Human Papillomavirus 16 E6 Oncoprotein Interferences with Insulin Signaling Pathway by Binding to Tuberin*

Tuberous sclerosis complex (TSC) is a genetic disorder caused by mutations in either TSC1 or TSC2 tumor suppressor gene. TSC1 and TSC2 products, Harmatin and Tuberin, form the functional complex to serve as the negative regulator for insulin-induced phosphorylation of S6 kinase and elF4E-binding protein 1. High-risk human papillomavirus (HPV) infection is the necessary cause for cervical cancer. E6 oncoprotein encoded by HPV plays a pivotal role in carcinogenesis by interference with the host intracellular protein functions. In this study, we show that HPV16 E6 interacts with tumor suppressor gene TSC2 product, Tuberin, and results in the phosphorylation of S6 kinase and S6 even in the absence of insulin. The overexpression of Tuberin overcomes the effect of E6 on S6 kinase phosphorylation. Binding with HPV16 E6 causes the proteasome-mediated degradation of Tuberin. A DILG motif and an ELVG motif located in the carboxyl-terminal of Tuberin are required for E6 binding. In addition, the Tuberin interaction region in E6 has been mapped in the amino-terminal portion of HPV16 E6, which is different from the binding domain with p53. These results provide a possible link between E6-induced oncogenesis and the insulin-stimulated cell proliferation signaling pathway.

Human papillomaviruses (HPVs) 1 infect epithelial cells and induce epithelial tumors or benign lesions (1). Based on their clinical outcomes, HPVs are grouped into high-risk and lowrisk types. The high-risk types, such as HPV16, HPV18, HPV31, HPV33, have been identified as the cause of cervical carcinoma (2). Two oncoproteins encoded by HPVs E6 and E7 are able to immortalize human epithelial cells in vitro (3). The transgenic mice carrying the E6 gene under the control of the keratin promoter developed skin cancer (4). Moreover, E6 and E7 oncoproteins were found in HPV-infected malignant tumors (5), indicating that E6 and E7 functions are required for tumorigenesis.
E6 interacts with a variety of host cell proteins and interferes with their functions. The high-risk HPV E6 proteins bind to tumor suppressor protein p53 for degradation (6). This observation provides an important functional clue in which E6 contributes to cell transformation. Several studies showed that additional E6 activities are required to reach full transformation potential (7,8). A growing number of E6 targets involved in the broad spectrum of cellular functions have been identified (9) including the transcriptional coactivator p300/CREB-binding protein (10,11), hMcm7 (12,13), IRF-3 (14), and E6TP1 (15) and others. Thus, E6 interferes with multiple cellular pathways leading to malignant transformation. However, the mechanism that coordinates E6 targeting cellular activities and leads to tumorigenesis currently remains unclear.
Tuberous sclerosis complex (TSC) is an inheritable genetic disorder characterized by the formation of benign tumors in multiple organs (16). Genetic studies show that TSC is caused by mutations in either the TSC1 or TSC2 tumor suppressor gene (17,18). TSC1 encodes a 130-kDa protein, Hamartin, whereas TSC2 encodes Tuberin, which is ϳ200 kDa with a carboxyl-terminal homology to the Rap GTPase-activating protein (19). Studies in both mammalian (20,21) and Drosophila (22) indicate that Hamartin and Tuberin form a functional complex in vivo and negatively control cell growth and proliferation (23,24). The Hamartin-Tuberin complex plays a negative regulator role in the insulin signaling pathway downstream of Akt (25)(26)(27). Akt phosphorylates Tuberin and disrupts the formation of the Hamartin-Tuberin complex, leading to the relief of its inhibition of S6 kinase (S6K) and to the activation of the elF4E-binding protein 1 (4EBP1) (26,28,29). The direct downstream target of Tuberin is the Ras family GTP-binding protein Rheb (30 -33). The GTP-bound Rheb promotes the phosphorylation of mammalian target of rapamycin and ultimately stimulates S6 kinase and 4EBP1 phosphorylation (22,34,35). Tuberin suppresses the Rheb activity by reducing the GTP bound Rheb level in vivo (31).
In this study, we identified Tuberin as an intracellular target for HPV16 E6 protein. E6 binds and induces the degradation of Tuberin, leading to the destruction of its inhibitory function in regulation of S6 kinase activation in response to insulin stimulation.

