The conformation of hepatitis C virus NS3 proteinase with and without NS4A: a structural basis for the activation of the enzyme by its cofactor.

BACKGROUND: Hepatitis C virus (HCV) NS3 proteinase activity is required for the release of HCV nonstructural proteins and is thus a potential antiviral target. The enzyme requires a protein cofactor NS4A, located downstream of NS3 on the polyprotein, for activation and efficient processing. OBJECTIVES: Comparison of the proteinase three-dimensional structure before and after NS4A binding should help to elucidate the mechanism of NS4A-dependent enzyme activation. STUDY DESIGN: We determined the crystal structure of NS3 proteinase of HCV BK isolate (genotype 1b; residues 1-189) and also the crystal structure of this proteinase complexed with HCV BK-NS4A (residues 21-34). RESULTS: The core region (residues 30-178) of the enzyme without cofactor (NS3P) or with bound cofactor (NS3P/4A) is folded into a trypsin-like conformation and the substrate P1 specificity pocket is essentially unchanged. However, the D1-E1 beta-loop shifts away from the cofactor binding site in NS3P/4A relative to NS3P, thereby accommodating NS4A. One result is that catalytic residues His-57 and Asp-81 move closer to Ser-139 and their sidechains adopt more 'traditional' (trypsin-like) orientation. The N-terminus (residues 1-30), while extended in NS3P, is folded into an alpha-helix and beta-strand that cover the bound cofactor of NS3P/4A. A new substrate-binding surface is formed from both the refolded N-terminus and NS4A, potentially affecting substrate residues immediately downstream of the cleavage site. CONCLUSIONS: Direct comparison of the crystal structures of NS3P and NS3P/4A shows that the binding of NS4A improves the anchoring and orientation of the enzyme's catalytic triad. This is consistent with the enhancement of NS3P's weak residual activity upon NS4A binding. There is also significant refolding of the enzyme's N-terminus which provides new interactions with P'-side substrate residues. The binding surface for P'-side substrate residues, including the P1 specificity pocket, changes little after NS4A binding. In summary, we observe a structural basis for improved substrate turnover and affinity that follows complexation of NS3P with its NS4A cofactor.

The NS3 proteinase domain of hepatitis C virus is a zinc-containing enzyme.

NS3 proteinase of hepatitis C virus (HCV), contained within the N-terminal domain of the NS3 protein, is a chymotrypsin-like serine proteinase responsible for processing of the nonstructural region of the HCV polyprotein. In this study, we examined the sensitivity of the NS3 proteinase to divalent metal ions, which is unusual behavior for this proteinase class. By using a cell-free coupled transcription-translation system, we found that HCV polyprotein processing can be activated by Zn2+ (and, to a lesser degree, by Cd2+, Pb2+, and Co2+) and inhibited by Cu2+ and Hg2+ ions. Elemental analysis of the purified NS3 proteinase domain revealed the presence of zinc in an equimolar ratio. The zinc content was unchanged in a mutated NS3 proteinase in which active-site residues His-57 and Ser-139 were replaced with Ala, suggesting that the zinc atom is not directly involved in catalysis but rather may have a structural role. Based on data from site-directed mutagenesis combined with zinc content determination, we propose that Cys-97, Cys-99, Cys-145, and His-149 coordinate the structural zinc in the HCV NS3 proteinase. A similar metal binding motif is found in 2A proteinases of enteroviruses and rhinoviruses, suggesting that these 2A proteinases and HCV NS3 proteinase are structurally related.

The crystal structure of hepatitis C virus NS3 proteinase reveals a trypsin-like fold and a structural zinc binding site.

During replication of hepatitis C virus (HCV), the final steps of polyprotein processing are performed by a viral proteinase located in the N-terminal one-third of nonstructural protein 3. The structure of NS3 proteinase from HCV BK strain was determined by X-ray crystallography at 2.4 angstrom resolution. NS3P folds as a trypsin-like proteinase with two beta barrels and a catalytic triad of His-57, Asp-81, Ser-139. The structure has a substrate-binding site consistent with the cleavage specificity of the enzyme. Novel features include a structural zinc-binding site and a long N-terminus that interacts with neighboring molecules by binding to a hydrophobic surface patch.

Structure-function analysis of the mammalian DNA polymerase beta active site: role of aspartic acid 256, arginine 254, and arginine 258 in nucleotidyl transfer.

The crystal structure of the catalytic domain of rat DNA polymerase beta revealed that Asp256 is located in proximity to the previously identified active site residues Asp190 and Asp192. We have prepared and kinetically characterized the nucleotidyl transfer activity of wild type and several mutant forms of human and rat pol beta. Herein we report steady-state kinetic determinations of KmdTTP, Km(dT)16, and kcat for mutants in residue Asp256 and two neighboring residues, Arg254 and Arg258, all centrally located on strand beta 7 in the pol beta structure. Mutation of Asp256 to alanine abolished the enzymatic activity of pol beta. Conservative replacement with glutamic acid (D256E) led to a 320-fold reduction of kcat compared to wild type. Replacement of Arg254 with an alanine (R254A) resulted in a 50-fold reduction of kcat compared to wild type. The Km(dT)16 of D256E and R254A increased about 18-fold relative to wild type. Replacement of Arg254 with a lysine resulted in a 15-fold decrease in kcat, and a 5-fold increase in the Km(dT)16. These kinetic observations support a role of Asp256 and Arg254 in the positioning of divalent metal ions and substrates in precise geometrical orientation for efficient catalysis. The mutation of Arg258 to alanine (R258A) resulted in a 10-fold increase in KmdTTP and a 65-fold increase in Km(dT)16 but resulted in no change of kcat. These observations are discussed in the context of the three-dimensional structures of the catalytic domain of pol beta and the ternary complex of pol beta, ddCTP, and DNA.

