Activation of the CKI-CDK-Rb-E2F Pathway inFull Genome Hepatitis C Virus-expressing Cells*

Received for publication, November 24, 2003, and in revised form, January 12, 2004Published, JBC Papers in Press, January 26, 2004, DOI 10.1074/jbc.M312822200

Kyoko Tsukiyama-Kohara,a,b Shigenobu Tone,a,c Isao Maruyama,a Kazuaki Inoue,a,d

Asao Katsume,a,e Hideko Nuriya,a Hiroshi Ohmori,e Jun Ohkawa,f Kazunari Taira,g

Yutaka Hoshikawa,h Futoshi Shibasaki,h Michael Reth,i Yohsuke Minatogawa,c

and Michinori Koharaa, j

From the Departments of aMicrobiology and Cell Biology and hCell Physiology, Tokyo Metropolitan Institute of MedicalScience, 3-18-22 Honkomagome, Bunkyo-ku, Tokyo 113-8613, Japan, the cDepartment of Biochemistry, Kawasaki MedicalSchool, 577 Matsushima, Kurashiki City, Okayama 701-0192, Japan, the dDivision of Gastroenterology, Showa UniversityFujigaoka Hospital, Aoba-ku, Fujigaoka 1-30, Yokohama 227-8501, Japan, the eFuji Gotemba Research Laboratory,Chugai Pharmaceutical Company, Limited, 135 Komakado 1-chome, Gotemba-shi, Shizuoka 412, Japan, the fAIDSResearch Center, National Institute of Health, 1-23-1 Toyama-cho, Shinjuku-ku, Tokyo 162-8640, Japan, the gDepartmentof Chemistry and Biotechnology, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan, and theiMax-Planck-Institut fur Immunbiologie, Stubeweg 51, D-79108 Freiburg, Germany

Hepatitis C virus (HCV) causes persistent infection inhepatocytes, and this infection is, in turn, strongly asso-ciated with the development of hepatocellular carci-noma. To clarify the mechanisms underlying these ef-fects, we established a Cre/loxP conditional expressionsystem for the precisely self-trimmed HCV genome inhuman liver cells. Passage of hepatocytes expressingreplicable full-length HCV (HCR6-Rz) RNA caused up-regulation of anchorage-independent growth after 44days. In contrast, hepatocytes expressing HCV struc-tural, nonstructural, or all viral proteins showed no sig-nificant changes after passage for 44 days. Only cellsexpressing HCR6-Rz passaged for 44 days displayed ac-celeration of CDK activity, hyperphosphorylation of Rb,and E2F activation. These results demonstrate that fullgenome HCV expression up-regulates the CDK-Rb-E2Fpathway much more effectively than HCV proteins dur-ing passage.

Hepatitis C virus (HCV)1 causes the persistent infectionchronic hepatitis in most infected patients. This disorder even-tually progresses to cirrhosis and hepatocellular carcinoma(HCC). Numerous studies have provided evidence supporting alink between chronic HCV infection and HCC (1, 2). However,exactly how HCV infection could be directly involved in thedevelopment of HCC remains unclear because of the lack of anefficient in vitro infection system.

HCV is a member of the Flaviviridae family and has a pos-

itive-strand RNA genome (�9.6 kb). Viral proteins are synthe-sized as a single polyprotein, which is then cleaved into struc-tural (core, E1, and E2) and nonstructural (NS2, NS3, NS4A,NS4B, NS5A, and NS5B) proteins. In vitro expression of HCVcore protein reportedly influences cell growth regulation, andthe protein can interact with the cytoplasmic tail of a lympho-toxin-� receptor (3), the death domain of tumor necrosis factorreceptor-1 (4), and NF-�B (5), resulting in enhancement orinhibition of Fas- and tumor necrosis factor-�-mediated apo-ptosis. HCV structural protein inhibits Fas-mediated apoptosisby inhibiting cytochrome c release from mitochondria in mice(6). Cellular transformation has been shown to be caused byHCV core protein in the presence of the ras gene (7, 8) and byloss of function of LZIP (9) and the presence of STAT3 (signaltransducer and activator of transcription-3) (10). NS3 (11) andNS4B (12) proteins reportedly display tumorigenicity in thepresence of ras. These results indicate that expression of indi-vidual HCV proteins does not cause cellular transformation invitro.

HCV core protein transgenic mice reportedly show inducedsteatosis and, after 16 months of age, develop HCC (13). Incontrast, HCV structural protein transgenic mice do not dis-play neoplastic or cancerous lesions in the liver by 20 months ofage (14). Moreover, conditional expression of an HCV struc-tural protein region (nucleotides 294–3435) causes hepatic in-jury in transgenic mice (15), but HCC is not observed by atleast 16 months of age (data not shown). The frequency of HCCoccurrence is reportedly higher in full-length HCV polyproteintransgenic mice than in those with the structural protein only(16). However, whether HCV proteins can represent directtriggers of transformation in hepatocytes remains unclear.

Cirrhosis and irregular regeneration have been reported asrisk factors for HCC (17). During persistent HCV infection,hepatic injury and regeneration repeatedly occur in the liver.An efficient system of HCV infection is required to clarify theeffects of HCV on cell growth. Efficient replication systemshave recently been established using HCV replicons andHuH-7 cells (18, 19). HuH-7 cells display a point mutation atcodon 220 of p53 (20), and chemosensitivity is decreased com-pared with other cell lines with wild-type p53 (21). To clarifyhow HCV infection modifies hepatocyte growth, we establishedan expression cassette of replicable full genome HCV, as con-

* This work was supported by grant-in-aids from the Ministry ofEducation, Culture, Sports, Science, and Technology of Japan and fromthe Ministry of Health, Labor, and Welfare of Japan and by the Orga-nization for Pharmaceutical Safety and Research of Japan. The costs ofpublication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked “advertisement”in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

b Present address: Lab. Animal Research Center, Inst. of MedicalScience, University of Tokyo, 4-6-1 Shirokane-dai, Minato-ku, Tokyo108-8639, Japan.

j To whom correspondence should be addressed. Tel.: 81-3-3823-2101;Fax: 81-3-3828-8945; E-mail: mkohara@rinshoken.or.jp.

1 The abbreviations used are: HCV, hepatitis C virus; HCC, hepato-cellular carcinoma; DHFR, dihydrofolate reductase; IFN, interferon;CDK, cyclin-dependent kinase; FACS, fluorescence-activated cell sort-er; dsRNA, double-stranded RNA; kbp, kilobase pair; CKI, cyclin-de-pendent kinase inhibitor.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 279, No. 15, Issue of April 9, pp. 14531–14541, 2004© 2004 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

This paper is available on line at http://www.jbc.org 14531

by guest on April 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

firmed in IMY cells and Tupaia belangeri,2 and HepG2 cellsoriginating from human hepatoblastoma and exhibiting char-acteristics of differentiated hepatocytes in that the response togrowth factors and p53 functions is retained (22, 23). Liver celllines in which HCV genomes can be conditionally expressedusing the Cre/loxP switching system (2) and precisely self-trimmed at the 5� and 3� termini using ribozyme sequenceswere thus established in this study. HCV-expressing cells werepassaged and characterized according to changes in cell growthregulation.

