Efficient rescue of hepatitis C virus RNA replication by trans-complementation with nonstructural protein 5A

Abstract

Studies of Hepatitis C virus (HCV) RNA replication have become possible with the development of subgenomic replicons. This system allows the functional analysis of the essential components of the viral replication complex, which so far are poorly defined. In the present study we wanted to investigate whether lethal mutations in HCV nonstructural genes can be rescued by trans-complementation. Therefore, a series of replicon RNAs carrying mutations in NS3, NS4B, NS5A, and NS5B that abolish replication were transfected into Huh-7 hepatoma cells harboring autonomously replicating helper RNAs. Similar to data described for the Bovine viral diarrhea virus (C. W. Grassmann, O. Isken, N. Tautz, and S. E. Behrens, J. Virol. 75:7791-7802, 2001), we found that only NS5A mutants could be efficiently rescued. There was no evidence for RNA recombination between helper and mutant RNAs, and we did not observe reversions in the transfected mutants. Furthermore, we established a transient complementation assay based on the cotransfection of helper and mutant RNAs. Using this assay, we extended our results and demonstrated that (i) inactivating NS5A mutations affecting the amino-terminal amphipathic helix cannot be complemented in trans; (ii) replication of the helper RNA is not necessary to allow efficient trans-complementation; and (iii) the minimal sequence required for trans-complementation of lethal NS5A mutations is NS3 to -5A, whereas NS5A expressed alone does not restore RNA replication. In summary, our results provide the first insight into the functional organization of the HCV replication complex.

FIG. 1.

(A) Structures of replicon RNAs used for trans-complementation assays. Adaptive mutations localized in NS3 (E1202G and T1280I), NS4B (K1846T), NS5A (S2197P), and NS5B (R2884G) are indicated by asterisks. Numbers refer to amino acid positions of the HCV Con-1 complete polyprotein. In the case of the selectable replicons (upper panel), expression of the first cistron is mediated by the HCV IRES, whereas translation of the nonstructural proteins NS3 to -5B is directed by the EMCV IRES (EI). For the luciferase (luc) replicons, translation of this reporter is mediated by the PV IRES (PI), which was fused to the 3′ end of the HCV IRES. Translation of the HCV replicase is again mediated by the EMCV IRES. (B) Schematic drawing of the experimental approach used to analyze trans-complementation. Mutations were introduced into a subgenomic neo replicon carrying one weak adaptive mutation in NS3 (E1202G, indicated by an asterisk). Huh-7 cells containing hyg helper replicons were transfected with mutated RNAs and subjected to selection with G418. Possible outcomes are depicted at the bottom. G418-resistant colonies may arise in case of reversion, recombination, or trans-complementation. Neo or N, neomycin phosphotransferase; Hyg or H, hygromycin phosphotransferase; Luc, P. vulgaris luciferase gene; ΔGDD, deletion of the NS5B GDD motif.

FIG. 2.

Representative results of trans-complementation experiments using selectable replicon mutants. Hygromycin replicon cells (Huh-7-hyg/ET and Huh-7-hyg/5.1) and naive Huh-7 cells were transfected with the neo replicons specified at the top. After G418 selection for 6 weeks, cells were fixed and colonies were stained with Coomassie blue solution.

FIG. 3.

(A) Analysis of HCV RNA species in G418-resistant cell clones by Northern hybridization. Huh-7 cells carrying stably replicating hyg helper RNAs (hyg-ET or hyg-5.1) were transfected with neo replicon mutants carrying lethal mutations in NS5A. After several months of G418 selection, single-cell clones (clones 1 to 9) were established and analyzed for HCV RNAs by using neo- or hyg-specific (negative-sense) riboprobes. As controls for specificity, neo and hyg replicon transcripts (10⁸ molecules) as well as total RNA of naive Huh-7 cells were analyzed in parallel. (B) Conserved mutations in neo replicons isolated from cell clones 3 to 7 and 9. After RT-PCR, two independent DNA clones were sequenced for each given cell clone. Conserved mutations are indicated by a asterisk in the case of the adaptive NS3 mutation and by vertical lines. The minimum region examined by sequence analysis is indicated by dotted lines. In the case of cell clone 4, the entire SfiI fragment was sequenced, and no further mutation was detected.

FIG. 4.

