Enzymatic characterization of the full-length and C-terminally truncated hepatitis C virus RNA polymerases: function of the last 21 amino acids of the C terminus in template binding and RNA synthesis

The nonstructural protein NS5B of hepatitis C virus (HCV) is an RNA-dependent RNA polymerase (RdRp), which plays a central role in viral replication. Most of the reported studies on HCV polymerase in vitro have used a truncated form of the enzyme lacking the C-terminal 21 amino acids (DeltaC(21)-NS5B). In this study, we compared the enzymatic properties of the full-length NS5B (FL-NS5B) and this truncated form. Removal of the C(21) domain enhanced the enzyme stability. Both enzymes are capable of performing de novo and primer-dependent RNA syntheses, but each possesses a unique set of biochemical requirements for optimal RdRp activity. Whereas RNA synthesis by FL-NS5B remained relatively constant at 12-100 mM KCl, synthesis by DeltaC(21)-NS5B rapidly decreased at KCl concentrations greater than 12 mM. The different salt requirement for overall RNA synthesis by these two polymerases can in part be explained by the effect of monovalent ion concentration at the step of template binding, where binding by DeltaC(21)-NS5B but not FL-NS5B decreased proportionally as the KCl concentration increased from 25 to 200 mM. Thus, the C(21) domain appears to contribute to NS5B-RNA template binding, probably through the hydrophobic stacking interaction between its aromatic amino acids and the nucleotide bases of the RNA. This interpretation was supported by the observation that the C(21) polypeptide by itself could also bind to RNA to form binary complexes that were resistant to changes in the KCl concentration. Though both enzymes exhibited similar K(s) values for each of the four NTPs (1-5 microM), DeltaC(21)-NS5B generally required lower NTP concentrations than FL-NS5B for optimal synthesis. Interestingly, DeltaC(21)-NS5B became severely inhibited at elevated NTP concentrations, which most likely is due to competitive binding of the noncomplementary nucleotide to the polymerase catalytic center. Finally, the terminal transferase activity of DeltaC(21)-NS5B was found to be distinct from that of FL-NS5B on several different RNA templates. Together, these findings indicated that the HCV NS5B C(21) domain, in addition to being a membrane anchor, functions in template binding, NTP substrate selection, and modulation of terminal transferase activity.     

RNA synthesis in nuclei isolated from early embryos of Xenopus laevis.

1. Rates of RNA synthesis in isolated Xenopus embryo nuclei decrease from blastula through gastrula and neurula stages to hatching tadpoles. 2. In blastula and gastrula nuclei, net synthesis of RNA continues for over 30 min, both in the presence of KCl at 0.4 M and in its absence. In nuclei from later stages, net synthesis continues for only about 10 min in the absence of KCl. 3. At low ionic strength, RNA synthesis in all nuclei is greater with optimum Mg-2+ (6 mM) than with optimum Mn-2+ (1 mM). At high ionic strength the reverse is true. 4. An unusual feature, which gradually disappears as development proceeds, is that curves relating RNA synthesis to KCl concentration show a peak at 0.1 M KCl. In blastula nuclei, RNA synthesis is more rapid at 0.1 M KCl than at 0.4 M. 5. This peak at low ionic strength is not observed in the presence of the initiation inhibitor rifamycin AF/013. It is concluded that the peak arises from initiation of RNA synthesis by an excess of RNA polymerases bound non-specifically to the isolated nuclei. The residual synthesis, representing elongation of chains that were initiated in vivo, still declines as development progresses. 6. In blastula nuclei, over half of the RNA synthesis is effected by polymerase II (inhibited by alpha-amanitin), the proportion remaining roughly constant with increasing ionic strength. In neurula nuclei, the proportion rises from about one-half to three-quarters. The initiation-dependent peak in blastula and gastrula nuclei is contributed by both alpha-amanitin-sensitive and alpha-amanitin-resistant enzymes.     

Reinitiation of synthesis of small cytoplasmic RNA species K and L in isolated HeLa cell nuclei in vitro.

Isolated HeLa cell nuclei were used to synthesize low molecular weight RNA species in-vitro. The labelled RNA released from the nuclei during the incubation mainly consists of 5S RNA, pre-tRNA and small cytoplasmic RNA species K and L. All these low molecular weight RNA species are synthesized by RNA polymerase C (or III). The polyanion heparin was applied to study the reinitiation of these RNA molecules in-vitro. A comparison of the kinetics of RNA synthesis in the absence and in the presence of this inhibitor demonstrates a highly efficient in-vitro reinitiation of scRNA species K and L as well as 5S and pre-tRNA by RNA polymerase C. These results indicate a general competence of this enzyme to catalyze the de-novo formation of specific gene products in-vitro.     

