Human Papillomavirus Type 18 E6 Protein Binds the
http://www.100md.com
病菌学杂志 2005年第7期
Laboratoire de Biologie Moléculaire de la Cellule, CNRS UMR 5161, Ecole Normale Supérieure de Lyon, Lyon, France
ABSTRACT
Several viral proteins expressed by DNA or RNA transforming viruses have the particular property of binding via their C-terminal end to various cellular proteins with PDZ domains. This study is focused on the PDZ protein TIP-2/GIPC, which was originally identified in two-hybrid screens performed with two different baits: the human T-cell leukemia virus type 1 Tax oncoprotein and the regulator of G signaling RGS-GAIP. Further studies have shown that TIP-2/GIPC is also able to associate with the cytoplasmic domains of various transmembrane proteins. In this report we show that TIP-2/GIPC interacts with the E6 protein of human papillomavirus type 18 (HPV-18). This event triggers polyubiquitination and proteasome-mediated degradation of the cellular protein. In agreement with this observation, silencing of E6 by RNA interference in HeLa cells causes an increase in the intracellular TIP-2/GIPC level. This PDZ protein has been previously found to be involved in transforming growth factor ? (TGF-?) signaling by favoring expression of the TGF-? type III receptor at the cell membrane. In line with this activity of TIP-2/GIPC, we observed that depletion of this protein in HeLa cells hampers induction of the Id3 gene by TGF-? treatment and also diminishes the antiproliferative effect of this cytokine. Conversely, silencing of E6 increases the expression of Id3 and blocks proliferation of HeLa cells. These results support the notion that HPV-18 E6 renders cells less sensitive to the cytostatic effect of TGF-? by lowering the intracellular amount of TIP-2/GIPC.
INTRODUCTION
In addition to the function they exert in virus replication, viral proteins often strongly interfere with cell physiology by interacting with cellular factors involved in important regulatory networks. Hence, the viral proteins, especially those expressed by DNA or RNA transforming viruses, represent interesting models to better understand cellular regulatory pathways. Along this line, the discovery of the ability of proteins expressed by different viruses to bind various cellular proteins with PSD-95-Discs-large-ZO1 (PDZ) domains has open a new and interesting field of investigation. The PDZ domain is indeed present in many human proteins, as either single or multiple copies, associated or not with other protein domains (for recent reviews see references 4 and 25). The PDZ proteins exert diverse functions, generally having an important organizing role by establishing specific interactions with multiple partners. The archetypal interaction is association of the PDZ domain with the C-terminal end of another protein, with the specificity of the interaction depending on its C-terminal four amino acids and the consensus motif being XS/TXV-COOH for class I PDZ domains (29, 32, 45). In addition to this general rule, it is now well established that the PDZ domain can also interact with an internal sequence motif and also with another PDZ domain (4, 25). Relying on these properties, PDZ proteins play important roles in transmembrane receptors assembly and clustering, as well as in organization of cellular junctions. This family of proteins also intervenes in signal transduction, and examples of transcriptional regulation activities exist (24, 49). Unexpectedly, it has been observed that three different viral transforming proteins, i.e., E4 open reading frame 1 of adenovirus type 9, E6 of human papillomavirus type 18 (HPV-18), and Tax of human T-cell leukemia virus type 1 (HTLV-1), were able to bind PDZ proteins (31, 35, 43). These three proteins include a canonical class I PDZ binding site (BS). For E4 and E6, deletion of this C-terminal PDZ BS triggers a loss of in vitro transforming activity (31). The PDZ BS of Tax of HTLV-1 also plays a role in in vitro transformation of Rat-1 cells (23). In line with these observations, it has been observed that a PDZ BS is present at the C terminus of E6 in malignant HPV types, such as HPV-18 or HPV-16, but not in the benign ones, such as HPV-11 (31). Similarly, Tax of HTLV-2, which is not associated with aggressive leukemia, does not include a functional PDZ BS (23). Collectively, these observations strongly support the notion that a relationship exists between in vivo transformation and interference with the normal functioning of PDZ proteins. Hence, it appears to be important to characterize the complete set of cellular PDZ proteins targeted by these viral oncoproteins. By performing a two-hybrid screen of a human B-lymphocyte cDNA library with Tax as bait, we have previously identified seven different PDZ proteins: TIP-1, -2, -15, -33, -40, and -43 and hDlg-1 (14, 43). Other reports have shown that hDlg-1 is bound by the three viral proteins Tax, E6, and E4 open reading frame 1 (17, 31, 35, 46). E6 has also been shown to interact with hScrib, MUPP1, and MAGI-1, -2, and -3 (18, 34, 38, 48), and recently the interaction of Tax with MAGI-3 was reported (40).
In this work, we examined whether TIP-2, which has been identified as interacting with Tax, also represents a target for the HPV-18 E6 protein. A more precise analysis of the relationship of this protein with viral oncoproteins is interesting, as various publications have shown that TIP-2 is able to interact with a great variety of cytoplasmic or transmembrane proteins. It has indeed been shown that TIP-2 interacts with the GTPase-activating protein for Gi GAIP (11). From this property, TIP-2 was named GIPC (for GAIP-interacting protein, C terminus) (11). Genes coding for homologous proteins have been identified in humans and named GIPC2 and GIPC3 (30, 44). This cellular protein has also been identified as a possible partner of the following proteins: the rat glucose transporter GLUT1 (under the name GLUT1CBP) (7), the mouse semaphorin MSemF (under the name SEMCAP-1) (50), the rat neuropilin 1 (under the name NIP) (8), the mouse syndecan-4 proteoglycan (under the name synectin) (16), TrkA and TrkB (37), the melanosomal protein gp75 (36), the 5 and 6 integrins (13, 47), the Xenopus IGF1 receptor (6), the human 5T4 oncofetal antigen (2), myosin VI (1, 7), the dopamine D2 and D3 receptors (26), the lutropin receptor (22), the ?1-adrenergic receptor, and the transforming growth factor ? (TGF-?) type III receptor (T?RIII) (5). For the sake of clarity, the name TIP-2/GIPC is used here.
Evidence that E6 of HPV-18 binds TIP-2/GIPC and triggers its degradation by the proteasome is presented in this report. Silencing of TIP-2/GIPC by RNA interference also showed that this PDZ protein is important for the effect of TGF-? on gene expression and cell proliferation. Our observations support the notion that destruction of TIP-2/GIPC by HPV-18 E6 contributes to cell transformation by hampering TGF-? signaling.
MATERIALS AND METHODS
Constructs. The pTL1-TIP-2 expression vector was generated as follows. The TIP-2/GIPC sequence was amplified from an expressed sequence tag (IMAGE clone 2958330) by using the sense and antisense primers 5'-GACCCCACTTCTCGCTGCTCATG-3' and 5'-GGGGGATCCTAGTAGCGGCCG-3', respectively. The PCR fragment was cloned in PCR-Script (Stratagene), and the NotI-PstI restriction fragment containing the TIP-2/GIPC coding sequence was inserted between the NotI and PstI restriction sites of pTL1, which is a pSG5 derivative (19). The pSGF-E6 vector, expressing the N-terminally FLAG-tagged E6 protein from HPV-18, was constructed as follows: the E6-coding sequence was amplified from plasmid pHPV18 (kindly provided by F. Thierry) with the sense and antisense primers 5'-GCGAATTCGCTCGAGTTGAGGATCCAACACGG-3' and 5'-CATAGTCGACATTATACTTGTGTTTCTC-3', respectively, and then digested with XhoI and SalI and inserted in the XhoI restriction site of pSG-FLAG (10) in the sense orientation. The vectors expressing the FLAG-tagged E6 proteins mutated in the PDZ binding site were generated as follows. The E6-coding sequence was amplified from pSGF-E6 with the sense primer 5'-GTAATACGACTCACTATAGGG-3' and the antisense primers 5'-CATAGTCGACATTATACTTGTGCTTCTCTGCGTC-3', 5'-CATAGTCGACATTATGCTTGTGTTTCTC-3', 5'-CATAGTCGACATTATGCTTGTGCTTCTCTGCGTC-3', and 5'-CATAGTCGACATTATCTGCGTCGTTGGAG-3' to generate the T156A, V158A, TAVA, and 4C mutants, respectively. The PCR fragments were digested with EcoRI and XhoI and inserted between the EcoRI and SalI restriction sites of pSGF. Plasmids expressing hemagglutinin (HA)-tagged ubiquitin, either wild type or with all lysines except lysine 48 or lysine 63 mutated to arginine (kindly provided by V. Dixit), have been previously described (52).
Two-hybrid assay. Plasmid pGBT9-E6, which was used as bait, was generated as described above for pSGF-E6, except that the PCR fragment was digested with the EcoRI and SalI restriction enzymes and inserted in the pGBT9 plasmid (Clontech) between the EcoRI and SalI restriction site. The clone –1, –2, –15, –33, –40, and –43 prey plasmids were those obtained from the two-hybrid screen against Tax (43). Two-hybrid assay, analysis of growth on minimal medium lacking histidine, and examination of ?-galactosidase expression were performed as previously described (43).
Cell culture and transfection. HeLa, COS7, and CV-1 cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (FBS), 100 U of penicillin ml–1, and 100 μg of streptomycin ml–1 at 37°C in a 5% CO2 humidified atmosphere. For plasmid transfection, the amount of FBS used in the culture medium was reduced to 5%. Transfection experiments were performed by the calcium phosphate precipitation method in 100-mm-diameter petri dishes with the amounts of expression vector indicated in the figure legends. The total amount of DNA was adjusted to 15 μg with pBlueScript plasmid. For transfection of small interfering RNAs (siRNAs) (12), 50,000 cells were seeded in 12-well plates in Dulbecco's modified Eagle's medium without antibiotics supplemented with 5% FBS the day before. Five microliters of 20 μM siRNA duplex was mixed with 40 μl of Opti-MEM I. Separately, 4.5 μl of Oligofectamine reagent (Invitrogen) was added to 24 μl of Opti-MEM I, and the mixture was incubated for 8 min at room temperature. The two mixtures were combined, incubated at room temperature for 25 min, and added to the cells. When transfection was done with different sizes of tissue culture supports, the amounts of cells, siRNAs, and Oligofectamine were adjusted proportionally to the surface area. The sequence of the sense siRNA for TIP-2/GIPC was 5'-GCCAACUGCCGAGGUGAUGTT-3'. The siRNA duplex used for E6/E7 is the one reported previously (20).
Antibodies. Rabbit polyclonal antibodies were generated against a peptide corresponding to the C-terminal 19 amino acids of TIP-2/GIPC. The peptide was chemically synthesized and coupled to ovalbumin. The TIP-2/GIPC antibodies obtained were affinity purified against a protein corresponding to the C-terminal 85 amino acids of TIP-2/GIPC. The clone 2 plasmid (43) was digested with XmaI, and the fragment coding for the C terminus of TIP-2/GIPC was inserted in the XmaI restriction site of PinPoint Xa1 (Promega) in the sense orientation. The HB101 Escherichia coli strain was transformed with the resulting plasmid. Expression of the fusion protein was induced in a 500-ml overnight culture in presence of 2 μM biotin by treatment with 100 μM IPTG (isopropyl-?-D-thiogalactopyranoside) for 4 h at 37°C. Bacteria were lysed by sonication in lysis buffer consisting of 50 mM Tris (pH 7.5), 50 mM NaCl, and 5% glycerol. After centrifugation, the bacterial lysate was loaded on a column containing 4 ml of streptavidin agarose beads equilibrated in phosphate-buffered saline (PBS). The lysate was passed three times through the column. After a 10-ml wash with lysis buffer, the column was loaded with 10 ml of antiserum to TIP-2/GIPC. The column was then successively loaded with 10 ml of 0.3 M NaCl, 10 ml of PBS-0.1% Tween 20, and 10 ml of PBS. The retained immunoglobulins were eluted with 15 ml of 0.1 M glycine (pH 3.5). The purified antibodies were collected in 1.5-ml tubes containing 100 μl of 1 M Tris (pH 8.0). A rabbit polyclonal antibody was similarly raised against a peptide corresponding to amino acids 323 to 341 of TIP-40 but was used without affinity purification. For FLAG-tagged proteins we used the mouse monoclonal antibody M2 (Sigma), and for HA-tagged proteins we used the mouse monoclonal antibody clone 12CA5 (Roche). Rabbit polyclonal antibody to ubiquitin-protein conjugates was purchased from Affiniti (UG9510), and monoclonal antibody to ?-actin was purchased from Sigma.