EXPERIMENTAL PROCEDURES

cDNA Cloning and Plasmid Construction—Full genome HCV cDNA(nucleotides 1–9611; DDBJ/GenBankTM/EBI accession numberAY045702) was cloned from the serum of a chronic hepatitis patient (HCVgenotype 1b; 50% chimpanzee infectious dose, 104.5 copies/ml; PCR titer,106.2 copies/ml). Complementary DNA clones with a consensus sequencein more than three clones were utilized for construction of a full-lengthclone. This was then subcloned under the control of a CAG promoter (24)and the Cre/loxP conditional expression cassette using a neomycin resist-ance gene with a poly(A) signal as a stuffer (see Fig. 1A) (2). The preciseHCV RNA can be exactly excised by the presence of hammerhead ri-bozyme sequences (25) at the 5� terminus and the hepatitis D virusribozyme (26) at the 3� terminus (see Fig. 1A). Mutant HCR6-Fse wasconstructed by digestion of the HCR6-Rz clone with FseI (at nucleotide3379; Takara), deleted using T4 polymerase (Takara). This mutationintroduces a stop codon at nucleotide 3606. Mutant HCR6-Age was con-structed by digestion of the HCR6-Rz clone with AgeI (at nucleotide 155;Nippon Gene) and NotI (at nucleotide 1967; Takara) and filled in with T4polymerase (see Fig. 4A). HCR6-CN5 was constructed by removal of the5�-untranslated region (nucleotides 1–155).

Additional expression was performed by further transfection of thepCAG-PURO-Mer-Cre-Mer vector (see Fig. 1B), the addition of 4-hy-droxytamoxifen (100 nM), or infection with the AxCANCre virus (2).AxCANCre was prepared by inserting the structural genes of Cre re-combinase into adenovirus E1A- and E1B-deleted regions under thecontrol of the CAG promoter. The control adenoviral vector AxCAw1lacks these inserted genes.

The pCAG-PURO-Mer-Cre-Mer vector was constructed from pAN-Mer-Cre-Mer (27) and pCAG-PURO, for which the puromycin gene wasderived from pBabe-PURO under the control of the CAG promoter (28).The dihydrofolate reductase (DHFR)-luciferase reporter plasmid hasthe DHFR promoter in the pGL3 vector (Promega).

Cells and Reagents—HepG2 cells were transfected with HCR6-Rz,HCR6-Fse, HCR6-Age, and HCR6-CN5 clones using a modified calciumphosphate method and selected with G418 (800 �g/ml bioactive; Invitro-gen). Cell lines Rz2-8, Rz2-18, Rz2-22, Rz24, Fse28, Age8, and CN5-1were established (Table I). These cells were further transfected withpCAG-PURO-Mer-Cre-Mer DNA and selected with puromycin (1.0 �g/ml). RzM6 and RzM13 were derived from Rz2-18, and RzM24 wasderived from Rz24. RzM2-8 and RzM2-9 were derived from Rz2-8 andRz2-9, respectively. FseM28 and AgeM8 were derived from Fse28 andAge8, respectively. Switching expression was performed by the additionof 4-hydroxitamoxifen (100 nM) for 4 days or by infection of CN5-1 withthe AxCANCre virus (multiplicity of infection of 20). For characteriza-tion of cell growth speed, 2 � 106 or 4.7 � 106 cells were plated onto 75-or 175-cm2 Falcon tissue culture bottles and passaged every 4 days for44 days. Cell numbers were counted at every time point of passage, andaverages of three experiments were calculated.

Immunoblot Analysis and Core Protein Quantitation—HCV-express-ing cells (1 � 106) were lysed with 100 �l of lysis buffer A (1% SDS, 0.5%Nonidet P-40, 0.15 M NaCl, 0.5 mM EDTA, 1 mM dithiothreitol, and 10mM Tris, pH 7.4), and 30 �g of total protein was electrophoresed onSDS-polyacrylamide gel and transferred to a polyvinylidene difluoridemembrane (Immobilon-P, Millipore Corp.). HCV proteins were detectedusing the originally established anti-core (515) (29), anti-E1 (384),anti-E2 (544), anti-NS4A (C14II-2-1), anti-NS4B (4B52), anti-NS5A(5A32), and anti-NS5B (5B14) monoclonal antibodies or rabbit anti-NS3(R212) and anti-NS4A/B (RR10) polyclonal antibodies. HCV core pro-tein was quantitated in cell lysates using the fluorescent enzyme-linkedimmunosorbent assay (30). Antibodies against p53 (Novo Castra);p21WAF1/CIP1 and EB1 (Transduction Laboratories); RhoA, interferon

(IFN) regulatory factor-1, Rho guanine nucleotide dissociation inhibi-tor, p27KIP1, p16INK, cyclin A, cyclin D1, cyclin E, CDK2, and CDK4(Santa Cruz Biotechnology); phospho-CDC2 (New England BiolabsInc.); actin (Roche Applied Science); and Rb (Pharmingen) were pur-chased and utilized according to the manufacturers’ protocols.

ATLAS cDNA Array Assay—Using 2–5 �g of total RNA from Rz2-18,RzM6, and M13 cells with or without 4-hydroxytamoxifen treatment,32P-labeled cDNA probes were synthesized; cultured for 0, 12, and 44days, respectively; and hybridized to filters (ATLAS human cancer cDNAexpression array, Clontech) according to the manufacturer’s protocol.After extensive washing, hybridization intensities were quantitated usinga Fuji BAS 2000 image intensifier. Values for each regulatory gene werestandardized to the average for nine housekeeping genes.

Fluorescence-activated Cell Sorter (FACS) Analysis—Cells (0.5–1 �106) in the exponential growth phase were trypsinized and stainedusing propidium iodide (Cycle TESTTM Plus DNA reagent kit, BDBiosciences) according to the manufacturer’s protocol. The population ofcells in each cell cycle phase was characterized using Modifit LT soft-ware in FACSCalibur (BD Biosciences) by counting 10,000 cells.

Anchorage-independent Growth Assay—Soft agar dishes were pre-pared using an underlayer of 0.5% low melting point agarose (FMCCorp.) in Dulbecco’s modified Eagle’s medium (Nissui) supplementedwith 20% fetal calf serum (Invitrogen). Rz-Hep, Fse-Hep, Age-Hep, andCN5-Hep cell lines with and without HCV genome expression werepassaged, and 103 cells were plated in the same medium containing0.3% agarose in each well of 6-well Falcon plates. One day after plating,1 ml of Dulbecco’s modified Eagle’s medium was supplemented. Thenumber of colonies (diameter � 0.1 mm) was determined after 10 days.The results represent the mean of three experiments performed induplicate.

E2F Promoter Assay—E2F activity was detected using the DHFRpromoter-luciferase reporter plasmid subcloned into the pGL3 vector(31). Plasmid DNA (100 ng/reaction) was transfected into each cellusing a modified calcium phosphate method in a 24-well Falcon plate.Each reaction was standardized by the Renilla luciferase reporter plas-mid (1 ng/reaction). Luciferase activity was measured using a dualluciferase reporter assay system (Promega) according to the manufac-turer’s protocol. The -fold increase ratio was calculated by division ofeach relative luciferase activity in cells prior to HCV expression. Theresults represent the mean of 3 wells from two experiments.

