Linking β-Catenin to Androgen-signaling Pathway
2002; Elsevier BV; Volume: 277; Issue: 13 Linguagem: Inglês
10.1074/jbc.m111962200
ISSN1083-351X
AutoresFajun Yang, Xiaoyu Li, Manju Sharma, Carl Y. Sasaki, Dan L. Longo, Bing Lim, Zijie Sun,
Tópico(s)Polyamine Metabolism and Applications
ResumoThe androgen-signaling pathway is important for the growth and progression of prostate cancer cells. The growth-promoting effects of androgen on prostate cells are mediated mostly through the androgen receptor (AR). There is increasing evidence that transcription activation by AR is mediated through interaction with other cofactors. β-Catenin plays a critical role in embryonic development and tumorigenesis through its effects on E-cadherin-mediated cell adhesion and Wnt-dependent signal transduction. Here, we demonstrate that a specific protein-protein interaction occurs between β-catenin and AR. Unlike the steroid hormone receptor coactivator 1 (SRC1), β-catenin showed a strong interaction with AR but not with other steroid hormone receptors such as estrogen receptor α, progesterone receptor β, and glucocorticoid receptor. The ligand binding domain of AR and the NH2terminus combined with the first six armadillo repeats of β-catenin were shown to be necessary for the interaction. Through this specific interaction, β-catenin augments the ligand-dependent activity of AR in prostate cancer cells. Moreover, expression of E-cadherin in E-cadherin-negative prostate cancer cells results in redistribution of the cytoplasmic β-catenin to the cell membrane and reduction of AR-mediated transcription. These data suggest that loss of E-cadherin can elevate the cellular levels of β-catenin in prostate cancer cells, which may directly contribute to invasiveness and a more malignant tumor phenotype by augmenting AR activity during prostate cancer progression. The androgen-signaling pathway is important for the growth and progression of prostate cancer cells. The growth-promoting effects of androgen on prostate cells are mediated mostly through the androgen receptor (AR). There is increasing evidence that transcription activation by AR is mediated through interaction with other cofactors. β-Catenin plays a critical role in embryonic development and tumorigenesis through its effects on E-cadherin-mediated cell adhesion and Wnt-dependent signal transduction. Here, we demonstrate that a specific protein-protein interaction occurs between β-catenin and AR. Unlike the steroid hormone receptor coactivator 1 (SRC1), β-catenin showed a strong interaction with AR but not with other steroid hormone receptors such as estrogen receptor α, progesterone receptor β, and glucocorticoid receptor. The ligand binding domain of AR and the NH2terminus combined with the first six armadillo repeats of β-catenin were shown to be necessary for the interaction. Through this specific interaction, β-catenin augments the ligand-dependent activity of AR in prostate cancer cells. Moreover, expression of E-cadherin in E-cadherin-negative prostate cancer cells results in redistribution of the cytoplasmic β-catenin to the cell membrane and reduction of AR-mediated transcription. These data suggest that loss of E-cadherin can elevate the cellular levels of β-catenin in prostate cancer cells, which may directly contribute to invasiveness and a more malignant tumor phenotype by augmenting AR activity during prostate cancer progression. Prostate cancer is the most commonly diagnosed malignancy among males in western countries (1.Landis S.H. Murray T. Bolden S. Wingo P.A. CA-Cancer J. Clin. 1999; 49: 8-31Crossref PubMed Scopus (3135) Google Scholar). However, in contrast to some other tumors, the molecular events involved in the development and progression of prostate cancer remain largely unknown. Androgen ablation, used as an effective treatment for the majority of advanced prostate cancers, indicates that androgen plays an essential role in regulating the growth of prostate cancer cells. The growth-promoting effects of androgen in prostate cells are mediated mostly through the androgen receptor (AR). 