Active site mutants of Escherichia coli dethiobiotin synthetase: effects of mutations on enzyme catalytic and structural properties.

Five active site residues, Thr11, Glu12, Lys15, Lys37, and Ser41, implicated by the protein crystal structure studies of Escherichia coli DTBS, were mutated to determine their function in catalysis and substrate binding. Nine mutant enzymes, T11V, E12A, E12D, K15Q, K37L, K37Q, K37R, S41A, and S41C, were overproduced in an E. coli strain lacking a functional endogenous DTBS gene and purified to homogeneity. Replacement of Thr11 with valine resulted in a 24,000-fold increase in the Km(ATP) with little or no change in the Kd(ATP), KM(DAPA) and DTBS k(cat), suggesting an essential role for this residue in the steady-state affinity for ATP. The two Glu12 mutants showed essentially wild-type DTBS activity (slightly elevated k(cat)'s). Unlike wild-type DTBS, E12A had the same apparent KM(DAPA) at subsaturating and saturating ATP concentrations, indicating a possible role for Glu12 in the binding synergy between DAPA and ATP. The mutations in Lys15 and Lys37 resulted in loss of catalytic activity (0.01% and <0.9% of wild-type DTBS k(cat) for K15Q and the Lys37 mutant enzymes, respectively) and higher KM's for both DAPA (40-fold and >100-fold higher than wild-type for the K15Q and Lys37 mutant enzymes, respectively) and ATP (1800-fold and >10-fold higher than wild-type for K15Q and the K37 mutant enzymes, respectively). These results strongly suggest that Lys15 and Lys37 are crucial to both catalysis and substrate binding. S41A and S41C had essentially the same k(cat) as wild-type and had moderate increases in the DAPA and ATP KM and Kd (ATP) values. Replacement of Ser41 with cysteine resulted in larger effects than replacement with alanine. These data suggest that the H-bond between N7 of DAPA and the Ser41 side chain is not very important for catalysis. The catalytic behavior of these mutant enzymes was also studied by pulse-chase experiments which produced results consistent with the steady-state kinetic analyses. X-ray crystallographic studies of four mutant enzymes, S41A, S41C, K37Q, and K37L, showed that the crystals were essentially isomorphous to that of the wild-type DTBS. The models of these mutant enzymes were well refined (1.9 -2.6 A) and showed good similarity to the wild-type enzyme (rmsd of C alpha atoms: 0.16-0.24 A). The crystal structure of S41C complexed with DAPA, Mn2+/Mg2+, and AMPPCP revealed a localized conformational change (rotations of side chains of Cys41 and Thr11) which can account for the changes in the kinetic parameters observed for S41C. The crystal structures of the Lys37 mutant enzymes showed that the positive charge of the side chain of Lys37 is indispensable. Mutations of Lys37 to either glutamine or leucine resulted in a shift of the metal ion (up to 0.5 A) together with side chains of other active site residues which could disrupt the subtle balance between the positive and negative charges in the active site. The conformational change of the phosphate binding loop (Gly8-X-X-X-X-X-Gly14-Lys15-Thr16) upon nucleotide binding observed previously [Huang, W., Jia, J., Gibson, K. J., Taylor, W. S., Rendina, A. R., Schneider, G., & Lindqvist, Y. (1995) Biochemistry 34, 10985] appears to be important to attain the proper active site scaffold.

Dethiobiotin synthetase: the carbonylation of 7,8-diaminonanoic acid proceeds regiospecifically via the N7-carbamate.

