Hoppe-Seyler's Z. Physiol. Chem. Bd. 357, S. 1605 - 1622, November 1976

Mitochondrial Adenosinetriphosphatase from Yeast, Saccharomyces cerevisiae Purification, Subunit Structure and Kinetics Koichiro TAKESHIGE, Benno HESS, Margret B OHM, and Hildegard ZIMMERMANN-TELSCHOW Max-Planck-Institut f r Ern hrungsphysiologie, Dortmund (Received 20 July/26 October 1976)

Summary: 1. A procedure for the purification of ATPase extracted by chloroform from baker's yeast (Saccharomyces cerevisiae) is reported. The yield based on submitochondrial particles was 55% and the purification was 100-fold. The isolated complex was homogeneous as assessed by gel filtration, ion-exchange chromatography, sedimentation in sucrose gradient and in the analytical ultracentrifuge. The molecular weight determined by gel filtration was 400000 ± 20000. Ultracentrifugation yielded s50>w = 12.50 ± 0.13 S and the laser light scattering study gave a diffusion coefficient of£>20)W = 2.92 χ 10-7 cm2 s~ J . The amino acid composition as well as absorption, fluorescence, and circular dichroism spectra, from which the helicity of 39% was evaluated, are given. 2. On polyacrylamide gel electrophoresis in the presence of sodium dodecylsulfate, six components with molecular weights of 58500 (a),

55000 (ft, 42000, 34000(γ), 10000 (δ), and 8600 (e) were observed with a stoichiometry of 3:3:1:1:1:1. The amino acid composition is given for α + β and 7 as well as δ and e components. 3. The maximum specific activity of the enzyme was 200 U/mg under the optimum conditions. The enzyme was inactivated by incubation at 0 °C and strongly inhibited by the antibiotic Dio-9 but not by oligomycin and./V,Af'-dicyclohexylcarbodiimide. The effects of kinetic parameters and anions on the enzyme are reported. Two active sites for Mg-ATP with Km values of 0.045mM and 0.37mM and a single active site for Mg-ITP with Km = 0.45OmM and for MgGTP with Km = 0.179mM were found. A study of the temperature dependence of the maximum activity revealed a straight line in the Arrhenius plots with an activation energy of 11.0 kcal/mol (= 46 kJ/mol).

Enzymes: Adenosinetriphosphatase (ATPase), ATP phosphohydrolase (EC 3.6.1.3); Alcohol dehydrogenase, alcohol:NADe oxidoreductase (EC 1.1.1.1); Catalase, hydrogen-peroxide:hydrogenperoxide oxidoreductase (EC 1.11.1.6); Lactate dehydrogenase, L-lactate:NAD® oxidoreductase (EC 1.1.1.27); Pyruvate kinase, ATP:pyruvate 2-O-phosphotransferase (EC 2.7.1.40); Abbreviation: HEPPS = Λ^-2-hydroxyethylpiperazineW-2-propane sulfonic acid.

Brought to you by | The University of A Authenticated Download Date | 6/17/15 4:5

1606

K. Takeshige, B. Hess, M. Böhm and H. Zimmermann-Telschow

Bd. 357 (1976)

MitochondrialeAdenosintriphosphatase aus Hefe, Saccharomyces cerevisiae: Reinigung, Struktur der Untereinheiten und Kinetik Zusammenfassung:

1. Es wird über ein Chloroformextraktions-Reinigungsverfahren (l00fach) von Hefe-ATPase bei SOproz. Ausbeute berichtet. Der Komplex ist bei Gelfiltration, lonenaustauschchromatographie, Sedimentation im Sacchaiosegradienten sowie in der analytischen Ultrazentrifuge homogen. Das Molekulargewicht von 400000 ± 20000 wurde durch Gelfiltration bestimmt. Ultrazentrifugation ergab s°o,w = 12.50 ± 0.13 S sowie Laserlichtstreuungsanalysen einen Diffusionskoeffizienten von£>2o,w = 2.92 10~7 cm2 s"1. Weiter werden die Aminosäure-Zusammensetzung, Absorptions-, Fluoreszenz- und CD-Spektren mitgeteilt. Aus dem CD-Spektrum wird ein helikaler Anteil von 39% vermittelt. 2. Polyacrylamid-Gel-Elektrophorese in Gegenwart von Dodecylsulfat ergibt 6 Komponenten mit Molekulargewichten von 58 500 (a),

55000 (0), 42000,34000 ( ), 10000 ( ) und 8600 (e) mit einer Stöchiometrie von 3:3:1:1:1:1. Für die a- +13-, - sowie 5- und e-Komponenten wird die Aminosäurezusammensetzung angegeben. 3. Unter optimalen Bedingungen findet sich eine maximale spezifische Aktivität von 200 U/mg. Das Enzym wird bei Inkubation bei 0 °C inaktiviert und durch das Antibiotikum Dio-9, jedoch nicht durch Oligomycin und./V,-/V'-Dicyclohexylcarbodiimid, inhibiert. Kinetische Parameter und Anioneneffekte werden mitgeteilt. Es fanden sich zwei Bindungsstellen für Mg-ATP mit MichaelisKonstanten von 0.045mM und 0.37mM sowie einer Bindungsstelle für Mg-ITP mit Km = 0.45mM und Mg-GTP mit Km = 0.179mM. Die Temperaturabhängigkeit der Maximalaktivität ergab eine lineare Beziehung nach Arrhenius mit einer Aktivierungsenergie von 11.0 kcal/mol (= 46 kJ/mol).