Comprehensive chemical modification interference and nucleotide substitution analysis of an RNA pseudoknot inhibitor to HIV-1 reverse transcriptase.

We had previously used in vitro RNA selection techniques to describe a consensus RNA pseudoknot that binds and inhibits HIV-1 reverse transcriptase (HIV-RT). In this work we constructed variants of this consensus pseudoknot in order to evaluate the contributions of individual nucleotide identities and secondary structure to affinity for HIV-RT. We have also used chemical modification of ligand RNAs to corroborate the predicted structure of the pseudoknot, to discover which modifiable groups are protected from chemical attack when bound to HIV-RT, and to find which modifications interfere with binding to HIV-RT. A novel interference study is presented which involves selection of ligands from a pool created by mixed reagent oligonucleotide synthesis in order to rapidly determine allowed substitutions of 2'-OCH3 groups for the usual 2'-OH group in such RNA ligands.

Structure of and kinetic channelling in bifunctional dihydrofolate reductase-thymidylate synthase.

The bifunctional enzyme dihydrofolate reductase-thymidylate synthase catalyses both the reductive methylation of 2'-deoxyuridylate and the subsequent reduction of dihydrofolate to yield 2'-deoxythymidylate and tetrahydrofolate at two spacially discrete sites situated on different protein domains. The X-ray structure of dihydrofolate reductase-thymidylate synthase from Leishmania major indicates that transfer of dihydrofolate between these sites does not occur by transient binding at both sites but rather by movement of dihydrofolate across the surface of the protein. The enzyme has an unusual surface charge distribution that could account for this channelling of dihydrofolate between active sites.

2.3 A crystal structure of the catalytic domain of DNA polymerase beta.

The crystal structure of the catalytic domain of rat DNA polymerase beta (pol beta) has been determined at 2.3 A resolution and refined to an R factor of 0.22. The mixed alpha/beta protein has three subdomains arranged in an overall U shape reminiscent of other polymerase structures. The folding topology of pol beta, however, is unique. Two divalent metals bind near three aspartic acid residues implicated in the catalytic activity. In the presence of Mn2+ and dTTP, interpretable electron density is seen for two metals and the triphosphate, but not the deoxythymidine moiety. The principal interaction of the triphosphate moiety is with the bound divalent metals.

RNase D, a reported new activity associated with HIV-1 reverse transcriptase, displays the same cleavage specificity as Escherichia coli RNase III.

RNase D was recently reported as a new enzymatic activity associated with HIV-1 reverse transcriptase (RT), cleaving RNA at two positions within the double-stranded region of the tRNA primer-viral RNA template complex (Ben-Artzi et al., Proc. Natl. Acad. Sci. USA 89 (1992) 927-931). This would make RNase D a fourth distinct activity of HIV-1 RT, in addition to RNA- and DNA-dependent DNA polymerase and RNase H. Using a specific substrate containing tRNA(Lys,3) hybridized to the primer binding site, we were able to detect the reported RNase D activity in our preparations of recombinant HIV-1 RT. This activity was also present in several active-site mutants of RT, suggesting that it is independent of the RNase H and polymerase functionalities of RT. Furthermore, we found that the cleavage specificity of RNase D is the same as that of RNase III isolated from E.coli. A likely explantation of these results--that the observed RNase D activity is attributable to traces of RNase III contamination--was further strengthened by the finding that the recombinant preparations of HIV-1 RT can specifically cleave a phage T7-derived double-stranded RNA processing signal, which has been used as a model substrate for detection of E.coli RNase III. Moreover, RT purified from an RNase III- strain of E.coli displayed no cleavage of the tRNA primer-RNA template complex.

Heterologous expression and purification of active human phosphoribosylglycinamide formyltransferase as a single domain.

We report here for the first time that the GART domain of the human trifunctional enzyme possessing GARS, AIRS, and GART activities can be expressed independently in Escherichia coli at high levels as a stable protein with enzymatic characteristics comparable to those of native trifunctional protein. Human trifunctional enzyme is involved in de novo purine biosynthesis, and has long been recognized as a target for antineoplastic intervention. The GART domain was expressed in E. coli under the control of bacteriophage T7 promotor and isolated by a three-step chromatographic procedure. Two residues, Asp 951 and His 915, were shown to be catalytically crucial by site-directed mutagenesis and subsequent characterization of purified mutant proteins. The active monofunctional GART protein produced in E. coli can serve as a valuable substitute of trifunctional enzyme for structural and functional studies which have been until now hindered because of insufficient quantity, instability, and size of the trifunctional GART protein.

