Signal Peptide Cleavage and Internal Targeting Sig
http://www.100md.com
病菌学杂志 2005年第24期
Institute of Molecular and Cellular Biology and Astbury Centre for Structural Molecular Biology, University of Leeds, Leeds LS2 9JT, United Kingdom
ABSTRACT
The hepatitis C virus (HCV) p7 protein forms an amantadine-sensitive ion channel required for viral replication in chimpanzees, though its precise role in the life cycle of HCV is unknown. In an attempt to gain some insights into p7 function, we examined the intracellular localization of p7 using epitope tags and an anti-p7 peptide antibody, antibody 1055. Immunofluorescence labeling of p7 at its C terminus revealed an endoplasmic reticulum (ER) localization independent of the presence of its signal peptide, whereas labeling the N terminus gave a mitochondrial-type distribution in brightly labeled cells. Both of these patterns could be visualized within individual cells, suggestive of separate pools of p7 where the N and C termini differed in accessibility to antibody. These patterns were disrupted by preventing signal peptide cleavage. Subcellular fractionation revealed that p7 was enriched in a heavy membrane fraction associated with mitochondria as well as normal ER-derived microsomes. The complex regulation of the intracellular distribution of p7 suggests that p7 plays multiple roles in the HCV life cycle either intracellularly or as a virion component.
INTRODUCTION
Hepatitis C virus (HCV) is a major cause of chronic liver disease worldwide and is now the most common reason for liver transplants in Western countries. Unusually for an RNA virus, the majority of patients become persistently infected after a mild acute episode, and clinical intervention occurs in late-stage symptomatic patients. Current therapy comprises high-dose pegylated alpha interferon (IFN) combined with the guanosine analogue ribavirin. The efficacy of this regimen is largely dependent on the viral genotype; the most prevalent genotype 1 viruses possess high levels of innate resistance to IFN, and reservoirs of resistance in other genotypes are building due to the highly variable nature of HCV (40). Interestingly, a recent meta-analysis of clinical trials in which patients were treated with a triple combination of IFN, ribavirin, and amantadine showed that this approach gave improved sustained viral responses in patients that previously did not respond to dual therapy, most often those infected with genotype 1 HCV (16).
HCV has a single-stranded positive-sense RNA of around 9.6 kb and is the prototype member of the Hepacivirus genus of the Flaviviridae family (13, 35). While, until very recently, no robust in vitro replication system has existed for HCV (27, 43, 46), many functions of the viral nonstructural proteins have been elucidated using replicons (7, 8, 23, 28, 33). The inability of these systems to produce extracellular virus has limited studies on structural proteins to virus-like particles (VLPs) made either within insect cells (4, 15) or by mammalian expression (5, 6), or more recently, to the use of pseudotyped retrovirus systems to investigate receptor tropism and cell entry (2, 3, 22).
The p7 protein of HCV is not required for RNA replication or the formation of VLPs in insect cells, and it is uncertain whether it is a virion component. p7 is a small hydrophobic protein of 63 amino acids located within the HCV genome at the junction between the structural and nonstructural proteins (26, 30). We previously showed that p7 from genotype 1b HCV forms an oligomeric ion channel in planar lipid bilayers that can be blocked by amantadine at micromolar concentrations, leading to our proposal that the potential antiviral effect of amantadine described above may be due to its action on p7 (19). Others have subsequently confirmed this ion channel activity for different HCV genotypes and have identified other channel-blocking compounds (32, 34). The important finding that p7 is required for replication of HCV in chimpanzees confirms the protein as a target for antiviral chemotherapy, yet its role in viral replication is unknown (38). The homologous p7 protein from bovine viral diarrhea virus is known to be necessary for the generation of infectious virus particles, though whether virions are able to assemble or are secreted in an immature form is not known, and attempts to detect the protein in purified particles were unsuccessful (18, 21). Furthermore, preventing the already inefficient cleavage of p7 from its precursor E2-p7 had a similar deleterious effect on virus spread, though whether this occurred at the same point in the virus life cycle is unknown. A role in assembly for HCV p7 is also suggested by our finding that p7 was able to replace the influenza A virus M2 protein in maintaining the pH-sensitive, receptor-binding conformation of the viral hemagglutinin during transport to the cell surface (20). The apparent localization of p7 to mitochondrial membranes in our study, however, seemed counterintuitive given its ability to replace M2 and was contrary to the findings of other investigators that p7 localized to the endoplasmic reticulum (ER) of transfected cells (11). This paradox is important to resolve, as localization could give vital clues to the function of p7.
Here, we have combined indirect immunofluorescence with subcellular fractionation to clarify the intracellular localization of HCV p7. We have examined the effect of the upstream signal peptide from E2 on the distribution of p7 using a novel rabbit polyclonal antibody to the C terminus of the protein in combination with epitope-tagged proteins. We show that native or tagged p7 is able to target the ER independently of its signal peptide, presumably in a posttranslational manner. For cells expressing native or epitope-tagged p7, we have identified a population of brightly labeled cells that contain two separate pools of p7 differing in the accessibility of their N or C termini to antibody: N-terminally stained p7 localizing in ER membranes associated with mitochondria and C-terminally stained molecules residing in the normal ER. Fractionation of cellular homogenates confirms that p7 localizes both to the normal ER and also to a heavy membrane fraction associated with mitochondria.
MATERIALS AND METHODS
Plasmid constructs. pCDNAp7 and pCDNAFLAGp7 have been described previously (20). PCR amplimers for pCDNASP-p7, pCDNASPm-p7, pCDNAMYCp7, and pCDNAp7FLAG were generated using Vent DNA polymerase (New England BioLabs, Inc.) with the J4 infectious molecular clone pCVJ46LS as the template (45). Specific primers generated amplimers corresponding to positions 2526 to 2768 for SP constructs (containing signal peptide) or positions 2580 to 2768 for p7 open reading frame constructs, and epitope tags were incorporated at either terminus by modifying the appropriate primer. All amplimers for pCDNA constructs were digested with EcoRI and NotI and ligated into an appropriately digested vector, pCDNA3.1 (Invitrogen). The SPm construct mutations A743N A745R were generated using a long forward primer containing the mutation in its sequence. Primers containing these mutations were also used to generate pCDNAE1-SPmp7 by overlap PCR of two amplimers: the first from the E1 signal peptide (the same as that used to generate the E1-p7 baculovirus [see below]) to the mutated N terminus of p7 and the second from the mutated N terminus of p7 to the same p7 reverse primer used above. All constructs were verified by double-stranded DNA sequencing. Primer sequences are available on request.
Recombinant baculoviruses. Baculoviruses expressing the HCV structural proteins were generated using the Bac-2-Bac baculovirus expression system (Invitrogen) incorporating a modified transfer vector containing a constitutive mammalian promoter sequence (CAG promoter comprising the cytomegalovirus immediate-early enhancer controlling the chicken ?-actin promoter from pBacMam-2 [Novagen]), pFBM. HCV sequences from core to p7 or E1 to p7 (incorporating the E1 signal peptide) were amplified using specific primers (sequences available on request) and cloned into pFBM digested with EcoRI. Viruses were amplified in and virus titers were determined on Spodoptera frugiperda Sf-9 cells and used at 107 PFU/ml for transduction of 293T cells seeded the previous day at 1 x 106 cells in a 10-cm dish.
Mammalian cell culture and transfection. Human embryonic kidney 293T cells and African green monkey kidney Vero cells were passaged in Dulbecco's modified Eagle medium (Invitrogen) supplemented with 10% fetal calf serum, 100 IU/ml penicillin, and 100 μg/ml streptomycin. Human hepatoblastoma HepG2 cells were passaged in minimal essential medium (Invitrogen) supplemented with 10% fetal calf serum, 100 IU/ml penicillin, 100 μg/ml streptomycin, 5 mM glutamine, and nonessential amino acids.
All transfections were carried out using Lipofectamine (Invitrogen) at 5 μl/μg DNA according to the manufacturer's instructions, overnight in Optimem (Invitrogen) serum-free medium. Cells for fluorescence analysis were mounted on poly-L-lysine-coated glass coverslips in 12-well plates at approximately 4 x 104 cells/coverslip and transfected with 0.5 μg DNA. Cells to be transfected for heavy/light membrane separation were seeded at approximately 2 x 105 cells/well of a six-well plate and transfected as described above with 1 μg DNA.
Antibodies. Antibody 1055 affinity-purified rabbit polyclonal antibody was raised against a peptide representing the C-terminal six residues of p7 from the J4 infectious clone of HCV genotype 1b, PPRAYA. The peptide CGGGPPRAYA was coupled to tuberculin purified protein derivative, and four rabbits were immunized and then boosted four times. Rabbit 1055 gave positive results in an enzyme-linked immunosorbent assay against the immunogen peptide and also recombinant histidine-tagged glutathione S-transferase-p7 prepared as described previously (19) using preimmune serum as an animal-specific negative control. The final working antibody was affinity purified with peptide to a concentration of 307 μg/ml. Antibody 1055 was detected by indirect fluorescence using Alexa Fluor 488-conjugated goat anti-rabbit or Alexa Fluor 594-conjugated chicken anti-rabbit secondary antibodies (Molecular Probes). FLAG-tagged proteins were detected by indirect fluorescence using a fluorescein isothiocyanate (FITC)-conjugated mouse anti-FLAG monoclonal antibody, M2 from Sigma. MYC-tagged protein was detected using mouse monoclonal antibody 9E10 (12) and a FITC-conjugated goat anti-mouse secondary antibody (Sigma). Mouse monoclonal antibody to HCV E2 protein, ALP98 was a gift from Arvind Patel (MRC Virology Unit, Glasgow, United Kingdom) and was detected by fluorescence using an Alexa Fluor 594-conjugated goat anti-mouse secondary antibody. The rabbit anti-NS5a antiserum was a gift from Ralf Bartenschlager and was detected using the same Alexa Fluor 594-conjugated chicken anti-rabbit secondary antibody described above. HCV core protein was detected using a mouse monoclonal antibody 215/07 from Biogenesis. A mouse monoclonal antibody to cytochrome oxidase (Oxphos IV) subunit 1 was obtained from Molecular Probes, and a rabbit polyclonal antiserum to human calreticulin was from Calbiochem. Detection of antibodies on Western blots was achieved using appropriate horseradish peroxidase-conjugated secondary antibodies (Sigma).
Indirect fluorescence detection of HCV p7. Cells to be labeled with Mitotracker CMXros (Molecular Probes) were incubated in a 200 nM solution of the dye in Dulbecco's modified Eagle medium for 1 h prior to fixation at 16 h posttransfection. Cells transfected on coverslips were washed three times in phosphate-buffered saline (PBS), fixed with 4% paraformaldehyde in PBS for 20 min at room temperature, and then washed twice more in PBS. Cells to be analyzed by immunocytochemistry were permeabilized with 0.1% Triton X-100 in PBS for 5 min. Coverslips were washed three times and then incubated with primary antibody diluted in PBS containing 10% fetal calf serum for 1 h at room temperature in a dark humidified container. Coverslips were washed three more times and incubated with the appropriate secondary antibody under the same conditions along with, where appropriate, Texas Red-conjugated concanavalin A (Molecular Probes) as a marker for the ER. Nuclei were labeled with Hoechst stain or TRO-PRO 3 (Molecular Probes) diluted 1/10,000 in PBS. The cells were then washed three times in PBS and once in distilled water prior to analysis. Images were captured using a DeltaVision restoration system (Applied Precision, Inc., Issaquah, WA) and an Olympus IX-70 inverted microscope. Optical sections of 0.2 microns were captured with a CoolSNAPHQ charge-coupled device camera (Roper Scientific, Tucson, AZ). Digital deconvolution and image analyses were then performed on three-dimensional data sets using 15 iterations of a constrained iterative deconvolution algorithm with SoftWoRx deconvolution software (Applied Precision, Inc.). Subsequent quantification of FLAG fluorescence from images of equal exposure times and neutral density filter settings was performed on individual color channels from deconvoluted sections using the Image J program; an average of seven individual images was taken.
Subcellular fractionation. 293T cells were washed three times in PBS, then scraped into 1 ml of ice-cold fractionation buffer (5 mM HEPES [pH 6.8], 1 mM EDTA, 250 mM sucrose, protease inhibitors), and homogenized on ice with 50 strokes of a loose-fitting Dounce homogenizer. Homogenates were then cleared of nuclei and unbroken cells by spinning at 1,000 x g in a microcentrifuge at 4°C for 5 min. For separation of crude heavy mitochondrial and light microsomal membranes, homogenates were spun in a microcentrifuge at 10,000 x g for 10 min at 4°C to obtain the heavy membrane pellet. This pellet was then washed twice by resuspension in 0.5 ml fractionation buffer and respinning. The supernatant from the first spin was respun at 10,000 x g for another 10 min to remove any residual mitochondria. This clarified supernatant was then spun in a Sorvall S100 AT3 rotor at 100,000 x g for 1 h to obtain a microsomal pellet. Both pellets were lysed in 50 μl of EBC lysis buffer (50 mM Tris-HCl [pH 8.0], 140 mM NaCl, 100 mM NaF, 200 μM Na3VO4, 0.1% sodium dodecyl sulfate [SDS], 0.5% NP-40), and protein concentrations were normalized by the bicinchoninic acid (BCA) protein assay (Bio-Rad) prior to SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting.