Immunoprecipitation and immunoblotting. Transfected COS7 cells were lysed for 20 min in 500 μl of radioimmunoprecipitation assay (RIPA) buffer (21) supplemented with 10 mM iodoacetamide. Lysates were centrifuged for 10 min at 15,000 x g, and the protein concentration was measured with the Dc protein assay kit (Bio-Rad). The supernatants, adjusted to equal amounts of total protein, were incubated with antibodies diluted 1:250. After incubation at 4°C for 2 h, 30 μl of protein A beads equilibrated in RIPA buffer was added to the mixtures, which were further incubated for an additional hour. Protein A beads were collected by centrifugation at 425 x g for 2 min and washed three times in RIPA buffer plus iodoacetamide. Proteins were eluted in sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis loading buffer at 80°C for 10 min. After separation by SDS-polyacrylamide gel electrophoresis, proteins were transferred to a polyvinylidene difluoride membrane. For immunoblotting, primary antibodies were used diluted 1:1,000 (for antibodies to TIP-2/GIPC, FLAG, HA, and ubiquitin-protein conjugates), 1:4,000 (for antibody to ?-actin), or 1:500 (for antibodies to p53 and TIP-40), and secondary antibodies were diluted 1:6,000. Visualization was performed by chemiluminescence with the ECL reagent (Amersham Pharmacia Biotech).
Real-time quantitative RT-PCR. After transfection of siRNAs, HeLa cells were collected and total RNAs were extracted by using the RNeasy minikit (Qiagen). One-step reverse transcription-PCRs (RT-PCRs) were performed with the use of the QuantiTect SYBR Green RT-PCR kit (Qiagen) and the LightCycler apparatus (Roche) according to the cycling conditions specified in the handbook for the kit. Gene-specific primers were designed by using the Primer3 software. The sequences of the sense and antisense primers, respectively, used for quantitative PCR were as follows: Id3, 5'-GGAGCTTTTGCCACTGACTC-3' and 5'-TTCAGGCCACAAGTTCACAG-3'; lamin, 5'-CCGAGTCTGAAGAGGTGGTC-3' and 5'-AGGTCACCCTCCTTCTTGGT-3'; Hmg1, 5'-AAGCACCCAGATGCTTCAGT-3' and 5'-GTGCATTGGGATCCTTGAA-3'; cyclin B1, 5'-CGGGAAGTCACTGGAAACAT-3' and 5'-AAACATGGCAGTGACACCAA-3'; Set, 5'-GCAAGAAGCGATTGAACACA-3' and 5'-GCAGTGCCTCTTCATCTTCC-3'; Plk1, 5'-AAGAGATCCCGGAGGTCCTA-3' and 5'-GCTGCGGTGAATGGATATTT-3'; Slc20A, 5'-AGCGTGGACTTGAAAGAGGA-3' and 5'-TCTTTGTACAGGCCGGAATC-3'; and TIP-2/GIPC, 5'-CTCCACCACTTTCCACCATC-3' and 5'-GAGGTAACAGGCTCCACAGG-3'.
Thymidine incorporation assays. After transfection of siRNAs and 4 h before harvest, 5 μCi of [methyl-3H]thymidine was added to the cell culture medium. The cells were washed twice in ice-cold trichloroacetic acid and lysed in 0.2 N NaOH. The protein concentration was measured with the Dc protein assay kit (Bio-Rad). Cell lysates (20 μl) were added to Supermix (Perkin-Elmer), and radioactivity was counted in a Wallac apparatus with 3H mode. Radioactivity values were normalized with respect to protein concentrations.
RESULTS
Interaction of HPV-18 E6 with TIP-2/GIPC in two-hybrid assays. As a first attempt to determine whether HPV-18 E6 can bind TIP-2/GIPC, two-hybrid assays were performed with E6 fused to the GAL4 DNA binding domain as bait, along with the various clones coding for PDZ proteins obtained in the Tax two-hybrid screen (clones –1, –2, –15, –33, –40, and –43) (43) as prey. The results of these assays performed with the HF7C strain were evaluated by both growth on minimal medium lacking histidine and expression of the ?-galactosidase reporter protein. E6 showed no association with the GAL4 DNA binding domain alone (Table 1). Growth on medium without histidine was positive for all clones, with the exception of that encoding TIP-40. However, ?-galactosidase expression was clearly positive only for TIP-15 and TIP-2/GIPC. These observations strongly suggested that TIP-2/GIPC is also targeted by E6 of HPV-18.
Binding to E6 triggers degradation of TIP-2/GIPC by the proteasome. Considering the positive result of the two-hybrid assay, whether E6 and TIP-2/GIPC interact in the context of mammalian cells was further tested. As it is well documented that E6 can cause degradation of cellular proteins to which it binds, including those with a PDZ domain, analysis of the interaction between the two proteins was performed by immunoprecipitation experiments either with or without treatment of the cells with the proteasome inhibitors MG132 and lactacystine. COS7 cells were transfected with vectors expressing a FLAG-tagged HPV-18 E6, as well as TIP-2/GIPC, either in the wild-type form or fused to enhanced green fluorescent protein (GFP) at its N-terminal end. Immunoblot analysis with an antibody directed against TIP-2/GIPC of the extracts used in these experiments showed that the proteasome inhibitor treatment did not modify the amount of either TIP-2/GIPC or GFP-TIP-2 (Fig. 1B). This was also the case for FLAG-E6 expression (Fig. 1D). Extracts of these transfected cells were used for immunoprecipitation with the M2 monoclonal antibody to FLAG. Immunoblot analysis using the polyclonal antibody directed against TIP-2/GIPC revealed its presence in the precipitated proteins, but this was evident mainly when cells were treated with MG132 (Fig. 1A, lanes 1 and 2). Use of a different antibody in the immunoprecipitation did not trigger any TIP-2/GIPC, indicating that the reaction was specific (Fig. 1A, lane 3). The GFP-TIP-2 fusion protein was also specifically precipitated by FLAG-E6 (Fig. 1A, lanes 4, 5, 7, and 8). In this case the signal was clearly visible in the absence of proteasome inhibitor treatment but was reinforced when cells were incubated with either MG132 or lactacystine (Fig. 1A, compare lanes 4 and 5 and lanes 7 and 8). These experiments were also performed in the opposite way, i.e., by precipitating TIP-2/GIPC and analyzing the presence of FLAG-E6 in immunoprecipitated proteins. The results obtained were symmetrical. FLAG-E6 was coprecipitated with TIP-2/GIPC, but only when cells were treated with MG132 (Fig. 1C, compare lanes 1 and 2), and coprecipitation with GFP-TIP-2 was augmented by either MG132 or lactacystine treatment (Fig. 1C, compare lanes 4 and 5 and lanes 7 and 8). It was verified that the interaction between E6 and TIP-2/GIPC relies on the E6 PDZ BS by altering the C-terminal end of the latter protein. Either the C-terminal valine or the threonine at position –2 was mutated to alanine. These two mutations did not impair interaction of E6 with TIP-2/GIPC as evaluated by MG132 treatment (Fig. 1E, lanes 1 and 2). However mutation of both residues markedly reduced it (Fig. 1E, lane 3), and deletion of the C-terminal four amino acids led to a complete disappearance of E6 coprecipitated with TIP-2/GIPC (Fig. 1E, lane 4). These four mutants were expressed at equal levels in cells (Fig. 1F). These data clearly show that E6 binds TIP-2/GIPC via its PDZ BS in mammalian cells, but they also indicate that this event cause proteasome-mediated degradation of the cellular protein. The GFP-TIP-2 fusion protein is probably less prone to degradation than wild-type TIP-2/GIPC, but some proteolysis also occurred, since the proteasome inhibitor treatment increased the amount of coprecipitated protein.
It has been established for several factors associating with E6 that this viral protein induces their polyubiquitination by binding concomitantly to the E3 ubiquitin ligase E6AP (15, 38, 42). In this regard, we examined whether E6 induces polyubiquitination of TIP-2/GIPC. Cells were transfected with expression vectors for both proteins together with vectors producing HA-tagged ubiquitin either in the wild-type form or with each lysine except lysine 48 or lysine 63 mutated to arginine (52). Immunoprecipitation was done with the antibody to TIP-2/GIPC, and immunoblot analysis was performed with a monoclonal antibody directed against HA. These experiments clearly showed that E6 stimulates polyubiquitination of TIP-2/GIPC (Fig. 2). This was seen with wild-type ubiquitin but also with ubiquitin allowing branching only on lysine 48 (Fig. 2A, lanes 2 and 4). The smear of polyubiquitinated forms of TIP-2/GIPC was less intense with ubiquitin bearing a single lysine at position 63 (Fig. 2A, lane 6). This result suggests that E6 may also stimulate polyubiquitination by K63 branching. Alternatively, it is possible that this observation results from incorporation of tagged mutated ubiquitin in a chain bearing endogenous wild-type molecules. In order to exclude that this polyubiquitination of TIP-2/GIPC resulted from ubiquitin overexpression, CV-1 cells were transfected with either the control vector pSGF or a vector expressing FLAG-tagged E6, and endogenous TIP-2/GIPC was immunoprecipitated and analyzed by immunoblotting with an antibody raised against ubiquitin-protein conjugates. This showed that expression of E6 in these cells increases the intensities of the bands corresponding to TIP-2/GIPC polyubiquitinated forms (Fig. 2B, compare lanes 3 and 4), with this being observed in the absence of ubiquitin overexpression. These observations clearly show that E6 binding to TIP-2/GIPC causes its polyubiquitination, mainly with lysine 48 branching.
To establish that E6 can destabilize TIP-2/GIPC under physiological conditions, we looked at the effect of E6 silencing on the intracellular level of this cellular protein in HeLa cells. It has recently been reported that depleting these cells of E6 and E7 by the RNA interference approach causes a rapid increase in the amount of p53 (20) (Fig. 3A, compare lanes 1 and 2). When an extract of cells treated with siRNAs directed against E6 and E7 were analyzed by immunoblotting for TIP-2/GIPC, a clear increase in the amount of this protein was also observed (Fig. 3B, lane 2). Such an effect was not observed when cells were transfected with control siRNAs (Fig. 3B, lane 1, and data not shown). In sum, by associating with TIP-2/GIPC, E6 triggers its polyubiquitination and proteasome-mediated degradation. This activity decreases the normal level of endogenous TIP-2/GIPC in cells expressing E6.
TIP-2/GIPC silencing inhibits Id3 expression. As a first step towards understanding the consequences of a decrease in TIP-2/GIPC for the cell, we examined the effect of its silencing by RNA interference on gene expression. A microarray analysis performed with a set of 1,853 genes involved in cell transformation showed that expression of Id3, which encodes a helix-loop-helix protein, was reduced in TIP-2/GIPC-silenced cells. To confirm this observation, the level of the Id3 mRNA in these cells was evaluated by quantitative PCR, along with that of a set of unrelated genes taken as controls. The TIP-2 mRNA was also included in the analysis. It was first verified by immunoblotting that the siRNA duplex directed against TIP-2 efficiently decreased the amount of protein (Fig. 4B, compare lane 2 with lanes 1 and 3). Immunoblot analysis using antibodies directed against another PDZ protein, TIP-40, or ?-actin showed that this effect was specific (Fig. 4C and D).
