doi: 10.1371/journal.ppat.0030139.
Hepatitis C virus induces E6AP-dependent degradation of the retinoblastoma protein
Affiliations
- PMID:17907805
- PMCID: PMC2323300
- DOI: 10.1371/journal.ppat.0030139
Item in Clipboard
Hepatitis C virus induces E6AP-dependent degradation of the retinoblastoma protein
Tsubasa Munakata et al. PLoS Pathog..
Display options
Format
Display options
Format
Abstract
Hepatitis C virus (HCV) is a positive-strand RNA virus that frequently causes persistent infections and is uniquely associated with the development of hepatocellular carcinoma. While the mechanism(s) by which the virus promotes cancer are poorly defined, previous studies indicate that the HCV RNA-dependent RNA polymerase, nonstructural protein 5B (NS5B), forms a complex with the retinoblastoma tumor suppressor protein (pRb), targeting it for degradation, activating E2F-responsive promoters, and stimulating cellular proliferation. Here, we describe the mechanism underlying pRb regulation by HCV and its relevance to HCV infection. We show that the abundance of pRb is strongly downregulated, and its normal nuclear localization altered to include a major cytoplasmic component, following infection of cultured hepatoma cells with either genotype 1a or 2a HCV. We further demonstrate that this is due to NS5B-dependent ubiquitination of pRb and its subsequent degradation via the proteasome. The NS5B-dependent ubiquitination of pRb requires the ubiquitin ligase activity of E6-associated protein (E6AP), as pRb abundance was restored by siRNA knockdown of E6AP or overexpression of a dominant-negative E6AP mutant in cells containing HCV RNA replicons. E6AP also forms a complex with pRb in an NS5B-dependent manner. These findings suggest a novel mechanism for the regulation of pRb in which the HCV NS5B protein traps pRb in the cytoplasm, and subsequently recruits E6AP to this complex in a process that leads to the ubiquitination of pRb. The disruption of pRb/E2F regulatory pathways in cells infected with HCV is likely to promote hepatocellular proliferation and chromosomal instability, factors important for the development of liver cancer.
Conflict of interest statement
Figures
(A) Schema for experiment shown in (B–E). Huh-7.5 cells were seeded into culture vessels and infected with JFH1 virus (MOI ∼1) at intervals, and subsequently lysed simultaneously for immunoblot analysis. Solid bars show infected; shaded bars, uninfected. (B) Immunoblots for total pRb and NS5B in cells infected with JFH1 virus according to the protocol outlined in (A). GAPDH was used as a loading control. (C) Mean ± range of values for abundance of pRb in cells infected with JFH1 virus in replicate experiments, relative to abundance in uninfected cells. (D) Immunoblots for phospho-pRb (residues 807/811) and NS5B in cells infected with JFH1 virus. GAPDH was used as a loading control. Lanes 6–7 demonstrate the specificity of the phospho-pRb: lysate from uninfected Huh-7.5 cells was subjected to immunoblotting before (lane 6) and after (lane 7) digestion with lambda protein phosphatase (λ-PPase). (E) Mean ± range of values for abundance of phospho-pRb (residues 807/811) in cells infected with JFH1 virus in replicate experiments, relative to abundance in uninfected cells.
(A) pRb accumulates in the cytoplasm of cells 48 h after infection with genotype 2a JFH1 virus. Cells were labeled with polyclonal antibody to the HCV NS5A protein (Alexa-594; red) and monoclonal anti-pRb (fluorescein; green), and counterstained with DAPI (blue) to visualize nuclei. Two cells near the center of the field are infected with virus and show cytoplasmic displacement of pRb. Frames: i, DAPI only; ii, pRb; iii, NS5A; iv, merge. (B) Confocal microscopic images of Huh-7.5 cells 84 h after infection with the genotype 1a H77S virus [20] at low MOI. Labeling of HCV antigens with polyclonal human antibody (fluorescein; green) revealed both infected and uninfected cells. pRb was labeled with monoclonal antibody (Alexa-568; red), and in infected cells has been either partially (frame iv, arrow) or completely (frame ii, arrow) redirected to the cytoplasm. Nuclei were visualized by DAPI staining. Frames: i, DAPI only; ii, pRb; iii, HCV; iv, merge. (C) The cytoplasmic and nuclear abundance of pRb was determined quantitatively in confocal microscopic images of H77S virus–infected Huh-7.5 cells using MetaMorph software (Molecular Devices,http://www.moleculardevices.com/ ). The average integrated intensities of the pRb fluorescence signal were determined for both the cytoplasm and nucleus (delineated by DAPI staining) in individual infected and immediately adjacent uninfected cells. These data were used to calculate a cytoplasmic–nuclear pRb ratio for each individual cell. Infected cells (identified by labeling of HCV antigens as in [A]) and uninfected cells were analyzed separately, revealing a striking difference in the ratio of cytoplasmic–nuclear pRb staining intensity in infected (0.65) versus uninfected (0.17) cells (p < 0.001 by Studentt test). (D) Confocal microscopy images of phospho-pRb expression in uninfected Huh-7.5 cells (frames i and ii), and 96 h following infection with the genotype 2a JFH1 virus (frames iii–vi). Frames i, iii, and v show labeling with monoclonal antibody to phospho-pRb 807/811 (red) only, whereas frames ii, iv, and vi show merged images of phospho-pRb (red), HCV (human polyclonal antibody; green), and DAPI labeling. The red channel gain was increased in frames iii–vi in order to normalize the intensity of phospho-pRb labeling within the nuclei of noninfected cells with that in frames i and ii. Nuclear phospho-pRb expression was ablated by JFH1 infection, and only partially restored after 20 h of treatment with 50 nM epoxomicin (frames v and vi). The arrow in frame vi indicates an accumulation of phospho-pRb within the cytoplasm of an infected cell following epoxomicin treatment.