To prepare purified mitochondria, cleared homogenates from cells previously labeled with Mitotracker CMXros (see above) were pelleted first at 5,000 x g, then at 10,000 x g, and finally at 100,000 x g and washed as described above. Pellets were resuspended in fractionation buffer such that input samples for each stage could be taken (550 μl for the 5,000 x g pellet; other pellets in 100 μl). Five hundred microliters of the 5,000 x g pellet suspension was then layered over a 20 to 50% weight/volume continuous sucrose gradient cushion in 5 mM HEPES (pH 6.8) and spun at 52,000 x g in a Sorvall AH650 rotor for 45 min to pellet mitochondria and remove attached membranes. The pellet was then resuspended in fractionation buffer and layered over a 20 to 50% weight/weight continuous sucrose gradient in 5 mM HEPES (pH 6.8) and spun under the same conditions. The gradient was harvested using a cannula into 12 equal fractions. Fifty microliters of each sample was then analyzed by fluorimetry using a standard rhodamine filter set to detect the peak fluorescence of Mitotracker CMXros. Peak samples were then diluted to a sucrose concentration of 250 mM and spun at 20,000 x g for 30 min to pellet mitochondria in a microcentrifuge at 4°C. This pellet was then washed in 0.5 ml fractionation buffer and respun at 6,500 x g at 4°C for 15 min to give the final mitochondrial pellet. Pellets from all stages of purification were resuspended in 60 μl fractionation buffer, and protein concentrations were normalized with input samples by BCA prior to SDS-PAGE and Western blotting.
RESULTS
Characterization of a novel anti-p7 peptide antiserum, antibody 1055. We previously generated small quantities of a mouse antipeptide antiserum to the N terminus of p7 that we have since exhausted (19). We therefore attempted to generate new antibodies against either terminus of the protein and were successful in producing a rabbit polyclonal antibody, antibody 1055, against a peptide corresponding to the C-terminal six amino acids (PPRAYA) of p7 from the J4 infectious clone of HCV genotype 1b (45). To characterize this antibody for use in detecting p7 by immunofluorescence, human embryonic kidney 293T cells expressing green fluorescent protein (GFP), an HCV NS5A-GFP fusion protein, or p7 (in the context of E1-p7 in the example shown) were fixed, permeabilized, and stained with antibody 1055 as described in Materials and Methods (Fig. 1A). Antibody 1055 showed no cross-reactivity with cellular or expressed protein in all cell types tested except when p7 was present. When expressed with HCV E1 and E2, the p7 signal overlapped significantly with the staining for E2 and showed localization consistent with ER membrane association. p7 was also specifically detected when expressed alone in various forms (see below and see Fig. 2). In addition to 293T cells, p7 was readily detectable when HepG2 or Vero cells were used (data not shown), though the level of expression and efficiency of transfection were significantly reduced in these cells, requiring coinfection with recombinant fowlpox virus expressing T7 polymerase. For this reason, the results hereafter comprise images and blots from 293T cells, though the other cell lines showed similar phenotypes in all cases.
We also used antibody 1055 to detect p7 expression by immunoblotting (Fig. 1B). p7 was expressed in 293T cells both with and without its upstream signal peptide, with epitope tags at either terminus, or where the signal peptidase recognition site had been mutated (see Materials and Methods). Interestingly, the migration of p7 on gels did not appear to directly correspond with the expected size of the protein, and this was observed for both 10 to 20% morpholinepropanesulfonic acid (MOPS) gradient gels and 15% Tris-glycine gels. Detection with antibody 1055 showed that the presence of an N-terminal epitope tag in place of the signal peptide (Fig. 1B, lanes 3 and 5) caused p7 to migrate significantly more slowly than unmodified p7 (lane 2) on both types of gel. This was most notable for FLAG-tagged p7 (FLAG-p7). In contrast, the presence of a C-terminal FLAG tag did not alter migration (lane 4). Furthermore, prolonged exposures of p7-FLAG lysates on Tris-glycine gels (Fig. 1B, bottom two blots, lane 3) revealed that the protein in fact migrates as two bands; the larger band is more reactive to antibody 1055, and the smaller band is more detectable with anti-FLAG. The bands could not be due to degradation of the protein, as they were both recognized by two antibodies, the epitopes for which were present at opposite termini of the protein. This would also argue against the expressed p7 being cleaved intracellularly at the cytosolic loop. The fact that the C-terminally tagged p7-FLAG did not appear to migrate differently than native p7 suggested that this doublet could be a feature unique to modifying the p7 N terminus; this feature could potentially be due to a change in the overall charge of the protein upon incorporation of a highly basic tag or perhaps due to a posttranslational modification or binding of a cellular factor. Two forms could not be discerned for MYC-tagged p7 (MYC-p7), as this protein migrated closer to the size of native p7. p7 expressed with its upstream signal peptide, SP-p7, appeared to migrate as the native protein did, although in light of the migration of p7-FLAG, this does not necessarily indicate the efficiency of signal peptide cleavage. p7 with a mutated signal peptidase recognition site, SPm-p7, did, however migrate as a species that was visibly larger than the native protein on Tris-glycine gels, yet due to the fact that its migration was akin to that of native p7 on MOPS gels, the effect of the mutation on signal peptidase activity was assessed by alternate means (Fig. 2C).
Staining p7 with an antibody directed to its C terminus shows ER localization independent of the signal peptide. As a result of inefficient signal peptidase-mediated cleavage, p7 has been shown to exist in cells both as the native protein and as an E2-p7 precursor (26, 30). The membranes of the endoplasmic reticulum are the initial site of production of these proteins, and others have detected p7 in an ER-like compartment in HepG2 cells (11). Conversely, we demonstrated that tagged p7 lacking the signal peptide from E2 localized to a mitochondrion-like compartment in transfected 293T cells, suggesting that p7 might be actively targeted to this cellular organelle (20). We reasoned that the lack of the E2-derived signal peptide might result in aberrant targeting of p7, as the newly translated protein would not be cotranslationally inserted into the ER membrane. To investigate this targeting in more detail, we therefore generated p7 expression constructs containing either the wild-type signal peptide from E2 upstream of p7 (SP-p7) or constructs where the signal peptidase recognition site had been altered to prevent cleavage (SPm-p7) by introducing mutations at the –3, –1 loci relative to the junction of the two proteins, A743N and A745R. This double mutation is based on that previously shown to abrogate E2-p7 cleavage and viral infectivity in bovine viral diarrhea virus (21). 293T cells were fixed in 4% paraformaldehyde 16 h posttransfection, permeabilized, and stained using antibody 1055 and a marker for the ER, Texas Red-conjugated concanavalin A (Molecular Probes) prior to analysis on a digital deconvolution microscope (DeltaVision). Sixteen hours was chosen as the time point, as cell viability was higher at this time than at the 20 h used in our previous study.
The ER distribution of p7, as detected by antibody 1055, was not affected by the presence of the signal peptide or an epitope tag and overlapped with the concanavalin A signal (Fig. 2A) but not with Mitotracker CMXros (data not shown) in all cases. The appearance did not vary between brightly labeled or dimly labeled cells. Thus, native or tagged p7 appears capable of targeting to the ER posttranslationally in the absence of the signal peptide from E2, and the presence of epitope tags does not cause erroneous targeting. The only difference in distribution of C-terminally stained protein occurred where the signal peptide cleavage site had been mutated, SPm-p7 (Fig. 2B). In this case, only dimly labeled cells showed a normal ER-type distribution, whereas brightly labeled cells showed accumulation of cytoplasmic aggregates of p7, which appear to comprise, at least in part, modified ER membranes as judged by concanavalin A staining (Fig. 2B). To confirm that this mutation blocked signal peptidase cleavage in cells, we introduced it into a construct expressing E1 through to p7, pCDNAE1-SPm-p7. The staining for p7 (antibody 1055) strongly overlapped with that for E2 (ALP98) in transfected cells expressing this construct. Immunoblot analysis confirmed the lack of signal peptide cleavage: a band migrating more slowly than E2 was detected by both ALP98 and antibody 1055. However, in the wild-type sequence, the majority of the p7 was cleaved from E2 (Fig. 2C).
Staining the N or C terminus of p7 defines separate intracellular pools. We had previously shown that the use of an anti-FLAG antibody to detect an N-terminally FLAG-tagged p7, p7-FLAG, revealed a significant overlap with a mitochondrial marker (20). The localization pattern of p7-FLAG obtained with antibody 1055, however, was not consistent with this pattern of fluorescence. This raised the possibility that two separate pools of p7 existed within the cell and that these two pools differ in both subcellular localization and accessibility of the N and C termini to antibody.
We therefore reexamined the staining pattern for this construct in 293T cells with the anti-FLAG antibody at the earlier time point to investigate any temporal effects on p7 localization. At 16 h posttransfection we still obtained a population of brightly labeled cells showing a ring-like staining pattern consistent with localization to mitochondria or adjacent membranes that did indeed overlap with Mitotracker CMXros (Fig. 3A and C). We noted, however, that a population of dimly labeled cells was also present at this time point showing an ER localization for p7-FLAG (Fig. 3A). Close inspection of the mitochondrial pattern revealed the staining to localize predominantly to the outer surfaces of these Mitotracker CMXros-stained organelles, resulting in a ring-like distribution (Fig. 3C, top panels). To control for an effect attributable to the presence of the FLAG tag, we also analyzed a construct with an N-terminal MYC tag, MYC-p7, which displayed the same phenotype as the FLAG-tagged construct (data not shown). Quantification of the FLAG fluorescence for the cell populations exhibiting each staining pattern confirmed that the cells showing a mitochondrial distribution of p7-FLAG stained with a directly conjugated anti-FLAG monoclonal antibody conjugated to FITC were, indeed, up to twice the intensity (measured as mean fluorescence intensity per pixel [see Materials and Methods]) as those exhibiting an ER pattern, suggesting that the localization was directly related to the level of expression of the protein (Fig. 3B). This analysis is more revealing than visual inspection of individual fluorescence images, which are automatically adjusted to optimize signal/noise for better contrast by digital image capture systems. For this reason, images showing both phenotypes are often overexposed for the brighter mitochondrial distribution or show the dimmer fluorescence poorly (Fig. 3A).
To reconcile this difference in labeling patterns, cells expressing FLAG- or MYC-tagged constructs (data not shown) were dual labeled with both antibody 1055 and the antibody to the epitope tag (Fig. 4). In both cases, brightly labeled cells showed little or no overlap between the ring-like N-terminal staining and the C-terminal ER pattern. At lower levels, however, there was an almost complete overlap (Fig. 4A). To control for the possibility that the epitope tags were cleaving or targeting the protein nonspecifically to intracellular membranes, a construct in which the FLAG tag was located at the C terminus, p7-FLAG, was examined. This construct displayed an ER-type distribution at all levels of labeling with either antibody; dual-labeled cells showed a complete overlap (Fig. 4B). In all cases, staining with antibody 1055 showed an ER distribution irrespective of the fluorescence intensity. Thus, the apparent difference in localization patterns was likely due to differences in the accessibility of the p7 termini to antibody when p7 was present in different intracellular compartments. This masking effect could not be removed by permeabilization in 0.1% SDS (data not shown).
To investigate whether this phenomenon occurred when p7 was expressed with its signal peptide, constructs with an N-terminal epitope tag immediately downstream of the signal peptide or with five amino acids inserted into the p7 sequence (to preserve the native cleavage site) were generated, as well as versions containing the K778A R780A mutations to reduce possible toxicity. Unfortunately, upon transfection into 293T cells, these constructs gave similar phenotypes to that of SPm-p7 with no evidence of ring-like staining. Western blot analysis did show these constructs migrating as doublets, though we could not confirm this was due to inefficient signal peptide cleavage or to the N-terminal FLAG tag (data not shown). To address this issue, we are currently trying again to generate an antibody to the p7 N terminus.
p7 is enriched in heavy membrane fractions and cofractionates with both normal ER and membranes associated with mitochondria. Antibody 1055 also allowed us to investigate p7 localization by Western blotting of fractionated cellular homogenates. Initially, clarified homogenates from transfected 293T cells were subjected to a crude separation comparing a mitochondrially enriched heavy membrane fraction, pelleted at 10,000 x g, to light ER-derived microsomes pelleted at 100,000 x g (Fig. 5A). Samples were blotted in parallel for protein markers of the ER (calreticulin) and an inner mitochondrial membrane protein (cytochrome oxidase subunit 1). As evident from Fig. 5A, p7 was enriched in the heavy membrane fractions compared to microsomes whether or not it retained its upstream signal peptide (compare H and L lanes for p7 to H and L lanes for SP-p7). The presence of an epitope tag was also inconsequential, though again p7-FLAG appeared to migrate more slowly than p7-FLAG or native p7; the smaller p7-FLAG band was not visible on this exposure (Fig. 5A, H and L lanes for Fp7). The enrichment of the p7-FLAG (H and L lanes for p7-F) protein in the heavy membranes was more apparent on reduced exposures of the gel (data not shown). SPm-p7 again migrated more slowly than did p7 (though not at the same rate as p7-FLAG) in these experiments. SPm-p7 was also entirely present in the heavy membrane fraction, consistent with the formation of aggregates seen by fluorescence. This crude technique does not separate mitochondria from their associated ER-derived membranes; hence, both heavy and light membrane fractions contained an approximately equal level of calreticulin, whereas only the heavy fractions contained cytochrome oxidase.
