Association of the human papillomavirus type 16 E7 oncoprotein with the 600-kDa retinoblastoma protein-associated factor, p600

Association of the human papillomavirus type 16 E7 oncoprotein with the 600-kDa retinoblastoma protein-associated factor, p600
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Association of the human papillomavirus type 16 E7oncoprotein with the 600-kDa retinoblastomaprotein-associated factor, p600 Kyung-Won Huh*, Joseph DeMasi † , Hidesato Ogawa ‡§ , Yoshihiro Nakatani ‡ , Peter M. Howley † , and Karl Mu ¨ nger* ¶ *The Channing Laboratory, Brigham and Women’s Hospital and Department of Medicine,  † Department of Pathology, and  ‡ Dana–Farber Cancer Institute,Harvard Medical School, Boston, MA 02115Contributed by Peter M. Howley, June 24, 2005 The human papillomavirus type 16 (HPV-16) E7 gene encodes amultifunctional oncoprotein that can subvert multiple cellularregulatory pathways. The best-known cellular targets of theHPV-16 E7 oncoprotein are the retinoblastoma tumor suppressorprotein pRB and the related pocket proteins p107 and p130.However, there is ample evidence that E7 has additional cellulartargets that contribute to its transforming potential. We isolatedHPV-16E7associatedcellularproteincomplexesbytandemaffinitypurification and mass spectrometry and identified the 600-kDaretinoblastoma protein associated factor, p600, as a cellular targetof E7. Association of E7 with p600 is independent of the pocketproteins and is mediated through the N terminal E7 domain, whichis related to conserved region 1 of the adenovirus E1A protein andimportantly contributes to cellular transformation independentof pRB binding. Depletion of p600 protein levels by RNA interfer-ence substantially decreased anchorage-independent growth inHPV-positive and -negative human cancer cells. Therefore, p600is a cellular target of E7 that regulates cellular pathways thatcontribute to anchorage-independent growth and cellulartransformation. apoptosis    cervical carcinogensis    retinoblastoma tumor suppressor H uman papillomaviruses (HPVs) are small DNA viruses withdouble-stranded circular genomes that exhibit a tropism forepithelial cells. Of the  200 HPVs that have been identified, asubgroup of    30 HPVs specifically infects mucosal tissues.These HPVs are classified as ‘‘low-risk’’ or ‘‘high-risk’’, depend-ing on the clinical prognosis of the lesions they cause. More than99% of cervical carcinomas and   20% of oral cancers areassociated with high-risk HPV infections (reviewed in ref. 1).The HPV E6 and E7 proteins are consistently expressed inHPV-associated cervical cancers, and persistent expression of these viral oncoproteins is necessary for maintenance of thetransformed phenotype. High-risk HPV E7 proteins are suffi-cient to transform NIH 3T3 cells, can cooperate with the rasoncogene to transform primary baby rat kidney cells, extend thelife span of primary human epithelial cells, and in combination with E6 facilitate their immortalization (reviewed in ref. 2).Furthermore, high-risk HPV E7 proteins importantly contributeto malignant progression through induction of centrosome ab-normalities that cause genomic instability, thereby fosteringdevelopment of aneuploidy (reviewed in ref. 3).HPV-16 E7 encodes a small, 98-aa multifunctional protein.The N-terminal portion of HPV-16 E7 shares amino acidsequence similarity to a portion of conserved regions (CR) 1 and2 of the adenovirus E1A protein and related sequences inpolyomavirus large tumor antigens (4, 5). High-risk HPV E7proteins exert their biological activities chiefly by interacting with host cellular protein complexes, thereby dysregulating theirnormal physiological functions. The best-defined cellular targetof HPV E7 proteins is the retinoblastoma tumor suppressorprotein pRB and the related ‘‘pocket proteins’’ p107 and p130(reviewed in ref. 6). High-risk HPV E7 proteins associate withthe pocket proteins and induce their proteasomal degradation(7, 8). The LXCXE motif within the CR2 homology domain of E7 is sufficient for pocket protein binding (9, 10), but additionalsequences located in the immediate N-terminal CR1 homologydomain of E7 are required for pocket protein degradation (8,11), and these sequences are also necessary for the transformingactivities