As expected, quantitation of the TIP-2 mRNA showed a marked decrease in siRNA-transfected cells, to 18.5% of its normal level (Fig. 4E). Under these experimental conditions, approximately 85% of the cells were transfected as evaluated by immunofluorescence analysis (data not shown). This indicates that the procedure almost completely removed TIP-2 RNA from the transfected cells. With the control genes no significant effect was observed, whereas the level of Id3 mRNA was only 38% (Fig. 4E). After correction for nontransfected cells, this meant that the TIP-2/GIPC depletion causes a fivefold decrease of the Id3 mRNA in transfected cells. Interestingly, it has been reported that Id3, which is known to be important in the control of cell growth by interfering with key basic helix-loop-helix factors, is regulated by transforming growth factor ? (27, 28). As TIP-2/GIPC has been implicated in the response to this cytokine (5), we analyzed further the effect of TGF-?1 treatment on Id3 expression. We observed that in HeLa cells addition of TGF-?1 increased the expression of Id3 (Fig. 5A, bar 2). This effect was impaired by TIP-2/GIPC silencing (Fig. 5A, bar 3). Interestingly, E6 silencing also increased Id3 expression (Fig. 5A, bar 4), and its combination with TGF-?1 treatment slightly augmented the effect of TGF-?1 alone (Fig. 5A, bar 5). This effect is likely to result from the increase in the TIP-2/GIPC level that can be seen by analyzing the level of expression of this protein in these cells (Fig. 5B). Collectively, these results establish that TIP-2/GIPC is an important factor for regulation of Id3 expression and that its presence is necessary for a full effect of TGF-?1 on expression of this gene.
TIP-2/GIPC silencing activates cell growth. In line with these observations, we examined whether TIP-2/GIPC can intervene in the regulation of cell proliferation. As reported previously, silencing of E6/E7 in HeLa cells markedly alters multiplication of HeLa cells as evaluated by their ability to incorporate tritium-labeled thymidine (20) (Fig. 6). At 72 h after transfection of the siRNAs, the amount of incorporated radioactivity was less than 20% of that measured for control cells, corresponding to an almost complete block of cell division, considering that approximately 15% of the cells were nontransfected. Conversely, silencing of TIP-2/GIPC led to a clear increase in the proliferation of HeLa cells (Fig. 6). This experiment showed that this protein negatively controls cell proliferation. Its degradation by E6 is therefore likely to participate in the induction of an active proliferation by HeLa cells. We further examined the effect of TIP-2/GIPC in combination with TGF-?1 on cell proliferation. As expected, treatment of HeLa cells with TGF-?1 inhibited proliferation, which dropped to 52% of its normal value (Fig. 7A). When E6 and E7 were depleted by RNA interference, TGF-?1 treatment did not additionally reduce proliferation, but as mentioned above, removal of the viral oncoproteins blocked cell division. As shown above, TIP-2/GIPC silencing (Fig. 7B) accelerated cell division (Fig. 7A). Addition of TGF-?1 to these TIP-2/GIPC-silenced cells reduced this effect, but only by 22%. This indicated that loss of TIP-2/GIPC renders the cells less sensitive to the effect of TGF-?1. Therefore, as for Id3 expression, removal of TIP-2/GIPC at least partially impairs the effect of TGF-?1.
DISCUSSION
TIP-2/GIPC has been initially characterized as a PDZ protein interacting with HTLV-1 Tax and GAIP (11, 43). As Tax and HPV-18 E6 share the property of interacting with PDZ proteins, we asked whether TIP-2/GIPC also interacts with E6. Results from two-hybrid assays as well as from coimmunoprecipitation experiments clearly showed that this is indeed the case. These data show that TIP-2/GIPC must be added to the list of cellular PDZ proteins bound by E6 via its C-terminal end. This oncoprotein has already been shown to associate with hDlg-1, MUPP1, hScrib, and MAGI-1, -2, and -3 (17, 18, 31, 34, 35, 38, 48). Similarly to what has been observed for these proteins, HPV-18 E6 induces polyubiquitination and degradation of TIP-2/GIPC. For many targets of E6, including p53 and also PDZ proteins such as hScrib, it has been shown that this degradation induction results from E6 acting as an adaptor between the E3 ubiquitin ligase E6AP and the protein to be degraded (38, 42). By analogy, it is likely that the effect of E6 on TIP-2/GIPC also involves E6AP; however, this point remains to be firmly established, as the role of E6AP in the degradation of hDlg remains controversial (38, 41). Whatever the exact nature of the E3 ligase involved in the process, the experiments involving silencing of E6/E7 in HeLa cells provide clear evidence that E6 mediates continuous degradation of TIP-2/GIPC. As has been noted before for hDlg or other PDZ proteins, this does not entirely eliminate the protein from the cell. This can be due to the presence of a limited amount of E6 or to partial inactivation of its C-terminal PDZ BS by PKA phosphorylation, as it has been shown that this modification impairs binding of E6 to hDlg (33).
Considering the wide diversity of the proteins which have been shown to interact with TIP-2/GIPC, it is difficult at this stage to associate this protein with a precise function. Almost all characterizations of these interactions originated from two-hybrid screens. They generally involve the PDZ domain or a larger region of TIP-2/GIPC, along with the PDZ BS corresponding to the C terminus of the partner protein or, more rarely, to an internal PDZ BS, as for TrkA (37). One exception to these two-hybrid analyses is a study of the relationship between TIP-2/GIPC and TGF-? signaling. In that work TIP-2/GIPC was identified in a retroviral screen for factors rendering cells resistant to the cytostatic effect of TGF-? (5). Further analysis paradoxically showed that overexpression of TIP-2/GIPC sensitizes the cells to the effect of TGF-?. Those authors explained this activity by an association of TIP-2/GIPC with a TGF-? type III receptor which exhibits a PDZ BS at its C terminus. Overexpression of TIP-2/GIPC increases the amount of T?RIII at the cell surface and inhibits its degradation by the proteasome (5). In our analysis of the effect of TIP-2/GIPC silencing on expression of cellular genes, the effect observed on Id3 supports the notion of a role of this PDZ protein in TGF-? signaling. This cytokine is indeed known to regulate Id genes (27, 28). In different epithelial cell lines it has been shown that TGF-? represses transcription of the Id genes, including Id3 (27). By contrast, in HeLa cells we observed that TGF-? activates Id3. This has also been observed in B-lymphocyte progenitors (28). In these cells, Id3 protein overexpression triggers arrest of cell growth and apoptosis (28). The data obtained with these cells indicate a key role of Id3 in the effect of TGF-?. Repression of the Id1 promoter in epithelial cell lines such as HaCaT or MCF-10A is due to ATF3, which acts as a repressor in combination with Smad3 (27). The exact mechanism is not known for Id3, but by analogy it is possible that in HeLa and B-lymphocyte progenitors, a repressor factor such as ATF3 is not induced by TGF-?, and the Smads therefore act positively. In HeLa cells, depletion of TIP-2/GIPC by RNA interference clearly decreased the amount of Id3 mRNA. It also impaired its increase in response to treatment of cells with TGF-?. In line with these observations, silencing of E6, which increased the amount of TIP-2/GIPC, led to an increase of Id3 expression and also slightly augmented the effect of the cytokine. These observations support the notion that TIP-2/GIPC is necessary for the effect of TGF-? on Id3 transcription, and this can be explained by a positive effect of the PDZ protein on TGF-? receptors. However, an intriguing aspect is that TIP-2/GIPC has an effect in the absence of the cytokine, and also the microarray data do not show an effect on other Id genes. Therefore, there is something particular to Id3, and it cannot be excluded at this stage that TIP-2/GIPC mediates another signaling event that would specifically activate this gene. A detailed analysis of the promoter elements mediating the effects of both TIP-2/GIPC silencing and TGF-? treatment should allow us to clarify this point and to understand what is specific to Id3.
Besides the effect on Id3, involvement of TIP-2/GIPC in TGF-? signaling is also supported by the role that it plays in regulation of proliferation. Silencing of this PDZ protein in HeLa cells had a marked effect on their proliferation rate, which increased progressively starting from 48 h after transfection of siRNAs. By contrast, silencing of E6, which increased TIP-2/GIPC, was associated with a clear reduction of proliferation. Interestingly, the cytostatic effect of TGF-? was limited by depletion of TIP-2/GIPC. This likely results from a decreased sensitivity to the cytokine, which can be explained, similarly to the effect on Id3, by the demonstrated effect of TIP-2/GIPC on T?RIII (5). The effect of TIP-2/GIPC depletion in the absence of TGF-? could be explained by the possible presence of cytostatic cytokines of the TGF-? family in the medium. Collectively, these observations show that TIP-2/GIPC plays an important role in the control of proliferation and hence that its decrease due to interaction with E6 might directly promote cell growth. In future studies it will be interesting to determine whether this activity results only from the effect of TIP-2/GIPC on TGF-? signaling or involves other pathways. Regulation of the sensitivity of cells to TGF-? by TIP-2/GIPC might involve effects on proteins other than T?RIII. In this regard, it has been reported that TIP-2/GIPC interacts with syndecan-4 (16), and recently it has been proposed that it also interacts with syndecan-2, a transmembrane heparan sulfate proteoglycan which, like T?RIII, facilitates the effect of TGF-? (9).
In conclusion, there is now strong evidence that the E6 protein of the high-risk HPV type 18 affects a group of different PDZ proteins. By targeting proteins such as hDlg, MAGIs, hScrib, and MUPP1, it probably alters epithelium morphology by disrupting cell-cell contacts. It has been reported that deletion of the HPV-18 E6 PDZ BS impairs the morphological change induced by this protein in keratinocytes (51). In mouse it has also been established that this motif is necessary for epithelial hyperplasia (39). According to our results, alteration of TIP-2/GIPC by E6 possibly directly participates in this increased proliferation. Our data strengthen the importance of the PDZ proteins as cellular targets of the transforming activity of E6 and offer a new connection with the action of TGF-?. This cytokine has recently been shown to downregulate expression of E6/E7 through Ski and NFI (3). Both the role of TIP-2/GIPC in TGF-? signaling and the effect of E6 on this protein raise the possibility of a positive feedback loop in which degradation of TIP-2/GIPC renders cells less sensitive to TGF-?, thereby increasing the expression of E6. Our findings reinforce the interest in seeking therapeutic means to disrupt the interaction of high-risk HPV E6s with cellular PDZ proteins in order to impair the deleterious effects of these oncoproteins.
ACKNOWLEDGMENTS
We are grateful to F. Thierry and V. Dixit for the generous gift of plasmids, to A. Roisin for assistance with cell culture, and to S. Gonin-Giraud and O. Gandrillon for advice on TGF-? assays. We thank C. Morris for help with microarray and real-time quantitative PCR experiments.
This work was supported by grants from the Ligue Nationale contre le Cancer (Programme Equipes Labelisées) and the Comité du Rh?ne de la Ligue Nationale contre le Cancer (to A.F.-B.).
Present address: Institut de Biochimie et Chimie des Protéines, UMR 5086 CNRS UCBL, 69367 Lyon Cedex 07, France.
Present address: Centre de Génétique Moléculaire et Cellulaire, CNRS UMR 5534, 69622 Villeurbanne Cedex, France.
REFERENCES
Aschenbrenner, L., T. Lee, and T. Hasson. 2003. Myo6 facilitates the translocation of endocytic vesicles from cell peripheries. Mol. Biol. Cell 14:2728-2743.
Awan, A., M. R. Lucic, D. M. Shaw, F. Sheppard, C. Westwater, S. A. Lyons, and P. L. Stern. 2002. 5T4 interacts with TIP-2/GIPC, a PDZ protein, with implications for metastasis. Biochem. Biophys. Res. Commun. 290:1030-1036.
Baldwin, A., L. Pirisi, and K. E. Creek. 2004. NFI-Ski interactions mediate transforming growth factor beta modulation of human papillomavirus type 16 early gene expression. J. Virol. 78:3953-3964.
Bezprozvanny, I., and A. Maximov. 2001. PDZ domains: more than just a glue. Proc. Natl. Acad. Sci. USA 98:787-789.
Blobe, G. C., X. Liu, S. J. Fang, T. How, and H. F. Lodish. 2001. A novel mechanism for regulating transforming growth factor beta (TGF-beta) signaling. Functional modulation of type III TGF-beta receptor expression through interaction with the PDZ domain protein, GIPC. J. Biol. Chem. 276:39608-39617.