(A) Pulse-chase labeling of pRb in 2–3 replicon (squares) and cured 2–3c (circles) cells. Cells were pulse labeled with [35S]-methionine/cysteine for 30 min, then chased with DMEM supplemented with cysteine and methionine and lysed at the times indicated. pRb was immunoprecipitated from lysates and separated by SDS-PAGE. [35S] was quantified by PhosphoImager analysis. Results represent the mean ± range % [35S] label recovered from each cell type at 0 h. (B) Inhibition of the proteasome with epoxomicin (EPX) partially restores pRb abundance in NNeo-C5B/2–3 cells containing an autonomously replicating, genome-length HCV RNA replicon. 2–3 cells and their interferon-cured, HCV-negative progeny, 2–3c cells [58], were treated with 0, 50, or 500 nM EPX for 20 h, followed by lysis and immunoblot analysis of pRb and NS5B. GAPDH was used as a loading control. (C) Inhibition of the proteasome with MG115 increases cytoplasmic pRb abundance in 2–3 replicon cells. 2–3 (top and middle frames) and 2–3c (lower frames) cells were treated with 0 μM (top frames) or 20 μM (middle and lower frames) MG115 for 8 h prior to fixation, labeling with anti-pRb and DAPI, and examination with a fluorescence microscope. Frames on the left show pRb labeling (fluorescein; green), while frames on the right show merged pRb and nuclear (DAPI; blue) labeling. Arrows denote MG115-treated 2–3 cells in which pRb expression is entirely cytoplasmic, with no apparent nuclear staining. (D) EPX treatment restores (lanes 1–2) pRb and (lanes 3–4) phospho-pRb abundance, and increases the abundance of high-molecular-mass pRb-immunoreactive protein (probable polyubiquitinated pRb) in HCV-infected cells. Cells were infected for 96 h with JFH1 virus and treated with 250 nM EPX for 20 h prior to lysis and immunoblot analysis.
(A) Polyubiquitination of pRb in HCV-infected hepatoma cells. Huh-7.5 cells were infected with JFH1 virus (+) or mock-infected (−) for 96 h, and treated (+) or not treated (−) with EPX (250 nM) for 20 h prior to lysis, precipitation with anti-pRb, and immunoblotting with either anti-ubiquitin (Ubi; top panel) or anti-pRb antibody (bottom panels). (B) Polyubiquitination of pRb in HCV replicon cells. 2–3 and 2–3c cells were transfected with a Flag-ubiquitin expression vector and cultured in the presence or absence of MG115 prior to preparation of lysates. Lysates were immunoprecipitated with anti-Flag antibody, followed by immunoblotting with anti-pRb (top panel) or anti-NS5B (middle panel). The lower panel shows direct NS5B immunoblotting of the cell lysates. Ubiquitination of endogenous pRb is evident in the replicon cells even in the absence of MG115 treatment. NS5B is associated with the ubiquitinated Rb. (C) Ectopic expression of wild-type (wt) NS5B induces polyubiquitination of pRb in normal Huh-7 cells. Cells were transfected with vectors expressing Flag-tagged NS3-4A, NS4B, NS5A, NS5B, and a mutant NS5B, D318N/D319N, then treated with MG115. Cell extracts were immunoprecipitated with anti-pRb, followed by immunoblotting with anti-ubiquitin (top panel). The bottom panels show immunoblots of the cell extracts carried out with antibodies to pRb, Flag (expression control), and actin (loading control).