of E7 (12, 13). Experiments with tissue culture andanimal model systems have shown that some of the transformingactivitiesofHPV-16E7areatleastinpartindependentofpocketprotein inactivation (14, 15). In addition, the ability of HPV-16E7 to induce centrosome-associated mitotic abnormalities is alsoindependent of pocket protein inactivation (16). To determinethe biochemical basis for the diverse biological activities of high-risk HPV E7 oncoproteins, we isolated HPV-16 E7 onco-protein-associated cellular protein complexes from cervical car-cinoma cells by tandem affinity purification (TAP) and identi-fied individual protein components by mass spectrometry. Here we report the isolation of p600 as a cellular target of E7 andprovideevidencethatp600maybeinvolvedinregulatingcellularpathways that are necessary for cellular transformation. Materials and Methods Cell Lines.  HeLa S3 suspension cells (obtained from Y.N.) weremaintained in DMEM containing 10% FBS. Large-scale HeLaS3 suspension cultures were grown in Joklik-modified minimumessential medium (GIBCO Invitrogen), containing 10% calf serum, at 37°C with constant stirring. U2OS, IMR90, IMR90-E7(17), and Phoenix cells were maintained in DMEM  10% FBS.CaSki cells were grown in RPMI medium 1640 (GIBCO Invitro-gen)  10% FBS, and NIH 3T3 cells were grown in DMEM  10%calf serum. Generation of Plasmids and HPV E7-Expressing HeLa Cell Lines.  HPVE7 sequences were inserted into the Xho and NotI sites of pOZ-C or pOZ-N retroviral vectors in frame with the N- or theC-terminal FLAG  hemaglutinin (HA) epitope tags (18). Toconstruct pLXSN CE7 and pLXSN NE7, each coding sequence was PCR amplified from pOZ-CE7 and pOZ-NE7, respectively.HeLa cells expressing E7 proteins were generated by retro- viral infection followed by three consecutive rounds of selectionusing IL-2R antibody conjugated to magnetic beads (18). Ex-pression of HPV E7 proteins was verified by immunoblottingand  or immunoprecipitation and immunofluorescence. Immunological Methods.  The following antibodies were used:HPV-16 E7 (8C9, Zymed, South San Francisco, CA; ED17, Freely available online through the PNAS open access option.Abbreviations: HPV, human papillomavirus; CR, conserved region; TAP, tandem affinitypurification; HA, hemagglutinin; shRNA, small hairpin RNA. § Present address: Department of Developmental Biology, National Institute for BasicBiology, Okazaki, 444-8585 Aichi, Japan. ¶ To whom correspondence should be addressed. E-mail:© 2005 by The National Academy of Sciences of the USA 11492–11497    PNAS    August 9, 2005    vol. 102    no. 32  cgi  doi  10.1073  pnas.0505337102  Santa Cruz Biotechnology), pRB (M153, Santa Cruz Biotech-nology), and HA (Y11, Santa Cruz Biotechnology). The p600-specific polyclonal antibody was generated by Y.N. M2 FLAGantibody resin was from Sigma, and HA-antibody resin (HA probeF7)wasfromSantaCruzBiotechnology.HA(HA.11)and3   FLAG peptides were from Babco (Richmond, CA) andSigma, respectively. IL-2R antibody (clone 7G7  B6, UpstateBiotechnology, Lake Placid, NY) was conjugated to Dynalbeads(M-450, Dynal, Great Neck, NY).For immunoprecipitations, cells were lysed by two freeze–thaw cycles (liquid N 2  37°C) in three volumes of 0.3B buffer (20mM Tris  HCl, pH 8.0  0.3 M KCl  5 mM MgCl 2  10% glycerol  0.1% Tween-20  10 mM 2-mercaptoethanol  0.2 mM PMSF)followed by incubation at 4°C for 30 min and centrifugation at16,000   g   at 4°C, for 30 min. The supernatant was recentrifugedat 16,000    g   at 4°C for 10 min followed by preclearing withprotein G PLUS agarose (Santa Cruz Biotechnology). Immu-noprecipitations with primary antibodies were performed for3–4 h at 4°C. Immune complexes were purified by using proteinG PLUS agarose and washed three times with 1 ml of 0.3Bbuffer.For localization of E7 and p600 by confocal fluorescencemicroscopy, CaSki cells grown on coverslips were fixed in 4%paraformaldehyde, 0.025% glutaraldehyde in BRB 80 (80 mMPipes, pH 6.8  1 mM MgCl 2  1 mM EGTA) for 15–20 min at37°C, and rinsed three times with BRB 80 and two times withantibody dilution solution (0.1% Triton X-100  2% BSA in BRB80). Fixed cells were permeabilized in BRB 80 containing 0.1%TritonX-100for10minatroomtemperatureandincubatedwithantibody dilution solution for 30 min