Next we examined the effects of expressing the other structural proteins of HCV by transducing 293T cells with recombinant baculoviruses expressing either Core-E1-E2-p7 or E1-E2-p7 (Fig. 5B). Again, p7 and to a more variable extent, E2-p7 were found to be enriched in the heavy membrane fraction. E2 showed an approximately equal distribution between the fractions, yet core protein was also enriched in the heavy membrane fraction; consistent with the recent report that it too has the ability to localize to mitochondria (39).
To determine whether the apparent mitochondrial localization of p7 was due to its presence in the organelle itself or in associated ER-derived membranes, mitochondrially enriched fractions were subjected to further purification by pelleting through a sucrose cushion followed by separation on a continuous sucrose gradient, a protocol modified from the method of Schwer et al. (39). Mitochondria were tracked through the process via fluorimetric detection of Mitotracker CMXros with which the cells had been labeled prior to harvesting (data not shown). p7 in 293T cells transduced with a baculovirus expressing Core-E1-E2-p7 was not found in purified mitochondria, but it was predominant in the associated ER-derived membranes (Fig. 5C). This membrane fraction remained in suspension following the initial 5,000 x g spin that removes the majority of the mitochondria (Fig. 5C, lanes 2), but it subsequently pelleted at 10,000 x g (lanes 3). In addition, a significant proportion of p7 was found in microsomes pelleted at 100,000 x g, consistent with an ER-resident pool (lanes 4). Despite a low level of calreticulin indicating traces of ER membrane in the purified mitochondrial fractions (lanes 5 and 6), no p7 was detectable in these samples. The relative levels of cytochrome oxidase to calreticulin combined with the very high sensitivity of the calreticulin antibody can be taken as meaning that the overall level of ER membrane in these fractions was minute; lanes 2, 5, and 6 displayed a significant enrichment for mitochondrial proteins. E2-p7 was also present mainly in the 10,000 x g pellet, though the levels of this species were low overall in these experiments. This would seemingly argue against cleavage of the signal peptide directing transport of p7 to membranes around mitochondria; however, it is likely that the 10,000 x g pellet contains other membrane compartments as well as those responsible for the mitochondrial localization phenotype. In support of this argument, indirect fluorescence with ALP98 (anti-E2) in 293T or HepG2 cells transduced with the same baculovirus showed no mitochondrial labeling (Fig. 1B and data not shown). Interestingly, very little core protein was present in gradient-purified mitochondria (requiring a long exposure to be visualized; Fig. 5C, lanes 5 and 6) and what there was appeared to be p21 rather than fully processed p19, though the relevance of this observation is not clear. We conclude that p7 is therefore present in both normal ER and also in a subset of ER closely associated with the mitochondrial outer membrane, potentially ER cisternae wrapped around the organelle.
DISCUSSION
Our data are consistent with there being two distinct populations of p7 distinguished by the accessibility of either the N or C terminus. p7 detected using antibodies to its C terminus is exclusively ER localized, whereas p7 detected with an N-terminal tag can either show an ER or an apparent mitochondrial localization, depending on the level of labeling and presumably, therefore, of expression. This work provides evidence that the trafficking of HCV p7 is a complex process potentially regulated by both the cleavage from its upstream signal peptide and targeting signals present within the protein sequence.
It appears that many HCV genotypes show an inefficient cleavage of the E2-p7 precursor protein relative to the other structural proteins. It has recently been suggested that this may regulate the generation of native p7 and so the formation of ion channel complexes and may provide a means of including p7 in virions, a notion with which we are in complete agreement (10). Interestingly, signal peptide cleavage between E2 and p7 in our baculovirus system appeared for the most part to be reasonably efficient. This may be due to our using an HCV genotype 1b protein which has been shown to cleave more efficiently than HCV genotype 1a sequences (24). Furthermore, in contrast to the other HCV structural proteins, the signal peptidase responsible for the E2-p7 cleavage appears to be unique to mammalian cells and is absent from insect cell systems, suggesting that the virus may have evolved to regulate this cleavage event via this specific pathway to achieve specific E2/E2-p7 ratios.
The development of an antibody to the C terminus of p7 has allowed visualization of native protein for the first time. We found p7 to be present in the ER using this antibody regardless of the presence or absence of the upstream signal peptide. In this regard, p7 exhibits targeting similar to that exhibited by HCV NS2, which has been shown to target the ER in the absence of its upstream signal peptide, the C-terminal helix of p7 (44). Labeling the N terminus, however, revealed a change in localization from an ER distribution in dimly labeled cells to mitochondrially adjacent membranes in brightly labeled cells. Assuming that the level of labeling directly correlates with the level of protein present, epitope-tagged p7 appears to move from the ER to membranes around mitochondria as labeling/expression increases. Transport is not complete, however, as C-terminally labeled protein is still detectable in the ER and this signal does not overlap with that of the N-terminally labeled protein. This implies that the pools of p7 protein in these compartments differ in the extent to which their termini are accessible to antibody postfixation. Folding of the protein within an oligomeric structure or a possible interaction with a cellular factor may explain this observation, causing occlusion of the epitope. As this process appears to occur independently of the signal peptide from E2, its determinants must presumably reside within the p7 amino acid sequence. The C-terminal helix of p7 has been shown to be capable of acting as a signal peptide when fused to the HCV E1 protein, facilitating correct glycosylation of the protein in HepG2 cells (11). It is possible that the C-terminal helix of p7 may also direct in part the localization of the protein itself or be part of a multipartite signal. It is becoming increasingly apparent that single translation products can be targeted to distinct membrane compartments by competition between two internal signals, the outcome of which can be affected through a variety of mechanisms ranging from protein folding to the metabolic status of the cell (reviewed in reference 25). Interestingly, p7 has been proposed to adopt two transmembrane topologies; its C terminus can present to either side of the ER membrane (24). It is possible that this process plays a role in or may be a consequence of targeting of p7 to different membrane compartments.
Insertion of an epitope tag at the junction of the p7 N terminus and its signal peptide appeared to interfere with processing; giving the same phenotype as the signal peptide cleavage site mutation did. Unfortunately, this made it impossible to tell by indirect fluorescence whether this transport event occurs for native p7 after it has been targeted to the ER by its signal peptide. This is contrary to a recent report where insertion of a MYC tag near the p7 N terminus improved cleavage efficiency for HCV genotype 1a p7 (10). That report also identified structural elements that appear to regulate processing at the E2-p7 junction. It is possible that disruption of these signals has different effects in separate virus genotypes; the termini of p7 have been shown to have a genotype-specific role in chimeric HCV infectivity studies in chimpanzees in which the p7 protein of genotype 2a was introduced into genotype 1a background, as the only viable chimera was where the termini remained as genotype 1a sequence (38). In addition, the genotype 1a and 1b p7 signal peptides appear to cleave with different efficiencies as determined by their amino acid sequences (24). Our finding that mutation of the signal peptidase recognition site disrupted the localization of p7 is consistent with this proteolytic cleavage event playing an important role in regulating p7 function. Having a kinetically slow cleavage might allow for direction of p7 into virions by retaining a pool in the ER as E2-p7, while cleaved protein passes to mitochondrially associated membranes, potentially forming active ion channel complexes. E2p7-MYC has been successfully detected in VLPs made in insect cells by immunoelectron microscopy (24). It would be of interest to determine whether uncleaved E2-p7 displays ion channel activity.
E2-p7 produced in the context of the other HCV structural proteins via baculovirus transduction ought to retain authentic signal peptidase processing. In this case, as well as by transfection, p7 was found to be enriched in the heavy membrane fraction of 293T cell postnuclear homogenates which contain almost all the mitochondria as well as their associated ER-derived membranes. ER cisternae are known to wrap closely around mitochondria, facilitating rapid signaling and transport between the two organelles (42). In particular, the transmission of calcium ion fluxes from ER cisternae to mitochondria has been shown to be a pivotal process in the regulation of apoptotic signaling (reviewed in reference 17). It is tempting to speculate that the presence of p7 in these membranes may interfere with such signals, rendering the cell insensitive to proapoptotic stimuli from immune cells or the effects of other viral gene products. The 2B protein of coxsackievirus B is structurally quite similar to p7 and has recently been shown to have an antiapoptotic effect via its influence on intracellular calcium levels (9). Coincidentally, we found that p7 showed increased in vitro ion channel activity with a calcium electrolyte in planar lipid bilayers (19). Both 2B and p7 are purported to belong to the same family of virus ion channels, viroporins.
Our analysis indicates that, as well as microsomes, p7 most likely associates with adjacent ER cisternae rather than with mitochondria per se; gradient-purified mitochondria contained no detectable p7, and the majority of the p7 protein remained with the 10,000 x g fraction after it had been substantially depleted of mitochondria by a 5,000 x g spin. The 10,000 x g pellet is also likely to contain specialized areas of ER membrane that directly contact mitochondrial membranes known collectively as mitochondrion-associated membranes, or MAMs (36, 42). These are biochemically distinct from both normal ER and mitochondrial membranes, being enriched in enzymes concerned with fatty acid metabolism, such as phosphatidylserine synthase (41, 42). The HCV core protein has also recently been reported to localize to both mitochondria and MAMs (39) following processing to p19 by signal peptide peptidase in the ER, which also permits transport of the protein to lipid droplets (29). We also detected core in purified mitochondria, though the protein appeared to migrate as unprocessed p21 and it was present at low levels; its detection required overexposure of relevant Western blots (Fig. 5C, middle blot). Like p7, the majority of the p19 core protein appeared to reside in the cleared MAM fraction. It is possible, however, that some or all of the outer mitochondrial membrane had been removed during purification, resulting in the apparent absence of these proteins from the organelle, though the presence of a low level of calreticulin in these purified fractions would argue against this. Nevertheless, the association of both core and p7 with mitochondria and/or associated membranes as well as the localization of core protein to lipid droplets points to additional functions for these proteins in the HCV life cycle. HCV nonstructural proteins have also been observed to partially localize to ER cisternae around mitochondria in replicon cells (31), and it is notable that ultrastructural changes in mitochondria have been observed in chronic HCV patients (1).
As well as acting in membranes around mitochondria, the pool of p7 residing in the ER may also act in processes such as viral assembly. Pseudotyped retroviruses presenting HCV E1 and E2 on their surface are known to show pH-dependent cell entry (22), likely due to E2 adopting a fusogenic conformation prematurely. It is conceivable that p7 may protect E2 from such pH-induced changes during assembly/entry in the same way that both it and M2 can protect influenza A virus hemagglutinin (14, 20, 37).
It is possible to construct a model for the regulation of p7 localization by combining the role of the signal peptide, its transmembrane topology, and the presence of internal signal sequences (Fig. 6). Upon translation in the rough ER, a proportion of p7 remains with its C-terminal helix on the cytosolic side of the membrane. As levels of protein increase, a potential signal in the C terminus of the protein binds to a cellular factor that then directs this population of protein to membranes around mitochondria. Conversely, the remaining protein with both termini on the luminal side of the membrane is bound by a factor that causes ER retention. Signal peptide cleavage could theoretically occur in any of the ER-derived membranes, though it is perhaps delayed by binding of an ER retention factor such that a pool of E2-p7 remains in the ER for incorporation into virions. If signal peptide cleavage occurs in the mitochondrial ER cisternae, E2 might be channelled back to the rough ER, whereas p7 remains and adopts its dual-spanning topology. Cleaved p7 in both sets of membranes would then be free to oligomerize and form ion channels in either membrane. In addition, p7 incorporated into virions as E2-p7 may also be processed during exocytosis to allow the formation of channels that protect E2 from fusogenic change and/or function during virus entry. The recently available HCV replication systems based on the HCV genotype 2a JFH-1 strain that produces infectious virus particles would be an ideal system in which to determine any potential role for p7 in virus exit/entry. Unfortunately, however, antibody 1055 will not be employable as a tool in this regard due to differences in the C termini of p7 proteins of genotype 1b and 2a virus isolates. We are, however, pursuing this line of investigation with a view to developing new anti-p7 antibodies specific to other HCV genotypes.
