Hepatitis C virus (HCV), a member of the Flaviviridaefamily, has been known to be the major etiologic agent ofposttransfusion non-A, non-B hepatitis and associatedwith a high frequency of chronic persistent infectionand hepatocarcinogenesis (23). The viral genome, single-stranded, positive-sense RNA of approximately 9.6 kb,exhibits a considerable degree of sequence variation,based on which HCV is currently classified into at least6 clades (also called genotypes) and more than 60 sub-types (7, 29, 43). Geographic distribution and clinico-pathological features including interferon responsive-ness and carcinogenicity appear to vary with differentHCV clades and subtypes (2, 7, 29, 49) and even with

different strains of the same subtype (8, 33, 50). TheHCV genome encodes a polyprotein of about 3,000amino acids (aa), which is cleaved by the signal peptidaseof the host cell and two virally-encoded proteinases togenerate at least 10 viral proteins; core, envelope 1 (E1),E2, p7, nonstructural protein 2 (NS2), NS3, NS4A,NS4B, NS5A and NS5B (41).

HCV core protein, cleaved from the N-terminus of thepolyprotein, is considered to form the viral capsid (41).There are at least two forms of HCV core protein, onebeing the initial product of 191 aa and the other generatedby additional cleavage at its C-terminus between aa 174and 191, with the latter being likely a component of

Hepatitis C Virus Core Protein Selectively InhibitsSynthesis and Accumulation of p21/Waf1 and Certain Nuclear Proteins

Kiyomasa Oka�, Motoko Nagano-Fujii, Isao Yoshida, Rachmat Hidajat, Lin Deng, Masato Akutsu, and Hak Hotta*

Department of Microbiology, Kobe University Graduate School of Medicine, Kobe, Hyogo 650–0017, Japan

Received November 1, 2002; in revised form, January 17, 2003. Accepted March 3, 2003

Abstract: By using a vaccinia virus-T7 expression system, possible effects of hepatitis C virus (HCV) coreprotein on synthesis and accumulation of host cellular proteins transiently expressed in cultured cells wereanalyzed. Immunoblot and immunofluorescence analyses revealed that synthesis and accumulation of cer-tain nuclear proteins, such as p21/Waf1, p53, proliferating cell nuclear antigen and c-Fos, were stronglyinhibited by HCV core protein. On the other hand, synthesis and accumulation of cytoplasmic proteins, suchas 2'-5'-oligoadenylate synthetase (2'-5'-OAS), RNase L and MEK1, were barely affected by HCV core pro-tein. Northern blot analysis showed that the degrees of mRNA expression for those proteins did not differbetween HCV core protein-expressing cells and the control, suggesting that the inhibition occurred at thepost-transcription level. Pulse-labeling analysis suggested that HCV core protein strongly inhibited syn-thesis of p21/Waf1 at the translation level. Once being accumulated in the nucleus, p21/Waf1 stability wasnot significantly affected by HCV core protein. Mutants of HCV core protein C-terminally deleted by 18or 41 amino acids (aa), which were localized almost exclusively in the nucleus, lost their ability to inhibit syn-thesis/accumulation of p21/Waf1 whereas another mutant C-terminally deleted by 8 aa still maintained thesame properties (subcellular localization and the inhibitory effect) as the full-length HCV core protein of191 aa. Taken together, our present results suggest that expression of HCV core protein in the cytoplasmselectively inhibits synthesis of p21/Waf1 and some other nuclear proteins at the translation level.

Key words: Hepatitis C virus, Core protein, p21/Waf1, Translation inhibition

429

Microbiol. Immunol., 47(6), 429–438, 2003

Abbreviations: aa, amino acid; CHX, cycloheximide; DTT,dithiothreitol; ER, endoplasmic reticulum; FITC, fluoresceinisothiocyanate; HCV, hepatitis C virus; NS, nonstructural protein;2'-5'-OAS, 2'-5'-oligoadenylate synthetase; PAGE, polyacryl-amide gel electrophoresis; PBS, phosphate-buffered saline;PCNA, proliferating cell nuclear antigen; RIPA, radioimmuno-precipitation; SDS, sodium dodecyl sulfate.

*Address correspondence to Dr. Hak Hotta, Department ofMicrobiology, Kobe University Graduate School of Medicine,7–5–1 Kusunoki-cho, Chuo-ku, Kobe, Hyogo 650–0017, Japan.Fax: �81–78–382–5519. E-mail: hotta@kobe-u.ac.jp

�Present address: Department of Bacterial Toxinology, ResearchInstitute for Microbial Diseases, Osaka University, 3–1 Yamada-oka, Suita, Osaka 565–0871, Japan.

native viral particles (24, 55). HCV core protein islocalized predominantly in the cytoplasm, with a minorfraction with a different higher-order structure beinglocalized in the nucleus (55). HCV core protein exhibitsversatile functions: it enhances or suppresses apoptosis,depending on the apoptosis-inducing stimuli and thecell type used (28, 37, 38, 44, 59). HCV core protein hasalso been shown to cooperate with the ras oncogene totransform rodent cells into a tumorigenic phenotype (6,53) and transgenic mice expressing HCV core protein inthe liver tend to develop hepatocellular carcinoma (31).HCV core protein regulates transcription, either nega-tively or positively, of certain cellular and viral genes,such as c-myc, c-fos, p53, and hepatitis B virus (25, 35,36, 39, 40, 47), probably through differential activationof transcription factors and other intracellular signalingpathways (1, 48, 51, 57). It has also been reported thatHCV core protein inhibits translation of capped mRNAin vitro (27). Thus, HCV core protein influences a vari-ety of cellular functions, such as gene expression and pro-tein synthesis, and is likely responsible for certain phe-notypic changes of HCV-infected cells.