at 20°C. Cells wereincubated with rabbit polyclonal p600 antibody (1:1,000) andmonoclonal E7 antibody (1:10) for 2 days at 4°C. After washing with antibody dilution solution and BRB 80 for 30 min each, cells were incubated with FITC-conjugated anti-mouse (1:500) andrhodamine-conjugated anti-rabbit (1:2,000) antibodies for 3 h at20°C. After rinsing with antibody dilution solution and BRB 80,cells were mounted and analyzed. TAPandMassSpectrometry. Cellularproteincomplexesassociated with E7 were isolated from 5–10 liters of stable HeLa suspensioncell lines (18). After a first round of affinity purification on M2FLAG antibody resin (Sigma), proteins were eluted with 0.5mg  ml 3   FLAG peptide. For subsequent purification on HA antibody resin, samples were incubated with HA antibody resin(HA-probe F-7, Santa Cruz Biotechnology) for 3 h at 4°C withrotation. Beads were washed three times with 0.1B buffer (20mM Tris  HCl, pH 8.0  0.1 M KCl  5 mM MgCl 2  10% glycerol  0.1% Tween-20  10 mM 2-mercaptoethanol  0.2 mM PMSF).Protein complexes were eluted with HA peptide (0.5 mg  ml) for30 min at 20°C and separated on SDS  4–12% Bis-Tris poly-acrylamide gradient gels (Invitrogen). Individual bands were visualized by colloidal blue (Bio-Rad) staining, excised, andanalyzed by mass spectrometry at the Taplin Biological MassSpectrometry Facility (Harvard Medical School). Knockdown Experiments.  The human p600-specific small hairpinRNA (shRNA) expression plasmid was generated by using thesequence GCAGTACGAGCCATTCTAC expressed frompRetro  Super (19). A reverse orientation shRNA expression vector CATCTTACCGAGCATGACG was used as a control.ThepRetro  Superbasedmousep600-specificshRNAexpression vector was generated by Y.N. Recombinant p600 shRNA ex-pressing retroviruses were generated by transfecting Phoenix cells using FuGENE 6 (Roche Diagnostics). NIH 3T3 cells wereinfected with p600 shRNA or control shRNA expressing retro- virus and selected in 2   g  ml puromycin. To generate HPV-16E6 and  or E7 expressing populations, selected stable p600 orreverse orientation control shRNA expression vector transducedlines were infected with pLXSN, pLXSN E7, or pLXSN E6  E7retroviral supernatants followed by selection in 500  g  ml G418for   2 weeks. The stable NIH 3T3 cell lines generated weremaintained in puromycin (2   g  ml) and G418 (500   g  ml).Stable p600 and control knockdown U2OS and CaSki cell lines were similarly established after selection with 2   g  ml puromy-cin. Anchorage-Independent Growth Assays.  Cells (2,500 per well of asix-well plate) were suspended in 0.3% Agar Noble (Difco)dissolved in tissue culture medium and layered onto dishescoated with 0.5% Agar Noble. Colony formation was evaluatedin duplicate wells after 2–3 weeks by phase-contrast microscopy. Results Construction and Characterization of HA  FLAG Epitope-TaggedHPV-16 E7 Proteins.  To isolate and identify cellular protein com-plexes associated with the HPV-16 E7 protein in human cervicalcancer cell lines by TAP, HPV-16 E7 coding sequences werecloned into pOZ-N and pOZ-C to generate FLAG  HA taggedHPV-16 E7 at the N (N-E7) or C (C-E7) terminus, respectively.In each case, the epitope tags are separated from the E7 codingsequence by a linker and the FLAG and HA tags are alsoseparated by linkers (Fig. 1  A ). Because some N-terminallytagged HPV-16 E7 proteins are functionally defective (20), weperformed transformation assays in NIH 3T3 cells to assessfunctionality. Consistent with our previous findings, the N-terminally double-tagged HPV-16 E7 protein (N-E7) was im-paired for NIH 3T3 cell transformation, whereas the C-terminally double tagged E7 protein (C-E7) was competent fortransformation at a level similar to the untagged HPV-16 E7protein (Fig. 1  B ). Isolation of Cellular Protein Complexes Associated with HPV-16 E7 inthe HeLa Cervical Carcinoma Cell Line.  With these results in mind, we next generated populations of the HPV-18-positive cervicalcarcinoma cell line HeLa with stable expression of the C-E7 orN-E7 proteins. HeLa cells were chosen for these experimentsbecause they are derived from a cervical cancer, contain aninherently intact retinoblastoma tumor suppressor pathway (21, Fig. 1.  Construction and characterization of epitope-tagged HPV-16 E7proteins.