PURIFICATION AND PROPERTIES OF FRUCTOKINASE Bovine Liver Fructokinase: Purification and Kinetic Properties? Frank M. Raushel and W. W. Cleland* ABSTRACT: Fructokinase from beef liver has been purified tagatose, D-psicose, D-xylulose, ~-ribulose,D-sedoheptulose, 2300-fold by acid and heat treatment, ammonium sulfate L-galactoheptulose, D-mannoheptulose, 5-keto-D-fructose, fractionation, and chromatography on Sephadex G- 100, D-ribose, 2,5-anhydro-D-mannitol, 2,5-anhydro-~-glucitol, DEAE- and CM-cellulose. The purified enzyme is homoge- 2,5-anhydro-~-mannose,2,5-anhydro-D-lyxitoI, and D-ri- neous by all criteria examined, has a molecular weight of bono-y-lactone. 5-Thio-D-fructose was not a substrate, but was 56 000, and is a dimer of equal molecular weight subunits. The a competitive inhibitor vs. D-fructose. Thus the minimum isoelectric point is 5.7. The Michaelis constant for activation molecule for substrate activity seems to be (2R)-2-hydroxy- by K+ is 15 mM, and the enzyme is also activated by Na+, methyl-3,4-dihydroxytetrahydrofuran.The configuration of Rb+, Cs+, NH4+, and TI+. The kinetic mechanism has been the substituents at carbons 3,4, and 5 appears not to be critical, determined at pH 7.0,25 "C. The initial velocity, product, and but the hydroxymethyl group must have the configuration dead-end inhibition patterns for CrATP, CrADP, and 1- corresponding to p-D- (or a-l-) keto sugars. The anomeric deoxy-D-fructose are consistent with a random kinetic mech- hydroxyl on carbon 2 is not required (although it contributes anism with the formation of two dead-end complexes. Sub- to binding), and a wide variety of groups may be present at strates for fructokinase 'include: D-fructose, L-sorbose, D- carbon 5. Fructokinase (ATP:D-fructose 1-phosphotransferase, EC zation. The concentration of D-xyhlose was determined by 2.7.1.3) catalyzes the following reaction: assay with fructokinase. L-Ribulose was prepared by epimer- K+ ization of L-arabinose by sodium aluminate (Haack et al., D-fructose + MgATP +MgADP + D-fructose-1-P (1) 1964). The yield, based on assay with fructokinase, was 9%. L-Arabinose and D-XylOSe were neither substrates nor inhibi- The enzyme has been purified by a number of workers but has tors of fructokinase. CrATP and CrADP were synthesized by never been prepared in a homogeneous form (Sfinchez et al., the procedure of DePamphilis and Cleland (1 973). 1-Deoxy- 197 1; Parks et al., 1957; Adelman et al., 1967). Fructokinase D-fructose was made according to Ishizu et al. (1 967). The will phosphorylate a large number of sugars (Raushel and procedure of James et al. (1967) was used to prepare D-psicose. Cleland, 1973), but only ATP (or 2'-dATP and 3'-dATP) can 2,5-Anhydro-~-mannitol,2,5-anhydro-D-lyxitoI, 2S-anhy- function as the phosphate donor (Parks et al., 1957; Adelman dro-D-glucitol, and 2,5-anhydro-~-mannosewere prepared as et al., 1967). It also has an absolute requirement for a mono- previously described by Raushel and Cleland (1 973). L-Ga- valent cation for activity (Parks et al., 1957; Sinchez et al., lactoheptulose, D-mannoheptulose, L-glucoheptulose, and 1971). In this paper we present a method for preparing ho- D-sedoheptulose hexaacetate were obtained from Dr. E. Zissis, mogeneous enzyme and report on its physical and chemical 5-thio-D-fructose was from Dr. R. Whistler, 5-keto-D-fructose properties. was from Dr. S. Englard, and 1,2-0-isopropylidene-D-psicose Raushel and Cleland (1973) concluded from the fact that was from Dr. A. Perlin. D-Sedoheptulose hexaacetate was 2,5-anhydro-~-mannitolwas a good substrate for fructokinase, deacylated with cold sodium methoxide (Richtmeyer, 1962). while 2,5-anhydro-~-glucitolwas phosphorylated on carbon Diisopropylidene-D-apiose was obtained from Pfanstiehl 6, and 2,6-anhydro-D-mannitol and 2,6-anhydro-~-glucitol Laboratories and converted to D-apiose with 0.5 N HzS04 were not substrates, that fructokinase was specific for the p- (Bell, 1962). furanose anomer of D-fructose and the a-furanose anomer of Enzyme Assays. Fructokinase activity was measured L-sorbose. We have now extended these studies and report the spectrophotometrically using a pyruvate kinase-lactate de- complete kinetic parameters for 16 substrates of fructokinase. hydrogenase coupling system. A Beckman DU monochrometer In addition we report product and dead-end inhibition studies equipped with a Gilford 200 optical density converter and a which establish the kinetic mechanism to be random. 