(A) siRNA pools specific for the E3 ubiquitin ligases E6AP, MDM2, and NEDD4, and control siRNAs C1 and C2 were transfected into cured 2–3c cells (lanes 1–5) and 2–3 replicon cells (lanes 6–10) using a liposome-mediated procedure. Cell extracts were prepared 72 h later, separated by SDS-PAGE, and immunoblotted using antibodies to pRb and each of the target E3 ligases. While the each specific siRNA pool effectively reduced the abundance of its E3 ligase target, only the E6AP pool caused a reproducible restoration of pRb abundance in the 2–3 cells. (B) Effect of E6AP knockdown on pRb abundance using each of the four different E6AP-specific siRNAs (E5, E6, E7, and E8) comprising the siRNA pool tested in (A). C1 and C2 are control siRNAs. Lysates were prepared 120 h following transfection. (C) Transfection of a mutated E6AP siRNA (E5mut) fails to induce restoration of pRb abundance in 2–3 replicon cells. Nucleotide sequences of the E5 and E5mut siRNAs are shown at the bottom of the panel, with the two base substitutions underlined.
(A) Endogenous pRb interacts with E6AP in 2–3 replicon cells. Lysates from 2–3 and 2–3c cells were immunoprecipitated with anti-Flag (control), anti-pRb, or anti-MDM2 monoclonal antibody, followed by immunoblotting with anti-pRb and anti-E6AP. All monoclonal antibodies were derived from the same isotype, IgG1. Ig heavy chain (HC) and nonspecific bands served as loading controls for the pRb and E6AP immunoblots, respectively. (B) Endogenous pRb interacts with NS5B in 2–3 replicon cells in an EPX-sensitive fashion. 2–3 and 2–3c cells were cultured in the presence or absence of EPX prior to preparation of lysates. Lysates were immunoprecipitated with monoclonal anti-pRb antibody, followed by immunoblotting with anti-pRb and anti-NS5B antibodies. Note that EPX treatment reduces the apparent interaction between pRb and NS5B. (C) Ectopic expression of wild-type NS5B induces interaction between E6AP and pRb in normal Huh-7 cells. Cells were transfected with vectors expressing Flag-tagged NS3-4A, NS4B, NS5A, NS5B, and a mutant NS5B, D318N/D319N, that does not bind pRb [8], and cultured in the presence of MG115. Cell extracts were immunoprecipitated with anti-pRb antibody, followed by immunoblotting with anti-pRb and anti-E6AP antibodies (top panels). The bottom panels show immunoblots of the cell extracts with the same antibodies.
(A) Immunoblot analysis showing that ectopic expression of a dominant-negative E6AP mutant (E6AP-C840A) partially restores pRb abundance in HCV replicon cells. 2–3 and cured 2–3c cells were transfected with vectors expressing wild-type E6AP and E6AP-C840A, which contains a mutation within the catalytic site of the ubiquitin ligase and acts as dominant-negative [35]. Cell extracts were immunoblotted with anti-pRb and anti-E6AP antibodies. Actin was used as a loading control. (B) Mean ± standard deviation of values for abundance of pRb in cells transfected with wild-type and mutant E6AP expression vectors obtained in three replicate experiments carried out as shown in (A), relative to abundance in cells transfected with an empty vector. (C) Quantitative pRb immunoblots of serial 2-fold dilutions of lysates of 2–3 cells transfected with expression vectors for wt E6AP, E6AP-C840A, or empty vector.
References
- Pawlotsky JM. Therapy of hepatitis C: From empiricism to eradication. Hepatology. 2006;43:S207–S220. - PubMed
- Levrero M. Viral hepatitis and liver cancer: The case of hepatitis C. Oncogene. 2006;25:3834–3847. - PubMed
- Farazi PA, Depinho RA. Hepatocellular carcinoma pathogenesis: From genes to environment. Nat Rev Cancer. 2006;6:674–687. - PubMed
- Lemon SM, Walker C, Alter MJ, Yi M. Hepatitis C viruses. In: Knipe D, Howley P, Griffin DE, Martin MA, Lamb RA, et al., editors. Fields' virology. 5th edition. Philadelphia: Lippincott Williams and Wilkins; 2007. pp. 1253–1304.
- Honda M, Kaneko S, Shimazaki T, Matsushita E, Kobayashi K, et al. Hepatitis C virus core protein induces apoptosis and impairs cell-cycle regulation in stably transformed Chinese hamster ovary cells. Hepatology. 2000;31:1351–1359. - PubMed
Publication types
MeSH terms
Substances
Grants and funding
LinkOut - more resources
Full Text Sources
Other Literature Sources