Dethiobiotin synthetase (DTBS) catalyzes the penultimate step in biotin biosynthesis, the formation of the ureido ring of dethiobiotin from (7R,8S)-7,8-diaminononanoic acid (7,8-diaminopelargonic acid, DAPA), CO2, and ATP. Solutions of DAPA at neutral pH readily formed a mixture of the N7- and N8-carbamates in the presence of CO2. However, four lines of evidence together indicated that only the N7-carbamate of DAPA was an intermediate in the reaction catalyzed by DTBS. (1) Addition of diazomethane to mixtures of DAPA and [14C]CO2 yielded a mixture of the N7- and N8-methyl carbamate esters, consistent with carbamate formation in free solution. In the presence of excess DTBS (over DAPA), the ratio of N7:N8-methyl carbamate esters recovered was roughly doubled, suggesting that the enzyme preferentially bound the N7-DAPA-carbamate. (2) Both N7- and N8-DAPA-carbamates were observed directly by 1H and 13C NMR in solutions containing DAPA and [13C]CO2. In the presence of excess DTBS (over DAPA) only one carbamate was observed, showing that carbamate binding to the enzyme was regiospecific. 13C NMR of mixtures containing enzyme, [7-15N]DAPA, and [13C]CO2 showed that the enzyme-bound carbamate was at N7 of DAPA. In addition, pulse-chase experiments showed that the binary complex of DTBS and N7-DAPA-carbamate became kinetically committed upon addition of MgATP. (3) The N7-DAPA-carbamate mimic, 3-(1-aminoethyl)nonanedioic acid, in which the carbamate nitrogen was replaced with a methylene group, cyclized to the corresponding lactam in the presence of DTBS and ATP; ADP and P(i) were also formed.(ABSTRACT TRUNCATED AT 250 WORDS)

Mechanism of an ATP-dependent carboxylase, dethiobiotin synthetase, based on crystallographic studies of complexes with substrates and a reaction intermediate.

The crystal structures of six complexes of homodimeric Escherichia coli dethiobiotin synthetase with a variety of substrates, substrate analogs, and products have been determined to high resolution. These include (1) the binary complex of dethiobiotin synthetase and the N7-carbamate of 7,8-diaminononanoic acid, (2) the binary complex of enzyme and the alternate substrate, 3-(1-aminoethyl)-nonanedioic acid, (3) the binary complex of enzyme with the product ADP, (4) the quaternary complex of enzyme, ADP, the N7-carbamate of 7,8-diaminononanoic acid, and Ca2+, (5) the ternary complex of enzyme, the ATP analog adenylyl (beta, gamma-methylene)diphosphonate, and the N7-carbamate of 7,8-diaminononanoic acid, and (6) the quaternary complex of enzyme, the ATP analog adenylyl (beta, gamma-methylene)diphosphonate, 7,8-diaminononanoic acid, and Mn2+. One molecule of each substrate binds to one monomer of the enzyme. ADP and the ATP analogue bind to the classical mononucleotide binding fold with the phosphate groups close to the phosphate binding loop Gly8--Thr16 between beta-strand beta 1 and the N-terminus of alpha-helix alpha 1. The adenine ring is bound in a pocket between beta-strands beta 6 and beta 7. In the quaternary complex with Mn2+, the metal binding site is found in the vicinity of the beta- and gamma-phosphate groups. Two oxygen atoms from the phosphates and oxygen atoms from the side chains of Asp54, Thr16, and Glu115 are ligands to the Mn2+ ion in the quaternary complex. In the complex with ADP and the N7-carbamate of 7,8-diaminononanoic acid prepared in the presence of Ca2+ ions, a different metal binding site is found. The Ca2+ ion is coordinated to an oxygen atom of the alpha-phosphate group of the nucleotide, the side chain of Asp54, and solvent molecules. The 7,8-diaminononanoic acid substrate molecule interacts with residues from both subunits, making the dimer the minimal functional unit. The diamino group binds between the loops after beta 2 and beta 4, and the terminal carboxyl group at the hydrophobic tail of the substrate interacts with the amino terminus of helix alpha 5 and with the side chain of Tyr187 in helix alpha 6 of the second subunit at the monomer-monomer interface. Strong additional electron density close to the N7 nitrogen atom of the 7,8-diaminononanoic acid substrate in some complexes indicates that, even in the absence of added bicarbonate in the crystallization mixture, the carbamylated intermediate is formed in the crystal.(ABSTRACT TRUNCATED AT 400 WORDS)

Phosphorylated thiosugars: synthesis, properties, and reactivity in enzymatic reactions.