We previously reported that HCV core protein, but notNS3 or NS5B, selectively inhibited p21/Waf1 expressionat the post-transcriptional level while not inhibitingexpression of 2'-5'-oligoadenylate synthetase (2'-5'-OAS) (56). In the present study, we further extended theanalysis to clarify the following issues: whether or notexpression of other cellular proteins were affected byHCV core protein, the mechanism by which p21/Waf1expression was inhibited, and which portion(s) of HCVcore protein was required for the inhibition of p21/Waf1expression. We report here that HCV core proteinstrongly inhibited synthesis and accumulation ofp21/Waf1 and other nuclear proteins, such as p53, pro-liferating cell nuclear antigen (PCNA) and c-Fos, butonly weakly, if any, that of cytoplasmic proteins includ-ing 2'-5'-OAS, RNase L and MEK. HCV core protein-mediated inhibition of p21/Waf1 expression was shownto occur principally at the translation level. We alsoreport that the inhibition was mediated by the full-lengthHCV core protein and a mutant C-terminally deletedby 8 aa [Core(1–183)] that was localized in the cyto-plasm, but was not mediated by mutants further deletedby 18 aa [Core(1–173)] or 41 aa [Core(1–150)] thatwere localized exclusively in the nucleus.

Materials and Methods

Expression plasmids. The entire coding sequence ofHCV core protein of HCV C980 cDNA (subtype 1b)(14) was subcloned into pBlueScript II SK� (Strata-gene Cloning Systems, La Jolla, Calif., U.S.A.) to gen-

erate pBS-Core (54). Plasmids for C-terminally deletedmutants of HCV core protein, pBS-Core(1–150), pBS-Core(1–173) and pBS-Core(1–183), were constructed byintroducing a stop codon after aa 150, 173 and 183,respectively. Expression plasmids for human p21/Waf1(pBS-p21) and human p53 (pBS-p53) were describedpreviously (18–20, 32, 54). The entire coding sequencesfor human PCNA (10), human c-Fos (22) and mouseMEK1 (34) were subcloned into pBlueScript II to gen-erate pBS-PCNA, pBS-c-Fos and pBS-MEK1, respec-tively. The entire coding sequence for mouse 2'-5'-OAS(17) and the entire coding sequence for human RNase L(58) were tagged with an HA epitope of influenza Avirus at the N-terminus and subcloned into pBlueScriptII to generate pBS-HA-2'-5'-OAS and pBS-HA-RNase L,respectively.

Cell culture and protein expression. HeLa cells weregrown in Dulbecco’s modification of Eagle’s minimumessential medium supplemented with 10% fetal calfserum in each well (35 mm in diameter) of a 6-well tis-sue culture dish. The cells were infected with vTF7-3, arecombinant vaccinia virus expressing T7 RNA poly-merase, and then transfected with 2 µg each of theexpression plasmids for cellular proteins mixed with anequal amount of pBS-Core or the control plasmid, pBlue-Script II SK�, using Lipofectin reagent (Life Tech-nologies, Inc.). When smaller tissue culture plates wereused, total amounts of plasmid DNA for transfectionwere reduced in proportion to the size. After cultivationfor 14–16 hr, the protein expression in the cells wasanalyzed by immunoblot, immunofluorescence andradioimmunoprecipitation (RIPA) analyses, as describedbelow. In some experiments, cells were cultivated toallow protein expression for 10 hr, and then treated withcycloheximide (CHX; 20 µM) to block de novo proteinsynthesis. The decrease in the amounts of expressed pro-teins was monitored thereafter by immunoblot analysis.

Immunoblotting. Cells were lysed in gel-loadingbuffer containing 50 mM Tris-HCl (pH 6.8), 100 mM

dithiothreitol (DTT), 2% sodium dodecyl sulfate (SDS),0.1% bromophenol blue and 10% glycerol, which wereresolved by SDS-polyacrylamide gel electrophoresis(PAGE) and electrically blotted onto a polyvinylidenedifluoride filter (Millipore). The filters were blocked withphosphate-buffered saline (PBS) containing 5% skimmilk and incubated with an appropriate mouse mono-clonal antibody (see below). After being washed withPBS containing 0.5% Tween 20, the filters were incu-bated with peroxidase-labeled goat antiserum againstmouse IgG. The protein bands were visualized by anenhanced chemiluminescence method (ECL; Amer-sham-Pharmacia Biotech).

Indirect immunofluorescence. Cells were washed

430 K. OKA ET AL

with PBS, fixed with cold methanol and incubated withan appropriate mouse monoclonal antibody at roomtemperature for 1 hr. After being washed with PBS,the cells were incubated with fluorescein isothiocyanate(FITC)-conjugated goat antiserum against mouse IgG.The samples were then washed with PBS and observedunder a fluorescent microscope (Optiphoto EFD2; Nikon,Tokyo).

Radioimmunoprecipitation. Cells transfected withthe plasmids were labeled at 12 hr postinfection with100–400 µCi of 35S-methionine/cysteine (Amersham-Pharmacia) per ml in methionine/cysteine-free mediumfor the indicated duration of time. For pulse-chaseanalysis, cells were labeled for 1 hr as described above,washed with medium to remove unincorporated 35S-methionine/cysteine, and chased thereafter in the 35S-free medium. Cell lysates were prepared in RIPA bufferconsisting of 10 mM Tris-HCl (pH 7.4), 150 mM NaCl,1% Triton X-100, 1% sodium deoxycholate and 0.1 mM

phenylmethylsulfonyl fluoride. The cell lysates wereclarified by centrifugation at 12,000�g for 20 min at 4 C,precleared with 20 µl of protein G-Sepharose beads(Amersham-Pharmacia) for 30 min, and incubated for 1hr at 4 C with either one of the appropriate antibodies.The immune complexes formed were precipitated with10 µl of protein G-Sepharose beads. After being washedfour times with PBS, the immunoprecipitates were sub-jected to SDS-PAGE and visualized by pictrography(BAS-Pictrography, Fujix, Tokyo).

