Isolation and various properties of nucleoside monophosphate kinases from Escherichia coli

A technique is proposed for isolation of nucleosidemonophosphate kinases--AMP-kinase (EC 2.7.4.11), GMP-kinase (EC 2.7.4.8), CMP-kinase (EC 2.7.4.14), UMP-kinase (EC 2.7.4.14) and TMP-kinase (EC 2.7.4.9)--from E. coli MRE-600. It involves cell destroying, precipitation of nucleic acids with polyethyleneimine, fractionation with ammonium sulphate followed by chromatography on different carriers (DEAE-Toyopearl-650 M, Matrex gel Blue A, Matrex gel Red A). The technique enables all the five enzymes to be obtained separately and without contaminations with nucleotide dephosphorylating enzymes. For all the enzymes the pH optimum was found to range from 6.5 to 8.0, and Mg2+ ions were found to be the best activator for all the enzymes studied. The substrate specificity was investigated with respect to acceptors and donors of the phosphate groups. The enzymes showed strict specificity to the heterocyclic base of the acceptor phosphate group. AMP-, GMP- and CMP-kinases phosphorylated the corresponding deoxynucleoside monophosphates less effectively than ribonucleoside monophosphates. ATP was found to be the most effective phosphate donor for all the enzymes under study.     

Comparative modelling and immunochemical reactivity of Escherichia coli UMP kinase

Bacterial UMP kinases do not exhibit any sequence homology with other nucleoside monophosphate kinases described so far, and appear under oligomeric forms, submitted to complex regulation by nucleotides. We propose here a structural model of UMP kinase from Escherichia coli based on the conservation of the fold of carbamate kinase whose crystal structure was recently solved. Despite sequence identity of only 18% over 203 amino acids, alignment of UMP kinase from E. coli with carbamate kinase from Enterococcus faecalis by hydrophobic cluster analysis and threading suggested the conservation of the overall structure, except for a small subdomain (absent in UMP kinase). The modelled dimer suggested conservation of the dimer interface observed in carbamate kinase while interaction of UMP kinase with a monoclonal antibody (Mab 44-2) suggests a three in-plane dimer subunit arrangement. The model was analyzed in light of various modified forms of UMP kinase obtained by site-directed mutagenesis.     

The structure of the Escherichia coli nucleoside diphosphate kinase reveals a new quaternary architecture for this enzyme family

Nucleoside diphosphate kinase (NDPK) catalyzes the transfer of gamma-phosphate from nucleoside triphosphates to nucleoside diphosphates. The subunit folding and the dimeric basic structural unit are remarkably the same for available structures but, depending on species, dimers self-associate to form hexamers or tetramers. The crystal structure of the Escherichia coli NDPK reveals a new tetrameric quaternary structure for this protein family. The two tetramers differ by the relative orientation of interacting dimers, which face either the convex or the concave side of their central sheet as in either Myxococcus xanthus (type I) or E. coli (type II), respectively. In the type II tetramer, the subunits interact by a new interface harboring a zone called the Kpn loop as in hexamers, but by the opposite face of this loop. The evolutionary conservation of the interface residues indicates that this new quaternary structure seems to be the most frequent assembly mode in bacterial tetrameric NDP kinases.     

Structure and reaction mechanism of L-rhamnulose kinase from Escherichia coli

Bacterial L-rhamnulose kinase participates in the degradation of L-rhamnose, which is ubiquitous and particularly abundant in some plants. The enzyme catalyzes the transfer of the gamma-phosphate group from ATP to the 1-hydroxyl group of L-rhamnulose. We determined the crystal structures of the substrate-free kinase and of a complex between the enzyme, ADP and L-fructose, which besides rhamnulose is also processed. According to its chainfold, the kinase belongs to the hexokinase-hsp70-actin superfamily. The closest structurally known homologue is glycerol kinase. The reported structures reveal a large conformational change on substrate binding as well as the key residues involved in catalysis. The substrates ADP and beta-L-fructose are in an ideal position to define a direct in-line phosphoryl transfer through a bipyramidal pentavalent intermediate. The enzyme contains one disulfide bridge at a position where two homologous glycerol kinases are regulated by phosphorylation and effector binding, respectively, and it has two more pairs of cysteine residues near the surface that are poised for bridging. However, identical catalytic rates were observed for the enzyme in reducing and oxidizing environments, suggesting that regulation by disulfide formation is unlikely.     