1The abbreviations used are: ARandrogen receptorGRglucocorticoid receptorERαestrogen receptor αPRβprogesterone receptor βVDRvitamin D receptorTADtranscription activation domainDBDDNA binding domainLBDligand binding domainDHTdihydrotestosteronePSAprostate specific antigenAREandrogen responsive elementMMTVmouse mammary tumor virusGSTglutathione S-transferaseARA70androgen receptor-associated protein 70β-galβ-galactosidaseAPCadenomatous polyposis coli there is increasing evidence that the nuclear hormone receptors, including AR, interact with other signal transduction pathways (2.Jenster G. Semin. Oncol. 1999; 26: 407-421PubMed Google Scholar). The regulation by cofactors can modulate AR activities, which may contribute to the development and progression of prostate cancer. androgen receptor glucocorticoid receptor estrogen receptor α progesterone receptor β vitamin D receptor transcription activation domain DNA binding domain ligand binding domain dihydrotestosterone prostate specific antigen androgen responsive element mouse mammary tumor virus glutathione S-transferase androgen receptor-associated protein 70 β-galactosidase adenomatous polyposis coli β-Catenin plays a pivotal role in cadherin-based cell adhesion and in the Wnt-signaling pathway (3.Polakis P. Genes Dev. 2000; 14: 1837-1851Crossref PubMed Google Scholar, 4.Morin P.J. Bioessays. 1999; 21: 1021-1030Crossref PubMed Scopus (817) Google Scholar). Corresponding to its dual functions in the cells, β-catenin is localized to two cellular pools. Most of the β-catenin is located in the cell membrane where it is associated with the cytoplasmic region of E-cadherin, a transmembrane protein involved in homotypic cell-cell contacts (5.Ozawa M. Baribault H. Kemler R. EMBO J. 1989; 8: 1711-1717Crossref PubMed Scopus (1152) Google Scholar). A smaller pool of β-catenin is located in the nucleus and cytoplasm and mediates Wnt signaling. In the absence of a Wnt signal, β-catenin is constitutively down-regulated by a multicomponent destruction complex containing GSK3β, axin, and a tumor suppressor, adenomatous polyposis coli (APC). These proteins promote the phosphorylation of serine and threonine residues in the NH2-terminal region of β-catenin and thereby target it for degradation by the ubiquitin proteasome pathway (6.Aberle H. Bauer A. Stappert J. Kispert A. Kemler R. EMBO J. 1997; 16: 3797-3804Crossref PubMed Scopus (2172) Google Scholar). Wnt signaling inhibits this process, which leads to an accumulation of β-catenin in the nucleus and promotes the formation of transcriptionally active complexes with members of the Tcf/LEF family (7.Molenaar M. van de Wetering M. Oosterwegel M. Peterson-Maduro J. Godsave S. Korinek V. Roose J. Destree O. Clevers H. Cell. 1996; 86: 391-399Abstract Full Text Full Text PDF PubMed Scopus (1625) Google Scholar). Activation of Tcf/LEF and β-catenin targets has been shown to induce neoplastic transformation in cells, suggesting a potential role of β-catenin in tumorigenesis (8.Aoki M. Hecht A. Kruse U. Kemler R. Vogt P.K. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 139-144Crossref PubMed Scopus (155) Google Scholar). The link between stabilized β-catenin and tumor development and progression was considerably strengthened by discoveries of mutations in both β-catenin and components of the destruction complex in a wide variety of human cancers, which cause increased cellular levels of β-catenin (3.Polakis P. Genes Dev. 2000; 14: 1837-1851Crossref PubMed Google Scholar, 9.Korinek V. Barker N. Morin P.J. van Wichen D. de Weger R. Kinzler K.W. Vogelstein B. Clevers H. Science. 