Key words: Adenosinetriphosphatase, yeast, subunits, amino acid composition, kinetics

The mechanism of oxidative phosphorylation is our results are compared with the properties of linked to the structure and function of ATPase in ATPase from other sources. the mitochondrial inner membrane and to its interactions with other components of the reMaterial and Methods spiratory pathway. These aspects led to the extraction and purification of the enzyme by difMaterial. All nucleotides and auxiliary enzymes were ferent procedures from various sources!1"7] with obtained from Boehringer Mannheim GmbH. Buffers variations in the specific activity of the final and other salts were of analytical grade. The reagents products and in the subunit structure^8"11!. In for electrophoresis were obtained from Serva, Heidelberg and DEAE-cellulose DE 23 from Vetter KG, Wiesloch, the course of a study of solvent extraction techniques, we obtained a highly active purified prep- Sepharose 6 B and Sephadex G-50 from Pharmacia, Uppsala. Ultrogel AcA22 and Ultrogel AcA34 were obtained aration by a chloroform extraction procedure, from LKB, Sweden. Lecithin and ferritin were purchased which recently was also applied to the purificafrom Merck AG, Darmstadt. Dio-9 was a product of tion of ATPase from beef heart submitochondrial Royal Netherlands Fermentation Industries Ltd, Holparticles yielding, however, a product with low land. HEPPS (A^-2-hydroxyethylpiperazineW-2-prospecific activity 1121. panesulfonic acid) was obtained from Calbiochem. Here we report the extraction by chloroform and Assay of ATPase activity. The activity was measured as purification procedure as well as some properties described by Pullman et alJ13l and modified by Ebel of the purified enzyme compared with the prop- and Lardyl14'. The measurements were carried out at erties of three other forms of ATPase from yeast, 25 °C in a total volume of 1.0 ml containing 50mM the sonicated, the purified oligomycin-sensitive, HEPPS/KOH, pH 8.0, ImM free magnesium as MgCl2, 2mM phosphoeno/pyruvate (K®), O.SmM NADH, ImM and the submitochondrial enzyme. In addition,

Brought to you by | The University of A Authenticated Download Date | 6/17/15 4:5

Bd. 357 (1976)

Yeast Mitochondrial Adenosinetriphosphatase

1607

KCN, 50 μ§ each of pyruvate kinase and lactate dehydrogenase suspended in 50%glycerol and 5mM magnesium-ATP complex or other nucleoside triphosphatemagnesium complexes. During the purification, only the method of Pullman et alJ 13 1 was used with the following modification: MgCl2, NADH and KCN were 3mM, 0.4mM and ImM, respectively, temperature = 25 °C. The activity obtained by this method was approximately 15% lower than that observed by the assay system given above. The reactions were initiated by the addition of the samples. Ammonium sulfate, ATP and EDTA were removed from the samples by centrifugation prior to the assay. Because of the lack of specificity of pyruvate kinase for nucleoside diphosphates, the same assay system was used in case of the other nucleotides 1 l5 1. The concentrations of ATP, GTP and ITP were determined spectrophotometrically at 259 (pH 7), 252 (pH 7), and 248.5 (pH 6) nm, assuming millimolar extinction coefficients of 15.4, 13.6, and 12.2, respectively[16,17]

Analytical gel filtration. The procedure was performed at room temperature in a column of Ultrogel AcA22 (1.0 χ 77 cm, 6 m//h), which was equilibrated with the buffer solution containing 0.25M sucrose, 2mM EDTA, 4mM ATP and 50mM Tris/S04,pH 7.5. Reference proteins for molecular weight determination were as follows: horse spleen ferritin (MT 540000), rabbit muscle pyruvate kinase (Mr 237000), yeast alcohol dehydrogenase (MT 148000).

Protein was determined according to the method of Lowry et alJ 18 l Crystalline bovine serum albumin or Precimat (Boehringer Mannheim GmbH) served as standards. ATP, ADP and AMP were determined according to the method of ref.l 19 l

Purification of sonicated and oligomycin-sensitive ATPase from baker's yeast. The sonicated enzyme was purified by the method of refJ27!, except that the procedure of Sepharose 6B chromatography was added after the step of DEAE-cellulose chromatography, instead of glycerol gradient centrifugation. Oligomycinsensitive ATPase was purified according to r

Ultracentrifugation was performed in a Beckman Spinco Model E analytical ultracentrifuge and evaluated in principle as described elsewhere^20,21] Tne enzyme was equilibrated in buffer I of Penefsky and Warner^22!.

PURIFICATION BY CHLOROFORM EXTRACTION

Amino acid analysis was performed using an automatic amino acid analyzer, Biotronik. Before analysis, samples were hydrolyzed in 6N HC1 at 110 °C for 20 - 24 h. Polarity of peptides was computed as mole fractions of polar amino acids, according to ref.l26). Sedimentation in sucrose density gradients was performed according to refJ2! using International Ultracentrifuge B-60, SB405 rotor. Catalase was used as a marker (s 2 ° 0>w =l 1.3).