End-product regulation and kinetic mechanism of guanosine-inosine kinase from Escherichia coli

Escherichia coli guanosine-inosine kinase was overproduced, purified, and characterized. The native and subunit molecular weights were 85,000 and 45,000, respectively, indicating that the enzyme was a dimer. A pI of 6.0 was obtained by isoelectric focusing. In addition to ATP, it was found that deoxyadenosine 5'-triphosphate, UTP, and CTP could serve as phosphate donors. The phosphate acceptors were guanosine, inosine, deoxyguanosine and xanthosine, but not adenosine, cytidine, uridine, or deoxythymidine. Maximum activity was attained at an ATP/Mg2+ concentration ratio of 0.5. In the presence of pyrimidine nucleotides, enzyme activity was slightly increased, while it was markedly inhibited by GDP and GTP. Initial velocity and product inhibition studies support an ordered Bi Bi mechanism in which guanosine was the first substrate to bind and GMP was the last product to be released. Guanosine kinase may be a regulatory enzyme that has a role in modulating nucleotide levels.     

Kinetics and equilibria of pyrimidine nucleoside monophosphate kinase from human erythrocytes.

The common type of pyrimidine nucleoside monophosphate kinase (ATP:CMP phosphotransferase, EC 2.7.4.14), purified 50 000-fold from human erythrotes, reacted with a wide variety of nucleotides, but only ATP, dATP, UMP and CMP were good substrates. The optimum Mg2+ concentration, 2-3 mM, was generally independent of substrate concentration, of the nature of the substrate, and of the direction of the reaction. Kinetic studies indicated that a ternary complex was formed, that the substrates were bound at two unlike sites, and that the order of addition of substrates was random. Equilibrium constants were ATP + UMP 0.98, ATP + CMP 1.59, dATP + UMP 1.13, and ATP + AMP 1.20.     

A membrane-bound diacylglycerol kinase that selectively phosphorylates arachidonoyl-diacylglycerol. Distinction from cytosolic diacylglycerol kinase and comparison with the membrane-bound enzyme from Escherichia coli

The membrane-bound diacylglycerol kinase from Swiss 3T3 cells (M-DG kinase) was characterized with a mixed micellar assay system, and compared with the cytosolic diacylglycerol kinase from 3T3 cells and with the membrane-bound diacylglycerol kinase from Escherichia coli. M-DG kinase selectively phosphorylated arachidonoyl-diacylglycerols, at a rate 2- to 8-fold higher than that for other naturally occurring long-chain diacylglycerols. In contrast, the cytosolic 3T3 enzyme exhibited little or no selectivity among long-chain diacylglycerols but had higher activity with more soluble substrates such as 1,2-didecanoylglycerol. Comparison of the properties of M-DG kinase with those of the bacterial membrane-bound enzyme revealed that selectivity for arachidonoyl-diacylglycerol was unique to the mammalian enzyme. All three kinases were activated by phosphatidylserine, but activation did not alter the arachidonoyl selectivity of M-DG kinase. Phosphatidylserine activated M-DG kinase by increasing Vm and decreasing the apparent Km for diacylglycerol. High concentrations of diacylglycerol reduced the Ka for phosphatidylserine, but did not abolish the phosphatidylserine requirement for maximum activity. Examination of the thermal lability of M-DG kinase revealed that this enzyme was rapidly and selectively inactivated by preincubation with its preferred substrate. This novel effect may have obscured previous attempts to discern substrate selectivity. Taken together, the results provide evidence that M-DG kinase is an arachidonoyl-diacylglycerol kinase that may participate in the formation of arachidonoyl-enriched species of phosphatidylinositol.     

