INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Human papillomaviruses (HPVs) are responsible for hyperproliferation of cutaneous and mucocutaneous epithelial cells that can lead to propagation of benign (30) or malignant (81) lesions depending on the virus type. The E6 and E7 proteins encoded by mucocutaneous high-risk types interact with p53 (65, 77) and the retinoblastoma (Rb) protein family (50), respectively, and transform cells in culture (10, 35, 39, 49, 74). In contrast, the E6 and E7 proteins encoded by low-risk viruses do not interact with these proteins and are not typically associated with events that lead to cell transformation (6).

The E6 proteins encoded by HPVs contain about 150 amino acids and possess two Cys-X-X-Cys zinc fingers that bind zinc (5). While host proteins that interact with the E6 protein from both low- and high-risk HPVs (22, 44, 45, 54, 69) or only from high-risk HPVs have been identified (16, 26-28, 40, 42, 47, 57, 60, 70), no specific interaction between low-risk E6 and host proteins has been described. Here we report that low-risk E6 from HPV type 6 (₆E6) interacts with zyxin, a focal adhesion protein (7).

Focal adhesion plaques are discrete areas on the cell membrane where the cells contact the underlying substratum or each other (36, 75). They are also the sites where multiple protein complexes involved in signaling assemble (15). Focal adhesions appear to represent transmembrane connections between the extracellular matrix and the cytoskeleton. Thus, it is not surprising that disrupted focal adhesions are frequently associated with the transformed phenotype (14). The E6 proteins from bovine papillomavirus and high-risk HPV interact with paxillin, another focal adhesion protein (13, 70, 71). This interaction may in part account for the disruption of actin fiber organization when bovine papillomavirus type 1 E6 is overexpressed in cells (70).

Zyxin has features reminiscent of a signaling protein. Relative to the structural components of focal adhesions such as vinculin and -actinin it is present at low abundance in cells and it is phosphorylated at multiple sites in vivo (18). Structurally, it has a proline-rich domain at its N terminus and multiple LIM (Lin-11, Isl-1, and Mec-3 [25]) domains in its carboxy-terminal half (8). Both domains are thought to be involved in protein binding (59, 66). The proline-rich domain associates with SH3 domains that are found in a number of protein components in signal transduction pathways such as the human proto-oncogene product Vav (38). The LIM domain is a double-zinc-finger motif that is present in a number of proteins involved in the regulation of cell proliferation and differentiation (29, 61, 63). Zyxin also possesses a nuclear export sequence, and chicken zyxin shuttles between the nucleus and focal adhesions (52). These characteristics suggest a role for zyxin as a messenger that relays information from sites of cell adhesion to the nucleus.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Plasmids. (i) HPV E6 plasmids. The E6 constructs ₆E6-Gal4-BD and ₁₈E6-Gal4-BD for yeast two-hybrid assays, glutathione S-transferase (GST)-₆E6, GST-₁₁E6, GST-₁₆E6, and GST-₁₈E6 for GST binding assays, ₆E6-pET and ₁₈E6-pET for in vitro translation, and ₆E6-Myc and ₁₈E6-Myc for mammalian cell transfections were described previously (22). The EcoRI and BamHI fragments containing the ₁₁E6 and ₁₆E6 coding sequences were cloned into pAS2-1 (Clontech, Palo Alto, Calif.) to generate ₁₁E6-Gal4-BD and ₁₆E6-Gal4-BD. An NcoI/BamHI fragment from ₁₁E6-Gal4-BD was cloned into pET21 (Novagen, Madison, Wis.) to generate ₁₁E6-pET. The sequence encoding ₁₁E6 was released from ₁₁E6-pET as an NcoI/NotI fragment, end filled, and cloned between the PmeI and NotI sites in pmycpl.1 (11) to construct ₁₁E6-Myc. The EcoRI/BsmI fragment containing the N-terminal 37 amino acids of ₁₁E6 was ligated to the BsmI/BamHI fragment containing amino acids 37 to 150 of ₆E6 and the resulting _11/6E6 fusion was cloned between EcoRI and BamHI sites in pAS2-1 to construct _11/6E6-Gal4-BD. The NcoI fragment from _11/6E6-Gal4-BD was filled with Klenow enzyme and cloned into the PmeI site in pmycpl.1 to generate _11/6E6-Myc. The MfeI/NotI fragment from _11/6E6-Myc was cloned between the EcoRI and NotI sites in the vector pALEX (53) to construct GST-_11/6E6.

(ii) Zyxin constructs. Full-length zyxin cDNA in pBluescript KS() (Stratagene, La Jolla, Calif.) was kindly provided by Mary Beckerle (Department of Biology, University of Utah, Salt Lake City, Utah). An EcoRI fragment from this construct was cloned into pET21b (Novagen) to create zyxin-pET for in vitro transcription and translation. The same fragment was also cloned into a eukaryotic expression vector, pCF2H (58), to generate zyxin-Flag. The NcoI-EcoRI fragment was cloned into pAS2-1 for zyxin-Gal4-BD and into pACTII (Clontech) for zyxin-Gal4-AD. The NcoI-HincII fragment of full-length zyxin cDNA was ligated between the NcoI and SmaI sites of vectors pAS2-1 and pACTII to construct zyxin deletion mutants Zy-5'-Gal4-BD and Zy-5'-Gal4-AD, respectively. The insert in the construct Zy-5'-Gal4-AD was released with NcoI and EcoRI and cloned into pET21d to generate Zy-5'-pET for in vitro translation. Other zyxin deletion mutants containing only LIM domains were all generated using PCR with primers having an NcoI site in the 5' primers and an EcoRI site in the 3' primers. The DNA sequence of each product was determined to ensure that there were no mutations. The primers for LIM-1+2+3 were 5'-GGCAGACCATGGCTGTCAACGAACTCTGC-3' and 5'-GGGCCTGAATTCACTCAGGTCTGGGCTCTAG-3'. To generate LIM-1+2, the 5' primer was the same as for LIM-1+2+3, but the 3' primer was 5'-GCAGACGGAATTCCTCGGGGCGTACTGCTTG-3'. For LIM-2+3, the 3' primer was the same as that used for generating LIM-1+2+3, but the 5' primer was 5'-CGTACTCCATGGGCTGTTACACTGACACCC-3'. For LIM-3, the same 3' primer as for LIM-1+2+3 was used in conjunction with the 5' primer 5'-ACTGTGCCATGGACTACCACAAGCAGTACGCC-3'. For LIM-1, the 5' primer was the same as that used for LIM-1+2+3, and the 3' primer was 5'-GCAGGTGGAATTCTTCTCCAGGGTGTCAGTG-3'. The NcoI-EcoRI fragments of these deletion mutants were cloned into pAS2-1 to make Gal4-BD fusion proteins and pACTII to make Gal4-AD fusion proteins. The NcoI-EcoRI fragment of LIM-1+2+3 was also cloned into pET21d for in vitro translation. The shortest zyxin clone from the two-hybrid library screen, containing zyxin amino acids 299 to 572, was fused in frame with Gal4-AD in pGAD10 (Clontech) and was named 6(3)-Gal4-AD. The insert was released with NdeI and SalI and cloned into pAS2-1 to construct 6(3)-Gal4-BD. For activation assays in mammalian cells, zyxin-Gal4-AD was digested with NcoI, the ends were filled with Klenow, and the plasmid was digested again with SacI to release the insert to be ligated between the SmaI and SacI sites in pSG424 (62). These manipulations generated FL-M-GBD, a full-length zyxin-Gal4-BD fusion that is expressed in mammalian cells. Zy-5'-M-GBD was constructed the same way. The LIM-1+2+3 DNA fragment was released with BglII, end filled, digested with SacI, and ligated into the same SmaI and SacI sites in pSG424 to create LIM-M-GBD.

(iii) Other constructs. The PKC-Flag construct was from Jae-Woe Soh (Department of Genetics, Columbia University, New York, N.Y.). The Gps2-Flag construct was previously described (22). RL-TK has a Renilla luciferase gene under the control of a basic thymidine kinase promoter and was purchased from Promega (Madison, Wis.). PG5-luc was described before (80).

Yeast two-hybrid assays. (i) Yeast two-hybrid library screen. A human foreskin keratinocyte cDNA library containing 5 × 10⁶ independent clones that was constructed using both oligo-dT and random priming and cloned in pGAD10 to create Gal4-AD fusions was purchased from Clontech. The library was screened as previously described (22).

(ii) Yeast strains and transformation. Saccharomyces cerevisiae strains YGH1 and L40 were used for transformation of Gal4-BD fusion proteins and LexA fusion proteins, respectively. Strains Y187 and Y190 were from Clontech. All strains were maintained at 30°C on YPD (20 g of Difco Peptone/l, 10 g of yeast extract/l, 2% glucose) plates. Transformation and selection on Leu Trp SD (6.7 g of amino acid-free yeast nitrogen base/l, 2% dextrose, 100 ml of 10× dropout solution/l) plates was performed as described in the Clontech Matchmaker System manual.

(iii) Filter lift assay for -galactosidase (-Gal) activity. Four to six days after transformation, the yeast colonies were lifted onto nitrocellulose membranes (Schleicher & Schuell, Keene, N.H.), and the cells were lysed by freezing at 80°C for 20 min and thawing at room temperature. The filter disks were placed onto Whatman paper soaked in 2 ml of Z buffer (60 mM Na₂HPO₄, 40 mM NaH₂PO₄, 10 mM KCl, 1 mM MgSO₄, 0.3% -mercaptoethanol) containing 0.33 mg of 5-bromo-4-chloro-3-indolyl--D-galactopyranoside (X-Gal) per ml and incubated at 37°C for 1 to 3 h.

(iv) Liquid -Gal assay. Individual yeast colonies were picked into 3 ml of Leu Trp SD medium and grown to stationary phase. They were then diluted 1:30 in medium and grown to an optical density at 600 nm (OD₆₀₀) of 0.4 to 0.6. The cells were collected by centrifugation at 1,500 × g for 5 min at 4°C and washed once with Z buffer. The cell pellets were resuspended in 250 µl of Z buffer, and the cells were lysed following the addition of 200 µl of ice-cold acid-washed glass beads by vigorous mixing three times for 1 min each. The cell lysates were clarified by centrifugation at 3,000 × g for 5 min at 4°C. Total protein concentrations were determined by the method of Bradford (12). For -Gal assays, 200 µl of 4-mg/ml o-nitrophenyl--D-galactopyranoside (ONPG) was added to 150 µl of cell extract diluted in 800 µl of Z buffer, and the mixture was incubated at 30°C until a pale yellow color developed. The reaction was stopped by adding 500 µl of 1 M Na₂CO₃, and the OD₄₂₀ was determined. Z buffer alone incubated with ONPG was used as a control for measuring the OD₄₂₀. The -Gal units were calculated from the equation U = (1,000 × OD₄₂₀)/(t × mg), where t is the reaction time in minutes and mg is the total amount of lysate protein in milligrams that was added to the reaction (4).

(v) In vitro transcription and translation. pET21 clones were subjected to in vitro transcription and translation using TNT Coupled Reticulocyte Lysate systems for zyxin or TNT Coupled Wheat Germ systems for E6 proteins as described by the manufacturer (Promega). For HPV E6, [³⁵S]cysteine (Amersham Pharmacia Biotech, Piscataway, N.J.) was used for labeling, while zyxin was labeled with [³⁵S]methionine (Amersham Pharmacia Biotech).

GST protein purification and GST pull-down assays. (i) GST protein purification. Escherichia coli strain BL21 was the host for production of GST-E6 fusion proteins. The cultures were grown at 30°C. Five hundred milliliters of cell culture was induced with 0.8 mM isopropyl--D-thiogalactopyranoside (IPTG) when the OD₆₀₀ reached 0.5 to 0.8. After four to six more hours, cells were harvested and resuspended in 20 ml of cold phosphate-buffered saline (PBS) containing 1 mM phenylmethylsulfonyl fluoride, 0.1 mM dithiothreitol, and 0.2 mg of lysozyme per ml. All subsequent steps were performed at 4°C. Cells were lysed by sonication (three times for 20 s each) after the addition of Triton X-100 to 1%. The cell lysate was clarified by centrifugation at 20,000 × g for 15 min. Six hundred microliters of preswollen glutathione-agarose beads (Sigma, St. Louis, Mo.) was added to the lysate and mixed at 4°C for 2 h. The beads were then washed four times with 45 ml of PBS plus 1% Triton.

(ii) GST pull-down assays. Forty microliters of a 50% suspension of glutathione-agarose beads containing 3 to 5 µg of GST fusion protein was blocked at 4°C for 0.5 h in 500 µl of binding buffer (20 mM Tris-HCl [pH 7.5], 50 mM KCl, 0.5 mM EDTA, 3% bovine serum albumin [BSA], 1% NP-40). The beads were pelleted at 7,000 rpm for 20 s in an Eppendorf 5415C microcentrifuge and resuspended in 100 µl of binding buffer. Two microliters of ³⁵S-labeled in vitro translation product was added to the mix and the binding was performed at 4°C for 1 h. The beads were then washed six times with 1 ml of binding buffer and boiled for 5 min in 15 µl of sodium dodecyl sulfate (SDS) loading buffer prior to analysis by SDS-polyacrylamide gel electrophoresis (PAGE) (46). After electrophoresis, the bound ³⁵S-labeled protein was detected by autoradiography.

Far Western analysis. Three to five micrograms of GST-E6 proteins was subjected to SDS-PAGE and then transferred onto a nitrocellulose membrane. After transfer, the membrane was blocked overnight in PBS containing 0.1% Tween 20 (Sigma) with 4% BSA to renature the proteins. The membrane was then washed with HEPES binding buffer (25 mM HEPES [pH 7.9], 50 mM NaCl, 1 mM EDTA, 1 mM dithiothreitol, 0.5% BSA) and blotted with in vitro translated ³⁵S-zyxin in the same buffer for 1 h at room temperature. The membrane was then washed five times with HEPES binding buffer, dried, and exposed to film.

Cell culture and transfections. Cos7, Mewo, and 293T cells were grown and maintained in Dulbecco's modified Eagle's medium (Gibco BRL, Grand Island, N.Y.) containing 10% fetal bovine serum (HyClone Laboratories Inc., Logan, Utah). Mewo cells are a human melanoma cell line that was obtained from Charles Grose, University of Iowa. 293T cells were transfected by the calcium phosphate precipitation method as previously described (78). Cos7 and Mewo cells were transfected with Lipofectamine Plus reagents from Gibco BRL following the manufacturer's protocol.

Coimmunoprecipitation. For coimmunoprecipitation assays, 293T cells in 10-cm-diameter dishes were transfected with 20 µg of zyxin-Flag, Gps2-Flag, PKC-Flag, or vector DNA. Thirty-six hours after transfection, cells were collected into ice-cold PBS and pelleted at 1,700 × g for 5 min at 4°C. All of the following steps were performed on ice. The cells were resuspended in 1 ml of lysis buffer (50 mM Tris-HCl [pH 7.5], 150 mM NaCl, 1% NP-40, 100 mM NaF, 10% glycerol, 200 µM Na₃VO₄, 10 µg of aprotinin per ml, 10 µg of leupeptin per ml, 1 mM phenylmethylsulfonyl fluoride) and sonicated for 1 min. After centrifugation at 14,000 rpm in a microcentrifuge for 5 min, the cell lysate was precleared by incubation with 50 µl of a 10% suspension of IgGsorb (The Enzyme Center, Malden, Mass.) in lysis buffer for 15 min and then the beads were pelleted. The supernatant was transferred to a fresh tube, and the total protein concentration was determined. One milligram of cell protein in a total volume of 500 µl of lysis buffer was mixed with 10 µl of [³⁵S]cysteine-labeled in vitro-translated E6 proteins for 2 h to overnight at 4°C. One hundred microliters of anti-Flag M2 affinity gel (Kodak, New Haven, Conn.) was added to the mix, and incubation was continued for another hour. The beads were harvested and washed five times with 1 ml of lysis buffer per wash by rotating at 4°C for 5 min and then centrifuging the beads in a microcentrifuge at 7,000 rpm for 20 s. After the final wash, the beads were boiled for 5 min in SDS loading buffer and subjected to SDS-PAGE. The coprecipitated E6 proteins were visualized with a phosphorimager.

Luciferase assay. Cos7 cells in 35-mm-diameter plates were harvested 36 h after transfection into 300 µl of 1× lysis buffer (Dual-Luciferase Reporter Assay system; Promega) per well. The luciferase activities from both firefly and Renilla luciferase constructs were quantified with a Berthold Lumat LB9501 luminometer (Wallac Inc., Gaithersburg, Md.) using the reagents and protocol provided by the manufacturer.

Immunofluorescence. Transfected cells grown on cover slides in six-well plates were fixed in 4% paraformaldehyde for 10 min at room temperature. They were then permeabilized with 0.5% NP-40 in PBS. After blocking in 10% normal goat serum for 0.5 h, the cells were stained with anti-zyxin to detect zyxin or anti-Myc (BAbCO, Richmond, Calif.) to detect E6-Myc. Anti-zyxin is an antibody made by immunizing a rabbit with a peptide corresponding to zyxin amino acids 138 to 207. For anti-zyxin the dilution was 1/50 while for anti-Myc the dilution was 1/100. To detect GBD fusions a 1/50 dilution of a monoclonal anti-GBD serum (Clontech) was used. After 0.5 h of incubation with the primary antibodies, cells were washed six times with PBS before incubating in a 1/200 dilution of anti-rabbit-rhodamine and anti-mouse-fluorescein isothiocyanate (Kirkegaard & Perry) for 0.5 h. Cells were washed six times before mounting onto slides. Preparations were viewed using a Leitz Dialux microscope with optical systems for the selective visualization of fluorescein or rhodamine. Representative fields of cells were photographed using a Kodak DC120 digital camera adapted for photomicroscopy (Eastman Kodak, Rochester, N.Y.). Images were acquired on a Macintosh computer using the Kodak plug-in for Adobe Photoshop. Digitized images were assembled using Adobe Illustrator, v. 9.02, on a Macintosh computer and printed on a Shinko dye sublimation printer. For confocal imaging, an Alexa Fluor 546-conjugated anti-rabbit immunoglobulin G and an Alexa Fluor 488-labeled anti-mouse immunoglobulin G (Molecular Probes, Inc., Eugene, Oreg.) were used to enhance the sensitivity of the detection. A Zeiss LSM 510 NLO Multiphoton confocal microscope and its associated software was used to capture the images, which were transferred to Adobe Photoshop, v. 5.5, for compiling before assembly using Adobe Illustrator, v. 9.02, and were then printed on an Epson 900 inkjet printer.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

A yeast two-hybrid library screen identifies zyxin as a protein partner for ₆E6. To identify novel proteins that interact with high-risk or low-risk E6 proteins we conducted a yeast two-hybrid library screen using low-risk ₆E6 or high-risk ₁₈E6 fused to the DNA binding domain of Gal4 (Gal4-BD) as baits. The yeast two-hybrid library contained Gal4 activation domain (Gal4-AD) fusions of cDNAs from normal human foreskin keratinocytes, which are the natural host cells for HPV. Yeast were cotransformed with the Gal4-BD vector or a Gal4-BD fusion with laminin and the putative interacting clones to eliminate false positives and the positive clones that survived this screen were tested further. Positive interactors were tested in two other yeast strains, Y190 and Y187, besides the YGH1 strain that was used for screening the library. The DNA sequences of the clones that were positive in all three strains of yeast were determined. Of the 10 clones that interacted with ₆E6-Gal4-BD 9 encoded various lengths of zyxin and the other encoded Gps2 (22). However, zyxin did not appear even once among the positive clones using ₁₈E6-Gal4-BD as bait. When cotransformed with ₁₈E6-Gal4-BD, none of the Gal4-AD fusions of the various zyxin clones showed positive interaction as judged by a -Gal filter lift assay (data not shown), suggesting that zyxin interacted with ₆E6 but not ₁₈E6. A full-length cDNA of zyxin was acquired (48) and tested for interaction with E6 proteins in the yeast two-hybrid system. Again, only yeast colonies cotransformed with the zyxin-Gal4-AD fusion and ₆E6-Gal4-BD but not ₁₈E6-Gal4-BD or Gal4-BD vector alone turned blue in the -Gal lift assay (data not shown), indicating that zyxin interacts with ₆E6 specifically.

Zyxin and ₆E6 interact in vitro. To test if zyxin could interact with ₆E6 directly, in vitro-translated, ³⁵S-labeled full-length zyxin was allowed to interact with identical amounts of immobilized GST and GST fusions with ₆E6 or ₁₈E6. After binding in the presence of 50 mM KCl, the beads with bound proteins were washed extensively with buffers containing increasing concentrations of salt. At a 50 mM [K⁺] 6.9% of the ³⁵S-labeled zyxin was retained by beads containing GST-₆E6, and at 250 mM salt 4.9% of the input remained on the beads (Fig. 1). In contrast, <1% of zyxin was retained by beads containing GST-₁₈E6 even at 50 mM [K⁺]. These results demonstrate that the interaction between ₆E6 and zyxin is stable and can exist under physiological conditions. Moreover, in this assay system zyxin is preferably bound by ₆E6 (Fig. 1).

View larger version (15K): [in this window] [in a new window]   FIG. 1.   Zyxin binding to GST-E6 proteins. In vitro-translated and 35S-labeled zyxin was allowed to interact with identical amounts of purified GST-6E6, GST-18E6, or GST alone. After binding at 50 mM [K+], the beads were washed extensively with buffers containing 50, 150, or 250 mM salt. Bound proteins were eluted and separated by SDS-PAGE. Zyxin was visualized by autoradiography of the dried gel. The numbers above each lane are the millimolar salt concentrations used for washing the beads.

View larger version (71K): [in this window] [in a new window]   FIG. 2.   Coimmunoprecipitation of 6E6 with zyxin. 293T cells were transfected with DNA encoding zyxin-Flag, Gps2-Flag, PKC-Flag, or the Flag vector alone. Thirty-six hours after transfection, cell lysates were prepared and mixed with in vitro-translated and 35S-labeled 6E6 or 18E6 (A). Flag-tagged proteins were then immunoprecipitated with an anti-Flag antibody and the coprecipitated E6 proteins were visualized with a phosphorimager after separation by SDS-PAGE. The Flag proteins retained on the Sepharose beads after immunoprecipitation were detected by Western analysis with an anti-Flag antibody after boiling the beads in SDS loading buffer and separating the bound proteins by SDS-PAGE (B). Cos7 cells were transfected with DNA encoding Flag-tagged zyxin or Gps2 alone, E6-Myc constructs alone, or the combination of a Flag-tagged protein construct and an E6 construct. Thirty-six hours after transfection, cells were labeled with 35S-translabel for 3 h, and the cells were harvested and lysed. Transfected cell lysates with equal amounts of total protein were reacted with an M2 anti-Flag affinity gel. E6 proteins that were spun out with the gel were visualized by autoradiography following SDS-PAGE separation of the bound proteins. The bands corresponding to zyxin, Gps2, and the E6 proteins are marked with arrows (C). (The protein levels of E6 in each of the transfected cell lysates were detected by Western analysis using an anti-Myc antibody after SDS-PAGE and are shown in the lower part of this panel.) This experiment was repeated without labeling the cells. After separating the proteins that eluted off of the gel by SDS-PAGE, the E6 proteins and the Flag-tagged proteins were detected by Western analysis using an anti-Myc antibody (upper panel) and an anti-Flag antibody (lower panel), respectively (D).

Zyxin and ₆E6 interact in vivo. To test if the interaction between ₆E6 and zyxin occurs in mammalian cells, Cos7 cells were transfected with DNA constructs that expressed Flag-tagged zyxin or Gps2, Myc-tagged ₆E6 or ₁₈E6, or combinations of a Flag-tagged and a Myc-tagged construct. Thirty-six hours posttransfection, cells were labeled with ³⁵S for 3 h and then cell lysates were prepared. Flag-tagged proteins were immunoprecipitated with an anti-Flag affinity gel. Proteins that coprecipitated with the Flag-tagged proteins were separated by SDS-PAGE and E6 proteins were identified based on their migration after autoradiography. The beads did not retain either ₆E6 or ₁₈E6 alone (Fig. 2C). In the cells cotransfected with zyxin, ₆E6 but not ₁₈E6 was detected on the beads, even though ₁₈E6 was readily detected on the beads incubated with the extract from cells cotransfected with Gps2 (Fig. 2C). Western analysis using an anti-Myc antibody revealed that E6 protein levels were similar in the presence or absence of the Flag-tagged proteins (Fig. 2C, lower panel). Furthermore, ₁₈E6 was present at higher levels than ₆E6 in the lysates but did not coprecipitate with zyxin (Fig. 2C, lower panel). These experiments demonstrate that ₆E6 but not ₁₈E6 interacts with zyxin in vivo. Because identification of E6 in the previous experiment was based on migration and its presence or absence in the appropriate lanes we sought to confirm this observation using an alternative approach. Accordingly, the experiment was repeated and E6-Myc proteins that interacted with Flag-tagged zyxin or Gps2 were identified by Western blot analysis using an anti-Myc antibody. Analysis of the resulting gel (Fig. 2D) demonstrates that ₆E6 but not ₁₈E6 interacts with zyxin, confirming the results of the experiment shown in Fig. 2C.

Zyxin interacts with ₆E6 but not with other E6 proteins. Because zyxin was found to interact with low-risk ₆E6 but not with high-risk ₁₈E6, we next asked if it interacted with other E6 proteins. To test for interactions, yeast cells were cotransformed with DNA constructs expressing full-length zyxin-Gal4-AD and -Gal4-BD fusions with ₆E6, ₁₈E6, ₁₁E6, or ₁₆E6. Assays for -Gal activity were performed on extracts from the transformants and the -Gal units from this assay were used to determine if there was an interaction between zyxin and the various E6 proteins. The results of this assay suggest that only ₆E6 interacts with zyxin and that none of the other E6 proteins, including ₁₁E6, another low-risk E6 protein, interact (Fig. 3A). In addition, cotransformation of yeast with each of the E6-Gal4-BD constructs and Gal4-AD vector DNA failed to activate the -Gal reporter (data not shown). To verify this finding, we tested the binding between zyxin and different E6 proteins in another system. Full-length zyxin was labeled with ³⁵S by coupled in vitro transcription-translation and mixed with identical amounts of GST-E6 fusion proteins bound to agarose beads. Again, zyxin was retained significantly only by GST-₆E6 (3.3% of input) and not by any of the other GST fusion proteins tested, including GST-₁₁E6 (<0.5%) (Fig. 3B). Figure 3C demonstrates that similar amounts of GST fusion proteins were used in the binding assay. The results of these experiments suggest that the interaction between zyxin and ₆E6 might be unique and not a general property of low-risk E6 proteins.

View larger version (19K): [in this window] [in a new window]   FIG. 3.   Interaction between E6 proteins and zyxin. (A) Plasmid DNAs encoding Gal4-BD fusions with 6E6, 11E6, 16E6, or 18E6 were cotransformed with a construct expressing a zyxin-Gal4-AD fusion into yeast strain YGH1. Liquid -Gal assays were performed on the cotransformed colonies as described in Materials and Methods. The numbers for the -Gal units are the averages of three separate experiments. (B) In vitro-translated 35S-labeled zyxin was allowed to bind to the same amounts of purified GST-E6 fusion proteins or GST alone as described in Materials and Methods. The bound zyxin protein was visualized by autoradiography after SDS-PAGE analysis. One-twentieth of the zyxin input was also electrophoresed on the gel. (C) Coomassie brilliant blue staining of the gel after SDS-PAGE was used to detect the amount of each GST fusion protein used for this assay.

View larger version (38K): [in this window] [in a new window]   FIG. 4.   Specificity of zyxin interactions with 6E6. (A) The amino acid sequences of 6E6 and 11E6 are aligned for comparison and only the amino acids in 11E6 that differ are shown. The vertical line denotes the junction between the sequences from 11E6 and the sequences from 6E6 in the chimeric protein 11/6E6. (B) Schematic representations of E6 proteins and their abilities to interact with zyxin in the yeast two-hybrid system or in the far Western analysis are shown. +, zyxin interaction is seen; , no interaction; ND, not tested. (C) Equal amounts of GST fusion proteins with 6E6, 11/6E6, or 11E6 were tested for binding to in vitro-translated 35S-zyxin in a far Western analysis as described in Materials and Methods, and the resulting autoradiograph is shown in the upper panel. The GST-E6 fusion proteins used for this assay were detected by Coomassie brilliant blue staining and are shown in the lower panel.

The LIM domains in zyxin are responsible for ₆E6 binding. Zyxin has two well-described protein-binding motifs, the proline-rich domain at the N terminus and the three adjacent LIM domains that occupy the C-terminal 189 amino acids (Fig. 5). Each of the zyxin clones identified in the yeast two-hybrid library screen contained all three of the LIM domains, but most lacked an intact proline-rich domain. To identify regions in zyxin that are responsible for ₆E6 binding, deletion mutants were constructed and tested for ₆E6 binding. Zy-5' contains all of the N-terminal amino acids that precede the LIM domain region. LIM-1+2+3 contains all three LIM domains, while LIM-1+2 contains LIM domains 1 and 2, LIM-2+3 contains LIM domains 2 and 3, LIM-1 contains the N-terminal LIM domain 1, and LIM-3 contains the C-terminal LIM domain 3 (Fig. 5). Each of these deletion mutants was cloned as a Gal4-AD fusion and cotransformed into yeast with ₆E6-Gal4-BD. Filter lift assays revealed that the yeast colonies cotransformed with the ₆E6-Gal4-BD construct and a full-length zyxin construct (FL) or zyxin construct LIM-1+2+3, LIM-2+3, or LIM-3 turned blue in the presence of X-Gal. However, cotransformation with zyxin constructs lacking the LIM-3 domain, such as Zy-5', LIM-1+2, and LIM-1, resulted in colonies that lacked detectable -Gal activity (data not shown). Thus, the LIM domains, but not the 5' proline-rich domain in zyxin, were required for binding. Notably, it was LIM-3, the most C-terminal LIM domain, that seemed to be essential for interaction with ₆E6. To examine this further, colonies from each cotransformation were picked and liquid -Gal assays were performed. These analyses supported our initial observation that the LIM domains were required for the interaction and that the Zy-5' clone did not produce a protein that interacted with ₆E6 (Fig. 6A). They also revealed a requirement for the LIM-3 domain because LIM-2+3 interacted while LIM-1+2 did not. While LIM-3 is required, the other two LIM domains seemed to facilitate the interaction, as the level of interaction with ₆E6 (in descending order) was LIM-1+2+3 > LIM-2+3 > LIM-3 (Fig. 6A). Although colonies cotransformed with ₆E6 and LIM-3 eventually turned blue in the filter lift assay this interaction was so weak that the level of -Gal activity was barely above background in the liquid -Gal assay (Fig. 6A). Because the Gal4-AD vector used for cloning zyxin has a hemagglutinin (HA) tag, we were able to monitor the expression levels of the deletion mutants in yeast by Western analysis using an anti-HA antibody. All of the zyxin deletion mutants were expressed in yeast (Fig. 6B).

View larger version (14K): [in this window] [in a new window]   FIG. 5.   Schematic diagram of zyxin. The boundaries of zyxin and the various zyxin deletion mutants constructed for this study are shown, with 6(3) being the shortest clone that interacted with 6E6 in the two-hybrid library screen. The FPPPP homology, the 5' proline-rich domain, and the three LIM domains at the C terminus are identified.

View larger version (35K): [in this window] [in a new window]   FIG. 6.   Identification of 6E6 interactive domains in zyxin. (A) Yeast strain YGH1 was cotransformed with DNA constructs encoding Gal4-AD fusions to zyxin deletions and 6E6-Gal4-BD (+E6) or Gal4-BD vector (E6). Data for the LIM proteins LIM-1+2+3 (1+2+3), LIM-2+3 (2+3), LIM-3 (3), LIM-1+2 (1+2), and LIM-1 (1) are shown. Liquid -Gal assays were performed on the cotransformants as described in Materials and Methods. The values are the average -Gal units from three separate experiments. (B) Yeast cotransformants were grown in Leu Trp liquid medium until the OD600 reached 0.4 to 0.6. Cells were pelleted and boiled in SDS loading buffer. Equal amounts of total cell extracts were subjected to SDS-PAGE. The expression of the Gal4-AD fusion proteins was detected by Western analysis using an anti-HA antibody. (C) Binding between zyxin deletion proteins and GST-6E6. In vitro-translated 35S-labeled full-length zyxin (FL) or zyxin deletion mutant proteins Zy-5' and LIM-1+2+3 were allowed to interact with the same amount of GST-6E6 fusion protein or GST alone as described in Materials and Methods. The bound proteins were separated by SDS-PAGE and visualized by autoradiography. One-twentieth of the input of each 35S-labeled protein was loaded on the same gel to quantify relative binding.

Zyxin proteins are transactivators in yeast. When the full-length zyxin was fused to Gal4-BD it strongly activated the -Gal reporter in yeast. A similar level of activation was observed when zyxin was fused to LexA and transformed into the L40 strain (data not shown). To identify the regions in zyxin that confer the transactivation activity, the zyxin deletion mutants were fused to the Gal4-BD and transformed into yeast YGH1, and liquid -Gal assays were performed on the transformants (Fig. 5 and 7). The full-length clone (FL) and a construct containing amino acids 1 to 382 (Zy-5') displayed transactivation activity, whereas a construct that expressed only the three LIM domains (Lim-1+2+3) did not activate transcription when expressed as a Gal4-BD fusion protein (Fig. 7). Thus, the region 5' of the LIM domains in zyxin has transactivation activity in yeast. Western analysis using an anti-Gal4-BD antibody showed that all of the constructs containing LIM domains were expressed (data not shown).

View larger version (13K): [in this window] [in a new window]   FIG. 7.   Analysis of the transcriptional activator activity of zyxin in yeast. Yeast strain YGH1 was transformed with DNAs expressing the Gal4-BD fused to full-length zyxin (FL), zyxin amino acids 1 to 382 (Zy-5'), and LIM-1+2+3 (1+2+3). The nomenclature used to describe the proteins is described in the legend for Fig. 5. Liquid -Gal assays were performed on the yeast transformants and -Gal units are plotted on the y axis.

Zyxin proteins have transactivation activity in mammalian cells. We next asked if zyxin can be an activator in mammalian cells. DNAs expressing full-length zyxin or Zy-5', each of which functioned as an activator in yeast, were cloned into a mammalian expression vector as Gal4-BD fusions and cotransfected into Cos7 cells with G5-luc, a luciferase reporter with multiple Gal4 binding sites in its promoter. These constructs exhibited minimal or no transactivation activity when expressed by themselves (Table 1). While the ₆E6-Myc and ₁₁E6-Myc clones each stimulated reporter activity, inclusion of ₆E6-Myc in the cotransfections resulted in stimulation of the transactivation activity of the full-length zyxin construct but not of Zy-5' (Table 1). Indeed, Zy-5' alone had no higher level of activation than empty vector. When cotransfected with ₆E6 the level of activation was less than what was detected with ₆E6 alone. In contrast, cotransfection of these constructs with ₁₁E6 resulted in much lower levels of gene activation. The action of ₁₁E6-Myc in the presence of FL-GBD is additive whereas the readout from cotransfections with ₆E6-Myc and FL-GBD reflect a synergistic response. This experiment suggests that exogenously expressed zyxin, when bound to Gal4-BD, has transcriptional activation potential and that this activity is synergistically enhanced by ₆E6 when the interacting LIM domain is present in the zyxin construct.

₆E6 alters the cellular localization of ectopically expressed zyxin proteins. Because ₆E6 stimulated transactivation by FL but not Zy-5' the cellular localization of the respective Gal4-BD fusions was analyzed in Cos7 cells cotransfected with constructs encoding Myc-tagged ₆E6 or the Myc epitope. Twenty-four hours after transfection, cells were fixed and stained for zyxin and ₆E6 using anti-zyxin and anti-Myc antibodies, respectively. As shown in Fig. 8A and C the FL and Zy-5' fusion proteins were found to localize predominantly to the cytoplasm. However, following cotransfection with ₆E6, a portion of FL fusion protein was detected in the nucleus where most of the ₆E6 was localized (Fig. 8B and F). Redistribution of the Zy-5' fusion protein did not occur (compare panels C and D in Fig. 8), perhaps because it lacks the LIM domains and therefore does not interact with ₆E6 (Fig. 6).

View larger version (34K): [in this window] [in a new window]   FIG. 8.   Cellular localization of zyxin. One-half of a microgram of FL-M-GBD or Zy-5'-M-GBD DNA was cotransfected into Cos7 cells with 1.5 µg of a construct expressing 6E6-Myc (B, D) or the Myc vector (A, C). Thirty-six hours after transfection, cells were fixed and zyxin proteins were detected with a polyclonal rabbit anti-zyxin antibody and visualized after interaction with a rhodamine-conjugated anti-rabbit serum (A to D), while E6 proteins were detected with a mouse monoclonal anti-Myc antibody followed by fluorescein isothiocyanate-labeled anti-mouse antibody (E to H).

View larger version (101K): [in this window] [in a new window]   FIG. 9.   Distribution of endogenous zyxin in Mewo cells cotransfected with E6 proteins. Mewo cells were transfected with DNA constructs expressing 6E6-Myc (A to F), 11E6-Myc (G to I), 18E6-Myc (J to L), or 11/6E6-Myc (M to O). Thirty-six hours after transfection, cells were fixed and the distribution of zyxin (A, B, C, E, H, K, and N) and E6 (A, B, C, D, G, J, and M) in the cells was detected by confocal microscopy as described in Materials and Methods. Panels A, B, and C are the same image photographed at three different depths of field to demonstrate that the E6 signal is nuclearly located. Panels F, I, L, and O are the merged images of panels D and E, G and H, J and K, and M and N, respectively.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

E6 and E7, the major transforming proteins from HPV, do not possess any known enzymatic activities. Rather, they promote cell proliferation by interacting with host cell factors and interfering with their normal functions. Several cell proteins have been identified as interaction partners for HPV E6. However, most of these only interact with high-risk HPV E6 (for review, see reference 43). Bak, Gps2, Mcm7, and p300 are the only four proteins that have been described as interacting with E6 from both high-risk and low-risk HPVs (22, 44, 54, 69). Here we report the identification and characterization of a protein that interacts specifically with ₆E6. This protein is zyxin, a focal adhesion molecule. As a result of this interaction a portion of the cell's zyxin translocates to the nucleus and can now serve as a transcriptional activator.

We first identified the interaction between ₆E6 and zyxin by screening a yeast two-hybrid library using ₆E6 as bait. Subsequent analyses using GST binding and coimmunoprecipitation demonstrated that ₆E6 and zyxin interact both in vitro and in vivo (Fig. 1 and 2). However, we were unable to detect any significant interaction between ₁₆E6 or ₁₈E6 and zyxin in any of these assays (Fig. 3). This is the first instance of a protein that interacts preferably with low-risk and not high-risk HPV E6. However, zyxin is not a general protein partner for low-risk E6 proteins. In both the two-hybrid system and GST binding assays ₁₁E6, another closely related low-risk E6 protein, also failed to interact with zyxin. This is not the first description of a protein that interacts with one type of HPV E6 but not with the whole group of high-risk or low-risk HPV E6s. E6-AP was found to interact with ₁₈E6 with a much lower affinity than with ₁₆E6 (40) and IRF-3 interacts selectively with ₁₆E6 (60).

HPV6 is detected in a higher percentage of genital warts than any other HPV (30, 31). Although HPV6 is rarely found in cervical tumors, tumors at other sites and Buschke-Lowenstein condylomata have been reported to harbor HPV6 (9, 30). We do not know if the interaction of ₆E6 with zyxin enhances replication of HPV6 in epithelial tissue. However, Hirota et al. recently demonstrated that zyxin can control mitosis progression by forming a regulatory complex with the h-warts/LATS1 tumor suppressor on the mitotic apparatus (37).

Zyxin is a protein with distinct structural features. The N-terminal two-thirds consists of a proline-rich domain that interacts with Src homology 3 domains and serves as a binding site for -actinin and Ena/VASP family members (19, 23, 38). These interactions enhance the production of actin-rich structures at the apical surface of cells and contribute to their positioning (23, 24). The C-terminal one-third of zyxin contains three LIM domains. The LIM domain is a cysteine- and histidine-rich motif of approximately 60 amino acids in length (for review, see reference 66). Each LIM domain coordinates two Zn²⁺ ions. The LIM motif is present in a number of proteins that are involved in the control of gene expression and cell differentiation (2, 68, 73, 76). LIM motifs can also serve as an interface for protein interactions (66) and this motif in zyxin interacts with cysteine-rich protein family members that are involved in differentiation (20). By analyzing the ability of a series of deletion mutants of zyxin (Fig. 5) to interact with ₆E6 we demonstrated that the LIM domains are responsible for binding this HPV protein (Fig. 6). In particular, LIM-3, the most C-terminal LIM domain, is required for ₆E6 binding, as demonstrated by the ability of LIM-2+3 but not LIM-1+2 to bind ₆E6 (Fig. 6). The other two LIM domains enhance binding to ₆E6, as LIM-1+2+3 binds ₆E6 better than LIM-2+3 and LIM-2+3 binds better than LIM-3 (Fig. 6).

What is the function of zyxin? As noted by others the functional conservation of the FPPPP motifs in zyxin and the ActA protein of Listeria monocytogenes and the ability of zyxin to bind to -actinin support a role for this protein in the organization of actin (19, 32). However, certain features of zyxin also suggest that it may function as a signaling protein by communicating between the adhesion plaque and the nucleus. It is much less abundant than the structural components of the adhesion plaques and it has multiple protein-binding motifs that could mediate interactions with other proteins, including SH3 domain-bearing proteins that are important signal transducers (for a review, see reference 8). It also has a nuclear export sequence and chicken zyxin can shuttle between the nucleus and focal adhesions (52). These properties are compatible with a role for zyxin as a messenger to relay information from the extracellular matrix to the nucleus or even to directly participate in the regulation of gene expression. In support of the latter proposal we note that many LIM domain proteins are well-characterized transcription factors (3, 21). Fusion of zyxin to the Gal4-BD resulted in transactivation in yeast (Fig. 7). Zyxin also exhibited transactivation activity in mammalian cells (Table 1). The preliminary deletion analysis conducted in yeast supports a role for the proline-rich region that is 5' of the LIM domains in transactivation.

While the experiments reported here make use of overexpression vectors to demonstrate in vivo interactions, coexpression of ₆E6 with the Gal4-BD fusion of FL in mammalian cells resulted in enhanced reporter activity (Table 1). While ₆E6 alone enhanced luciferase expression we consistently observed greater stimulation when ₆E6 was coexpressed with FL than when it was expressed with Zy-5'. This might be explained by our observation that while both of these proteins localize to the cytoplasm in the absence of ₆E6 (Fig. 8A and C) only FL is found in the nucleus of cells cotransfected with ₆E6 (Fig. 8B versus D). The Zy-5' protein probably remained predominantly cytoplasmic, even in the presence of ₆E6 (Fig. 8C and D), because it lacks the LIM domains that are involved in interacting with ₆E6 (Fig. 6). E6 proteins have been reported to localize in the membranous compartments of the nucleus and cytoplasm (1, 33). We found that ₆E6 predominantly localized in the nucleus (Fig. 8F and H; Fig. 9D, G, J, and M). Nuclear staining of either endogenous zyxin or the product of a cotransfected zyxin allele was only observed in cells transfected with ₆E6. Thus, ₆E6 might also act as a charon and ferry excess zyxin to the nucleus where it can be available to stimulate transcription.

lpp is a gene that is involved in lipomas (56). It encodes a zyxin family member that shares 41% sequence identity with zyxin, contains a proline-rich domain and three LIM domains, and also localizes to focal adhesion plaques and points of cell-to-cell contact. LPP was recently shown to shuttle between the nucleus and cytoplasm and to function as a transactivator (55). We speculate that one of the physiological functions of zyxin family members is to shuttle to the nucleus from the cellular membrane in response to certain extracellular stimuli. Once there, they have the potential to activate transcription of specific genes. ₆E6 may potentiate this by transporting and retaining these proteins in the nucleus (Fig. 8 and 9).

Cell-to-cell communication and cell adhesion to an extracellular matrix can affect cell proliferation and differentiation (17, 34). An imbalance in these activities can lead to cellular transformation. In tumor cells, disruption of adhesion plaques is a hallmark of the transformed phenotype (14). Two other zyxin family members have been linked to tumors. The gene encoding LPP was first identified during an analysis of chromosomal translocation events that are associated with lipomas, a group of common benign mesenchymal tumors in humans (67). The TRIP6 gene was assigned to a segment of chromosome 7q22 that is commonly deleted in malignant myeloid diseases and uterine leiomyoma (79). Zyxin itself has been found to interact with differentiation proteins and proto-oncogenes (for review, see reference8). These findings raise the possibility that zyxin might also be linked directly or indirectly to transformation. High-risk but not low-risk E6 proteins interact with paxillin (70-72), another focal adhesion protein (64). Here, we have described the interaction of zyxin with ₆E6, the E6 protein from the most common HPV type associated with genital warts.