(  A )SchematicrepresentationoftheFLAG  HAdoubletaggedHPV-16E7fusionproteinsconstructedforthisstudy.C-E7andN-E7containFLAG  HAtags fused to their C and N termini, respectively. The amino acid sequencescorresponding to the epitope tags are underlined. ( B ) Transforming activitiesof C-E7 and N-E7 compared to untagged HPV-16 E7 (E7) as determined byinduction of anchorage independent growth in NIH 3T3 cells. Values shownwereobtainedaftersubtractingthenumberofcoloniesobservedwithvectortransduced NIH 3T3 cells (273), even though these colonies were smaller thanthose in HPV oncoprotein expressing cells. Averages and standard deviationsfrom an experiments performed with duplicate plates are shown. Huh  et al.  PNAS    August 9, 2005    vol. 102    no. 32    11493      M     I     C     R     O     B     I     O     L     O     G     Y  22), and the ectopic expression of HPV-16 E7 in these cells canrescue the biological and biochemical defects that are caused byloss of expression of endogenous HPV-18 E7 (23, 24). Theresulting HeLa cell populations expressed the tagged E7 pro-teins at low levels; similar to those detected in the SiHa cervicalcarcinoma cell line where E7 is expressed from integratedlow-copy partial HPV-16 genomes (25–27) (data not shown).Low-level expression of the tagged E7 protein is desirable forthese experiments because it minimizes the possibility of for-mation of aberrant complexes (18). A total of 5–10 liters of C-E7-expressing or control HeLa S3 cells were grown in spinnerflasks, and C-E7-associated protein complexes were purified byconsecutive rounds of immuno-affinity purification using HA and FLAG antibody beads. To control for nonspecific proteinbinding to the antibody columns, we performed the samepurification steps with an equal amount of vector transducedHeLa S3 cells. Protein complexes were resolved on SDS-polyacrylamide gels and visualized by Colloidal Blue staining(Fig. 2  A ). Individual bands were excised, digested with trypsin,and identified by mass spectrometry. These experiments yieldedevidence for  100 different cellular proteins as potential com-ponents of HPV-16 E7 associated host cellular protein com-plexes. Among the proteins identified are previously describedcellulartargetsofE7includingpRB,p107,p130,severalE2FandDP family members, cyclin A, cyclin E, CDC2, and CDK2 (Fig.2  A ), thus clearly supporting the viability of the TAP approach. The p600 Protein Associates with HPV-16 E7 N-Terminal SequencesThat Are Important for Transformation Independent of pRB Binding. Some of the other proteins may presumably represent additionalcellular E7 targets. For this study, we focused on a C-E7associatedproteinbandthatmigratesatamolecularsizeof   188kDa. In multiple experiments, we obtained   100 peptidescorresponding to p600, a 5,183-aa protein (GenBank accessionnumber AF348492; Fig. 2  A ).Because p600 has been identified as a component of pRB-associated protein complexes in the HPV-18-positive HeLacervical carcinoma cells (Y.N., unpublished results), we deter-mined whether association of E7 with p600 may be indirectthrough pRB and, thus, depends on the integrity of the corepRB-binding site of HPV-16 E7. A series of HeLa-based celllines with stable expression of various C-terminally HA   FLAGtagged HPV-16 E7 mutants were generated and their ability toassociate with p600 and pRB was compared by immunoprecipi-tation  immunoblot experiments (Fig. 2  B ). These experimentsshowed that HPV-16 E7 mutants H2P and  6–10 that target theN-terminal E7 CR1 homology domain and retain the ability toassociate with pRB are defective for association with p600 (Fig.2  B ). In contrast the pRB-binding-deficient HPV-16 E7  21–24mutant associated with p600 at levels similar to wild-typeHPV-16 E7 (Fig. 2  B ) Similarly, HPV-16 E7 mutants within aC-terminal domain that is important for cellular transformationindependent of pRB binding and degradation (14) retained thecapacity to associate with p600, albeit at reduced levels (Fig. 2  B ).The cell lines expressing the different E7 mutants containedsimilar p600 protein levels (data not shown).The transformation-defective N-terminally HA   FLAGtagged HPV-16 E7 fusion protein (N-E7), was unable to interact with p600, even though mass spectrometric analyses