Conformations of nicotinamide coenzymes bound to dehydrogenases determined by transferred nuclear Overhauser effects

Transferred nuclear Overhauser enhancement was used to examine the conformation of NAD+ and NADP+ bound to glucose-6-phosphate dehydrogenase and glutamate dehydrogenase and of NAD+ bound to lactate dehydrogenase. The results demonstrate that the conformation of the nicotinamide-ribose bond is anti for dehydrogenases with A stereospecificity and syn for dehydrogenases with B stereospecificity. In those dehydrogenases that bind both NAD+ and NADP+, significant differences occur in the conformations of the bound nicotinamide coenzymes.

Crystallization and preliminary x-ray data for glucose-6-phosphate dehydrogenase from Leuconostoc mesenteroides

The enzyme glucose-6-phosphate dehydrogenase from Leuconostoc mesenteroides has been crystallized from phosphate buffer in a form suitable for x-ray crystallographic studies. The crystals diffract to better than 2.4 A. The spacegroup is P3121 (P3221) a = 105.8 A, c = 225.1 A, V = 2.18 X 10(6) A3. The asymmetric unit probably contains a single dimer.

Glucose-6-phosphate dehydrogenase from Leuconostoc mesenteroides: revised kinetic mechanism and kinetics of ATP inhibition

The kinetic mechanisms of the NAD- and NADP-linked reactions catalyzed by glucose-6-phosphate dehydrogenase from Leuconostoc mesenteroides were examined using product inhibition, dead-end inhibition and alternate substrate experiments. The results are consistent with a steady-state random mechanism for the NAD-linked and an ordered, sequential mechanism with NADP+ binding first for the NADP-linked reaction. Thus, the enzyme can bind NADP+, NAD+, and glucose 6-phosphate, but the enzyme-glucose 6-phosphate complex can react only with NAD+, not with NADP+. This affects the rate equation for the NADP-linked reaction by introducing a term for a dead-end enzyme-glucose 6-phosphate complex. The kinetic mechanisms represent revisions of those proposed previously (C. Olive, M.E. Geroch, and H.R. Levy, 1971, J. Biol. Chem. 246, 2047-2057) and provide a kinetic basis for the regulation of coenzyme utilization of the enzyme by glucose 6-phosphate concentration (H.R. Levy, and G.H. Daouk, 1979, J. Biol. Chem. 254, 4843-4847) and NADPH/NADP+ concentration ratios (H.R. Levy, G.H. Daouk, and M.A. Katopes, 1979, Arch, Biochem. Biophys. 198, 406-413). The kinetic mechanisms were found to be the same at pH 6.2 and pH 7.8. The kinetics of ATP inhibition of the NAD- and NADP-linked reactions were examined at pH 6.2 and pH 7.8. The results are interpreted in terms of ATP addition to binary enzyme-coenzyme and enzyme-glucose 6-phosphate complexes.

Glucose-6-phosphate dehydrogenase from Leuconostoc mesenteroides. Isolation and sequence of a peptide containing an essential lysine

Interaction of glucose-6-phosphate dehydrogenase from Leuconostoc mesenteroides with pyridoxal 5'-phosphate and sodium borohydride leads to inactivation and modification of two lysine residues per enzyme dimer that are thought to bind glucose 6-phosphate [Milhausen, M., & Levy, H.R. (1975) Eur. J. Biochem. 50, 453-461]. The amino acid sequence surrounding this lysine residue is reported. Following tryptic hydrolysis of the modified enzyme, two peptides, each containing one pyridoxyllysine residue, were purified to homogeneity and subjected to automated Edman degradation. The sequences revealed that one of these, a heptapeptide, was derived from the other, containing 11 amino acids. Supporting evidence for the role of the modified lysine is provided in the following paper [Haghighi, B., & Levy, H.R. (1982) Biochemistry (second paper of three in this issue)]. End-group analysis of the native enzyme revealed that valine is the N-terminal and glycine the C-terminal amino acid and provides support for the identity of the enzyme's two subunits.

