Highly permissive cell lines for subgenomic and genomic hepatitis C virus RNA replication

Abstract

Hepatitis C virus (HCV) replication appears to be restricted to the human hepatoma cell line Huh-7, indicating that a favorable cellular environment exists within these cells. Although adaptive mutations in the HCV nonstructural proteins typically enhance the replicative capacity of subgenomic replicons in Huh-7 cells, replication can only be detected in a subpopulation of these cells. Here we show that self-replicating subgenomic RNA could be eliminated from Huh-7 clones by prolonged treatment with alpha interferon (IFN-alpha) and that a higher frequency of cured cells could support both subgenomic and full-length HCV replication. The increased permissiveness of one of the cured cell lines allowed us to readily detect HCV RNA and antigens early after RNA transfection, eliminating the need for selection of replication-positive cells. We also demonstrate that a single amino acid substitution in NS5A is sufficient for establishing HCV replication in a majority of cured cells and that the major phosphate acceptor site of subtype 1b NS5A is not essential for HCV replication.

FIG. 1.

Schematic representation of HCV RNAs used in this study. The 5′ and 3′ NTR structures are shown, and ORFs are depicted as open boxes with the polyprotein cleavage products indicated. The first 12 amino acids of the core-coding region (solid box), the neo gene (Neo; shaded box), the EMCV IRES (EMCV; solid line), and ubiquitin (cross-hatched box) are illustrated. Locations of the NS5A adaptive mutations S2204I (*) and Δ47aa are indicated.

FIG. 2.

Identification of Huh-7 lines highly permissive for HCV replication. Huh-7 cells that had been cured of self-replicating subgenomic RNAs by extended IFN-α treatment were electroporated with 1 μg of the subgenomic replicons SG-Neo (S2204I), SG-Neo (5AΔ47), and SG-Neo (wt). Forty-eight hours later, cells were subjected to G418 selection, and the resulting colonies were fixed and stained with crystal violet. Representative plates are illustrated, with the number of transfected cells seeded per 100-mm-diameter dish shown on the left. Percentages below each dish refer to the calculated G418 transduction efficiency of the replicon. To determine the G418 transduction efficiency, transfected cells were serially titrated from 5 × 10⁵ to 10³ cells per 100-mm-diameter dish, together with feeder cells electroporated with the pol⁻ replicon. The resulting G418-resistant foci were counted for at least three cell densities, and the relative G418 transduction efficiency was expressed as a percentage, after dividing the number of colonies by the number of electroporated cells initially plated. Similar transduction efficiencies were obtained in two independent transfections. A poliovirus subgenomic replicon expressing GFP (see Materials and Methods) was electroporated in parallel. Based on both the fraction of GFP-positive cells and replicon-induced cytopathogenicity, ∼90% of cells were routinely transfected. NT, not tested

FIG. 3.

Detection of HCV proteins and RNA in Huh-7.5 and Huh-7 cells transiently transfected with HCV RNA. Top panel, Huh-7.5 and Huh-7 cells were transfected with the subgenomic replicons pol⁻ (lanes 1 and 7), SG-5′HE (S2204I) (lanes 2 and 8), SG-5′HE (5AΔ47) (lanes 3 and 9), SG-Neo (S2204I) (lanes 4 and 10), SG-Neo (5AΔ47) (lanes 5 and 11), and FL (S2204I) HCV RNA (lanes 6 and 12). At 96 h posttransfection, monolayers were incubated for 10 h in the presence of [³⁵S]methionine and [³⁵S]cysteine. Labeled cells were lysed and immunoprecipitated with HCV-positive human serum (anti-NS3, NS4B, and NS5A), and labeled proteins were separated by SDS-10% PAGE. Note that twice the amount of immunoprecipitated sample was loaded in lanes 6 and 12. The mobilities of molecular weight standards (in thousands) are indicated on the left, and the migration of NS3, NS4B, NS5A, and 5AΔ47 is shown on the right. For data in the panel directly below the gel, total cellular RNA was extracted at 96 h posttransfection and quantified for HCV RNA levels as described in the Materials and Methods. The ratio of HCV RNA relative to the pol⁻ defective replicon is shown (HCV RNA/pol⁻). HCV RNA levels relative to the pol⁻ control were comparable in three independent experiments. For data in the bottom two panels, 96 h after transfection cells were fixed with 4% paraformaldehyde, permeabilized with 0.1% saponin, stained for either HCV core or NS3 antigens, and analyzed by FACS. The percentage of cells expressing core and NS3 relative to an isotype matched irrelevant IgG is displayed. Values <1.5% were considered negative (−).

FIG. 4.

HCV RNA accumulation after transfection of Huh-7.5 cells with full-length HCV RNA. One microgram of in vitro-transcribed RNA was electroporated into Huh-7.5, and 2 × 10⁵ cells were plated into 35-mm-diameter wells. Total cellular RNA was isolated at 24, 48, and 96 h posttransfection, and HCV RNA levels were quantified as described in the Materials and Methods. The ratio of HCV RNA relative to the pol⁻ defective subgenomic RNA (HCV RNA/pol⁻) was plotted against the time posttransfection, and similar results were obtained when this experiment was repeated

FIG. 5.

