Biochemical characterization of a recombinant Japanese encephalitis virus RNA-dependent RNA polymerase

Abstract

Background: Japanese encephalitis virus (JEV) NS5 is a viral nonstructural protein that carries both methyltransferase and RNA-dependent RNA polymerase (RdRp) domains. It is a key component of the viral RNA replicase complex that presumably includes other viral nonstructural and cellular proteins. The biochemical properties of JEV NS5 have not been characterized due to the lack of a robust in vitro RdRp assay system, and the molecular mechanisms for the initiation of RNA synthesis by JEV NS5 remain to be elucidated.

Results: To characterize the biochemical properties of JEV RdRp, we expressed in Escherichia coli and purified an enzymatically active full-length recombinant JEV NS5 protein with a hexahistidine tag at the N-terminus. The purified NS5 protein, but not the mutant NS5 protein with an Ala substitution at the first Asp of the RdRp-conserved GDD motif, exhibited template- and primer-dependent RNA synthesis activity using a poly(A) RNA template. The NS5 protein was able to use both plus- and minus-strand 3'-untranslated regions of the JEV genome as templates in the absence of a primer, with the latter RNA being a better template. Analysis of the RNA synthesis initiation site using the 3'-end 83 nucleotides of the JEV genome as a minimal RNA template revealed that the NS5 protein specifically initiates RNA synthesis from an internal site, U81, at the two nucleotides upstream of the 3'-end of the template.

Conclusion: As a first step toward the understanding of the molecular mechanisms for JEV RNA replication and ultimately for the in vitro reconstitution of viral RNA replicase complex, we for the first time established an in vitro JEV RdRp assay system with a functional full-length recombinant JEV NS5 protein and characterized the mechanisms of RNA synthesis from nonviral and viral RNA templates. The full-length recombinant JEV NS5 will be useful for the elucidation of the structure-function relationship of this enzyme and for the development of anti-JEV agents.

Figure 1

Purification of recombinant JEV NS5 protein. JEV NS5 protein was expressed in E. coli and purified by Ni-NTA chromatography, gel filtration chromatography, and ion-exchange chromatography using an SP-Sepharose column. (A) Imidazole elution profile of JEV NS5 from Ni-NTA resin. (B) Gel filtration chromatography elution profile of JEV NS5. (C) NaCl-elution profile of JEV NS5 from an SP-Sepharose column. (A-C) Fractions from each purification step were resolved by SDS-12% PAGE and stained with Coomassie brilliant blue. (D) JEV NS5 and its mutant NS5_D668A from a peak fraction eluted from an SP-Sepharose column were resolved by SDS-12% PAGE and visualized by silver staining. (E) Western blot analysis of the purified JEV NS5 and NS5_D668A using an anti-His₆ antibody. The sizes of protein markers are indicated in kilodaltons. Closed and open arrowheads indicate the full-length JEV NS5, and its major cleaved form identified by MALDI-TOF analysis, respectively.

Figure 2

RdRp assay using a poly(A) template and oligo(U)₂₀ primer. (A) Primer-dependent RNA synthesis. RdRp assays were performed with the purified JEV NS5 using a poly(A) RNA template in the presence (+) or absence (-) of the primer oligo(U)₂₀. (B) RdRp assays were performed with the purified wild-type NS5 (GDD) and its mutant NS5_D668A (GAD) in the presence (+) or absence (-) of a poly(A) RNA template. Relative RdRp activities (%), which were obtained by comparing the ³²P-UMP incorporation measured by liquid scintillation counting with that obtained for the reaction with the template and primer, 3.0 × 10⁵ cpm, are presented.

Figure 3

Optimization of JEV RdRp assay conditions. Effect of temperature (A), pH (B), K⁺ ion (C), and Mn²⁺ ion (D) on JEV RdRp activity. RdRp assays were performed with the poly(A)/(U)₂₀ template under the indicated conditions. The RdRp activity was measured as in Figure 2 and is presented as the percentage of that observed under each optimal condition. Shown is the mean and standard error from three independent experiments.