Glucose-6-phosphate dehydrogenase from Leuconostoc mesenteroides. Kinetics of reassociation and reactivation from inactive subunits

Glucose-6-phosphate dehydrogenase from Leuconostoc mesenteroides is denatured in 8 M urea and dissociated into its two inactive subunits. Denaturation leads to an approximately 80% decrease in protein fluorescence and a 20-nm red shift in the emission maximum. Upon dilution, the urea-treated enzyme regains catalytic activity (approximately 70%). The reactivated enzyme is indistinguishable from the native enzyme based on a number of physicochemical and enzymological criteria. The kinetics of renaturation and reactivation were monitored by measuring the rates of regain of native fluorescence and appearance of activity and the accessibility of histidine residues toward diethyl pyrocarbonate modification. Regain of native fluorescence was too rapid to measure at 25 degrees C; at 5 degrees C the initial phase was also too rapid, but a slower phase was monitored and shown to obey first-order kinetics with k = (5.9 +/- 1.3) x 10(-3) s-1. Reappearance of activity was measured at several protein concentrations; reactivation followed second-order kinetics with k = (4.85 +/- 0.47) x 10(-3) M-1 min-1. Reactivation was stimulated to different degrees by either the initial or delayed addition of NAD+, NADP+, or glucose 6-phosphate. During the initial, rapid phase of renaturation, approximately 3 of the enzyme's 12 histidine residues become unreactive toward diethyl pyrocarbonate; concomitant with the subsequent reactivation, approximately 7 more histidines become inaccessible to diethyl pyrocarbonate. The data are consistent with a model for enzyme renaturation and reactivation in which the unfolded subunits rapidly refold to an inactive structure that can dimerize slowly to generate native enzyme. Specific ligands stimulate reactivation by binding to refolded subunits or incompletely folded dimers.

Simultaneous analysis of NAD- and NADP-linked activities of dual nucleotide-specific dehydrogenases. Application to Leuconostoc mesenteroides glucose-6-phosphate dehydrogenase

A method is described which enables one to assay simultaneously the NAD- and NADP-linked reactions of dehydrogenases which can utilize both coenzymes. The method is based on the fact that the thionicotinamide analogs of NADH and NADPH absorb light maximally at 400 nm, a wavelength sufficiently far removed from the absorbance maximum of NADH and NADPH to permit measurements of the simultaneous reduction of NAD+ (or NADP+) and the thionicotinamide analog of NADP+ (or NAD+). Application of the method to glucose-6-phosphate dehydrogenase from Leuconostoc mesenteroides reveals differential effects of glucose 6-phosphate concentration on the NAD- and NADP-linked reactions catalyzed by this enzyme which can not be detected by conventional assay procedures and which may have regulatory significance.

Glucose-6-phosphate dehydrogenase from lactating rat mammary gland and R3230AC adenocarcinoma

A number of properties of glucose-6-phosphate dehydrogenase from lactating rat mammary gland and R3230AC rat mammary adenocarcinoma are compared. The main electrophoretic forms of the enzyme from these sources are indistinguishable with respect to charge and molecular weight whereas the minor forms show differences in these properties. The subunit molecular weight and steroid inhibition of the enzymes from the lactating gland and tumor are not significantly different. These results are contrasted with similar studies in mice.

Metal ion requirement and tryptophan inhibition of normal and variant anthranilate synthase-anthranilate 5-phosphoribosylpyrophosphate phosphoribosyltransferase complexes from Salmonella tyrhimrium