Northern blotting. Total cellular RNA (3 µg) extract-ed from the transfected cells using TRIZOL reagent(Life Technologies) was subjected to 1% formaldehyde-agarose gel electrophoresis and blotted onto a nylon fil-ter. The filters were hybridized with each of 32P-labeledcDNA probes in a hybridization buffer (50% formamide,5� SSC, 1% SDS, 1% Protein blocker; Amersham-Pharmacia) at 42 C overnight. After being washedunder stringent conditions (0.2� SSC, 0.1% SDS; 55 C),the RNA band that hybridized to the probe was visualizedby autoradiography.

Antibodies. Mouse monoclonal antibodies againstp21/Waf1 (187; Santa Cruz Biotech., Santa Cruz, Calif.,U.S.A.), p53 (PAb421; Calbiochem-NovabiochemCorp.), PCNA (PC10; Calbiochem-Novabiochem), c-Fos (Ab-1; Calbiochem-Novabiochem), MEK1 (C18;Santa Cruz) and HA tag (16B12; BAbCO, Richmond,Calif., U.S.A.) were purchased. A mouse monoclonalantibody against HCV core protein was kindly providedby Dr. M. Kohara, Tokyo Metropolitan Institute of Med-ical Science, Tokyo.

Results

HCV Core Protein Selectively Inhibits Accumulation ofp21/Waf1 and Other Nuclear Proteins

Using a vaccinia virus-T7 hybrid expression system,we examined possible effects of HCV core protein onectopically expressed cellular proteins in HeLa cells.Immunoblot analysis revealed that expression of cer-tain nuclear proteins, such as p21/Waf1, p53, PCNAand c-Fos, was markedly inhibited in HCV core protein-expressing cells compared with that in the non-express-ing control cells (Fig. 1a). This inhibition was repro-ducibly observed in repeated experiments and p21/Waf1was the most strongly affected among the four proteinstested. It should also be noted that constitutive expressionof endogenous PCNA was unaffected by coexpressedHCV core protein. Endogenous expression of p21/Waf1,p53 and c-Fos was barely detected under these experi-mental conditions (data not shown), due to their lowerexpression levels compared to the plasmid-mediatedectopic expression. Despite its strong inhibitory effect on

431TRANSLATION INHIBITION BY HCV CORE PROTEIN

Fig. 1. Inhibition of certain nuclear, but not cytoplasmic, proteinsby HCV core protein. HeLa cells transiently expressing (a)nuclear proteins, such as p21/Waf1, p53, PCNA and c-Fos, and (b)cytoplasmic proteins, such as 2'-5'-OAS, RNase L and MEK1, inthe absence (�) or presence (�) of HCV core protein, wereanalyzed by immunoblotting using monoclonal antibodies againstthe respective proteins. Arrowheads indicate the respective pro-teins expressed and an asterisk depicts constitutively expressedendogenous PCNA. Relative expression levels of the respectiveproteins, with those in the absence of HCV core protein beingarbitrarily expressed as 100, are shown at the bottom.

432 K. OKA ET AL

Fig. 2. Immunofluorescence analysis. HeLa cells tran-siently expressing p21/Waf1 alone (a), HCV core pro-tein alone (c) or both p21/Waf1 and HCV core protein(b, d) were stained with monoclonal antibody againsteither p21/Waf1 (a, b) or HCV core protein (c, d). Theimmunofluorescence images of p53 (e, f) and c-Fos(g, h) in the absence (e, g) or presence of HCV core pro-tein (f, h) were also shown.

Fig. 3. Lack of transcriptional inhibition by HCV coreprotein. HeLa cells transiently transfected with variousexpression plasmids in the absence (�) or presence (�) ofthe HCV core protein expression plasmid were analyzed forsteady-state mRNA levels by Northern blot analysis usingspecific cDNA probes for p21/Waf1, p53, PCNA, c-Fos, 2'-5'-OAS, RNase L or MEK1. To verify that equal amounts oftotal cellular RNA were loaded in each set of experiments,18S ribosomal RNA stained with ethidium bromide isshown.

the nuclear proteins expression, HCV core protein bare-ly inhibited the expression of cytoplasmic proteins, suchas HA-tagged 2'-5'-OAS, HA-tagged RNase L andMEK1 (Fig. 1b). Consistent results were obtained alsowith HuH-7 human hepatoma cells (data not shown).

Immunofluorescence analysis revealed that HCV coreprotein was localized predominantly in the cytoplasmwhereas p21/Waf1 almost exclusively in the nucleus (Fig.2). Consistent with the results of immunoblotting analysis,p21/Waf1 expression was much weaker when coexpressedwith HCV core protein than when expressed alone (Fig. 2,a and b). On the other hand, the expression level of HCVcore protein was not affected by coexpression of p21/Waf1(Fig. 2, c and d). Similar results were obtained with p53and c-Fos: expression of p53 and c-Fos proteins werereproducibly decreased in HCV core protein-expressingcells compared with the control (Fig. 2, e to h). It shouldalso be noted that colocalization of HCV core proteinwith p53 or c-Fos was not observed (data not shown).

HCV Core Protein Inhibits p21/Waf1 Expression Post-Transcriptionally

To look into the mechanism of the HCV core protein-

mediated inhibition of protein expression, we examinedsteady-state mRNA levels by Northern blot analysis.As shown in Fig. 3, mRNA expression for any of the pro-teins tested was not significantly inhibited by coex-pressed HCV core protein. This result suggests thatthe decreased accumulation of such proteins as p21/Waf1,p53, PCNA and c-Fos was not due to transcriptionalsuppression by HCV core protein, but was mediatedpost-transcriptionally.

