Search

The dynamics of the C-terminus of the dUTPases from Escherichia coli and equine infectious anaemia virus (EIAV) were studied by 1H-(15)N nuclear magnetic resonance spectroscopy. The two enzymes differ with regard to flexibility in the backbone of the 15 most C-terminal amino acid residues, some of which are conserved and essential for enzymic activity. In the bacterial enzyme, the residues closest to the C-terminus are highly flexible and display a correlation time in the nanosecond time range. No similar high flexibility could be detected for the C-terminal part of EIAV dUTPase, indicating a different time range of flexibility.

Transient kinetics of the equine infectious anemia virus deoxyuridine 5'-triphosphate nucleotide hydrolase were characterized by monitoring the fluorescence of the protein. Rate constants for the association and dissociation of substrate and inhibitors were determined and found to be consistent with a one-step mechanism for substrate binding. A C-terminal part of the enzyme presumed to be flexible was removed by limited trypsinolysis. As a result, the activity of the dUTPase was completely quenched, but the rate constants and fluorescent signal of the truncated enzyme were affected only to a minor degree. We conclude that the flexible C-terminus is not a prerequisite for substrate binding, but indispensable for catalysis.

The X-ray structures of dUTPase from equine infectious anaemia virus (EIAV) in unliganded and complexed forms have been determined to 1.9 and 2.0 A resolution, respectively. The structures were solved by molecular replacement using Escherichia coli dUTPase as search model. The exploitation of a relatively novel refinement approach for the initial model, combining maximum likelihood refinement with stereochemically unrestrained updating of the model, proved to be of crucial importance and should be of general relevance.EIAV dUTPase is a homotrimer where each subunit folds into a twisted antiparallel beta-barrel with the N and C-terminal portions interacting with adjacent subunits. The C-terminal 14 and 17 amino acid residues are disordered in the crystal structure of the unliganded and complexed enzyme, respectively. Interactions along the 3-fold axis include a water-containing volume (size 207 A3) which has no contact with bulk solvent. It has earlier been shown that a divalent metal ion is essential for catalysis. For the first time, a putative binding site for such a metal ion, in this case Sr2+, is established. The positions of the inhibitor (the non-hydrolysable substrate analogue dUDP) and the metal ion in the complex are consistent with the location of the active centre established for trimeric dUTPase structures, in which subunit interfaces form three surface clefts lined with evolutionary conserved residues. However, a detailed comparison of the active sites of the EIAV and E. coli enzymes reveals some structural differences. The viral enzyme undergoes a small conformational change in the uracil-binding beta-hairpin structure upon dUDP binding not observed in the other known dUTPase structures., (Copyright 1999 Academic Press.)

Kinetic properties of the monomeric enzyme dUTPase from herpes simplex virus type 1 (HSV) were investigated and compared to those previously determined for homotrimeric dUTPases of bacterial and retroviral origins. The HSV and Escherichia coli dUTPases are equally potent as catalysts towards the native substrate dUTP with a kcat/K(M) of about 10(7) M(-1) s(-1) and a K(M) of 0.3 microM. However, the viral enzymes are less specific than the bacterial enzyme. The HSV and E. coli dUTPases show the same stereospecificity towards the racemic substrate analogue dUTPalphaS (2'-deoxyuridine 5'-(alpha-thio)triphosphate), suggesting that they have identical reaction mechanisms.