EXPERIMENTAL PROCEDURES
Plasmids, Antibodies, and Reagents-The different HPV GST-E6s were constructed by PCR amplification and insertion of the full-length E6s into pGEX-4T-2 (Amersham Biosciences) at EcoRI/XhoI sites. The pCMV-Myc-16E6 was constructed by the insertion of PCR-amplified full-length HVP16 E6 into pCMV-Myc vector (Clontech). The resulting construct is in-frame with a Myc tag placed at the amino terminus. The GST-HPV16 E6 deletion mutants illustrated in Fig. 3A were generated by PCR and subcloned into pGEX-4T-2. The resulting clones were confirmed by DNA sequencing. The TSC2-C-Gal4-AD contains the amino acid residues 1315-1807 of TSC2 and was acquired from a HeLa cDNA two-hybrid library (Clontech) by screening with the 16E6-Gal4-BD bait. The pCMV-HA-TSC2-C was generated by insertion of the EcoRI/XhoI fragment from the TSC2-C-Gal4-AD into the pCMV-HA plasmid (Clontech). The TSC2 cDNA fragments encoding amino acids 1315-1560 (T1), 1560 -1695 (T2), and 1695-1807 (T3) were subcloned into the pCMV-HA vector with the HA tag placed at the amino terminus ( Fig. 2A). The pCMV-FLAG-TSC2 was generated by PCR with the FLAG tag in the amino terminus and subcloned into the pCDNA 3.0 vector (Invitrogen). All of the constructs generated by PCR were confirmed by DNA sequencing. Plasmids used for transfection were purified by the Qiagen plasmid mega kit. The HPV6, HPV11, HPV16, and HPV18 were provided by Dr. zur Hausen. The full-length TSC2 expression vector with Xpress tag was provided by Dr. Jin Q. Chen. The plasmid pEGFP-N1 used as the control in the experiments was purchased from Clontech.
Yeast Two-hybrid Screening-A Gal4-based Matchmaker yeast twohybrid screen was performed according to the manufacturer's protocol (Clontech). The full-length HPV16 E6 gene constructed in-frame with the Gal4 DNA-binding domain in the pAS2-1 vector was used as bait. The HeLa cDNA library was cloned into the Gal4-activating domain of the pACT2. The yeast strain Y190 was cotransfected with the pAS2-1-16E6 and pACT2-library DNA. HPV16 E6-interacting proteins were identified by growth on selection medium and expression of the ␤-galactosidase. Clones were retested for E6-specific interaction by comparing their interaction with the pAS2-1-16E6 and a negative control bait pLam5Ј (human lamin-Gal4 DNA-binding domain hybrid, Clontech).
GST-E6 Pull-down of Tuberin from Cold NIH 3T3 and BGC823 Cell Extracts-GST-E6s were expressed and purified as described in the manufacturer's instructions. The purified fusion protein was conjugated to glutathione-Sepharose. Cells were lysed in lysis buffer (15 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.1% Tween 20, 1 mM phenylmethylsulfonyl fluoride, 1 mM dithiothreitol supplemented with protease inhibitors). Equal amounts of GST or GST-E6 proteins (ϳ5 g) and 250 g of cell lysate were used in precipitation. The mixture of GST-E6 beads and cell lysate was incubated with rotation at 4°C for 4 h. The beads then were washed three times with lysis buffer. The complexes were separated on 6% SDS-PAGE. Western blot was performed with anti-Tuberin antibody.
In Vitro Binding Assays-The FLAG-tagged C-terminal portion of Tuberin was in vitro translated using the rabbit reticulocyte lysate system (TNT system, Promega) and pre-cleared with glutathione-Sepharose beads. 20 l of the in vitro translation product then was incubated with 25 l of the different GST-E6 fusion protein beads. The reaction was carried out in total of 250 l of reaction buffer as described above. The beads were washed three times with lysis buffer (see above) and separated by SDS-PAGE. The gel was transferred to polyvinylidene difluoride membrane, and HPV E6-associated Tuberin was detected by anti-FLAG antibody.
Cell Culture and Transfection-All of the cells were cultured in Dulbecco's modified Eagle medium containing 10% fetal bovine serum. Transfections were carried out using LipofectAMINE 2000 (Invitrogen) according to the manufacturer's instructions. The total amount of plasmids added to each plate was kept constant by including the appropriate amount of empty vector.