HCV Core Protein Inhibits p21/Waf1 Protein SynthesisWe then performed pulse-label and pulse-chase exper-

iments to analyze protein synthesis and degradation.Since p21/Waf1 was most strongly affected (see Fig.1a), we focused on the analysis of p21/Waf1. Cellswere pulse-labeled with 35S-methionine/cysteine for 5, 15and 30 min and subjected to immunoprecipitation analy-sis. The result obtained clearly demonstrated that theamount of p21/Waf1 that had been newly synthesized andaccumulated was much smaller in HCV core protein-expressing cells than in the control (Fig. 4a). On theother hand, inhibition of 2'-5'-OAS synthesis by HCVcore protein was only minimal (Fig. 4b). Pulse-chase

433TRANSLATION INHIBITION BY HCV CORE PROTEIN

Fig. 4. Analysis of synthesis and degradation of p21/Waf1 and 2'-5'-OAS in the absence or presence of HCV core protein. (a)Cells expressing p21/Waf1 in the absence or presence of HCVcore protein were pulse-labeled with 100 µCi of 35S-methio-nine/cysteine per ml for 5, 15 and 30 min and subjected toimmunoprecipitation analysis using anti-p21/Waf1 monoclonalantibody. Relative amounts of p21/Waf1 in the absence (�) andpresence of HCV core protein (�) are plotted. The amount ofp21/Waf1 after labeling for 30 min in the absence of HCV coreprotein was arbitrarily expressed as 100%. (b) Cells expressingHA-tagged 2'-5'-OAS in the absence or presence of HCV coreprotein were analyzed in the same way as in (a) except for usinganti-HA monoclonal antibody. Relative amounts of HA-tagged 2'-5'-OAS in the absence (�) and presence of HCV core protein (�)are plotted. (c) Cells expressing p21/Waf1 in the absence orpresence of HCV core protein were pulse-labeled with 100 µCi of35S-methionine/cysteine per ml for 1 hr. After being labeled, thecells were washed to remove unincorporated 35S-methionine/cys-teine and chased in 35S-free medium for 1 and 3 hr. The cells werethen subjected to immunoprecipitation analysis using anti-p21/Waf1 monoclonal antibody. Relative amounts of p21/Waf1 inthe absence (�) and presence of HCV core protein (�) are plot-ted. The amount of p21/Waf1 at 0 hr in the absence of HCV coreprotein was expressed as 100%. (d) Cells that had been allowedto express and accumulate p21/Waf1 for 10 hr were treated withCHX to block de novo protein synthesis and degradation patternsof p21/Waf1 were monitored by immunoblot analysis at 0, 1.5 and3 hr after initiation of CHX treatment. Mean values�S.D. ofrelative amounts of p21/Waf1 in the absence (�) and presence ofHCV core protein (�) obtained from three independent experi-ments are plotted. The amount of p21/Waf1 at 0 hr in the absenceof HCV core protein was expressed as 100%.

experiments revealed that p21/Waf1 was degraded witha half-life of approximately 1.5 hr in the absence ofHCV core protein (Fig. 4c). Degradation of p21/Waf1 inHCV core protein-expressing cells could not be moni-tored accurately, due to inhibited, nearly backgroundlevels of initial accumulation of 35S-labeled p21/Waf1.We then performed experiments in which CHX wasused to block de novo protein synthesis so that proteindegradation could be monitored thereafter. Cells wereallowed to express p21/Waf1 for 10 hr, treated withCHX and then analyzed for p21/Waf1 protein levels byimmunoblotting. The results revealed that an equalamount of p21/Waf1 was degraded in HCV core protein-expressing cells and the control during a given period oftime (Fig. 4d), suggesting the possibility that, oncep21/Waf1 is accumulated in the cell, its degradationpattern was not significantly affected by HCV core pro-tein. During the first 1.5 hr, p21/Waf1 decreased to thelevel approximately 60% of the initial amount at 0 hr,with the degradation pattern being similar to what wasobserved in the pulse-chase experiments.

The Full-Length or Nearly a Full-Length HCV CoreProtein Is Required for Inhibition of p21/Waf1 Synthesisand Accumulation

Two forms of HCV core protein were reported, onewith the molecular mass of 23 kDa and the other 21kDa, with the latter being generated probably through thecleavage near the C-terminus of the full-length HCVcore protein (between aa 174 and 191) (55). In order tosee which form(s) was responsible for the decreasedexpression of p21/Waf1, we expressed C-terminallydeleted mutants of HCV core protein, Core(1–183),Core(1–173) and Core(1–150), in addition to the full-length HCV core protein of 191 aa. Immunofluorescenceanalysis revealed that, whereas the full-length HCV coreprotein and Core(1–183) were localized predominantlyin the cytoplasm, Core(1–173) and Core(1–150) werelocalized almost exclusively in the nucleus (Fig. 5a).We also noticed that expression levels of Core(1–173)and Core(1–150) were lower than that of the full-lengthHCV core protein when the same amount of the plasmidswas transfected (data not shown). Therefore, we trans-fected pBS-Core(1–173) or pBS-Core(1–150) of anamount double that of pBS-Core to ensure the sameexpression level as that of the full-length HCV coreprotein. The result obtained demonstrated that neitherCore(1–173) nor Core(1–150) significantly inhibitedsynthesis and accumulation of p21/Waf1, although thefull-length HCV core protein of 191 aa and Core(1–183)at the same expression level clearly did (Fig. 5, b and c).We also observed that Core(1–183) significantly inhibitedsynthesis and accumulation of p53 and c-Fos proteins, as

434 K. OKA ET AL

Fig. 5. Requirement of the full-length or nearly a full-lengthHCV core protein for the inhibition of synthesis and accumulationof p21/Waf1. (a) Schematic representation and subcellular local-ization of the full-length (C191) or C-terminally deleted mutantsof HCV core protein (C183, C173 and C150). The immunofluo-rescent microscopic images depict intracellular localization of thefull-length and C-terminally deleted mutants of HCV core protein.(b) Immunoblot analysis. HeLa cells transiently expressingp21/Waf1 in the absence (Vector) or presence of the full-length(C191) or C-terminally deleted mutants of HCV core protein(C183, C173 and C150) were analyzed for accumulation ofp21/Waf1 protein (arrow). Molecular masses of marker proteins(in kDa) are shown on the left. (c) The relative intensity of thebands in (b) was measured. The amount of p21/Waf1 in theabsence of HCV core protein was expressed as 100%.

did the full-length HCV core protein of 191 aa (datanot shown).

