The ESCRT system is required for hepatitis C virus production

Abstract

Background: Recently, lipid droplets have been found to be involved in an important cytoplasmic organelle for hepatitis C virus (HCV) production. However, the mechanisms of HCV assembly, budding, and release remain poorly understood. Retroviruses and some other enveloped viruses require an endosomal sorting complex required for transport (ESCRT) components and their associated proteins for their budding process.

Methodology/principal findings: To determine whether or not the ESCRT system is needed for HCV production, we examined the infectivity of HCV or the Core levels in culture supernatants as well as HCV RNA levels in HuH-7-derived RSc cells, in which HCV-JFH1 can infect and efficiently replicate, expressing short hairpin RNA or siRNA targeted to tumor susceptibility gene 101 (TSG101), apoptosis-linked gene 2 interacting protein X (Alix), Vps4B, charged multivesicular body protein 4b (CHMP4b), or Brox, all of which are components of the ESCRT system. We found that the infectivity of HCV in the supernatants was significantly suppressed in these knockdown cells. Consequently, the release of the HCV Core into the culture supernatants was significantly suppressed in these knockdown cells after HCV-JFH1 infection, while the intracellular infectivity and the RNA replication of HCV-JFH1 were not significantly affected. Furthermore, the HCV Core mostly colocalized with CHMP4b, a component of ESCRT-III. In this context, HCV Core could bind to CHMP4b. Nevertheless, we failed to find the conserved viral late domain motif, which is required for interaction with the ESCRT component, in the HCV-JFH1 Core, suggesting that HCV Core has a novel motif required for HCV production.

Conclusions/significance: These results suggest that the ESCRT system is required for infectious HCV production.

Figure 1. ESCRT components are required for the infectious HCV production.

(A) Inhibition of TSG101, Alix, Vps4B, or CHMP4b protein expression by shRNA-producing lentiviral vectors. The results of the Western blot analysis of cellular lysates with anti-TSG101, anti-Alix, anti-Vps4B, anti-CHMP4b, anti-Chk2, anti-HSP70, or anti-β-actin antibody in RSc cells expressing shRNA targeted to TSG101 (TSG101i), Alix (Alixi), Vps4B (Vps4Bi), or CHMP4b (CHMP4bi) as well as in RSc cells transduced with a control lentiviral vector (Con) are shown. Real-time LightCycler RT-PCR for TSG101 (B), Alix (C), Vps4B (D), or CHMP4b mRNA (E) was performed as well as for β-actin mRNA in triplicate. Each mRNA level was calculated relative to the level in RSc cells transduced with a control lentiviral vector (Con) which was assigned as 100%. Error bars in this panel and other figures indicate standard deviations. (F) MTT assay of each knockdown RSc cells at the indicated time. (G) The levels of HCV Core in the culture supernatants from the stable knockdown RSc cells 97 h after inoculation of HCV-JFH1 at an MOI of 0.1 were determined by ELISA. Experiments were done in triplicate and columns represent the mean Core protein levels. (H) The infectivity of HCV in the culture supernatants from the stable knockdown RSc cells 97 hrs after inoculation of HCV-JFH1 at an MOI of 0.1 was determined by a focus-forming assay at 48 hrs post-infection. Experiments were done in triplicate and each virus titer was calculated relative to the level in RSc cells transduced with a control lentiviral vector (Con) which was assigned as 100%. Asterisks indicate significant differences compared to the control treatment. *P<0.05; **P<0.01. (I) The level of intracellular genome-length HCV-JFH1 RNA in the cells at 97 hrs post-infection was monitored by real-time LightCycler RT-PCR. Results from three independent experiments are shown. (J) Inhibition of TSG101, Alix, Vps4B, or CHMP4b protein expression by 72 hrs after transient transfection of RSc cells with a pool of control siRNA (Con) or a pool of siRNA specific for Alix, Vps4B, or CHMP4b (50 nM). The results of the Western blot analysis of cellular lysates with anti-TSG101, anti-Alix, anti-Vps4B, anti-CHMP4b, or anti-β-actin antibody is shown. (K) MTT assay of each knockdown RSc cells at the indicated time. (L) The levels of HCV Core in the culture supernatants were determined by ELISA 24 hrs after inoculation of HCV-JFH1. RSc cells were transiently transfected with a pool of control siRNA (Con) or a pool of siRNA specific for TSG101, Alix, Vps4B, or CHMP4b (50 nM). At 48 hrs after transfection, the cells were inoculated with HCV-JFH1 at an MOI of 5 and incubated for 2 hrs. Then, culture medium was changed and incubated for 22 hrs. Experiments were done in triplicate and each Core level was calculated relative to the level in the culture supernatants from the control cells and indicated below. (M) The infectivity of HCV in the culture supernatants from the transient knockdown RSc cells 24 hrs after inoculation of HCV-JFH1 at an MOI of 5 was determined by a focus-forming assay at 48 hrs post-infection. Experiments were done in triplicate and each virus titer was calculated relative to the level in RSc cells transfected with a control siRNA (Con) which was assigned as 100%. Asterisks indicate significant differences compared to the control treatment. *P<0.05; **P<0.01.