revealedthat N-E7 efficiently interacted with pRB and other pocketproteins (data not shown). Hence, association of HPV-16 E7 with p600 is independent of the ability to interact with pRB. Low-Risk HPV-6b and HPV-11 E7 Proteins Can Associate with p600. Low-risk HPV-6b and HPV-11 E7 proteins are impaired forcellular transformation, associate with pRB at a decreased effi-ciency, and are defective for inducing its degradation (reviewed inref.6).TodeterminewhethertheHPV-6bandHPV-11E7proteinscan associate with p600, we generated HeLa-based cell lines withstable expression of C-terminally HA   FLAG-tagged HPV-6b orHPV-11 E7 proteins and performed immunoprecipitation  immunoblotexperiments.HPV-6bandHPV-11E7associatedwithp600 as efficiently as HPV-16 E7, whereas consistent with earlierreports (9, 28), HPV-6b and -11 E7 proteins interacted with pRBat a decreased efficiency (Fig. 2 C ). Association of HPV-16 E7 with p600 in the HPV-16-Positive CaSkiCervical Cancer Line.  To evaluate association of untaggedHPV-16 E7 with endogenous p600, we performed coimmu-noprecipitation experiments in the HPV-16 positive CaSkicervical carcinoma cell line. The HPV-negative C33A cervicalcancer cell line was used as a negative control. Immunoblotexperiments revealed that CaSki and C33A cells expressedsimilar amounts of p600. Immunoprecipitation experiments were performed by using HPV-16 E7-specific monoclonalantibodies, and p600 was coprecipitated in the HPV-16-positive CaSki cells but not in HPV-negative C33A cells (Fig.3  A ). Therefore, association of HPV-16 E7 with p600 is de-tected in a cervical carcinoma cell line under conditions of endogenous p600 and HPV-16 E7 expression.To further document HPV-16 E7  p600 association  in vivo , wealso performed confocal immunofluorescence experiments inCaSki cells. HPV-16 E7 and p600 staining was observed incytoplasmic as well as nuclear structures with partial colocal-ization of the HPV-16 E7 and p600 signals (Fig. 3  B ). Association of HPV-16 E7 with p600 Does Not Contribute to E7-Mediated pRB Degradation.  A computer search of the conserveddomain database (29) revealed that the p600 protein contains aRING finger-like structure in its N-terminal domain (amino acidresidues 1,660–1,717). RING fingers are cysteine  histidine- Fig. 2.  Identification of p600 as an interactor of HPV-16 E7 and mapping ofE7 domains necessary for p600 association. (  A ) SDS  4–12% PAGE analysis ofcellular protein complexes associated with C-E7 in HeLa cells after TAP. CdenotesTAPofanequivalentamountofcelllysatesfromcellstransducedwithempty vector. Gel was stained with colloidal blue; protein bands containingC-E7, known HPV-16 E7 associated proteins and p600 are indicated. See textfor details. ( B ) Mapping of the sequences on HPV-16 E7 that are necessary forp600association.HeLacellswithstableexpressionofC-E7andthecorrespond-ingE7mutantsdoubletaggedwithFLAG  HAattheirCterminiweresubjectedto FLAG immunoprecipitation followed by immunoblot analysis for p600( Top ) and pRB ( Middle ) as a control. Expression of the corresponding E7mutants is documented by immunoblot using HA antibody ( Bottom ). Thedifferentpanelsarederivedfromthesamegelandx-rayfilm.( C  )Thelow-riskHPV-6bandHPV-11E7proteinsassociatewithp600asefficientlyasHPV-16E7.HeLa cells with stable expression of the corresponding E7 proteins taggedwithFLAG  HAattheirCterminiweresubjectedtoFLAGimmunoprecipitationfollowedbyimmunoblotanalysisforp600( Top )andpRB( Middle ).Expressionof the corresponding E7 mutants is documented by immunoblot using HAantibody ( Bottom ). 11494    cgi  doi  10.1073  pnas.0505337102 Huh  et al.  containing zinc-binding structures that are also detected in someubiquitin ligases (reviewed in ref. 30). The RING finger in p600ismostcloselyrelatedtotheRINGfingerinN-recognin(31,32),a specificity factor and ubiquitin ligase in the N-end rule pathway(reviewed in ref. 33). Therefore, p600 may also be associated with E3 ubiquitin ligase activity. Because our mapping studiesrevealed that the association with p600 requires N-terminalHPV-16 E7 sequences (Fig. 2  B ) that are necessary for E7-mediated pRB degradation (11), we next determined whetherthe interaction of E7 with p600 may be involved in HPV-16E7-mediated pRB destabilization. We used small