Fig. 1. Hepatisis C virus (HCV) features. Genome organization and schematic representation of the HCV structural proteins (C, E1 and E2) and non-structural proteins (p7, NS2, NS3, NS4A, NS4B, NS5A and NS5B) as the proteins associate with the endoplasmic reticulum. The function of the non-structural proteins is also indicated. Another HCV protein named “F” might also be synthesized by a ribosomal frameshift in the core coding sequence. E1 and E2 are N-glycosylated. 5′ and 3′NCR, non coding region; IRES, internal ribosome entry site. Initially, the HCV virus recognizes and is incorporated into human liver cells. The internalized virus then dissociates, liberating the viral RNA genome. The HCV RNA is then translated by the host ribosomes, producing the HCV polypeptide. This polypeptide is subsequently processed, first by host peptidases then by the HCV proteases (NS2 and NS3) into 10 different HCV proteins. The non-structural proteins (NS2–NS5b) are next assembled and localized within the liver cell to form a replication complex, which produces multiple copies of the HCV RNA genome. These RNA copies are then able to re-enter the life cycle, producing more HCV proteins. Eventually, the HCV structural proteins (C, E1 and E2) along with copies of the HCV RNA are packaged as infectious virus particles, released from the liver cell, and are able to infect new cells.

Fig. 2. Schematic drawing illustrating the proposed role of mitochondria in the innate immunity and the strategy used by HCV to evade it. Viral infection results in the activation of multiple pathways. The MAVS (mitochondrial antiviral signalling) protein coordinates upstream events leading to activation of the interferon-β (IFN-β). MAVS (also known as IPS-1, VISA and CARDIF) is localized at the outer mitochondrial membrane where it is anchored by a transmembrane helix. The presence of viral RNA replication products is sensed by the cytosolic protein RIG-I (retinoic acid induced gene I) which contains a RNA helicase domain that binds to dsRNA. RIG-I also contains tandem N-terminal caspase recruitment domains (CARDs) that interact with another CARD domain present in the MAVS protein. Binding of RIG-I complexed to dsRNA to MAVS activates downstream signalling pathways. These include activation of the IKK (inhibitor κB kinase) complex (which regulate NF-κB (nuclear factor κB)) and of the non “canonical” kinase TBK-1 (which regulates IRF3/7 (IFN regulatory factor 3/7)). Activation of these transcription factors and their translocation to the nucleus leads to assembly of a multiprotein enhancer complex, the enhanceosome, which drives expression of IFN-β gene. Importantly, the mitochondrial localization of MAVS is essential for its signalling function, because the removal of the mitochondrial targeting domain of MAVS abolishes its function to induce IFNs. The HCV protein NS3/4A (which is a serine protease) cleaves MAVS at Cys-508, resulting in dislocation of the N-terminal fragment of MAVS from the mitochondria. Point mutation of MAVS at Cys-508 renders MAVS resistant to cleavage by NS3/4A, thus maintaining the ability of MAVS to induce interferons in HCV replicon cells. The physical interaction between the ER-bound NS3/4A and the mitochondria-bound MAVS might be provided by contiguity at localised contact sites of the ER-mitochondria networks or by transient membrane fusion-driven intermembrane protein transfer. Drawn from the evidence reported in [7], [8], [9], [10] and [11].

Fig. 3. Regulated HCV-gene expression in U-2 OS-derived cell line. (A) Tetracycline-regulated gene expression system. The system consists of a tetracycline-controlled transactivator (rTA), which is composed of the tetracycline repressor (tet R) fused to the activating domain of VP16 of herpes simplex virus, and of a tTA-dependent promoter, which is composed of a minimal sequence derived from the cytomegalovirus intermediate early promoter (CMV P) combined with heptameric tetracycline operator (tet O) sequences. The tTA-dependent promoter is virtually silent in many cell types in the presence of low concentrations of tetracycline (TET), which prevent the tTA from binding to tet O sequences. In the absence of tetracycline, the tTA binds to the tet O sequences to activate transcription from the minimal promoter. (B) Tightly regulated expression of HCV proteins in U-2 OS cell line. U-2 OS cells were cultured in DMEM supplemented with 10% FBS in the presence of 1 μg/ml tetracycline. Subsequently, in one sample the medium was changed with withdrawal of tetracycline (TET−) and in the other the medium was replaced without omitting tetracycline. After 48 h the cell samples were harvested and assayed. Panel on the left: proteins from cell lysate were separated by 12% SDS-PAGE and analysed by immunoblot with a pool of monoclonal antibodies against core, NS3, NS5b and β-actin. Right panel: immuno-cytochemical detection of HCV-core protein by laser scanning confocal microscopy (LSCM). U-2 OS cells, cultured at low density onto fibronectin coated 35 mm glass bottom dishes as described before, were fixed (4% paraformaldehyde), permeabilised (0.2% Triton X-100), blocked (3% BSA in PBS), and then sequentially incubated with diluted mouse monoclonal anti-core and secondary Rhodamine-conjugated goat anti-mouse IgG. Subsequently cells were treated for nuclear staining with 1 μM Topro-1. Then after cells were examined by a Nikon TE 2000 microscope (images collected using a 60X objective 1.4 NA) coupled to a Radiance 2100 dual laser scanning confocal microscopy system (Biorad). The fluorescence signals of Rhodamine and Topro-1 were monitored sequentially exciting first with the He–Ne laser beam (λ_ex = 543 nm) and then with the Ar–Kr laser beam (λ_ex = 514 nm) respectively. Magnification of the selected area (indicated by the white frame) is shown for the HCV-induced U-2 OS cell sample. The images are representative of three different preparations. Analogous results were obtained for the immuno-localization of the HCV-NS2 protein. Scale bars: 20 μm.

