Alternative approaches for efficient inhibition of hepatitis C virus RNA replication by small interfering RNAs

Abstract

Persistent infection with hepatitis C virus (HCV) is a leading cause of chronic hepatitis, liver cirrhosis, and hepatocellular carcinoma. It has recently been shown that HCV RNA replication is susceptible to small interfering RNAs (siRNAs), but the antiviral activity of siRNAs depends very much on their complementarity to the target sequence. Thus, the high degree of sequence diversity between different HCV genotypes and the rapid evolution of new quasispecies is a major problem in the development of siRNA-based gene therapies. For this study, we developed two alternative strategies to overcome these obstacles. In one approach, we used endoribonuclease-prepared siRNAs (esiRNAs) to simultaneously target multiple sites of the viral genome. We show that esiRNAs directed against various regions of the HCV coding sequence as well as the 5' nontranslated region (5' NTR) efficiently block the replication of subgenomic and genomic HCV replicons. In an alternative approach, we generated pseudotyped retroviruses encoding short hairpin RNAs (shRNAs). A total of 12 shRNAs, most of them targeting highly conserved sequence motifs within the 5' NTR or the early core coding region, were analyzed for their antiviral activities. After the transduction of Huh-7 cells containing a subgenomic HCV replicon, we found that all shRNAs targeting sequences in domain IV or nearby coding sequences blocked viral replication. In contrast, only one of seven shRNAs targeting sequences in domain II or III had a similar degree of antiviral activity, indicating that large sections of the NTRs are resistant to RNA interference. Moreover, we show that naive Huh-7 cells that stably expressed certain 5' NTR-specific shRNAs were largely resistant to a challenge with HCV replicons. These results demonstrate that the retroviral transduction of HCV-specific shRNAs provides a new possibility for antiviral intervention.

FIG. 1.

Target sequences of esiRNAs and design of a retroviral vector system for the delivery of shRNAs. (A) Schematic representation of the subgenomic HCV replicon I₃₈₉/NS3-3′/LucUbiNeo-ET that persistently replicates in cells of the Huh-7 cell clone 9B. The replicon is composed of the HCV 5′ NTR; nucleotides 342 to 389 of the core coding region (core) fused to the coding sequences of the firefly luciferase gene (Luc), the ubiquitin gene (Ubi), and the neomycin phosphotransferase gene (Neo^r); the IRES of the encephalomyocarditis virus (EMCV IRES); the coding region of the HCV nonstructural proteins NS3 to NS5B; and the HCV 3′ NTR. Black bars indicate viral and replicon nonviral sequences targeted by shRNAs. Note that numbers refer to nucleotide positions of the Con1 consensus genome. (B) Schematic representation of pBABE/H1/SV40/EGZ/ΔU3. The vector contains two heterologous promoter elements that operate in a bidirectional manner. The polymerase III H1-RNA promoter (H1) is used to direct the transcription of shRNAs, whereas an SV40 promoter (SV) directs the transcription of an mRNA which encodes an EGFP-zeocin resistance fusion protein (EGFP-Zeo^r). Note that the 3′ LTR carries a large deletion in the U3 region in order to inactivate U3-dependent gene expression of the integrated provirus. Labeled arrows indicate the start and stop sites of the RNA polymerase III. (C) Predicted secondary structure of the HCV-specific shRNA HCV-321. (D) Predicted secondary structure of the β-galactosidase-specific shRNA LacZ.

FIG. 2.

Effect of different esiRNAs on the replication of a subgenomic HCV replicon. (A) Dose-dependent inhibition of replicon-dependent luciferase reporter gene expression by HCV-specific esiRNAs. (B) Antiviral effect of esiRNAs targeting different regions of the HCV genome. About 2 × 10⁵ cells of the Huh-7 clone 9B (containing replicon I₃₈₉/NS3-3′/LucUbiNeo-ET) were transfected with the indicated amounts of HCV genotype 1b-specific esiRNAs (ORF1, ORF3, or 5′ NTR), unrelated esiRNAs (EGFP or LacZ), tRNA, or poly(IC). As additional controls, cells were treated with the transfection reagent only or left untreated. At the times indicated, cells were lysed and luciferase activities were determined, at least in duplicate. RLU, relative light units.

FIG. 3.

Long-term effect of esiRNAs on the replication of a subgenomic HCV replicon. The effects of single (A) and consecutive (B) esiRNA transfections on replicon-dependent luciferase reporter gene expression are shown. About 2 × 10⁵ cells of the Huh-7 clone 9B (containing the HCV 1b replicon I₃₈₉/NS3-3′/LucUbiNeo-ET) were transfected with 1 μg of HCV genotype 1b-specific esiRNAs (ORF1 and ORF3) or unrelated esiRNAs (EGFP) or were left untreated. At the times indicated, cells were lysed and luciferase activities were determined, at least in duplicate. Arrows and dotted lines indicate times of transfection and cell splitting (ratio of 1:3), respectively. RLU, relative light units. (C) Effect of genotype 1b-specific esiRNA transfections on the replicon copy number. RNAs were prepared from cells that had been lysed 11 days after the first esiRNA transfection, and the numbers of replicon molecules were determined by quantitative RT-PCR. The concentrations of viral RNAs are given in replicon copy numbers per microgram of total RNA. Hatched columns, viral RNA concentration in control cells; black columns, viral RNA concentration after transfection of HCV-specific esiRNAs. The figure shows the results from a single representative experiment.

FIG. 4.