Evidence for two forms of RNA-dependent DNA polymerase in Visna virus.

The visna viral RNA-dependent DNA polymerase has been resolved into two forms by affinity chromatography. Glycerine gradient centrifugation of the two forms showed that one form sedimented at 6.9 S corresponding to an apparent molecular weight of 135 000 and the other at 6.3 S corresponding to 118 000. Sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis of the two forms indicated that the 6.9 S enzyme is composed of 2 molecules of 68 000 mol. wt. chain and the 6.3 S is a single chain enzyme. The latter form has been identified as a glycoprotein. The 6.9 S form can be completely inactivated in 20 min at 45 degrees C, prefers poly(rC) over poly(rA) as template and has high efficiency in utilizing visna 70 S RNA as template. The 6.3 S form is stable at 45 degrees C, active with 70 S viral RNA as template, prefers poly(rA) over poly(rC), and requires higher concentration of Mn2+ (0.4 mM) for maximum activity than the 6.9 S form does (0.1 mM) with synthetic homopolymers as templates. However, both 6.9 S and 6.3 S forms prefer Mg2+ over Mn2+ regardless of the nature of the templates.     

Two distinct poly(A) polymerases isolated from the cytoplasm of Ehrlich ascites tumour cells

The poly(A) polymerases from the cytosol and ribosomal fractions of Ehrlich ascites tumour cells are isolated and partially purified by DEAE-cellulose and phosphocellulose column chromatography. Two distinct enzymes are identified: (a) a cytosol Mn2+-dependent poly(A) polymerase (ATP:RNA adenylyltransferase) and (b) a ribosome-associated enzyme defined tentatively as ATP(UTP): RNA nucleotidyltransferase. The cytosol poly(A) polymerase is strictly Mn2+-dependent (optimum at 1 mM Mn2+) and uses only ATP as substrate, poly(A) is a better primer than ribosomal RNA. The purified enzyme is free of poly(A) hydrolase activity, but degradation of [3H]poly(A) takes place in the presence of inorganic pyrophosphate. Most likely this enzyme is of nuclear origin. The ribosomal enzyme is associated with the ribosomes but it is found also in free state in the cytosol. The purified enzyme uses both ATP and UTP as substrates. The substrate specificity varies depending on ionic conditions: the optimal enzyme activity with ATP as substrate is at 1 mM Mn2+, while that with UTP as substrate is at 10--20 mM Mg2+. The enzymes uses both ribosomal RNA and poly(A) [but not poly(U)] as primers. The purified enzyme is free of poly(A) hydrolase activity.     

Presence of a thiol protease in regenerating rat-liver nuclei. Partial purification and some properties

1. Nuclei of regenerating rat liver washed with Triton X-100 were found to contain a new protease. Since the enzymatic activity for degrading ribosomal proteins was inhibited in vivo by administration of E-64, a thiol protease inhibitor, the enzyme may participate in the degradation of newly synthesized ribosomal proteins and histones in regenerating rat liver nuclei as reported previously by us [Biochem. Biophys. Res. Commun. 75, 525-531 (1077)]. The optimum pH was 5.5. 2. The enzyme was extracted from washed nuclei and partially purified by gel filtration through Sepharose 6B. Its molecular weight was about 40 000. A maximal activity of partially purified enzyme was observed in the presence of 1 mM EDTA and 2 mM dithiothreitol at pH 5.5 It was inhibited by thio reagents, E-64, leupeptin and hevy metal ions. The enzyme degraded ribosomal proteins endoproteolytically and degraded most proteins tested as substrates, although liver cell sap proteins and serum albumin were less degraded than ribosomal proteins and histones, alpha-N-Benzoylarginine-beta-naphthylamide and benzoylarginine amide were not hydrolyzed.     