Deoxyuridine 5'-triphosphate nucleotidohydrolase (dUTPase, E.C. 3.6. 1.23) catalyzes the hydrolysis of dUTP to dUMP and pyrophosphate and is involved in nucleotide metabolism and DNA synthesis. A crystal of the recombinant E. coli enzyme, precipitated from polyethylene glycol mixtures in the presence of succinate at pH 4.2, was used to collect synchrotron diffraction data to 1.9 A resolution, in space group R3, a = b = 86.62, c = 62.23 A. Mercury and platinum derivative data were collected at wavelengths to optimize the anomalous contribution. The resulting 2.2 A MIRAS phases differed from the final set by 40 degrees on average and produced an excellent map which was easy to interpret. The model contains 132 water molecules and refined to an R value of 13.7%. 136 residues have clear electron density out of 152 expected from the gene sequence. The 16 C-terminal residues are presumably disordered in the crystal lattice. The monomer is a 'jelly-roll' type, containing mostly beta-sheet and only one short helix. The molecule is a tight trimer. A long C-terminal arm extends from one subunit and encompasses the next one within the trimer contributing to its beta-sheet. Conserved sequence motifs common among dUTPases, previously suggested to compose the active site and confirmed in a recent study of the dUDP complex, are located at subunit-subunit interfaces along the threefold axis, in parts of the beta-sheet and in loop regions. A similar molecular architecture has recently been found in two other trimeric dUTPases.

The molecular mechanism of substrate analogue interaction with Escherichia coli dUTPase was investigated, using the non-hydrolyzable 2'-deoxyuridine 5'-(alpha,beta-imido)triphosphate (alpha,beta-imido-dUTP). Binding of this analogue induces a difference in the far UV circular dichroism (CD) spectrum arguing for a significant change in protein conformation. The spectral shift is strictly Mg2+-dependent, does not appear with dUDP instead of alpha,beta-imido-dUTP and is not elicited if the flexible C-terminal arm is deleted from the protein by limited tryptic digestion. Involvement of the C-terminal arm in alpha,beta-imido-dUTP binding is consistent with the finding that this analogue protects against tryptic hydrolysis at Arg-141. Near UV CD of ligand-enzyme complexes reveals a characteristic difference in the microenvironments of enzyme-bound dUDP and alpha,beta-imido-dUTP, a difference not observable in C-terminally truncated dUTPase. The results suggest that (i) closing of the active site during the catalytic cycle, through the movement of the C-terminal arm, requires the presence of the complete triphosphate moiety of the substrate in complex with Mg2+, and (ii) after catalytic cleavage the active site pops open to facilitate product release.

The kinetic properties of dUTPase from equine infectious anemia virus (EIAV) were investigated. K(M) (1.1 +/- 0.1 microM) and k(cat) (25 s(-1)) were found to be independent of pH in the neutral pH range. Above pH 8.0, K(M) increases slightly. Below pH 6.0, the enzyme is rapidly deactivated. Detergent was found to enhance activity, leaving K(M) and k(cat) unaffected. Compared to the Escherichia coli dUTPase, the EIAV enzyme is equally potent in hydrolyzing dUTP, but less specific. Inhibition of the viral enzyme by the nucleotides dTTP, dUMP and a synthetic analogue, 2'-deoxyuridine 5'-(alpha,beta-imido)triphosphate, is stronger by one order of magnitude.

The enzyme dUTPase catalyses the hydrolysis of dUTP to dUMP and pyrophosphate, thereby suppressing incorporation of uracil into DNA and providing a pool of dUMP, the precursor of dTTP. Hydrolysis of other nucleotides similar in structure to dUTP would conceivably be physiologically detrimental and high specificity of the reaction seems to be necessary. In this work, we characterize the substrate specificity of the dUTPases from herpes simplex virus type 1 (HSV-1) and mouse mammary tumour virus (MMTV) in comparison to the Escherichia coli enzyme. We tested dCTP, dTTP, UTP and dUDP as substrates. Significantly higher reactivity was observed for the HSV-1 enzyme with dCTP and dTTP and for the MMTV enzyme with dTTP and UTP. The lower substrate specificity of the two virus enzymes compared with the bacterial enzyme is discussed in relation to the DNA precursor metabolism during virus replication.