Booth, R. A., C. Cummings, M. Tiberi, and X. J. Liu. 2002. GIPC participates in G protein signaling downstream of insulin-like growth factor 1 receptor. J. Biol. Chem. 277:6719-6725.
Bunn, R. C., M. A. Jensen, and B. C. Reed. 1999. Protein interactions with the glucose transporter binding protein GLUT1CBP that provide a link between GLUT1 and the cytoskeleton. Mol. Biol. Cell 10:819-832.
Cai, H., and R. R. Reed. 1999. Cloning and characterization of neuropilin-1-interacting protein: a PSD-95/Dlg/ZO-1 domain-containing protein that interacts with the cytoplasmic domain of neuropilin-1. J. Neurosci. 19:6519-6527.
Chen, L., C. Klass, and A. Woods. 2004. Syndecan-2 regulates transforming growth factor-beta signaling. J. Biol. Chem. 279:15715-15718.
Desbois, C., R. Rousset, F. Bantignies, and P. Jalinot. 1996. Exclusion of Int-6 from PML nuclear bodies by binding to the HTLV-I Tax oncoprotein. Science 273:951-953.
De Vries, L., X. Lou, G. Zhao, B. Zheng, and M. G. Farquhar. 1998. GIPC, a PDZ domain containing protein, interacts specifically with the C terminus of RGS-GAIP. Proc. Natl. Acad. Sci. USA. 95:12340-12345.
Elbashir, S. M., J. Harborth, W. Lendeckel, A. Yalcin, K. Weber, and T. Tuschl. 2001. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411:494-498.
El Mourabit, H., P. Poinat, J. Koster, H. Sondermann, V. Wixler, E. Wegener, E. Laplantine, D. Geerts, E. Georges-Labouesse, A. Sonnenberg, and M. Aumailley. 2002. The PDZ domain of TIP-2/GIPC interacts with the C-terminus of the integrin alpha5 and alpha6 subunits. Matrix Biol. 21:207-214.
Fabre, S., C. Reynaud, and P. Jalinot. 2000. Identification of functional PDZ domain binding sites in several human proteins. Mol. Biol. Rep. 27:217-224.
Gao, Q., A. Kumar, L. Singh, J. M. Huibregtse, S. Beaudenon, S. Srinivasan, D. E. Wazer, H. Band, and V. Band. 2002. Human papillomavirus E6-induced degradation of E6TP1 is mediated by E6AP ubiquitin ligase. Cancer Res. 62:3315-3321.
Gao, Y., M. Li, W. Chen, and M. Simons. 2000. Synectin, syndecan-4 cytoplasmic domain binding PDZ protein, inhibits cell migration. J. Cell Physiol. 184:373-379.
Gardiol, D., C. Kuhne, B. Glaunsinger, S. S. Lee, R. Javier, and L. Banks. 1999. Oncogenic human papillomavirus E6 proteins target the discs large tumour suppressor for proteasome-mediated degradation. Oncogene 18:5487-5496.
Glaunsinger, B. A., S. S. Lee, M. Thomas, L. Banks, and R. Javier. 2000. Interactions of the PDZ-protein MAGI-1 with adenovirus E4-ORF1 and high-risk papillomavirus E6 oncoproteins. Oncogene 19:5270-5280.
Green, S., I. Issemann, and E. Sheer. 1988. A versatile in vivo and in vitro eukaryotic expression vector for protein engineering. Nucleic Acids Res. 16:369.
Hall, A. H., and K. A. Alexander. 2003. RNA interference of human papillomavirus type 18 E6 and E7 induces senescence in HeLa cells. J. Virol. 77:6066-6069.
Harlow, E., and D. Lane. 1988. Antibodies: a laboratory manual, p. 447-470. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
Hirakawa, T., C. Galet, M. Kishi, M. Ascoli, L. A. Hu, W. Chen, N. P. Martin, E. J. Whalen, R. T. Premont, and R. J. Lefkowitz. 2003. GIPC binds to the human lutropin receptor (hLHR) through an unusual PDZ domain binding motif, and it regulates the sorting of the internalized human choriogonadotropin and the density of cell surface hLHR. J. Biol. Chem. 278:49348-49357.
Hirata, A., M. Higuchi, A. Niinuma, M. Ohashi, M. Fukushi, M. Oie, T. Akiyama, Y. Tanaka, F. Gejyo, and M. Fujii. 2004. PDZ domain-binding motif of human T-cell leukemia virus type 1 Tax oncoprotein augments the transforming activity in a rat fibroblast cell line. Virology 318:327-336.
Hsueh, Y. P., T. F. Wang, F. C. Yang, and M. Sheng. 2000. Nuclear translocation and transcription regulation by the membrane-associated guanylate kinase CASK/LIN-2. Nature 404:298-302.
Hung, A. Y., and M. Sheng. 2002. PDZ domains: structural modules for protein complex assembly. J. Biol. Chem. 277:5699-5702.
Jeanneteau, F., J. Diaz, P. Sokoloff, and N. Griffon. 2004. Interactions of GIPC with dopamine D2, D3 but not D4 receptors define a novel mode of regulation of G protein-coupled receptors. Mol. Biol. Cell 15:696-705.
Kang, Y., C. R. Chen, and J. Massague. 2003. A self-enabling TGFbeta response coupled to stress signaling: Smad engages stress response factor ATF3 for Id1 repression in epithelial cells. Mol. Cell 11:915-926.
Kee, B. L., R. R. Rivera, and C. Murre. 2001. Id3 inhibits B lymphocyte progenitor growth and survival in response to TGF-beta. Nat. Immunol. 2:242-247.
Kim, E., M. Niethammer, A. Rothschild, Y. N. Jan, and M. Sheng. 1995. Clustering of Shaker-type K+ channels by interaction with a family of membrane-associated guanylate kinases. Nature 378:85-88.
Kirikoshi, H., and M. Katoh. 2002. Molecular cloning and characterization of human GIPC2, a novel gene homologous to human GIPC1 and Xenopus Kermit. Int. J. Oncol. 20:571-576.
Kiyono, T., A. Hiraiwa, M. Fujita, Y. Hayashi, T. Akiyama, and M. Ishibashi. 1997. Binding of high-risk human papillomavirus E6 oncoproteins to the human homologue of the Drosophila discs large tumor suppressor protein. Proc. Natl. Acad. Sci. USA 94:11612-11616.
Kornau, H. C., L. T. Schenker, M. B. Kennedy, and P. H. Seeburg. 1995. Domain interaction between NMDA receptor subunits and the postsynaptic density protein PSD-95. Science 269:1737-1740.
Kuhne, C., D. Gardiol, C. Guarnaccia, H. Amenitsch, and L. Banks. 2000. Differential regulation of human papillomavirus E6 by protein kinase A: conditional degradation of human discs large protein by oncogenic E6. Oncogene 19:5884-5891.
Lee, S. S., B. Glaunsinger, F. Mantovani, L. Banks, and R. T. Javier. 2000. Multi-PDZ domain protein MUPP1 is a cellular target for both adenovirus E4-ORF1 and high-risk papillomavirus type 18 E6 oncoproteins. J. Virol. 74:9680-9693.
Lee, S. S., R. S. Weiss, and R. T. Javier. 1997. Binding of human virus oncoproteins to hDlg/SAP97, a mammalian homolog of the Drosophila discs large tumor suppressor protein. Proc. Natl. Acad. Sci. USA 94:6670-6675.
Liu, T. F., G. Kandala, and V. Setaluri. 2001. PDZ domain protein GIPC interacts with the cytoplasmic tail of melanosomal membrane protein gp75 (tyrosinase-related protein-1). J. Biol. Chem. 276:35768-35777.
Lou, X., H. Yano, F. Lee, M. V. Chao, and M. G. Farquhar. 2001. GIPC and GAIP form a complex with TrkA: a putative link between G protein and receptor tyrosine kinase pathways. Mol. Biol. Cell 12:615-627.
Nakagawa, S., and J. M. Huibregtse. 2000. Human scribble (Vartul) is targeted for ubiquitin-mediated degradation by the high-risk papillomavirus E6 proteins and the E6AP ubiquitin-protein ligase. Mol. Cell. Biol. 20:8244-8253.
Nguyen, M. L., M. M. Nguyen, D. Lee, A. E. Griep, and P. F. Lambert. 2003. The PDZ ligand domain of the human papillomavirus type 16 E6 protein is required for E6's induction of epithelial hyperplasia in vivo. J. Virol. 77:6957-6964.
Ohashi, M., M. Sakurai, M. Higuchi, N. Mori, M. Fukushi, M. Oie, R. J. Coffey, K. Yoshiura, Y. Tanaka, M. Uchiyama, M. Hatanaka, and M. Fujii. 2004. Human T-cell leukemia virus type 1 Tax oncoprotein induces and interacts with a multi-PDZ domain protein, MAGI-3. Virology 320:52-62.
Pim, D., M. Thomas, R. Javier, D. Gardiol, and L. Banks. 2000. HPV E6 targeted degradation of the discs large protein: evidence for the involvement of a novel ubiquitin ligase. Oncogene 19:719-725.
Rolfe, M., P. Beer-Romero, S. Glass, J. Eckstein, I. Berdo, A. Theodoras, M. Pagano, and G. Draetta. 1995. Reconstitution of p53-ubiquitinylation reactions from purified components: the role of human ubiquitin-conjugating enzyme UBC4 and E6-associated protein (E6AP). Proc. Natl. Acad. Sci. USA 92:3264-3268.
Rousset, R., S. Fabre, C. Desbois, F. Bantignies, and P. Jalinot. 1998. The C-terminus of the HTLV-1 Tax oncoprotein mediates interaction with the PDZ domain of cellular proteins. Oncogene 16:643-654.
Saitoh, T., T. Mine, and M. Katoh. 2002. Molecular cloning and characterization of human GIPC3, a novel gene homologous to human GIPC1 and GIPC2. Int. J. Oncol. 20:577-582.
Songyang, Z., A. S. Fanning, C. Fu, J. Xu, S. M. Marfatia, A. H. Chishti, A. Crompton, A. C. Chan, J. M. Anderson, and L. C. Cantley. 1997. Recognition of unique carboxyl-terminal motifs by distinct PDZ domains. Science 275:73-77.
Suzuki, T., Y. Ohsugi, M. Uchida-Toita, T. Akiyama, and M. Yoshida. 1999. Tax oncoprotein of HTLV-1 binds to the human homologue of Drosophila discs large tumor suppressor protein, hDLG, and perturbs its function in cell growth control. Oncogene 18:5967-5972.
Tani, T. T., and A. M. Mercurio. 2001. PDZ interaction sites in integrin alpha subunits. T14853, TIP/GIPC binds to a type I recognition sequence in alpha 6A/alpha 5 and a novel sequence in alpha 6B. J. Biol. Chem. 276:36535-36542.
Thomas, M., R. Laura, K. Hepner, E. Guccione, C. Sawyers, L. Lasky, and L. Banks. 2002. Oncogenic human papillomavirus E6 proteins target the MAGI-2 and MAGI-3 proteins for degradation. Oncogene 21:5088-5096.
Thomas, M. K., K. M. Yao, M. S. Tenser, G. G. Wong, and J. F. Habener. 1999. Bridge-1, a novel PDZ-domain coactivator of E2A-mediated regulation of insulin gene transcription. Mol. Cell. Biol. 19:8492-8504.
Wang, L. H., R. G. Kalb, and S. M. Strittmatter. 1999. A PDZ protein regulates the distribution of the transmembrane semaphorin, M-SemF. J. Biol. Chem. 274:14137-14146.
Watson, R. A., M. Thomas, L. Banks, and S. Roberts. 2003. Activity of the human papillomavirus E6 PDZ-binding motif correlates with an enhanced morphological transformation of immortalized human keratinocytes. J. Cell Sci. 116:4925-4934.