As more insights into p7 function and behavior in cells are gained, it is clear that targeting p7 in future antiviral therapies could potentially act by blocking HCV at multiple points in its life cycle, perhaps using compounds based on amantadine derivatives. Further experiments on ion channel function, ideally in systems where HCV virions can be produced, will be required to define the precise function of p7 in HCV replication.
ACKNOWLEDGMENTS
We thank Helen Bright and Tony Carroll (Glaxo-Smith-Kline, Stevenage, United Kingdom) for help in developing antibody 1055. We also thank Andrew Street (University of Leeds) for the NS5A-GFP construct, as well as Matthew Bentham and Gareth Howell (University of Leeds) for useful discussions. Fluorescence microscopy was undertaken in the Wellcome Trust Bio-imaging facility in the Faculty of Biological Sciences of the University of Leeds.
This work was supported by grants from the Wellcome Trust (067125 and 074023). Dean Clarke was supported by a Medical Research Council Ph.D. studentship.
REFERENCES
Barbaro, G., G. Di Lorenzo, A. Asti, M. Ribersani, G. Belloni, B. Grisorio, G. Filice, and G. Barbarini. 1999. Hepatocellular mitochondrial alterations in patients with chronic hepatitis C: ultrastructural and biochemical findings. Am. J. Gastroenterol. 94:2198-2205.
Bartosch, B., J. Bukh, J. C. Meunier, C. Granier, R. E. Engle, W. C. Blackwelder, S. U. Emerson, F. L. Cosset, and R. H. Purcell. 2003. In vitro assay for neutralizing antibody to hepatitis C virus: evidence for broadly conserved neutralization epitopes. Proc. Natl. Acad. Sci. USA 100:14199-14204.
Bartosch, B., A. Vitelli, C. Granier, C. Goujon, J. Dubuisson, S. Pascale, E. Scarselli, R. Cortese, A. Nicosia, and F. L. Cosset. 2003. Cell entry of hepatitis C virus requires a set of co-receptors that include the CD81 tetraspanin and the SR-B1 scavenger receptor. J. Biol. Chem. 278:41624-41630.
Baumert, T. F., S. Ito, D. T. Wong, and T. J. Liang. 1998. Hepatitis C virus structural proteins assemble into viruslike particles in insect cells. J. Virol. 72:3827-3836.
Blanchard, E., D. Brand, S. Trassard, A. Goudeau, and P. Roingeard. 2002. Hepatitis C virus-like particle morphogenesis. J. Virol. 76:4073-4079.
Blanchard, E., C. Hourioux, D. Brand, M. Ait-Goughoulte, A. Moreau, S. Trassard, P. Y. Sizaret, F. Dubois, and P. Roingeard. 2003. Hepatitis C virus-like particle budding: role of the core protein and importance of its Asp111. J. Virol. 77:10131-10138.
Blight, K. J., A. A. Kolykhalov, and C. M. Rice. 2000. Efficient initiation of HCV RNA replication in cell culture. Science 290:1972-1974.
Blight, K. J., J. A. McKeating, J. Marcotrigiano, and C. M. Rice. 2003. Efficient replication of hepatitis C virus genotype 1a RNAs in cell culture. J. Virol. 77:3181-3190.
Campanella, M., A. S. de Jong, K. W. Lanke, W. J. Melchers, P. H. Willems, P. Pinton, R. Rizzuto, and F. J. van Kuppeveld. 2004. The coxsackievirus 2B protein suppresses apoptotic host cell responses by manipulating intracellular Ca2+ homeostasis. J. Biol. Chem. 279:18440-18450.
Carrere-Kremer, S., C. Montpellier, L. Lorenzo, B. Brulin, L. Cocquerel, S. Belouzard, F. Penin, and J. Dubuisson. 2004. Regulation of hepatitis C virus polyprotein processing by signal peptidase involves structural determinants at the p7 sequence junctions. J. Biol. Chem. 279:41384-41392.
Carrere-Kremer, S., C. Montpellier-Pala, L. Cocquerel, C. Wychowski, F. Penin, and J. Dubuisson. 2002. Subcellular localization and topology of the p7 polypeptide of hepatitis C virus. J. Virol. 76:3720-3730.
Chan, S., H. Gabra, F. Hill, G. Evan, and K. Sikora. 1987. A novel tumour marker related to the c-myc oncogene product. Mol. Cell Probes 1:73-82.
Choo, Q. L., G. Kuo, A. J. Weiner, L. R. Overby, D. W. Bradley, and M. Houghton. 1989. Isolation of a cDNA clone derived from a blood-borne non-A, non-B viral hepatitis genome. Science 244:359-362.
Ciampor, F., P. M. Bayley, M. V. Nermut, E. M. Hirst, R. J. Sugrue, and A. J. Hay. 1992. Evidence that the amantadine-induced, M2-mediated conversion of influenza A virus hemagglutinin to the low pH conformation occurs in an acidic trans Golgi compartment. Virology 188:14-24.
Clayton, R. F., A. Owsianka, J. Aitken, S. Graham, D. Bhella, and A. H. Patel. 2002. Analysis of antigenicity and topology of E2 glycoprotein present on recombinant hepatitis C virus-like particles. J. Virol. 76:7672-7682.
Deltenre, P., J. Henrion, V. Canva, S. Dharancy, F. Texier, A. Louvet, S. De Maeght, J. C. Paris, and P. Mathurin. 2004. Evaluation of amantadine in chronic hepatitis C: a meta-analysis. J. Hepatol. 41:462-473.
Demaurex, N., and C. Distelhorst. 2003. Cell biology. Apoptosis-the calcium connection. Science 300:65-67.
Elbers, K., N. Tautz, P. Becher, D. Stoll, T. Rumenapf, and H. J. Thiel. 1996. Processing in the pestivirus E2-NS2 region: identification of proteins p7 and E2p7. J. Virol. 70:4131-4135.
Griffin, S. D., L. P. Beales, D. S. Clarke, O. Worsfold, S. D. Evans, J. Jaeger, M. P. Harris, and D. J. Rowlands. 2003. The p7 protein of hepatitis C virus forms an ion channel that is blocked by the antiviral drug, Amantadine. FEBS Lett. 535:34-38.
Griffin, S. D., R. Harvey, D. S. Clarke, W. S. Barclay, M. Harris, and D. J. Rowlands. 2004. A conserved basic loop in hepatitis C virus p7 protein is required for amantadine-sensitive ion channel activity in mammalian cells but is dispensable for localization to mitochondria. J. Gen. Virol. 85:451-461.
Harada, T., N. Tautz, and H. J. Thiel. 2000. E2-p7 region of the bovine viral diarrhea virus polyprotein: processing and functional studies. J. Virol. 74:9498-9506.
Hsu, M., J. Zhang, M. Flint, C. Logvinoff, C. Cheng-Mayer, C. M. Rice, and J. A. McKeating. 2003. Hepatitis C virus glycoproteins mediate pH-dependent cell entry of pseudotyped retroviral particles. Proc. Natl. Acad. Sci. USA 100:7271-7276.
Ikeda, M., M. Yi, K. Li, and S. M. Lemon. 2002. Selectable subgenomic and genome-length dicistronic RNAs derived from an infectious molecular clone of the HCV-N strain of hepatitis C virus replicate efficiently in cultured Huh7 cells. J. Virol. 76:2997-3006.
Isherwood, B. J., and A. H. Patel. 2005. Analysis of the processing and transmembrane topology of the E2p7 protein of hepatitis C virus. J. Gen. Virol. 86:667-676.
Karniely, S., and O. Pines. 2005. Single translation-dual destination: mechanisms of dual protein targeting in eukaryotes. EMBO Rep. 6:420-425.
Lin, C., B. D. Lindenbach, B. M. Pragai, D. W. McCourt, and C. M. Rice. 1994. Processing in the hepatitis C virus E2-NS2 region: identification of p7 and two distinct E2-specific products with different C termini. J. Virol. 68:5063-5073.
Lindenbach, B. D., M. J. Evans, A. J. Syder, B. Wolk, T. L. Tellinghuisen, C. C. Liu, T. Maruyama, R. O. Hynes, D. R. Burton, J. A. McKeating, and C. M. Rice. 2005. Complete replication of hepatitis C virus in cell culture. Science 309:623-626.
Lohmann, V., F. Korner, J. Koch, U. Herian, L. Theilmann, and R. Bartenschlager. 1999. Replication of subgenomic hepatitis C virus RNAs in a hepatoma cell line. Science 285:110-113.
McLauchlan, J., M. K. Lemberg, G. Hope, and B. Martoglio. 2002. Intramembrane proteolysis promotes trafficking of hepatitis C virus core protein to lipid droplets. EMBO J. 21:3980-3988.
Mizushima, H., M. Hijikata, S. Asabe, M. Hirota, K. Kimura, and K. Shimotohno. 1994. Two hepatitis C virus glycoprotein E2 products with different C termini. J. Virol. 68:6215-6222.
Mottola, G., G. Cardinali, A. Ceccacci, C. Trozzi, L. Bartholomew, M. R. Torrisi, E. Pedrazzini, S. Bonatti, and G. Migliaccio. 2002. Hepatitis C virus nonstructural proteins are localized in a modified endoplasmic reticulum of cells expressing viral subgenomic replicons. Virology 293:31-43.
Pavlovic, D., D. C. Neville, O. Argaud, B. Blumberg, R. A. Dwek, W. B. Fischer, and N. Zitzmann. 2003. The hepatitis C virus p7 protein forms an ion channel that is inhibited by long-alkyl-chain iminosugar derivatives. Proc. Natl. Acad. Sci. USA 100:6104-6108.
Pietschmann, T., V. Lohmann, A. Kaul, N. Krieger, G. Rinck, G. Rutter, D. Strand, and R. Bartenschlager. 2002. Persistent and transient replication of full-length hepatitis C virus genomes in cell culture. J. Virol. 76:4008-4021.
Premkumar, A., L. Wilson, G. D. Ewart, and P. W. Gage. 2004. Cation-selective ion channels formed by p7 of hepatitis C virus are blocked by hexamethylene amiloride. FEBS Lett. 557:99-103.
Robertson, B., G. Myers, C. Howard, T. Brettin, J. Bukh, B. Gaschen, T. Gojobori, G. Maertens, M. Mizokami, O. Nainan, S. Netesov, K. Nishioka, I. T. Shin, P. Simmonds, D. Smith, L. Stuyver, and A. Weiner. 1998. Classification, nomenclature, and database development for hepatitis C virus (HCV) and related viruses: proposals for standardization. Arch. Virol. 143:2493-2503.
Rusinol, A. E., Z. Cui, M. H. Chen, and J. E. Vance. 1994. A unique mitochondria-associated membrane fraction from rat liver has a high capacity for lipid synthesis and contains pre-Golgi secretory proteins including nascent lipoproteins. J. Biol. Chem. 269:27494-27502.
Sakaguchi, T., G. P. Leser, and R. A. Lamb. 1996. The ion channel activity of the influenza virus M2 protein affects transport through the Golgi apparatus. J. Cell Biol. 133:733-747.
Sakai, A., M. S. Claire, K. Faulk, S. Govindarajan, S. U. Emerson, R. H. Purcell, and J. Bukh. 2003. The p7 polypeptide of hepatitis C virus is critical for infectivity and contains functionally important genotype-specific sequences. Proc. Natl. Acad. Sci. USA 100:11646-11651.
Schwer, B., S. Ren, T. Pietschmann, J. Kartenbeck, K. Kaehlcke, R. Bartenschlager, T. S. Yen, and M. Ott. 2004. Targeting of hepatitis C virus core protein to mitochondria through a novel C-terminal localization motif. J. Virol. 78:7958-7968.
Simmonds, P. 2004. Genetic diversity and evolution of hepatitis C virus-15 years on. J. Gen. Virol. 85:3173-3188.
Stone, S. J., and J. E. Vance. 2000. Phosphatidylserine synthase-1 and -2 are localized to mitochondria-associated membranes. J. Biol. Chem. 275:34534-34540.
Vance, J. E. 1990. Phospholipid synthesis in a membrane fraction associated with mitochondria. J. Biol. Chem. 265:7248-7256.
Wakita, T., T. Pietschmann, T. Kato, T. Date, M. Miyamoto, Z. Zhao, K. Murthy, A. Habermann, H. G. Krausslich, M. Mizokami, R. Bartenschlager, and T. J. Liang. 2005. Production of infectious hepatitis C virus in tissue culture from a cloned viral genome. Nat. Med. 11:791-796.
Yamaga, A. K., and J. H. Ou. 2002. Membrane topology of the hepatitis C virus NS2 protein. J. Biol. Chem. 277:33228-33234.
Yanagi, M., M. St. Claire, M. Shapiro, S. U. Emerson, R. H. Purcell, and J. Bukh. 1998. Transcripts of a chimeric cDNA clone of hepatitis C virus genotype 1b are infectious in vivo. Virology 244:161-172.