1. Both Mn2+ and Co2+ can replace Mg2+ as the required divalent cation for all activities of the enzyme complex between anthranilate synthase (chorismate pyruvate-lyase (amino-accepting), EC 4.1.3.27) and anthranilate-5-phosphoribosylpyrophosphate phosphoribosyltransferase (N(5'-phosphoribosyl)-anthranilate:pyrophosphate phosphoribosytransferase, EC 2.4.2.18) from Salmonella typhimurium. They have much lower apparent Km values than Mg2+, both for glutamine-dependent anthranilate synthase (Mn2+ = 1.1 muM, Co2+ - 2.6 muM, Mg2+ = 83 muM) and for phosphoribosyltransferase (Mn2+ = 16 muM, Co2+ = 14.6 muM, Mg2+ = 133 muM). The ratio of total Mg2+ to total Mn2+ found in a cell extract of S. typhimurium trpE2 , the source of normal enzyme complex, was found to be 350, suggesting that Mg2+ is probably utilized by the enzyme complex in vivo under our growth conditions. 2. An enzyme complex has been isolated from a mutant strain of S. typhimurium (SO-515) that has a variation in the anthranilate synthase subunit which is thought to be a single amino acid substitution. This variation causes glutamine-dependent anthranilate synthase to be hypersensitive to feedback inhibition by tryptophan (Ki = 0.4 muM compared to Ki = 20 muM for normal enzyme complex). The phosphoribosyltransferase in the variant enzyme complex is also hypersensitive to tryptophan but the kinetics are complex and involve activation by tryptophan in the presence of low amounts of 5-phosphoribosyl 1-pyrophosphate. 3. In the variant enzyme complex the apparent Km for Mg2+ is elevated to 360 muM for glutamine-linked anthranilate synthase but reduced to 75 muM for phosphoribosyltransferase. 4. These results suggest that the variant enzyme complex has altered tertiary and quaternary structures and that regulation of both activities is effected by tryptophan binding to only anthranilate synthase.

Anthranilate synthase-anthranilate 5-phosphoribosylpyrophosphate phosphoribosyltransferases from Salmonella typhimurium. Purification of the enzyme complex and analysis of multiple forms

1. The anthranilate synthase-anthranilate 5-phosphoribosylpyrophosphate phosphoribosyltransferase enzyme complex (chorismate pyruvatelyase (amino-accepting), EC 4.1.3.27) - (N-(5'-phosphoribosyl)-anthranilate: pyrophosphate phosphoribosyltransferase, EC 2.4.2.18), from Salmonella typhimurium has been purified with high yields to homogeneity. Sodium dodecyl sulfate gel electrophoresis of the purified enzyme complex revealed one major band containing 96% of the protein. The final yield of enzyme complex activity ranged from 30 to 60%. The absorbance spectrum of enzyme complex showed a peak at 280 nm and fine structure with peaks at 253, 259, 266 and 269 nm. These latter wavelengths correspond closely with the known absorbance maxima of phenylalanine. 2. When purified enzyme complex was subjected to standard gel electrophoresis, a four band pattern of protein peaks was consistently observed. The major enzyme complex band was apparently the native tetramer, having a molecular weight 280 000 and containing ammonia- and glutamine-dependent anthranilate synthase activity. The other three bands were molecular weight isomers of the major enzyme complex band. Two forms of molecular weight isomers were present: dimers and an aggregate of the native enzyme complex. The molecular weight isomers of the enzyme complex may represent forms generated by aggregation and denaturation of the native enzyme complex. 3. A new and highly sensitive spectrophotometric assay for phosphoribosyl-transferase is described. The method is based upon the difference in extinction coefficients between anthranilate and N-(5'-phosphoribosyl)anthranilate.

Anthranilate synthetase-anthranilate 5-phosphoribosylpyrophosphate phosphoribosyl-transferase from Salmonella typhimurium. Inactivation of glutamine-dependent anthranilate synthetase by agarose-bound anthranilate

Exposure of the anthranilate synthetase-anthranilate phosphoribosyltransferase enzyme complex (chorismate pyruvate-lyase (amino-accepting) EC 4.1.3.27 and N-(5-phosphoribosyl)-anthranilate pyrophosphate phosphoribosyl-transferase, EC 2.4.2.18) from Salmonella typhimurium to agarose-bound anthranilate led to the slow inactivation of glutamine-dependent anthranilate synthetase activity, an activity dependent on protein-protein interaction in the enzyme complex. Region I of phosphoribosyltransferase, the location of the enzyme complex glutaminase activity, is the site of alteration. Phosphoribosyltransferase and NH3-dependent anthranilate synthetase activities and trypto phan regulation of phosphoribosyltransferase were unaffected by the anthranilate matrix. The molecular weight (280 000) and subunit molecular weight (62 000) of the enzyme complex eluted from an anthranilate matrix were identical to those of enzyme complex purified by classical methodology. The enzyme complex could be partially protected against inactivation by storiing in 0.1 M L-glutamine or 30% glycerol and completely protected by storing in 50% glycerol at -18 degrees C. Evidence is presented that the anthranilate matrix acts as a hydrophobic matrix and may be binding to and altering a hydrophobic region in the enzyme complex. The anthranilate matrix provides a convenient tool for altering a specific region of an enzyme complex without covalent modification. At the same time, the results demonstrate that affinity matrices are not necessarily innocuous but may subject macromolecules to an adverse environment not previously recognized.