Discussion

We previously reported selective inhibition ofp21/Waf1, but no inhibition of 2'-5'-OAS, by HCV coreprotein using a recombinant vaccinia virus-T7 hybridexpression system (56). In the present study using thesame expression system, we observed that, in addition top21/Waf1, expression of the other nuclear proteins, suchas p53, PCNA and c-Fos, was strongly inhibited whentransiently coexpressed with HCV core protein (Fig.1a). It should be noted that p21/Waf1 was the moststrongly affected among the proteins tested. On theother hand, expression of some cytoplasmic proteins,such as 2'-5'-OAS, RNase L and MEK1, was barelyaffected by HCV core protein under the same experi-mental conditions (Fig. 1b). The expression levels ofHCV core protein did not differ significantly betweencells coexpressing any of the proteins tested and thoseexpressing HCV core protein alone (Fig. 2, c and d;data not shown). Even 2'-5'-OAS or RNase L did notsignificantly inhibit HCV core protein synthesis, whichcould be reasonably explained by the fact that vacciniavirus E3L protein counteracts the inhibitory functionof 2'-5'-OAS (42) and, presumably, RNase L as well. Weobserved previously that NS3 or NS5B did not inhibitp21/Waf1 expression while HCV core protein did (56).Using a system that did not utilize recombinant vac-cinia virus, we also observed that HCV core protein,NS4A and NS4B, but not NS3, NS5A or NS5B, inhib-ited p21/Waf1 expression post-transcriptionally (9). Itthus appears that HCV core protein selectively inhibitsexpression of some nuclear proteins.

The decreased p21/Waf1 expression in HCV coreprotein-expressing cells was not due to transcriptionalsuppression by HCV core protein, as demonstrated byNorthern blot analysis that showed the equal levels ofmRNA for the respective proteins irrespective of HCVcore protein expression (Fig. 3). Pulse-label experi-ments strongly suggested inhibited synthesis of p21/Waf1in HCV core protein-expressing cells (Fig. 4a). HCVcore protein-mediated inhibition of capped, but notuncapped, RNA translation was recently demonstratedusing an in vitro translation system (27). It is feasiblethat accumulation of HCV core protein in endoplasmicreticulum (ER), which is thought to occur commonly inoverexpressing cells, leads to activation of certainenzymes, such as double-stranded RNA-activated proteinkinase (PKR), PKR-like ER kinase (PERK) (12, 13,30) and ER transmembrane kinase/ribonuclease (Ire1)(21). Activation of PKR or PERK phosphorylates

eukaryotic initiation factor (eIF)-2α to inhibit proteinsynthesis, and Ire1 is thought to repress translationthrough 28S ribosomal RNA cleavage in response to ERstress. All of the above mechanisms, however, are like-ly to exert general inhibitory effects on synthesis of anykind of protein. With a result that demonstrates differ-ential inhibition of protein synthesis by HCV core protein(Fig. 1), one must consider the possible involvement ofan additional mechanism(s) mediated by HCV core pro-tein, such as selective translational inhibition of particularproteins, impaired intracellular/nuclear transport orimpaired folding/maturation of certain proteins imme-diately after their synthesis. As for the selective trans-lational inhibition of particular proteins, it was recentlyreported that a nascent peptide sequence of a bacterialprotein could influence translation continuation or ter-mination within a translating ribosome (11).

The accumulation of unfolded proteins in ER inducestwo other distinct cellular responses, in addition to theinhibition of protein synthesis. One is the unfoldedprotein response (UPR) that induces transcriptional acti-vation of genes encoding the ER-resident chaperoneproteins, which help proper folding of the target proteins(4, 30). The other response is ER-associated degradation(ERAD) of misfolded proteins, which requires a mem-brane-anchored ubiquitin ligase, such as Hrd1p/Der3p (3,52). Such a ubiquitin ligase(s) has an apparent preferencefor misfolded forms of certain proteins. The ERADsystem may be functioning in the HCV core protein-expressing cells to degrade newly synthesized, misfold-ed proteins. On the other hand, HCV core protein doesnot seem to affect the ordinary ubiquitin-proteasomesystem, which is known to play an important role indegradation of properly folded p21/Waf1 (5, 26). Therationale for this statement is that, once being accumu-lated in the cell, p21/Waf1 stability was not significant-ly affected by HCV core protein (Fig. 4d). This mightalso explain why only transiently expressed PCNA, butnot endogenous PCNA that had been accumulated as astable form, was affected by HCV core protein (Fig.1a).

Of the C-terminally deleted mutants of HCV coreprotein tested, Core(1–150) and Core(1–173), but notCore(1–183), failed to inhibit synthesis or accumula-tion of p21/Waf1 even at the same expression level as thatof the full-length HCV core protein (Fig. 5). Whereas thefull-length HCV core protein and Core(1–183) werelocalized predominantly in the cytoplasm, Core(1–173)and Core(1–150) were found exclusively in the nucleus.The different subcellular localization of the mutantswould certainly exert differential effects on a variety ofcellular responses. In this connection, the equivalentmutant(s) of Core(1–173) was reported to be localized

435TRANSLATION INHIBITION BY HCV CORE PROTEIN

either in the nucleus (24), the cytoplasm (16, 45), orboth (28), probably depending on the cell lines used. Inany case, our results suggest that the full-length or near-ly a full-length sequence (aa 1–183) is required for HCVcore protein to inhibit synthesis and accumulation ofcertain proteins.