Structural and functional characterization of Escherichia coli UMP kinase in complex with its allosteric regulator GTP

Bacterial UMP kinases are essential enzymes involved in the multistep synthesis of UTP. They are hexamers regulated by GTP (allosteric activator) and UTP (inhibitor). We describe here the 2.8 angstroms crystal structure of Escherichia coli UMP kinase bound to GTP. The GTP-binding site, situated at 15 angstroms from the UMP-binding site and at 24 angstroms from the ATP-binding site, is delineated by two contiguous dimers. The overall structure, as compared with those bound to UMP, UDP, or UTP, shows a rearrangement of its quaternary structure: GTP induces an 11 degrees opening of the UMP kinase dimer, resulting in a tighter dimer-dimer interaction. A nucleotide-free UMP kinase dimer has an intermediate opening. Superposition of our structure with that of archaeal UMP kinases, which are also hexamers, shows that a loop appears to hamper any GTP binding in archeal enzymes. This would explain the absence of activating effect of GTP on this group of UMP kinases. Among GTP-binding residues, the Asp-93 is the most conserved in bacterial UMP kinases. In the previously published structures of E. coli UMP kinase, this residue was shown to be involved in hydrogen bonds between the subunits of a dimer. Its substitution by an alanine decreases the cooperativity for UTP binding and suppresses the reversal by GTP of UTP inhibition. This demonstrates that the previously described mutual exclusion of these two nucleotides is mediated by Asp-93.     

Diglyceride kinase from Escherichia coli. Modulation of enzyme activity by glycosphingolipids

Diglyceride kinase was purified from membranes of Escherichia coli K-12 using organic solvents. The enzyme apoprotein depended on lipids, such as cardiolipin (diphosphatidylglycerol), phosphatidylcholine or 1-monooleoylglycerol, for activity with 1,2-dipalmitoylglycerol. Mixed brain cerebrosides and gangliosides as well as defined ganglioside fractions and synthetic lactocerebroside were devoid of lipid cofactor activity. However, all these glycosphingolipids were strong inhibitors of activation by phosphatidylcholine. When cardiolipin was used as lipid activator with the detergent, Triton X-100, as solubilizing agent, the addition of mixed or purified gangliosides first (at about 0.4 mM) resulted in additional activation, but higher ganglioside concentrations were strongly inhibitory. Both effects were absolutely dependent on the presence of lipid-bound sialic acid and were not given by cerebrosides, by free sialic acid or by sialyl-lactose. The stimulating and inhibitory effects of glycosphingolipids could also be demonstrated when 1-monooleoylglycerol was used as substrate, lipid activator and solubilizing agent at the same time. The modulation of kinase activity by glycosphingolipids is discussed at the level of lipid/protein interactions.     

Diglyceride kinase from Escherichia coli: Modulation of enzyme activity by glycosphingolipids

Diglyceride kinase was purified from membranes of Escherichia coli K-12 using organic solvents. The enzyme apoprotein depended on lipids, such as cardiolipin (diphosphatidylglycerol), phosphatidylcholine or 1-monooleoylglycerol, for activity with 1,2-dipalmitoylglycerol. Mixed brain cerebrosides and gangliosides as well as defined ganglioside fractions and synthetic lactocerebroside were devoid of lipid cofactor activity. However, all these glycosphingolipids were strong inhibitors of activation by phosphatidylcholine. When cardiolipin was used as lipid activator with the detergent, Triton X-100, as solubilizing agent, the addition of mixed or purified gangliosides first (at about 0.4 mM) resulted in additional activation, but higher ganglioside concentrations were strongly inhibitory. Both effects were absolutely dependent on the presence of lipid-bound sialic acid and were not given by cerebrosides, by free sialic acid or by sialyl-lactose. The stimulating and inhibitory effects of glycosphingolipids could also be demonstrated when 1-monooleoylglycerol was used as substrate, lipid activator and solubilizing agent at the same time. The modulation of kinase activity by glycosphingolipids is discussed at the level of lipid/protein interactions.     

Behavior of various ribo- and deoxyribonucleosides, nucleoside monophosphate kinases, and nucleoside diphosphokinase on Blue Sepharose affinity columns.