Zhou, H., I. Wertz, K. O'Rourke, M. Ultsch, S. Seshagiri, M. Eby, W. Xiao, and V. M. Dixit. 2004. Bcl10 activates the NF-kappaB pathway through ubiquitination of NEMO. Nature 427:167-171.(Arnaud Favre-Bonvin, Caro)
ABSTRACT
Several viral proteins expressed by DNA or RNA transforming viruses have the particular property of binding via their C-terminal end to various cellular proteins with PDZ domains. This study is focused on the PDZ protein TIP-2/GIPC, which was originally identified in two-hybrid screens performed with two different baits: the human T-cell leukemia virus type 1 Tax oncoprotein and the regulator of G signaling RGS-GAIP. Further studies have shown that TIP-2/GIPC is also able to associate with the cytoplasmic domains of various transmembrane proteins. In this report we show that TIP-2/GIPC interacts with the E6 protein of human papillomavirus type 18 (HPV-18). This event triggers polyubiquitination and proteasome-mediated degradation of the cellular protein. In agreement with this observation, silencing of E6 by RNA interference in HeLa cells causes an increase in the intracellular TIP-2/GIPC level. This PDZ protein has been previously found to be involved in transforming growth factor ? (TGF-?) signaling by favoring expression of the TGF-? type III receptor at the cell membrane. In line with this activity of TIP-2/GIPC, we observed that depletion of this protein in HeLa cells hampers induction of the Id3 gene by TGF-? treatment and also diminishes the antiproliferative effect of this cytokine. Conversely, silencing of E6 increases the expression of Id3 and blocks proliferation of HeLa cells. These results support the notion that HPV-18 E6 renders cells less sensitive to the cytostatic effect of TGF-? by lowering the intracellular amount of TIP-2/GIPC.
INTRODUCTION
In addition to the function they exert in virus replication, viral proteins often strongly interfere with cell physiology by interacting with cellular factors involved in important regulatory networks. Hence, the viral proteins, especially those expressed by DNA or RNA transforming viruses, represent interesting models to better understand cellular regulatory pathways. Along this line, the discovery of the ability of proteins expressed by different viruses to bind various cellular proteins with PSD-95-Discs-large-ZO1 (PDZ) domains has open a new and interesting field of investigation. The PDZ domain is indeed present in many human proteins, as either single or multiple copies, associated or not with other protein domains (for recent reviews see references 4 and 25). The PDZ proteins exert diverse functions, generally having an important organizing role by establishing specific interactions with multiple partners. The archetypal interaction is association of the PDZ domain with the C-terminal end of another protein, with the specificity of the interaction depending on its C-terminal four amino acids and the consensus motif being XS/TXV-COOH for class I PDZ domains (29, 32, 45). In addition to this general rule, it is now well established that the PDZ domain can also interact with an internal sequence motif and also with another PDZ domain (4, 25). Relying on these properties, PDZ proteins play important roles in transmembrane receptors assembly and clustering, as well as in organization of cellular junctions. This family of proteins also intervenes in signal transduction, and examples of transcriptional regulation activities exist (24, 49). Unexpectedly, it has been observed that three different viral transforming proteins, i.e., E4 open reading frame 1 of adenovirus type 9, E6 of human papillomavirus type 18 (HPV-18), and Tax of human T-cell leukemia virus type 1 (HTLV-1), were able to bind PDZ proteins (31, 35, 43). These three proteins include a canonical class I PDZ binding site (BS). For E4 and E6, deletion of this C-terminal PDZ BS triggers a loss of in vitro transforming activity (31). The PDZ BS of Tax of HTLV-1 also plays a role in in vitro transformation of Rat-1 cells (23). In line with these observations, it has been observed that a PDZ BS is present at the C terminus of E6 in malignant HPV types, such as HPV-18 or HPV-16, but not in the benign ones, such as HPV-11 (31). Similarly, Tax of HTLV-2, which is not associated with aggressive leukemia, does not include a functional PDZ BS (23). Collectively, these observations strongly support the notion that a relationship exists between in vivo transformation and interference with the normal functioning of PDZ proteins. Hence, it appears to be important to characterize the complete set of cellular PDZ proteins targeted by these viral oncoproteins. By performing a two-hybrid screen of a human B-lymphocyte cDNA library with Tax as bait, we have previously identified seven different PDZ proteins: TIP-1, -2, -15, -33, -40, and -43 and hDlg-1 (14, 43). Other reports have shown that hDlg-1 is bound by the three viral proteins Tax, E6, and E4 open reading frame 1 (17, 31, 35, 46). E6 has also been shown to interact with hScrib, MUPP1, and MAGI-1, -2, and -3 (18, 34, 38, 48), and recently the interaction of Tax with MAGI-3 was reported (40).
In this work, we examined whether TIP-2, which has been identified as interacting with Tax, also represents a target for the HPV-18 E6 protein. A more precise analysis of the relationship of this protein with viral oncoproteins is interesting, as various publications have shown that TIP-2 is able to interact with a great variety of cytoplasmic or transmembrane proteins. It has indeed been shown that TIP-2 interacts with the GTPase-activating protein for Gi GAIP (11). From this property, TIP-2 was named GIPC (for GAIP-interacting protein, C terminus) (11). Genes coding for homologous proteins have been identified in humans and named GIPC2 and GIPC3 (30, 44). This cellular protein has also been identified as a possible partner of the following proteins: the rat glucose transporter GLUT1 (under the name GLUT1CBP) (7), the mouse semaphorin MSemF (under the name SEMCAP-1) (50), the rat neuropilin 1 (under the name NIP) (8), the mouse syndecan-4 proteoglycan (under the name synectin) (16), TrkA and TrkB (37), the melanosomal protein gp75 (36), the 5 and 6 integrins (13, 47), the Xenopus IGF1 receptor (6), the human 5T4 oncofetal antigen (2), myosin VI (1, 7), the dopamine D2 and D3 receptors (26), the lutropin receptor (22), the ?1-adrenergic receptor, and the transforming growth factor ? (TGF-?) type III receptor (T?RIII) (5). For the sake of clarity, the name TIP-2/GIPC is used here.
Evidence that E6 of HPV-18 binds TIP-2/GIPC and triggers its degradation by the proteasome is presented in this report. Silencing of TIP-2/GIPC by RNA interference also showed that this PDZ protein is important for the effect of TGF-? on gene expression and cell proliferation. Our observations support the notion that destruction of TIP-2/GIPC by HPV-18 E6 contributes to cell transformation by hampering TGF-? signaling.
MATERIALS AND METHODS
Constructs. The pTL1-TIP-2 expression vector was generated as follows. The TIP-2/GIPC sequence was amplified from an expressed sequence tag (IMAGE clone 2958330) by using the sense and antisense primers 5'-GACCCCACTTCTCGCTGCTCATG-3' and 5'-GGGGGATCCTAGTAGCGGCCG-3', respectively. The PCR fragment was cloned in PCR-Script (Stratagene), and the NotI-PstI restriction fragment containing the TIP-2/GIPC coding sequence was inserted between the NotI and PstI restriction sites of pTL1, which is a pSG5 derivative (19). The pSGF-E6 vector, expressing the N-terminally FLAG-tagged E6 protein from HPV-18, was constructed as follows: the E6-coding sequence was amplified from plasmid pHPV18 (kindly provided by F. Thierry) with the sense and antisense primers 5'-GCGAATTCGCTCGAGTTGAGGATCCAACACGG-3' and 5'-CATAGTCGACATTATACTTGTGTTTCTC-3', respectively, and then digested with XhoI and SalI and inserted in the XhoI restriction site of pSG-FLAG (10) in the sense orientation. The vectors expressing the FLAG-tagged E6 proteins mutated in the PDZ binding site were generated as follows. The E6-coding sequence was amplified from pSGF-E6 with the sense primer 5'-GTAATACGACTCACTATAGGG-3' and the antisense primers 5'-CATAGTCGACATTATACTTGTGCTTCTCTGCGTC-3', 5'-CATAGTCGACATTATGCTTGTGTTTCTC-3', 5'-CATAGTCGACATTATGCTTGTGCTTCTCTGCGTC-3', and 5'-CATAGTCGACATTATCTGCGTCGTTGGAG-3' to generate the T156A, V158A, TAVA, and 4C mutants, respectively. The PCR fragments were digested with EcoRI and XhoI and inserted between the EcoRI and SalI restriction sites of pSGF. Plasmids expressing hemagglutinin (HA)-tagged ubiquitin, either wild type or with all lysines except lysine 48 or lysine 63 mutated to arginine (kindly provided by V. Dixit), have been previously described (52).
Two-hybrid assay. Plasmid pGBT9-E6, which was used as bait, was generated as described above for pSGF-E6, except that the PCR fragment was digested with the EcoRI and SalI restriction enzymes and inserted in the pGBT9 plasmid (Clontech) between the EcoRI and SalI restriction site. The clone –1, –2, –15, –33, –40, and –43 prey plasmids were those obtained from the two-hybrid screen against Tax (43). Two-hybrid assay, analysis of growth on minimal medium lacking histidine, and examination of ?-galactosidase expression were performed as previously described (43).
Cell culture and transfection. HeLa, COS7, and CV-1 cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (FBS), 100 U of penicillin ml–1, and 100 μg of streptomycin ml–1 at 37°C in a 5% CO2 humidified atmosphere. For plasmid transfection, the amount of FBS used in the culture medium was reduced to 5%. Transfection experiments were performed by the calcium phosphate precipitation method in 100-mm-diameter petri dishes with the amounts of expression vector indicated in the figure legends. The total amount of DNA was adjusted to 15 μg with pBlueScript plasmid. For transfection of small interfering RNAs (siRNAs) (12), 50,000 cells were seeded in 12-well plates in Dulbecco's modified Eagle's medium without antibiotics supplemented with 5% FBS the day before. Five microliters of 20 μM siRNA duplex was mixed with 40 μl of Opti-MEM I. Separately, 4.5 μl of Oligofectamine reagent (Invitrogen) was added to 24 μl of Opti-MEM I, and the mixture was incubated for 8 min at room temperature. The two mixtures were combined, incubated at room temperature for 25 min, and added to the cells. When transfection was done with different sizes of tissue culture supports, the amounts of cells, siRNAs, and Oligofectamine were adjusted proportionally to the surface area. The sequence of the sense siRNA for TIP-2/GIPC was 5'-GCCAACUGCCGAGGUGAUGTT-3'. The siRNA duplex used for E6/E7 is the one reported previously (20).
Antibodies. Rabbit polyclonal antibodies were generated against a peptide corresponding to the C-terminal 19 amino acids of TIP-2/GIPC. The peptide was chemically synthesized and coupled to ovalbumin. The TIP-2/GIPC antibodies obtained were affinity purified against a protein corresponding to the C-terminal 85 amino acids of TIP-2/GIPC. The clone 2 plasmid (43) was digested with XmaI, and the fragment coding for the C terminus of TIP-2/GIPC was inserted in the XmaI restriction site of PinPoint Xa1 (Promega) in the sense orientation. The HB101 Escherichia coli strain was transformed with the resulting plasmid. Expression of the fusion protein was induced in a 500-ml overnight culture in presence of 2 μM biotin by treatment with 100 μM IPTG (isopropyl-?-D-thiogalactopyranoside) for 4 h at 37°C. Bacteria were lysed by sonication in lysis buffer consisting of 50 mM Tris (pH 7.5), 50 mM NaCl, and 5% glycerol. After centrifugation, the bacterial lysate was loaded on a column containing 4 ml of streptavidin agarose beads equilibrated in phosphate-buffered saline (PBS). The lysate was passed three times through the column. After a 10-ml wash with lysis buffer, the column was loaded with 10 ml of antiserum to TIP-2/GIPC. The column was then successively loaded with 10 ml of 0.3 M NaCl, 10 ml of PBS-0.1% Tween 20, and 10 ml of PBS. The retained immunoglobulins were eluted with 15 ml of 0.1 M glycine (pH 3.5). The purified antibodies were collected in 1.5-ml tubes containing 100 μl of 1 M Tris (pH 8.0). A rabbit polyclonal antibody was similarly raised against a peptide corresponding to amino acids 323 to 341 of TIP-40 but was used without affinity purification. For FLAG-tagged proteins we used the mouse monoclonal antibody M2 (Sigma), and for HA-tagged proteins we used the mouse monoclonal antibody clone 12CA5 (Roche). Rabbit polyclonal antibody to ubiquitin-protein conjugates was purchased from Affiniti (UG9510), and monoclonal antibody to ?-actin was purchased from Sigma.