Step 1: Preparation of'yeast mitochondria. Commercial baker's yeast (1.2 kg) was suspended in 600 ml of buffer l (0.4M sucrose, ImM EDTA. 50mM Tris/acetate, pH 8.2) and 0.1 ml of Silicon Antifoam Emulsion, and homogenized with glass beads in a colloid mill (Vibrogen Cell Mill, B hler, T bingen). The mill was cooled to 0 °C Polyacrylamide gel electrophoresis in sodium dodecylsul- and operated under nitrogen gas. The overall time for fate was carried out as described in ref.l23!, except cell breaking was 1.5 h. The homogenate was adjusted that the gels were stained with 1 % Amido black disto pH 7.5 with 30% KOH and centrifuged twice at solved in methanol/acetic acid/water (5:1:5) and 2500 χ g for 15 min to remove unbroken cells and debris. destained in the same solvent. The gels were scanned at The turbid supernatant solution was centrifuged at 540 nm with a Spectrodensitometer, Model SD 3000, 45900 x£ for 30 min. The mitochondrial paste was Schoeffel Instrument-Corp. suspended in 900 ml of buffer 2 (0.25M sucrose, lOmM Tris/acetate, pH 7.5) with a Potter-Elvehjem homogeniser Absorption, fluorescence, and circular dichroism spectra. and washed twice by centrifugation at 45 900 χ g for 1 h. Spectra were obtained with the enzymes washed 3 times The pellets were suspended in 420 ml of buffer 2 and by ammonium sulfate precipitation or chromatographed stored at 0 °C overnight. on a column of Sephadex G-50, which was equilibrated with a buffer system containing 50 %glycerol as described The stored particles were washed twice with buffer 2 by centrifugation at 140000 χ g for 30 min. The washed elsewhere^24!. Fluorescence spectra were obtained in a particles were suspended in 300 ml of buffer 2 (30 mg Hitachi-Perkin-Elmer fluorescence spectrometer (Type protein/m/) and stored at - 20 °C for at least 2 days. MPF-3). Circular dicroism spectra were run in a DichroThe gravity values refer to the bottom of the centrifuge graph III of Jobin Yvon and processed with the help of a Nicolet Instrument Corp., Model 1074. The calculation tube. Step 2: Preparation of submitochondria. Mitochondria of helicity was based on a mean residual weight of the were thawed at 25 °C in a water bath and sonicated in a amino acids of 100 according tol25!.

Diffusion properties were analysed in a laser light scattering digital correlator of Precision Devices, Malvern, England.

Brought to you by | The University of A Authenticated Download Date | 6/17/15 4:5

1608

K. Takeshige, B. Hess, M. B hm and H. Zimmermann-Telschow

Branson sonicator B30 setting at step 2 and pulsed (80% sonication and 20% rest) for 108 s at 200 W. The suspension was centrifuged at 20 200 χ g for 15 min and the supernatant solution was centrifuged (in A-l 70 rotor of International Ultracentrifuge B-60) at 140000 χ g for 90 min. The entire pellet was suspended in 420 m/ of buffer 2 and recentrifuged at 140000 xg- for 60 min. The washed particles were suspended in 130 ml of buffer 2 to give a final protein concentration of 25 mg/m/ and stored at 0 °C overnight. Step 3: Solubilization oftheATPase by chloroform. Submitochondrial particles were suspended in a solution containing 0.25M sucrose, lOmM Tris/SO4, pH 7.3, ImM EDTA and 2mM ATP to give a final protein concentration of 10 mg/m/. From here on, all manipulations were carried out strictly at 15 - 20 °C. Aliquots (50 m/) of the suspension were placed into beakers containing 25 ml chloroform and mixed vigorously for 25 s by magnetic stirrer. The emulsion was broken by centrifugation in a Christ centrifuge at low speed. The white supernatant was centrifuged with metal tubes at 45900 xg for 10 min to remove chloroform completely and then the slightly turbid supernatant was recentrifuged at 140000 xg for 30 min. The clear, slightly yellow supernatant solution was brought to pH 7.5 with dilute NaOH. Step 4: DEAE-cellulose chromatography. A column of DEAE-cellulose DE 23 (3 χ 4.8 cm, 3.5 m//h) was equilibrated with solution A (0.25M sucrose, lOmM Tris/SO4, pH 7.5, ImM EDTA and 2mM ATP). The enzyme preparation from step 3 was applied and washed into the column with 50 ml of solution A. The column was washed with 200 ml of the same solution containing 20mM K2SO4. ATPase activity was then sharply eluted with 200 ml of solution A containing 75mM K2SO4. 10-m/ fractions were collected and the 4 or 5 tubes con-

Bd. 357 (1976)

taining 70% or more of the eluted activity were pooled. 2M Tris/SO4, pH 7.5, 0.2M ATP, O.lM EDTA were added to bring the concentrations to 50, 4, and 2mM, respectively. Step 5: Ammonium sulfate fractionation. The enzyme solution was brought to 50% saturation by addition of solid ammonium sulfate. Following removal of a small amount of precipitated protein by centrifugation (occasionally no turbidity was evident at this point and centrifugation was omitted), the supernatant solution was brought to 70% saturation in ammonium sulfate and stored at 0 - 5 °C. Step 6: Gel filtration on UltrogelAcA 34. The ammonium sulfate suspension was centrifuged and protein was dissolved in 2.0 ml of solution Β (0.25Μ sucrose, 4mM ATP, 2mM EDTA and 50mM Tris/SO4, pH 8.0). The enzyme solution was then applied to a column of Ultrogel AcA 34 (120 χ 2.5 cm, 30 m//h), which had been equilibrated with solution B. 5 ml fractions were collected and the 5 or 6 tubes containing 60% or more of the eluted activity were pooled. The enzyme solution was brought to 70% saturation in ammonium sulfate after concentrated by ultrafiltration by using Diaflo XM 50 membrane filter and stored at 0 - 5 °C. A summary of a typical purification is given in Table 1. The yield based on submitochondria was 55 % with a specific activity of 175.6 U/mg.

