Essential role of domain III of nonstructural protein 5A for hepatitis C virus infectious particle assembly

Abstract

Persistent infection with the hepatitis C virus (HCV) is a major risk factor for the development of liver cirrhosis and hepatocellular carcinoma. With an estimated about 3% of the world population infected with this virus, the lack of a prophylactic vaccine and a selective therapy, chronic hepatitis C currently is a main indication for liver transplantation. The establishment of cell-based replication and virus production systems has led to first insights into the functions of HCV proteins. However, the role of nonstructural protein 5A (NS5A) in the viral replication cycle is so far not known. NS5A is a membrane-associated RNA-binding protein assumed to be involved in HCV RNA replication. Its numerous interactions with the host cell suggest that NS5A is also an important determinant for pathogenesis and persistence. In this study we show that NS5A is a key factor for the assembly of infectious HCV particles. We specifically identify the C-terminal domain III as the primary determinant in NS5A for particle formation. We show that both core and NS5A colocalize on the surface of lipid droplets, a proposed site for HCV particle assembly. Deletions in domain III of NS5A disrupting this colocalization abrogate infectious particle formation and lead to an enhanced accumulation of core protein on the surface of lipid droplets. Finally, we show that mutations in NS5A causing an assembly defect can be rescued by trans-complementation. These data provide novel insights into the production of infectious HCV and identify NS5A as a major determinant for HCV assembly. Since domain III of NS5A is one of the most variable regions in the HCV genome, the results suggest that viral isolates may differ in their level of virion production and thus in their level of fitness and pathogenesis.

Figure 1. Schematic representation of constructs used in this study.

(A) Structure of chimeric HCV genomes consisting of the coding region of core up to the first transmembrane segment of NS2 derived from the genotype 2a isolate J6CF and the remainder of the JFH-1 isolate. Jc1/ΔE1E2 carries an in-frame deletion of 350 codons within the E1-E2 coding region (indicated by dotted boxes). Jc1/ΔE1E2+Δ2328-2435 carries an additional deletion of domain III of the NS5A coding sequence. The structure of the genomic luciferase reporter virus genome is drawn below. It is a bicistronic RNA with the first cistron encoding the firefly luciferase gene (Luc) expressed via the HCV IRES whereas the polyprotein is expressed via the IRES of the encephalomyocarditis virus (E–I). This construct design is analogous to the one of the subgenomic helper RNA (sgLuc/wt; schematic in the bottom). However, it lacks the coding region from core to NS2 and was used for trans-complementation assays. (B) Domain structure of NS5A (upper panel) and deletion mutations in domains II and III (lower panels). For details see text. The structures of the deletion mutants lacking all or parts of domain II or domain III are shown below and their designation is given in the right. Numbers indicate the last and the first amino acid position flanking the deletion or the N- and the C-terminus of JFH-1 NS5A.

Figure 2. Transient replication of NS5A deletion mutants and release of virus from cells transfected with these genomes.

(A) Huh7-Lunet cells were transfected with luciferase reporter virus constructs specified in the bottom. Cells were lysed at given time points after transfection and luciferase activity was determined. A representative experiment of three independent repetitions, each value measured in duplicate is shown. Note that the luciferase reporter genomes have a delayed replication kinetic compared to reporter-free genomes and that Huh7-Lunet cells support HCV spread only poorly . Therefore, luciferase levels in transfected cultures are not affected by virus release and spread and thus reflect almost exclusively RNA replication. (B) Kinetic of intracellular accumulation of core protein (left panel) and release of core protein (right panel) from Huh7-Lunet cells transfected with reporter-free Jc1 constructs. Core amounts were determined by using core-specific ELISA. Intracellular core amounts were normalized to the 4h value that reflects transfection efficiency. A representative experiment of at least three independent repetitions is given in each panel. (C) Efficiency of core protein release from cells transfected with reporter-free Jc1 or given mutants. The percentage of released core protein in relation to total core protein (the sum of intra and extracellular core protein) was calculated for each time point. A representative experiment of three independent repetitions is shown. (D) Release of infectivity from cells transfected with reporter-free Jc1 or mutants specified in the top by using a limiting dilution assay. Cell free culture fluids were harvested at different time points after transfection and titrated to determine viral infectivity of the indicated genomes.The grey bar represents the detection limit of the assay.

Figure 3. Efficiency of virus assembly and release of Jc1 carrying deletions in NS5A.

Huh7-Lunet cells were transfected with the genomes specified in the bottom and 48 h post transfection total infectious particle production was determined. Supernatants and cells were treated by freeze-thaw cycles and viral infectivity was determined by a limiting dilution assay. Mean values of four independent experiments are shown and the standard deviations of the means are presented. The grey bar represents the detection limit of the assay.

Figure 4. Analysis of the phosphorylation status of NS5A expressed from Jc1, Jc1/ΔE1E2 and Jc1-derived NS5A mutants.

Huh7-Lunet cells were transfected with the indicated genomes and lysed in RIPA buffer 72 h post transfection. Cell lysates were treated with aceton/methanol and one-half of the precipitated protein was incubated with λ-phosphatase whereas the other half was mock treated in phosphatase buffer. Both samples were separated by 8% SDS-PAGE and NS5A was detected by Western blot. The positions of p56 and p58 are indicated in the left.