In Vivo Degradation Assay-Cells were transfected with the indicated DNA balanced with empty vector to keep an equal amount and harvested 48 h after transfection. 40 g of total protein was resolved by SDS-PAGE and transferred to polyvinylidene difluoride membrane. The immunoblotting was carried out as described below.
Cell Lysate Preparation and Immunoblotting-Cells were lysed in lysis buffer described above. Protein concentration was estimated by BCA assay (Pierce), and equal amounts of proteins were resolved in SDS-PAGE. Gels were transferred to polyvinylidene difluoride and probed with the indicated antibodies. Western blots were developed using peroxidase-conjugated secondary antibody and the enhanced chemiluminescence substrates (Amersham Biosciences).

HPV16 E6
Interacts with Tuberin-During a yeast two-hybrid screen using HPV16 E6 open reading frame fused to the Gal4 DNA-binding domain as the bait, we identified several previous known E6-interacting proteins including IRF-3 (14) and E6TP1 (15). Interestingly, another positive clone turned out from the carboxyl-terminal 492 amino acids of Tuberin.
To confirm the interaction between Tuberin and HPV16 E6 and to determine whether this association is HPV type-specific, we performed a GST-E6 pull-down experiment to precipitate Tuberin from NIH 3T3 and BGC823 (a gastric carcinoma cell line) (for review see Ref. 36) cell extracts. GST-E6 fusion proteins from HPV6, HPV11, HPV16, and HPV18 were incubated with cell extracts, respectively (Fig. 1A). Tuberin was expressed in both NIH 3T3 and BGC823 cells (Fig. 1A, input lines). GST-E6 from HPV16 was strongly associated with the endogenous Tuberin from both NIH 3T3 and BGC823 cells. In contrast, GST-E6 proteins from HPV6, HPV11, HPV18, and GST alone did not bind to Tuberin. These results suggest that HPV16 E6 specifically interacts with Tuberin. Two bands between 170-and 200-kDa molecular mass markers were detected by anti-Tuberin antibody in NIH 3T3 cells. This result is consistent with a previous report suggesting that the upper The fusion proteins were purified and conjugated to glutathione-Sepharose. NIH 3T3 cells and BGC823 cells were lysed and incubated with indicated GST-E6 beads. The complexes were washed and resolved by SDS-PAGE. Tuberin was detected by anti-Tuberin antibody. The input corresponds to 10% of protein used in the pull-down experiments. The lower panel is the stain of GST-E6 fusions applied in the above assay. B, the BGC823 cell lysate was incubated with the indicated GST-E6 proteins and detected with anti-p53 antibody. C, the FLAG-tagged carboxyl-terminal portion of TSC was in vitro transcribed and translated and then incubated with the indicated GST-E6 fusions. The complexes were separated and detected with anti-FLAG antibody. band represents the phosphorylated Tuberin (37) . Fig. 1A, lower panel, represents the amount of GST-E6 fusion protein used in the above experiment. As the positive control, the association of p53 with different HPV E6s was performed using the NIH 3T3 cell extracts (Fig. 1B). In contrast to Tuberin, p53 binds to both HPV16 and 18 E6s, although the binding affinity to HPV16 E6 was stronger. In our current system, p53 also showed very low binding activity to HPV11 E6. To further verify the association between HPV16 E6 and Tuberin, an in vitro binding assay was carried out. Consistent with the endogenous Tuberin, the in vitro translated carboxyl-terminal portion of Tuberin also bound to HPV16 E6 but not to HPV 6, HPV11, and HPV18 E6s (Fig. 1C).
Expression of E6 Interferences with the S6 Kinase Phosphorylation Induced by Insulin-We next sought to determine the biological significance of the association of Tuberin with HPV16 E6. Recent studies (29,38,39) demonstrate that Tuberin negatively controls the activation of the p70 ribosomal protein S6 kinase 1 in response to insulin stimulation. We examined whether the activation of S6 kinase was affected by the expression of HPV16 E6. Previous studies demonstrate (40) that S6 kinase is activated by phosphorylation on several residues in which the anti-phosphospecific antibody was used to analyze S6 kinase activity. COS-7 cells were transfected with TSC2 or E6 alone, or TSC2 plus E6. 24 h after transfection the cells were cultured in 0.5% serum overnight. Insulin was added to stimulate the phosphorylation of S6 kinase and S6. The phosphorylation of S6 kinase and S6 was determined by antiphospho-S6 kinase and anti-phospho-S6 antibodies. The addition of insulin stimulated the phosphorylation of S6 kinase and S6 in COS-7 cells transfected with the vector (Fig. 2A, lanes 1  and 2). Overexpression of TSC2 inhibited insulin stimulated S6 kinase and S6 phosphorylation ( Fig. 2A, lanes 3 and 4). Interestingly, the expression of HPV16 E6 alone resulted in the phosphorylation of both S6 kinase and S6 even in the absence of insulin (Fig. 2A, lane 5). The addition of insulin to the cells expressing E6 did not further increase the phosphorylation of S6 kinase and S6 (Fig. 2A, lane 6), thus suggesting the effect of E6 on the phosphorylation of S6 kinase and S6 was via insulin signaling pathway. More importantly, the effect of E6 on S6 kinase and S6 phosphorylation was overcome by coexpression of Tuberin ( Fig. 2A, lanes 7 and 8), suggesting that E6 targets insulin signaling by eliminating Tuberin suppressor function.