As for the possible effects of HCV core protein on p53and p21/Waf1 expression, controversial observationshave been reported. Ray et al. (39, 40) observed thatHCV core protein transcriptionally suppressed p53 andp21/Waf1 gene expression. On the other hand, it wasrecently reported that HCV core protein enhanced p53function that resulted in increased p21/Waf1 expression(15, 25, 35). Those results were obtained with culturedcells. In humans, p21/Waf1 expression was reduced inHCV-associated hepatocellular carcinoma throughimpaired p53 function (46). At the molecular level,HCV core protein was shown to associate physicallywith p53 (35) and p21/Waf1 (54) using in vitro glu-tathione S-transferase fusion protein binding assay.Their physical association in vivo, however, has yet to bedemonstrated. We have demonstrated in the presentstudy that HCV core protein inhibits synthesis and accu-mulation of p21/Waf1 and p53. The virological signifi-cance, as well as the underlying mechanism, of our find-ing needs further investigation.

We are grateful to Drs. M. Kohara (Tokyo Metropolitan Insti-tute of Medical Science, Tokyo, Japan), K. Shimotohno (KyotoUniversity, Kyoto, Japan), T. Tsurimoto (Nara Institute of Scienceand Technology, Ikoma, Japan), Y. Sokawa (Kyoto Institute ofTechnology, Kyoto, Japan), R. Silverman (Cleveland ClinicFoundation, Cleveland, Ohio., U.S.A.) and K. Okazaki (Bio-molecular Engineering Research Institute, Osaka, Japan) forproviding the mouse monoclonal antibody against HCV coreprotein, HCV C980 cDNA and the full-length cDNAs for PCNA(pT7-PCNA), 2'-5'-OAS (MA25), RNase L (pZC5) and MEK1(pactEFMek1), respectively. This work was supported in partby a Grant-in-Aid for Scientific Research from Japan Society forthe Promotion of Science, and a research grant from ResearchFoundation of Viral Hepatitis, Japan.

References

1) Aoki, H., Hayashi, J., Moriyama, M., Arakawa, Y., andHino, O. 2000. Hepatitis C virus core protein interacts with14-3-3 protein and activates the kinase Raf-1. J. Virol. 74:1736–1741.

2) Apichartpiyakul, C., Chittivudikarn, C., Miyajima, H.,Homma, M., and Hotta, H. 1994. Analysis of hepatitis Cvirus isolates among healthy blood donors and drug addictsin Chiang Mai, Thailand. J. Clin. Microbiol. 32: 2276–2279.

3) Bays, N.W., Gardner, R.G., Seelig, L.P., Joazeiro, C.A., andHampton, R.Y. 2001. Hrd1p/Der3p is a membrane-anchoredubiquitin ligase required for ER-associated degradation.

Nat. Cell Biol. 3: 24–29.4) Benedetti, C., Fabbri, M., Sitia, R., and Cabibbo, A. 2000.

Aspects of gene regulation during the UPR in human cells.Biochem. Biophys. Res. Commun. 278: 530–536.

5) Blagosklonny, M.V., Wu, G.S., Omura, S., and El-Deiry, W.S.1996. Proteasome-dependent regulation of p21WAF1/CIP1expression. Biochem. Biophys. Res. Commun. 227: 564–569.

6) Chang, J., Yang, S.-H., Cho, Y.-G., Hwang, S.B., Hahn,Y.S., and Sung, Y.C. 1998. Hepatitis C virus core from twodifferent genotypes has an oncogenic potential but is not suf-ficient for transforming primary rat embryo fibroblasts incooperation with the H-ras oncogene. J. Virol. 72:3060–3065.

7) Doi, H., Apichartpiyakul, C., Ohba, K., Mizokami, M., andHotta, H. 1996. Hepatitis C virus (HCV) subtype preva-lence in Chiang Mai, Thailand, and identification of novelsubtypes of HCV major type 6. J. Clin. Microbiol. 34:569–574.

8) Enomoto, N., Sakuma, I., Asahina, Y., Kurosaki, M.,Murakami, T., Yamamoto, C., Ogura, Y., Izumi, N., Marumo,F., and Sato, C. 1996. Mutations in the nonstructural protein5A gene and response to interferon in patients with chronichepatitis C virus 1b infection. N. Engl. J. Med. 334: 77–81.

9) Florese, R., Nagano-Fujii, M., Iwanaga, Y., Hidajat, R., andHotta, H. 2002. Inhibition of protein synthesis by the non-structural proteins NS4A and NS4B of hepatitis C virus.Virus Res. 90: 119–131.

10) Fukuda, K., Morioka, H., Imajou, S., Ikeda, S., Ohtsuka, E.,and Tsurimoto, T. 1995. Structure-function relationship of theeukaryotic DNA replication factor, proliferating cell nuclearantigen. J. Biol. Chem. 270: 22527–22534.

11) Gong, F., and Yanofsky, C. 2002. Instruction of translatingribosome by nascent peptide. Science 297: 1864–1867.

12) Harding, H.P., Zhang, Y., Bertolotti, A., Zeng, H., and Ron,D. 2000. Perk is essential for translational regulation and cellsurvival during the unfolded protein response. Mol. Cell 5:897–904.

13) Harding, H.P., Zhang, Y., and Ron, D. 1999. Protein trans-lation and folding are coupled by an endoplasmic-reticulum-resident kinase. Nature 397: 271–274.

14) Hijikata, M., Kato, N., Ootsuyama, Y., Nakagawa, M., andShimotohno, K. 1991. Gene mapping of the putative struc-tural region of the hepatitis C virus genome by in vitroprocessing analysis. Proc. Natl. Acad. Sci. U.S.A. 88:5547–5551.

15) Honda, M., Kaneko, S., Shimazaki, T., Matsushita, E.,Kobayashi, K., Ping, L.-H., Zhang, H.-C., and Lemon, S.M.2000. Hepatitis C virus core protein induces apoptosis andimpairs cell-cycle regulation in stably transformed Chinesehamster ovary cells. Hepatology 31: 1351–1359.

16) Hope, R.G., and McLauchlan, J. 2000. Sequence motifsrequired for lipid droplet association and protein stability areunique to the hepatitis C virus core protein. J. Gen. Virol. 81:1913–1925.