Fig. 4. Functional analysis of HCV protein induction in U-2 OS-derived cell line on the mitochondrial OXPHOS system. (A) Measurements of oxygen consumption in intact cells. 5–7 10⁶ viable cell/ml was assayed polarographycally in 50 mM KPi, 10 mM HEPES, 1 mM EDTA, pH 7.4; after establishment of a stationary endogenous respiratory rate 2 μg/ml of oligomycin was added. Both rates of O₂ consumption were corrected for 2 mM KCN-insensitive respiration. RCR: respiratory control ratio obtained dividing the rates of oxygen consumption attained before and after addition of oligomycin. The values reported are mean (±S.E.M.) of 6 independent preparations. Two tailed Student's t test was applied to evaluate the significance between the rates measured with the non induced (+TET) and induced (−TET) cells. The P value for significant differences is shown. (B) Imaging of ΔΨ-generating mitochondria by LSCM. Cells were seeded at low density onto fibronectin coated 35 mm glass bottom dishes. After adhesion, living cells were directly incubated for 20 min at 37 °C with MitoCapture (1/1000 dilution). Stained cells were washed with PBS and examined by a Nikon TE 2000 microscope (images collected using a 60X objective 1.4 NA) coupled to a Radiance 2100 dual laser scanning confocal microscopy system (Biorad). The fluorescent signal of the MitoCapture double-emitter probe was examined sequentially, exciting first with the Ar–Kr laser beam (λ_ex = 488 nm) and then with the He–Ne laser beam (λ_ex = 543 nm). MitoCapture is a lipophilic cation that accumulates electrophoretically in mitochondria; depending on the extent of the trans-membrane electrical potential the dye accumulates as monomer (green fluorescence emitter) or dimer (red fluorescence emitter, in iper-polarized mitochondria). Magnification of selected areas (indicated by the white frame are shown below each panel. The images are representative of five different preparations. Scale bars: 20 μm. (C) Measurements of the specific activities of NADH-CoQ oxidoreductase (CI), cytochrome c reductase (CIII) and cytochrome c oxidase (CIV). The activities were determined spectrophotometrically on mitoplast fraction of ultrasound-treated U-2 OS cells in 10 mM Tris, 1 mg/ml serum albumin, pH 7.4. CI was assayed (in the presence of antimycin A plus KCN) by following the initial rotenone-sensitive rate of NADH oxidation (ΔA at 340 nm) in the presence of UQ₂ as electron acceptor; CIII was assayed (in the presence of rotenone plus KCN) by following the initial antimycin A-sensitive rate of cytochrome c reduction (ΔA at 550 nm) in the presence of UQ₂ as electron donor; CIV was assayed by following (in the presence of antimycin A) the initial KCN-sensitive rate of cytochrome c oxidation (ΔA at 550 nm) under aerobic conditions. The values reported for the induced cells (TET−) are expressed as percentage of the non-induced (TET+) control cells. Averages (±S.E.M.) of at least 5 different preparations; N.S., not significant.

Fig. 5. LSCM analysis of reactive oxygen species (ROS) production in non-induced and HCV-induced U-2 OS-derived cell line. Imaging of intracellular ROS production by the O₂⁻ and H₂O₂ sensitive probes MitoSOX (A) and DCF (B) respectively. Cells were seeded at low density onto fibronectin coated 35 mm glass bottom dishes. After adhesion, living cells were directly incubated for 20 min at 37 °C with 3 μM MitoSOX or 10 μM dichlorofluorescin-diacetate DCF-DA. Stained cells were washed with PBS and examined as indicated in the legend to Fig. 5. The red fluorescence of MitoSOX, highly selective for detection of superoxide in mitochondria of live cells, was monitored exciting with the He–Ne laser beam (λ_ex = 543 nm). DCF-DA is a membrane permeant probe which is hydrolised by intracellular esterases and converted to the membrane-impermeant and ROS (mainly H₂O₂)-reacting product DCF. The green fluorescence of oxidised DCF was analysed by exciting the sample with the Ar–Kr laser beam (λ_ex = 488 nm). Magnification of selected areas (indicated by the white frame are shown below each panel. The images are representative of five different preparations. Scale bars: 20 μm.

Fig. 6. Schematic working model for the mitochondrial dysfunction caused by HCV infection. Black-outlined boxes summarize the experimental evidence shown in this work with the downward distribution of mitochondrial dysfunctions suggesting a temporary progression of events. In blue possible connection with the HCV-induced ER-stress via Ca²⁺ ER-efflux/mitochondrial overload is shown. See text for further explanation.