Effect(s) of alternative substitutions at position 2204 in NS5A on HCV RNA replication. Huh-7.5 cells were transfected with 1 μg of the SG-5′HE replicons carrying the indicated amino acid substitutions and 2 × 10⁵ cells plated in 35-mm-diameter wells. After 24, 48, and 96 h in culture, total cellular RNA was extracted and HCV RNA levels were measured as described in Materials and Methods. The ratio of HCV RNA relative to the pol⁻ defective subgenomic RNA (HCV RNA/pol⁻) was plotted against the time posttransfection. The increase in HCV RNA above pol⁻ is indicated above each bar. In this figure the levels of HCV RNA relative to the pol⁻ are the highest we have achieved so far. When these RNAs were transfected into Huh-7.5 cells a second time, a similar trend in HCV RNA accumulation was observed.

FIG. 6.

Effect(s) of combining NS5A adaptive mutations on HCV RNA replication. Subgenomic replicons (SG-5′HE) carrying the indicated mutations were transfected into Huh-7.5 cells and HCV RNA levels were quantitated as described in Fig. 5. The ratio of HCV RNA relative to the pol⁻ defective subgenomic RNA (HCV RNA/pol⁻) was plotted against the time posttransfection, and the increase in HCV RNA above pol⁻ is indicated above each bar. An additional transfection experiment yielded HCV RNA/pol⁻ ratios similar to those illustrated here.

FIG. 7.

Effect(s) of combining NS3 and NS5A mutations on HCV RNA replication. Subgenomic replicons lacking neo (SG-5′HE) were generated carrying S2204I with further mutations in NS3. (A) For the gel shown at top, 96 h after RNA transfection of Huh-7.5 cells, monolayers were labeled with ³⁵S-protein labeling mixture; cells were lysed; and NS3, NS4A, and NS5A were analyzed by immunoprecipitation, SDS-10% PAGE, and autoradiography. Positions of the molecular weight standards (in thousands) are given on the left, and HCV-specific proteins are indicated to the right. For data in the panel directly below the gel, total cellular RNA was extracted at 96 h posttransfection and HCV RNA levels were quantified as described in Materials and Methods. The ratio of HCV RNA relative to the pol⁻ negative control is shown (HCV RNA/pol⁻). Comparable ratios were obtained in two independent experiments. For data in the bottom two panels, 96 h after transfection, cells were fixed with 4% paraformaldehyde, permeabilized with 0.1% saponin, stained for HCV NS3 and NS5B antigens, and analyzed by FACS. The percentage of cells expressing NS3 and NS5B relative to an isotype-matched irrelevant IgG is displayed. Values <1.5% were considered negative (−). (B) Transfected cells seeded in eight-well chamber slides were fixed, permeabilized, and stained for NS5B by immunofluorescence after 96 h in culture. Nuclei were counterstained with Hoechst 33342, and stained cells visualized by fluorescence microscopy (magnification, × 40).

FIG. 7.

Effect(s) of combining NS3 and NS5A mutations on HCV RNA replication. Subgenomic replicons lacking neo (SG-5′HE) were generated carrying S2204I with further mutations in NS3. (A) For the gel shown at top, 96 h after RNA transfection of Huh-7.5 cells, monolayers were labeled with ³⁵S-protein labeling mixture; cells were lysed; and NS3, NS4A, and NS5A were analyzed by immunoprecipitation, SDS-10% PAGE, and autoradiography. Positions of the molecular weight standards (in thousands) are given on the left, and HCV-specific proteins are indicated to the right. For data in the panel directly below the gel, total cellular RNA was extracted at 96 h posttransfection and HCV RNA levels were quantified as described in Materials and Methods. The ratio of HCV RNA relative to the pol⁻ negative control is shown (HCV RNA/pol⁻). Comparable ratios were obtained in two independent experiments. For data in the bottom two panels, 96 h after transfection, cells were fixed with 4% paraformaldehyde, permeabilized with 0.1% saponin, stained for HCV NS3 and NS5B antigens, and analyzed by FACS. The percentage of cells expressing NS3 and NS5B relative to an isotype-matched irrelevant IgG is displayed. Values <1.5% were considered negative (−). (B) Transfected cells seeded in eight-well chamber slides were fixed, permeabilized, and stained for NS5B by immunofluorescence after 96 h in culture. Nuclei were counterstained with Hoechst 33342, and stained cells visualized by fluorescence microscopy (magnification, × 40).

FIG. 8.

Effect(s) of S2194A and S2194D mutations on HCV RNA replication. S2194 was replaced with Ala or Asp in the selectable bicistronic replicon SG-Neo (S2204I), and RNA was transcribed in vitro. (A) RNA transcripts were transfected into Huh-7 cells, and G418-selected colonies were fixed and stained with crystal violet. The relative G418 transduction efficiencies are indicated below each dish. (B) Ninety-six hours posttransfection Huh-7.5 cells were labeled with [³⁵S]methionine and [³⁵S]cysteine for 10 h. Cells were lysed, and HCV proteins were isolated by immunoprecipitation using a patient serum specific for NS3, NS4B, and NS5A. HCV proteins and the positions of protein molecular weight standards (in thousands) are shown. The ratio of HCV RNA relative to the pol⁻ negative control at 96 h posttransfection is shown below each track (HCV RNA/pol⁻). The results illustrated are representative of two independent transfections.