Proteins required for the binding of mitochondrial ATPase to the mitochondrial inner membrane

1. Isolated F₁ (mitochondrial ATPase) binds to urea-treated submitochondrial particles suspended in sucrose/Tris/EDTA with a dissociation constant of 0.1 μM.

2. About one-third of the F₁ and the oligomycin-sensitivity conferring protein (OSCP) are lost during preparation of submitochondrial particles prepared at high pH (A particles). None is lost from particles treated with trypsin (T particles).

3. After further treatment with alkali of urea-treated particles, binding of F₁ requires the addition of OSCP. Maximum binding is reached when both OSCP and Fc₂ are added. The concentration of F₁-binding sites in the presence of both OSCP and Fc₂ is about the same as that in TU particles.

4. After further extraction with silicotungstate of urea- and alkali-treated particles, OSCP no longer induces binding of F₁, unless Fc₂ is also present. Fc₂ induces binding in the absence of OSCP but with a lower binding constant and, in contrast to results under all the other conditions studied in this paper, the ATPase activity is oligomycin insensitive.

5. It is tentatively concluded that OSCP is the binding site for F₁ and Fc₂ is the binding site for OSCP.     

Studies on the activation of rat liver pyruvate dehydrogenase and 2-oxoglutarate dehydrogenase by adrenaline and glucagon. Role of increases in intramitochondrial Ca2+ concentration.

The association of different phospholipids with a lipid-depleted oligomycin-sensitive ATPase from bovine cardiac mitochondria [Serrano, Kanner & Racker (1976) J. Biol. Chem. 251, 2453-2461] has been examined using three approaches. First, reconstitution of the ATPase with different synthetic diacyl phospholipids resulted in a 2-10-fold stimulation of ATPase specific activity depending upon the particular phospholipid employed. The phospholipid headgroup region displayed the following order of ATPase reactivation potential: dioleoylphosphatidylglycerol greater than dioleoylphosphatidic acid greater than dioleoylphosphatidylcholine. Furthermore, the ATPase showed higher levels of specific activity when reconstituted with dioleoyl phospholipid derivatives compared with dimyristoyl derivatives. Second, examination of the phospholipid remaining associated with the lipid-depleted ATPase upon purification showed that phosphatidylcholine, phosphatidylethanolamine, and diphosphatidylglycerol were present. No relative enrichment of any of these phospholipids (compared with their distribution in submitochondrial particles) was noted. Therefore, no preferential association between the ATPase and any one phospholipid could be found in the mitochondrial ATPase. Third, the sodium cholate-mediated phospholipid exchange procedure was employed for studying the phospholipid requirements of the ATPase. Replacement of about 50% of the mitochondrial phospholipid remaining with the lipid-depleted ATPase could be achieved utilizing either synthetic phosphatidic acid or phosphatidylcholine. Examination of the displaced mitochondrial phospholipid showed that phosphatidylcholine, phosphatidylethanolamine, and diphosphatidylglycerol were replaced with equal facility.     

Purification and properties of ATPase inhibitor from rat liver mitochondria

(1) The ATPase inhibitor protein has been isolated from rat liver mitochondria in purified form. The molecular weight determined by sodium dodecyl sulfate gel electrophoresis is approximately 9500, and the isoelectric point is 8.9.

(2) The protein inhibits both the soluble ATPase and the particle-bound ATPase from rat liver mitochondria. It also inhibits ATPase activities of soluble F₁, and inhibitor-depleted submitochondrial particles derived from bovine heart mitochondria.

(3) On particle-bound ATPase the inhibitor has its maximal effect if incubated in the presence of Mg²⁺. ATP at slightly acidic pH.

(4) The inhibitor has a minimal effect on P_i-ATP exchange activity in sonicated submitochondrial particles. However, unexpectedly the inhibitor greatly stimulates P_i-ATP exchange activity in whole mitochondria while the low ATPase activity of the mitochondria is not affected. The possible mechanism of action of the inhibitor on intact mitochondria is offered.     

F1-ATPase from different submitochondrial particles.