interferingRNA (siRNA) technology to deplete p600 in HPV-16 E7-expressingcells.Ifp600wasarate-limitingfactorinE7-mediatedpRB degradation, it may be expected that pRB levels be elevatedin p600-depleted, E7-expressing cells. However, experiments inE7-expressing IMR90 human diploid lung fibroblasts (17) andHPV-16-positive CaSki human cervical cancer cells by stableexpression of a p600-specific shRNA expression vector providedno evidence for elevated pRB levels upon p600 depletion (Fig.3 C ). The HPV-16 E7 protein has a short half-life and is degradedthroughaproteasome-dependentpathwayinvolvingconjugationof ubiquitin to the N-terminal residue (34). Depletion of p600had no effect on HPV-16 E7 steady state levels (data not shown)and, hence, it is unlikely that p600 plays a rate-limiting role inE7-mediated pRB destabilization or HPV-16 E7 turnover. Depletion of p600 Decreases Anchorage-Independent Growth.  Asso-ciation of p600 with HPV-16 E7 is mediated through N-terminalE7 sequences that are necessary for cellular transformation,including anchorage-independent growth in NIH 3T3 cells (12).To determine whether p600 depletion in CaSki cells may affecttheir transformed phenotype, we assessed their potential to formcolonies in soft agar-containing growth medium. Compared toCaSki cells that express the reverse orientation shRNA expres-sion vector, p600-depleted CaSki cells displayed a   10-folddecreased capacity to form colonies in soft agar-containingmedium (Fig. 4  A ). Similarly, p600 depletion in HPV-16 E7 orHPV-16E6  E7transformedNIH3T3cellsresultedina30–40%decrease in anchorage-independent growth (Fig. 4  B  and  C ).ThisfindingissomewhatlessdramaticthanintheCaSkicellline, which may reflect genetic differences between these two celltypes. To determine whether this effect was specific for HPVE7-transformed cells, we also depleted p600 in U2OS humanosteosarcoma cells, which do not contain HPV sequences.Similar to HPV-16-transformed CaSki cells, p600 depletion inU2OS cells resulted in an  10-fold decreased capacity to formcolonies in soft agar (Fig. 4  D ). Importantly, p600 depletion didnot markedly reduce growth of U2OS cells in monolayer culture,suggesting that p600 depletion does not generally interfere withcellular proliferation (data not shown). Likewise, p600 depletionin parental NIH 3T3 cells caused a decrease in anchorage-independent growth. Hence, p600 depletion in HPV-positive or-negative cells inhibits their capacity to grow in the absence of asubstratum, which represents an important hallmark of cellulartransformation (reviewed in ref. 35). Discussion Here we report the isolation of host cellular protein complexesassociatedwiththeHPV-16E7oncoproteinintheHeLacervical Fig. 3.  Association and colocalization of HPV-16 E7 with p600 in cervicalcancer cells. (  A ) HPV-16 E7 associates with p600 in the HPV-16 positive CaSkihumancervicalcarcinomacellline.ImmunoprecipitationwasperformedwithE7-specific antibodies followed by detection of p600 by immunoblot. TheHPV-negativecervicalcancercelllineC33expressessimilarp600levelsasCaSkiandwasusedasaspecificitycontrolfortheseexperiments.( B )LocalizationofHPV-16 E7 (green,  Left  ) and p600 (red,  Center  ) in CaSki cells in nuclear andcytoplasmic structures by confocal fluorescence microscopy. The merged pic-ture (yellow,  Right  ) documents colocalization of the two proteins. ( C  ) Deple-tion of p600 ( Top ) by stable expression of a p600-specific shRNA expressionplasmid(600i)inHPV-16E7expressingIMR-90humandiploidfibroblasts( Left  )and the HPV-16-positive CaSki cervical cancer cell line ( Right  ) does not affectintracellularpRBlevels( Middle ).GAPDHproteinlevels( Bottom )areshowntodocument equal loading. Cells expressing the reverse orientation shRNAexpression vector were used as controls (Ci). Fig.4.  Depletionofp600bystableexpressionofspecificshRNAexpressionplasmids(600i)decreasesanchorageindependentgrowthinHPV-16positiveCaSkicervical cancer (  A ), HPV-16 E7 expressing NIH 3T3 ( B ), and HPV-16 E6  E7 expressing NIH 3T3 ( C  ) and HPV-negative U2OS osteosarcoma ( D ) cells. Decreased p600expressionisdocumentedbyimmunoblot;GAPDHproteinlevelsareshownasaloadingcontrol.CellsexpressingthereverseorientationshRNAexpressionvectorwere used as controls (Ci). The bar graphs denote averages and standard deviations of three (NIH 3T3 E7, E6  E7) or two (CaSki, U2OS) experiments, eachperformed with duplicate plates. Huh  et al.  