The enzyme dUTPase catalyzes the hydrolysis of dUTP to dUMP and pyrophosphate, thereby preventing a deleterious incorporation of uracil into DNA. The best known dUTPase is that from Escherichia coli, which, like the human enzyme, consists of three identical subunits. In the present work, the catalytic properties of the E. coli dUTPase were investigated in the pH range 5-11. The enzyme was found to be highly specific for dUTP and discriminated both base and sugar as well as the phosphate moiety (bound dUDP was not hydrolyzed). The second best substrate among the nucleotides serving as building blocks for DNA was dCTP, which was hydrolyzed an astonishing 10(5) times less efficiently than dUTP, a decline largely accounted for by a higher Km for dCTP. With dUTP.Mg as substrate, kcat was found to vary little with pH and to range from 6 to 9 s-1. Km passed through a broad minimum in the neutral pH range with values approaching 10(-7) M. It increased with deprotonation of the uracil moiety of dUTP and showed dependence on two ionizations in the enzyme, exhibiting pKa values of 5.8 and 10.3. When excess dUTPase was reacted with dUTP middle dotMg at pH 8, the two protons transferred to the reaction medium were released in a concerted mode after the rate-limiting step. The Mg2+ ion enhances binding to dUTPase of dUTP by a factor of 100 and dUDP by a factor of 10. Only one enantiomer of the substrate analog 2'-deoxyuridine-5'-(alpha-thio)-triphosphate was hydrolyzed by the enzyme. These results are interpreted to favor a catalytic mechanism involving magnesium binding to the alpha-phosphate, rate-limiting hydrolysis by a shielded and activated water molecule and a fast ordered desorption of the products. The results are discussed with reference to recent data on the structure of the E. coli dUTPase.UDP complex.

We have determined the structure of the homotrimeric dUTPase from Escherichia coli, completed with an inhibitor and substrate analogue, dUDP. Three molecules of dUDP are found symmetrically bound per trimer, each in a shallow cleft between adjacent subunits, interacting with evolutionary conserved residues. The interactions of the uracil ring and the deoxypentose with the protein are consistent with the high specificity of the enzyme with respect to these groups. The positions of the two phosphate groups and adjacent water molecules are discussed in relation to the mechanism and kinetics of catalysis. The role that dUTPase plays in DNA metabolism makes the enzyme a potential target for chemotherapeutic drugs: the results presented here will aid in the design and development of inhibitory compounds.

The dUDP analogue, 2'-deoxyuridine 5'-(alpha,beta-imido)diphosphate (dUPNP) was synthesized. The corresponding triphosphate analogue (dUPNPP) was prepared by enzymic phosphorylation of dUPNP using the enzyme pyruvate kinase and phosphoenolpyruvate as the phosphate donor. This method was successful in phosphorylating the imidodiphosphate analogue of 2'-deoxythymidine (dTPNP) to 2'-deoxythymidine 5'-(alpha, beta-imido)triphosphate (dTPNPP), in contradiction to a previous report. The properties of dUPNPP have been tested using the enzyme dUTPase from Escherichia coli. This enzyme, having a crucial role in nucleotide metabolism, is strictly specific for its substrate (dUTP) and catalyzes the hydrolysis of the alpha, beta-bridge, resulting in dUMP and pyrophosphate. Replacement of the alpha, beta-bridging oxygen in dUTP with an imido group resulted in a nonhydrolyzable substrate analogue and a potent competitive inhibitor of dUTPase (Ki = 5 microM). The analogue prepared (dUPNPP) may be utilized in crystallographic studies of the active site of dUTPase to provide knowledge about specific interactions involved in substrate binding and as a parental compound in design of dUTPase inhibition for medical purposes.

Selective modification of one (of three) tyrosine residue per enzyme monomer leads to inactivation of dUTPase of the retrovirus equine infectious anemia virus (EIAV). The substrate dUMP and the cofactor Mg2+ protect against inactivation and modification, in agreement with the study on E. coli dUTPase (Vertessy et al. (1994) Biochim. Biophys. Acta 1205, 146-150). Amino acid analyses of nitrated dUTPases confirmed Tyr-selectivity of modification. The nitrated residue in E. coli dUTPase was identified as the evolutionary highly conserved Tyr-93. The modifiable residue is shown to be the only Tyr exposed in both E. coli and EIAV dUTPases. As a consequence of Tyr-93 derivatization, the Mg2+-dependent interaction between the substrate-analogue dUDP and E. coli dUTPase becomes impaired as shown by circular dichroism spectroscopy, here presented as a tool for monitoring ligand binding to the active site.