Effect of esiRNAs on the replication of a full-length HCV replicon. (A) Effect of esiRNAs on HCV RNA copy number. About 2 × 10⁵ cells of the Huh-7 clone 20-1 (containing the replicon I₃₈₉/Core-3′/5.1) were transfected with 1 μg of genotype 1b-specific esiRNAs (ORF1, ORF3, or EGFP) or were left untreated. Half of the cells were harvested 2 days later. The other half was passaged (split ratio, 1:4), seeded into new cell culture dishes, and transfected a second time or left untreated. Two days later, these cells were also harvested, RNAs were prepared, and the replicon copy numbers per microgram of total RNA were determined by quantitative RT-PCR. Hatched columns, viral RNA concentration in control cells; black columns, viral RNA concentration after the transfection of HCV-specific esiRNAs. The figure shows the results from a single representative experiment. (B) Effect of esiRNAs on the expression of NS5A. About 2 × 10⁵ cells of clone 20-1 were transfected with 1.5 μg of esiRNAs (ORF1, ORF3, or EGFP) or were left untreated. Two days later, the cells were split in a ratio of 1:3 and seeded onto glass coverslips. After 2 days of further cultivation, cells were fixed, permeabilized, and immunostained for NS5A (red) and counterstained for DNA (blue) by using an NS5A-specific mouse monoclonal antibody and DAPI, respectively.

FIG. 5.

Sequence specificity of esiRNAs. (A) Effect of genotype 1a-, 1b-, and 2a-specific esiRNAs on the replication of an HCV genotype 1b replicon. About 2 × 10⁵ cells of Huh-7 clone 9B (containing the replicon I₃₈₉/NS3-3′/LucUbiNeo-ET) were transfected with 1.5 μg of esiRNAs, as indicated, or were left untreated. Two days later, cells were lysed and luciferase activities were determined, at least in duplicate. RLU, relative light units. (B) Effect of genotype 1a- and 1b-specific esiRNAs on the replication of a chimeric HCV replicon with genotype 1a coding sequences. About 2 × 10⁵ cells of the Huh-7 clone A-3/3 (containing the replicon I₃₈₉/NS3-3′/H77/DR) were transfected with 1 μg of esiRNAs, as indicated, or were left untreated. Half of the cells were harvested 2 days later. The other half was passaged (split ratio, 1:4), seeded into new cell culture dishes, and transfected a second time or left untreated. Two days later, these cells were also harvested, RNAs were prepared, and the replicon copy numbers per microgram of total RNA were determined by quantitative RT-PCR. Hatched columns, viral RNA concentration in control cells; black columns, viral RNA concentration after the transfection of HCV-specific esiRNAs. The figure shows the results from a single representative experiment.

FIG. 6.

Target structures of HCV-specific shRNAs. A schematic representation of the secondary structure of the HCV 5′ NTR, including adjacent coding sequences (modified from the work of Honda et al. [25]), is shown. Domains and subdomains are labeled with roman numerals and lowercase letters, respectively. The position of the AUG start codon within domain IV is indicated by an arced arrow. A pseudoknot structure that is generated by base pairing between domain IIIf and the intervening sequence connecting domains IIIf and IV is indicated by three horizontal lines. Furthermore, the positions of two regions (ranging from nucleotides 24 to 38 and 428 to 444) that are involved in a long-range RNA-RNA interaction (31) are marked with striped bars. Red lines indicate sequences targeted by shRNAs. Note that numbers refer to nucleotide positions of the Con1 consensus genome.

FIG. 7.

Target sequences of HCV-specific shRNAs. The CLUSTAL W algorithm (61) was used to align the first 389 nucleotides of the HCV Con1 consensus genome (genotype 1b) with homologous sequences of six other genomes representing genotypes 1a (strain H77), 2a (strain HC-J6CH), 3a (strain NZL1), 4a (strain HEMA51), 5a (strain EUH1480), and 6b (strain Th580). Black lines indicate sequences targeted by shRNAs. Coding nucleotides are highlighted in gray.

FIG. 8.

Antiviral activity of HCV-specific shRNAs. (A) Effect of shRNAs on replicon-dependent luciferase reporter gene expression in cells persistently replicating a subgenomic HCV replicon. About 6 × 10⁴ cells of clone 9B (containing replicon I₃₈₉/NS3-3′/LucUbiNeo-ET) were transduced thrice with retroviral vectors encoding either an HCV-specific shRNA or an unrelated control shRNA (as indicated). Control cells were transduced with the parental retroviral vector (BABE) or incubated with medium from untransfected 293T cells. About 24 h after the last transduction, the cells were passaged (split ratio, 1:4) and further cultivated. About 96 and 120 h later, the cells were lysed and luciferase activities were determined, at least in duplicate. The mean luciferase activities of nontransduced control cells were set at 100% and used to normalize luciferase activities of transduced cells. The resulting relative luciferase values are shown. (B) Transduction of HCV-specific shRNAs confers resistance against a subgenomic HCV replicon. Naive Huh-7 cells were transduced twice with retroviral vectors that encode either an HCV-specific shRNA or an unrelated control shRNA (as indicated). After an incubation period of 13 days, during which the cells were cultured in the presence of zeocin and split at regular intervals, the cells were electroporated with the subgenomic genotype 1b replicon I₃₈₉/NS3-3′/Luc-ET, seeded into multiple cell culture dishes, and harvested at given time points. The average luciferase activities of nontransduced control cells were set at 100% and used to normalize the corresponding activities of transduced cells. The resulting relative luciferase values are shown.