Figure 4

Dependence of JEV RdRp activity on Mn²⁺ion. (A) RdRp assays were performed with the poly(A)/(U)₂₀ template (A) in the absence (lane 1) or in the presence of increasing concentrations of MgCl₂ or MnCl₂ (lanes 2–6 and 7–11; 0.5, 1.0, 2.5, 5.0, and 10 mM of MgCl₂ and MnCl₂, respectively). (B) RdRp assays were performed with the 83-nt RNA representing the 3' end of the plus-strand JEV genome, in the absence of metal ions (lane 1) or in the presence of 2.5 mM of the divalent metal ion indicated above the autoradiogram (lanes 2 and 3). RdRp products were denatured and resolved on a medium size (20 × 20 cm) denaturing 5% polyacrylamide gel, and subjected to autoradiography.

Figure 5

De novo initiation of RNA synthesis from the plus- and minus-strand 3'-UTR of the JEV genome. RdRp assays were performed with the purified wild-type NS5 (WT) and mutant NS5_D668A (Mt) in the presence (+) or absence (-) of JEV 3'(+)UTR RNA (A and C), 3' (-)UTR RNA (B and C), and the 83-nt RNA (D) template. RdRp products were analyzed as in Figure 4 by autoradiography. Arrowheads indicate the template positions visualized by ethidium-bromide staining of the gels.

Figure 6

Nuclease S1 treatment of the RNA products synthesized from the 83-nt RNA template. Heat-denatured (+Δ) or untreated (-Δ) RdRp products synthesized from the 83-nt RNA, which represents the 3'-terminal region of JEV genome, were left untreated (-) or digested with nuclease S1 (S1), and resolved on a denaturing polyacrylamide gel. Nuclease S1 treatments were performed in 50 mM NaCl (L; low salt) or 500 mM NaCl (H; high salt). Arrowhead indicates the position of the 83-nt RNA template.

Figure 7

Mapping of the RNA synthesis initiation site on the 83-nt RNA template. The RdRp assay was performed with the 83-nt RNA template. (A) An autoradiogram showing the major RNA product synthesized by JEV NS5 using the 83-nt RNA template. Products were resolved on a 5% polyacrylamide sequencing gel (20 × 40 cm) containing 8 M urea. The RNA size markers, 5'-end labeled RNA template (End), and a set of labeled RNA fragments generated by alkaline hydrolysis of the 5'-end labeled RNA template (End/OH), were resolved on the same gel. Arrowhead indicates the internally initiated 81-nt RNA product. (B) The close-up autoradiogram of the same gel shown in (A). (C) Secondary structure of the 83-nt RNA template predicted by the Mfold program. Bent arrow denotes the predicted RNA synthesis initiation site.

Figure 8

Effect of nucleotide addition and deletion at the lower part of the stem-loop structure of the 83-nt RNA template on RNA synthesis initiation. (A) Sequences and secondary structures of wild-type 83-nt RNA and its derivatives. Names and ρ G are shown below the structures predicted by the Mfold program. Partial structures representing the lower stem-region are shown. The deleted and added nucleotides are indicated by ρ and boxed, respectively. (B) RNA synthesis directed by the wild-type 83-nt RNA and its derivatives. RdRp assays were performed with the RNA templates indicated above the autoradiogram, and products were resolved a 5% polyacrylamide sequencing gel (20 × 40 cm) containing 8 M urea. Arrowhead indicates the 81-nt internally initiated RNA product. Representative data from three independent experiments are shown.

Figure 9

TNTase activity of JEV NS5. The 83-nt RNA was used as a template for RdRp and TNTase activity assays. For TNTase activity assays, reactions were performed in the presence of cold UTP and [α-³²P] UTP (lane 3), or in the presence of single [α-³²P] UTP (lane 4). An RNA product from the standard RdRp reaction mixture is shown as a control (lane 2). Lane 1, 5'-end labeled 83-nt RNA size marker.