A previous study (41) shows that high-risk E6 proteins degraded the membrane-associated guanylate kinase homologues MAGI-2 and MAGI-3. MAGI-2 and MAGI-3 were associated with PTEN (phosphatase and tensin homologue deleted on chromosome 10) and enhanced its ability to suppress Akt activation (42). Therefore, E6 may indirectly activate Akt. To confirm that the phosphorylation of S6 kinase by expression of HPV16 E6 is a result of Tuberin elimination and not the indirect activation of Akt, the phosphorylation of Akt in these transfections was tested by Western blot using an anti-phospho-Akt antibody. In contrast to S6 kinase and S6, phosphorylation of Akt was not affected by expression of HPV16 E6 ( Fig. 2A), indicating that the effect of E6 on phosphorylation of S6 kinase and S6 was mainly through abolishing TSC complex function.
To test whether the other E6s that do not bind to Tuberin cause the activation of S6 kinase independent of insulin stimulation, HEK293 cells were transfected with the E6 constructs. The S6 kinase phosphorylation was detected by Western blot. In contrast to HPV16 E6, the expression of 6E6 and 18E6 did not result in the phosphorylation of S6 kinase (Fig. 2B) The Expression of HPV16 E6 Promotes the Degradation of Tuberin in HEK293 Cells-HPV E6 interferes with p53 func-tion by degradation of the protein via ubiquitin-mediated pathway (6). To investigate the mechanism showing that HPV16 E6 interrupts Tuberin function, we determined whether the expression of E6 affects the stability of Tuberin. HEK293 cells were transfected with an increased amount of the HPV16 E6 construct. The expressions of the endogenous Tuberin and p53 were determined by a Western blot. The levels of Tuberin and p53 were decreased, whereas the E6 expression level was increased (Fig. 3A). Interestingly, the phosphorylation but not the expression of endogenous S6 was increased simultaneously with the decreased Tuberin expression, suggesting that the phosphorylation of S6 might result from the degradation of Tuberin. To determine that the decreased expression of Tuberin upon HPV16 E6 expression occurred after transcription, a Northern blot was performed. The expressions with different amounts of E6 did not affect the TSC2 transcription (Fig. 3B).
The half-life of the Tuberin in the cells expressing E6 was compared with that in the cells without E6. HEK293 cells were transfected with either Xpress-tagged TSC2 or TSC2 plus E6. 48 h after transfection, cells were treated with cycloheximide for the indicated hours. The Tuberin level was detected at different time points. The result is shown in Fig. 3C. The amount of Tuberin significantly reduced 4 h after cycloheximide treatment and completely vanished 6 h after treatment in the cells with E6. In contrast, without E6 expression, Tuberin slightly reduced 4 h after treatment and dramatically reduced but did not vanish 6 h after treatment (Fig. 3C).
Previous studies revealed that HPV E6 proteins target and degrade intracellular partners through proteasomal degradation (43)(44)(45)(46). To determine whether the Tuberin degradation is caused by this mechanism, HEK293 cells were transfected with Xpress-tagged Tuberin alone or with E6. 48 h after transfection, cells were treated with the proteasome inhibitor MG132 for 3 h prior to protein extraction. The expression of exogenous Tuberin was high in the cells without E6 in both the presence and absence of MG132. The presence of E6 greatly decreased the Tuberin level. However, the Tuberin expression was recovered in the presence of MG132. In the presence of MG132, degradation of Tuberin was inhibited (Fig. 3D).