1997; 275: 1784-1787Crossref PubMed Scopus (2950) Google Scholar). About 85% of all sporadic and hereditary colorectal tumors show loss of APC function, which correlates with the increased levels of free β-catenin found in these cancer cells (10.Kinzler K.W. Vogelstein B. Cell. 1996; 87: 159-170Abstract Full Text Full Text PDF PubMed Scopus (4286) Google Scholar, 11.Munemitsu S. Albert I. Souza B. Rubinfeld B. Polakis P. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 3046-3050Crossref PubMed Scopus (957) Google Scholar, 12.Bienz M. Clevers H. Cell. 2000; 103: 311-320Abstract Full Text Full Text PDF PubMed Scopus (1310) Google Scholar). It appears that inappropriate high cellular levels of β-catenin play a fundamentally important role in tumorigenesis. In normal epithelial tissues, E-cadherin complexes with actin cytoskeleton via cytoplasmic catenins to maintain the functional characteristics of epithelia. Disruption of this complex, due primarily to the loss or decreased expression of E-cadherin, is frequently observed in many advanced, poorly differentiated carcinomas (13.Gayther S.A. Gorringe K.L. Ramus S.J. Huntsman D. Roviello F. Grehan N. Machado J.C. Pinto E. Seruca R. Halling K. MacLeod P. Powell S.M. Jackson C.E. Ponder B.A. Caldas C. Cancer Res. 1998; 58: 4086-4089PubMed Google Scholar, 14.Guilford P. Hopkins J. Harraway J. McLeod M. McLeod N. Harawira P. Taite H. Scoular R. Miller A. Reeve A.E. Nature. 1998; 392: 402-405Crossref PubMed Scopus (1384) Google Scholar). There is a strong correlation between decreased expression of E-cadherin and an invasive and metastatic phenotype of human prostate cancers (15.Richmond P.J. Karayiannakis A.J. Nagafuchi A. Kaisary A.V. Pignatelli M. Cancer Res. 1997; 57: 3189-3193PubMed Google Scholar). Besides playing a role in retaining normal cell-cell contact, E-cadherin can also modulate the cytoplasmic pools of β-catenin for signaling (16.Sasaki C.Y. Lin H. Morin P.J. Longo D.L. Cancer Res. 2000; 60: 7057-7065PubMed Google Scholar). Here, we demonstrated a specific protein-protein interaction between β-catenin and AR. Importantly, unlike the steroid receptor cofactor 1 (SRC1), β-catenin selectively binds to AR in a ligand-dependent manner but not to other steroid hormone receptors such as the estrogen receptor α (ERα), the progesterone receptor β (PRβ), and glucocorticoid receptor (GR). The ligand binding domain (LBD) of AR and the central region spanning the armadillo repeats 1–6 of β-catenin were found to be responsible for the interaction. Using transient transfection experiments, we further demonstrated that β-catenin augments the ligand-dependent activity of AR in prostate cancer cells through this specific interaction. These data identify a new role for β-catenin in nuclear hormone receptor-mediated transcription. Moreover, transfection of an E-cadherin expression construct into an E-cadherin-negative prostate cancer cell line, TSU.pr-1, resulted in redistribution of β-catenin to the cell membrane and reduction of AR-dependent transcriptional activity. They suggest that reduced expression of E-cadherin can elevate the cellular levels of β-catenin in prostate cancer cells, which may directly contribute to the invasiveness and more malignant tumor phenotype by augmenting AR activity during the progression of prostate cancer. Yeast two-hybrid experiments were basically performed as described previously (17.Yang F. Li X. Sharma M. Zarnegar M. Lim B. Sun Z. J. Biol. Chem. 2001; 276: 15345-15353Abstract Full Text Full Text PDF PubMed Scopus (132) Google Scholar). The LBD of human AR (amino acids 629–919) was fused in frame to the GAL4 DBD in the pGBT9 vector (CLONTECH, Palo Alto, CA). The construct was transformed into a modified yeast strain PJ69-4A (18.James P. Halladay J. Craig E.A. Genetics. 