Results General properties. The purification procedure allows the preparation of a stable ATPase complex with high specific activity in a yield of roughly 50%. Table 2 indicates that other sol-

Table 1. Purification of chloroform ATPase. Steps

Mitochondria Submitochondria Chloroform extract DEAE-Cellulose chromatography (NH 4 ) 2 SO 4 50-70% fractionation UltrogelAcA 34 chromatography

Volume

Total protein

[ml]

[mg]

300 470 385

9186 4916

54

Total activity IU]

Spec. act. [U/mg]

Yield [%]

1.63 1.73 18.84

100.0

720

14973 8521 13567

107

7510

69.88

88.1

159.2

2

71.7

7224

100.81

84.8

34.5

26.6

4665

175.60

54.7

Brought to you by | The University of A Authenticated Download Date | 6/17/15 4:5

Yeast Mitochondrial Adenosinetriphosphatase

Bd. 357 (1976)

1609

Table 2. Extraction of ATPase by various organic solvents. The conditions were the same as described in Materials and Methods except the total volume was 1.0 ml. Organic solvents Chloroform Diethyl ether Carbon tetrachloride Iso -octane Cyclohexane 1-Butanol 2-Methy 1-1 -propanol 1-Propanol Benzylalcohol

Protein [mg]

0.87 1.45 1.50 1.30 1.35 1.20 1.25 1.40 0*

Total activity [U] 11.00 12.70 10.30 3.04 3.04 0 0 0 0

Spec. act. [U/mg] 12.64 8.48 6.87 2.34 2.25 0 0 0 0

The sample became mushy and no supernatant was observed after centrifugation.

vents were less effective for extraction of the enzyme. Among them diethyl ether and carbon tetrachloride yielded crude extracts with equal yields, but with low specific activities. The addition of proteolytic inhibitors to the buffer systems proved not to be necessary. Stabilizing ligands such as ATP, ADP, EDTA, and inorganic phosphate did not affect the yield or the specific activity. It was found, however, that ATP (~ 2mM ) was necessary to stabilize the enzyme. The optimum solubilization occured at pH 7.3 and 10mg/ ml of protein concentration in lOmM Tris/SO4 and 0.25M sucrose. There was negligible ATPase activity remaining in the membrane fraction after chloroform treatment. The purified enzyme was stable at room temperature at a concentration of 1.0 mg/m/ in solution B for up to 24 h. Incubation of the purified enzyme at 0 °C at a concentration of 0.1 mg/m/ in 20mM Tris/acetate buffer, pH 7.5 resulted in 87% loss of activity after 90 min. However, the enzyme was stable for at least a month at 0 °C in the presence of 50% glycerol and lOOmM Tris/ S04, pH 7.5. The enzyme was also stable for months at 0 °C in 70% ammonium sulfate solution with solution B, or under storage at — 70 °C without ammonium sulfate after quick freezing in liquid nitrogen. The purified enzyme was completely inactivated after 10 min at 70 °C, and addition of 4mM ATP could not stabilize the enzyme against heating. The activity of the purified enzyme was insensitive to oligomycin

(6 χ 104mol /mol Fj*) andW,W-dicyclohexylcarbodiimide (40 mol/mol F!). On incubation overnight withTV^V'-dicyclohexylcarbodiimide F t ATPase was inhibited by about 95%. The oligomycin-sensitive enzyme was immediately inhibited up to 70% by 2.5 χ ΙΟ 3 to 2.5 χ ΙΟ4 mol oligomycin per mol enzyme and by about 28% by 70 molA^'-dicyclohexylcarbodiimide per mol enzyme after an incubation of 25 h at 4 °C. Ultracentrifugation. The purified ATPase sedimented homogeneously in the analytical ultracentrifuge, as shown in Fig. 1. Experiments at concentrations ranging from 4.0 to 1.0 mg/m/ indicated that the sedimentation coefficient was not concentration-dependent in this concentration range. Averaging these data, a value of sJo.w = 12.50 ± 0.13 S was obtained. Sedimentation of the enzyme in a sucrose gradient resulted in a single peak of protein coincident with a peak of enzymatic activity, as shown in Fig. 2. A comparison of the sedimentation rate with that of catalase sedimented under identical condition indicated a sedimentation coefficient of S2o,w= 12.6 ±0.4S. Laser light scattering. The application of the laser light scattering technique demonstrated the presence of a component (> 90%) with a correlation time of 45 MS from which-D2o,w °f 2.92 χ 10~7 cm2 s-1 was computed. = soluble mitochondrial ATPase.

Brought to you by | The University of Au Authenticated Download Date | 6/17/15 4:50

1610

K. Takeshige, B. Hess, M. Böhm and H. Zimmermann-Telschow

Bottom

W

Bd. 357 (1976)

20 Fract. no.—)

Fig. 2. Sucrose density gradient analysis of chloroform ATPase. The sedimentation position of catalase is marked by the arrow.

d)

e)

f)

Fig. 1. Sedimentation velocity analysis of chloroform ATPase. The enzyme was equilibrated with buffer I of refJ 22 '. Final protein concentration, 2.1 mg/m/. Rotor speed, 68000 r.p.m. at 25 °C. Phase angle, 60°. Sedimentation, from left to right. The photographs were taken at 4-min intervals using Schlieren optics.