A number of phosphorylated thiosugars have been prepared and tested as substrates for metabolic reactions. 6-Thioglucose-6-P is readily synthesized by reaction of 6-tosylglucose with trisodium thiophosphate at pH 10 in aqueous solution; the product has only sulfur between carbon and phosphorus. When ethyl glycerate is tosylated and treated similarly with thiophosphate, a 5:1 mixture of 3-thioglycerate-3-P and the 2-isomer is formed. 6-Thioglucose-6-P is converted by glycolytic enzymes to triose phosphates, 3-thioglycerol-3-P and 3-thioglycerate-3-P, and is oxidized by enzymes of the hexose monophosphate shunt to 5-thioribulose-5-P, which can be converted via phosphoribulokinase and ribulose-bis-P carboxylase into 3-P-glycerate and 3-thioglycerate-3-P. For most of the non-phosphoryl-transferring enzymes there are only moderate effects on Vmax and Km. Phosphoglucoisomerase, however, is very sensitive to the sulfur for oxygen change, with Vmax decreasing 60-fold and Km increasing 15-fold. Surprisingly, phosphoribulokinase has a V/K value for 5-thioribulose-5-P that is over 3 orders of magnitude less than for ribulose-5-P. 6-Thio-glucose-6-P was found to be a substrate for several enzymes that transfer the phosphoryl group. It is as good a substrate for alkaline phosphatase as glucose-6-P, and with phosphoglucomutase it is converted to 6-thioglucose-1-P with a rate that is 11% of the rate of reaction of glucose-1-P, with a Keq value of 45.6. The free energy of hydrolysis of the phosphorylated thiol is thus -7.2 kcal/mol at pH 7.(ABSTRACT TRUNCATED AT 250 WORDS)

Kinetic characterization, stereoselectivity, and species selectivity of the inhibition of plant acetyl-CoA carboxylase by the aryloxyphenoxypropionic acid grass herbicides.

The selective grass herbicides diclofop, haloxyfop, and trifop were found to be potent reversible inhibitors of acetyl-CoA carboxylase from the susceptible species barley, corn, and wheat. Kis values with variable concentrations of acetyl-CoA ranged from 0.01 to 0.06 microM at pH 8.5 depending on the species of grass. Inhibition of the wheat enzyme by diclofop was noncompetitive versus acetyl-CoA with Kis less than Kii and noncompetitive versus MgATP and bicarbonate, but with Kis approximately equal to Kii. Since the apparent inhibition constant was most sensitive to the level of acetyl-CoA, these compounds probably interact with the transcarboxylase site rather than the biotin carboxylation site. With the wheat enzyme the Kis value for the R-(+)-enantiomer of trifop was 1.98 +/- 0.22 times lower than that of the racemic mixture. This confirms the stereoselectivity observed in the whole plant. The enzyme from tolerant broadleaf species (spinach and mung bean) was much less sensitive to these herbicides (Kis values varied from 16 to 515 microM). These data confirm that acetyl-CoA carboxylase is the site of action for the aryloxyphenoxypropionic acid herbicides and may explain their selectivity for monocotyledenous species.

Cyclohexanedione Herbicides Are Selective and Potent Inhibitors of Acetyl-CoA Carboxylase from Grasses.

Biochemical studies of plant species susceptible to the cyclohexanedione herbicides, alloxydim, sethoxydim, and clethodim, have demonstrated that these selective grass herbicides inhibit acetyl-coenzyme A carboxylase, the second enzyme common to both fatty acid and flavonoid biosynthetic pathways. The K(i)s for the cyclohexanediones tested ranged from 0.02 to 1.95 micromolar, depending on the species. The enzyme isolated from broadleaf plants was much less sensitive to inhibition by these herbicides (K(i)s from 53 micromolar to 2.2 millimolar). These results may explain the mechanism of action of these herbicides and their selectivity for monocotyledonous species.

Determination of the screw sense specificity of bovine liver fructokinase.

Fructokinase from beef liver showed a clear reversal in specificity when the two isomers of ATP beta S were used as substrates with Mg2+ and Cd2+, with the Sp isomer having the higher V/K value with Mg2+ and the Rp isomer the higher value with Cd2+. The delta isomer of MgATP is thus the active form of the substrate. The substitution of sulfur for oxygen in the noncoordinated position of the beta-phosphate caused a 102-fold decrease in V/K over the value seen with MgATP, while substitution in the coordinated position gave a 21-fold decrease over the V/K value seen with CdATP. The Km values were little affected by sulfur substitution, showing that the wrong screw sense isomers were nonproductively bound almost as well as the correct ones. When ADP alpha S was used as a substrate in the reverse reaction, the Sp isomer showed the highest V/K value with both Mg2+ and Cd2+, suggesting that the metal ion is not coordinated to the alpha-phosphate during transphosphorylation. The failure of CrATP to act as a substrate for fructokinase suggests that the enzyme inserts one of its side chains into the inner coordination sphere of the metal ion during the reaction.

