Voltar
Artigo Acesso aberto Revisado por pares
BIBTEX RIS

ARM Domain-dependent Nuclear Import of Adenomatous Polyposis Coli Protein Is Stimulated by the B56α Subunit of Protein Phosphatase 2A

2001; Elsevier BV; Volume: 276; Issue: 49 Linguagem: Inglês

10.1074/jbc.m107149200

ISSN

1083-351X

Autores

Melanie Galea, Alexandra Eleftheriou, Beric R. Henderson,

Tópico(s)

Genetic factors in colorectal cancer

Resumo

Inactivating mutations in the adenomatous polyposis coli (APC) gene correlate with progression of colon cancer and familial adenomatous polyposis. The APC tumor suppressor contributes to chromosome segregation and turnover of the oncogenic transcriptional activator β-catenin, and these activities are impaired by truncating cancer mutations. APC was recently identified as a shuttling protein whose subcellular distribution is regulated by two nuclear localization signals (NLSs) and multiple nuclear export signals (NESs). Here, we show that mutant disease-linked truncated forms of APC, most of which lack the two central NLSs and certain NES sequences, retain nuclear-cytoplasmic shuttling activity. Nuclear export of truncated APC is mediated by a dominant N-terminal NES. Nuclear import of NLS-deficient APC mutants is facilitated by the N-terminal ARM domain. Furthermore, co-expression of the ARM-binding protein, B56α, increased the nuclear localization of mutant and wild-type APC. The minimal B56α-responsive sequence mapped to APC amino acids 302–625. B56α is a regulatory subunit of protein phosphatase 2A; however, its ability to shift APC to the nucleus was independent of phosphatase activity. We conclude that APC nuclear import is regulated by the ARM domain through its interaction with B56α and postulate that APC/B56α complexes target the dephosphorylation of specific proteins within the nucleus. Inactivating mutations in the adenomatous polyposis coli (APC) gene correlate with progression of colon cancer and familial adenomatous polyposis. The APC tumor suppressor contributes to chromosome segregation and turnover of the oncogenic transcriptional activator β-catenin, and these activities are impaired by truncating cancer mutations. APC was recently identified as a shuttling protein whose subcellular distribution is regulated by two nuclear localization signals (NLSs) and multiple nuclear export signals (NESs). Here, we show that mutant disease-linked truncated forms of APC, most of which lack the two central NLSs and certain NES sequences, retain nuclear-cytoplasmic shuttling activity. Nuclear export of truncated APC is mediated by a dominant N-terminal NES. Nuclear import of NLS-deficient APC mutants is facilitated by the N-terminal ARM domain. Furthermore, co-expression of the ARM-binding protein, B56α, increased the nuclear localization of mutant and wild-type APC. The minimal B56α-responsive sequence mapped to APC amino acids 302–625. B56α is a regulatory subunit of protein phosphatase 2A; however, its ability to shift APC to the nucleus was independent of phosphatase activity. We conclude that APC nuclear import is regulated by the ARM domain through its interaction with B56α and postulate that APC/B56α complexes target the dephosphorylation of specific proteins within the nucleus. adenomatous polyposis coli protein chromosome region maintenance 1 green fluorescent protein glycogen synthase-3β hemagglutinin leptomycin B nuclear export sequence nuclear localization sequence polymerase chain reaction protein phosphatase 2A yellow fluorescent protein familial adenomatous polyposis Mutational inactivation of theAPC 1 gene is a key event in the development of colon cancer and the intestinal polyp disorder, familial adenomatous polyposis (FAP) (1Kinzler K.W. Vogelstein B. Cell. 1996; 87: 159-170Abstract Full Text Full Text PDF PubMed Scopus (4264) Google Scholar, 2Polakis P. Genes Dev. 2000; 14: 1837-1851Crossref PubMed Google Scholar, 3van Es J.H. Giles R.H. Clevers H.C. Exp. Cell Res. 2001; 264: 126-134Crossref PubMed Scopus (114) Google Scholar). In colon cancers, the APC gene is frequently targeted by germ-line and somatic mutations that delete the axin-binding motifs required to promote degradation of β-catenin (1Kinzler K.W. Vogelstein B. Cell. 1996; 87: 159-170Abstract Full Text Full Text PDF PubMed Scopus (4264) Google Scholar, 2Polakis P. Genes Dev. 2000; 14: 1837-1851Crossref PubMed Google Scholar). Thus, β-catenin accumulates in APC-mutated tumor cells and forms active transcription complexes with T cell factor/lymphoid enhancer factor (TCF/LEF-1) (2Polakis P. Genes Dev. 2000; 14: 1837-1851Crossref PubMed Google Scholar, 4Morin P.J. BioEssays. 