Eur. J . Biochem. 59, 335- 345 (1975)

The Pyruvate-Dehydrogenase Complex from Azotobacter vinelandii 1. Purification and Properties Tjarda W. BRESTERS, Ronney A . D E ABREU, Arie DE KOK, Jaap VISSER, and Cees VEEGER Department of Biochemistry, Agricultural University, Wageningen (Received June 25/August 27,1975)

The pyruvate dehydrogenase complex from Azotobacter vinelandii was isolated in a five-step procedure. The minimum molecular weight of the pure complex is 600000, as based on an FAD content of 1.6 nmol . mg protein-'. The molecular weight is 1.0- 1.2 x lo6, indicating 1 mole of lipoamide dehydrogenase dimer per complex molecule. Sodium dodecylsulphate gel electrophoretical patterns show that apart from pyruvate dehydrogenase ( M ,89 000) and lipoamide dehydrogenase ( M ,monomer 56000) two active transacetylase isoenzymes are present with molecular weight on the gel 82000 and 59 000 but probably actually lower. The pure complex has a specific activity of the pyruvate-NAD+ reductase (overall) reaction of 10 units . mg protein-' at 25 "C.The partial reactions have the following specific activities in units. mg protein- at 25 "C under standard conditions : pyruvate-K,Fe(CN), reductase 0.14, transacetylase 3.6 and lipoamide dehydrogenase 2.9. The properties of this complex are compared with those from other sources. NADPH reduced the FAD of lipoamide dehydrogenase as well in the complex as in the free form. NADP' cannot be used as electron acceptor. Under aerobic conditions pyruvate oxidase reaction, dependent on Mg+and thiamine pyrophosphate, converts pyruvate into COz and acetate; Vis 0.2 pmol O2.min-' mg-', K,,, (pyruvate) 0.3 mM. The kinetics of this reaction shows a linear l/velocity- l/[pyruvate] plot. K,Fe(CN)6 competes with the oxidase reaction. The oxidase activity ist stimulated by AMP and sulphate and is inhibited by acetyl-CoA. The partially purified enzyme contains considerable phosphotransacetylase activity. The pure complex does not contain this activity. The physiological significance of this activity is discussed. It is generally accepted that the pyruvate dehydrogenase complex is responsible for the aerobic oxidation of pyruvate by a complex sequence of at least four reactions, catalyzed by three enzymes : pyruvate dehydrogenase, transacetylase and lipoamide dehydrogenase. The complex has been isolated from a number of sources [l - 71. These multi-enzyme complexes are functional units with molecular weights of 4- 9million, Ahbreviutions. TPP. thiamine pyrophosphate; tricine, N-tris(h ydroxymethyl)mcthyIglycine. Lnzymus. Pyruvate dehydrogenase complex is pyruvate-NAD' reductase. decarboxylating, CoA acetylating (EC 1.2.4.1); pyruvate dehydrogenase is pyruvate-K,Fe(CN), reductase, decarboxylating (EC 1.2.2.2); transacetylase or lipoate acetyltransferase (EC 2.3.1.12);lipoamide dehydrogenase (EC 1.6.4.3);transhydrogenase or pyridine nucleotide transhydrogenasc (EC 1.6.1.1); phosphotransacetylase or phosphate acetyltransferase (EC 2.3.1.8): acetate kinase (EC 2.7.2.1); acetatc thiokinase is acetyl-CoA synthetase (EC 6.2.1.1); citrate synthase (EC 4.1.3.7); malate dehydrogenase (EC 1.1.1.37);lactate dehydrogenase (EC 1.1.1.27).

but the complex of Escherichia coli seems to be somewhat smaller than those isolated from mammals. Each type of complex has been separated into the three components [8,9] and reconstruction of the complex with return of the overall activity after mixing the three enzymes has been reported. Investigations have revealed that the complexes from eukaryotes (mammals, Neurospora) contain in addition a protein kinase and a phosphatase, apparently as regulatory subunits of the complex [lo- 131. Covalent modification of the pyruvate dehydrogenase component is part of the regulation of the complex. The pyruvate dehydrogenase complex of E.coli was found to be devoid of the latter activities [14]. Biochemical and electron microscopic analysis suggest that the E.coli complex contains a 'core' of transacetylase molecules, consisting of 24 identical polypeptide chains ( M , 65 000- 70000), which are