Zhong, J., P. Gastaminza, G. Cheng, S. Kapadia, T. Kato, D. R. Burton, S. F. Wieland, S. L. Uprichard, T. Wakita, and F. V. Chisari. 2005. Robust hepatitis C virus infection in vitro. Proc. Natl. Acad. Sci. USA 102:9294-9299.(Stephen Griffin, Dean Cla)
ABSTRACT
The hepatitis C virus (HCV) p7 protein forms an amantadine-sensitive ion channel required for viral replication in chimpanzees, though its precise role in the life cycle of HCV is unknown. In an attempt to gain some insights into p7 function, we examined the intracellular localization of p7 using epitope tags and an anti-p7 peptide antibody, antibody 1055. Immunofluorescence labeling of p7 at its C terminus revealed an endoplasmic reticulum (ER) localization independent of the presence of its signal peptide, whereas labeling the N terminus gave a mitochondrial-type distribution in brightly labeled cells. Both of these patterns could be visualized within individual cells, suggestive of separate pools of p7 where the N and C termini differed in accessibility to antibody. These patterns were disrupted by preventing signal peptide cleavage. Subcellular fractionation revealed that p7 was enriched in a heavy membrane fraction associated with mitochondria as well as normal ER-derived microsomes. The complex regulation of the intracellular distribution of p7 suggests that p7 plays multiple roles in the HCV life cycle either intracellularly or as a virion component.
INTRODUCTION
Hepatitis C virus (HCV) is a major cause of chronic liver disease worldwide and is now the most common reason for liver transplants in Western countries. Unusually for an RNA virus, the majority of patients become persistently infected after a mild acute episode, and clinical intervention occurs in late-stage symptomatic patients. Current therapy comprises high-dose pegylated alpha interferon (IFN) combined with the guanosine analogue ribavirin. The efficacy of this regimen is largely dependent on the viral genotype; the most prevalent genotype 1 viruses possess high levels of innate resistance to IFN, and reservoirs of resistance in other genotypes are building due to the highly variable nature of HCV (40). Interestingly, a recent meta-analysis of clinical trials in which patients were treated with a triple combination of IFN, ribavirin, and amantadine showed that this approach gave improved sustained viral responses in patients that previously did not respond to dual therapy, most often those infected with genotype 1 HCV (16).
HCV has a single-stranded positive-sense RNA of around 9.6 kb and is the prototype member of the Hepacivirus genus of the Flaviviridae family (13, 35). While, until very recently, no robust in vitro replication system has existed for HCV (27, 43, 46), many functions of the viral nonstructural proteins have been elucidated using replicons (7, 8, 23, 28, 33). The inability of these systems to produce extracellular virus has limited studies on structural proteins to virus-like particles (VLPs) made either within insect cells (4, 15) or by mammalian expression (5, 6), or more recently, to the use of pseudotyped retrovirus systems to investigate receptor tropism and cell entry (2, 3, 22).
The p7 protein of HCV is not required for RNA replication or the formation of VLPs in insect cells, and it is uncertain whether it is a virion component. p7 is a small hydrophobic protein of 63 amino acids located within the HCV genome at the junction between the structural and nonstructural proteins (26, 30). We previously showed that p7 from genotype 1b HCV forms an oligomeric ion channel in planar lipid bilayers that can be blocked by amantadine at micromolar concentrations, leading to our proposal that the potential antiviral effect of amantadine described above may be due to its action on p7 (19). Others have subsequently confirmed this ion channel activity for different HCV genotypes and have identified other channel-blocking compounds (32, 34). The important finding that p7 is required for replication of HCV in chimpanzees confirms the protein as a target for antiviral chemotherapy, yet its role in viral replication is unknown (38). The homologous p7 protein from bovine viral diarrhea virus is known to be necessary for the generation of infectious virus particles, though whether virions are able to assemble or are secreted in an immature form is not known, and attempts to detect the protein in purified particles were unsuccessful (18, 21). Furthermore, preventing the already inefficient cleavage of p7 from its precursor E2-p7 had a similar deleterious effect on virus spread, though whether this occurred at the same point in the virus life cycle is unknown. A role in assembly for HCV p7 is also suggested by our finding that p7 was able to replace the influenza A virus M2 protein in maintaining the pH-sensitive, receptor-binding conformation of the viral hemagglutinin during transport to the cell surface (20). The apparent localization of p7 to mitochondrial membranes in our study, however, seemed counterintuitive given its ability to replace M2 and was contrary to the findings of other investigators that p7 localized to the endoplasmic reticulum (ER) of transfected cells (11). This paradox is important to resolve, as localization could give vital clues to the function of p7.
Here, we have combined indirect immunofluorescence with subcellular fractionation to clarify the intracellular localization of HCV p7. We have examined the effect of the upstream signal peptide from E2 on the distribution of p7 using a novel rabbit polyclonal antibody to the C terminus of the protein in combination with epitope-tagged proteins. We show that native or tagged p7 is able to target the ER independently of its signal peptide, presumably in a posttranslational manner. For cells expressing native or epitope-tagged p7, we have identified a population of brightly labeled cells that contain two separate pools of p7 differing in the accessibility of their N or C termini to antibody: N-terminally stained p7 localizing in ER membranes associated with mitochondria and C-terminally stained molecules residing in the normal ER. Fractionation of cellular homogenates confirms that p7 localizes both to the normal ER and also to a heavy membrane fraction associated with mitochondria.
MATERIALS AND METHODS
Plasmid constructs. pCDNAp7 and pCDNAFLAGp7 have been described previously (20). PCR amplimers for pCDNASP-p7, pCDNASPm-p7, pCDNAMYCp7, and pCDNAp7FLAG were generated using Vent DNA polymerase (New England BioLabs, Inc.) with the J4 infectious molecular clone pCVJ46LS as the template (45). Specific primers generated amplimers corresponding to positions 2526 to 2768 for SP constructs (containing signal peptide) or positions 2580 to 2768 for p7 open reading frame constructs, and epitope tags were incorporated at either terminus by modifying the appropriate primer. All amplimers for pCDNA constructs were digested with EcoRI and NotI and ligated into an appropriately digested vector, pCDNA3.1 (Invitrogen). The SPm construct mutations A743N A745R were generated using a long forward primer containing the mutation in its sequence. Primers containing these mutations were also used to generate pCDNAE1-SPmp7 by overlap PCR of two amplimers: the first from the E1 signal peptide (the same as that used to generate the E1-p7 baculovirus [see below]) to the mutated N terminus of p7 and the second from the mutated N terminus of p7 to the same p7 reverse primer used above. All constructs were verified by double-stranded DNA sequencing. Primer sequences are available on request.
Recombinant baculoviruses. Baculoviruses expressing the HCV structural proteins were generated using the Bac-2-Bac baculovirus expression system (Invitrogen) incorporating a modified transfer vector containing a constitutive mammalian promoter sequence (CAG promoter comprising the cytomegalovirus immediate-early enhancer controlling the chicken ?-actin promoter from pBacMam-2 [Novagen]), pFBM. HCV sequences from core to p7 or E1 to p7 (incorporating the E1 signal peptide) were amplified using specific primers (sequences available on request) and cloned into pFBM digested with EcoRI. Viruses were amplified in and virus titers were determined on Spodoptera frugiperda Sf-9 cells and used at 107 PFU/ml for transduction of 293T cells seeded the previous day at 1 x 106 cells in a 10-cm dish.
Mammalian cell culture and transfection. Human embryonic kidney 293T cells and African green monkey kidney Vero cells were passaged in Dulbecco's modified Eagle medium (Invitrogen) supplemented with 10% fetal calf serum, 100 IU/ml penicillin, and 100 μg/ml streptomycin. Human hepatoblastoma HepG2 cells were passaged in minimal essential medium (Invitrogen) supplemented with 10% fetal calf serum, 100 IU/ml penicillin, 100 μg/ml streptomycin, 5 mM glutamine, and nonessential amino acids.
All transfections were carried out using Lipofectamine (Invitrogen) at 5 μl/μg DNA according to the manufacturer's instructions, overnight in Optimem (Invitrogen) serum-free medium. Cells for fluorescence analysis were mounted on poly-L-lysine-coated glass coverslips in 12-well plates at approximately 4 x 104 cells/coverslip and transfected with 0.5 μg DNA. Cells to be transfected for heavy/light membrane separation were seeded at approximately 2 x 105 cells/well of a six-well plate and transfected as described above with 1 μg DNA.
Antibodies. Antibody 1055 affinity-purified rabbit polyclonal antibody was raised against a peptide representing the C-terminal six residues of p7 from the J4 infectious clone of HCV genotype 1b, PPRAYA. The peptide CGGGPPRAYA was coupled to tuberculin purified protein derivative, and four rabbits were immunized and then boosted four times. Rabbit 1055 gave positive results in an enzyme-linked immunosorbent assay against the immunogen peptide and also recombinant histidine-tagged glutathione S-transferase-p7 prepared as described previously (19) using preimmune serum as an animal-specific negative control. The final working antibody was affinity purified with peptide to a concentration of 307 μg/ml. Antibody 1055 was detected by indirect fluorescence using Alexa Fluor 488-conjugated goat anti-rabbit or Alexa Fluor 594-conjugated chicken anti-rabbit secondary antibodies (Molecular Probes). FLAG-tagged proteins were detected by indirect fluorescence using a fluorescein isothiocyanate (FITC)-conjugated mouse anti-FLAG monoclonal antibody, M2 from Sigma. MYC-tagged protein was detected using mouse monoclonal antibody 9E10 (12) and a FITC-conjugated goat anti-mouse secondary antibody (Sigma). Mouse monoclonal antibody to HCV E2 protein, ALP98 was a gift from Arvind Patel (MRC Virology Unit, Glasgow, United Kingdom) and was detected by fluorescence using an Alexa Fluor 594-conjugated goat anti-mouse secondary antibody. The rabbit anti-NS5a antiserum was a gift from Ralf Bartenschlager and was detected using the same Alexa Fluor 594-conjugated chicken anti-rabbit secondary antibody described above. HCV core protein was detected using a mouse monoclonal antibody 215/07 from Biogenesis. A mouse monoclonal antibody to cytochrome oxidase (Oxphos IV) subunit 1 was obtained from Molecular Probes, and a rabbit polyclonal antiserum to human calreticulin was from Calbiochem. Detection of antibodies on Western blots was achieved using appropriate horseradish peroxidase-conjugated secondary antibodies (Sigma).
Indirect fluorescence detection of HCV p7. Cells to be labeled with Mitotracker CMXros (Molecular Probes) were incubated in a 200 nM solution of the dye in Dulbecco's modified Eagle medium for 1 h prior to fixation at 16 h posttransfection. Cells transfected on coverslips were washed three times in phosphate-buffered saline (PBS), fixed with 4% paraformaldehyde in PBS for 20 min at room temperature, and then washed twice more in PBS. Cells to be analyzed by immunocytochemistry were permeabilized with 0.1% Triton X-100 in PBS for 5 min. Coverslips were washed three times and then incubated with primary antibody diluted in PBS containing 10% fetal calf serum for 1 h at room temperature in a dark humidified container. Coverslips were washed three more times and incubated with the appropriate secondary antibody under the same conditions along with, where appropriate, Texas Red-conjugated concanavalin A (Molecular Probes) as a marker for the ER. Nuclei were labeled with Hoechst stain or TRO-PRO 3 (Molecular Probes) diluted 1/10,000 in PBS. The cells were then washed three times in PBS and once in distilled water prior to analysis. Images were captured using a DeltaVision restoration system (Applied Precision, Inc., Issaquah, WA) and an Olympus IX-70 inverted microscope. Optical sections of 0.2 microns were captured with a CoolSNAPHQ charge-coupled device camera (Roper Scientific, Tucson, AZ). Digital deconvolution and image analyses were then performed on three-dimensional data sets using 15 iterations of a constrained iterative deconvolution algorithm with SoftWoRx deconvolution software (Applied Precision, Inc.). Subsequent quantification of FLAG fluorescence from images of equal exposure times and neutral density filter settings was performed on individual color channels from deconvoluted sections using the Image J program; an average of seven individual images was taken.
Subcellular fractionation. 293T cells were washed three times in PBS, then scraped into 1 ml of ice-cold fractionation buffer (5 mM HEPES [pH 6.8], 1 mM EDTA, 250 mM sucrose, protease inhibitors), and homogenized on ice with 50 strokes of a loose-fitting Dounce homogenizer. Homogenates were then cleared of nuclei and unbroken cells by spinning at 1,000 x g in a microcentrifuge at 4°C for 5 min. For separation of crude heavy mitochondrial and light microsomal membranes, homogenates were spun in a microcentrifuge at 10,000 x g for 10 min at 4°C to obtain the heavy membrane pellet. This pellet was then washed twice by resuspension in 0.5 ml fractionation buffer and respinning. The supernatant from the first spin was respun at 10,000 x g for another 10 min to remove any residual mitochondria. This clarified supernatant was then spun in a Sorvall S100 AT3 rotor at 100,000 x g for 1 h to obtain a microsomal pellet. Both pellets were lysed in 50 μl of EBC lysis buffer (50 mM Tris-HCl [pH 8.0], 140 mM NaCl, 100 mM NaF, 200 μM Na3VO4, 0.1% sodium dodecyl sulfate [SDS], 0.5% NP-40), and protein concentrations were normalized by the bicinchoninic acid (BCA) protein assay (Bio-Rad) prior to SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting.