Deoxyuridine 5'-triphosphate nucleotidohydrolase (dUTPase, EC 3.6.1.23) catalyzes the hydrolysis of dUTP to dUMP and pyrophosphate, and plays important roles in nucleotide metabolism and DNA replication. The dUTPase gene of the retrovirus equine infectious anemia virus (EIAV) was cloned and overexpressed in Escherichia coli using the T7 RNA polymerase expression system. The recombinant vector (pET-3a/EDU), constructed by mutagenic PCR, was transformed into E. coli BL21 (DE3) pLysS cells, resulting in expression of EIAV dUTPase at about 40% of the extracted protein. This level of overproduction is very high compared to previous reports on heterologous expression of dUTPases in E. coli. A one-step purification procedure using phosphocellulose chromatography results in a homogeneous preparation of the enzyme in a yield of 45 mg liter-1 of bacterial culture. The purified EIAV dUTPase, run on a sodium dodecyl sulfate-polyacrylamide gel electrophoresis, shows an apparent molecular mass of 15.1 kDa in accordance with the gene structure. The isoelectric point (pI) was determined to 5.6. Gel filtration under nondenaturating conditions gives a retention volume corresponding to a molecular mass of 40.6 kDa, suggesting a trimeric organization of the enzyme. The amino acid composition and amino-terminal sequence of the recombinant dUTPase are in agreement with predictions from the DNA sequence.

A ribosomal frameshift at the gag-pro junction of mouse mammary tumor virus (MMTV) gives rise to the protein p30. The protein consists of two domains, the zinc-finger-containing nucleocapsid (NC) protein portion with 95 residues and a C-terminal extension comprising 154 residues. The C-terminal domain shows similarity in sequence with the enzyme dUTPase from other sources. In this paper, we demonstrate that p30 is a functional dUTPase. Overproduction of the NC protein in Escherichia coli, using the native frameshift sequence at the gag stop codon, caused a detectable expression of dUTPase ascribed to a low frequency of readthrough. By a 1-base insertion, eliminating the gag stop codon and fusing the gag and pro reading frames, a plasmid, pET-3d-NCDU, directing overexpression of p30, was constructed. The overproduced protein, purified by phosphocellulose chromatography, shows both zinc-binding and dUTPase activity. Analytical gel filtration and sequence homology to other dUTPases suggest a trimeric assembly of p30 subunits. MMTV thus possesses two different forms of the nucleocapsid protein, the ordinary NC protein and the p30, having the NC protein connected to a domain of dUTPase.

dUTP nucleotidohydrolase (dUTPase, EC 3.6.1.23) from E. coli contains a total of six tyrosine residues per trimer. About half of them were found to be susceptible to acetylation with N-acetylimidazole or to nitration with tetranitromethane with concomitant loss of activity. Deacetylation with N-hydroxylamine leads to full reactivation. Inhibitory products of dUTP hydrolysis, i.e., dUMP and inorganic pyrophosphate together with the cofactor Mg2+ protect significantly against inactivation and chemical modification. In the Cu(2+)-dUTPase complex, charge transfer from Cu2+ to the tyrosinate anion was perturbed by the presence of the substrate dUTP. These results, together with the occurrence of one tyrosine residue in a strictly conserved sequence motif suggest the critical importance of this residue for the function of the enzyme.