10-mV recorder was used to follow the reaction at 340 nm. Unless otherwise noted, all assays were done at 25 "C. Materials and Methods For activity measurements during the purification of fruc- Aldolase B was isolated from beef liver using the procedure tokinase, each 3-mL cuvette contained 50 mM piperazine- of Gracy et al. (1 969). x xylulose was synthesized by the ep- N,N'-bis(2-ethanesulfonic acid), pH 7.0, 100 pg each of salt- irnerization of D-xylose with anhydrous pyridine (Touster, free lactate dehydrogenase and pyruvate kinase, 1 mM phos- 1962) and was used as a mixture of D-xylose and D-xylulose phoenolpyruvate, 0.2 mM NADH, 100 mM KC1, 5.0 mM after as much D-xylose as possible was removed by crystalli- ATP (pH 7.0), 5.0 mM fructose, and 6.0 mM MgC12. A unit of fructokinase is that amount of enzyme that produces 1 pmol of MgADP per min in the above assay system. During the early From the Department of Biochemistry, University of Wisconsin, Madison, Wisconsin 53706. Received October 28, 1976. This work was stages of the purification, activities corresponding to ATPase, supported by grants from the National Science Foundation (BMS-75- myokinase, and sorbitol dehydrogenase had to be subtracted 16134) and the National Institutes of Health (GM 18938). from the observed rate using an appropriate blank. When BIOCHEMISTRY, VOL. 16, NO. 10, 1977 2169 2ot -0' ['Im 5L fraction f IGURE I: Chromatography of bovine liver fructokinase on Sephadex G-100. The fraction volume w'as 7 mL: (closed circles) protein (Azso); 40 (open circles) fructokinase activity. fraction FIGLRE 2: Elution of fructokinase from DEAE-cellulose. The fraction volume was 7 mL:(closed circles) protein (Azso); (open circles) fructo- studying the activation of monovalent cations on fructokinase, kinase activity. activity was measured as above except that all acidic compo- nents were neutralized with either Tris' or cyclohexylamine, and fructose and MgATP were held at 25 and 3.0 mM. I In the presence of added ADP, fructokinase activity in the forward reaction was measured spectrophotometrically using a liver aldolose B and a-glycerophosphate dehydrogenase coupling system. Each cuvette contained in 3 mL total volume, 50 mM piperazine-N,N'-bis(2-ethanesulfonic acid)-KOH (pH 7.0), 1. I units of aldolase B, 5 units of a-glycerophosphate dehydrogenase, 0.2 mM NADH, 100 mM KCI, 2 mM excess MgC12, fructokinase, and various amounts of substrates and inhibitors. The back reaction was measured using a hexokinase-glu- cose-6-phosphate dehydrogenase coupling system. Each cu- vette contained in 3 mL, 50 mM piperazine-N,N'-bis(2-eth- anesulfonic acid)-KOH (pH 7.0), 10 units each of hexokinase and glucose-6-P dehydrogenase, 0.3 mM NADP, 2.0 mM iIGURE 3: Elution of fructokinase from CM-cellulose. The fraction glucose, 2 mM excess MgC12, 100 mM KCI, fructokinase, and volume was 1.5 mL: (closed circles) protein (A280); (open circles) fruc- tokinase activity. various amounts of D-fructose- 1-Pand ADP. All assays were at 25 OCand in all cases fructokinase was added last with an adder mixer. Data Analysis. Reciprocal velocities were plotted graphi- an extinction coefficient, 1?1~~''''", of 8.0 at 280 nm when protein cally against reciprocals of substrate concentration and the was determined by the method of Lowry et al. (1951) using data were fitted to eq 2, assuming equal variance for the ve- BSA as a standard. locities and using the Fortran programs of Cleland (1967). Purification of Fructokinase. The purification was carried Data for a sequential initial velocity pattern were fitted to eq out at 0-4 OC. All centrifugations were for 30 min at 8000g 3, for competitive inhibition to eq 4, and for noncompetitive and all buffers contained 1 .O mM dithiothreitol and 0.1 mM inhibition to eq 5. The nomenclature used in this paper is that EDTA. of Cleland (1 963). Five hundred grams of beef liver were homogenized in 750 mL of 0.25 M mannitol in a Waring blender for 1 min. After VA centrifugation, the pH was adjusted to 5.0 with cold 1 N HCI. I' = -- K+A The acid-treated homogenate was centrifuged and the pH of VAB the supernatant solution was adjusted to 7.3 with cold 1 N L'= (3) KOH (fraction I). Fraction 1 has heated in a stainless steel K,,,Kb i-Kb.4 K,B AB + + beaker, with overhead stirring, to 65 "C for 1 min in an 85 OC V4 water bath. The contents were cooled in an ice bath and the L'= (4) K(1 + (I/KI\))+ A precipitated protein was removed by centrifugation and dis- VA carded. The supernatant solution (fraction 11) was brought to L'= (5) 40% saturation with solid ammonium sulfate (0.23 g/mL). The K(1 + (I/Kld) + A(1 + (I/K,l)) suspension was stirred for 30 min and centrifuged. The su- Protein Determination. The protein content of solutions pernatant solution was brought to 60% saturation by the slow during purification was determined using the method of addition of solid ammonium sulfate (0.123 g/mL) and cen- Warburg and Christian (1941) or Lowry et al.