CDK Assay—Cells (1 � 107 for CDK4/6 and 2.5 � 106 for CDK2) werelysed as described previously (51). In brief, 1 ml of lysis buffer B (50 mM

HEPES, pH 7.5, 250 mM NaCl, 1 mM EDTA, 0.1% Tween 20, 1 mM

aprotinin, 1 �g/ml leupeptin, 0.1 mM phenylmethylsulfonyl fluoride, 10mM �-glycerophosphate, 0.1 mM NaVO4, 1 mM NaF, and 0.1 mM sodiumorthovanadate) was sonicated twice for 10 s and then centrifuged at15,000 rpm for 10 min. Supernatants were pretreated either with 10 �lof 50% protein G-Sepharose (Amersham Biosciences) for CDK4/6 orwith 50 �l of 50% protein A-agarose (Bio-Rad) for CDK2 at 4 °C for 1 h.Supernatants were reacted with 20 �l of 50% protein G-Sepharoseprecoated with 2 �g of anti-cyclin D1 monoclonal antibody (DCS-11,BIOMOL Research Labs Inc.) or with 30 �l of protein A-agarose withrabbit anti-CDK2 polyclonal antibody (Santa Cruz Biotechnology) at4 °C for 3–6 h. Immune complexes were precipitated and washed fourtimes with lysis buffer B and twice with 50 mM HEPES, pH 7.5,containing 1 mM dithiothreitol. Immunoprecipitates were incubated at30 °C for 60 min with 25 �l of kinase buffer (50 mM HEPES, pH 7.5, 10mM MgCl2, 1 mM dithiothreitol, 2.5 mM EGTA, 10 mM �-glycerophos-phate, 0.1 mM sodium orthovanadate, 1 mM NaF, and 20 �M ATP), 10�Ci of [�-32P]ATP (6000 Ci/mmol; PerkinElmer Life Sciences), andeither 0.5 �g of truncated Rb (QED Bioscience) for CDK4/6 or 1 �g ofhistone H1 (Roche Applied Science) for CDK2. Reaction products wereelectrophoresed on 12% SDS-polyacrylamide gel and exposed to x-rayfilm or a Fuji BAS 2000 phospho-imager plate.

Stability Assay for p21WAF1/CIP1 and Rb Proteins—The half-lives ofthe p21WAF1/CIP1 and Rb proteins were characterized after treatmentwith cycloheximide (10 �g/ml) for the indicated times at 37 °C as de-scribed previously (32). The p21WAF1/CIP1 and Rb protein expressionlevels were characterized by Western blotting as described above.

Quantitation of 2�–5�-Oligoadenylate Synthetase, PKR, IFN, andHCV Negative-strand RNA—Quantitation of 2�–5�-oligoadenylate syn-thetase, the double-stranded RNA (dsRNA)-dependent kinase PKR,and HCV negative-strand RNA was performed as described previously(33). The primers and probes for 2�–5�-oligoadenylate synthetase were5�-CTCAGAAATACCCCAGCCAAATC-3� (sense primer), 5�-GTGGTG-AGAGGACTGAGGAA-3� (antisense primer), and 5�-CCAGGTCAGCG-TCAGATCGGCCTC-3� (probe); and those for PKR were 5�-CCTGTCC-TCTGGTTCTTTTG-3� (sense primer), 5�-CATGTCAGGAAGGTCAAA-

2 Y. Amako, A. Katsume, K. Tsukiyama-Kohara, K. Satoh, Y.Hayashi, N. Funata, and M. Kohara, submitted for publication.

Full Genome HCV Up-regulates the CKI-CDK-Rb-E2F Pathway14532

by guest on April 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

FIG. 1. Structure of HCR6-Rz vectors, conditional expression of full genome HCV RNA and proteins, and chronological changesin cell growth properties upon HCR6-Rz gene expression. A, the HCR6-Rz clone contains full-length HCV cDNA surrounded by ribozyme(Rz; gray box) and hepatitis D virus ribozyme (HDV-Rz; white box) sequences under the control of the CAG promoter (CAG-Pro.; cytomegalovirusIE enhancer (gray box) and chicken �-actin promoter (white box)) with a Cre/loxP switching expression cassette comprising the neomycin resistancegene (Neo) as a stuffer surrounded by loxP sequences. The region for the probe used in genomic Southern blotting is also indicated (black box). UTR,untranslated region; C, core. B, shown is the structure of pCAG-PURO-Mer-Cre-Mer. The Cre enzyme (gray box) is surrounded by tamoxifen-binding domains (TBD; black arrows) (27). C, switching expression of full genome HCV RNA was detected by Northern blotting in RzM6 cells. Themolecular size markers are shown in an RNA ladder (Invitrogen). D, HCV protein before (�) and after (�) tamoxifen treatment was detected byWestern blotting in RzM6 cells. E, DNA from serially passaged RzM6 cells was subjected to genomic Southern blot analysis. Rz-Hep cells weretreated with tamoxifen (100 �M), and XbaI- and SalI-digested genomic DNA was probed using the HCV core to the E1 (KpnI-SalI) region (A),yielding an unrecombined 2.9-kbp fragment (day 0) and a recombined 1.8-kbp fragment after 8, 16, 28, and 44 days. F, RzM6 cells were treatedwith tamoxifen and passaged for 44 days at an interval of 4 days. Expression of core protein was quantitated by fluorescent enzyme-linkedimmunosorbent assay (30), and the results were divided by the total protein concentration at each time point.

Full Genome HCV Up-regulates the CKI-CDK-Rb-E2F Pathway 14533

by guest on April 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

TCTGGGTG-3� (antisense primer), and 5�-CTACGTGTGAGTCCCAA-AGCAAC-3� (probe). The primers and probes used for quantitation ofHCV negative-strand RNA, IFN, and glyceraldehyde-3-phosphate de-hydrogenase will be described elsewhere.3 A reporter dye (6-carboxy-fluorescein) was covalently attached to the 5�-end, and a quencher dye(6-carboxytetramethylrhodamine) was joined to the 3�-end of the probesequence using an amino-linked method.

RESULTS

Establishment of Conditional Full Genome HCV-expressingHepatocytes—The conditional expression system of the full ge-nome HCV cDNA clone (HCR6-Rz) was established using theCre/loxP system (Fig. 1A). The precise HCV RNA was trimmedusing the ribozyme sequence at the 5� and 3� termini (Fig. 1, Aand C). Infectivity of the HCR6-Rz cDNA clone was observed insusceptible human liver cell lines and animals.3 Four inde-pendent HCR6-Rz-expressing HepG2 cell lines (Rz2-8, Rz2-18,Rz2-22, and Rz2-24) were established. HCV expression wasinduced using Cre recombinase from a modified Mer-Cre-Merexpression cassette (Fig. 1B) (27) in the presence of 4-hy-droxytamoxifen (100 nM; referred to as tamoxifen below). Thesetamoxifen-inducible Rz-Hep cells were designated RzM6 andM13 (parental strain Rz2-18), RzM2-8 (parental strain Rz2-8),RzM2-9 (parental strain Rz2-9), and RzM24 (parental strainRz24) (Table I). In Rz-Hep cells, expression of full-length HCVRNA (9.6 kb) was induced, and full-length HCV RNA wastrimmed precisely at the 5� and 3� termini by the ribozymesequence, which was confirmed by Northern blotting (Fig. 1C)and 5�-dA tailing, the 3�-oligo(A) tailing cloning method, andsequencing (data not shown). All HCV protein expression wasinduced by tamoxifen treatment (Fig. 1D).