1. F1-ATPase has been extracted by the diphosphatidylglycerol procedure from mitochondrial ATPase complexes that differ in ATPase activity, cold stability, ATPase inhibitor and magnesium content. 2. The ATPase activity of the isolated enzymes was dependent upon the activity of the original particles. In this respect, F1-ATPase extracted from submitochondrial particles prepared in ammonia (pH 9.2) and filtered through Sephadex G-50 was comparable to the enzyme purified by conventional procedures (Horstman, L.L. and Racker, E. (1970) J. Biol. Chem. 245, 1336--1344), whereas F1-ATPase extracted from submitochondrial particles prepared in the presence of magnesium and ATP at neutral pH was similar to factor A (Andreoli, T.E., Lam, K.W. and Sanadi, D.R. (1965) J. Biol. Chem. 240, 2644--2653). 3. No systematic relationship has been found in these F1-ATPase preparations between their ATPase inhibitor content and ATPase activity. Rather, a relationship has been observed between this activity and the efficiency of the ATPase inhibitor-F1-ATPase association within the membrane. 4. It is concluded that the ATPase activity of isolated F1-ATPase reflects the properties of original ATPase complex provided a rapid and not denaturing procedure of isolation is employed.     

Physiological and genetic modifications of the expression of the yeast mitochondrial adenosine triphosphatase inhibitor

1. The oligomycin-sensitive ATPase activity of submitochondrial particles of the glycerol-grown “petite-negative” yeast: Schizosaccharomyces pombe is markedly stimulated by incubation at 40°C and by trypsin activations are treatment. Both increased in Triton-X 100 extracts of the submitochondrial particles.

2. A trypsin-sensitive inhibitory factor of mitochondrial ATPase with properties similar to that of beef heart has been extracted and purified from glycerolgrown and glucose-grown S. pombe wild type, from the nuclear pleiotropic respiratory-deficient mutant S. pombe M126 and from Saccharomyces cerevisiae.

3. ATPase activation by heat is more pronounced in submitochondrial particles isolated from glycerol-grown than from glucose-grown S. pombe. An activation of lower extent is observed in rat liver mitochondrial particles but is barely detectable in the “petite-positive” yeast: S. cerevisiae. No activation but inhibition by heat is observed in the pleitotropic respiratory-deficient nuclear mutant S. pombe M126.

4. The inhibition of S. pombe ATPase activity by low concentrations of dicyclohexylcarbodiimide dissapears at inhibitor concentrations above 25 μM. In Triton-extract of submitochondrial particles net stimulation of ATPase activity is observed at 100 μM dicyclohexylcarbodiimide. The pattern of stimulation of ATPase activity by dicyclohexylcarbodiimide in different genetic and physiological conditions parallels that produced by heat and trypsin. A similar mode of action is therefore proposed for the three agents: dissociation or inactivation of an ATPase inhibitory factor.

5. We conclude that “petite-positive” and “petite-negative” yeasts contain an ATPase inhibitor factor with properties similar to those of the bovine mitochondrial ATPase inhibitor. The expression of the ATPase inhibitor, measured by ATPase activation by heat, trypsin or high concentrations of dicyclohexylcarbodiimide, is sensitive to alterations of the hydrophobic membrane environment and dependent on both physiological state and genetic conditions of the yeast cells.     

The number and localisation of adenine nucleotide-binding sites in beef-heart mitochondrial ATPase (F1) determined by photolabelling with 8-azido-ATP and 8-azido-ADP

1. When irradiated 8-azido-ATP becomes covalently bound (as the nitreno compound) to beef-heart mitochondrial ATPase (F₁) as the triphosphate, either in the absence or presence of Mg²⁺, label covalently bound is not hydrolysed.

2. In the presence of Mg²⁺ the nitreno-ATP is bound to both the and β subunits, mainly (63%) to the subunits.

3. After successive photolabelling of F₁ with 8-azido-ATP (no Mg²⁺) and 8-azido-ADP (with Mg²⁺) 4 mol label is bound to F₁, 2 mol to the and 2 mol to the β subunits.

4. When the order of photolabelling is reversed, much less 8-nitreno-ATP is bound to F₁ previously labelled with 8-nitreno-ADP. It is concluded that binding to the -subunits hinders binding to the β subunits.

5. F₁ that has been photolabelled with up to 4 mol label still contains 2 mol firmly bound adenine nucleotides per mol F₁.

6. It is concluded that at least 6 sites for adenine nucleotides are present in isolated F₁.     

Further studies on F1-ATPase inhibition by local anesthetics

We have measured the inhibitory potencies of several local anesthetics (procaine, lidocaine, tetracaine and dibucaine) and related compounds (chlorpromazine, procainamide and propranolol) on the ATPase activities of bovine heart submitochondrial particles and purified F₁ extracted from these particles. All of these agents cause inhibition of ATPase in F₁ as well as in submitochondrial particles. A linear relationship is found between the log of the octanol/water partition coefficients and the log of the concentrations required for 50% inhibition of F₁. Sedimentation velocity ultracentrifugation and polyacrylamide gel electrophoresis showed that 1.0 mM tetracaine caused partial dissociation of the F₁ complex. Complete reversibility of the enzyme inhibitory effects was demonstrated, however. This work shows that local anesthetics can affect protein structure and enzyme activity without the mediation of lipid.     

Respiration-dependent uncoupler-stimulated ATPase activity in castor bean endosperm mitochondria and submitochondrial particles

1. The uncoupler-stimulated ATPase activity of castor bean endosperm mitochondria and submitochondrial particles has been studied. The rate of ATP hydrolysis catalyzed by intact mitochondria was slow and little enhanced by addition of uncouplers at the concentration required for uncoupling the oxidative phosphorylation. ATPase activity was stimulated at higher concentrations of uncouplers.

2. 1-Anilinonaphthalene 8-sulfonate fluorescence was decreased when the mitochondria were oxidizing succinate. Carbonylcyanide-p-trifluoromethoxyphenylhydrazone and antimycin reversed the succinate-induced fluorescence diminution. ATP did not induce the fluorescence response.

3. The addition of succinate, NADH or ascorbate/N,N,N′,N′-tetramethyl-p-phenylenediamine as electron donor induced high ATPase activity in the presence of low concentrations of uncouplers. Stimulating effect of uncouplers was completely abolished by further addition of antimycin.

4. Submitochondrial particles were prepared by sonication. The particles catalyzed a rapid hydrolysis of ATP and carbonylcyanide-p-trifluoromethoxyphenylhydrazone at 10^-8 M did not stimulate the ATPase activity. Addition of succinate induced uncoupler-stimulated ATPase activity. The effect of succinate was completely abolished by further addition of antimycin.

5. The treatment of submitochondrial particles by trypsin or high pH also induced uncoupler-stimulated ATPase activity.

6. The above results were interpreted to indicate that ATPase inhibitor regulated the back-flow reaction of mitochondrial oxidative phosphorylation.     

Reconstitution of mitochondrial F0.F1-ATPase with phosphatidylcholine using the nonionic detergent, octylglucoside

A reconstitution procedure has been developed for the incorporation of the mitochondrial F0.F1-ATPase into the bilayer of egg phosphatidylcholine vesicles. The nonionic detergent, octylglucoside, egg phosphatidylcholine, and the lipid-deficient, oligomycin-sensitive F0.F1-ATPase (Serrano, R., Kanner, B., and Racker, E. (1976) J. Biol. Chem. 251, 2453-2461) were combined in a 4770:320:1 detergent/phospholipid/protein molar ratio and then centrifuged on a discontinuous sucrose gradient to isolate the F0.F1-phosphatidylcholine complex. The specific activity of the reconstituted F0.F1-ATPase was as high as 14.5 mumol/min/mg protein, whereas with no added lipid the activity ranged between 1.4 and 2.2 mumol/min/mg protein. This reconstituted preparation exhibited greater than 90% oligomycin sensitivity which demonstrated the intactness of the multisubunit enzyme complex. The phosphatidylcholine/protein molar ratio of the reconstituted F0.F1 was 250:1 with less than 0.4% of the added octylglucoside remaining. Titrations with both phosphatidylcholine and octylglucoside demonstrated that the specific activity and oligomycin sensitivity were highly dependent on the concentrations of both phospholipid and detergent in the original reconstitution mixture. Analysis of the reconstituted ATPase by electron microscopy demonstrated that the catalytic portion of the enzyme complex projected from the phospholipid bilayer with an orientation similar to that observed with submitochondrial particles. The F0.F1-phosphatidylcholine complex was able to trap inulin, which suggests a vesicular structure impermeable to macromolecules. The electrophoretic mobility of the complex was identical to that for liposomes of egg phosphatidylcholine alone. The reconstitution conditions utilized give rise to an enzyme-phospholipid complex with very low ionic charge that demonstrates high oligomycin-sensitive ATPase activity.     