336

linked together non-covalently. One molecule of lipoic acid is covalently bound to each chain [6,15]. Twentyfour identical pyruvate dehydrogenase units ( M , 96000) and 12 units of lipoamide dehydrogenase ( M , 56000) are thought to be distributed in a regular manner along the edges of the transacetylase cube in non-covalent interaction. Vogel et al. [16,17] have shown that the 'core' complex from E. coli K12 after removing excess pyruvate dehydrogenase, consists of 16 chains of this enzyme, 16 chains of transacetylase and 16 chains of lipoamide dehydrogenase. The mammalian complex differs from the E. coli complex not only with respect to its molecular weights, the stoichiometry of the components and the presence of a kinase and a phosphatase, but also bccause it contains two non-identical pyruvate dehydrogenase subunits [2,18]. The appearance of the mammalian transacetylase in the electron micrographs also differs from that of thc E. coli enzyme [3]. It was of interest to investigate whether the complex from Azotohacter vinelandii also shows the typical characteristic of a bacterial enzyme system, particularly since it has been reported that the partially purified complex contains phosphotransacetylase which is connected with ATP synthesis [19,20]. MATERIALS AND METHODS Large-scale production of Azotobacter vinelandii (ATCC 478) was kindly performed by the Royal Yeast and Fermentation Industries (Delft, The Netherlands), and by Diosynth B.V. (Oss, The Netherlands), on a nitrogen-free medium (cf. [19]). A 3000-1 fermenter filled with 2500 1 of growth medium was inoculated with a 2 % inoculum and the cells were allowed to grow for 20 h at 30°C under aeration (0.5 1 air. 1 medium-' . min-') and stirring (maximal 100 rev./min). After 20 h (end of the logarithmic phase), the suspension was cooled to 10- 15 "C and the cells were collected in a Sharpless centrifuge at a speed of 180 l/h. The yield (8.5 kg wet weight) was stored, without previous washing at - 20 "C. When experiments were performed with extracts prepared from freshly grown cells, cells were grown for 24 h on a rotary shaker in 5-1 flasks (final volume 1 1) under air and harvested by centrifugation. Pyruvate-NAD' reductase (overall) activity of the complex was measured according to Schwartz and Reed [14] at 25°C with 0.6 mM NAD+ . The enzyme was diluted in either phosphate or Tris-HC1 buffer pH 7.0 containing 1 % ovalbumin. Due to the strong inhibition of the reaction by acetyl-CoA, initial velocity measurements were performed on the AmincoChance dual-wavelength spectrophotometer (380345 nm) using the 5 - 20 % transmission scale. Pyruvate-K,Fe(CN), reductase (pyruvate dehydrogenase) activity was assayed at 25°C in 20 mM tricine buffer

Pyruvate-Dehydrogenase Complex from A . vinelandii

(pH 7.59, 5 mM pyruvate, 10 mM MgCl,, 0.1 mM thiamine pyrophosphate (TPP) and 1mM K,Fe(CN),. The reaction was started by adding enzyme in an appropriate dilution and the decrease in K,Fe(CN), was followed at 430 nm ( E = 1030 M-' . cm-'). Reduced lipoate-CoA transacetylase activity was assayed at 30°C according to Reed and Willms [21]. Lipoamide dehydrogenase activity was assayed at 25 "C under the optimal conditions for this enzyme according to Van den Broek [22]. Phosphotransacetylase activity was assayed at 25°C according to a modification of the method of Bergmeyer (cf: [23]), acetate kinase according to Rose [24] and transhydrogenase activity according to Van den Broek [22]. A unit of activity is defined as the amount of enzyme required to produce 1 pmol of product min-'. Specific activities are given as units mg protein-'. Protein concentrations were determined by the (micro)biuret method [25,26]. The FAD content of the complex was determined according to Beinert and Page [27] by measuring the absorbance of neutralized trichloroacetic acid extracts at 450 nm ( E = 10300 M-' . cm-') before and after reduction with dithionite. In preparations containing FMN plus FAD, both nucleotides were determined according to Wassink and Mayhew [28]. Pyruvate oxidase activity was determined by measuring oxygen consumption at 30°C on the Gilson Oxygraph K M equipped with a Clark electrode. Acetate was determined enzymically according to Bergmeyer and Moellering [29], with acetate thiokinase instead of acetate kinase, citrate synthase and malate dehydrogenase; the malate dehydrogenase is used for the 'preceded indicator reaction'. The concentration of pyruvate was determined with lactate dehydrogenase according to Bergmeyer [30]. Small concentrations of acetate and pyruvate were measured on the Aminco-Chance dual-wavelength spectrophotometer, using the 5- 10 transmission scale (340 - 380 nm). Absorption spectra were recorded on a Cary-14 recording spectrophotometer at 25 "C. The spectra were corrected by the method of Tanford [31] for Rayleigh scattering, taking the refractive index increment independent of the wavelength. Spectra thus calculated compared very well with the ones recorded against a glycogen solution as a blank. The latter spectra were obtained by adjusting the absorbance of the glycogen solution at 700nm to the one of the enzyme solution. Anaerobic conditions were obtained by evacuating and refilling the cuvette with oxygen-free nitrogen, repeating this procedure at least five times. Sedimentation and diffusion patterns were obtained using a MSE analytical ultracentrifuge. Sedimentation velocity runs were performed in 10-mm and 20-mm double-sector cells at 20 "C and at various rotor speeds (33000-45000 rev./min) in 50 mM phos-