PNAS    August 9, 2005    vol. 102    no. 32    11495      M     I     C     R     O     B     I     O     L     O     G     Y  cancer cell line. In addition to well established cellular targetproteins such as pRB and E2F family members, cyclins A and E,cdk2, and cdk1 (reviewed in ref. 2), we discovered additionalpotential cellular proteins in association with HPV-16 E7 (Fig.2  A ). These include the 600-kDa retinoblastoma protein associ-ated factor, p600. This protein was initially discovered in HeLacells by TAP as a pRB-associated protein (Y.N., unpublishedresults). Our results show that the p600  HPV-16 E7 interactionis mediated through an N-terminal E7 domain that is related toa portion of CR1 in Ad E1A (4) (Fig. 2  B ). Association of p600 with HPV-16 E7 is independent of pRB because the pRBbinding-deficient HPV-16 E7   21–24 mutant associates withp600 at similar levels as wild-type HPV-16 E7. In addition, thebovine papillomavirus type 1 (BPV-1) E7 protein, which lacksthe LXCXE core pRB-binding domain and is pRB-bindingdeficient (9), was shown in an independent study to associate with p600 through related N-terminal sequences (36). Addi-tional studies are necessary to determine whether the HPV-16E7  p600 complex also contains pRB or whether the HPV-16E7  pRB and HPV-16 E7  p600 complexes represent biochemi-cally and biologically distinct entities. In addition, it remains tobe determined whether E7 associates directly with p600 or whethercomplexformationismediatedthroughanothercellularprotein.However, our experiments strongly suggest that p600 is animportant cellular target that mediates some of the transformingactivities of HPV-16 E7. For one, p600 binding-deficientHPV-16 E7 mutants within the N-terminal CR1 homologydomain of E7 have a reduced transforming potential (12, 13).Similarly, the ability of BPV-1 E7 to cooperate with BPV-1 E6to induce anchorage-independent growth (37) maps to relatedN-terminal BPV-1 E7 sequences that are necessary for p600association (36). Moreover, p600 depletion in HPV-16 E7 andE6  E7-transformed NIH 3T3 mouse fibroblasts and HPV-16-positive CaSki cervical cancer cells interfered with the trans-formed phenotype, as evidenced by a marked reduction of anchorage independent growth (Fig. 4). However, this effect wasnot specific to HPV oncogene-transformed cells, and p600depletion in the HPV-negative U2OS human osteosarcoma cellline resulted in a similar decrease in anchorage-independentgrowth (Fig. 4). Hence, p600 may be an important regulator of cellular processes that modulate cellular proliferation and  orsurvival in response to detachment from the matrix.In addition, we found that, similar to HPV-16 E7-expressingcells (11), p600-depleted IMR90 human diploid fibroblasts andNIH 3T3 murine fibroblasts are sensitized to cell death underconditions of confluence and growth factor deprivation (K.W.H.and K.M., unpublished observations). These observations sug-gest that p600 may be a component of the trophic sentinel-signaling pathway, an important intrinsic tumor suppressorpathway that functions to thwart outgrowth of early neoplasticcell clones that have suffered mutations in oncogene pathways(reviewed in ref. 38).Even though p600 is a very large protein, it does not containmany conserved domains that yield much information regardingits biological function. Of note is a RING finger domain betweenamino acid residues 1660 and 1717 that is most closely related toa similar motif in N-recognin, a ubiquitin ligase in the N-end rulepathway that directly binds to N-terminal destabilizing aminoacidresidues(31,32).Hence,itistemptingtospeculatethatp600may also function as a regulator of ubiquitin-mediated proteol- ysis. Despite the fact that p600 associates with HPV-16 E7through N-terminal sequences that are necessary for E7 inducedpRB destabilization, our siRNA experiments suggest that p600is not a rate-limiting factor for HPV-16 E7-mediated pRBdegradation (Fig. 3 C ) or proteasome-mediated turnover of HPV-16 E7 (data not shown).Proteins related to p600 have also been isolated from otherorganisms.Theseincludethe  Arabidopsisthaliana