Electron microscopy of the chloroform ATPase preparation negatively stained with 1 % uranyloxalate revealed a homogeneous preparation of unique spherical particles with a diameter of ~ 110 A. Some particles could be seen to be a hexagonal arrangement of about six subunits (see Fig. 3). Gel filtration. The homogeneity of the purified enzyme was further assessed by gel filtration on Ultrogel AcA 34 column and by ion exchange on a DEAE-cellulose column. In both experiments the protein and enzyme activity eluted in a single coincident peak, as shown in Fig. 4 and Fig. 5, respectively. The purified enzyme was chromatographed on Ultrogel AcA 22 at 25 °C simultaneously with ferritin, pyruvate kinase and yeast alcohol dehydrogenase and the positions of the peaks were determined. As shown in Fig. 6, a plot of Fig. 3. Electron micrograph of chloroform ATPase negalog molecular weight of the marker proteins vertively stained with 1 % uranyloxalate, pH 6.6. Magnificasus the midpoint of the peak gave a straight line. tion 400000:1.

Brought to you by | The University of A Authenticated Download Date | 6/17/15 4:5

1611

Yeast Mitochondrial Adenosinetriphosphatase

Bd. 3157 (1976)

Fract. no. Fig. 4. Gel filtration analysis of chloroform ATPase on Ultrogel AcA 34. The enzyme (1.0 ml, 2.34 mg) in solution B was layered on a column (2.5 X 120 cm) equilibrated with the same solution. The enzyme was eluted at a flow rate of 3.5 m//h at 20 °C. Arrows indicate the elution positions of dextran blue and 2,4-dinitrophenol.

The value of 400000 ± 20000 was obtained for the molecular weight of chloroform ATPase. The molecular weight of sonicated ATPase was also determined under the identical conditions and the value of 370000 ± 20000 was obtained. Subunit composition. The protein components of yeast chloroform ATPase were analysed electrophoretically after depolymerization with dodecylsulfate and compared with those of the sonicated and the oligomycin-sensitive ATPases. The results are shown in Fig. 7. Chloroform ATPase was composed of six subunits, but a very small peak, which was prominent in the oligomycin-sensitive enzyme (subunit 5) was also detected in the densitogram. Ail seven subunits are displayed in oligomycin-sensitive ATPase. In the sonicated enzyme, peaks 3 and 5 were missing. It should be noted that peak 6 of the chloroform ATPase migrated slightly faster than those of the two other forms of ATPase. All six subunits of chloroform ATPase were stable components and did not disappear even after further purification by sucrose density gradient centrifugation or rechromatography on Ultrogel AcA 34 column. In

Fig. 5. DEAE-cellulose chromatography of chloroform ATPase. 5 mg of enzyme was applied to the column (1.5 X 3 cm) equilibrated with solution A. The elution was performed with solution A containing a continuous gradient of K2SO4.

Table 3 the molecular weights of the subunit proteins as well as their relative intensities in the gel densitogram are summarized. In eight different gels the molecular weights estimated did not differ by more than 4% from the value reported. The stoichiometric relationship was computed from densitometry as follows: 1(α):2(β):3:4(7): 6(5):7(e) = 3:3:l:l:l:l.

105

110

115 Fract. no. -

120

125

Fig. 6. Molecular weight of chloroform ATPase (·) obtained by gel filtration. 1) Ferritin, 2) pyruvate kinase, 3) alcohol dehydrogenase.

Brought to you by | The University of A Authenticated Download Date | 6/17/15 4:5

1612

K. Takeshige, B. Hess, M. B hm and H. Zimmermann-Telschow

7α

Bd. 357 (1976)

7b

12

7c 7α

/**-*—^

Fig. 7. Polyacrylamide gel electrophoresis in the presence of sodium dodecylsulfate. a, 110 Mg chloroform ATPase; a', 15 Mg chloroform ATPase; b, 60 Mg oligomycin-sensitive ATPase; c, 110 jug sonicated ATPase.

Brought to you by | The University of A Authenticated Download Date | 6/17/15 4:5

Yeast Mitochondrial Adenosinetriphosphatase

Bd. 357 (1976)

1613

Table 3. Molecular weights of subunit proteins. A stamdard curve was plotted from the relative migration of bovine serum albumin, trypsin and cytochrome c. The rnutnber of the subunits corresponds to the peaks shown in Fig. 7.

Subunit * 1 2 3 4 5 6 7

Chloroform-ex tract ed a [%!**

(Ύ)

58500 55000 42000 34000

39.5 39.5 2.7 11.9

(δ) (β)

10000 8600

4.0 2.4

( Mg-ITP > Mg-GTP > Mg-CTP > Mg-UTP. The type of inhibition of Dio-9 for Mg-ITP and Mg-GTP was also noncompetitive (data not shown). The values of K^ were determined to be 0.25 Mg/m/ and 0.55 Mg/m/ at low and high concentrations of Mg-ATP, respectively. In the presence of Dio-9, the low velocity of chloroform ATPase was increased by the activating anion by the same factor as the velocity in the absence of Dio-9. Discussion

Fig. 12. Effect of Dio-9 on the ATPase activities. o-o-o Chloroform-extracted; ·-·-· sonicated; Δ-Δ-Δ oligomycin-sensitive; α-ο-α submitochondrial.

Fig. 12 shows the dependence of the activity on the concentration of Dio-9 for the four enzyme preparations investigated. The activities of the chloroform, sonicated and oligomycin-sensitive forms were strongly inhibited, but submitochondrial ATPase was less sensitive to the inhibitor. The Lineweaver-Burk plots in Fig. 13 show that

The experiments reported here described the 100-fold purification of yeast mitochondrial ATPase extracted by chloroform and chromatographed to homogeneity with high specific activity (200 U/mg), 55% yield and high stability. For the routine work, chloroform was used because of its technical advantages, although the use of ether yielded similar results. In addition it should be noted that during the purification, the temperature must be strictly controlled at 15 - 20 °C. Otherwise the enzyme activity was rapidly lost and the purification yield appreciably lowered. The high homogeneity of the enzyme was given by the sedimentation velocity analysis, the sucrose density gradient analysis, the gel filtration

Fig. 13. Lineweaver-Burk plots showing the effect of Dio-9 on the activity of chloroform ATPase. The concentration of Dio-9 was 1.25 Mg/m/. [Mg-AJP]-1[l/mmol]-

Brought to you by | The University of Au Authenticated Download Date | 6/17/15 4:50

Bd. 357 (1976)

Yeast Mitochondrial Adenosinetriphosphatase

1619

and Ion-exchange chiomatography as well as the laser light scattering response. The preparation did not show any adenylate kinase activity. Chloroform has also been used for extraction of ATPase from beef heart submitochondria to yield a cold-stable enzyme, however, with a low specific activity of 20 U/mg at 30 °C. The crude extracted enzyme from baker's yeast submitochondria reported here was cold-labile, indicating a difference in structure between the enzymes from beef heart and baker's yeast. Such differences between the extracts from yeast and beef heart submitochondria were also observed when Triton X-100 was used for the extraction of the oligomycin-sensitive ATPasel31!.

mycin-sensitive enzyme. The molecular weights of the subunits of our sonicated ATPase were similar to those reported for Schizosaccharomyces pombe^33^, but different from those reported for the diploid strain of Saccharomyces cerevisiae^21^. In the latter case a subunit of a molecular weight of 8600 has not been reported. In addition, the authors have reported a subunit of a molecular weight of 38500, which might correspond to the subunit of 42000 observed in chloroform ATPase. Also, it should be mentioned that the molecular weight of the subunits of the oligomycinsensitive enzyme described here differed from those observed by these authors, where two more bands (29000 and 18 500) have been reported.

The molecular weight of chloroform ATPase was approximately 400000. This value was slightly higher than that (340000) of the ATPase obtained from diploid strain of baker's yeast'27'. This difference might be due to the missing component with the molecular weight of 8 600 and to the differences in the molecular weights of subunits. With the Svedberg equation, the diffusion coefficient was calculated to be 2.90 χ 10~7 cm2 χ s"1 from the data of molecular weight, the sedimentation coefficient and the partial specific volume ( V = 0.738 cm3 χ g"1) which was evaluated from the amino acid composition according to the method of I 32 ). This derived value was comparable to the value (£>2o,w = 2.92 χ 10~7 cm2 χ s-1) computed from the correlation time measured by the laser light scattering method.

The subunit 7(e) might be the ATPase inhibitor protein, since the molecular weight of subunit 7(e) (8600) was very similar to the data on the inhibitor proteins purified from two strains of yeastl34'35! and beef heartl36! with the molecular weight of 6000 and 10500, respectively, but the amino acid composition of subunit 7(e) was different from those obtained from the two strains of yeast and beef heart.

The amino acid composition of the native chlorofrom ATPase was surprisingly similar to that of five enzymes from unrelated sources, except that chloroform ATPase from yeast contained less methionine when our data were normalized to 30000 g of protein and compared with the other data which were also normalized to 30000 g of proteini2»37!. With respect to the amino acid composition of the subunits, it was notable that The homogeneity of the preparation was furtherin subunits δ and e methionine could not be more supported by the results obtained with found under the mild conditions used. The amino electron microscope. Although this technique acid composition of subunit γ was relatively simisuggested that the enzyme is a spherical particle, lar to that of beef heart ATPase, although there a detailed conclusion cannot be drawn at present were remarkable differences in the compositions with respect to the arrangement of the subunits. of subunit δ and e in chloroform and beef heart ATPasel38!. The distribution of polar and nonThe analysis of the subunit composition of chloropolar residues of about 50% each was of interest form ATPase demonstrated six subunits. Their and indicated the excellent solubility of the ensuggested stoichiometric relationship of: l(a): zyme in aqueous solutions with the suitable ionic 2(/3):3:4(γ):6(δ):7(ε) = 3:3:1:1:1:1 gave a mostrength. lecular weight of 435100, which is compatible The spectral data and the enzymatic analysis inwith the molecular weight determined by gel dicated the presence of tightly bound adenine filtration and the value evaluated from the amino nucleotides. The ratio ATP:ADP:AMP = 1:1:1 acid compositions. Chloroform ATPase from per mole of enzyme was found in oligomycinbaker's yeast might not contain subunit 5 of the sensitive ATPase. Although the source of AMP is seven subunits which were observed in the oligo-

Brought to you by | The University of Au Authenticated Download Date | 6/17/15 4:50

1620

K. Takeshige, B. Hess, M. B hm and H. Zimmermann-Telschow

not known at present, it should be noted that no activity of adenylate kinase was found in the final preparation.

Bd. 357 (1976)

two active sites for Mg-ATP and the active sites for other nucleoside triphosphates, since the inhibition type was noncompetitive and two different inhibitor constants were found.