Use of multiple isotope effects to study the mechanism of 6-phosphogluconate dehydrogenase.

The multiple isotope effect method of Hermes et al. [Hermes, J. D., Roeske, C. A., O'Leary, M. H., & Cleland, W. W. (1982) Biochemistry 21, 5106-5114] has been used to study the mechanism of the oxidative decarboxylation catalyzed by 6-phosphogluconate dehydrogenase from yeast. 13C kinetic isotope effects of 1.0096 and 1.0081 with unlabeled or 3-deuterated 6-phosphogluconate, plus a 13C equilibrium isotope effect of 0.996 and a deuterium isotope effect on V/K of 1.54, show that the chemical reaction after the substrates have bound is stepwise, with hydride transfer preceding decarboxylation. The kinetic mechanism of substrate addition is random at pH 8, since the deuterium isotope effect is the same when either NADP or 6-phosphogluconate or 6-phosphogluconate-3-d is varied at fixed saturating levels of the other substrate. Deuterium isotope effects on V and V/K decrease toward unity at high pH at the same time that V and V/K are decreasing, suggesting that proton removal from the 3-hydroxyl may precede dehydrogenation. Comparison of the tritium effect of 2.05 with the other measured isotope effects gives limits of 3-4 on the intrinsic deuterium and of 1.01-1.05 for the intrinsic 13C isotope effect for C-C bond breakage in the forward direction and suggests that reverse hydride transfer is 1-4 times faster than decarboxylation.

A novel method for determining rate constants for dehydration of aldehyde hydrates.

A method has been developed for calculating rate constants for dehydration of aldehydes that induce ATPase reactions by kinases and where 18O is transferred from the aldehyde or its hydrate to inorganic phosphate during the reaction. The method involves measurement of the fraction of 18O in phosphate by 31P NMR after the ATPase reaction has proceeded for several minutes with zero-order kinetics. The reaction is started by addition of the aldehyde in a small volume of H2 18O, and the speed of washout of 18O by reversible dehydration relative to the rate of the ATPase reaction allows calculation of the rate constants if the hydration equilibrium constant is known from the proton NMR spectrum of the aldehyde. Dehydration rate constants (s-1 at pH 8-8.5, 0.1 M buffer, 25 degrees C) for the following aldehydes (all over 95% hydrated) and kinases used are as follows: D-glyceraldehyde with glycerokinase, 0.03; 2,5-anhydro-D-mannose 6-phosphate with fructose-6-phosphate kinase, 0.025; 2,5-anhydro-D-mannose or 2,5-anhydro-D-talose with fructokinase, 0.029 and 0.017, respectively; D-gluco-hexodialdose with hexokinase, 0.068. With betaine aldehyde and choline kinase or glyoxylate and pyruvate kinase, no 18O was transferred to phosphate during the ATPase reactions. However, the dehydration rate constant for glyoxylate (0.007 s-1 at pH 7 extrapolated to zero buffer concentration and up to 0.11 s-1 at pH 9.0 with 0.3 M buffer) was determined by extrapolating the initial rate of reduction of the free aldehyde catalyzed by lactate dehydrogenase to infinite enzyme levels.(ABSTRACT TRUNCATED AT 250 WORDS)

Mechanisms of aldehyde-induced adenosinetriphosphatase activities of kinases.