To prepare purified mitochondria, cleared homogenates from cells previously labeled with Mitotracker CMXros (see above) were pelleted first at 5,000 x g, then at 10,000 x g, and finally at 100,000 x g and washed as described above. Pellets were resuspended in fractionation buffer such that input samples for each stage could be taken (550 μl for the 5,000 x g pellet; other pellets in 100 μl). Five hundred microliters of the 5,000 x g pellet suspension was then layered over a 20 to 50% weight/volume continuous sucrose gradient cushion in 5 mM HEPES (pH 6.8) and spun at 52,000 x g in a Sorvall AH650 rotor for 45 min to pellet mitochondria and remove attached membranes. The pellet was then resuspended in fractionation buffer and layered over a 20 to 50% weight/weight continuous sucrose gradient in 5 mM HEPES (pH 6.8) and spun under the same conditions. The gradient was harvested using a cannula into 12 equal fractions. Fifty microliters of each sample was then analyzed by fluorimetry using a standard rhodamine filter set to detect the peak fluorescence of Mitotracker CMXros. Peak samples were then diluted to a sucrose concentration of 250 mM and spun at 20,000 x g for 30 min to pellet mitochondria in a microcentrifuge at 4°C. This pellet was then washed in 0.5 ml fractionation buffer and respun at 6,500 x g at 4°C for 15 min to give the final mitochondrial pellet. Pellets from all stages of purification were resuspended in 60 μl fractionation buffer, and protein concentrations were normalized with input samples by BCA prior to SDS-PAGE and Western blotting.
RESULTS
Characterization of a novel anti-p7 peptide antiserum, antibody 1055. We previously generated small quantities of a mouse antipeptide antiserum to the N terminus of p7 that we have since exhausted (19). We therefore attempted to generate new antibodies against either terminus of the protein and were successful in producing a rabbit polyclonal antibody, antibody 1055, against a peptide corresponding to the C-terminal six amino acids (PPRAYA) of p7 from the J4 infectious clone of HCV genotype 1b (45). To characterize this antibody for use in detecting p7 by immunofluorescence, human embryonic kidney 293T cells expressing green fluorescent protein (GFP), an HCV NS5A-GFP fusion protein, or p7 (in the context of E1-p7 in the example shown) were fixed, permeabilized, and stained with antibody 1055 as described in Materials and Methods (Fig. 1A). Antibody 1055 showed no cross-reactivity with cellular or expressed protein in all cell types tested except when p7 was present. When expressed with HCV E1 and E2, the p7 signal overlapped significantly with the staining for E2 and showed localization consistent with ER membrane association. p7 was also specifically detected when expressed alone in various forms (see below and see Fig. 2). In addition to 293T cells, p7 was readily detectable when HepG2 or Vero cells were used (data not shown), though the level of expression and efficiency of transfection were significantly reduced in these cells, requiring coinfection with recombinant fowlpox virus expressing T7 polymerase. For this reason, the results hereafter comprise images and blots from 293T cells, though the other cell lines showed similar phenotypes in all cases.
We also used antibody 1055 to detect p7 expression by immunoblotting (Fig. 1B). p7 was expressed in 293T cells both with and without its upstream signal peptide, with epitope tags at either terminus, or where the signal peptidase recognition site had been mutated (see Materials and Methods). Interestingly, the migration of p7 on gels did not appear to directly correspond with the expected size of the protein, and this was observed for both 10 to 20% morpholinepropanesulfonic acid (MOPS) gradient gels and 15% Tris-glycine gels. Detection with antibody 1055 showed that the presence of an N-terminal epitope tag in place of the signal peptide (Fig. 1B, lanes 3 and 5) caused p7 to migrate significantly more slowly than unmodified p7 (lane 2) on both types of gel. This was most notable for FLAG-tagged p7 (FLAG-p7). In contrast, the presence of a C-terminal FLAG tag did not alter migration (lane 4). Furthermore, prolonged exposures of p7-FLAG lysates on Tris-glycine gels (Fig. 1B, bottom two blots, lane 3) revealed that the protein in fact migrates as two bands; the larger band is more reactive to antibody 1055, and the smaller band is more detectable with anti-FLAG. The bands could not be due to degradation of the protein, as they were both recognized by two antibodies, the epitopes for which were present at opposite termini of the protein. This would also argue against the expressed p7 being cleaved intracellularly at the cytosolic loop. The fact that the C-terminally tagged p7-FLAG did not appear to migrate differently than native p7 suggested that this doublet could be a feature unique to modifying the p7 N terminus; this feature could potentially be due to a change in the overall charge of the protein upon incorporation of a highly basic tag or perhaps due to a posttranslational modification or binding of a cellular factor. Two forms could not be discerned for MYC-tagged p7 (MYC-p7), as this protein migrated closer to the size of native p7. p7 expressed with its upstream signal peptide, SP-p7, appeared to migrate as the native protein did, although in light of the migration of p7-FLAG, this does not necessarily indicate the efficiency of signal peptide cleavage. p7 with a mutated signal peptidase recognition site, SPm-p7, did, however migrate as a species that was visibly larger than the native protein on Tris-glycine gels, yet due to the fact that its migration was akin to that of native p7 on MOPS gels, the effect of the mutation on signal peptidase activity was assessed by alternate means (Fig. 2C).
Staining p7 with an antibody directed to its C terminus shows ER localization independent of the signal peptide. As a result of inefficient signal peptidase-mediated cleavage, p7 has been shown to exist in cells both as the native protein and as an E2-p7 precursor (26, 30). The membranes of the endoplasmic reticulum are the initial site of production of these proteins, and others have detected p7 in an ER-like compartment in HepG2 cells (11). Conversely, we demonstrated that tagged p7 lacking the signal peptide from E2 localized to a mitochondrion-like compartment in transfected 293T cells, suggesting that p7 might be actively targeted to this cellular organelle (20). We reasoned that the lack of the E2-derived signal peptide might result in aberrant targeting of p7, as the newly translated protein would not be cotranslationally inserted into the ER membrane. To investigate this targeting in more detail, we therefore generated p7 expression constructs containing either the wild-type signal peptide from E2 upstream of p7 (SP-p7) or constructs where the signal peptidase recognition site had been altered to prevent cleavage (SPm-p7) by introducing mutations at the –3, –1 loci relative to the junction of the two proteins, A743N and A745R. This double mutation is based on that previously shown to abrogate E2-p7 cleavage and viral infectivity in bovine viral diarrhea virus (21). 293T cells were fixed in 4% paraformaldehyde 16 h posttransfection, permeabilized, and stained using antibody 1055 and a marker for the ER, Texas Red-conjugated concanavalin A (Molecular Probes) prior to analysis on a digital deconvolution microscope (DeltaVision). Sixteen hours was chosen as the time point, as cell viability was higher at this time than at the 20 h used in our previous study.
The ER distribution of p7, as detected by antibody 1055, was not affected by the presence of the signal peptide or an epitope tag and overlapped with the concanavalin A signal (Fig. 2A) but not with Mitotracker CMXros (data not shown) in all cases. The appearance did not vary between brightly labeled or dimly labeled cells. Thus, native or tagged p7 appears capable of targeting to the ER posttranslationally in the absence of the signal peptide from E2, and the presence of epitope tags does not cause erroneous targeting. The only difference in distribution of C-terminally stained protein occurred where the signal peptide cleavage site had been mutated, SPm-p7 (Fig. 2B). In this case, only dimly labeled cells showed a normal ER-type distribution, whereas brightly labeled cells showed accumulation of cytoplasmic aggregates of p7, which appear to comprise, at least in part, modified ER membranes as judged by concanavalin A staining (Fig. 2B). To confirm that this mutation blocked signal peptidase cleavage in cells, we introduced it into a construct expressing E1 through to p7, pCDNAE1-SPm-p7. The staining for p7 (antibody 1055) strongly overlapped with that for E2 (ALP98) in transfected cells expressing this construct. Immunoblot analysis confirmed the lack of signal peptide cleavage: a band migrating more slowly than E2 was detected by both ALP98 and antibody 1055. However, in the wild-type sequence, the majority of the p7 was cleaved from E2 (Fig. 2C).
Staining the N or C terminus of p7 defines separate intracellular pools. We had previously shown that the use of an anti-FLAG antibody to detect an N-terminally FLAG-tagged p7, p7-FLAG, revealed a significant overlap with a mitochondrial marker (20). The localization pattern of p7-FLAG obtained with antibody 1055, however, was not consistent with this pattern of fluorescence. This raised the possibility that two separate pools of p7 existed within the cell and that these two pools differ in both subcellular localization and accessibility of the N and C termini to antibody.
We therefore reexamined the staining pattern for this construct in 293T cells with the anti-FLAG antibody at the earlier time point to investigate any temporal effects on p7 localization. At 16 h posttransfection we still obtained a population of brightly labeled cells showing a ring-like staining pattern consistent with localization to mitochondria or adjacent membranes that did indeed overlap with Mitotracker CMXros (Fig. 3A and C). We noted, however, that a population of dimly labeled cells was also present at this time point showing an ER localization for p7-FLAG (Fig. 3A). Close inspection of the mitochondrial pattern revealed the staining to localize predominantly to the outer surfaces of these Mitotracker CMXros-stained organelles, resulting in a ring-like distribution (Fig. 3C, top panels). To control for an effect attributable to the presence of the FLAG tag, we also analyzed a construct with an N-terminal MYC tag, MYC-p7, which displayed the same phenotype as the FLAG-tagged construct (data not shown). Quantification of the FLAG fluorescence for the cell populations exhibiting each staining pattern confirmed that the cells showing a mitochondrial distribution of p7-FLAG stained with a directly conjugated anti-FLAG monoclonal antibody conjugated to FITC were, indeed, up to twice the intensity (measured as mean fluorescence intensity per pixel [see Materials and Methods]) as those exhibiting an ER pattern, suggesting that the localization was directly related to the level of expression of the protein (Fig. 3B). This analysis is more revealing than visual inspection of individual fluorescence images, which are automatically adjusted to optimize signal/noise for better contrast by digital image capture systems. For this reason, images showing both phenotypes are often overexposed for the brighter mitochondrial distribution or show the dimmer fluorescence poorly (Fig. 3A).
To reconcile this difference in labeling patterns, cells expressing FLAG- or MYC-tagged constructs (data not shown) were dual labeled with both antibody 1055 and the antibody to the epitope tag (Fig. 4). In both cases, brightly labeled cells showed little or no overlap between the ring-like N-terminal staining and the C-terminal ER pattern. At lower levels, however, there was an almost complete overlap (Fig. 4A). To control for the possibility that the epitope tags were cleaving or targeting the protein nonspecifically to intracellular membranes, a construct in which the FLAG tag was located at the C terminus, p7-FLAG, was examined. This construct displayed an ER-type distribution at all levels of labeling with either antibody; dual-labeled cells showed a complete overlap (Fig. 4B). In all cases, staining with antibody 1055 showed an ER distribution irrespective of the fluorescence intensity. Thus, the apparent difference in localization patterns was likely due to differences in the accessibility of the p7 termini to antibody when p7 was present in different intracellular compartments. This masking effect could not be removed by permeabilization in 0.1% SDS (data not shown).
To investigate whether this phenomenon occurred when p7 was expressed with its signal peptide, constructs with an N-terminal epitope tag immediately downstream of the signal peptide or with five amino acids inserted into the p7 sequence (to preserve the native cleavage site) were generated, as well as versions containing the K778A R780A mutations to reduce possible toxicity. Unfortunately, upon transfection into 293T cells, these constructs gave similar phenotypes to that of SPm-p7 with no evidence of ring-like staining. Western blot analysis did show these constructs migrating as doublets, though we could not confirm this was due to inefficient signal peptide cleavage or to the N-terminal FLAG tag (data not shown). To address this issue, we are currently trying again to generate an antibody to the p7 N terminus.
p7 is enriched in heavy membrane fractions and cofractionates with both normal ER and membranes associated with mitochondria. Antibody 1055 also allowed us to investigate p7 localization by Western blotting of fractionated cellular homogenates. Initially, clarified homogenates from transfected 293T cells were subjected to a crude separation comparing a mitochondrially enriched heavy membrane fraction, pelleted at 10,000 x g, to light ER-derived microsomes pelleted at 100,000 x g (Fig. 5A). Samples were blotted in parallel for protein markers of the ER (calreticulin) and an inner mitochondrial membrane protein (cytochrome oxidase subunit 1). As evident from Fig. 5A, p7 was enriched in the heavy membrane fractions compared to microsomes whether or not it retained its upstream signal peptide (compare H and L lanes for p7 to H and L lanes for SP-p7). The presence of an epitope tag was also inconsequential, though again p7-FLAG appeared to migrate more slowly than p7-FLAG or native p7; the smaller p7-FLAG band was not visible on this exposure (Fig. 5A, H and L lanes for Fp7). The enrichment of the p7-FLAG (H and L lanes for p7-F) protein in the heavy membranes was more apparent on reduced exposures of the gel (data not shown). SPm-p7 again migrated more slowly than did p7 (though not at the same rate as p7-FLAG) in these experiments. SPm-p7 was also entirely present in the heavy membrane fraction, consistent with the formation of aggregates seen by fluorescence. This crude technique does not separate mitochondria from their associated ER-derived membranes; hence, both heavy and light membrane fractions contained an approximately equal level of calreticulin, whereas only the heavy fractions contained cytochrome oxidase.