T. W. Brestcrs, R. A. de Abreu, A. de Kok, J. Visscr, and C. Vcegcr

phatc pH 7.0. Sedimentation coefficients (s) were determined grafically or calculated according to the relation of Svedberg and Pederson [32] and corrected to standard conditions (cf. [33]). Scans were made with intervals of 6 min. The diffusion coefficient was calculated according to Elias [33]; corrections were made to standard conditions. The molecular weight of the complex was calculated assuming a partial specific volume of 0.75 ml/g. Sedimentation equilibrium experiments were carried out at a rotor speed of 4600 rev./min in a double-sector synthetic-boundary cell at 6 "C. Light-scattering data were obtained with a CencoTNO apparatus at room temperature in 50mM phosphate pH 7.0. Measurements were kindly performed by Mr Van Markwijk (N.I.Z.O., Ede, The Netherlands). The molecular weight has been calculated according to the methods of Zimm [34] and Yang [35], as given by Van den Broek et al. [36]. The samples were filtered before use with a 100-nm filter and enzyme solutions centrifuged before dilution. Values obtained were corrected for contributions of the solvent. NAD', NADH, NADP', NADPH, FAD, FMN, ATP, ADP, AMP, lipoic acid, ovalbumin, bovine serum albumine, TPP, acetyl CoA, dithiothreitol (Cleland's reagent), potassium pyruvate, were obtained from Sigma Chemical Co. ; acetylphosphate was purchased from Boehringer-Mannheim ; K,Fe(CN), and oxaloacetate were from the British Drug House; sodium dithionite and phenylmethylsulphonyl fluoride from Merck; Sepharose 6B and 4B, Sephadex G-25 and DEAE-Sephadex from Pharmacia (Uppsala) ; protamine sulphate and DEAE-cellulose from Serva ; argon from Loos and Co. All other chemicals used were analytical grade and solutions were made up in bidistilled water. RESULTS AND DISCUSSION Purification of the Complex; the 'Pure' Complex Like most procedures concerning the isolation and purification of the pyruvate dehydrogenase complex, the one for A . vinelundii is very similar with that developed and improved by Reed et al. for E. coli Crookes strain [6,21]. The group of Henning [17] has used a different and very elegant procedure, which can not be used satisfactorily in our case. All steps were performed at 4"C, unless otherwise stated. Step 1: Preparation o f t h e Cell-Free Extract. After being washed twice with 1 1 deionized water and twice with 1 I 0.05 M phosphate buffer (pH 7.0), 500 g (wet weight) frozen cell paste was suspended in the same buffer using 2 - 3 ml for 1 g of cells. The cell suspension was treated in a 100-W ultrasonic disintegrator (MSE London) in portions of 40ml for 4min and then centrifuged at 2100 x g for 45 min.