17) Ichii, Y., Fukunaga, R., Shiojiri, S., and Sokawa, Y. 1986.Mouse 2-5A synthetase cDNA: nucleotide sequence andcomparison to human 2-5A synthetase. Nucleic Acids Res.14: 10117.

18) Ishido, S., and Hotta, H. 1998. Complex formation of the

436 K. OKA ET AL

nonstructural protein 3 of hepatitis C virus with the p53tumor suppressor. FEBS Lett. 438: 258–262.

19) Ishido, S., Fujita, T., and Hotta, H. 1998. Complex formationof NS5B with NS3 and NS4A proteins of hepatitis C virus.Biochem. Biophys. Res. Commun. 244: 35–40.

20) Ishido, S., Muramatsu, S., Fujita, T., Iwanaga, Y., Tong,W.-Y., Katayama, Y., Itoh, M., and Hotta, H. 1997. Wild-type, but not mutant-type, p53 enhances nuclear accumula-tion of the NS3 protein of hepatitis C virus. Biochem. Bio-phys. Res. Commun. 230: 431–436.

21) Iwawaki, T., Hosoda, A., Okuda, T., Kamigori, Y., Nomura-Furuwatari, C., Kimata, Y., Tsuru, A., and Kohno, K. 2001.Translational control by the ER transmembranekinase/ribonuclease IRE1 under ER stress. Nat. Cell Biol. 3:158–164.

22) Kuroki, Y., Shiozawa, S., Yoshihara, R., and Hotta, H. 1993.The contribution of human c-fos DNA to cultured synovialcells: a transfection study. J. Rheumatol. 20: 422–428.

23) Lauer, G.M., and Walker, B.D. 2001. Hepatitis C virusinfection. N. Engl. J. Med. 345: 41–52.

24) Liu, Q., Tackney, C., Bhat, R.A., Prince, A.M., and Zhang, P.1997. Regulated processing of hepatitis C virus core proteinis linked to subcellular localization. J. Virol. 71: 657–662.

25) Lu, W., Lo, S.-Y., Chen, M., Wu, K.-J., Fung, Y.K.T., and Ou,J.-H. 1999. Activation of p53 tumor suppressor by hepatitisC virus core protein. Virology 264: 134–141.

26) Maki, C.G., and Howley, P.M. 1997. Ubiquitination of p53and p21 is differentially affected by ionizing and UV radia-tion. Mol. Cell. Biol. 17: 355–363.

27) Mamiya, N., and Worman, H.J. 1999. Hepatitis C virus coreprotein binds to a DEAD box RNA helicase. J. Biol. Chem.274: 15751–15756.

28) Marusawa, H., Hijikata, M., Chiba, T., and Shimotohno, K.1999. Hepatitis C virus core protein inhibits Fas- and tumornecrosis factor alpha-mediated apoptosis via NF-κB activa-tion. J. Virol. 73: 4713–4720.

29) Mellor, J., Holmes, E.C., Jarvis, L.M., Yap, P.L., Simmonds,P., and The International HCV Collaborative Study Group.1995. Investigation of the pattern of hepatitis C virussequence diversity in different geographical regions: impli-cations for virus classification. J. Gen. Virol. 76: 2493–2507.

30) Mori, K. 2000. Tripartite management of unfolded proteinsin the endoplasmic reticulum. Cell 101: 451–454.

31) Moriya, K., Fujie, H., Shintani, Y., Yotsuyanagi, H., Tsutsumi,T., Ishibashi, K., Matsuura, Y., Kimura, S., Miyamura, T., andKoike, K. 1998. The core protein of hepatitis C virus induceshepatocellular carcinoma in transgenic mice. Nat. Med. 4:1065–1067.

32) Muramatsu, S., Ishido, S., Fujita, T., Itoh, M., and Hotta, H.1997. Nuclear localization of the NS3 protein of hepatitis Cvirus and factors affecting the localization. J. Virol. 71:4954–4961.

33) Ogata, S., Ku, Y., Yoon, S., Makino, S., Nagano-Fujii, M.,and Hotta, H. 2002. Correlation between secondary structureof an amino-terminal portion of the nonstructural protein 3(NS3) of hepatitis C virus and development of hepatocellu-lar carcinoma. Microbiol. Immunol. 46: 549–554.

34) Okazaki, K., and Sagata, N. 1995. The Mos/MAP kinasepathway stabilizes c-Fos by phosphorylation and augments

its transforming activity in NIH 3T3 cells. EMBO J. 14:5048–5059.

35) Otsuka, M., Kato, N., Lan, K.-H., Yoshida, H., Kato, J.,Goto, T., Shiratori, Y., and Omata, M. 2000. Hepatitis Cvirus core protein enhances p53 function through augmen-tation of DNA binding affinity and transcriptional ability. J.Biol. Chem. 275: 34122–34130.

36) Ray, R.B., Lagging, L.M., Meyer, K., Steele, R., and Ray, R.1995. Transcriptional regulation of cellular and viral pro-moters by the hepatitis C virus core protein. Virus Res. 37:209–220.

37) Ray, R.B., Meyer, K., and Ray, R. 1996. Suppression ofapoptotic cell death by hepatitis C virus core protein. Virol-ogy 226: 176–182.

38) Ray, R.B., Meyer, K., Steele, R., Shrivastava, A., Aggarwal,B.B., and Ray, R. 1998. Inhibition of tumor necrosis factor(TNF-α)-mediated apoptosis by hepatitis C virus core pro-tein. J. Biol. Chem. 273: 2256–2259.

39) Ray, R.B., Steele, R., Meyer, K., and Ray, R. 1997. Tran-scriptional repression of p53 promoter by hepatitis C viruscore protein. J. Biol. Chem. 272: 10983–10986.

40) Ray, R.B., Steele, R., Meyer, K., and Ray, R. 1998. Hepati-tis C virus core protein represses p21WAF1/Cip1/Sdi1 pro-moter activity. Gene 208: 331–336.