BIGprotein(39)and the  Drosophila melanogaster   calossin  pushover protein (40).The  Arabidopsis BIGproteinmediatespolarauxintransport(39)as well as hormone and light responses (41), whereas the  Drosophila calossin  pushover protein has been implicated in regulating neu-ronal excitability and transmitter release processes at neuromus-cular junctions as well as glial growth (40, 42). Thus, it is temptingtospeculatethatp600maybesimilarlyinvolvedinvesicletraffickingin mammalian cells. Given its large size, p600 is most likely amultifunctional protein; however, it is presently unclear whetherand how such potential vesicle trafficking functions may be relatedto the regulation of anchorage-dependent growth in mammaliancellsthatisinferredfromourstudies(Fig.4).Nevertheless,becauseHPV-16 E7 colocalizes and associates with cytoplasmic as well asnuclear p600 (Fig. 3  B ), association of the HPV-16 E7 protein mayaffect nuclear as well as cytoplasmic p600 activities.In summary, we have identified p600 as a cellular target of theHPV-16 E7 oncoprotein. Our results indicate that p600 maymediate some of the pRB-independent transforming activities of HPV-16 E7. The simultaneous identification of p600 as a potentialtransformation target of BPV-1 E7 (36) suggests that this mecha-nism is conserved with other papillomavirus types. In addition, theE7 sequences that are necessary for p600 association are alsoconservedwithintheadenovirusE1Aandsimianvirus40Tantigensequences (4, 5), and it will be interesting to determine whetherthese oncoproteins also target p600. Clearly, p600 association of HPV E7 proteins per se is not sufficient for cellular transformationbecause the transformation-defective low-risk HPV E7 proteinsefficiently associate with p600 (Fig. 2 C ). This finding is similar to whathasbeenpreviouslyreportedforE7  pRBcomplexformation, where cellular transformation has been linked to pRB degradation,and not mere binding (20). Hence, p600 association may benecessary but not sufficient for transformation by papillomavirusE7 proteins. Therefore, it will be important to determine thefunctional consequences of p600 association of transforming andnontransforming papillomavirus E7 proteins in more detail. We thank Dr. Galloway (Fred Hutchinson Cancer Center, Seattle) forthe pLXSN based expression vectors and Dr. Agami (Center forBiomedical Genetics, Utrecht, The Netherlands) for his kind gift of thepRetro  Super plasmid. We also thank Kei-ichiro Ishiguro and HideakiTagami,twopreviousmembersoftheNakatanilaboratory,fortheirkindhelp with various aspects of tandem affinity purification. This work wassupported by Public Health Service Grants R01-CA081135 (to K.M.)and P01-CA050661 (to P.M.H.). J.D. was supported by American Association Society Fellowship PF-0414701. 1. zur Hausen, H. (2002)  Nat. Rev. Cancer   2,  342–350.2. Mu¨nger, K., Baldwin, A., Edwards, K. M., Hayakawa, H., Nguyen, C. L.,Owens, M., Grace, M. & Huh, K. W. (2004)  J. Virol.  78,  11451–11460.3. Duensing, S. & Mu¨nger, K. (2004)  Int. J. Cancer   109,  157–162.4. Phelps, W. C., Yee, C. L., Mu¨nger, K. & Howley, P. M. (1988)  Cell  53,  539–547.5. Vousden, K. H. & Jat, P. S. (1989)  Oncogene  4,  153–158.6. Munger, K., Basile, J. R., Duensing, S., Eichten, A., Gonzalez, S. L., Grace, M.& Zacny, V. L. (2001)  Oncogene  20,  7888–7898.7. Boyer, S. N., Wazer, D. E. & Band, V. (1996)  Cancer Res.  56,  4620–4624.8. Jones, D. L. & Mu¨nger, K. (1997)  J. Virol.  71,  2905–2912.9. Mu¨nger, K., Werness, B. A., Dyson, N., Phelps, W. C., Harlow, E. & Howley,P. M. (1989)  EMBO J.  8,  4099–4105.10. Edmonds, C. & Vousden, K. H. (1989)  J. Virol.  63,  2650–2656.11. Jones, D. L., Thompson, D. A. & Mu¨nger, K. (1997)  Virology  239,  97–107.12. Banks, L., Edmonds, C. & Vousden, K. (1990)  Oncogene  5,  1383–1389.13. Phelps, W. C., Mu¨nger, K., Yee, C. L., Barnes, J. A. & Howley, P. M. (1992)  J. Virol.  66,  2418–2427.14. Helt, A. M. & Galloway, D. A. (2001)  J. Virol.  75,  6737–6747.15. Balsitis, S. J., Sage, J., Duensing, S., Mu¨nger, K., Jacks, T. & Lambert, P. F.(2003)  Mol. Cell. Biol.  23,  9094–9103.16. Duensing, S. & Mu¨nger, K. (2003)  J. Virol.  77,  12331–12335. 11496    cgi  doi  10.1073  pnas.0505337102 Huh  et al.
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks