Human papillomavirus type 18 E6 protein binds the cellular PDZ protein TIP-2/GIPC, which is involved in transforming growth factor beta signaling and triggers its degradation by the proteasome

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 beta (TGF-beta) signaling by favoring expression of the TGF-beta 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-beta 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-beta by lowering the intracellular amount of TIP-2/GIPC.

FIG. 1.

Interaction of HPV-18 E6 with TIP-2/GIPC triggers its degradation by the proteasome and is mediated by the C terminus of E6. (A) COS7 cells were transfected with expression vectors pSGF-E6 (2 μg) (lanes 1 to 8) together with either pTL1-TIP-2 (2 μg) (lanes 1 to 3) or pEGFPC1-TIP-2 (1 μg) (lanes 4 to 8). Cells were either not treated or treated with 20 μM MG132 (M) (lanes 2 and 5) or lactacystine (L) (lane 8) for 6 h before cell harvest. Using extracts of these transfected cells, immunoprecipitations (I.P.) were carried out in RIPA buffer with either the M2 monoclonal antibody to FLAG (α-FLAG) (lanes 1, 2, 4, 5, 7, and 8) or a monoclonal antibody directed against hDlg as a control (ctl) (lanes 3 and 6). After separation in an SDS-10% polyacrylamide protein gel, immunoblot analysis (I.B.) was performed with a polyclonal antibody to TIP-2/GIPC. The positions of the bands corresponding to TIP-2/GIPC and to the GFP-TIP-2 fusionprotein are marked T2 and G2, respectively, on the right. *n.s. indicates the position of a nonspecific band due to the immunoglobulin heavy chain as well as to some protein A released by the beads. (B) Ten micrograms (total protein amount) of extracts used for the immunoprecipitations shown in panel A (lanes 1, 2, 4, 5, 7, and 8) were analyzed by immunoblotting with the antibody to TIP-2/GIPC. The positions of TIP-2/GIPC and GFP-TIP-2 are indicated as in panel A. (C and D) The experiment was performed exactly as in panels A and B, with the same extracts, except that immunoprecipitations were carried out with either the polyclonal antibody directed against TIP-2/GIPC (C, lanes 1, 2, 4, 5, 7, and 8) or a polyclonal antibody directed against integrin α6 (C, lanes 3 and 6) as a control and that immunoblot analyses were done with the M2 monoclonal antibody to FLAG (D). The position of the FLAG-E6 is marked E6 on the right. Separation of the proteins was performed through an SDS-14% polyacrylamide protein gel. (E) COS7 cells were transfected with plasmid pTL1-TIP-2 and pSGF-E6 expression vectors including either mutation of threonine 156 to alanine (lane 1), mutation of valine 158 to alanine (lane 2), mutation of threonine 156 and valine 158 to alanine (lane 3), or deletion of the C-terminal four amino acids of E6 (lane 4). Cells were treated with 20 μM MG132 for 5 h before cell harvest, and immunoprecipitations were carried out with the antibody to TIP-2/GIPC and analyzed with the FLAG antibody as in described for panel C. (F) The extracts used for panel E were analyzed by immunoblotting with the antibody to FLAG as described for panel D.

FIG. 2.

E6 induces polyubiquitination with K48 branching of TIP-2/GIPC. (A) COS7 cells were transfected with plasmid pTL1-TIP-2 (lanes 1 to 6) together with either pSG5 (lanes 1, 3, and 5) or pSGF-E6 (lanes 2, 4, and 6) and a vector expressing HA-tagged ubiquitin (Ub), either wild type (wt) or including a single lysine at position 48 (lanes 3 and 4) or at position 63 (lanes 5 and 6) (52). The amount of each plasmid was 2 μg. Extracts of these transfected cells were used for immunoprecipitation (I.P.) with the antibody directed against TIP-2/GIPC (α-TIP-2), and precipitated proteins were analyzed by immunoblotting (I.B.) with a monoclonal antibody to the HA epitope. This revealed the smear of polyubiquitinated TIP-2/GIPC. The positions of bands of a molecular mass marker run in parallel are indicated on the right. (B) CV-1 cells were transfected with either pSGF (lanes 1 and 3) or pSGF-E6 (lanes 2 and 4). Extracts of these cells were analyzed by immunoblotting with an antibody directed against ubiquitin-protein conjugates (lanes 1 and 2) or used for immunoprecipitation with the antibody to TIP-2/GIPC. Immunoprecipitated proteins were analyzed by immunoblotting with an antibody directed against ubiquitin-protein conjugates to visualize the polyubiquitinated forms of TIP-2/GIPC.

FIG. 3.

E6 silencing by RNA interference increases the amount of TIP-2/GIPC. HeLa cells were transfected with E7 siRNA duplex, which also silences E6 (20). As a control (Ctl), cells were also transfected with a nonfunctional siRNA couple targeting an unrelated protein (I6.2). Extracts of cells transfected with control (lanes 1) or E7 (lanes 2) siRNAs were prepared 24 h after transfection and analyzed by immunoblotting (I.B.) with antibody to p53 (α-p53) (A) or TIP-2/GIPC (B). The positions of the bands corresponding to p53 and TIP-2/GIPC are indicated, together with those of a molecular mass marker run in parallel.

FIG. 4.

Silencing of TIP-2/GIPC by RNA interference decreases Id3 mRNA. (A) Schematic representation of the TIP-2 mRNA and of the siRNA couple used to silence TIP-2/GIPC. Its sequence starts at position 276 with respect to the first nucleotide of the initiation codon. (B) HeLa cells were either mock transfected or transfected with control (ctl) (I6.2) or TIP-2.1 siRNA couples. Extracts were prepared 72 h after transfection and analyzed by immunoblotting (I.B.) with the antibody to TIP-2/GIPC (α-TIP-2). The position of the band corresponding to TIP-2/GIPC is indicated, together with those of a molecular mass marker run in parallel. (C and D) As controls, the extracts used for panel B were analyzed by immunoblotting with antibodies directed against either the PDZ protein TIP-40 (C) or β-actin (D). (E) Total RNA from HeLa cells transfected with either control or TIP-2/GIPC siRNA was prepared and analyzed by real-time quantitative PCR for expression of various genes as indicated. The amount of mRNA present in TIP-2/GIPC-silenced cells is expressed as a percentage of that in cells transfected with control siRNA.

FIG. 5.

Regulation of Id3 mRNA amount by both TIP-2/GIPC and TGF-β1. (A) HeLa cells were transfected for 72 h with siRNAs targeted against TIP-2/GIPC or E6/E7 and treated or not with 200 pM TGF-β1 for 16 h before harvest as indicated. Total RNAs were prepared, the TIP-2/GIPC mRNA was analyzed by real-time quantitative PCR, and the results are presented as described in the legend to Fig. 4. Error bars indicate standard deviations. (B) Extracts of the cells used for panel A were analyzed by immunoblotting with the antibody directed against TIP-2/GIPC (α-TIP-2).

FIG. 6.

Silencing of TIP-2/GIPC increases proliferation of HeLa cells. HeLa cells were transfected with either control siRNAs (I6.2) or duplexes targeting TIP-2/GIPC (solid line) or E7 (dotted line). At 24, 48, and 72 h after transfection, incorporation of radioactively labeled thymidine for 4 h was measured. Means and standard deviations of values corresponding to triplicates are shown for TIP-2/GIPC- and E7-silenced cells as percentages of control values plotted against time.

FIG. 7.

Effects of TGF-β1 treatment on TIP-2/GIPC-silenced cells. (A) HeLa cells were transfected for 72 h with control (ctl), E7, or TIP-2/GIPC siRNAs and treated or not with 200 pM TGF-β1 for 5 h before harvest. Incorporation of radioactively labeled thymidine for 5 h was measured, and means of values of duplicates are shown as percentages of control values. The error bars correspond to half the difference between the two values. (B) Extracts of cells used for panel A were analyzed by immunoblotting (I.B.) with the antibody directed against TIP-2/GIPC (α-TIP-2).