Next we examined the effects of expressing the other structural proteins of HCV by transducing 293T cells with recombinant baculoviruses expressing either Core-E1-E2-p7 or E1-E2-p7 (Fig. 5B). Again, p7 and to a more variable extent, E2-p7 were found to be enriched in the heavy membrane fraction. E2 showed an approximately equal distribution between the fractions, yet core protein was also enriched in the heavy membrane fraction; consistent with the recent report that it too has the ability to localize to mitochondria (39).
To determine whether the apparent mitochondrial localization of p7 was due to its presence in the organelle itself or in associated ER-derived membranes, mitochondrially enriched fractions were subjected to further purification by pelleting through a sucrose cushion followed by separation on a continuous sucrose gradient, a protocol modified from the method of Schwer et al. (39). Mitochondria were tracked through the process via fluorimetric detection of Mitotracker CMXros with which the cells had been labeled prior to harvesting (data not shown). p7 in 293T cells transduced with a baculovirus expressing Core-E1-E2-p7 was not found in purified mitochondria, but it was predominant in the associated ER-derived membranes (Fig. 5C). This membrane fraction remained in suspension following the initial 5,000 x g spin that removes the majority of the mitochondria (Fig. 5C, lanes 2), but it subsequently pelleted at 10,000 x g (lanes 3). In addition, a significant proportion of p7 was found in microsomes pelleted at 100,000 x g, consistent with an ER-resident pool (lanes 4). Despite a low level of calreticulin indicating traces of ER membrane in the purified mitochondrial fractions (lanes 5 and 6), no p7 was detectable in these samples. The relative levels of cytochrome oxidase to calreticulin combined with the very high sensitivity of the calreticulin antibody can be taken as meaning that the overall level of ER membrane in these fractions was minute; lanes 2, 5, and 6 displayed a significant enrichment for mitochondrial proteins. E2-p7 was also present mainly in the 10,000 x g pellet, though the levels of this species were low overall in these experiments. This would seemingly argue against cleavage of the signal peptide directing transport of p7 to membranes around mitochondria; however, it is likely that the 10,000 x g pellet contains other membrane compartments as well as those responsible for the mitochondrial localization phenotype. In support of this argument, indirect fluorescence with ALP98 (anti-E2) in 293T or HepG2 cells transduced with the same baculovirus showed no mitochondrial labeling (Fig. 1B and data not shown). Interestingly, very little core protein was present in gradient-purified mitochondria (requiring a long exposure to be visualized; Fig. 5C, lanes 5 and 6) and what there was appeared to be p21 rather than fully processed p19, though the relevance of this observation is not clear. We conclude that p7 is therefore present in both normal ER and also in a subset of ER closely associated with the mitochondrial outer membrane, potentially ER cisternae wrapped around the organelle.
DISCUSSION
Our data are consistent with there being two distinct populations of p7 distinguished by the accessibility of either the N or C terminus. p7 detected using antibodies to its C terminus is exclusively ER localized, whereas p7 detected with an N-terminal tag can either show an ER or an apparent mitochondrial localization, depending on the level of labeling and presumably, therefore, of expression. This work provides evidence that the trafficking of HCV p7 is a complex process potentially regulated by both the cleavage from its upstream signal peptide and targeting signals present within the protein sequence.
It appears that many HCV genotypes show an inefficient cleavage of the E2-p7 precursor protein relative to the other structural proteins. It has recently been suggested that this may regulate the generation of native p7 and so the formation of ion channel complexes and may provide a means of including p7 in virions, a notion with which we are in complete agreement (10). Interestingly, signal peptide cleavage between E2 and p7 in our baculovirus system appeared for the most part to be reasonably efficient. This may be due to our using an HCV genotype 1b protein which has been shown to cleave more efficiently than HCV genotype 1a sequences (24). Furthermore, in contrast to the other HCV structural proteins, the signal peptidase responsible for the E2-p7 cleavage appears to be unique to mammalian cells and is absent from insect cell systems, suggesting that the virus may have evolved to regulate this cleavage event via this specific pathway to achieve specific E2/E2-p7 ratios.
The development of an antibody to the C terminus of p7 has allowed visualization of native protein for the first time. We found p7 to be present in the ER using this antibody regardless of the presence or absence of the upstream signal peptide. In this regard, p7 exhibits targeting similar to that exhibited by HCV NS2, which has been shown to target the ER in the absence of its upstream signal peptide, the C-terminal helix of p7 (44). Labeling the N terminus, however, revealed a change in localization from an ER distribution in dimly labeled cells to mitochondrially adjacent membranes in brightly labeled cells. Assuming that the level of labeling directly correlates with the level of protein present, epitope-tagged p7 appears to move from the ER to membranes around mitochondria as labeling/expression increases. Transport is not complete, however, as C-terminally labeled protein is still detectable in the ER and this signal does not overlap with that of the N-terminally labeled protein. This implies that the pools of p7 protein in these compartments differ in the extent to which their termini are accessible to antibody postfixation. Folding of the protein within an oligomeric structure or a possible interaction with a cellular factor may explain this observation, causing occlusion of the epitope. As this process appears to occur independently of the signal peptide from E2, its determinants must presumably reside within the p7 amino acid sequence. The C-terminal helix of p7 has been shown to be capable of acting as a signal peptide when fused to the HCV E1 protein, facilitating correct glycosylation of the protein in HepG2 cells (11). It is possible that the C-terminal helix of p7 may also direct in part the localization of the protein itself or be part of a multipartite signal. It is becoming increasingly apparent that single translation products can be targeted to distinct membrane compartments by competition between two internal signals, the outcome of which can be affected through a variety of mechanisms ranging from protein folding to the metabolic status of the cell (reviewed in reference 25). Interestingly, p7 has been proposed to adopt two transmembrane topologies; its C terminus can present to either side of the ER membrane (24). It is possible that this process plays a role in or may be a consequence of targeting of p7 to different membrane compartments.
Insertion of an epitope tag at the junction of the p7 N terminus and its signal peptide appeared to interfere with processing; giving the same phenotype as the signal peptide cleavage site mutation did. Unfortunately, this made it impossible to tell by indirect fluorescence whether this transport event occurs for native p7 after it has been targeted to the ER by its signal peptide. This is contrary to a recent report where insertion of a MYC tag near the p7 N terminus improved cleavage efficiency for HCV genotype 1a p7 (10). That report also identified structural elements that appear to regulate processing at the E2-p7 junction. It is possible that disruption of these signals has different effects in separate virus genotypes; the termini of p7 have been shown to have a genotype-specific role in chimeric HCV infectivity studies in chimpanzees in which the p7 protein of genotype 2a was introduced into genotype 1a background, as the only viable chimera was where the termini remained as genotype 1a sequence (38). In addition, the genotype 1a and 1b p7 signal peptides appear to cleave with different efficiencies as determined by their amino acid sequences (24). Our finding that mutation of the signal peptidase recognition site disrupted the localization of p7 is consistent with this proteolytic cleavage event playing an important role in regulating p7 function. Having a kinetically slow cleavage might allow for direction of p7 into virions by retaining a pool in the ER as E2-p7, while cleaved protein passes to mitochondrially associated membranes, potentially forming active ion channel complexes. E2p7-MYC has been successfully detected in VLPs made in insect cells by immunoelectron microscopy (24). It would be of interest to determine whether uncleaved E2-p7 displays ion channel activity.
E2-p7 produced in the context of the other HCV structural proteins via baculovirus transduction ought to retain authentic signal peptidase processing. In this case, as well as by transfection, p7 was found to be enriched in the heavy membrane fraction of 293T cell postnuclear homogenates which contain almost all the mitochondria as well as their associated ER-derived membranes. ER cisternae are known to wrap closely around mitochondria, facilitating rapid signaling and transport between the two organelles (42). In particular, the transmission of calcium ion fluxes from ER cisternae to mitochondria has been shown to be a pivotal process in the regulation of apoptotic signaling (reviewed in reference 17). It is tempting to speculate that the presence of p7 in these membranes may interfere with such signals, rendering the cell insensitive to proapoptotic stimuli from immune cells or the effects of other viral gene products. The 2B protein of coxsackievirus B is structurally quite similar to p7 and has recently been shown to have an antiapoptotic effect via its influence on intracellular calcium levels (9). Coincidentally, we found that p7 showed increased in vitro ion channel activity with a calcium electrolyte in planar lipid bilayers (19). Both 2B and p7 are purported to belong to the same family of virus ion channels, viroporins.
Our analysis indicates that, as well as microsomes, p7 most likely associates with adjacent ER cisternae rather than with mitochondria per se; gradient-purified mitochondria contained no detectable p7, and the majority of the p7 protein remained with the 10,000 x g fraction after it had been substantially depleted of mitochondria by a 5,000 x g spin. The 10,000 x g pellet is also likely to contain specialized areas of ER membrane that directly contact mitochondrial membranes known collectively as mitochondrion-associated membranes, or MAMs (36, 42). These are biochemically distinct from both normal ER and mitochondrial membranes, being enriched in enzymes concerned with fatty acid metabolism, such as phosphatidylserine synthase (41, 42). The HCV core protein has also recently been reported to localize to both mitochondria and MAMs (39) following processing to p19 by signal peptide peptidase in the ER, which also permits transport of the protein to lipid droplets (29). We also detected core in purified mitochondria, though the protein appeared to migrate as unprocessed p21 and it was present at low levels; its detection required overexposure of relevant Western blots (Fig. 5C, middle blot). Like p7, the majority of the p19 core protein appeared to reside in the cleared MAM fraction. It is possible, however, that some or all of the outer mitochondrial membrane had been removed during purification, resulting in the apparent absence of these proteins from the organelle, though the presence of a low level of calreticulin in these purified fractions would argue against this. Nevertheless, the association of both core and p7 with mitochondria and/or associated membranes as well as the localization of core protein to lipid droplets points to additional functions for these proteins in the HCV life cycle. HCV nonstructural proteins have also been observed to partially localize to ER cisternae around mitochondria in replicon cells (31), and it is notable that ultrastructural changes in mitochondria have been observed in chronic HCV patients (1).
As well as acting in membranes around mitochondria, the pool of p7 residing in the ER may also act in processes such as viral assembly. Pseudotyped retroviruses presenting HCV E1 and E2 on their surface are known to show pH-dependent cell entry (22), likely due to E2 adopting a fusogenic conformation prematurely. It is conceivable that p7 may protect E2 from such pH-induced changes during assembly/entry in the same way that both it and M2 can protect influenza A virus hemagglutinin (14, 20, 37).
It is possible to construct a model for the regulation of p7 localization by combining the role of the signal peptide, its transmembrane topology, and the presence of internal signal sequences (Fig. 6). Upon translation in the rough ER, a proportion of p7 remains with its C-terminal helix on the cytosolic side of the membrane. As levels of protein increase, a potential signal in the C terminus of the protein binds to a cellular factor that then directs this population of protein to membranes around mitochondria. Conversely, the remaining protein with both termini on the luminal side of the membrane is bound by a factor that causes ER retention. Signal peptide cleavage could theoretically occur in any of the ER-derived membranes, though it is perhaps delayed by binding of an ER retention factor such that a pool of E2-p7 remains in the ER for incorporation into virions. If signal peptide cleavage occurs in the mitochondrial ER cisternae, E2 might be channelled back to the rough ER, whereas p7 remains and adopts its dual-spanning topology. Cleaved p7 in both sets of membranes would then be free to oligomerize and form ion channels in either membrane. In addition, p7 incorporated into virions as E2-p7 may also be processed during exocytosis to allow the formation of channels that protect E2 from fusogenic change and/or function during virus entry. The recently available HCV replication systems based on the HCV genotype 2a JFH-1 strain that produces infectious virus particles would be an ideal system in which to determine any potential role for p7 in virus exit/entry. Unfortunately, however, antibody 1055 will not be employable as a tool in this regard due to differences in the C termini of p7 proteins of genotype 1b and 2a virus isolates. We are, however, pursuing this line of investigation with a view to developing new anti-p7 antibodies specific to other HCV genotypes.
As more insights into p7 function and behavior in cells are gained, it is clear that targeting p7 in future antiviral therapies could potentially act by blocking HCV at multiple points in its life cycle, perhaps using compounds based on amantadine derivatives. Further experiments on ion channel function, ideally in systems where HCV virions can be produced, will be required to define the precise function of p7 in HCV replication.