The kinetic analysis of the three forms of ATPase While this manuscript was in preparation Pedersen from yeast demonstrated a biphasic linear Line1431 reported that the Lineweaver-Burk plots of weaver-Burk relationship, and this finding sugATP hydrolysis by purified and submitochondrial gests strongly the existence of two different acATPase from rat liver mitochondria were biphasic tive sites (site I and site Π) for the binding and in Tris/Cl buffer, but monophasic in Tris/hydrohydrolysis of Mg-ATP, with quite different Km values and rather similar V. This interpretation is gencarbonate buffer. In our experiments sodium hydrogencarbonate did not affect the biphasic based on the assumption that the two sites are slope of Mg-ATP hydrolysis although it stimunot interdependent. However, the possibility, lated the activity. Two substrate sites for Mg-ATP should also be considered that there is one active have also been suggested by Ebel and Lardyl14! site controlled by an allosteric regulatory site. recently. However they failed to observe the Indeed, the computed data could also be fitted biphasic kinetic pattern for Mg-ATP hydrolysis with an additional dissociation constant for the catalysed by submitochondria. Schatz et al.t4! ATP binding at a regulatory site. The absence of did not observe the biphasic slope in the overall biphasic slopes in the plots for Mg-GTP and Mg-ITP, with the higher Km comparable to site II kinetics by sonicated ATPase from baker's yeast. In their experiments they did not remove ammofor Mg-ATP, indicated the possibility that only one of the sites was acceptable for the two ligands nium sulfate from the coupling enzymes before assays. And as the value of Km they reported Mg-GTP and Mg-ITP, or that both sites were (0.25mM) was similar to that observed in this paper equivalent and not interdependent. Also it is for chloroform ATPase (0.37mM), it was suggested possible that there are specific sites for Mg-GTP that high affinity site I might have disappeared in and Mg-ITP, which are different from the sites the presence of sulfate ions. for Mg-ATP. Here again, the allosteric model could be discussed*. At present it is difficult to discuss a correlation between the binding sites defined here and the The kinetic properties of chloroform ATPase several binding sites observed by other authors were strongly affected by anions, as observed in 39 40 42 [19,24] mitochondriali !, submitochondriall - ! and 3 5 30 isolated ATPasel » · ! from various sources. In The remarkable similarity of the values of Km addition to their activating and inhibiting funcfor Mg-ATP of the three forms of ATPase with tion, the differential effects on the high affinity two active sites and the similar values of Km for site I and low affinity site II for Mg-ATP were Mg-ITP and Mg-GTP, comparable to site Π for observed, where site I and site II disappeared in Mg-ATP, suggested that none of the differences the presence of sulfate and sulfite, respectively. observed in the subunit compositions and molecIt is also possible that the anions interfere with a ular weights between the three forms of ATPase regulatory site controlling allosterically the activ- was responsible for the functions displayed by ity of the enzyme. These observations indicate Km. Similarly, there were no differences between the possibility of further differentiation, by sonicated and chloroform ATPase with respect other compounds, between the two binding sites, to the effects of anions, Dio-9 and the temperaand experiments based on this idea are currently ture. It should be noted that Sone et aU5l have being carried out in this laboratory. It should also found that both the membrane-bound and also be mentioned that the antibiotic inhibitor soluble ATPase from yeast, Endomyces magnusii, Dio-9 did not seem to act simultaneously on the had the same Michaelis constant. These findings indicate that yeast ATPase may exist in a similar conformational state when bound to the membrane and when free in aqueous solution. On the : See note added in proof, p. 1621. contrary, significant differences in the values of

Brought to you by | The University of A Authenticated Download Date | 6/17/15 4:5

Bd. 357(1976)