A rapid and effective separation and purification of cytidylate-deoxycytidylate-uridylate kinase, adenylate kinase, and nucleoside diphosphokinase has been achieved with Blue Sepharose CL6B chromatography. When crude extracts of human crythrocytes or acute myelocytic leukemia cells were applied to the column, adenylate kinase, cytidylate-deoxycytidylate-uridylate kinase, and nucleoside diphosphokinase were retarded while guanylate kinase, cytidine kinase, uridine kinase, and deoxycytidine kinase were unabsorbed. The buffers required to elute the retarded kinases depended on the amount of sample applied to the column.     

Purification and characterization of a new ribopolynucleotide synthesizing enzyme from Escherichia coli.

A new type of ribopolynucleotide-synthesizing enzyme was found both on cytoplasmic membranes and in protein-DNA complexes isolated from Escherichia coli. The enzyme was purified by exploiting a specific, reversible enzyme aggregation with ATP and spermidine. The purified enzyme (more than 90% pure) was free from other enzymatic activities such as ATPase and polynucleotide phosphorylase. The enzyme (molecular weight 270,000 ± 15%) contains two kinds of polypeptide chain (molecular weights 91,000 ± 10%, and 45,000 ± 10%) and these polypeptides are not common with those of DNA-dependent RNA polymerase. The enzyme catalyses the synthesis of ribopolynucleotides from nucleoside triphosphates in the presence of 1 mm-MgCl₂. Rifampicin and streptolydigin do not affect the enzyme reaction.     

Regulation of nucleoside diphosphate kinase and an alternative kinase in Escherichia coli: role of the sspA and rnk genes in nucleoside triphosphate formation

We have previously reported that two genes cloned from a cosmid library of Escherichia coli can restore mucoidy to an algR2 mutant of Pseudomonas aeruginosa . AlgR2 is a protein involved in the regulation of nucleoside diphosphate kinase (Ndk) as well as alginate synthesis in P. aeruginosa . One of the E. coli genes, rnk , encodes a 14.9 kDa protein with no homology to any other proteins. The other gene, sspA , encodes the stringent starvation protein, a regulatory protein involved in stationary-phase regulation and the stringent response of E. coli . While both rnk and sspA restored alginate production to the P. aeruginosa algR2 mutant, only rnk restored Ndk activity to the mutant. In this report, we have examined the effect of mutations in rnk and sspA on the levels of Ndk in E. coli . We find that a mutation in rnk drastically reduces the level of Ndk in E. coli . A mutation in sspA , however, affects the level of another nucleoside diphosphate kinase distinct from Ndk. The proteins can be easily distinguished from each other by their different affinities for nucleoside diphosphates (NDPs) and also by the differential effect of anti-Ndk antibodies on the reactions they catalyse. The ability of either of these two proteins to restore alginate synthesis in the algR2 mutant of P. aeruginosa demonstrates the importance of nucleoside triphosphate synthesis and energy metabolism for alginate synthesis. Additionally, a role for the stringent starvation protein (SspA) in the modulation of nucleoside triphosphate (NTP) levels in E. coli is also suggested from these experiments.     

Degradation of L-glutamate dehydrogenase from Escherichia coli: allosteric regulation of enzyme stability.

L-glutamate dehydrogenase (GDH) is stable in exponentially growing Escherichia coli cells but is degraded at a rate of 20-30% per hour in cells starved for either nitrogen or carbon. GDH degradation is energy-dependent, and mutations in ATP-dependent proteases, ClpAP or Lon lead to partial stabilization. Degradation is inhibited by chloramphenicol and is completely blocked in relA mutant cells, suggesting that ribosome-mediated signaling may facilitate GDH degradation. Purified GDH has a single tight site for NADPH binding. Binding of NADPH in the absence of other ligands leads to destabilization of the enzyme. NADPH-induced instability and sensitivity to proteolysis is reversed by tri- and dicarboxylic acids or nucleoside di- and triphosphates. GTP and ppGpp bind to GDH at an allosteric site and reverse the destabilizing effects of NADPH. Native GDH is resistant to degradation by several purified ATP-dependent proteases: ClpAP, ClpXP, Lon, and ClpYQ, but denatured GDH is degraded by ClpAP. Our results suggest that, in vivo, GDH is sensitized to proteases by loss of a stabilizing ligand or interaction with an destabilizing metabolite that accumulates in starving cells, and that any of several ATP-dependent proteases degrade the sensitized protein.