1996; 144: 1425-1436Crossref PubMed Google Scholar). A cDNA library from human brain tissue was used in this screening (CLONTECH). Transformants were selected on Sabouraud dextrose medium lacking adenine, leucine, and tryptophan in the presence of 100 nm dihydrotestosterone (DHT). The specificity of interaction with AR was determined by a liquid β-galactosidase (β-gal) assay as described previously (17.Yang F. Li X. Sharma M. Zarnegar M. Lim B. Sun Z. J. Biol. Chem. 2001; 276: 15345-15353Abstract Full Text Full Text PDF PubMed Scopus (132) Google Scholar). β-Gal activities were measured using the Galacto-light Plus kit (Tropix Inc., Bedford, MA) and normalized by cell density (A600). pGBT9 constructs with three different AR fragments, including the partial TAD (amino acids 1–333), DBD (amino acids 505–676), and LBD were used to confirm the interaction. A yeast clone containing the full-length cDNA of human β-catenin was isolated in the screen. Using it as a template, the COOH-terminal and internal deletions of β-catenin clones were generated by PCR with specific primers containing the appropriate restriction enzyme sites. After cleavage, the fragments containing different portions of the β-catenin were cloned downstream of GAL4 TAD in the pGAD10 vector (CLONTECH). The LBD fragments of ERα (amino acids 250–602), PRβ (amino acids 633–952), VDR (amino acids 90–427) were generated by PCR with specific primers and subcloned in-frame to the GAL4 DBD in pGBT9. An antisense construct of β-catenin containing the NH2-terminal 513 bp was generated by PCR and cloned into the pcDNA3 vector at EcoRI site. All constructs were sequenced to confirm that there were no mutations introduced by PCR. The AR expression vector, pSV-hAR, was provided by Dr. Albert Brinkmann (Erasmus University, Rotterdam, The Netherlands). The expression constructs for human ERα and pERE-luc plasmid were generously given by Dr. Myles Brown (Dana-Farber Cancer Institute, Boston, MA). A human PRβ and PRE-luc reporter were provided by Dr. Kathryn B. Horwitz (University of Colorado). The expression constructs of human GR and VDR, and the pVDRE-luc reporter plasmid, were the kind gifts of Dr. David Feldman (Stanford University, Stanford, CA). pSV-β-gal, an SV40-driven β-galactosidase reporter plasmid (Promega, Madison, WI) was used in this study as an internal control. The pSG5-ARA70 plasmid and the reporter plasmid pARE-luc were the kind gifts of Dr. Chawnshang Chang (19.Yeh S. Chang C. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 5517-5521Crossref PubMed Scopus (532) Google Scholar). pMMTV-pA3-luc was provided by Dr. Richard Pestell (Albert Einstein College of Medicine, New York). The reporter plasmids, pPSA7kb-luc, with the luciferase gene under the control of promoter fragments of the human prostate-specific antigen was obtained from Dr. Jan Trapman (20.Cleutjens K.B. van Eekelen C.C. van der Korput H.A. Brinkman A.O. Trapman J. J. Biol. Chem. 1996; 271: 6379-6388Abstract Full Text Full Text PDF PubMed Scopus (353) Google Scholar). The monkey kidney cell line, CV-1, was maintained in Dulbecco's modified Eagle's medium supplemented with 5% fetal calf serum (HyClone, Denver, CO). An AR-positive prostate cancer cell line, LNCaP, was maintained in T-medium (Invitrogen) with 5% fetal calf serum. The two sublines derived from TSU.pr-1 prostate cancer cells (16.Sasaki C.Y. Lin H. Morin P.J. Longo D.L. Cancer Res. 2000; 60: 7057-7065PubMed Google Scholar) were maintained in RPMI 1640 medium with 10% fetal calf serum and G418 (500 μg/ml). Transient transfections were carried out using a LipofectAMINE transfection kit (Invitrogen) for CV1 and LipofectAMINE 2000 (Invitrogen) for TSU.pr-1 and LNCaP cells. Approximately 1.5–2 × 104 cells were plated in a 48-well plate 16 h before transfection. 12–16 h after transfection, the cells were washed and fed medium containing 5% charcoal-stripped fetal calf serum (HyClone) in the presence or absence of steroid hormones. Cells were incubated for another 24 h, and luciferase activity was measured as relative light units (21.Sharma M. Zarnegar M. Li X. Lim B. Sun Z. J. Biol. Chem. 2000; 275: 35200-35208Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar). The relative light units from individual transfections were normalized by β-galactosidase activity in the same samples. Individual transfection experiments were done in triplicate and the results are reported as mean relative light units/β-galactosidase (+S.D.). GST-β-catenin fusion proteins were constructed in the pGEX-4T-1 vector (Amersham Biosciences, Inc.). Expression and purification of GST fusion proteins were performed according to the manufacturer's instructions. Full-length human AR proteins were generated and 35S-labeled in vitro by the TNT-coupled reticulocyte lysate system (Promega, Madison, WI). Equal amounts of GST fusion proteins coupled to glutathione-Sepharose beads were incubated with 35S-labeled proteins at 4 °C for 2 h in the lysis buffer as described above. Beads were carefully washed three times with washing buffer (50 mm Tris-HCl, pH 8.0, 150 mm NaCl, 1% Nonidet P-40). GST fusion proteins were then eluted by incubating with buffer containing 10 mm glutathione and 50 mmTris-HCl, pH 8.0, for 10 min at room temperature. The bound proteins were analyzed by SDS-PAGE followed by autoradiography. CV-1 cells were plated onto gelatin-coated (2%) coverslips the day before transfection. The pcDNA3-AR and the wild type or mutants of β-catenin plasmids were cotransfected into cells with the LipofectAMINE-PLUS reagent (Invitrogen). After 2 h, transfected cells were fed with fresh medium plus/minus 10 nm DHT, incubated for 4 h, and then fixed for 10 min with 3% paraformaldehyde in phosphate-buffered saline and washed with 0.1% Nonidet P-40/phosphate-buffered saline buffer. Nonspecific sites were blocked with 5% skim milk powder in phosphate-buffered saline for 30 min. The cells were then incubated with either anti-FLAG monoclonal or anti-AR polyclonal antibody for 1 h at room temperature. Cells were washed three times followed by incubation with fluorescein isothiocyanate-conjugated anti-mouse or rhodamine-conjugated anti-rabbit secondary antibody (Santa Cruz Biotechnology). Indirect immunofluorescence staining was performed according to the procedure described previously (16.Sasaki C.Y. Lin H. Morin P.J. Longo D.L. Cancer Res. 2000; 60: 7057-7065PubMed Google Scholar). In TSU cells, E-cadherin was stained with the rat monoclonal antibody against uvomorulin (6 μg/ml; Sigma) and donkey anti-rat immunoglobulin conjugated to Alexa-488 (20 μg/ml; Molecular Probes, Eugene, OR). β-Catenin was stained with a mouse monoclonal anti-β-catenin antibody conjugated to TRITC (10 μg/ml; Transduction Laboratories). Using a bait construct containing the LBD and hinge region of the AR, we employed a modified yeast two-hybrid system to identify proteins that interact with AR in an androgen-dependent manner (17.Yang F. Li X. Sharma M. Zarnegar M. Lim B. Sun Z. J. Biol. Chem. 2001; 276: 15345-15353Abstract Full Text Full Text PDF PubMed Scopus (132) Google Scholar, 18.James P. Halladay J. Craig E.A. Genetics. 1996; 144: 1425-1436Crossref PubMed Google Scholar). Of 2 × 107 transformants, 73 grew under selective conditions and showed increased adenine and β-gal productions in medium containin g 100 nm DHT. Rescue of the plasmids and sequencing of the inserts revealed several different cDNAs, including the previously identified SRC1 (22.Onate S.A. Tsai S.Y. Tsai M.J. O'Malley B.W. Science. 