Transient trans-complementation assay. (A) A replication-deficient luciferase replicon (hatched box) was cotransfected with a neo helper RNA (open box) into cells of a highly permissive Huh-7 cell passage. Equal numbers of cells were seeded and harvested at given time points posttransfection. Replication of the mutant was monitored by luciferase assays, whereas replication of the helper RNA was controlled by Northern hybridization. (B) Possible outcomes of trans-complementation after cotransfection of various helper and mutant replicon RNAs. The jagged arrows indicate inactivating mutations in NS5A. Asterisks indicate the positions of adaptive mutations in NS3, NS4A, NS5A, or NS5B. Outcomes of trans-complementation based on the exchange of the NS5A protein and expected replication efficiencies are shown in the right. Numbers in parentheses indicate rows referred to in the text. +, poor replication; ++, moderate replication; +++ high replication.

FIG. 5.

trans-complementation of the replicon mutant luc-2197 + 2201 by different helper RNAs. (A) The inactivating mutations S2197A+S2201A were introduced into the luciferase replicon luc-T (Fig. 1A) carrying two weak adaptive mutations in NS3. After cotransfection with given helper RNAs, cells were harvested 4, 24, 48 and 72 h later and luciferase activities were determined. Values are the ratios of the values at 48 and 4 h, reflecting transfection efficiency. For comparison, the mutant was transfected without helper RNA (hatched box). Luciferase activities measured upon transfection of the parental replicon luc-T served as a positive control (open box). (B) trans-complementation of mutant luc-2197 + 2201-ET with given helper RNAs as described for panel A. Data are means and standard deviations from two independent experiments, each with duplicate measurements. (C) Analysis of the helper RNAs specified in panel B by Northern hybridization. Twenty micrograms of total RNA was analyzed with a neo-specific riboprobe. Total RNA of naive Huh-7 cells served as a negative control. Known amounts of replicon in vitro transcripts were used as positive controls. p.t., posttransfection.

FIG. 6.

trans-complementation of NS5A mutants in a transient assay. (A) Schematic drawing of the NS5A protein structure. The region spanning the N-terminal amphipathic α-helix is shaded. Regions important for the basal phosphorylation of NS5A are indicated by crosshatched boxes. The major phosphoacceptor site (S2194) and the potential hyperphosphorylation sites (S2197, S2201, and S2204) are marked by black dots. (B) Transient replication of NS5A mutants after rescue with helper RNA neo-ET (Fig. 1A). Luciferase activities were determined as described in the legend to Fig. 5. Hatched bars represent the replication levels of the mutant in the absence of helper RNA. Bars are means and error ranges of quadruplicate results.

FIG. 7.

Rescue of NS5A mutants by a nonreplicating helper RNA. (A) Huh-7 cells were cotransfected with the indicated mutants and neo-ΔGDD as helper RNA. The transfection efficiency of the latter was confirmed by Northern hybridization with total RNA prepared from cells harvested 4 h posttransfection (p.t.) (upper panel). For further details, see the legend to Fig. 5C. (B) Huh-7 cells were cotransfected with the mutant luc-2197 + 2201-ET and given helper RNAs. Cells were harvested at 4, 24, and 48 h after transfection, and luciferase activities derived from the mutant were determined. Data were normalized for transfection efficiency by using the 4 h values. A representative result from three independent experiments is shown.

FIG. 8.

Rescue of an NS5A mutant by coexpression of NS3 to NS5A in trans. (A) The structure of the mutant is shown at the top, and the those of the different helper RNAs are given below the mutant structure. All RNAs carry the EMCV IRES at the 5′ end to allow efficient protein expression. Helper RNAs expressing NS3 to NS5Bcontain the HCV 3′ NTR, which is not present with RNAs encoding for NS3 to NS5A or NS5A alone. GND indicates the position of the inactivating amino acid substitution in the NS5B RdRp. Adaptive mutations in NS3, NS4B, and NS5A are marked by asterisks. (B) Huh-7 cells were cotransfected with mutant luc-2201 + 2204-ET and a given helper RNA carrying adaptive mutations as shown in panel A. Cells were harvested at 4, 24 and 48 h after transfection. Shown are the 24-h values normalized to the 4-h value. Data are means and standard deviations from at least three independent experiments. The hatched bar represents the replication efficiency of the mutant in the absence of the helper RNA. (C) Detection of helper RNAs by Northern hybridization. Positions of helper RNAs are indicated on the right. As a negative control, total RNA from naive Huh-7 cells was used (left lane). Note that the NS3 to NS5A helper RNA had a size similar to that of the 28S RNA and therefore could not be detected in this analysis. p.t., posttransfection.