Yeast Mitochondrial Adenosinetriphosphatase

1621

5 Sone, N., Furuya, E. & Hagiwara, B. (1969) /. Biochem. (Tokyo} 65, 935 - 943. 6 Munoz, E., Salton, M. R. J., Ng, M. H. & Schor, M. T. (1969) Eur. J. Biochem. 7, 490 - 501. 7 Abrams, A. (1965) /. Biol. Chem. 240, 3675 - 3681. The difference between chloroform ATPase and 8 Senior, A. E. (1973) Biochim. Biophys. A eta 301, oligomycin-sensitive ATPase with respect to the 249 - 277. sensitivity to oligomycin should be due to the 9 Beechy, R.B. (1974) Biochem. Soc. Spec. Publ. 4, lack of lipid and the oligomycin-sensitivity con41-62. 10 Munn, E.A. (1974) The structure of Mitochondria, ferring protein (OSCP), for which a molecular Academic Press, London. weight of 18500 is reported!27!. Chloroform 11 Kozlov, I.A. & Mikelsaar, H.N. (1974) FEBSLett. ATPase did not contain the subunit (subunit 5) 43,212-214. which was observed in the oligomycin-sensitive 12 Beechey, R.B., Hubbard, S.A., Linnett, P.E, ATPase of yeast. This subunit was analysed here Mitchell, A.D. & Munn, E. A. (1975) Biochem. J. 148,533-537. as having Mt = 23 500, and it mi^it be the OSCP, 13 Pullmann, M.E., Penefsky, H.S., Datta, A. & Racker, E. although the 18500 component might be hidden (1960) / Biol. Chem. 235, 3322 - 3329. in the broad line. The addition of phospholipids 14 Ebel, R.E. & Lardy, H.A. (1975)/. Biol. Chem. 250, (lecithin) to chloroform ATPase could not re191-196. 15 Kayne, F. J. (1973) in The Enzymes (Boy er, P.D., store the sensitivity to oligomycin. ed.) Vol. VIII, pp. 353 - 382, Academic Press, The influence of the quaternary structure and New York. the lipid environment on the function of the 16 Bock, R.M., Ling, N.S., Morel, S.A. & Lipton, S.H. (1956) Arch. Biochem. Biophys. 62, 253 - 264. ATPase was reflected in the segmented Arrhenius 17 Beaven, G.H., Holiday, E.R. & Johnson, E. A. (1955) plots for the oligomycin-sensitive and the subin The Nucleic Acids (Chargaff, E. & Davidson, J.N., mitochondrial ATPase, indicating the temperaeds.) Vol. 1, pp. 493 - 553, Academic Press, New York. ture-dependent conformation changes or the 18 Lowry, O.K., Rosebrough, N.J., Farr, A.L. & lipid-phase transitions of the enzymel45'47!. Randall, R. J. (1951) / Biol. Chem. 193, 265 - 275. 19 Harris, D.A., van de Stadt, R. J. & Slater, E.G. (1973) We are grateful to Drs. D. Kuschmitz, N. Tamaki and Biochim. Biophys. Acta 314, 1285 -1291. £>. Recktenwald for helpful discussions during the 20 Bischofberger, H., Hess, B. & Röschlau, P. (1971) preparation of the manuscript. The excellent technical thisJ. 352,1139-1150. assistance of Mrs. U. Schacknies and Mrs. U. Jennrich 21 Yphantis, D.A. (1964) Biochemistry 3, 297 - 317. is gratefully recognized. We are also indebted to 22 Penefsky, H. S. & Warner, R.C. (1965) /. Biol. Chem. Dr. T. Reed and Mr. W. Doster for taking the circular 240, 4694 - 4702. dichroism spectra and for carrying out the laser light 23 Catterall, W. A., Coty, W.A. & Pedersen, P.L. (1973) scattering analysis, respectively. Furthermore we acJ. Biol. Chem. 248, 7427 - 7431. knowledge the help of Dr. D. Schäfer, who took the 24 Garrett, N. E. & Penefsky, H. S. (1975) /. Biol Chem. electron microscope pictures. 250, 6640 - 6647. 25 Chen, Y.-H. & Yang, J.T. (1971) Biochem. Biophys. Res. Commun. 44,1285 -1291. Note added in proof: 26 Capaldi, R. A. & Vanderkooi, G. (1972) Proc. Nat. Acad. Sei. USA 69, 930 - 932. Recent experiments indicate the allosteric nature of the 27 Tzagoloff, A. & Meagher, P. (1971) J. Biol Chem. steady state kinetics of ATPase with a negative cooper246,7328-7336. ativity of a Hill coefficient being 0.5. 28 Pedersen, P.L. (1975) Bioenergetics 6, 243 - 275. 29 Layne, E. (1957) in Methods in Enyzmology (Colowick, S.P., and Kaplan, N.O., eds.) Vol. Ill, pp. 447 - 454, Academic Press, New York. 30 Racker, E. (1962) Fed. Proc. 21, 54. Literature 31 Linnett, P.E., Mitchell, A.D. & Beechey, R.B. (1975) FEBS Lett. 53,180-183. 1 Knowles, A. H. & Penefsky, H. S. (1972) /. Biol 32 Cohn, E. J. & Edsall, J.T. (1943) in Proteins Chem. 247,6617-6623. Aminoacids and Peptides, Reinhold Publishing 2 Catterall, W. A. & Pedersen, P. L. (1971) /. Biol. Chem. Company, New York, pp. 370 - 381. 246,4987 - 4994. 33 Goffeau, A., Landry, Y., Froury, F. & Briguet, M. 3 Lambeth, D.O. & Lardy, H. A. (1971) Eur. J. (1973) /. Biol. Chem. 248, 7097 - 7105. Biochem. 22, 355 - 363. 34 Satre, M., Jerphanion, M.-B., Huet, J. & Vignais, P.V. 4 Schatz, G., Penefsky, H. S. & Racker, E. (l 967) (1975) Biochim. Biophys. Acta 387, 241 - 255. J. Biol. Chem. 242, 2552 - 2560.

Km between the purified and membrane-bound ATPase from beef heart and rat liver mitochondria have been reportedl43»44!.

Brought to you by | The University of Au Authenticated Download Date | 6/17/15 4:50

1622

K. Takeshige, B. Hess, M. Böhm and H. Zimmermann-Telschow

35 Ebner, E. (1974) 9th FEBS-Meet. Budapest, Abstr. No. s6j3. 36 Brooks, J.C. & Senior, A. R (1971) Arch. Biochem Biophys. 147, 467 - 470. 37 Schnebli, H.P., Vatter, A. F. & Abrams, A. J. (1970) /. Biol. Chem. 245, 1122-1127. 38 Knowles, A.F. & Penefsky, H. S. (1972) /. Biol. Chem. 247, 6624 - 6630. 39 Fanestü, D.D., Hastings, A.B. & Mahowald, T.A. (1963) /. Biol. Chem. 238, 836 - 842. 40 Ulrich, F. (1963) Biochem. J. 88,193 - 206. 41 Mitchell, P. & Moyle, J. (1971) Bioenergetics 2, l -11.

Bd. 357 (1976)

42 Pinto, E., Gomez-Puyou, A., Sandoval, F., Chavez, E. & Tuena, M. (1970) Biochim. Biophys. Acta 223, 436-438. 43 Pedersen, P.L. (1976) /. Biol. Chem. 251,934 - 940. 44 Kammes, G.G. & Hilborn, D. A. (1971) Biochim. Biophys. Acta 233, 580 - 590. 45 Raison, J.K., Lyons, J.M. & Thomson, W.W. (1971) Arch. Biochem. Biophys. 142, 83 - 90. 46 Mavis, R.D. & Vagelos, P.R. (1972) /. Biol. Chem. 247, 652 - 659. 47 Oldfield, E. & Chapman, D. (1972) FEBS Lett. 23, 285 - 297.

Prof. Dr. Benno Hess, Max-Planck-Institut für Ernährungsphysiologie, Rheinlanddamm 201, D-4600 Dortmund 1.

Brought to you by | The University of A Authenticated Download Date | 6/17/15 4:5