Figure 2. ESCRT system is not required for HCV RNA replication and assembly of intracellular infectious HCV.

(A) Schematic gene organization of subgenomic JFH1 (JRN/3-5B) RNA encoding Renilla luciferase gene. Renilla luciferase gene (RLuc) is depicted as a box and is expressed as a fusion protein with Neo. The HCV RNA replication level in each ESCRT knockdown OR6c JRN/3-5B cells by lentiviral vector-mediated RNA interference (shRNA) was monitored by RL assay. The RL activity (RLU) is shown. The results shown are means from three independent experiments. (B) 72 hrs after the transfection of OR6c JRN/3-5B polyclonal cells with each of the siRNA (50 nM), the HCV RNA replication level was monitored by RL assay as described in (A). (C) Schematic gene organization of genome-length HCV-O RNA encoding Renilla luciferase gene. The position of an adaptive mutation, K1609E, is indicated by a triangle. 72 hrs after the transfection of OR6 cells with each of the siRNA (50 nM), the HCV RNA replication level was monitored by RL assay as described in (A). (D) The infectivity of HCV in the culture supernatants from the transient knockdown RSc cells 24 hrs after inoculation of HCV-JFH1 at an MOI of 2 was determined by a focus-forming assay at 48 hrs post-infection. Experiments were done in triplicate and each virus titer was calculated relative to the level in RSc cells transfected with a control siRNA (Con) which was assigned as 100%. (E) Intracellular HCV infectivity was determined by a focus-forming assay at 48 hrs post-inoculation of lysates by repeated freeze and thaw cycles as described in (D). (F) RSc cells were transiently transfected with a pool of control siRNA (Con) or a pool of siRNA specific for Alix, Vps4B, or CHMP4b (50 nM). At 24 hrs after the transfection, the cells were inoculated with HCV-JFH1 at an MOI of 0.2 and incubated for 48 hrs. Then, culture medium was changed and incubated for 24 hours. Western blotting of cell lysates 72 hrs post-infection with anti-β-actin, anti-HCV NS5A, or anti-HCV Core antibody is shown.

Figure 3. HCV Core is targeted to lipid droplets even in the ESCRT knockdown cells.

The RSc cells transduced with a control lentiviral vector (Con), the TSG101 knockdown (TSG101i), the Alix knockdown (Alixi), the Vps4B knockdown (Vps4Bi), or the CHMP4b knockdown (CHMP4bi) cells were infected with HCV-JFH1. Cells were fixed 60 hrs post-infection and were then examined by confocal laser scanning microscopy. Cells were stained with anti-HCV Core (CP-9 and CP-11 mixture) and were then visualized with Cy3 (red). Lipid droplets and nuclei were stained with BODIPY 493/503 (green) and DAPI (blue), respectively. Images were visualized using confocal laser scanning microscopy. Colocalization is shown in yellow (Merged).