41) Reed, K.E., and Rice, C.M. 2000. Overview of hepatitis Cvirus genome structure, polyprotein processing, and pro-tein properties. Curr. Top. Microbiol. Immunol. 242: 55–84.

42) Rivas, C., Gil, J., Melková, Z., Esteban, M., and Díaz-Guer-ra, M. 1998. Vaccinia virus E3L protein is an inhibitor of theinterferon (i.f.n.)-induced 2-5A synthetase enzyme. Virolo-gy 243: 406–414.

43) Robertson, B., Myers, G., Howard, C., Brettin, T., Bukh, J.,Gaschen, B., Gojobori, T., Maertens, G., Mizokami, M.,Nainan, O., Netesov, S., Nishioka, K., Shin-i., T., Sim-monds, P., Smith, D., Stuyver, L., and Weiner, A. 1998.Classification, nomenclature, and database development forhepatitis C virus (HCV) and related virus: proposals forstandardization. Arch. Virol. 143: 2493–2503.

44) Ruggieri, A., Harada, T., Matsuura, Y., and Miyamura, T.1997. Sensitization to Fas-mediated apoptosis by hepatitis Cvirus core protein. Virology 229: 68–76.

45) Sabile, A., Perlemuter, G., Bono, F., Kohara, K., Demaugre,F., Kohara, M., Matsuura, Y., Miyamura, T., Brechot, C., andBarba, G. 1999. Hepatitis C virus core protein binds toapolipoprotein AII and its secretion is modulated by fibrates.Hepatology 30: 1064–1076.

46) Shi, Y.Z., Hui, A.M., Takayama, T., Li, X., Cui, X., andMakuuchi, M. 2000. Reduced p21(WAF1/CIP1) proteinexpression is predominantly related to altered p53 in hepa-tocellular carcinomas. Brit. J. Cancer 83: 50–55.

47) Shih, C.-M., Chen, C.-M., Chen, S.-Y., and Lee, Y.-H.W.1995. Modulation of the trans-suppression activity of hepati-tis C virus core protein by phosphorylation. J. Virol. 69:1160–1171.

48) Shrivastava, A., Manna, S.K., Ray, R., and Aggarwal, B.B.1998. Ectopic expression of hepatitis C virus core protein dif-ferentially regulates nuclear transcription factors. J. Virol. 72:9722–9728.

49) Soetjipto, Handajani, R., Lusida, M.I., Darmadi, S., Adi,

437TRANSLATION INHIBITION BY HCV CORE PROTEIN

P., Soemarto, Ishido, S., Katayama, Y., and Hotta, H. 1996.Differential prevalence of hepatitis C virus subtypes inhealthy blood donors, patients on maintenance hemodialysis,and patients with hepatocellular carcinoma in Surabaya,Indonesia. J. Clin. Microbiol. 34: 2875–2880.

50) Song, J., Fujii, M., Wang, F., Itoh, M., and Hotta, H. 1999.The NS5A protein of hepatitis C virus partially inhibits theantiviral activity of interferon. J. Gen. Virol. 80: 879–886.

51) Tai, D.-I., Tsai, S.-L., Chen, Y.-M., Chuang, Y.-L., Peng, C.-Y., Sheen, I.-S., Yeh, C.-T., Chang, K.S.S., Huang, S.-N.,Kuo, G.C., and Liaw, Y.-F. 2000. Activation of nuclear fac-tor κB in hepatitis C virus infection: implications for patho-genesis and hepatocarcinogenesis. Hepatology 31: 656–664.

52) Travers, K.J., Patil, C.K., Wodicka, L., Lockhart, D.J., Weiss-man, J.S., and Walter, P. 2000. Functional and genomicanalyses reveal an essential coordination between the unfold-ed protein response and ER-associated degradation. Cell101: 249–258.

53) Tsuchihara, K., Hijikata, M., Fukuda, K., Kuroki, T.,Yamamoto, N., and Shimotohno, K. 1999. Hepatitis C viruscore protein regulates cell growth and signal transductionpathway transmitting growth stimuli. Virology 258: 100–107.

54) Wang, F., Yoshida, I., Takamatsu, M., Ishido, S., Fujita, T.,Oka, K., and Hotta, H. 2000. Complex formation between

hepatitis C virus core protein and p21Waf1/Cip1/Sdi1.Biochem. Biophys. Res. Commun. 273: 479–484.

55) Yasui, K., Wakita, T., Tsukiyama-Kohara, K., Funahashi, S.-I., Ichikawa, M., Kajita, T., Moradpour, D., Wands, J.R.,and Kohara, M. 1998. The native form and maturationprocess of hepatitis C virus core protein. J. Virol. 72:6048–6055.

56) Yoshida, I., Oka, K., Hidajat, R., Nagano-Fujii, M., Ishido, S.,and Hotta, H. 2001. Inhibition of p21/Waf1/Cip1/Sdi1expression by hepatitis C virus core protein. Microbiol.Immunol. 45: 689–697.

57) You, L.-R., Chen, C.-M., and Lee, Y.-H.W. 1999. Hepatitis Cvirus core protein enhances NF-κB signal pathway trigger-ing by lymphotoxin-β receptor ligand and tumor necrosis fac-tor alpha. J. Virol. 73: 1672–1681.

58) Zhou, A., Hassel, B.A., and Silverman, R.H. 1993. Expres-sion cloning of 2-5A-dependent RNAase: a uniquely regu-lated mediator of interferon action. Cell 72: 753–765.

59) Zhu, N., Khoshnan, A., Schneider, R., Matsumoto, M., Den-nert, G., Ware, C., and Lai, M.M.C. 1998. Hepatitis C viruscore protein binds to the cytoplasmic domain of tumornecrosis factor (TNF) receptor 1 and enhances TNF-inducedapoptosis. J. Virol. 72: 3691–3697.

438 K. OKA ET AL