Mapping of the E6-binding Region within Tuberin-A previous study (47) that screened HPV16 E6-binding peptides demonstrates that a 27-amino acid peptide homology to Tuberin bound to E6 and that the binding domain was DILG. The sequence analysis showed that the DILG domain is mapped to the amino acids corresponding to the positions 1396 -1399 of Tuberin and located in the cDNA fragment of TSC2 identified from the yeast two-hybrid screen. In addition, a similar binding  1 and 2) or vector (lanes 3 and 4). 40 M MG132 was added (lanes 1 and 3) 3 h before cells were harvested. Tuberin was detected using anti-Xpress antibody. EGFP, enhanced green fluorescent protein; GAPDH, glyceraldehyde-3-phosphate dehydrogenase. domain, ELVG, is located between positions 1351 and 1354 of Tuberin. To confirm the binding domain for HPV16 E6, three truncated constructs were analyzed (Fig. 4A). The pCMV-HA-TSC2-C contains the cDNA fragment isolated from yeast twohybrid screen. The pCMV-HA-T1, T2, and T3 represent the different proportion of the pCMV-HA-TSC2-C as shown in Fig.  4A. COS-7 cells were transfected with the pCMV-HA-TSC2-C construct and the three truncated constructs. Cell lysates were made for GST-16E6 pull-down experiments. Consistent with previous experiments, GST-16E6 bound to the pCMV-HA-TSC2-C (Fig. 4B, lane 1). The truncated construct, pCMV-HA-T1, also associated with the GST-16E6 (Fig. 4B, lane 2), whereas the pCMV-HA-T2 (Fig. 4B, lane 3) and T3 (Fig. 4B,  lane 4) did not. These results indicate that the amino acids 1315-1560 of Tuberin containing both DILG and ELVG domains have the E6 binding activity. To determine the effect of DILG and ELVG domains on E6 binding, the binding domain mutations were generated (Fig. 4C). The pCMV-HA-TSC2-Mu1 containing the substitution for DILG and the pCMV-HA-TSC2-Mu2 containing the substitution for ELVG partially reduced the binding activity to HPV16 E6 (Fig. 4D, lanes 1 and 2  compared with lane C). However, the binding activity of pCMV-HA-TSC2-Mu3 was totally abolished in which both DILG and ELVG sites were replaced (Fig. 4D, lane 3). These results indicate that both DILG and ELVG domains are required for Tuberin to associate with HPV16 E6.
Identification of the Region in HPV16 E6 That Is Required for Interaction with Tuberin-To define the region of E6 re-sponsible for binding to Tuberin, a series of HPV16 E6 deletions were generated as illustrated in Fig. 5A. GST-HPV16 E6 deletion fusions were used to test the binding capability with Tuberin. As shown in Fig. 5B, top panel, Tuberin binds to the HPV16 E6 deletion mutants 16E6-2-GST and 16E6-4-GST (lanes 2 and 4). In contrast, 16E6-1-GST, 16E6-3-GST, 16E6-5-GST, and 16E6-6-GST were unable to bind to Tuberin (lanes 1, 3, and 5). The amino acids 78 -104 of 16E6 are overlapped in 16E6-2-GST and 16E6-4-GST and missed in the other constructs, indicating that the residues between 78 and 104 are the binding region for Tuberin. To further confirm the Tuberin binding region, a deletion mutant without the amino acids 78 -104 was generated from the full-length HPV16 E6. The resulting GST fusion protein, 16E6-6-GST, failed to bind to Tuberin (lane 6), supporting that Tuberin binding region is located between amino acids 78 and 104 on HPV16 E6. DISCUSSION We have demonstrated here that HPV16 E6, which is present in Ͼ50% of cervical cancers, interacts with the tumor suppressor protein, Tuberin, and interferes with its suppressor function in insulin-induced cell growth and proliferation signaling pathway, leading to the activation of S6 kinase downstream of Akt activation. The exogenous expression of E6 resulted in the degradation of the endogenous Tuberin. Expression of E6 in serum-starved cells was sufficient to induce S6 kinase and S6 phosphorylation without the addition of stimuli, whereas insulin stimulated the phosphorylations of S6 kinase and S6, suggesting that E6 activated the signaling component downstream of insulin receptor tyrosine kinase. Conversely, the overexpression of Tuberin blocked the S6 kinase phosphorylation and activation induced by E6, indicating that E6 targeted and interfered with Tuberin function in the insulin signaling pathway. The possible model for interference of Tuberin by HPV16 E6 leading to S6 kinase activation is illustrated in Fig. 6.
