cis-acting RNA signals in the NS5B C-terminal coding sequence of the hepatitis C virus genome

Abstract

The cis-replicating RNA elements in the 5' and 3' nontranslated regions (NTRs) of the hepatitis C virus (HCV) genome have been thoroughly studied before. However, no cis-replicating elements have been identified in the coding sequences of the HCV polyprotein until very recently. The existence of highly conserved and stable stem-loop structures in the RNA polymerase NS5B coding sequence, however, has been previously predicted (A. Tuplin, J. Wood, D. J. Evans, A. H. Patel, and P. Simmonds, RNA 8:824-841, 2002). We have selected for our studies a 249-nt-long RNA segment in the C-terminal NS5B coding region (NS5BCR), which is predicted to form four stable stem-loop structures (SL-IV to SL-VII). By deletion and mutational analyses of the RNA structures, we have determined that two of the stem-loops (SL-V and SL-VI) are essential for replication of the HCV subgenomic replicon in Huh-7 cells. Mutations in the loop and the top of the stem of these RNA elements abolished replicon RNA synthesis but had no effect on translation. In vitro gel shift and filter-binding assays revealed that purified NS5B specifically binds to SL-V. The NS5B-RNA complexes were specifically competed away by unlabeled homologous RNA, to a small extent by 3' NTR RNA, and only poorly by 5' NTR RNA. The other two stem-loops (SL-IV and SL-VII) of the NS5BCR domain were found to be important but not essential for colony formation by the subgenomic replicon. The precise function(s) of these cis-acting RNA elements is not known.

FIG.1.

(A) Genomic structure of full-length HCV RNA. The single-stranded RNA genome of HCV is shown with the 5′ NTR (single line), the structural and nonstructural proteins of the polyprotein (boxed), and the 3′ NTR (single line) indicated. The NS5B domain is shown enlarged both as a coding sequence for the NS5B protein and as an RNA sequence for potential cis-replicating RNA elements. (B) Genomic structure of the subgenomic HCV replicon RNA. The single-stranded RNA genome of the subgenomic replicon contains the HCV 5′ NTR (single line), the neo gene (boxed), the EMCV IRES (single line), the nonstructural proteins of HCV from NS3 to NS5B (boxed), and the 3′ NTR (single line). (C) Predicted RNA stem-loop structures in the HCV 3′ NTR and the C-terminal coding domain of NS5B. Illustrated are the predicted structures of stem-loops IV to VII contained in a 249-nt-long segment of the NS5B coding region (nt 9126 to 9374 of HCV RNA), designated NS5BCR. The single-stranded stretch of nucleotides between stem-loops VI and VII is designated L. Downstream from NS5BCR is the 3′ NTR, which contains a variable region, a poly(U/C) tract, and the 3′ X tail. The highly conserved 3′ X domain is predicted to consist of three stem-loops, I to III (15, 68, 69). (D) P-num values of a 257-nt-long segment (nt 9117 to 9374) of the NS5B C-terminal coding sequence, including all of the NS5BCR sequence, were obtained for HCV-J1 from the website www.virology.wisc.edu/acp/RNAFolds. Stem-loops were identified by consistently low P-num values. aa, amino acids.

FIG. 2.

Effects of deletions in NS5BCR on the colony-forming ability of the subgenomic HCV replicon. (A) Illustrated are the predicted stem-loop structures in the NS5BCR domain of NS5BCR and the deletions that were made: SL-V (nt 9261 to 9308), SL-VI plus L (nt 9189 to 9260), and L (nt 9189 to 9212). SL-V, SL-VI, and L are contained within the noncatalytic portion of the NS5B protein. (B) RNA transcripts of the constructs containing the deletions were transfected into Huh-7 cells, and the efficiency of colony formation was compared with that of the WT replicon (see Materials and Methods).

FIG. 3.