Aldehyde analogues of the normal alcohol substrates induce ATPase activities by glycerokinase (D-glyceraldehyde), fructose-6-phosphate kinase (2,5-anhydromannose 6-phosphate), fructokinase (2,5-anhydromannose or 2,5-anhydrotalose), hexokinase (D-gluco-hexodialdose), choline kinase (betaine aldehyde), and pyruvate kinase (glyoxylate). Since purified deuterated aldehydes give V and V/K isotope effects near 1.0 for glycerokinase, fructokinase with 2,5-anhydro[1-2H]talose, hexokinase, choline kinase, and pyruvate kinase, the hydrates of these almost fully hydrated aldehydes are the activators of the ATPase reactions. Fructose-6-phosphate kinase and fructokinase with 2,5-anhydro[1-2H]mannose show V/K deuterium isotope effects of 1.10 and 1.22, respectively, suggesting either that both hydrate and free aldehyde may be activators (predicted values are 1.37 if only the free aldehyde activates the ATPase) or, more likely, that the phosphorylated hydrate breaks down in a rate-limiting step on the enzyme while MgADP is still present and the back-reaction to yield free hydrate in solution is still possible. 18O was transferred from the aldehyde hydrate to phosphate during the ATPase reactions of glycerokinase, fructose-6-phosphate kinase, fructokinase, and hexokinase but not with choline kinase or pyruvate kinase. Thus, direct phosphorylation of the hydrates by the first four enzymes gives the phosphate adduct of the aldehyde, which decomposes nonenzymatically, while with choline kinase and pyruvate kinase the hydrates induce transfer to water (metal-bound hydroxide or water with pyruvate kinase on the basis of pH profiles). Observation of a lag in the release of phosphate from the glycerokinase ATPase reaction at 15 degrees C supports the existence of a phosphorylated hydrate intermediate with a rate constant for breakdown of 0.035-0.043 s-1 at this temperature. Kinases that phosphorylate creatine, 3-phosphoglycerate, and acetate did not exhibit ATPase activities in the presence of keto or aldehyde analogues (N-methylhydantoic acid, D-glyceraldehyde 3-phosphate, and acetaldehyde, respectively), possibly because of the absence of an acid-base catalytic group in the latter two cases. These analogues were competitive inhibitors vs. the normal substrates, and in the latter case, the hydrate of acetaldehyde was shown to be the inhibitory species on the basis of the deuterium isotope effect on the inhibition constant.

Characterization of 3-iron ferredoxins by means of the linear electric field effect in EPR.

We describe a method for the differentiation of 3-iron from 2-iron and 4-iron Fe/S proteins based on consideration of both the magnetic field dependence of shifts in g induced by an externally applied electric field (LEFE) and the continuous wave EPR spectra properties. The magnetic field dependence and the magnitude of the LEFE for 3-iron ferredoxins are similar to those for 4-iron ferredoxins but differ considerably from those for 2-iron ferredoxins or for high potential iron proteins. Furthermore, as 3-iron ferredoxins and high potential iron proteins are EPR-active when oxidized while 2-iron and 4-iron ferredoxins are only EPR-active when reduced, the differentiation among all of them can be made on the basis of both continuous wave EPR and LEFE properties, but not by each individually.

Substrate synergism and the kinetic mechanism of yeast hexokinase.

Michaelis constants for MgATP with yeast hexokinase vary from 28 microM with D-mannose to above 4 mM for the slow ATPase reaction, with the different values reflecting the degree of synergism in binding of MgATP and the sugar substrate. The best substrates show the greatest synergism, but the correlation is not exact. Similar synergistic binding between MgADP or its methylene analogue and phosphorylated sugars is seen. Product inhibiton of MgADP vs. MgATP and vice versa appears noncompetitive at low levels of variable substrate but becomes competitive at high levels. These patterns show that MgATP can combine with E-glucose-6-P (Ki = 4 mM) and MgADP with E-glucose (Ki = 1.6 mM). Isotope partitioning studies with glucose or glucose-6-P have determined the rates of release of these substrates from binary and ternary complexes and, together with reverse isotope exchange studies and the product inhibition studies mentioned above, have shown that the kinetic mechanism is a somewhat random one in which dissociation of sugars from productive ternary complexes is very slow, but release from nonproductive ternary complexes occurs at rates similar to those from binary enzyme-sugar complexes. D-Arabinose-5-P has a Km of 4.6 mM and a Vmax 5% that for glucose-6-P, confirming that the high Km for D-arabinose in the forward direction is caused by the low proportion in the furanose form. The dissociation constant of MgADP in the absence of sugars was determined from the Ki of 5.8 mM for MgADP as a competitive inhibitor vs. MgATP of the slow ATPase reaction.