Recognition of 16S rRNA by ribosomal protein S4 from Bacillus stearothermophilus

Protein S4 is essential for bacterial small ribosomal subunit assembly and recognizes the 5' domain (approximately 500 nt) of small subunit rRNA. This study characterizes the thermodynamics of forming the S4-5' domain rRNA complex from a thermophile, Bacillus stearothermophilus, and points out unexpected differences from the homologous Escherichia coli complex. Upon incubation of the protein and RNA at temperatures between 35 and 50 degrees C under ribosome reconstitution conditions [350 mM KCl, 8 mM MgCl2, and 30 mM Tris (pH 7.5)], a complex with an association constant of > or = 10(9) M(-1) was observed, more than an order of magnitude tighter than previously found for the homologous E. coli complex under similar conditions. This high-affinity complex was shown to be stoichiometric, in equilibrium, and formed at rates on the order of magnitude expected for diffusion-controlled reactions ( approximately 10(7) M(-1) x s(-1)), though at low temperatures the complex became kinetically trapped. Heterologous binding experiments with E. coli S4 and 5' domain RNA suggest that it is the B. stearothermophilus S4, not the rRNA, that is activated by higher temperatures; the E. coli S4 is able to bind 5' domain rRNA equally well at 0 and 37 degrees C. Tight complex formation requires a low Mg ion concentration (1-2 mM) and is very sensitive to KCl concentration [- partial differential[log(K)]/partial differential(log[KCl]) = 9.3]. The protein has an unusually strong nonspecific binding affinity of 3-5 x 10(6) M(-1), detected as a binding of one or two additional proteins to the target 5' domain RNA or two to three proteins binding a noncognate 23S rRNA fragment of the approximately same size. This binding is not as sensitive to monovalent ion concentration [- partial differential[log(K)]/partial differential(log[KCl]) = 6.3] as specific binding and does not require Mg ion. These findings are consistent with S4 stabilizing a compact form of the rRNA 5' domain.     

The release and reassociation of 5.8 S rRNA with yeast ribosomes.

Yeast 5.8 S rRNA is released from purified 26 S rRNA when it is dissolved in water or low salt buffer (50 mM KCl, 10mM Tris-HCl, pH 7.5); it is not released from 60 S ribosomal subunits under similar conditions. The 5.8 S RNA component together with 5 S rRNA can be released from subunits or whole ribosomes by brief heat treatment or in 50% formamide; the Tm for the heat dissociation of 5.8 S RNA is 47 degrees C. This Tm is only slightly lower when 5 S rRNA is released first with EDTA treatment prior to heat treatment. No ribosomal proteins are released by the brief heat treatment. A significant portion of the 5.8 S RNA reassociates with the 60 S subunit when suspended in a higher salt buffer (e.g.0.4 m KCl, 25 mM Tris-HCl, pH 7.5, 6 mM magnesium acetate, 5 mM beta-mercaptoethanol). The Tm of this reassociated complex is also 47 degrees C. The results indicate that in yeast ribosomes the 5.8 S-26 S rRNA interaction is stabilized by ribosomal proteins but that the association is sufficiently loose to permit a reversible dissociation of the 5.8 S rRNA molecule.     

Isolation and Characterization of RNA-Directed DNA Polymerase from a B-Type RNA Tumor Virus

RNA-directed DNA polymerase was isolated from milk-borne B-type murine mammary tumor virus of the RIII mouse strain. The several hundred-fold-purified enzyme sediments at 5.5 to 5.7S with an average molecular weight of approximately 100,000. The purified enzyme is completely template dependent and responds to RNA, DNA, and synthetic templates. Stability studies indicate differential lability dependent on the exogenous template used to monitor activity.     

Transcription in Isolated Wheat Nuclei: I. ISOLATION OF NUCLEI AND ELIMINATION OF ENDOGENOUS RIBONUCLEASE ACTIVITY

Nuclei isolated from embryos of wheat (var. Yamhill) incorporated [(3)H]UTP into a trichloroacetic acid-insoluble product linearly for 60 minutes. When the RNA synthesized in vitro was analyzed on a sucrose gradient, the amount of RNA in the 4S region increased with longer incubation times. These data and the absence of higher molecular weight RNA of specific size classes in our work (and previously published reports) suggested that nuclear fractions from plant tissue contained active nucleases. This was confirmed when wheat nuclei were mixed with [(3)H]yeast RNA (4, 18, 26S). All of the radioactive yeast RNA was degraded within 30 minutes to species sedimenting between 4 and 10S. The inclusion of high salt (125 millimolar (NH(4))(2)SO(4), 100 millimolar KCl), EGTA, and exogenous RNA or DNA reduced but did not eliminate endogenous RNase activity. Wheat embryo nuclei were further purified by centrifugation on a gradient of a polyvinylpyrrolidone-coated colloidal silica suspension (Percoll). These nuclei were ellipsoidal, free of cytoplasmic material, and lacked endogenous nuclease activity when assayed with [(3)H]yeast RNA. Sucrose gradients were not as effective as Percoll gradients in purifying nuclei free of RNase activity. The Percoll method of isolating nuclei and the RNase assay reported here will be useful in isolating plant nuclei that are capable of synthesizing distinct RNA species in vitro.