Cell Growth Properties of HCV-expressing Hepatocytes dur-ing Passage—To examine the modification of cell growth prop-erties by HCV, we passaged full genome HCR6-Rz-expressingHepG2 cells (RzM6, RzM13, RzM24, and the respective paren-tal strains) every 4 days for �40 days (Fig. 1, E and F; and Fig.2, A and B). After tamoxifen treatment on day 8, Cre-mediatedtransgene recombination occurred (Fig. 1E). The 2.9-kbp DNAfragment on day 0 and the already recombined 1.8-kbp frag-ment in response to tamoxifen treatment after day 8 (Fig. 1E,8, 16, 28, and 44 days) were detected with XbaI- and SalI-digested genomic DNA by Southern blot analysis (Fig. 1, A andE). HCV protein was persistently expressed in RzM6 cells up toat least 44 days (Fig. 1F), and this expression was quantitatedby fluorescent enzyme-linked immunosorbent assay and West-ern blotting (data not shown). The cell growth ratio of RzM6cells was examined for 44 days (Fig. 2B, panel a); it wasreduced to �50% in 8�12 days after the onset of HCV expres-sion by tamoxifen (core protein expression of 3.4 ng/mg of totalprotein) (Fig. 1F) and then unexpectedly recovered to 100% by20 days and exceeded 100% at 44 days of passage. In contrast,the cell growth of the parental cell line (Rz2-18) did not showany significant change during passage after tamoxifen treat-ment (Fig. 2B, panel a). RzM24 cells with 1.2 ng/mg coreprotein expression on day 8 showed suppression of cell growthto 50% on day 8, but recovered to 100% on day 16, which wasfaster than the recovery seen for RzM6 cells (Fig. 2B, panel b).Suppression of cell growth by HCR6-Rz on day 8 was alsoobserved in RzM13, RzM2-8, and RzM2-9 cells (data notshown). To define the essential HCV genome region for cellgrowth modification, truncated 5�-untranslated region-to-NS2protein (HCR6-Fse), E2-to-3�-untranslated region (HCR6-Age),and core-to-NS5 protein (HCR6-CN5) expression cassettes

were introduced into HepG2 cells (Fig. 2A). HCV core proteinexpression levels in FseM28 and CN5-1 cells were 3.0 and 6.7ng/mg of total protein, respectively (Table I). The expressionlevels of NS5B in AgeM8 and CN5-1 cells were almost equiva-lent to that in RzM6 cells as determined by Western blotting(data not shown). The cell growth properties upon persistentexpression of the truncated HCV genome were examined inHCR6-Fse-, HCR6-Age-, and HCR6-CN5-expressing HepG2cells (Fig. 2A) passaged every 4 days for �44 days (Fig. 2B,panels c–e). HCV protein was persistently expressed in thesecell lines after 44 days of passage (data not shown). The cellgrowth ratio decreased in FseM28 cells on day 8 and recoveredwithin 24 days (Fig. 2B, panel c). No significant change wasobserved in the growth of AgeM8 cells or parental cell lines(Fig. 2B, panel d). Expression of the full-length HCV proteinwas induced by the Cre adenovirus and compared with that incells infected with the Swa adenovirus as a control adenovirusand passaged for 44 days (Fig. 2B). The growth of CN5-1 cellsdid not show any significant change.

To further characterize the process of cell growth modifica-tion by HCV, we performed FACS analysis (Table II). Passagedcells were analyzed on days 0, 8, and 44, as these time pointsoccurred prior to HCV expression, at the time of the mostretarded cell growth, and when the growth ratio had returnedto �100% (Fig. 2A), respectively. Expression of HCR6-Rz andHCR6-Fse retarded the cell cycle at the G0/G1 phase after 8days. At 44 days of passage, the percentage of cells in the G0/G1

phase returned to the level seen prior to HCV expression (TableII). Neither parental nor other cell lines showed any significantchange in cell cycle on days 0, 8, and 44.

Anchorage-independent Growth Activity of HCV-expressingand Passaged Cells—To clarify changes in the character ofHCV gene-expressing cells during passage for 44 days, weexamined the anchorage-independent growth of these cells (Ta-ble III). RzM6 cells formed an average of 7.5 colonies on day 0,a value that had decreased to 2.5 colonies after 8 days of HCVgenome expression. Notably, cells passaged for 44 days withHCR6-Rz expression increased colony numbers to 65.3 (TableIII). Tamoxifen-treated Rz2-18 cells did not show any signifi-cant change in colony formation after 44 days of passage. Ta-moxifen-treated RzM24 cells (another HCR6-Rz-expressing cellline) also showed significantly elevated colony numbers after44 days of passage (Table III). We attempted to further char-acterize the anchorage-independent growth of Fse-Hep, Age-Hep, and CN5-Hep cells, but did not observe any significantincrease in anchorage-independent growth activity during pas-sage in comparison with RzM6 cells (Table III).

Target Host Factors of HCR6-Rz during Passage—To clarifythe molecular mechanisms underlying this modification of an-chorage-independent growth by HCR6-Rz, ATLAS cDNA array

3 M. Kohara, K. Tsukiyama-Kohara, M. Kaito, K. Higashi, Y. Amako,K. Inoue, H. Omori, A. Katsume, T. Itoh, T. Wakita, K. Yasui, J.Ohkawa, K. Taira, Y. Matsuura, and S. Watanabe, manuscript inpreparation.

TABLE IHCV-expressing HepG2 cell lines established in this study

The expression levels of the NS5B protein in AgeM8 and CN5-1 cellswere similar to that observed in RzM6 cells. MCM, Mer-Cre-Mer, ND,not detected.

DNA Parental MCM introduced Core proteina

ng⁄mg

HCR6-Rz Rz2-18 RzM6 3.5RzM13 3.0

Rz24 RzM24 1.2Rz2-8 RzM2-8 0.6Rz2-9 RzM2-9 0.2

HCR6-Fse Fse28 FseM28 3.0HCR6-Age Age8 AgeM8 NDHCR6-CN5 CN5-1 6.7

a Nanograms of core protein/mg of total protein after 4 days of induc-tion.

Full Genome HCV Up-regulates the CKI-CDK-Rb-E2F Pathway14534

by guest on April 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

FIG. 2. Structure of the truncated HCV gene and cell growth properties of Rz-Hep cells during passage. A, structures of HCR6-Fse,HCR6-Age, and HCR6-CN5 expression vectors. CAG-Pro, CAG promoter; Neo, neomycin resistance gene; Rz, ribozyme; UTR, untranslated region;C, core; HDV-Rz, hepatitis D virus ribozyme. B, panel a, growth of Rz2-18 (white bars) and RzM6 (black bars) cells. The growth ratio wasdetermined by dividing the number of cells with tamoxifen by the number of cells without tamoxifen based on the average of three independentexperiments. Panel b, growth properties of Rz24 (white bars) and RzM24 (black bars) cells. Panels c and d, serial passage of Fse28 (white bars) andFseM28 (black bars) cells and Age8 (white bars) and AgeM8 (black bars) cells, respectively. Cell growth properties are indicated by the multiplicityratio of cell growth (division of cell numbers with tamoxifen by those without tamoxifen). The results represent the average of two independentexperiments. Panel e, HepG2 (white bars) and CN5-1 (black bars) cells. The growth ratio was determined by dividing the number of cells infectedwith AxCANCre (cre) by the number of cells infected with AxCAw1 (swa).