Structures of apo and complexed Escherichia coli glycinamide ribonucleotide transformylase.

The three-dimensional structure of phosphoribosylglycinamide formyltransferase (10-formyltetrahydrofolate:5'-phosphoribosylglycinamide formyltransferase, EC 2.1.2.2) has been solved both as an apoenzyme at 2.8-A resolution and as a ternary complex with the substrate glycinamide ribonucleotide and a folate inhibitor at 2.5-A resolution. The structure is a modified doubly wound alpha/beta sheet with flexibility in the active site, including a disordered loop in the apo structure, which is ordered in the ternary complex structure. This enzyme is a target for anti-cancer therapy and now for structure-based drug design.

Reverse transcriptase of human immunodeficiency virus type 1: functionality of subunits of the heterodimer in DNA synthesis.

From an in vitro analysis of the DNA-synthesizing abilities of certain specifically mutated forms of the heterodimeric reverse transcriptase of human immunodeficiency virus type 1, we can conclude that in a heterodimer, the functionality of p66 is necessary while the functionality of the p51 subunit is not needed. Conversely, p51 is not able to catalyze DNA synthesis when associated with p66, and yet when the p66 protein is absent, p51 can function. These conclusions applied to DNA synthesis on heteropolymeric RNA and DNA templates.

Proteolytic release and crystallization of the RNase H domain of human immunodeficiency virus type 1 reverse transcriptase.

The RNase H domain of human immunodeficiency virus type 1 (HIV-1) reverse transcriptase was released from recombinant DHFR-RNase H fusion protein by the action of HIV-1 protease and crystallized as large trigonal prisms that diffract x-rays to at least 2.4-A resolution. The protease cleavage occurred 18 residues away from the Phe440-Tyr441 site reported to be processed during maturation of the reverse transcriptase heterodimer. Mutagenesis of the protease-sensitive region (residues 430-440), which is part of the crystallized domain, indicates that any alteration of the wild-type sequence results in increased proteolysis of the p66 subunit. A model of asymmetric processing in HIV-1 reserve transcriptase which involves partial unfolding of the RNase H domain is proposed based on these results and the recently reported three-dimensional structure of this domain.

Crystal structure of the ribonuclease H domain of HIV-1 reverse transcriptase.

The crystal structure of the ribonuclease (RNase) H domain of HIV-1 reverse transcriptase (RT) has been determined at a resolution of 2.4 A and refined to a crystallographic R factor of 0.20. The protein folds into a five-stranded mixed beta sheet flanked by an asymmetric distribution of four alpha helices. Two divalent metal cations bind in the active site surrounded by a cluster of four conserved acidic amino acid residues. The overall structure is similar in most respects to the RNase H from Escherichia coli. Structural features characteristic of the retroviral protein suggest how it may interface with the DNA polymerase domain of p66 in the mature RT heterodimer. These features also offer insights into why the isolated RNase H domain is catalytically inactive but when combined in vitro with the isolated p51 domain of RT RNase H activity can be reconstituted. Surprisingly, the peptide bond cleaved by HIV-1 protease near the polymerase-RNase H junction of p66 is completely inaccessible to solvent in the structure reported here. This suggests that the homodimeric p66-p66 precursor of mature RT is asymmetric with one of the two RNase H domains at least partially unfolded.

Reconstitution in vitro of RNase H activity by using purified N-terminal and C-terminal domains of human immunodeficiency virus type 1 reverse transcriptase.

Two constituent protein domains of human immunodeficiency virus type 1 (HIV-1) reverse transcriptase were expressed separately and purified to homogeneity. The N-terminal domain (p51) behaves as a monomeric protein exhibiting salt-sensitive DNA polymerase activity. The C-terminal domain (p15) on its own has no detectable RNase H activity. However, the combination of both isolated p51 and p15 in vitro leads to reconstitution of RNase H activity on a defined substrate. These results demonstrate that domains of HIV-1 reverse transcriptase are functionally interdependent to a much higher degree than in the case of reverse transcriptase from Moloney murine leukemia virus.

Ion-exchange derivatives of Spheron. V. Sulphate and sulpho derivatives.

Using various reactions, strongly acidic derivatives of Spheron of the sulphate and sulpho types were prepared with nominal capacities between 0.37 and 2.0 mequiv./g. They were characterized by their capacities for small ions, by their static and dynamic capacities for proteins, by elemental analysis, by their working volumes and specific internal surface areas. The differences between the nominal capacities calculated theoretically from the sulphur contents and the measured values are discussed, as are the relationships between the nominal capacities for small ions and the static and dynamic capacities for proteins. The chromatographic properties of the cation exchangers are illustrated by the separation of an artificial mixture of proteins (serum albumin, chymotrypsinogen and lysozyme) at various pH values, and of a natural mixture of egg-white proteins. The separation of active enzymes is also reported. Sulphobutyl-Spheron was tested also as a catalyst for the esterification of alcohols.