Deoxyuridine 5'-triphosphate nucleotidohydrolase (dUTPase), widespread in nature with a crucial role in the nucleotide metabolism, catalyzes the hydrolysis of dUTP to dUMP and pyrophosphate. The enzyme from herpes simplex virus type 1 (HSV-1 dUTPase) was overproduced in Escherichia coli by using the T7 RNA polymerase expression system. The coding region of the HSV-1 dUTPase gene, UL 50, was positioned downstream of the promoter and the ribosome-binding site of the phage T7 gene 10 on the expression vector pET-3a. The resulting recombinant plasmid, pET-3a/UL50, was transformed into E. coli BL21(DE3)pLysS cells, conferring expression of HSV-1 dUTPase as 2-3% of the soluble protein inducible by isopropyl thiogalactoside. By chromatography on phosphocellulose and Mono S (Pharmacia LKB) columns a nearly homogeneous preparation of the enzyme with a high specific activity (49 mumol per minute per milligram) was obtained. The recombinant protein was compared with the native dUTPase similarly purified from HSV-1-infected Vero cells (African green monkey kidney fibroblasts). The two proteins showed the same mobility on sodium dodecyl sulfate-polyacrylamide gel electrophoresis and the amino-terminal sequences were found to be identical. The molecular mass (39 kDa) and the amino acid composition of the recombinant enzyme are also in accordance with predictions from the DNA sequence. Thus, the overproducing system described here appears suitable for providing HSV-1 dUTPase for detailed studies of molecular properties.

The enzyme dUTPase catalyses the hydrolysis of dUTP and maintains a low intracellular concentration of dUTP so that uracil cannot be incorporated into DNA. dUTPase from Escherichia coli is strictly specific for its dUTP substrate, the active site discriminating between nucleotides with respect to the sugar moiety as well as the pyrimidine base. Here we report the three-dimensional structure of E. coli dUTPase determined by X-ray crystallography at a resolution of 1.9 A. The enzyme is a symmetrical trimer, and of the 152 amino acid residues in the subunit, the first 136 are visible in the crystal structure. The tertiary structure resembles a jelly-roll fold and does not show the 'classical' nucleotide-binding domain. In the quaternary structure there is a complex interaction between the subunits that may be important in catalysis. This possibility is supported by the location of conserved elements in the sequence.

Restriction endonuclease mapping of previously constructed dut plasmids has been carried out using the enzymes PvuI, PvuII and SacI. Various dut plasmids were also tested in the "maxicell" protein-synthesizing system. They all show two protein bands in common, one of Mr 16000 in agreement with the size previously reported for the purified dUTPase subunit (Shlomai and Kornberg, 1978). With the information obtained the structural gene for dUTPase can be assigned to a 950-bp SacI-PvuII fragment of the E. coli genome. Studies, described in the preceding paper, on the overproduction of dUTPase by bacterial strains carrying different dut plasmids strongly suggest that the dut gene is transcribed in the direction from the SacI site towards the PvuII site and that the SacI site is located within the dut control region. The second protein band observed in the "maxicell" experiments has an Mr of 23500. Its identity is unknown but it may represent a precursor of dUTPase or the product of a separate gene located between dut and pyrE.

We have constructed thermoinducible plasmids carrying the gene (dut) for the enzyme deoxyuridine 5'-triphosphate nucleotidohydrolase (dUTPase, EC 3.6.1.23) from Escherichia coli. A 9.4-kb BamHI restriction enzyme fragment carrying the dut gene was inserted into the runaway-replication plasmid pKN402A (Uhlin et al., Gene 6 (1979) 91-106). Strains carrying such plasmids increased their dUTPase activity considerably. In minimal medium a 200-fold increase was demonstrated. A smaller (1.5-kb) SacI-BamHI fragment from the dut region was also cloned into pKN402A. The dUTPase production in dut mutant strains carrying this plasmid (pKK141) was only at about wild-type level after temperature shift. To test the hypothesis that the SacI cleavage used affects a control region for the dut gene, we recloned the dut fragment by transferring it from pKK141 into pHUB2 (Bernard et al., Gene 5 (1979) 59-76), a plasmid carrying the phage lambda pL promoter. A 3.6-kb EcoRI-BamHI fragment from pKK141, including the 1.5-kb SacI-BamHI segment from the dut region, was inserted downstream from the pL promoter. When this plasmid was present in a strain containing a thermosensitive lambda repressor gene, thermoinduction of dUTPase was negligible, apparently due to the presence of some termination signals between pL and dut. Therefore, we removed a 1.9-kb EcoRI-SacI fragment from the region between pL and the dut gene and replaced it with a 0.22-kb EcoRI-SacI fragment, obtained from the b2 region of lambda. Strains carrying such a shortened dut-pHUB2 derivative and a temperature-sensitive lambda repressor overproduced dUTPase very dramatically after heat induction. The final level reached was 300-400 times the wild-type level, corresponding to 10% of the total soluble protein. The information obtained, together with analysis of plasmid-directed polypeptide products described by Lundberg et al. (Gene 22 (1983) 127-131) shows that the SacI site is indeed on the promoter-proximal side of the dut gene.