Effects of silent mutations in NS5BCR RNA on the colony-forming ability of the subgenomic replicon. (A) Illustrated are the silent mutations that were introduced into each of the four predicted stem-loops of NS5BCR. (B) Predicted secondary structures of each the mutated RNA elements. (C) RNA transcripts of the constructs containing mutations in one of the stem-loops were transfected into Huh-7 cells, and the colony-forming efficiency of the replicons was measured as described in Materials and Methods. The colony-forming efficiency of the mutant replicons is compared to that of the WT replicon (see Materials and Methods).

FIG. 4.

Effects of silent mutations in the central bulge of stem-loop VI of NS5BCR RNA. (A) Illustrated are the two mutations that were introduced into the predicted central bulge of SL-VI. (B) RNA transcripts containing the mutations were transfected into Huh-7 cells, and the colony-forming efficiency of the mutants was compared to that of the WT replicon (see Materials and Methods).

FIG. 5.

Effects of silent mutations in SL-V and VI on in vitro translation of replicon RNA. Transcripts of WT and mutant RNAs were translated in rabbit reticulocyte lysate as described in Materials and Methods. The samples were analyzed by SDS-12.5% PAGE. The control reaction mixture (C) contained no RNA. The positions of the major translation products (NS3 and HCV core-Neo fusion protein) are indicated. The values on the left indicate the positions of molecular size standards.

FIG. 6.

Effects of deletions and mutations in NS5BCR on HCV replicon RNA levels. Transfected cells were harvested daily for 5 days, and the total RNA was isolated. Five hundred nanograms of total RNA was analyzed by real-time RT-RCR as described in Materials and Methods. (A) Quantitation of the plus-strand replicon RNA present in cells on days 1 to 5 by real-time PCR. Similar results were obtained in three separate experiments. (B) Agarose gel electrophoresis of the final PCR products obtained in the real-time RT-PCR experiments. The PCR products collected after 40 cycles of PCR on days 1 to 5 (lanes 1 to 5) were analyzed by 2% native agarose gel electrophoresis. The upper bands are specific for HCV NS5B, and the lower bands are specific for glyceraldehyde 3-phosphate dehydrogenase, as indicated by the arrows. Lane M, DNA molecular weight marker.

FIG. 7.

Gel shift analysis of purified NS5B binding to RNA transcripts of NS5BCR. (A) Purification of HCV NS5B. The enzyme was expressed in baculovirus and purified to near homogeneity (see Materials and Methods). The purity of the enzyme was tested by SDS-PAGE, followed by Coomassie staining. Protein sizes are indicated on the left. (B) Gel shift analysis of purified NS5B (0 to 1 μM) binding to NS5BCR RNA was carried out in the presence of a 1,000-fold excess of tRNA as described in Materials and Methods. (C) Competition of NS5B (1 μM) binding to NS5BCR RNA (7 nM, 2 × 10⁴ cpm) by increasing amounts (0.5 to 2.0 μg) of unlabeled RNA transcripts of the 5′ NTR (5 to 20 μM), NS5BCR (6 to 24 μM), and the 3′ NTR (7 to 28 μM). Gel shift analysis was carried out in the presence of tRNA as described in Materials and Methods.

FIG. 8.

Specificity of NS5B binding to NS5BCR RNA fragments. (A) Gel shift analysis of NS5B (1 μM) binding to full-length NSBCR RNA (7 nM, 2 × 10⁴ cpm) in the presence of tRNA (see Materials and Methods). Unlabeled NS5BCR RNA fragments containing deletions (Δ) were used as competitors as indicated (NS5BCR WT, 13 μM; NS5BCR [ΔV+ΔVI+ΔL], 25 μM; NS5BCR [ΔV], 15 μM; NS5BCR [ΔVI+L], 17 μM; NS5BCR [ΔL], 14 μM). (B) Filter-binding assays of NS5B binding to full-length NS5BCR RNA (7 nM, 2 × 10⁴ cpm) or to NS5BCR RNAs containing deletions. The assays were done in the presence of tRNA (see Materials and Methods). The data are expressed as a percentage of the total NS5BCR RNA that is protein bound. The deleted regions of the RNA probes are indicated by dotted lines in the drawings on the right.