ACKNOWLEDGMENTS
We thank Helen Bright and Tony Carroll (Glaxo-Smith-Kline, Stevenage, United Kingdom) for help in developing antibody 1055. We also thank Andrew Street (University of Leeds) for the NS5A-GFP construct, as well as Matthew Bentham and Gareth Howell (University of Leeds) for useful discussions. Fluorescence microscopy was undertaken in the Wellcome Trust Bio-imaging facility in the Faculty of Biological Sciences of the University of Leeds.
This work was supported by grants from the Wellcome Trust (067125 and 074023). Dean Clarke was supported by a Medical Research Council Ph.D. studentship.
REFERENCES
Barbaro, G., G. Di Lorenzo, A. Asti, M. Ribersani, G. Belloni, B. Grisorio, G. Filice, and G. Barbarini. 1999. Hepatocellular mitochondrial alterations in patients with chronic hepatitis C: ultrastructural and biochemical findings. Am. J. Gastroenterol. 94:2198-2205.
Bartosch, B., J. Bukh, J. C. Meunier, C. Granier, R. E. Engle, W. C. Blackwelder, S. U. Emerson, F. L. Cosset, and R. H. Purcell. 2003. In vitro assay for neutralizing antibody to hepatitis C virus: evidence for broadly conserved neutralization epitopes. Proc. Natl. Acad. Sci. USA 100:14199-14204.
Bartosch, B., A. Vitelli, C. Granier, C. Goujon, J. Dubuisson, S. Pascale, E. Scarselli, R. Cortese, A. Nicosia, and F. L. Cosset. 2003. Cell entry of hepatitis C virus requires a set of co-receptors that include the CD81 tetraspanin and the SR-B1 scavenger receptor. J. Biol. Chem. 278:41624-41630.
Baumert, T. F., S. Ito, D. T. Wong, and T. J. Liang. 1998. Hepatitis C virus structural proteins assemble into viruslike particles in insect cells. J. Virol. 72:3827-3836.
Blanchard, E., D. Brand, S. Trassard, A. Goudeau, and P. Roingeard. 2002. Hepatitis C virus-like particle morphogenesis. J. Virol. 76:4073-4079.
Blanchard, E., C. Hourioux, D. Brand, M. Ait-Goughoulte, A. Moreau, S. Trassard, P. Y. Sizaret, F. Dubois, and P. Roingeard. 2003. Hepatitis C virus-like particle budding: role of the core protein and importance of its Asp111. J. Virol. 77:10131-10138.
Blight, K. J., A. A. Kolykhalov, and C. M. Rice. 2000. Efficient initiation of HCV RNA replication in cell culture. Science 290:1972-1974.
Blight, K. J., J. A. McKeating, J. Marcotrigiano, and C. M. Rice. 2003. Efficient replication of hepatitis C virus genotype 1a RNAs in cell culture. J. Virol. 77:3181-3190.
Campanella, M., A. S. de Jong, K. W. Lanke, W. J. Melchers, P. H. Willems, P. Pinton, R. Rizzuto, and F. J. van Kuppeveld. 2004. The coxsackievirus 2B protein suppresses apoptotic host cell responses by manipulating intracellular Ca2+ homeostasis. J. Biol. Chem. 279:18440-18450.
Carrere-Kremer, S., C. Montpellier, L. Lorenzo, B. Brulin, L. Cocquerel, S. Belouzard, F. Penin, and J. Dubuisson. 2004. Regulation of hepatitis C virus polyprotein processing by signal peptidase involves structural determinants at the p7 sequence junctions. J. Biol. Chem. 279:41384-41392.
Carrere-Kremer, S., C. Montpellier-Pala, L. Cocquerel, C. Wychowski, F. Penin, and J. Dubuisson. 2002. Subcellular localization and topology of the p7 polypeptide of hepatitis C virus. J. Virol. 76:3720-3730.
Chan, S., H. Gabra, F. Hill, G. Evan, and K. Sikora. 1987. A novel tumour marker related to the c-myc oncogene product. Mol. Cell Probes 1:73-82.
Choo, Q. L., G. Kuo, A. J. Weiner, L. R. Overby, D. W. Bradley, and M. Houghton. 1989. Isolation of a cDNA clone derived from a blood-borne non-A, non-B viral hepatitis genome. Science 244:359-362.
Ciampor, F., P. M. Bayley, M. V. Nermut, E. M. Hirst, R. J. Sugrue, and A. J. Hay. 1992. Evidence that the amantadine-induced, M2-mediated conversion of influenza A virus hemagglutinin to the low pH conformation occurs in an acidic trans Golgi compartment. Virology 188:14-24.
Clayton, R. F., A. Owsianka, J. Aitken, S. Graham, D. Bhella, and A. H. Patel. 2002. Analysis of antigenicity and topology of E2 glycoprotein present on recombinant hepatitis C virus-like particles. J. Virol. 76:7672-7682.
Deltenre, P., J. Henrion, V. Canva, S. Dharancy, F. Texier, A. Louvet, S. De Maeght, J. C. Paris, and P. Mathurin. 2004. Evaluation of amantadine in chronic hepatitis C: a meta-analysis. J. Hepatol. 41:462-473.
Demaurex, N., and C. Distelhorst. 2003. Cell biology. Apoptosis-the calcium connection. Science 300:65-67.
Elbers, K., N. Tautz, P. Becher, D. Stoll, T. Rumenapf, and H. J. Thiel. 1996. Processing in the pestivirus E2-NS2 region: identification of proteins p7 and E2p7. J. Virol. 70:4131-4135.
Griffin, S. D., L. P. Beales, D. S. Clarke, O. Worsfold, S. D. Evans, J. Jaeger, M. P. Harris, and D. J. Rowlands. 2003. The p7 protein of hepatitis C virus forms an ion channel that is blocked by the antiviral drug, Amantadine. FEBS Lett. 535:34-38.
Griffin, S. D., R. Harvey, D. S. Clarke, W. S. Barclay, M. Harris, and D. J. Rowlands. 2004. A conserved basic loop in hepatitis C virus p7 protein is required for amantadine-sensitive ion channel activity in mammalian cells but is dispensable for localization to mitochondria. J. Gen. Virol. 85:451-461.
Harada, T., N. Tautz, and H. J. Thiel. 2000. E2-p7 region of the bovine viral diarrhea virus polyprotein: processing and functional studies. J. Virol. 74:9498-9506.
Hsu, M., J. Zhang, M. Flint, C. Logvinoff, C. Cheng-Mayer, C. M. Rice, and J. A. McKeating. 2003. Hepatitis C virus glycoproteins mediate pH-dependent cell entry of pseudotyped retroviral particles. Proc. Natl. Acad. Sci. USA 100:7271-7276.
Ikeda, M., M. Yi, K. Li, and S. M. Lemon. 2002. Selectable subgenomic and genome-length dicistronic RNAs derived from an infectious molecular clone of the HCV-N strain of hepatitis C virus replicate efficiently in cultured Huh7 cells. J. Virol. 76:2997-3006.
Isherwood, B. J., and A. H. Patel. 2005. Analysis of the processing and transmembrane topology of the E2p7 protein of hepatitis C virus. J. Gen. Virol. 86:667-676.
Karniely, S., and O. Pines. 2005. Single translation-dual destination: mechanisms of dual protein targeting in eukaryotes. EMBO Rep. 6:420-425.
Lin, C., B. D. Lindenbach, B. M. Pragai, D. W. McCourt, and C. M. Rice. 1994. Processing in the hepatitis C virus E2-NS2 region: identification of p7 and two distinct E2-specific products with different C termini. J. Virol. 68:5063-5073.
Lindenbach, B. D., M. J. Evans, A. J. Syder, B. Wolk, T. L. Tellinghuisen, C. C. Liu, T. Maruyama, R. O. Hynes, D. R. Burton, J. A. McKeating, and C. M. Rice. 2005. Complete replication of hepatitis C virus in cell culture. Science 309:623-626.
Lohmann, V., F. Korner, J. Koch, U. Herian, L. Theilmann, and R. Bartenschlager. 1999. Replication of subgenomic hepatitis C virus RNAs in a hepatoma cell line. Science 285:110-113.
McLauchlan, J., M. K. Lemberg, G. Hope, and B. Martoglio. 2002. Intramembrane proteolysis promotes trafficking of hepatitis C virus core protein to lipid droplets. EMBO J. 21:3980-3988.
Mizushima, H., M. Hijikata, S. Asabe, M. Hirota, K. Kimura, and K. Shimotohno. 1994. Two hepatitis C virus glycoprotein E2 products with different C termini. J. Virol. 68:6215-6222.
Mottola, G., G. Cardinali, A. Ceccacci, C. Trozzi, L. Bartholomew, M. R. Torrisi, E. Pedrazzini, S. Bonatti, and G. Migliaccio. 2002. Hepatitis C virus nonstructural proteins are localized in a modified endoplasmic reticulum of cells expressing viral subgenomic replicons. Virology 293:31-43.
Pavlovic, D., D. C. Neville, O. Argaud, B. Blumberg, R. A. Dwek, W. B. Fischer, and N. Zitzmann. 2003. The hepatitis C virus p7 protein forms an ion channel that is inhibited by long-alkyl-chain iminosugar derivatives. Proc. Natl. Acad. Sci. USA 100:6104-6108.
Pietschmann, T., V. Lohmann, A. Kaul, N. Krieger, G. Rinck, G. Rutter, D. Strand, and R. Bartenschlager. 2002. Persistent and transient replication of full-length hepatitis C virus genomes in cell culture. J. Virol. 76:4008-4021.
Premkumar, A., L. Wilson, G. D. Ewart, and P. W. Gage. 2004. Cation-selective ion channels formed by p7 of hepatitis C virus are blocked by hexamethylene amiloride. FEBS Lett. 557:99-103.
Robertson, B., G. Myers, C. Howard, T. Brettin, J. Bukh, B. Gaschen, T. Gojobori, G. Maertens, M. Mizokami, O. Nainan, S. Netesov, K. Nishioka, I. T. Shin, P. Simmonds, D. Smith, L. Stuyver, and A. Weiner. 1998. Classification, nomenclature, and database development for hepatitis C virus (HCV) and related viruses: proposals for standardization. Arch. Virol. 143:2493-2503.
Rusinol, A. E., Z. Cui, M. H. Chen, and J. E. Vance. 1994. A unique mitochondria-associated membrane fraction from rat liver has a high capacity for lipid synthesis and contains pre-Golgi secretory proteins including nascent lipoproteins. J. Biol. Chem. 269:27494-27502.
Sakaguchi, T., G. P. Leser, and R. A. Lamb. 1996. The ion channel activity of the influenza virus M2 protein affects transport through the Golgi apparatus. J. Cell Biol. 133:733-747.
Sakai, A., M. S. Claire, K. Faulk, S. Govindarajan, S. U. Emerson, R. H. Purcell, and J. Bukh. 2003. The p7 polypeptide of hepatitis C virus is critical for infectivity and contains functionally important genotype-specific sequences. Proc. Natl. Acad. Sci. USA 100:11646-11651.
Schwer, B., S. Ren, T. Pietschmann, J. Kartenbeck, K. Kaehlcke, R. Bartenschlager, T. S. Yen, and M. Ott. 2004. Targeting of hepatitis C virus core protein to mitochondria through a novel C-terminal localization motif. J. Virol. 78:7958-7968.
Simmonds, P. 2004. Genetic diversity and evolution of hepatitis C virus-15 years on. J. Gen. Virol. 85:3173-3188.
Stone, S. J., and J. E. Vance. 2000. Phosphatidylserine synthase-1 and -2 are localized to mitochondria-associated membranes. J. Biol. Chem. 275:34534-34540.
Vance, J. E. 1990. Phospholipid synthesis in a membrane fraction associated with mitochondria. J. Biol. Chem. 265:7248-7256.
Wakita, T., T. Pietschmann, T. Kato, T. Date, M. Miyamoto, Z. Zhao, K. Murthy, A. Habermann, H. G. Krausslich, M. Mizokami, R. Bartenschlager, and T. J. Liang. 2005. Production of infectious hepatitis C virus in tissue culture from a cloned viral genome. Nat. Med. 11:791-796.
Yamaga, A. K., and J. H. Ou. 2002. Membrane topology of the hepatitis C virus NS2 protein. J. Biol. Chem. 277:33228-33234.
Yanagi, M., M. St. Claire, M. Shapiro, S. U. Emerson, R. H. Purcell, and J. Bukh. 1998. Transcripts of a chimeric cDNA clone of hepatitis C virus genotype 1b are infectious in vivo. Virology 244:161-172.
Zhong, J., P. Gastaminza, G. Cheng, S. Kapadia, T. Kato, D. R. Burton, S. F. Wieland, S. L. Uprichard, T. Wakita, and F. V. Chisari. 2005. Robust hepatitis C virus infection in vitro. Proc. Natl. Acad. Sci. USA 102:9294-9299.(Stephen Griffin, Dean Cla)