Figure 4. HCV Core interacts with CHMP4b.

(A–C) HCV Core colocalizes with CHMP4b. 293FT cells cotransfected with 100 ng of pcDNA3/core (JFH1) and either 100 ng of pCHMP4b-GFP (A) or pFLAG-CHMP4b (B) were examined by confocal laser scanning microscopy. Cells were stained with anti-HCV Core and anti-FLAG polyclonal antibody and were then visualized with FITC (FLAG-CHMP4b) or Cy3 (Core). Images were visualized using confocal laser scanning microscopy. The right panels exhibit the two-color overlay images (Merged). Colocalization is shown in yellow. (C) The Core or NS5A partially colocalizes with CHMP4b in HCV-JFH1-infected RSc cells. RSc cells transfected with 100 ng of pCHMP4b-GFP were infected with HCV-JFH1. Cells were fixed 60 hrs post-infection and were then examined by confocal laser scanning microscopy as shown in panel (A). (D) HCV Core binds to CHMP4b. 293FT cells transfected with 4 µg of pCHMP4b-GFP, pEGFP C3 (Clontech), pcDNA3-FLAG, pcDNA3-FLAG-Alix or pFLAG-CHMP4b and RSc cells 5 days after inoculation of HCV-JFH1 at an MOI of 4 were lysed. The mixtures of these lysates were immunoprecipitated with either anti-FLAG or Living Colors A.v. monoclonal antibody (anti-GFP antibody), followed by immunoblot analysis using anti-HCV Core, anti-HCV NS5A, anti-FLAG, and/or Living Colors A.v. monoclonal antibody.

Figure 5. Brox is required for HCV life cycle.

(A) Subcellular localization of Myc-tagged TSG101 in HCV-JFH1-infected RSc cells. RSc cells transfected with 100 ng of pBj-Myc-TSG101 were infected with HCV-JFH1. Cells were fixed 60 hrs post-infection and were then examined by confocal laser scanning microscopy as shown in Fig. 3. High magnification image of area 1 is shown. Subcellular localization of endogenous Alix in HCV-JFH1-infected RSc cells 60 hrs post-infection. Cells were stained with anti-Alix and anti-HCV Core antibodies and were examined by confocal laser scanning microscopy. High magnification images of area 1 and area 2 are shown. (B) HCV Core partially colocalizes with Brox. 293FT cells cotransfected with 100 ng of pcDNA3/core (JFH1) and 100 ng of pmGFP-Brox^WT were examined by confocal laser scanning microscopy. (C) The HCV Core partially colocalizes with Brox in HCV-JFH1-infected RSc cells. RSc cells transfected with 100 ng of pmGFP-Brox^WT were infected with HCV-JFH1. Cells were fixed 60 hrs post-infection and were then examined by confocal laser scanning microscopy. (D) Inhibition of Brox mRNA expression by the shRNA-producing lentiviral vector. Real-time LightCycler RT-PCR for Brox was performed as well as for β-actin mRNA in triplicate. Each mRNA level was calculated relative to the level in RSc cells transduced with a control lentiviral vector (Con) which was assigned as 100%. (E) The levels of HCV Core in the culture supernatants from the Brox knockdown RSc cells (Broxi) 72 hrs after inoculation of HCV-JFH1 were determined by ELISA. (F) The infectivity of HCV in the culture supernatants was determined by a focus-forming assay at 48 hrs post-infection. Experiments were done in triplicate and each virus titer was calculated relative to the level in RSc cells transduced with a control lentiviral vector (Con) which was assigned as 100%. Asterisks indicate significant differences compared to the control treatment. *P<0.05; **P<0.01. (G) The levels of intracellular genome-length HCV-JFH1 RNA in the cells used in (E) were monitored by real-time LightCycler RT-PCR.