The binding of aurovertin to mitochondria, and its effect on mitochondrial respiration

1. The fluorescence of aurovertin increases about 100-fold on binding to sub-mitochondrial particles.

2. The mitochondrial ATPase (F₁) binds one mole aurovertin/mole F₁ with a dissociation constant of 6·10⁻⁸ M.

3. The fluorescence of mitochondrion-bound aurovertin is maximal during State-3 respiration and is partially quenched on anaerobiosis, addition of respiratory inhibitor, oligomycin or uncoupler, or transition to State 4. This quenching is still present when the binding site is saturated with aurovertin, showing that the quantum yield of fluorescence is lowered.

4. Aurovertin is bound co-operatively to State-3 mitochondria.

5. The curve relating inhibition of State-3 respiration to aurovertin concentration is more sharply sigmoidal than the binding curve.

6. An analysis of the binding and inhibition data leads to the conclusion that aurovertin induces a conformation change in the binding site on F₁ in two ways: (i) directly by acting as an allosteric effector of an oligomeric system, (ii) indirectly by inhibiting State-3 respiration which changes the allosteric constant of the oligomeric system.

7. The concentration of the aurovertin-binding site in both rat-liver and rat-heart mitochondria is about the same as that of the antimycin-binding and oligomycin-binding sites.     

The mechanism of inhibition by oligomycin of oxidative phosphorylation in mitochondria

1. The concentration of specific oligomycin-binding sites in rat-liver mitochondria is 0.12 nmole/mg protein, whereas at least 10-times more oligomycin can be bound non-specifically.

2. The activity of oligomycin-inhibited processes in intact mitochondria and submitochondrial particles cannot be restored by treatment with egg lecithin or mitochondrial phospholipids.

3. Analysis of the kinetics of inhibition of State-3 respiration by oligomycin reveals that (i) after a certain lag period the inhibition by oligomycin is pseudo-first order with respect to the respiratory-control ratio, defined as the ratio of the respiratory rate at time t to that of the final inhibited site, (ii) the value of the pseudo-first-order rate constant (k₀) is dependent on the oligomycin: protein ratio, phospholipid: protein ratio, pH and temperature, (iii) the effects of various substrates and inhibitors of electron transfer on the kinetics of oligomycin inhibition can be explained by their effects on respiratory control.

4. A detailed model is proposed for the interaction of oligomycin with mitochondria. It is proposed that two conformations of the oligomycin-sensitive site are present, and that oligomycin specifically binds to the conformation that is involved in the induction of respiratory control.     

Studies on the activation of purified mitochondrial ATPase by phospholipids

1. An ATPase which is activated by phospholipids and inhibited by oligomycin, has been purified from beef heart submitochondrial particles using affinity chromatography. Phospholipid and detergent are removed by washing the enzyme with a solution of serum albumin while it is attached to the biospecific adsorbent.

2. The ATPase is activated up to 18-fold by lysolecithin and to a smaller extent by cardiolipin, phosphatidylinositol and phosphatidylethanolamine. The amount required of each of these phospholipids to give half-maximal activation is apparently inversely related to the number of fatty acid chains in the lipid. Lecithin, which is a poor activator of the ATPase, competitively inhibits the activation by cardiolipin.

3. The activation of the ATPase consists of an increase in both the maximal velocity of the reaction and the affinity for substrate ATP. The pH optimum of the reaction is not influenced by the charge of the lipid.

4. Arrhenius plots of ATPase activated with lysolecithin show a transition to a higher activation energy at temperatures below 19 °C. The sensitivity of the lysolecithin-activated enzyme to oligomycin is markedly reduced below the same temperature. With cardiolipin the transition is observed at 13 °C.