Full Genome HCV Up-regulates the CKI-CDK-Rb-E2F Pathway 14535

by guest on April 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

analysis of mRNA from RzM6 and Rz2-18 cells was performedon days 0, 12, and 44. Molecules for which expression levelsshowed significant changes during passage of RzM6 cells uponHCV expression and not during passage of Rz2-18 cells aftertamoxifen treatment were identified as cyclin D1,p21WAF1/CIP1, RhoA, Rho guanine nucleotide dissociation in-hibitor, and EB1. Protein expression levels were examined byWestern blotting in these regulators and in other cell cycleregulators: p53, IFN regulatory factor-1, MDM2, p16INK,p27KIP1, cyclin A, cyclin E, phosphorylated CDC2, and Rb (Fig.3A). Consistent with changes in cell cycle progression, theexpression levels of p21WAF1/CIP1 were up-regulated after 8days of passage and returned to base-line levels after 44 days(Table II). Expression of cyclin D1 was also up-regulated on day8, whereas the Rb protein was dephosphorylated (Fig. 3A),which may indicate that cyclin D1 was inactivated byp21WAF1/CIP1 (34, 35). In particular, in RzM6 cells passaged today 44, expression of hyperphosphorylated Rb protein wasincreased more dramatically than in cells on days 0 and 8(Fig. 3A).

After the onset of HCV protein expression, the cell growthratio decreased in FseM28 cells on day 8. In contrast, no sig-nificant changes were observed in AgeM8 cells (Fig. 2B). FACSanalysis revealed that the cell cycle progression of FseM28 cellswas decreased after 8 days of expression (Table II). However,the expression levels of the p21WAF1/CIP1 protein did not showany significant difference in either FseM28 or AgeM8 cells onday 8 (Fig. 3B). These results indicate that cell cycle retarda-tion by HCV structural protein does not mediate p21WAF1/CIP1.Moreover, the phosphorylation status of Rb was not acceleratedafter passage for 44 days in FseM28 or AgeM8 cells comparedwith RzM6 cells (Fig. 3, A and C). Therefore, we further char-acterized the events that occurred within 44 days of passage inRzM6 cells.

Examination of HCV Replication in HCR6-Rz-expressingCells—HCV possesses positive-strand RNA as a viral genomeand produces negative-strand RNA as a mold only for replica-tion. The quantity of negative-strand viral RNA is thus propor-tional to viral replication. Negative-strand RNA synthesis was

detected by strand-specific real-time detection-PCR (as de-scribed under “Experimental Procedures”) in RzM6 cells onboth days 8 and 44, but was not observed in Rz2-18 cells (Fig.4A) or in FseM28 or AgeM8 cells (data not shown). IFN induc-tion was then examined and was shown to be produced effi-ciently only in RzM6 cells on day 8 (Fig. 4B). Downstream2�–5�-oligoadenylate synthetase and PKR mRNA expressionlevels increased on day 8 and decreased to base-line levels after44 days in RzM6 cells (Fig. 4, C and D). The activity of PKR wasinitially increased on day 8 and decreased below base-linelevels after 44 days in these cells (Fig. 4E). Thus, replication ofHCV might produce negative-strand RNA for formation ofdsRNA, with the subsequent production of IFN and activationof downstream signaling pathways in HCR6-Rz-expressingcells on day 8.

Activation of E2F and CDK by HCR6-Rz—Replication ofHCV and activation of the IFN pathway were observed only inRz-Hep cells. However, the pathways that HCR6-Rz mightmodify to cause acceleration of Rb hyperphosphorylation afterpassage for 44 days remained unclear. A previous study re-ported that CKI, CDK, and Rb regulatory pathways play sig-nificant roles in many types of tumor formation, including HCC(36). The downstream target of p21WAF1/CIP1 was thereforecharacterized. The p21WAF1/CIP1 protein inhibits the activity of

TABLE IICell cycle progression of Rz-Hep, Fse-Hep, and Age-Hep cells

The percentages indicate the population of cells in each cell cyclephase. The results represent the means of three experiments performedin duplicate.

Passage G0/G1 S G2/M

% % %

Rz2-18Day 0 56.2 � 4.7 35.7 � 7.1 8.1 � 2.4Day 8 57.9 � 2.8 32.9 � 4.3 9.2 � 1.5Day 8 60.1 � 2.9 31.6 � 8.2 8.3 � 5.3

RzM6Day 0 57.8 � 0.3 36.2 � 1.7 6.0 � 1.4Day 8 70.4 � 1.5 20.0 � 0.0 9.6 � 1.5Day 44 53.2 � 1.1 33.7 � 2.0 13.1 � 3.1

Fse28Day 0 57.8 � 3.1 34.0 � 0.1 8.2 � 3.2Day 8 57.8 � 3.0 32.8 � 2.3 9.4 � 0.8Day 44 59.3 � 1.4 33.9 � 3.9 6.8 � 2.5

FseM28Day 0 55.8 � 0.1 34.1 � 0.1 10.1 � 0.1Day 8 70.7 � 3.7 19.2 � 0.3 10.1 � 4.0Day 44 53.6 � 2.5 36.0 � 2.0 10.4 � 0.5

Age8Day 0 56.6 � 6.8 36.0 � 7.3 7.4 � 0.9Day 8 55.8 � 10.2 38.3 � 10.6 5.9 � 0.4Day 44 56.8 � 6.3 36.6 � 8.0 6.6 � 1.7

AgeM8Day 0 55.5 � 7.6 36.3 � 7.7 8.2 � 0.1Day 8 58.2 � 10 34.8 � 9.0 7.0 � 1.1Day 44 56.5 � 6.9 35.4 � 7.2 8.1 � 0.3

TABLE IIIColony formation in HCV-expressing HepG2 cells

The results represent the means of three experiments performed induplicate.

PassageColony formation

No. Size

mm

Rz2-18Day 0 9.2 � 3.2 0.9 � 0.1Day 8 6.2 � 2.4 0.9 � 0.1Day 44 8.4 � 6.0 1.0 � 0.4

RzM6Day 0 7.5 � 4.0 0.9 � 0.4Day 8 2.5 � 1.7 0.4 � 0.4Day 44 65.3 � 11 2.0 � 0.6

Rz24Day 0 13.0 � 0.5 0.8 � 0.1Day 44 13.0 � 0.5 1.0 � 0.4

RzM24Day 0 10.0 � 0.01 0.8 � 0.1Day 44 40.0 � 0.8 1.0 � 0.4

Fse28Day 0 10.0 � 0.8 0.7 � 0.3Day 8 16.0 � 1.4 0.8 � 0.3Day 44 13.3 � 3.1 0.7 � 0.1

FseM28Day 0 11.7 � 1.7 0.8 � 0.1Day 8 1.7 � 0.9 0.5 � 0.2Day 44 11.3 � 2.5 0.8 � 0.4

Age8Day 0 13.7 � 1.7 0.8 � 0.2Day 8 11.3 � 1.2 0.9 � 0.3Day 44 14.0 � 2.9 0.7 � 0.3

AgeM8Day 0 12.3 � 4.2 0.6 � 0.1Day 8 13.3 � 4.2 0.9 � 0.3Day 44 13.7 � 1.9 0.8 � 0.3