1995; 270: 1354-1357Crossref PubMed Scopus (2063) Google Scholar), an AR-associated protein (ARA70) (19.Yeh S. Chang C. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 5517-5521Crossref PubMed Scopus (532) Google Scholar), and several other AR-interacting proteins identified recently by others or us (17.Yang F. Li X. Sharma M. Zarnegar M. Lim B. Sun Z. J. Biol. Chem. 2001; 276: 15345-15353Abstract Full Text Full Text PDF PubMed Scopus (132) Google Scholar, 21.Sharma M. Zarnegar M. Li X. Lim B. Sun Z. J. Biol. Chem. 2000; 275: 35200-35208Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar, 23.Moilanen A.M. Karvonen U. Poukka H. Yan W. Toppari J. Janne O.A. Palvimo J.J. J. Biol. Chem. 1999; 274: 3700-3704Abstract Full Text Full Text PDF PubMed Scopus (138) Google Scholar). Importantly, 23 of these clones perfectly matched the sequence of the full-length coding region of β-catenin. To confirm the interaction, we cotransformed one of these β-catenin clones with various constructs containing either GAL4DBD alone or the AR fusion proteins with a partial transactivation domain (pTAD), the DBD, and the LBD (Fig. 1A). pGAD10-β-catenin showed a specific interaction with GAL4DBD-AR-LBD by producing adenine in the presence of 100 nm DHT (data not shown). In the liquid β-gal assays, pGAD10-β-catenin showed an ∼97-fold induction with pGBT9-AR-LBD in the presence of DHT (Fig. 1B). This result demonstrated that the LBD of AR specifically interacts with β-catenin in a ligand-dependent manner. β-Catenin and its Drosophila homolog, armadillo, contain a central core domain of 12 armadillo repeats flanked by unique NH2 and COOH termini (7.Molenaar M. van de Wetering M. Oosterwegel M. Peterson-Maduro J. Godsave S. Korinek V. Roose J. Destree O. Clevers H. Cell. 1996; 86: 391-399Abstract Full Text Full Text PDF PubMed Scopus (1625) Google Scholar). To identify the region of β-catenin that interacts with AR, we generated several truncated mutants of β-catenin and assessed their ability to interact with AR using the yeast two-hybrid system (Fig. 2A). As shown in Fig. 2B, deletion of the COOH-terminal activation domain of β-catenin (β-cat-t671) alone, or in combination with the last five armadillo repeats (β-cat-t423), did not significantly affect the binding. However, a mutant in which the NH2-terminal activation domain alone (β-cat-t134–671), or in combination with the central armadillo domain (β-cat-t671–781) was deleted, showed no interaction. This result suggests that the primary binding region for AR spans the NH2 terminus and the first seven armadillo repeats of β-catenin. To precisely map the interacting region, a series of truncated mutants were made in which each single armadillo repeat was subsequently deleted (Fig. 3A). The deletion constructs containing the NH2-terminal region and the first six repeats (β-cat-t393) showed about two-thirds the activity of the full-length protein (β-cat-F) (Fig. 3A). However by deleting repeat 6 (β-cat-t350), the interaction was essentially abolished, indicating that armadillo repeat 6 is crucial for binding to AR. Deletion of repeats 1–5 obviously had little further effect. It has been shown that most β-catenin-binding proteins such as Tcf/LEF family members (24.van de Wetering M. Cavallo R. Dooijes D. van Beest M. van Es J. Loureiro J. Ypma A. Hursh D. Jones T. Bejsovec A. Peifer M. Mortin M. Clevers H. Cell. 1997; 88: 789-799Abstract Full Text Full Text PDF PubMed Scopus (1064) Google Scholar), axin (25.Behrens J. von Kries J.P. Kuhl M. Bruhn L. Wedlich D. Grosschedl R. Birchmeier W. Nature. 