1999; 21: 1021-1030Crossref PubMed Scopus (813) Google Scholar), thereby activating genes involved in cancer progression (5He T.C. Sparks A.B. Rago C. Hermeking H. Zawel L. da Costa L.T. Morin P.J. Vogelstein B. Kinzler K.W. Science. 1998; 281: 1509-1512Crossref PubMed Scopus (4059) Google Scholar, 6Tetsu O. McCormick F. Nature. 1999; 398: 422-426Crossref PubMed Scopus (3245) Google Scholar). Most gene mutations result in C-terminal truncations of the APC protein (1Kinzler K.W. Vogelstein B. Cell. 1996; 87: 159-170Abstract Full Text Full Text PDF PubMed Scopus (4264) Google Scholar, 2Polakis P. Genes Dev. 2000; 14: 1837-1851Crossref PubMed Google Scholar, 3van Es J.H. Giles R.H. Clevers H.C. Exp. Cell Res. 2001; 264: 126-134Crossref PubMed Scopus (114) Google Scholar). In addition to their effect on β-catenin turnover, APC C-terminal deletions prevent EB1-mediated association of APC with microtubules and regulation of chromosome segregation (1Kinzler K.W. Vogelstein B. Cell. 1996; 87: 159-170Abstract Full Text Full Text PDF PubMed Scopus (4264) Google Scholar, 7Fodde R. Kuipers J. Rosenberg C. Smits R. Kielman M. Gaspar C. van Es J.H. Breukel C. Wiegant J. Giles R.H. Clevers H. Nat. Cell Biol. 2001; 3: 433-438Crossref PubMed Scopus (586) Google Scholar, 8Kaplan K.B. Burds A.A. Swedlow J.R. Bekir S.S. Sorger P.K. Nathke I.S. Nat. Cell Biol. 2001; 3: 429-432Crossref PubMed Scopus (468) Google Scholar), binding of APC to the human homolog ofDrosophila discs large protein (DLG) and its inhibition of cell cycle regulation (9Matsumine A. Ogai A. Senda T. Okamura N. Satoh K. Baeg G.H. Kawahara T. Kobayashi S. Okada M. Toyoshima K. Akiyama T. Science. 1996; 272: 1020-1023Crossref PubMed Scopus (407) Google Scholar, 10Ishidate T. Matsumine A. Toyoshima K. Akiyama T. Oncogene. 2000; 19: 365-372Crossref PubMed Scopus (165) Google Scholar), and the ability of APC to bind DNA (11Deka J. Herter P. Sprenger-Haussels M. Koosh S. Franz D. Muller K-M. Kuhnen C. Hoffman I. Muller O. Oncogene. 1999; 18: 5654-5661Crossref PubMed Scopus (38) Google Scholar). Recently, we and others showed that APC is a nuclear-cytoplasmic shuttling protein (12Henderson B.R. Nat. Cell Biol. 2000; 2: 653-660Crossref PubMed Scopus (413) Google Scholar, 13Neufeld K.L. Nix D.A. Bogerd H. Kang Y. Beckerle M.C. Cullen B.R. White R.L. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 12085-12090Crossref PubMed Scopus (140) Google Scholar, 14Rosin-Arbesfeld R. Townsley F. Bienz M. Nature. 2000; 406: 1009-1012Crossref PubMed Scopus (307) Google Scholar), contributing to the nuclear export of β-catenin to stimulate its degradation in the cytoplasm (12Henderson B.R. Nat. Cell Biol. 2000; 2: 653-660Crossref PubMed Scopus (413) Google Scholar, 14Rosin-Arbesfeld R. Townsley F. Bienz M. Nature. 2000; 406: 1009-1012Crossref PubMed Scopus (307) Google Scholar, 15Neufeld K.L. Zhang F. Cullen B. White R.L. EMBO Rep. 2000; 1: 519-523Crossref PubMed Scopus (143) Google Scholar). In addition to two nuclear export signals (NESs) identified at the N terminus (12Henderson B.R. Nat. Cell Biol. 2000; 2: 653-660Crossref PubMed Scopus (413) Google Scholar, 13Neufeld K.L. Nix D.A. Bogerd H. Kang Y. Beckerle M.C. Cullen B.R. White R.L. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 12085-12090Crossref PubMed Scopus (140) Google Scholar), three different NESs were also detected in the central 20-amino acid repeats of APC (14Rosin-Arbesfeld R. Townsley F. Bienz M. Nature. 2000; 406: 1009-1012Crossref PubMed Scopus (307) Google Scholar). These latter export signals in addition to two nuclear localization signals (NLS) identified in the center of APC (16Zhang F. White R.L. Neufeld K.L. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 12577-12582Crossref PubMed Scopus (92) Google Scholar) are deleted by most cancer and FAP mutations, prompting the suggestion that APC nuclear transport is impaired in cancer cells (14Rosin-Arbesfeld R. Townsley F. Bienz M. Nature. 2000; 406: 1009-1012Crossref PubMed Scopus (307) Google Scholar). In this study, we show that this is not the case and demonstrate that cancer-mutated forms of APC retain nuclear-cytoplasmic shuttling activity. Using assays to compare nuclear transport activity, we show that different truncated APC mutants are still able to exit the nucleus efficiently due to the presence of a dominant N-terminal NES. Moreover, we identified an important role for the N-terminal ARM domain in the nuclear localization of APC, in particular those forms of APC that lack a nuclear localization signal. Of two known ARM-binding proteins, B56α (17Seeling J.M. Miller J.R. Gil R. Moon R.T. White R. Virshup D.M. Science. 1999; 283: 2089-2091Crossref PubMed Scopus (364) Google Scholar) and ASEF (18Kawasaki Y. Senda T. Ishidate T. Koyama R. Morishita T. Iwayama Y. Higuchi O. Akiyama T. Science. 2000; 289: 1194-1197Crossref PubMed Scopus (296) Google Scholar), only B56α, a subunit of protein phosphatase 2A (PP2A), strongly enhanced nuclear accumulation of APC in co-transfected cells. These findings reveal that APC nuclear transport can be modulated by certain APC-binding proteins and suggest a functional link between APC and targeted phosphatase activity in the nucleus. NIH 3T3 mouse fibroblasts, human HCT116, and SW480 colon carcinoma cells were cultured in Dulbecco's modified Eagle's medium with 10% fetal calf serum and were free of mycoplasma. Leptomycin B (LMB) (supplied by M. Yoshida, Tokyo) was added to a final concentration of 6 ng/ml. MG132 (Calbiochem) was used at a final concentration of 20 μm. Actinomycin D (Sigma) was used at a final concentration of 5 μg/ml. DNA transfection of cells (typically 1 μg of DNA per 2 ml of medium) was performed with FuGene reagent as directed (Roche), using cells at medium density seeded onto coverslips. Some transfections were performed with LipofectAMINE reagent (Life Technologies, Inc.). Several double-stranded oligonucleotide sequences were cloned between the BamHI andAge1 restriction sites of pRev(1.4)-GFP to test APC sequences for possible nuclear export activity as recently described in detail (19Henderson B.R. Eleftheriou A. Exp. Cell Res. 2000; 256: 213-224Crossref PubMed Scopus (350) Google Scholar). The oligonucleotides correspond to the NES protein sequences shown in Fig. 3. The expression vectors that encode full-length APC (pCMV-APC) and the C-terminal truncated APC mutants, pCMV-APC-(1–2644), pCMV-APC-(1–1941), and pCMV-APC-(1–1309), were described previously (20Smith K.J. Levy D.B. Maupin P. Pollard T.D. Vogelstein B. Kinzler K.W. Cancer Res. 1994; 54: 3672-3675PubMed Google Scholar, 21Morin P.J. Sparks A.B. Korinek V. Barker N. Clevers H. Vogelstein B. Kinzler K.W. Science. 1997; 275: 1787-1790Crossref PubMed Scopus (3488) Google Scholar). Additional APC truncation mutants (1–932, 1–625, and 1–302) were constructed in pCMV-APC by first removing the KpnI/XhoI fragment and replacing it with PCR-amplified fragments that introduce published FAP mutations (22Miyoshi Y. Ando H. Nagase H. Nishisho I. Horii A. Miki Y. Mori T. Utsunomiya J. Baba S. Petersen G. Hamilton S.R. Kinzler K.W. Vogelstein B. Nakamura Y. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 4452-4456Crossref PubMed Scopus (518) Google Scholar) and premature stop codons into the APC sequence. For PCR, a common forward primer was used (5′-TGGAATATGAAGCAA-3′), and theXhoI site-containing reverse primers were as follows: mutant (1), 5′-TCCGCTCGAGTAAGTGTTTAATGTGTATGG-3′; mutant (1), 5′-TCCGCTCGAGTGTTTGTCTGCTCCGGTAAG-3′; and mutant (1), 5′-TCCGCTCGAGTTGTCAGCCTTCAGGTGCAG-3′. The full-length APC cDNA containing site-directed mutations in NES1 or NES1+2 were described previously (12Henderson B.R. Nat. Cell Biol. 2000; 2: 653-660Crossref PubMed Scopus (413) Google Scholar), and the same NES1 mutation (L75A/L77A) was inserted into the APC deletion constructs APC-(1–302) and APC-(1–625) by replacing the SalI/KpnI fragment (N terminus of APC including start site and NESs). A series of FLAG-tagged APC vectors were constructed in which the