Hep-SwaDay 44 14.4 � 4.0 0.8 � 0.1

Hep-CreDay 44 22.0 � 6.0 0.8 � 0.4

CN5-1-SwaDay 0 10.0 � 5.0 0.7 � 0.3Day 44 13.0 � 1.0 0.8 � 0.1

CN5-1-CreDay 0 11.7 � 1.7 0.8 � 0.1Day 44 11.3 � 2.5 1.1 � 0.1

Full Genome HCV Up-regulates the CKI-CDK-Rb-E2F Pathway14536

by guest on April 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

FIG. 3. Modification of cell growth regulatory proteins by HCV. A, protein expression levels of cell growth regulatory genes on days 0, 8,and 44 of passage in exponentially growing Rz2-18 and RzM6 cells. The results are representative data from three experiments. IRF-1, IFNregulatory factor-1; Rho-GDI, Rho guanine nucleotide dissociation inhibitor. B, expression of the p21WAF1/CIP1 protein in Fse28, FseM28, Age8, andAgeM8 cells in the exponential cell growth stage on days 0, 8, and 44 of passage as determined by Western blotting. The results are representativepanels from two independent experiments. C, detection of the Rb protein by Western blotting in Rz-Hep, Fse-Hep, and Age-Hep cells in theexponential growth phase. The results from SDS-PAGE (6.5%) show representative data from two experiments.

Full Genome HCV Up-regulates the CKI-CDK-Rb-E2F Pathway 14537

by guest on April 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

CDK4/cyclin D and CDK2/cyclin E, resulting in hypophospho-rylation of Rb and suppression of E2F activity (37, 38). Toassess the possibility that these pathways are modified byHCR6-Rz expression, E2F activity was characterized usingthe DHFR-luciferase reporter plasmid. In RzM6 cells, E2Factivity showed a 0.6-fold decrease on day 8 and a 1.8-foldincrease after 44 days of passage, whereas in Fse-Hep andAge-Hep cells, no significant changes were observed (Fig.5A). This Rb-E2F pathway was thus suppressed on day 8and enhanced by 1.8-fold by day 44 in RzM6 cells (Figs. 3and 5A).

To clarify the hyperphosphorylation mechanism of Rb, wecharacterized CDK4/6 and CDK2 activities during passage ofRzM6 cells (Fig. 5B). The kinase activity of CDK4/6, which wasquantitated using a Fuji BAS 2000 phospho-imager, was 69%on day 8 and 139% on day 44 compared with day 0. CDK2activity was suppressed to 20% on day 8 and increased to 222%on day 44 compared with day 0. In contrast, in the HCV-non-expressing parental Rz2-18 cells, CDK4/6 and CDK2 activitiesdid not show any significant change during passage (Fig. 5B).The protein expression levels of CDK4 and CDK2 likewise didnot show any significant change in either RzM6 or Rz2-18 cells(Fig. 3A).

The stability of p21WAF1/CIP1 is reportedly decreased by ex-pression of HCV proteins, and degradation of hyperphosphory-lated Rb is apparently accelerated in HCC (39). The stability ofthe p21WAF1/CIP1 and Rb proteins was therefore characterizedat the indicated times after cycloheximide treatment (Fig. 5C).The results for the half-life experiment with the p21WAF1/CIP1

and Rb proteins did not show any significant differences be-tween HCV protein-non-expressing cells (0 day) and HCV pro-tein-expressing cells passaged for 44 days. Activation of E2Fwas therefore predominantly caused by activation of CDKs andthe resulting hyperphosphorylation of Rb.

DISCUSSION

HCV-induced cell growth modifications have not yet beenfully clarified because of the lack of an efficient in vitro infec-tion system. This study therefore established a Cre/loxP condi-tional expression system for the full-length HCV genome,which is self-trimmed by a double ribozyme. This enabled us toreproduce the state of hepatocytes persistently infected withHCV. As a result, passage of HCR6-Rz was found to causeup-regulation of anchorage-independent growth and the CDK-Rb-E2F pathway.

The Rb protein became hyperphosphorylated (Fig. 3A), andanchorage-independent growth was accelerated in full genomeHCV RNA-expressing hepatocytes that were passaged for 44days (Table III). Degradation of hyperphosphorylated Rb isreportedly accelerated and E2F is activated, and both correlateclosely with hepatocyte transformation (39). In fact, RzM6 cellspassaged for 44 days displayed accelerated tumorigenicity innude mice (data not shown).

In the case of RzM6 cells, the stability of p21WAF1/CIP1 andRb did not change; and thus, the increased hyperphosphoryla-tion of Rb was predominantly caused by activation of CDKs(Fig. 5). Recent findings have suggested that CDK4/6 is respon-sible only for phosphorylation of Rb and that CDK2 is neces-sary for hyperphosphorylation of Rb (35, 40). In HCR6-Rz-expressing cells, both CDKs were activated, particularly CDK2(Fig. 5B). Acceleration of tumorigenicity by HCR6-Rz may becaused during passage of growth-arrested cells, which mightinduce the disruption of cyclin-CDK-CKI complexes, therebyresulting in the activation of CDKs. A previous study reportedthat the papilloma virus oncoprotein E7 abrogates the inhibi-tory effect of p21WAF1/CIP1 on cyclin E-CDK2 complex activitywithout influencing p21WAF1/CIP1 expression levels (41). The

E8 protein from bovine papilloma virus reportedly acceleratesanchorage-independent growth by activation of cyclin A andCDK2 and deregulation of p27KIP1 (42). This was not observedin RzM6 cells after 44 days of passage (Fig. 3A).

Hyperphosphorylation of Rb results in the release of E2F-1,allowing the cell cycle to progress from G1 to S phase (43, 44) byovercoming restriction point and cellular transformation via co-operation with other oncogenes. Recent reports have noted theabsence or down-regulation of CKI (p16INK, p21WAF1/CIP1, andp27KIP1) expression in most HCCs (36) and have described asignificant correlation between hyperphosphorylation of Rb andHCC (39, 45). These results shed light on the mechanism respon-sible for the high incidence of HCC in HCV-infected patients.

After 8 days of passage, p21WAF1/CIP1 expression was acceler-ated in full genome HCV RNA (HCR6-Rz)-expressing hepato-cytes, and CDK4/6 and CDK2 activities and phosphorylation ofRb were suppressed (Figs. 3A and 5B). These events introducedcell growth retardation at the G0/G1 phase (Table II). AlthoughHCR6-Fse expression induced suppression of cell growth after 8days, no up-regulation of p21WAF1/CIP1 was observed during pas-sage (Fig. 3B). HCR6-Age expression did not influence cellgrowth during passage. This may indicate that HCV structuralprotein modifies another pathway to suppress cell growth ratherthan up-regulation of p21WAF1/CIP1. The precise pathway modi-fied by HCR6-Fse is now under investigation.

In RzM6 cells, up-regulation of RhoA, Rho guanine nucle-otide dissociation inhibitor, and EB1 mRNA transcriptionwas observed, but no increase in protein expression levelswas apparent (Fig. 3A). These inconsistencies might be due tothe existence of other modifications in post-transcriptionalsteps, and the exact reasons are currently under investiga-tion. Expression of p21WAF1/CIP1 was accelerated at the tran-scriptional level as detected by Northern blotting (1.3-fold)and p21WAF1/CIP1 promoter-luciferase reporter assay (3.5-fold) after 8 days of HCR6-Rz expression (data not shown).Modification of p21WAF1/CIP1 expression was not observed inHCR6-Fse- or HCR6-Age-expressing cells. The precise mech-anism underlying modification of p21WAF1/CIP1 expression byHCR6-Rz is not known at present. One possible inducer isIFN, given that it inhibits cell growth by transcriptionalup-regulation of p21WAF1/CIP1 (46–48). HCV negative-strandRNA is the replicative intermediate for HCV and was pro-duced only in HCR6-Rz-expressing cells, an effect that wasconfirmed by strand-specific real-time detection-PCR (Fig.4A). The HCV negative-strand can form dsRNA with positive-strand viral RNA and can stimulate production of IFN-� andIFN-� by activation of mitogen-activated protein kinases andIFN regulatory factor-3 (49). Consistent with these results,expression of IFN-� and IFN-� mRNAs was increased at 8days of passage and returned to basal levels in RzM6 cellsafter 44 days of passage (Fig. 4B). HCV dsRNA still existed inRzM6 cells passaged for 44 days (Fig. 4A), but IFN-�, PKR,2�5�-oligoadenylate synthetase, and p21WAF1/CIP1 expressionreturned to basal levels (Figs. 3A and 4, B–E). The existenceof dsRNA might thus play a significant role in activation ofCDKs in these cells. The detailed mechanisms underlying theactivation of the CDK-Rb-E2F pathway during passage ofHCR6-Rz-expressing cells for 44 days are now underinvestigation.

Synthesis of HCV negative-strand RNA was detected inHCR6-Rz-expressing and passaged cells, but not in truncatedHCV-expressing cells (Fig. 4A) (data not shown). The HCR6-Age and HCR6-CN5 constructs possess the NS5B protein,which is predicted to encode RNA polymerase. However, neg-ative-strand RNA synthesis was not detected. Frieb et al. (50)recently reported that the 5�-untranslated region sequences of

Full Genome HCV Up-regulates the CKI-CDK-Rb-E2F Pathway14538

by guest on April 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

nucleotides 1–125 and 296–341 are important for replication ofthe HCV replicon. HCR6-Age lacks nucleotides 196–341, soproduction of replicative forms in Age-Hep cells might be belowdetectable levels.

The present results indicate that passage of full genomeHCV-expressing cells activates the CKI-CDK-Rb-E2F path-way. Future studies on primary hepatocytes are necessary todelineate the effects of full genome HCV on the CKI-CDK-

FIG. 4. Detection of HCV negative-strand RNA, IFN-�, PKR, and 2�–5�-oligoadenylate synthetase mRNAs. HCV negative-strand RNA(A), IFN-� (B), PKR (C), and 2�–5�-oligoadenylate synthetase (2�-5�OAS; D) mRNAs were quantified using real-time detection-PCR on days 0, 8,and 44. Copies of each RNA/total RNA (�g) are indicated. The mean of three separate experiments is shown. Error bars indicate S.D. PKR kinaseactivity was detected in Rz-Hep cells, and the ratio of radioactivity on day 0 was taken as 100% (E).

Full Genome HCV Up-regulates the CKI-CDK-Rb-E2F Pathway 14539

by guest on April 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

FIG. 5. Activation of E2F and CDK occurs in RzM6 cells after 44 days of passage. A, DHFR promoter activity in RzM6, Fse-Hep, andAge-Hep cells. Each reaction was standardized using the Renilla luciferase reporter plasmid (relative luciferase activity). The ratio of the -foldincrease was calculated by the division of each relative luciferase activity in cells prior to HCV expression. -Fold activation of relative luciferaseactivity is given as a ratio. The results represent the average of 3 wells in two independent experiments. B, CDK2 and CDK4/6 activities in Rz2-18and RzM6 cells on days 0, 8, and 44. NRS, immunoglobulin derived from normal rabbit serum; NMS, immunoglobulin derived from normal mouseserum. The percentage of each kinase activity relative to day 0 as quantitated using a phospho-imager plate is shown. tr.Rb, truncated Rb. C,stability of the p21WAF1/CIP1 and Rb proteins characterized by Western blotting after cycloheximide treatment for the indicated times.

Full Genome HCV Up-regulates the CKI-CDK-Rb-E2F Pathway14540

by guest on April 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

Rb-E2F pathway. This may represent a novel mechanism forthe transformation of hepatocytes during persistent HCVinfection.

Acknowledgments—We thank G. Perlemuter and C. Brechot forthoughtful comments and suggestions, I. Saito for the suggestion toutilize the Cre/loxP system, B. Vogelstein for providing p21WAF1/CIP1

promoter-luciferase plasmid DNA, and Y. Hayasi and K. Tomita forpathological analyses.

REFERENCES

1. Saito, I., Miyamura, T., Ohbayashi, A., Harada, H., Katayama, T., Kikuchi, S.,Watanabe, Y., Koi, S., Onji, M., Ohta, Y., Choo, Q., Houghton, M., and Kuo,G. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 6547–6549

2. Kanegae, Y., Lee, G., Sato, Y., Tanaka, M., Nakai, M., Sakaki, T., Sugano, S.,and Saito, I. (1995) Nucleic Acids Res. 23, 3816–3821

3. Matsumoto, M., Hsieh, T. Y., Zhu, N., VanArsdale, T., Hwang, S. B., Jeng,K. S., Gorbalenya, A. E., Lo, S. Y., Ou, J. H., Ware, C. F., and Lai, M. M.(1997) J. Virol. 71, 1301–1309

4. Zhu, N., Khoshnan, A., Schneider, R., Matsumoto, M., Dennert, G., Ware, C.,and Lai, M. M. (1998) J. Virol. 72, 3691–3697

5. Marusawa, H., Hijikata, M., Chiba, T., and Shimotohno, K. (1999) J. Virol. 73,4713–4720

6. Machida, K., Tsukiyama-Kohara, K., Seike, E., Tone, S., Shibasaki, F.,Shimizu, M., Takahashi, H., Hayashi, Y., Funata, N., Taya, C., Yonekawa,H., and Kohara, M. (2001) J. Biol. Chem. 276, 12140–12146

7. Ray, R. B., Lagging, L. M., Meyer, K., and Ray, R. (1996) J. Virol. 70,4438–4443

8. Chang, J., Yang, S. H., Cho, Y. G., Hwang, S. B., Hahn, Y. S., and Sung, Y. C.(1998) J. Virol. 72, 3060–3065

9. Jin, D.-Y., Wang, H.-L., Zhou, Y., Chun, A. C. S., Kibler, K. V., Hou, Y.-D.,Kung, H.-F., and Jeang, K.-T. (2000) EMBO J. 19, 729–740

10. Yoshida, T., Hanada, T., Tokuhisa, T., Kosai, K., Sata, M., Kohara, M., andYoshimura, A. (2002) J. Exp. Med. 196, 641–653

11. Sakamuro, D., Furukawa, T., and Takegami, T. (1995) J. Virol. 69, 3893–389612. Park, J. S., Yang, J. M., and Min, M. K. (2000) Biochem. Biophys. Res.

Commun. 267, 581–58713. Moriya, K., Fujie, H., Shintani, Y., Yotsuyanagi, H., Tsutsumi, T., Ishibashi,

K., Matsuura, Y., Kimura, S., Miyamura, T., and Koike, K. (1998) Nat. Med.4, 1065–1067

14. Honda, A., Arai, Y., Hirota, N., Sato, T., Ikegaki, J., Koizumi, T., Hatano, M.,Kohara, M., Moriyama, T., Imawari, M., Shimotohno, K., and Tokuhisa, T.(1999) J. Med. Virol. 59, 281–289

15. Wakita, T., Taya, C., Katsume, A., Kato, J., Yonekawa, H., Kanegae, Y., Saito,I., Hayashi, Y., Koike, M., and Kohara, M. (1998) J. Biol. Chem. 273,9001–9006

16. Lerat, H., Honda, M., Beard, M. R., Loesch, K., Sun, J., Yang, Y., Okuda, M.,Gosert, R., Xiao, S. Y., Weinman, S. A., and Lemon, S. M. (2002) Gastroen-terology 122, 352–365

17. Shibata, M., Morizane, T., Uchida, T., Yamagami, T., Onozuka, Y., Nakano,M., Mitamura, K., and Ueno, Y. (1998) Lancet 351, 1773–1777

18. Lohmann, V., Korner, F., Koch, J., Herian, U., Theilmann, L., and Barten-schlager, R. (1999) Science 285, 110–113

19. Blight, K. J., McKeating, J. A., Marcotrigiano, J., and Rice, C. M. (2003)J. Virol. 77, 3181–3190

20. Hsu, I. C., Tokiwa, T., Bennett, W., Metcalf, R. A., Welsh, J. A., Sun, T., andHarris, C. C. (1993) Carcinogenesis 14, 987–992

21. Lee, T. K., Lau, T. C., and Ng, I. O. (2002) Cancer Chemother. Pharmacol. 49,78–86

22. Knowles, B. B., Howe, C. C., and Aden, D. P. (1980) Science 209, 497–49923. Zauberman, A., Oren, M., and Zipori, D. (1997) Oncogene 15, 1705–171124. Niwa, H., Yamamura, K., and Miyazaki, J. (1991) Gene (Amst.) 108, 193–19925. Shimayama, T., Nishikawa, S., and Taira, K. (1995) Biochemistry 34,

3649–365426. Suh, Y. A., Kumar, P. K., Nishikawa, F., Kayano, E., Nakai, S., Odai, O.,

Uesugi, S., Taira, K., and Nishikawa, S. (1992) Nucleic Acids Res. 20,747–753

27. Zhang, Y., Riesterer, C., Ayrall, A. M., Sablitzky, F., Littlewood, T. D., andReth, M. (1996) Nucleic Acids Res. 24, 543–548

28. Tsukiyama-Kohara, K., Poulin, F., Kohara, M., DeMaria, C. T., Cheng, A., Wu,Z., Gingras, A.-C., Katsume, A., Elchebly, M., Spiegelman, B. M., Harper,M.-E., Tremblay, M. L., and Sonenberg, N. (2001) Nat. Med. 7, 1128–1132

29. Yasui, K., Wakita, T., Tsukiyama-Kohara, K., Funahashi, S. I., Ichikawa, M.,Kajita, T., Moradpour, D., Wands, J. R., and Kohara, M. (1998) J. Virol. 72,6048–6055

30. Tanaka, T., Lau, J. Y., Mizokami, M., Orito, E., Tanaka, E., Kiyosawa, K.,Yasui, K., Ohta, Y., Hasegawa, A., Tanaka, S., and Kohara, M. (1995)J. Hepatol. 23, 742–745

31. Ikeda, M. A., Jakoi, L., and Nevins, J. R. (1996) Proc. Natl. Acad. Sci. U. S. A.93, 3215–3220

32. Patrick, G. N., Zhou, P., Kwon, Y. T., Howley, P. M., and Tsai, L. H. (1998)J. Biol. Chem. 273, 24057–24064

33. Takeuchi, T., Katsume, A., Tanaka, T., Abe, A., Inoue, K., Tsukiyama-Kohara,K., Kawaguchi, R., Tanaka, S., and Kohara, M. (1999) Gastroenterology 116,636–642

34. Grana, X., and Reddy, E. P. (1995) Oncogene 11, 211–21935. Lundberg, A. S., and Weinberg, R. A. (1998) Mol. Cell. Biol. 18, 753–76136. Hui, A. M., Makuuchi, M., and Li, X. (1998) Hepatogastroenterology 45,

1635–164237. Yin, M. B., Voigt, W., Panadero, A., Vanhoefer, U., Frank, C., Pajovic, S.,

Azizkhan, J., and Rustum, Y. M. (1997) Mol. Pharmacol. 51, 630–63638. Ma, Y., Prigent, S. A., Born, T. L., Monell, C. R., Feramisco, J. R., and

Bertolaet, B. L. (1999) Cancer Res. 59, 5341–534839. Higashitsuji, H., Itoh, K., Nagao, T., Dawson, S., Nonoguchi, K., Kido, T.,

Mayer, R. J., Arii, S., and Fujita, J. (2000) Nat. Med. 6, 96–9940. Harbour, J. W., Luo, R. X., Dei Santi, A., Postigo, A. A., and Dean, D. C. (1999)

Cell 98, 859–86941. Ruesch, M. N., and Laimins, L. A. (1997) J. Virol. 71, 5570–557842. O’Brien, V., and Campo, M. S. (1998) Oncogene 17, 293–30143. Nevins, J. R., Leone, G., DeGregori, J., and Jakoi, L. (1997) J. Cell. Physiol.

173, 233–23644. Weinberg, R. A. (1996) Cell 85, 457–45945. Woitach, J. T., Zhang, M., Niu, C. H., and Thorgeirsson, S. S. (1998) Nat.

Genet. 19, 371–37446. Sangfelt, O., Erickson, S., Einhorn, S., and Grander, D. (1997) Oncogene 14,

415–42347. Sangfelt, O., Erickson, S., Castro, J., Heiden, T., Gustafsson, A., Einhorn, S.,

and Grander, D. (1999) Oncogene 18, 2798–281048. Kirch, H. C., Putzer, B., Brockmann, D., Esche, H., and Kloke, O. (1997) Eur.

J. Biochem. 246, 736–74449. Servant, M. J., Grandvaux, N., and Hiscott, J. (2002) Biochem. Pharmacol. 64,

985–99250. Friebe, P., Lohmann, V., Krieger, N., and Bartenschlager, R. (2001) J. Virol.

75, 12047–1205751. Matsushime, H., Quelle, D. E., Shultleff, S. A., Shibuya, M., Sherr, C. J., and

Kato, J. Y. (1994) Mol. Cell Biol. 14, 2066–2076

Full Genome HCV Up-regulates the CKI-CDK-Rb-E2F Pathway 14541

by guest on April 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

KoharaHoshikawa, Futoshi Shibasaki, Michael Reth, Yohsuke Minatogawa and Michinori

Katsume, Hideko Nuriya, Hiroshi Ohmori, Jun Ohkawa, Kazunari Taira, Yutaka Kyoko Tsukiyama-Kohara, Shigenobu Toné, Isao Maruyama, Kazuaki Inoue, Asao

Virus-expressing CellsActivation of the CKI-CDK-Rb-E2F Pathway in Full Genome Hepatitis C

doi: 10.1074/jbc.M312822200 originally published online January 26, 20042004, 279:14531-14541.J. Biol. Chem. 

  10.1074/jbc.M312822200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/279/15/14531.full.html#ref-list-1

This article cites 51 references, 23 of which can be accessed free at

by guest on April 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from