Immunoprecipitation and immunoblotting. Transfected COS7 cells were lysed for 20 min in 500 μl of radioimmunoprecipitation assay (RIPA) buffer (21) supplemented with 10 mM iodoacetamide. Lysates were centrifuged for 10 min at 15,000 x g, and the protein concentration was measured with the Dc protein assay kit (Bio-Rad). The supernatants, adjusted to equal amounts of total protein, were incubated with antibodies diluted 1:250. After incubation at 4°C for 2 h, 30 μl of protein A beads equilibrated in RIPA buffer was added to the mixtures, which were further incubated for an additional hour. Protein A beads were collected by centrifugation at 425 x g for 2 min and washed three times in RIPA buffer plus iodoacetamide. Proteins were eluted in sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis loading buffer at 80°C for 10 min. After separation by SDS-polyacrylamide gel electrophoresis, proteins were transferred to a polyvinylidene difluoride membrane. For immunoblotting, primary antibodies were used diluted 1:1,000 (for antibodies to TIP-2/GIPC, FLAG, HA, and ubiquitin-protein conjugates), 1:4,000 (for antibody to ?-actin), or 1:500 (for antibodies to p53 and TIP-40), and secondary antibodies were diluted 1:6,000. Visualization was performed by chemiluminescence with the ECL reagent (Amersham Pharmacia Biotech).
Real-time quantitative RT-PCR. After transfection of siRNAs, HeLa cells were collected and total RNAs were extracted by using the RNeasy minikit (Qiagen). One-step reverse transcription-PCRs (RT-PCRs) were performed with the use of the QuantiTect SYBR Green RT-PCR kit (Qiagen) and the LightCycler apparatus (Roche) according to the cycling conditions specified in the handbook for the kit. Gene-specific primers were designed by using the Primer3 software. The sequences of the sense and antisense primers, respectively, used for quantitative PCR were as follows: Id3, 5'-GGAGCTTTTGCCACTGACTC-3' and 5'-TTCAGGCCACAAGTTCACAG-3'; lamin, 5'-CCGAGTCTGAAGAGGTGGTC-3' and 5'-AGGTCACCCTCCTTCTTGGT-3'; Hmg1, 5'-AAGCACCCAGATGCTTCAGT-3' and 5'-GTGCATTGGGATCCTTGAA-3'; cyclin B1, 5'-CGGGAAGTCACTGGAAACAT-3' and 5'-AAACATGGCAGTGACACCAA-3'; Set, 5'-GCAAGAAGCGATTGAACACA-3' and 5'-GCAGTGCCTCTTCATCTTCC-3'; Plk1, 5'-AAGAGATCCCGGAGGTCCTA-3' and 5'-GCTGCGGTGAATGGATATTT-3'; Slc20A, 5'-AGCGTGGACTTGAAAGAGGA-3' and 5'-TCTTTGTACAGGCCGGAATC-3'; and TIP-2/GIPC, 5'-CTCCACCACTTTCCACCATC-3' and 5'-GAGGTAACAGGCTCCACAGG-3'.
Thymidine incorporation assays. After transfection of siRNAs and 4 h before harvest, 5 μCi of [methyl-3H]thymidine was added to the cell culture medium. The cells were washed twice in ice-cold trichloroacetic acid and lysed in 0.2 N NaOH. The protein concentration was measured with the Dc protein assay kit (Bio-Rad). Cell lysates (20 μl) were added to Supermix (Perkin-Elmer), and radioactivity was counted in a Wallac apparatus with 3H mode. Radioactivity values were normalized with respect to protein concentrations.
RESULTS
Interaction of HPV-18 E6 with TIP-2/GIPC in two-hybrid assays. As a first attempt to determine whether HPV-18 E6 can bind TIP-2/GIPC, two-hybrid assays were performed with E6 fused to the GAL4 DNA binding domain as bait, along with the various clones coding for PDZ proteins obtained in the Tax two-hybrid screen (clones –1, –2, –15, –33, –40, and –43) (43) as prey. The results of these assays performed with the HF7C strain were evaluated by both growth on minimal medium lacking histidine and expression of the ?-galactosidase reporter protein. E6 showed no association with the GAL4 DNA binding domain alone (Table 1). Growth on medium without histidine was positive for all clones, with the exception of that encoding TIP-40. However, ?-galactosidase expression was clearly positive only for TIP-15 and TIP-2/GIPC. These observations strongly suggested that TIP-2/GIPC is also targeted by E6 of HPV-18.
Binding to E6 triggers degradation of TIP-2/GIPC by the proteasome. Considering the positive result of the two-hybrid assay, whether E6 and TIP-2/GIPC interact in the context of mammalian cells was further tested. As it is well documented that E6 can cause degradation of cellular proteins to which it binds, including those with a PDZ domain, analysis of the interaction between the two proteins was performed by immunoprecipitation experiments either with or without treatment of the cells with the proteasome inhibitors MG132 and lactacystine. COS7 cells were transfected with vectors expressing a FLAG-tagged HPV-18 E6, as well as TIP-2/GIPC, either in the wild-type form or fused to enhanced green fluorescent protein (GFP) at its N-terminal end. Immunoblot analysis with an antibody directed against TIP-2/GIPC of the extracts used in these experiments showed that the proteasome inhibitor treatment did not modify the amount of either TIP-2/GIPC or GFP-TIP-2 (Fig. 1B). This was also the case for FLAG-E6 expression (Fig. 1D). Extracts of these transfected cells were used for immunoprecipitation with the M2 monoclonal antibody to FLAG. Immunoblot analysis using the polyclonal antibody directed against TIP-2/GIPC revealed its presence in the precipitated proteins, but this was evident mainly when cells were treated with MG132 (Fig. 1A, lanes 1 and 2). Use of a different antibody in the immunoprecipitation did not trigger any TIP-2/GIPC, indicating that the reaction was specific (Fig. 1A, lane 3). The GFP-TIP-2 fusion protein was also specifically precipitated by FLAG-E6 (Fig. 1A, lanes 4, 5, 7, and 8). In this case the signal was clearly visible in the absence of proteasome inhibitor treatment but was reinforced when cells were incubated with either MG132 or lactacystine (Fig. 1A, compare lanes 4 and 5 and lanes 7 and 8). These experiments were also performed in the opposite way, i.e., by precipitating TIP-2/GIPC and analyzing the presence of FLAG-E6 in immunoprecipitated proteins. The results obtained were symmetrical. FLAG-E6 was coprecipitated with TIP-2/GIPC, but only when cells were treated with MG132 (Fig. 1C, compare lanes 1 and 2), and coprecipitation with GFP-TIP-2 was augmented by either MG132 or lactacystine treatment (Fig. 1C, compare lanes 4 and 5 and lanes 7 and 8). It was verified that the interaction between E6 and TIP-2/GIPC relies on the E6 PDZ BS by altering the C-terminal end of the latter protein. Either the C-terminal valine or the threonine at position –2 was mutated to alanine. These two mutations did not impair interaction of E6 with TIP-2/GIPC as evaluated by MG132 treatment (Fig. 1E, lanes 1 and 2). However mutation of both residues markedly reduced it (Fig. 1E, lane 3), and deletion of the C-terminal four amino acids led to a complete disappearance of E6 coprecipitated with TIP-2/GIPC (Fig. 1E, lane 4). These four mutants were expressed at equal levels in cells (Fig. 1F). These data clearly show that E6 binds TIP-2/GIPC via its PDZ BS in mammalian cells, but they also indicate that this event cause proteasome-mediated degradation of the cellular protein. The GFP-TIP-2 fusion protein is probably less prone to degradation than wild-type TIP-2/GIPC, but some proteolysis also occurred, since the proteasome inhibitor treatment increased the amount of coprecipitated protein.
It has been established for several factors associating with E6 that this viral protein induces their polyubiquitination by binding concomitantly to the E3 ubiquitin ligase E6AP (15, 38, 42). In this regard, we examined whether E6 induces polyubiquitination of TIP-2/GIPC. Cells were transfected with expression vectors for both proteins together with vectors producing HA-tagged ubiquitin either in the wild-type form or with each lysine except lysine 48 or lysine 63 mutated to arginine (52). Immunoprecipitation was done with the antibody to TIP-2/GIPC, and immunoblot analysis was performed with a monoclonal antibody directed against HA. These experiments clearly showed that E6 stimulates polyubiquitination of TIP-2/GIPC (Fig. 2). This was seen with wild-type ubiquitin but also with ubiquitin allowing branching only on lysine 48 (Fig. 2A, lanes 2 and 4). The smear of polyubiquitinated forms of TIP-2/GIPC was less intense with ubiquitin bearing a single lysine at position 63 (Fig. 2A, lane 6). This result suggests that E6 may also stimulate polyubiquitination by K63 branching. Alternatively, it is possible that this observation results from incorporation of tagged mutated ubiquitin in a chain bearing endogenous wild-type molecules. In order to exclude that this polyubiquitination of TIP-2/GIPC resulted from ubiquitin overexpression, CV-1 cells were transfected with either the control vector pSGF or a vector expressing FLAG-tagged E6, and endogenous TIP-2/GIPC was immunoprecipitated and analyzed by immunoblotting with an antibody raised against ubiquitin-protein conjugates. This showed that expression of E6 in these cells increases the intensities of the bands corresponding to TIP-2/GIPC polyubiquitinated forms (Fig. 2B, compare lanes 3 and 4), with this being observed in the absence of ubiquitin overexpression. These observations clearly show that E6 binding to TIP-2/GIPC causes its polyubiquitination, mainly with lysine 48 branching.
To establish that E6 can destabilize TIP-2/GIPC under physiological conditions, we looked at the effect of E6 silencing on the intracellular level of this cellular protein in HeLa cells. It has recently been reported that depleting these cells of E6 and E7 by the RNA interference approach causes a rapid increase in the amount of p53 (20) (Fig. 3A, compare lanes 1 and 2). When an extract of cells treated with siRNAs directed against E6 and E7 were analyzed by immunoblotting for TIP-2/GIPC, a clear increase in the amount of this protein was also observed (Fig. 3B, lane 2). Such an effect was not observed when cells were transfected with control siRNAs (Fig. 3B, lane 1, and data not shown). In sum, by associating with TIP-2/GIPC, E6 triggers its polyubiquitination and proteasome-mediated degradation. This activity decreases the normal level of endogenous TIP-2/GIPC in cells expressing E6.
TIP-2/GIPC silencing inhibits Id3 expression. As a first step towards understanding the consequences of a decrease in TIP-2/GIPC for the cell, we examined the effect of its silencing by RNA interference on gene expression. A microarray analysis performed with a set of 1,853 genes involved in cell transformation showed that expression of Id3, which encodes a helix-loop-helix protein, was reduced in TIP-2/GIPC-silenced cells. To confirm this observation, the level of the Id3 mRNA in these cells was evaluated by quantitative PCR, along with that of a set of unrelated genes taken as controls. The TIP-2 mRNA was also included in the analysis. It was first verified by immunoblotting that the siRNA duplex directed against TIP-2 efficiently decreased the amount of protein (Fig. 4B, compare lane 2 with lanes 1 and 3). Immunoblot analysis using antibodies directed against another PDZ protein, TIP-40, or ?-actin showed that this effect was specific (Fig. 4C and D).