5. ADP, Mg²⁺ and to a smaller extent ATP, Mg²⁺ enhance the activation of ATPase by suboptimal amounts of phospholipid.     

Restoration of ATP synthesis in urea-treated membranes prepared from pea cotyledon mitochondria

Submitochondrial particles were prepared from pea cotyledon mitochondria by sonication in a medium containing 5 mM MgCl₂. The resulting particles (Mg²⁺-submitochondrial particles) catalyzed oxidative phosphorylation at the rate of 100–200 nmol ATP formed / min per mg protein. Treatment of Mg²⁺-submitochondrial particles with 3.0 M urea resulted in a preparation of highly resolved particles with low ATPase activity and no capacity for oxidative phosphorylation. However, the resulting membranes were not capable of reconstitution of oxidative posphorylation with the purified mitochondrial F₁-ATPase. Urea particles capable of reconstitution of oxidative phosphorylation could be prepared by extracting Mg²⁺-submitochondrial particles with concentrations of urea ranging from 1.7 to 2.0 M. We have used 1.9 M urea for large-scale preparation of urea particles that could be stored in liquid nitrogen without any loss of reconstitution capacity. The residual oxidative phosphorylation rate of these particles was 6–8 nmol ATP / min per mg protein and this rate could increase to 60–70 nmol ATP / min per mg protein on incubation with saturating amounts of purified mitochondrial F₁-ATPase. In contrast to the mitochondrial F₁, purified activated pea chloroplast CF₁ was unable to stimulate ATP synthesis in 1.9 M urea particles.     

Characterization of a mutant of Schizosaccharomyces pombe lacking cytochrome b-566

1. A chromosomal respiration-deficient mutant of the petite-negative yeast Schizosaccharomyces pombe was isolated. Its mitochondria show respiration rates of about 7% of the wild-type respiration with NADH and succinate as substrate, and 45% with ascorbate in the presence of tetramethyl-p-phenylenediamine. Oxidation of NADH and succinate is insensitive to antimycin and cyanide and that of ascorbate is much less sensitive to cyanide than the wild type.

2. The amounts of cytochromes c₁ and aa₃ are similar in the mutant and wild type. Cytochrome b-566 could not be detected in low-temperature spectra after reduction with various substrates or dithionite. A b-558 is, however, present.

3. The b-cytochromes in the mutant are not reduced by NADH or succinate during the steady state even after addition of ubiquinone-1. QH₂-3: cytochrome c reductase activity is very low and succinate oxidation is highly stimulated by phenazine methosulphate.

4. Antimycin does not bind to either oxidized or reduced mitochondrial particles of the mutant.

5. In contrast to the b-cytochromes of the wild type, b-558 in the mutant reacts with CO.

6. Cytochromes aa₃, c and c₁ are partly reduced in aerated submitochondrial particles isolated from the mutant and the EPR signal of Cu (II), measured at 35°K, is detectable only after the addition of ferricyanide. In the mutant, a signal with a trough at g = 2.01 is found, in addition to the signal at g = 1.98 found in the wild type.

7. The ATPase activity of particles isolated from the mutant is much lower than in the wild type but is still inhibited by oligomycin.     

Mitochondrial H+-ATPase activation by an amine oxide detergent

Lauryl dimethylamine oxide activates ATP hydrolysis by the mitochondrial H+-ATPase. Activation is observed in systems with a high content of inhibitor protein as described by Pullman and Monroy (Pullman, M.E., and Monroy, G.C. (1963) J. Biol. Chem. 238, 3762-3769), i.e. Mg-ATP submitochondrial particles and a Triton X-100-solubilized H+-ATPase from the same particles. Detergent activation of ATP hydrolysis is also present in inhibitor-reconstituted systems, i.e. submitochondrial particles, Triton extracts, and soluble F1-ATPase. In submitochondrial particles depleted of inhibitor protein, lauryl dimethylamine oxide induced a biphasic response which is characterized by a drop-in activity induced by relatively low concentrations of LDAO; at higher concentrations the detergent activates to an extent never greater than the initial activity. In inhibitor protein-depleted oligomycin-sensitive Triton extracts, lauryl dimethylamine oxide stimulates ATP hydrolysis to very high values (30 mumol min-1 mg-1). These findings suggest that in addition to the inhibitor protein ATP hydrolysis is controlled by other subunit interactions.     

Rhodamine 6G, inhibitor of both H-ejections from mitochondria energized with ATP and with respiratory substrates

Rhodamine 6G inhibited ATP hydrolysis by oligomycin-sensitive ATPase, purified from rat liver mitochondria, in good accord with the dose-response curve for its inhibition of energy transduction of ATP synthesis in mitochondria, but it did not inhibit ATP hydrolysis by purified F₁. Rhodamine 6G also inhibited both H⁺-ejections from mitochondria energized with respiratory substrates and with ATP.

The present findings show that the inhibitory effect of rhodamine 6G on energy transduction is not due to a modification of the transport system for adenine nucleotides, P_i, and respiratory substrates, and that the inhibition sites of rhodamine 6G are on components related with H⁺-ejection by redox components and also on F₀.     

F1-ATPase of Micrococcus lysodeikticus is not a glycoprotein

It has been claimed (Andreu, J.M., Warth, R. and Muñoz, E. (1978) FEBS Lett. 86, 1–5) that the F₁-ATPase of Micrococcus lysodeikticus is a glycoprotein containing mannose and glucose as the principal sugars. Even after extensive purification of M. lysodeikticus F₁-ATPase by DEAE-Sephadex A25 chromatography, carbohydrate contents varying from 2.7 to 10.8% have been found. Concanavalin A-reactive components corresponding to the succinylated lipomannan have been detected and separated from the ATPase in purified F₁ preparations by immunoelectrophoresis (rocket and two-dimensional) through agarose gels containing concanavalin A. Passage of the purified F₁-ATPase through concanavalin A-Sepharose 4B columns removed the carbohydrate component(s) without loss of the specific activity of the ATPase. Mannose was the only sugar detectable by gas-liquid chromatography of the F₁-ATPase before Con A-Sepharose 4B chromatography and it was completely eliminated after chromatography. No qualitative or quantitative changes in the subunit (, β, γ, δ and ε) profiles were detectable when the sodium dodecyl sulfate polyacrylamide gels were scanned by densitometry of F₁-ATPase before and after Con A-Sepharose 4B chromatography. We conclude that there is no evidence of carbohydrate covalently linked to this F₁-ATPase and that this membrane protein is not a glycoprotein. The presence of carbohydrate is attributable to contamination with lipomannan.     

A light- and cysteine-activated ATP-Pi exchange reaction in chloroplasts and its relationship to ATPase and photophosphorylation

1. ‘Broken’ chloroplasts from spinach if illuminated for a period in the presence of cysteine and phenazine methosulphate develop an ATP-P_i exchange activity which can be observed in the dark. The conditions giving rise to ATP-P_i exchange activity are similar to those giving rise to the thiol-activated light-triggered ATPase.

2. ATP is not necessary during illumination for development of ATP-P_i exchange activity, but the activity declines if a period of time elapses between illumination and addition of ATP. This is accompanied by a similar decline in the cysteine-activated light-triggered ATPase.

3. The ATP-P_i exchange and ATPase show the same dependence on ATP concentration and are both inhibited by added ADP.

4. Both reactions are inhibited by Dio-9.

5. Desaspidin, 4-octyl-2,6-dinitrophenol and carbonyl cyanide 4-trifluoromethoxyphenylhydrazone, added immediately after illumination, inhibit the ATP-P_i exchange. The ATPase is initially stimulated under these conditions and then inhibited. If present during illumination, desaspidin and octyldinitrophenol inhibit the ATPase.

6. It is concluded that the ATP-P_i exchange reaction and the ATPase are activities of the same enzyme complex in the chloroplast and that this is probably part of the terminal enzyme system of photophosphorylation.     

Coupling factor activity of the purified pea mitochondrial F1-ATPase

The pea cotyledon mitochondrial F₁-ATPase was released from the submitochondrial particles by a washing procedure using 300 mM sucrose /2 mM Tricine (pH 7.4). The enzyme was purified by DEAE-cellulose chromatography and subsequent sucrose density gradient centrifugation. Using polyacrylamide gel electrophoresis under non-denaturing conditions, the purified protein exhibited a single sharp band with slightly lower mobility than the purified pea chloroplast CF₁-ATPase. The molecular weights of pea mitochondrial F₁-ATPase and pea chloroplast CF₁-ATPase were found to be 409 000 and 378 000, respectively. The purified pea mitochondrial F₁-ATPase dissociated into six types of subunits on polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate. Most of these subunits had mobilities different from the subunits of the pea chloroplast CF₁-ATPase. The purified mitochondrial F₁-ATPase exhibited coupling factor activity. In spite of the observed differences between CF₁ and F₁, the mitochondrial enzyme stimulated ATP formation in CF₁-depleted pea chloroplast membranes. Thus, the mitochondrial F₁ was able to substitute functionally for the chloroplast CF₁ in reconstituting photophosphorylation.