Deoxyuridine triphosphate nucleotidohydrolase (dUTPase), an enzyme in the nucleotide metabolism that is a pyrophosphatase hydrolyzing dUTP, has been crystallized. The crystals belong to the trigonal space group R3 and diffract beyond 2 A. The native dUTPase crystals and a mercury derivative are stable in the X-ray beam and are suitable for a high resolution X-ray structure analysis.

A recombinant plasmid, pHW1, directing the overproduction of the enzyme deoxyuridine 5'-triphosphate nucleotidohydrolase (dUTPase, EC 3.6.1.23) from Escherichia coli has been constructed. A 1900-base DNA fragment carrying the structural gene for the enzyme (dut) has been recloned into a runaway replication vector that also carries the strong leftward promoter (pL) of bacteriophage lambda. Upon temperature shift, an E. coli strain carrying the new plasmid gives an increase in dUTPase activity of about 600-fold in rich medium compared to wild-type bacteria. The 64-kDa protein corresponding to the mature form of the enzyme reaches 20% of the total protein content of the bacterial cell. Using this strain, a simplified procedure has been developed for the purification of dUTPase. The purification steps consist of extraction of the cytoplasmic proteins, ammonium sulfate precipitation, anion-exchange chromatography and gel filtration on FPLC. The new overproducing plasmid and the simplified purification procedure developed will make it possible to purify dUTPase in sufficient amounts for detailed characterization studies.

The nucleotide sequence of the dUTPase structural gene, dut, of Escherichia coli has been determined. The DNA sequence predicts a polypeptide chain of 150 amino acid residues (mol. wt. 16 006) corresponding in size and composition to the purified dUTPase subunit. In a tentative promoter region preceding the dut gene, the -35 and -10 regions are separated by a SacI (SstI) site. Cloning of the dut gene utilization this SacI site was previously shown to reduce dut expression dramatically. The nucleotide sequence also contains a 210-codon open reading frame 106 bp downstream of dut and co-directional with dut. Previous protein synthesis experiments using dut plasmids allocated the gene of a polypeptide of mol. wt. 23 500 to this DNA region. The open reading frame thus may correspond to a protein of unknown function co-transcribed with the dut gene.

Strains of Escherichia coli with a mutation in the sof (dnaS) locus show a higher than normal frequency of recombination (are hyper rec) and incorporate label into short (4-5S) DNA fragments following brief [3H]thymidine pulses [Konrad and Lehman, Proc. Natl. Acad. Sci. USA 72, 2150 (1975)]. These mutant strains have now been found to be defective in deoxyuridinetriphosphate diphosphohydrolase (dUTPase; deoxyuridinetriphosphatase, EC 3.6.1.23), the enzyme that catalyzes the hydrolysis of dUTP to dUMP and PPi. Reversion of one sof- mutation to sof+ restores dUTPase activity and abolishes the accumulation of labeled 4-5S DNA fragments. Mutants initially isolated as defective in dUTPase (dut-) are also hyper rec and show transient accumulation of short DNA fragments. Both the sof and dut mutations are located at 81 min on the E. coli map, closely linked to the pyrE locus. The sof and dut loci thus appear to be identical. A decrease in dUTPase as a consequence of a sof or dut mutation may result in the increased incorporation of uracil into DNA. Rapid removal of the uracil by an excision-repair process could then lead to the transient accumulation of short DNA fragments. It is possible that at least a portion of the Okazaki fragments seen in wild-type cells may originate in this way.