As expected, quantitation of the TIP-2 mRNA showed a marked decrease in siRNA-transfected cells, to 18.5% of its normal level (Fig. 4E). Under these experimental conditions, approximately 85% of the cells were transfected as evaluated by immunofluorescence analysis (data not shown). This indicates that the procedure almost completely removed TIP-2 RNA from the transfected cells. With the control genes no significant effect was observed, whereas the level of Id3 mRNA was only 38% (Fig. 4E). After correction for nontransfected cells, this meant that the TIP-2/GIPC depletion causes a fivefold decrease of the Id3 mRNA in transfected cells. Interestingly, it has been reported that Id3, which is known to be important in the control of cell growth by interfering with key basic helix-loop-helix factors, is regulated by transforming growth factor ? (27, 28). As TIP-2/GIPC has been implicated in the response to this cytokine (5), we analyzed further the effect of TGF-?1 treatment on Id3 expression. We observed that in HeLa cells addition of TGF-?1 increased the expression of Id3 (Fig. 5A, bar 2). This effect was impaired by TIP-2/GIPC silencing (Fig. 5A, bar 3). Interestingly, E6 silencing also increased Id3 expression (Fig. 5A, bar 4), and its combination with TGF-?1 treatment slightly augmented the effect of TGF-?1 alone (Fig. 5A, bar 5). This effect is likely to result from the increase in the TIP-2/GIPC level that can be seen by analyzing the level of expression of this protein in these cells (Fig. 5B). Collectively, these results establish that TIP-2/GIPC is an important factor for regulation of Id3 expression and that its presence is necessary for a full effect of TGF-?1 on expression of this gene.
TIP-2/GIPC silencing activates cell growth. In line with these observations, we examined whether TIP-2/GIPC can intervene in the regulation of cell proliferation. As reported previously, silencing of E6/E7 in HeLa cells markedly alters multiplication of HeLa cells as evaluated by their ability to incorporate tritium-labeled thymidine (20) (Fig. 6). At 72 h after transfection of the siRNAs, the amount of incorporated radioactivity was less than 20% of that measured for control cells, corresponding to an almost complete block of cell division, considering that approximately 15% of the cells were nontransfected. Conversely, silencing of TIP-2/GIPC led to a clear increase in the proliferation of HeLa cells (Fig. 6). This experiment showed that this protein negatively controls cell proliferation. Its degradation by E6 is therefore likely to participate in the induction of an active proliferation by HeLa cells. We further examined the effect of TIP-2/GIPC in combination with TGF-?1 on cell proliferation. As expected, treatment of HeLa cells with TGF-?1 inhibited proliferation, which dropped to 52% of its normal value (Fig. 7A). When E6 and E7 were depleted by RNA interference, TGF-?1 treatment did not additionally reduce proliferation, but as mentioned above, removal of the viral oncoproteins blocked cell division. As shown above, TIP-2/GIPC silencing (Fig. 7B) accelerated cell division (Fig. 7A). Addition of TGF-?1 to these TIP-2/GIPC-silenced cells reduced this effect, but only by 22%. This indicated that loss of TIP-2/GIPC renders the cells less sensitive to the effect of TGF-?1. Therefore, as for Id3 expression, removal of TIP-2/GIPC at least partially impairs the effect of TGF-?1.
DISCUSSION
TIP-2/GIPC has been initially characterized as a PDZ protein interacting with HTLV-1 Tax and GAIP (11, 43). As Tax and HPV-18 E6 share the property of interacting with PDZ proteins, we asked whether TIP-2/GIPC also interacts with E6. Results from two-hybrid assays as well as from coimmunoprecipitation experiments clearly showed that this is indeed the case. These data show that TIP-2/GIPC must be added to the list of cellular PDZ proteins bound by E6 via its C-terminal end. This oncoprotein has already been shown to associate with hDlg-1, MUPP1, hScrib, and MAGI-1, -2, and -3 (17, 18, 31, 34, 35, 38, 48). Similarly to what has been observed for these proteins, HPV-18 E6 induces polyubiquitination and degradation of TIP-2/GIPC. For many targets of E6, including p53 and also PDZ proteins such as hScrib, it has been shown that this degradation induction results from E6 acting as an adaptor between the E3 ubiquitin ligase E6AP and the protein to be degraded (38, 42). By analogy, it is likely that the effect of E6 on TIP-2/GIPC also involves E6AP; however, this point remains to be firmly established, as the role of E6AP in the degradation of hDlg remains controversial (38, 41). Whatever the exact nature of the E3 ligase involved in the process, the experiments involving silencing of E6/E7 in HeLa cells provide clear evidence that E6 mediates continuous degradation of TIP-2/GIPC. As has been noted before for hDlg or other PDZ proteins, this does not entirely eliminate the protein from the cell. This can be due to the presence of a limited amount of E6 or to partial inactivation of its C-terminal PDZ BS by PKA phosphorylation, as it has been shown that this modification impairs binding of E6 to hDlg (33).
Considering the wide diversity of the proteins which have been shown to interact with TIP-2/GIPC, it is difficult at this stage to associate this protein with a precise function. Almost all characterizations of these interactions originated from two-hybrid screens. They generally involve the PDZ domain or a larger region of TIP-2/GIPC, along with the PDZ BS corresponding to the C terminus of the partner protein or, more rarely, to an internal PDZ BS, as for TrkA (37). One exception to these two-hybrid analyses is a study of the relationship between TIP-2/GIPC and TGF-? signaling. In that work TIP-2/GIPC was identified in a retroviral screen for factors rendering cells resistant to the cytostatic effect of TGF-? (5). Further analysis paradoxically showed that overexpression of TIP-2/GIPC sensitizes the cells to the effect of TGF-?. Those authors explained this activity by an association of TIP-2/GIPC with a TGF-? type III receptor which exhibits a PDZ BS at its C terminus. Overexpression of TIP-2/GIPC increases the amount of T?RIII at the cell surface and inhibits its degradation by the proteasome (5). In our analysis of the effect of TIP-2/GIPC silencing on expression of cellular genes, the effect observed on Id3 supports the notion of a role of this PDZ protein in TGF-? signaling. This cytokine is indeed known to regulate Id genes (27, 28). In different epithelial cell lines it has been shown that TGF-? represses transcription of the Id genes, including Id3 (27). By contrast, in HeLa cells we observed that TGF-? activates Id3. This has also been observed in B-lymphocyte progenitors (28). In these cells, Id3 protein overexpression triggers arrest of cell growth and apoptosis (28). The data obtained with these cells indicate a key role of Id3 in the effect of TGF-?. Repression of the Id1 promoter in epithelial cell lines such as HaCaT or MCF-10A is due to ATF3, which acts as a repressor in combination with Smad3 (27). The exact mechanism is not known for Id3, but by analogy it is possible that in HeLa and B-lymphocyte progenitors, a repressor factor such as ATF3 is not induced by TGF-?, and the Smads therefore act positively. In HeLa cells, depletion of TIP-2/GIPC by RNA interference clearly decreased the amount of Id3 mRNA. It also impaired its increase in response to treatment of cells with TGF-?. In line with these observations, silencing of E6, which increased the amount of TIP-2/GIPC, led to an increase of Id3 expression and also slightly augmented the effect of the cytokine. These observations support the notion that TIP-2/GIPC is necessary for the effect of TGF-? on Id3 transcription, and this can be explained by a positive effect of the PDZ protein on TGF-? receptors. However, an intriguing aspect is that TIP-2/GIPC has an effect in the absence of the cytokine, and also the microarray data do not show an effect on other Id genes. Therefore, there is something particular to Id3, and it cannot be excluded at this stage that TIP-2/GIPC mediates another signaling event that would specifically activate this gene. A detailed analysis of the promoter elements mediating the effects of both TIP-2/GIPC silencing and TGF-? treatment should allow us to clarify this point and to understand what is specific to Id3.
Besides the effect on Id3, involvement of TIP-2/GIPC in TGF-? signaling is also supported by the role that it plays in regulation of proliferation. Silencing of this PDZ protein in HeLa cells had a marked effect on their proliferation rate, which increased progressively starting from 48 h after transfection of siRNAs. By contrast, silencing of E6, which increased TIP-2/GIPC, was associated with a clear reduction of proliferation. Interestingly, the cytostatic effect of TGF-? was limited by depletion of TIP-2/GIPC. This likely results from a decreased sensitivity to the cytokine, which can be explained, similarly to the effect on Id3, by the demonstrated effect of TIP-2/GIPC on T?RIII (5). The effect of TIP-2/GIPC depletion in the absence of TGF-? could be explained by the possible presence of cytostatic cytokines of the TGF-? family in the medium. Collectively, these observations show that TIP-2/GIPC plays an important role in the control of proliferation and hence that its decrease due to interaction with E6 might directly promote cell growth. In future studies it will be interesting to determine whether this activity results only from the effect of TIP-2/GIPC on TGF-? signaling or involves other pathways. Regulation of the sensitivity of cells to TGF-? by TIP-2/GIPC might involve effects on proteins other than T?RIII. In this regard, it has been reported that TIP-2/GIPC interacts with syndecan-4 (16), and recently it has been proposed that it also interacts with syndecan-2, a transmembrane heparan sulfate proteoglycan which, like T?RIII, facilitates the effect of TGF-? (9).
In conclusion, there is now strong evidence that the E6 protein of the high-risk HPV type 18 affects a group of different PDZ proteins. By targeting proteins such as hDlg, MAGIs, hScrib, and MUPP1, it probably alters epithelium morphology by disrupting cell-cell contacts. It has been reported that deletion of the HPV-18 E6 PDZ BS impairs the morphological change induced by this protein in keratinocytes (51). In mouse it has also been established that this motif is necessary for epithelial hyperplasia (39). According to our results, alteration of TIP-2/GIPC by E6 possibly directly participates in this increased proliferation. Our data strengthen the importance of the PDZ proteins as cellular targets of the transforming activity of E6 and offer a new connection with the action of TGF-?. This cytokine has recently been shown to downregulate expression of E6/E7 through Ski and NFI (3). Both the role of TIP-2/GIPC in TGF-? signaling and the effect of E6 on this protein raise the possibility of a positive feedback loop in which degradation of TIP-2/GIPC renders cells less sensitive to TGF-?, thereby increasing the expression of E6. Our findings reinforce the interest in seeking therapeutic means to disrupt the interaction of high-risk HPV E6s with cellular PDZ proteins in order to impair the deleterious effects of these oncoproteins.
ACKNOWLEDGMENTS
We are grateful to F. Thierry and V. Dixit for the generous gift of plasmids, to A. Roisin for assistance with cell culture, and to S. Gonin-Giraud and O. Gandrillon for advice on TGF-? assays. We thank C. Morris for help with microarray and real-time quantitative PCR experiments.
This work was supported by grants from the Ligue Nationale contre le Cancer (Programme Equipes Labelisées) and the Comité du Rh?ne de la Ligue Nationale contre le Cancer (to A.F.-B.).
Present address: Institut de Biochimie et Chimie des Protéines, UMR 5086 CNRS UCBL, 69367 Lyon Cedex 07, France.
Present address: Centre de Génétique Moléculaire et Cellulaire, CNRS UMR 5534, 69622 Villeurbanne Cedex, France.
REFERENCES
Aschenbrenner, L., T. Lee, and T. Hasson. 2003. Myo6 facilitates the translocation of endocytic vesicles from cell peripheries. Mol. Biol. Cell 14:2728-2743.
Awan, A., M. R. Lucic, D. M. Shaw, F. Sheppard, C. Westwater, S. A. Lyons, and P. L. Stern. 2002. 5T4 interacts with TIP-2/GIPC, a PDZ protein, with implications for metastasis. Biochem. Biophys. Res. Commun. 290:1030-1036.
Baldwin, A., L. Pirisi, and K. E. Creek. 2004. NFI-Ski interactions mediate transforming growth factor beta modulation of human papillomavirus type 16 early gene expression. J. Virol. 78:3953-3964.
Bezprozvanny, I., and A. Maximov. 2001. PDZ domains: more than just a glue. Proc. Natl. Acad. Sci. USA 98:787-789.
Blobe, G. C., X. Liu, S. J. Fang, T. How, and H. F. Lodish. 2001. A novel mechanism for regulating transforming growth factor beta (TGF-beta) signaling. Functional modulation of type III TGF-beta receptor expression through interaction with the PDZ domain protein, GIPC. J. Biol. Chem. 276:39608-39617.
