelson, 255. N. (1980) Eur. J. Biochem. 111, 535-543.

130