1996; 382: 638-642Crossref PubMed Scopus (2605) Google Scholar), APC (26.Hulsken J. Birchmeier W. Behrens J. J. Cell Biol. 1994; 127: 2061-2069Crossref PubMed Scopus (586) Google Scholar, 27.Hulsken J. Behrens J. Birchmeier W. Curr. Opin. Cell Biol. 1994; 6: 711-716Crossref PubMed Scopus (109) Google Scholar), and cadherins (26.Hulsken J. Birchmeier W. Behrens J. J. Cell Biol. 1994; 127: 2061-2069Crossref PubMed Scopus (586) Google Scholar, 28.Pai L.M. Kirkpatrick C. Blanton J. Oda H. Takeichi M. Peifer M. J. Biol. Chem. 1996; 271: 32411-32420Abstract Full Text Full Text PDF PubMed Scopus (85) Google Scholar) bind β-catenin mainly through the central armadillo repeats. In most of cases, the first 10 repeats are required for the interactions, and a minimum of 6–7 repeats are sufficient for detectable binding (29.Huber A.H. Nelson W.J. Weis W.I. Cell. 1997; 90: 871-882Abstract Full Text Full Text PDF PubMed Scopus (573) Google Scholar, 30.Graham T.A. Weaver C. Mao F. Kimelman D. Xu W. Cell. 2000; 103: 885-896Abstract Full Text Full Text PDF PubMed Scopus (346) Google Scholar). Those data are consistent with our finding that deletion of repeat 6 fully abolished the interaction with AR. To more precisely map the interaction region within the first six armadillo repeats, we used a PCR-based, site-directed mutagenesis techniques to generate several internal deletion mutants. As shown in the figure, the wild type β-catenin and the mutant with deletion of repeat 7 (ΔR7) or 12 (ΔR12) all reacted avidly with the AR-LBD. In contrast, mutants lacking repeat 6 (ΔR6) showed no interaction with the AR-LBD (Fig. 3B). Moreover, deletion of repeat 5 alone also fully abolished the interaction, indicating that the armadillo repeats 1–5 may be also involved in the interaction. To confirm this result, an additional internal deletion mutant lacking repeat 3 (ΔR3) was generated and tested. As we expected, the mutant also showed no interaction with AR (data not shown). Taken together, the results allow us to conclude that the region spanning armadillo repeats 1–6 is mainly responsible for binding to AR. Physical interaction between AR and β-catenin was further assessed by GST pull-down experiments. A series of GST fusion proteins with the full-length β-catenin and internal deletion mutants were generated and immobilized onto a glutathione-Sepharose matrix. The binding of [35S]methionine-labeled AR protein to GST-β-catenin fusion proteins was analyzed by SDS-PAGE and detected by autoradiography. As shown in Fig. 4A, AR protein bound to the GST fusion protein containing wild type β-catenin and its mutants lacking either repeat 7 or 12. The interaction is more pronounced in the presence of DHT than in the absence of DHT, and as much as 5% of the input protein was recovered (Fig. 4A). However, a significant reduction of binding was observed between the AR protein and the β-catenin mutants lacking repeat 6 when equal amounts of the GST fusion proteins were used in the experiments (Fig. 4B). These results are consistent with our observations from the yeast two-hybrid system and show a domain-dependent interaction in vitro. To confirm that endogenous AR and β-catenin are physically associated in intact cells, coimmunoprecipitation assays were carried out to detect a possible protein complex in a prostate cancer cell line, LNCaP. Using specific antibodies, we further confirmed that AR and β-catenin proteins form a protein complex in LNCaP cells and the formation of AR and β-catenin complexes in these cells was also enhanced by DHT (data not shown). These results are consistent with a recent report by Truica and colleagues (31.Truica C.I. Byers S. Gelmann E.P. Cancer Res. 2000; 60: 4709-4713PubMed Google Scholar). Next, we examined whether a dynamic interaction between β-catenin and AR existed