Booth, R. A., C. Cummings, M. Tiberi, and X. J. Liu. 2002. GIPC participates in G protein signaling downstream of insulin-like growth factor 1 receptor. J. Biol. Chem. 277:6719-6725.
Bunn, R. C., M. A. Jensen, and B. C. Reed. 1999. Protein interactions with the glucose transporter binding protein GLUT1CBP that provide a link between GLUT1 and the cytoskeleton. Mol. Biol. Cell 10:819-832.
Cai, H., and R. R. Reed. 1999. Cloning and characterization of neuropilin-1-interacting protein: a PSD-95/Dlg/ZO-1 domain-containing protein that interacts with the cytoplasmic domain of neuropilin-1. J. Neurosci. 19:6519-6527.
Chen, L., C. Klass, and A. Woods. 2004. Syndecan-2 regulates transforming growth factor-beta signaling. J. Biol. Chem. 279:15715-15718.
Desbois, C., R. Rousset, F. Bantignies, and P. Jalinot. 1996. Exclusion of Int-6 from PML nuclear bodies by binding to the HTLV-I Tax oncoprotein. Science 273:951-953.
De Vries, L., X. Lou, G. Zhao, B. Zheng, and M. G. Farquhar. 1998. GIPC, a PDZ domain containing protein, interacts specifically with the C terminus of RGS-GAIP. Proc. Natl. Acad. Sci. USA. 95:12340-12345.
Elbashir, S. M., J. Harborth, W. Lendeckel, A. Yalcin, K. Weber, and T. Tuschl. 2001. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411:494-498.
El Mourabit, H., P. Poinat, J. Koster, H. Sondermann, V. Wixler, E. Wegener, E. Laplantine, D. Geerts, E. Georges-Labouesse, A. Sonnenberg, and M. Aumailley. 2002. The PDZ domain of TIP-2/GIPC interacts with the C-terminus of the integrin alpha5 and alpha6 subunits. Matrix Biol. 21:207-214.
Fabre, S., C. Reynaud, and P. Jalinot. 2000. Identification of functional PDZ domain binding sites in several human proteins. Mol. Biol. Rep. 27:217-224.
Gao, Q., A. Kumar, L. Singh, J. M. Huibregtse, S. Beaudenon, S. Srinivasan, D. E. Wazer, H. Band, and V. Band. 2002. Human papillomavirus E6-induced degradation of E6TP1 is mediated by E6AP ubiquitin ligase. Cancer Res. 62:3315-3321.
Gao, Y., M. Li, W. Chen, and M. Simons. 2000. Synectin, syndecan-4 cytoplasmic domain binding PDZ protein, inhibits cell migration. J. Cell Physiol. 184:373-379.
Gardiol, D., C. Kuhne, B. Glaunsinger, S. S. Lee, R. Javier, and L. Banks. 1999. Oncogenic human papillomavirus E6 proteins target the discs large tumour suppressor for proteasome-mediated degradation. Oncogene 18:5487-5496.
Glaunsinger, B. A., S. S. Lee, M. Thomas, L. Banks, and R. Javier. 2000. Interactions of the PDZ-protein MAGI-1 with adenovirus E4-ORF1 and high-risk papillomavirus E6 oncoproteins. Oncogene 19:5270-5280.
Green, S., I. Issemann, and E. Sheer. 1988. A versatile in vivo and in vitro eukaryotic expression vector for protein engineering. Nucleic Acids Res. 16:369.
Hall, A. H., and K. A. Alexander. 2003. RNA interference of human papillomavirus type 18 E6 and E7 induces senescence in HeLa cells. J. Virol. 77:6066-6069.
Harlow, E., and D. Lane. 1988. Antibodies: a laboratory manual, p. 447-470. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
Hirakawa, T., C. Galet, M. Kishi, M. Ascoli, L. A. Hu, W. Chen, N. P. Martin, E. J. Whalen, R. T. Premont, and R. J. Lefkowitz. 2003. GIPC binds to the human lutropin receptor (hLHR) through an unusual PDZ domain binding motif, and it regulates the sorting of the internalized human choriogonadotropin and the density of cell surface hLHR. J. Biol. Chem. 278:49348-49357.
Hirata, A., M. Higuchi, A. Niinuma, M. Ohashi, M. Fukushi, M. Oie, T. Akiyama, Y. Tanaka, F. Gejyo, and M. Fujii. 2004. PDZ domain-binding motif of human T-cell leukemia virus type 1 Tax oncoprotein augments the transforming activity in a rat fibroblast cell line. Virology 318:327-336.
Hsueh, Y. P., T. F. Wang, F. C. Yang, and M. Sheng. 2000. Nuclear translocation and transcription regulation by the membrane-associated guanylate kinase CASK/LIN-2. Nature 404:298-302.
Hung, A. Y., and M. Sheng. 2002. PDZ domains: structural modules for protein complex assembly. J. Biol. Chem. 277:5699-5702.
Jeanneteau, F., J. Diaz, P. Sokoloff, and N. Griffon. 2004. Interactions of GIPC with dopamine D2, D3 but not D4 receptors define a novel mode of regulation of G protein-coupled receptors. Mol. Biol. Cell 15:696-705.
Kang, Y., C. R. Chen, and J. Massague. 2003. A self-enabling TGFbeta response coupled to stress signaling: Smad engages stress response factor ATF3 for Id1 repression in epithelial cells. Mol. Cell 11:915-926.
Kee, B. L., R. R. Rivera, and C. Murre. 2001. Id3 inhibits B lymphocyte progenitor growth and survival in response to TGF-beta. Nat. Immunol. 2:242-247.
Kim, E., M. Niethammer, A. Rothschild, Y. N. Jan, and M. Sheng. 1995. Clustering of Shaker-type K+ channels by interaction with a family of membrane-associated guanylate kinases. Nature 378:85-88.
Kirikoshi, H., and M. Katoh. 2002. Molecular cloning and characterization of human GIPC2, a novel gene homologous to human GIPC1 and Xenopus Kermit. Int. J. Oncol. 20:571-576.
Kiyono, T., A. Hiraiwa, M. Fujita, Y. Hayashi, T. Akiyama, and M. Ishibashi. 1997. Binding of high-risk human papillomavirus E6 oncoproteins to the human homologue of the Drosophila discs large tumor suppressor protein. Proc. Natl. Acad. Sci. USA 94:11612-11616.
Kornau, H. C., L. T. Schenker, M. B. Kennedy, and P. H. Seeburg. 1995. Domain interaction between NMDA receptor subunits and the postsynaptic density protein PSD-95. Science 269:1737-1740.
Kuhne, C., D. Gardiol, C. Guarnaccia, H. Amenitsch, and L. Banks. 2000. Differential regulation of human papillomavirus E6 by protein kinase A: conditional degradation of human discs large protein by oncogenic E6. Oncogene 19:5884-5891.
Lee, S. S., B. Glaunsinger, F. Mantovani, L. Banks, and R. T. Javier. 2000. Multi-PDZ domain protein MUPP1 is a cellular target for both adenovirus E4-ORF1 and high-risk papillomavirus type 18 E6 oncoproteins. J. Virol. 74:9680-9693.
Lee, S. S., R. S. Weiss, and R. T. Javier. 1997. Binding of human virus oncoproteins to hDlg/SAP97, a mammalian homolog of the Drosophila discs large tumor suppressor protein. Proc. Natl. Acad. Sci. USA 94:6670-6675.
Liu, T. F., G. Kandala, and V. Setaluri. 2001. PDZ domain protein GIPC interacts with the cytoplasmic tail of melanosomal membrane protein gp75 (tyrosinase-related protein-1). J. Biol. Chem. 276:35768-35777.
Lou, X., H. Yano, F. Lee, M. V. Chao, and M. G. Farquhar. 2001. GIPC and GAIP form a complex with TrkA: a putative link between G protein and receptor tyrosine kinase pathways. Mol. Biol. Cell 12:615-627.
Nakagawa, S., and J. M. Huibregtse. 2000. Human scribble (Vartul) is targeted for ubiquitin-mediated degradation by the high-risk papillomavirus E6 proteins and the E6AP ubiquitin-protein ligase. Mol. Cell. Biol. 20:8244-8253.
Nguyen, M. L., M. M. Nguyen, D. Lee, A. E. Griep, and P. F. Lambert. 2003. The PDZ ligand domain of the human papillomavirus type 16 E6 protein is required for E6's induction of epithelial hyperplasia in vivo. J. Virol. 77:6957-6964.
Ohashi, M., M. Sakurai, M. Higuchi, N. Mori, M. Fukushi, M. Oie, R. J. Coffey, K. Yoshiura, Y. Tanaka, M. Uchiyama, M. Hatanaka, and M. Fujii. 2004. Human T-cell leukemia virus type 1 Tax oncoprotein induces and interacts with a multi-PDZ domain protein, MAGI-3. Virology 320:52-62.
Pim, D., M. Thomas, R. Javier, D. Gardiol, and L. Banks. 2000. HPV E6 targeted degradation of the discs large protein: evidence for the involvement of a novel ubiquitin ligase. Oncogene 19:719-725.
Rolfe, M., P. Beer-Romero, S. Glass, J. Eckstein, I. Berdo, A. Theodoras, M. Pagano, and G. Draetta. 1995. Reconstitution of p53-ubiquitinylation reactions from purified components: the role of human ubiquitin-conjugating enzyme UBC4 and E6-associated protein (E6AP). Proc. Natl. Acad. Sci. USA 92:3264-3268.
Rousset, R., S. Fabre, C. Desbois, F. Bantignies, and P. Jalinot. 1998. The C-terminus of the HTLV-1 Tax oncoprotein mediates interaction with the PDZ domain of cellular proteins. Oncogene 16:643-654.
Saitoh, T., T. Mine, and M. Katoh. 2002. Molecular cloning and characterization of human GIPC3, a novel gene homologous to human GIPC1 and GIPC2. Int. J. Oncol. 20:577-582.
Songyang, Z., A. S. Fanning, C. Fu, J. Xu, S. M. Marfatia, A. H. Chishti, A. Crompton, A. C. Chan, J. M. Anderson, and L. C. Cantley. 1997. Recognition of unique carboxyl-terminal motifs by distinct PDZ domains. Science 275:73-77.
Suzuki, T., Y. Ohsugi, M. Uchida-Toita, T. Akiyama, and M. Yoshida. 1999. Tax oncoprotein of HTLV-1 binds to the human homologue of Drosophila discs large tumor suppressor protein, hDLG, and perturbs its function in cell growth control. Oncogene 18:5967-5972.
Tani, T. T., and A. M. Mercurio. 2001. PDZ interaction sites in integrin alpha subunits. T14853, TIP/GIPC binds to a type I recognition sequence in alpha 6A/alpha 5 and a novel sequence in alpha 6B. J. Biol. Chem. 276:36535-36542.
Thomas, M., R. Laura, K. Hepner, E. Guccione, C. Sawyers, L. Lasky, and L. Banks. 2002. Oncogenic human papillomavirus E6 proteins target the MAGI-2 and MAGI-3 proteins for degradation. Oncogene 21:5088-5096.
Thomas, M. K., K. M. Yao, M. S. Tenser, G. G. Wong, and J. F. Habener. 1999. Bridge-1, a novel PDZ-domain coactivator of E2A-mediated regulation of insulin gene transcription. Mol. Cell. Biol. 19:8492-8504.
Wang, L. H., R. G. Kalb, and S. M. Strittmatter. 1999. A PDZ protein regulates the distribution of the transmembrane semaphorin, M-SemF. J. Biol. Chem. 274:14137-14146.
Watson, R. A., M. Thomas, L. Banks, and S. Roberts. 2003. Activity of the human papillomavirus E6 PDZ-binding motif correlates with an enhanced morphological transformation of immortalized human keratinocytes. J. Cell Sci. 116:4925-4934.
Zhou, H., I. Wertz, K. O'Rourke, M. Ultsch, S. Seshagiri, M. Eby, W. Xiao, and V. M. Dixit. 2004. Bcl10 activates the NF-kappaB pathway through ubiquitination of NEMO. Nature 427:167-171.(Arnaud Favre-Bonvin, Caro)