Recent studies have placed the complex of Hamartin and Tuberin in the cell growth and proliferation pathway in which the complex negatively regulates both growth factor-and nutrient-dependent activation of TOR signaling to its downstream targets S6 kinase and 4EBP1 (48 -50). The negative regulator role of the complex is achieved by Tuberin GTPase- activating protein activity toward Rheb. Overexpression of Rheb bypasses the nutrient induction to activate TOR signaling (31,32,(51)(52)(53). The TSC2 mutations derived from TSC patients are defective in GTPase-activating protein activity, suggesting that failure in the control of Rheb plays an important role in the pathology of TSC. Abrogation of Tuberin function by HPV16 E6 may therefore mimic the naturally occurring TSC2 mutations and contribute to HPV16-induced proliferation and tumorigenesis.
Human papillomaviruses infect proliferating epithelial cells. The replication of high-risk types of HPV occurs in the differentiated cells, which contain little or none of the cellular replication components. To utilize the host replication machinery, the virus needs to modify the host cellular process and drive the differentiated cells to re-enter the cell cycle. HPV16 oncoproteins E6 and E7 have been found to independently modulate the ability of keratinocyte differentiation (54,55). The potentiality of virus oncoproteins to re-program the cellular process is gained by interfering with the key cellular protein functions. Targeting and degradation of p53 by E6 oncoprotein represent the most critical functional event, leading to an increase in host genetic instability. In addition to p53, more than a dozen cellular proteins have been identified as E6-interacting proteins (9). Identification of Tuberin as the cellular partner of HPV16 E6 provides the direct evidence that E6 targets the major growth and proliferation signal transduction pathway. In addition to Tuberin, HPV E6 proteins also target the membraneassociated guanylate kinase homologues MAGI-2 and MAGI-3 for degradation (41). MAGI-2 and MAGI-3 have been reported to be associated with PTEN (42), the other negative regulator of the insulin-induced growth pathway, and enhance the ability of PTEN to suppress Akt activation. Therefore, the degradation of MAGI-2 and MAGI-3 by E6 may result in the abolishment of PTEN function. Taken together, these results highlight the importance of interference with the growth factor-mediated signaling pathway in E6-induced immortalization. Although the disruption of p53 by E6 shall demolish the fidelity of DNA replication, the abrogation of the cell growth and proliferation pathway may lead to uncontrolled growth and immortalization. The diverse activities of HPV E6 may represent the different sides of the coin. How those activities of E6 are coordinated to achieve the full malignancy is currently unclear and needs to be elucidated.
In system used here, Tuberin bound to HPV16 E6 but not to HPV18 E6. The reason for this specificity might be due to the preferential binding motifs in different HPV E6 proteins. Several proteins are targeted by E6 through the PDZ motif. HPV18 E6 has a perfect PDZ binding motif, whereas HPV16 is suboptimal (56). Therefore, proteins with PDZ motif may bind HPV18 E6 more efficiently. In contrast, the other proteins, such as IRF-3 and Tuberin binding to E6 through ELLG motif identified as (E/D)L(L/V)G, may favor HPV16 E6 (14,47,57). HPV18 E6 also binds to the ELLG domain but with much lower efficiency compared with HPV16 E6. It cannot exclude the possibility that HPV18 E6 binds to Tuberin with very weak capability, which cannot be detected under current condition. Our results have shown that two E6 binding motifs in Tuberin are required to achieve the full binding activity. This is similar to the observation in tumor necrosis factor receptor 1 (57).
In contrast to its functional information, there is less clarity regarding the regions of E6 that are responsible for the association with the cellular partners. In this study, we have shown that residues 78 -104 of HPV16 E6 are responsible for Tuberin binding. The corresponding region of HPV18 E6 was identified previously for the interaction with E6BP (58). However, unlike Tuberin, the associations of E6BP with E6 do not cause the degradation of E6BP. In addition, p53 and its binding partner E6AP bind to the separated regions of E6 that are as different as the Tuberin (59). Moreover, some E6 mutants that are defective in p53 degradation remain the ability to induce cell immortalization (7), implying that multiple functions of E6 are involved in HPV-induced tumorigenesis. It would be interesting to investigate the link between the structure of E6 oncoprotein and its functional outcome. The biological activity of E6 mutants involved in Tuberin binding is undergoing exploration.