in cells. FLAG-tagged vectors containing either full-length or mutants of β-catenin were transfected into CV-1 cells, and the expressed protein showed a cytoplasmic and nuclear distribution, which was not altered by treatment with DHT (data not shown). Overexpressed β-catenin protein with AR vector, in the absence of DHT, showed the same cellular distribution as transfection of β-catenin plasmid alone, while transfected AR protein is localized mainly in the cytoplasm (Fig. 4C, panels 1, 3, and 5). In the presence of DHT, AR proteins are fully translocated into the nuclei (panels 2, 4, and 6). Importantly, both the wild type (panel 2) and the ΔR12 mutant of β-catenin (panel 6) showed increased levels of nuclear translocation when cotransfected with AR compared with cells transfected with the ΔR6 mutant of β-catenin in which cytoplasmic staining of β-catenin persisted (panel 4). These results provide the first evidence that β-catenin can translocate into the nucleus as part of a complex with AR by an interaction through armadillo repeat 6. To assess the possibility that β-catenin functions as a general coactivator of nuclear receptors, we examined the interaction of β-catenin with other members of the nuclear receptor family in yeast. The LBD of ERα, PRβ, and VDR were generated and fused to GAL-DBD in the pGBT9 vector. These plasmids were cotransformed with either pGAD10-β-catenin or pGAD10-SRC1 as a positive control in the presence of corresponding ligands. The yeast transformants were grown on the selective media, and a liquid β-gal assay was performed to quantify the interactions. All receptors were shown to have a ligand-dependent interaction with the SRC1 (Fig. 5A), which is consistent with the previous reports (22.Onate S.A. Tsai S.Y. Tsai M.J. O'Malley B.W. Science. 1995; 270: 1354-1357Crossref PubMed Scopus (2063) Google Scholar, 32.Shibata H. Spencer T.E. Onate S.A. Jenster G. Tsai S.Y. Tsai M.J. O'Malley B.W. Recent Prog. Horm. Res. 1997; 52: 141-165PubMed Google Scholar). However, β-catenin showed a strong interaction with AR but not with ERα and PRβ. VDR showed a weaker interaction with β-catenin in comparison to SRC1. These results indicate that β-catenin selectively interacts with AR. The specificity of interaction between β-catenin and AR proteins was further tested in CV1 cells. Since AR, GR, and PRβ all can activate the MMTV promoter, we examined whether β-catenin is able to enhance GR and PRβ activity under identical experimental condition. Transfection experiments were repeated with β-catenin and AR, GR, and PRβ expression plasmids, along with a luciferase reporter plasmid regulated by an MMTV promoter containing the steroid hormone-responsive elements (33.Hoeck W. Hofer P. Groner B. J. Steroid Biochem. Mol. Biol. 1992; 41: 283-289Crossref PubMed Scopus (23) Google Scholar, 34.Mink S. Ponta H. Cato A.C. Nucleic Acids Res. 1990; 18: 2017-2024Crossref PubMed Scopus (39) Google Scholar). As shown in Fig. 5B, all receptors showed a ligand-dependent transactivation with the MMTV promoter. However, β-catenin specifically augmented only AR-mediated transcription but not GR and PRβ (Fig. 5B). Taken together, our results suggest that β-catenin differs from SRC1 and selectively affects AR. Transient transfection assays were performed to further investigate the possible effect of β-catenin on AR-ediated transcription. Plasmids capable of expressing AR, wild type or mutants of β-catenin, and a luciferase reporter plasmid regulated by the MMTV-LTR (MMTVpA3-Luc), were transfected into CV-1 cells (35.Rovera G. Mehta S. Maul G. Exp. Cell Res. 1974; 89: 295-305Crossref PubMed Scopus (26) Google Scholar). A nearly 3-fold ligand-dependent transactivation was observed in the cells transfected with AR plasmid
Referência(s)