Figure 5. Deletion of domain III of NS5A abolishes NS5A colocalization with core on LDs.

Huh7-Lunet cells were transfected with Jc1 (panel I to V), Jc1/Δ2328-2435 (panel VI to X), Jc1/ΔE1E2 (panel XI to XV), and Jc1/ΔE1E2+Δ2328-2435 (panel XVI to XX) and fixed 72 h post transfection. Fixed cells were analyzed for the subcellular localization patterns of core (red) and NS5A (blue) by immunofluorescence whereas LDs were stained with BODIPY493/503 (green). Images were generated with a spinning disk confocal and they represent single optical planes. The locations of the 4 cropped sections shown for each of the constructs in the right are boxed in each overview panel. Scale bars represent 5 µm and 1 µm in single cell and cropped sections, respectivetly.

Figure 6. Mutations in domain III of NS5A impairing virus assembly also impair colocalization of NS5A with core protein on LDs.

(A) The quantitation of colocalization between NS5A and core protein, or NS5A and NS4B in a population of 50 cells per construct is shown in the upper panel. For the measurement of the degree of colocalization, the correlation coefficients between the signal obtained for the respective proteins were calculated by using the plugin ‘Intensity correlation analysis’ of Image J. The lower panel summarizes the association of core protein and LDs in cells transfected with the constructs specified between both panels. For calculation of the percentage of core protein associated with LDs about 2,000 to 2,500 LDs were counted for each construct and colocalization with core protein was analyzed by using the plugin ‘RG2B colocalization’ of Image J. Error bars in both panels represent the standard errors of the means of 10 to 15 cells per construct. (B) Representative confocal immunofluorescence images of cells transfected with the constructs as specified in panel A.

Figure 7. Disruption of NS5A – core colocalization on LDs by deletion of NS5A domain III.

(A) Cells transfected with the Jc1 wild type, the envelope deletion mutant (Jc1/ΔE1E2) or the domain III deletion mutant (Jc1/Δ2328-2435) were grown on coverslips and double labeled with core- and NS5A-specific antibodies; LDs were stained with BODIPY493/503 (green) and the nuclei with DAPI. Subcellular distribution of HCV proteins on LDs was analyzed by confocal fluorescence microscopy. For enhanced clarity of 3-dimensional distribution of markers, the image stacks were deconvolved as described in Materials and Methods. The deconvolved z-stacks are shown in a projection including a shadow cast to create a 3-dimensional impression. Arrows refer to LDs with core – NS5A colocalization. Numbers and arrows in the upper and middle panel indicate colocalization patterns as described in the text. (B) Detection of HCV core and NS5A in transfected cells by using immuno electron microscopy. Cells transfected with the Jc1 wild type, the envelope deletion mutant and the NS5A mutant were grown in 10 cm² cell culture dishes and fixed after 48 hours. Thawed cryosections were prepared (as described in Materials and Methods) and double-labeled with antibodies to HCV NS5A (10-nm-diameter protein A-gold) and HCV Core (15-nm-diameter protein A-gold). Bars, 200 nm. LD, lipid droplet. A quantification of the immunolabeling is summarized in Table 1.

Figure 8. Particle assembly of NS5A mutants can be rescued by trans-complementation.

(A) Schematic of the experimental approach. Genomic (Jc1) wild type or mutant RNA was cotransfected with one of the given helper RNAs (sg/lucwt or sg/lucΔ2328-2435). Replication of the helper was determined by luciferase assay. Release of infectious particles from transfected cells was determined by infection of naïve Huh7.5 cells with supernatants (sup.) from transfected cells and subsequent analysis of infected cells (1^st passage) by luciferase assay, immunofluorescence analysis or TCID50 assay. Infectivity release from these infected cells was determined by subsequent inoculation of naïve Huh7.5 cells (2^nd passage) with the corresponding supernatants (sup.) and infection was determined by luciferase assay or TCID₅₀ assay. (B) Transient replication of sg/luc helper RNAs 24, 48 and 72 h after co-transfection with Jc1 wild type or the Jc1/Δ2328-2435 mutant. (C) Supernatants from transfected cells (panel B) were harvested 24, 48 and 72 h post transfection and used for inoculation of naïve Huh7.5 cells. The level of trans-packaged or trans-complemented helper RNA was determined by luciferase assay with cells 72 h after inoculation. (D) Supernatants from transfected cells (panel B) were harvested 48 h post transfection and used for inoculation of naïve Huh7.5 cells. Seventy two hours after inoculation cells were fixed and NS5A or core protein were detected by immunofluorescence. Nuclei of cells were stained with DAPI. (E) Supernatants from cells after first passage were harvested 72 h post infection and used to inoculate naïve Huh7.5 cells. Infectivity in the supernatant of these cells was determined by TCID₅₀ assay with supernatant harvested 72 h post inoculation. Values shown in panels B, C and E represent the mean of two independent experiments, each measured in duplicate and the error bars.