Spin label and lanthanide binding sites on glyceraldehyde-3-phosphate dehydrogenase

The electron spin resonance spectrum of rabbit muscle D-glyceraldehyde-3-phosphate dehydrogenase spin-labelled with 4-(2-iodoacetamido)-2,2,6,6-tetramethylpiperidinooxyl has two components. One component is due to a spin label highly immobilized on the enzyme surface and the other to a nitroxyl group able to tumble more rapidly. The spin-labelled enzyme is inactive. Selective modification of the active site cysteine residue (149) and determinations of total sulphydryl content implicate this residue as the site of the immobile spin-label. The mobile spin label is attached to another sulphydryl group. Crystallographic studies on the human muscle enzyme (Watson, H.C., Duee, E. and Mercer, W.D. (1972) Nat. New Biol., 240, 130) have located a binding site for samarium ion in the active centre. Addition of the paramagnetic gadolinium ion to spin-labelled enzyme reduces the intensity of both the spin label signals (by 72% for the mobile and by 11% for the immobile component). This indicates that the metal ion site (Kd equals 0.7 mM) is close to both types of spin label. Measurements of the effect of gadolinium-protein binding on the relaxation rate of solvent water protons enable the enzyme-bound spin label-metal ion distances to be tentatively estimated as 15 angstrom.

Evidence for an essential lysine in glucose-6-phosphate dehydrogenase from Leuconostoc mesenteroides

1. Pyridoxal 5'-phosphate inhibits glucose-6-phosphate dehydrogenase from Leuconostoc mesenteroides reversibly which Ki equals 0.04-0.06 mM. 2. This inhibition is competitive with respect to glucose 6-phosphate and non-competitive with respect to NADP+ or NAD+. Interaction between enzyme and excess pyridoxal 5'-phosphate follows pseudo-first-order kinetics and indicates that one molecule of inhibitor reacts with each active unit of enzyme. 3. Substrate and coenzyme protect the enzyme from inhibition by pyridoxal 5'-phosphate. Dissociation constants for NADP+ and glucose 6-phosphate were determined from their effects on the kinetics of enzyme--inhibitor interaction. 4. Reaction of the enzyme with pyridoxal 5'-phosphate produces a typical Schiff-base absorbance peak at 430 nm. Subsequent reduction with sodium borohydride leads to spectral changes characteristic for the formation of a secondary amine. 5. The irreversibly inactivated enzyme thus produced contains two moles of inhibitor per mole of enzyme (two subunits per mole). After protein hydrolysis, N-6-pyridoxyllysine can be identified by paper chromatography. 6. The enzyme is inhibited irreversibly by 1-fluoro-2,4-dinitrobenzene, even in the presence of excess 2-mercaptoethanol. At least one dinitrophenyl group is bound per active unit of enzyme; 4 to 5 moles of dinitrophenyl group are bound per mole of enzyme. NADP+ AND GLUCOSE 6-PHOSPHATE PROTECT AGAINST INHIBITION BY 1-FLUORO-2,4-DINITROBENZENE. The absorption spectrum of dinitrophenyl-enzyme corresponds to that for dinitrophenylated amino groups. 7. These studies indicate that there is an essential lysine at the active site of the enzyme. It is suggested that the function of this lysine is to bind glucose 6-phosphate. 8. It is proposed that a group of "active lysine" proteins may exist (in analogy with the "active serine" enzymes), which share a common structural feature at their substrate-binding site and to which pyridoxal 5'-phosphate binds specifically.