337

Step 2. Fractionation of the Complex, The cell-free extract was brought to pH 6.1 under continuous stirring with a 1 % acetic acid solution. This extract was fractionated with a clear solution of 2 % (w/v) protamine sulphate of the same pH and kept at 25 "C. The effectiveness of this procedure is rather dependent on the batch of bacteria used. A pilot run was done before the large-scale purification, usually 6 "/,,'ofthe volume was added dropwise under stirring. After stirring for 20 min, the solution was centrifuged at 18000 x g for 30 min. This procedure was repeated by adding protamine sulphate to about 3 % of the total volume until the activity just precipitated. After centrifugation, the precipitate was stirred overnight in 200 ml 0.05 M phosphate buffer (pH 7.0). Then 20 ml of a neutralized 1 % (w/v) RNA solution was added slowly, precipitating excess protamine sulphate. After 2 h stirring the solution was ccntrifuged for 45 min at 36000 x g. The complex was precipitated from the clear yellow-brownish supernatant by 50 "4 ammonium sulphate, centrifuged at 10000 x g for 20 min and after dissolving in 35 mlO.05 M phosphate buffer (pH 7.0) dialyzed against the same buffer. Step 3 : Iso-Electric Precipitation. The pH of the dialyzed solution, diluted to 5 mg . ml-', was carefully lowered by adding 1 % acetic acid under stirring until a precipitate appeared, usually at pH 5.6 or slightly lower. The solution was stirred for 5 min and centrifuged at 30000 x g for 10 min. The precipitate was discarded. Upon further adjustment of the pH to 4.9 again a precipitate was formed, which after stirring and centrifugation as above was dissolved in about 30 ml 0.05 M phosphate buffer (pH 7.0). Insoluble material was removed by centrifugation at 30000 x g for 20 min. Step 4 : Ultracentrifugation. The clear solution of step 3 was subjected to differential centrifugation. After 30 rnin at 144000 x g the brown pellet was discarded and the supernatant centrifuged for another 4.5 h. The sediment was carefully collected by swirling with 0.05 M phosphate buffer, thus avoiding the brown material on the bottom of the pellet. The precipitate was homogenized in a Potter-Elvehjem using a small ,volume 0.05 M phosphate buffer (pH 7.0). Insoluble material was removed by centrifugation at 30000 x g for 20 min. If the specific activity was less than 6, this step was repeated. Alternatively, fractionation with solid ammonium sulphate to 30- 50% increases the specific activity somewhat. For many purpose the enzyme as isolated at this stage can be used. It contains, however, phosphotransacetylase activity in varying amounts. Especially when studying the role of acetyl-CoA, this can lead to complications and artefacts (qfi [19,37]). At this stage the enzyme also contains protein-bound FMN in varying amounts ( 5 - 15 This FMN-containing protein as well as the phosphotransacetylase can be

x).

Pyruvate-Dehydrogenase Complex from A . vinelandii

338

removed (step 5) by chromatography on a Biogel column (95 x 2.5 cm). This procedure also removes a brown contamination with a higher molecular weight than that of the complex, which is responsible for an absorption band at 415 nm in the spectrum of the complex. The identity of the FMN protein has not been established yet. The results indicate in contrast to earlier statements [ 19,371, that phosphotransacetylase, although it is closely associated with the complex, can be removed. The complex was stored in liquid N, ; up to 90 % of the activity was recovered after a few months of storage. Upon storage at 4°C a gradual decrease in activity was observed, especially at lower (< 1 mg . r n 1 - I ) concentrations. The complex is more stable in phosphate than in Tris buffer pH 7.0. The activity cannot be restored by the addition of Mg2+ plus TPP or thiols (cf: [7]). Stabilization of the dilute solutions can be achieved by the addition of 1 "/, (w/v) ovalbumin or bovine serum albumin. The purification procedure can be modified slightly by replacing the iso-electric precipitation by a fractionation with polyethylene glycol ( M , 6000) at 4 "C exactly according to Eley et al. [6]. Although the final specific activity is not enhanced, the recovery is somewhat higher. In Table 1 the procedure is summarized. Once a specific activity of about 10 units . mg protein-' was obtained, no further enhancement of activity could be obtained by various column procedures like adsorption chromatography (calcium phosphate gel), ion exchange (DEAE-cellulose, DEAE-Sephadex) and gel filtration (Sepharose 4B and 6B). Since the complex easily dissociates at high purity, these procedures often lead to inactivation. After incubating the complex 5 min at 100°C with 2 7" sodium dodecyl sulphate + 5 2-mercaptoethanol pH 6.8 and subjecting the mixture to dodecylsulphate-gel electrophoresis (10 76 gels, pH 8.8), four major bands were observed with molecular weights of 89000, 82000, 59000 and 56000. Perham and Thomas [38] have observed additional bands in the

The pyruvate-dehydrogenase complex from Azotobacter vinelandii.

The pyruvate dehydrogenase complex from Axotobacter vinelandii was isolated in a five-step procedure. The minimum molecular weight of the pure complex...
929KB Sizes 0 Downloads 0 Views