N-terminal amino acids 1–334 were removed. pCMV-FLAG-APC-(334–2843) was made by first removing the N-terminalBamHI/NheI fragment from pCMV-APC and then inserting a BamHI/NheI-digested DNA fragment containing a new start site and the FLAG epitope in-frame with the adjoining APC sequence (starting at codon 334). C-terminal deletions were then introduced into the pCMV-FLAG-APC-(334–2843) vector by transposing the NheI/XhoI fragments from pCMV-APC-(1–1309), APC-(1–932), and APC-(1–625) described above. All constructs were confirmed by DNA sequencing. The assay used to assess nuclear export activity was described recently in detail (19Henderson B.R. Eleftheriou A. Exp. Cell Res. 2000; 256: 213-224Crossref PubMed Scopus (350) Google Scholar). The assay is based on scoring the cellular distribution of GFP fusion proteins expressed following transfection into cells. The reference vector (pRev(1.4)-GFP) contains an export-deficient form of Rev-GFP, which retains a nuclear localization signal and therefore accumulates in the nucleus/nucleolus of cells. The ability to block nuclear import (with actinomycin D) or export (with leptomycin B) allows for accurate comparison of the export activity of test NESs inserted into the control vector. SW480 cells were separated into nuclear, cytoplasmic, and membrane fractions using the NE-PER kit (Pierce) as directed. Equivalent proportions of each cell fraction (60 μg of cytoplasm, 20 μg of nuclear extract) were separated on an 8% SDS-polyacrylamide gel and transferred to an immobilon-P membrane (Millipore). The filter was probed with APC monoclonal antibodies Ab1 and Ab7 (Oncogene Research), β-catenin antibody mAb C19220 (Transduction Laboratory), and a topoisomerase II antibody Ab1 (Santa Cruz Biotechnologies) as a fractionation control. APC was detected at the expected size (∼150-kDa band) as confirmed by size markers. Filters were developed with an ECL chemiluminescence reagent (Amersham Pharmacia Biotech) exposed to x-ray film and quantitated by densitometry. Cells were fixed and stained with various primary antibodies as described (12Henderson B.R. Nat. Cell Biol. 2000; 2: 653-660Crossref PubMed Scopus (413) Google Scholar). Samples were then incubated with a secondary antibody (1:120 dilution of a fluorescein isothiocyanate- or Texas Red-conjugated anti-rabbit or anti-mouse antibody from Sigma) prior to mounting on slides with Vectashield (Vector Laboratories, CA), and fluorescence microscopy. Cells were photographed with an Olympus fluorescence microscope at × 400 magnification, or digital images were captured using an OptiScan confocal microscope at × 600 magnification. For quantitation of nuclear and cytoplasmic fluorescence intensities, more than 100 confocal cell images were analyzed using the NIH Image software as previously described (12Henderson B.R. Nat. Cell Biol. 2000; 2: 653-660Crossref PubMed Scopus (413) Google Scholar). The shuttling of APC between nucleus and cytoplasm is likely to result from the combined action of multiple nuclear import and export signals (Fig.1 A). We therefore compared the subcellular distribution of different APC truncation peptides that result from frequent germline mutations associated with FAP (22Miyoshi Y. Ando H. Nagase H. Nishisho I. Horii A. Miki Y. Mori T. Utsunomiya J. Baba S. Petersen G. Hamilton S.R. Kinzler K.W. Vogelstein B. Nakamura Y. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 4452-4456Crossref PubMed Scopus (518) Google Scholar, 23Nagase H. Miyoshi Y. Horii A. Aoki T. Ogawa M. Utsunomiya J. Baba S. Sasazuki T. Nakamura Y. Cancer Res. 1992; 52: 4055-4057PubMed Google Scholar) or from somatic mutations observed in colon tumors (24Powell S.M. Zilz N. Beazer-Barclay Y. Bryan T.M. Mailton S.R. Thibodeau S.N. Vogelstein B. Kinzler K.W. Nature. 1992; 359: 235-237Crossref PubMed Scopus (1666) Google Scholar, 25Miyoshi Y. Nagase H. Ando H. Ichii S. Nakatsura S. Aoki T. Miki Y. Mori T. Nakamura Y. Hum. Mol. Gen. 1992; 1: 229-233Crossref PubMed Scopus (867) Google Scholar) (see Fig.1 A). APC constructs were transfected into HCT116 (APCwt/wt) and SW480 (APCmut/mut) colon tumor cells, and APC localization was analyzed by immunostaining and fluorescence microscopy. All forms of APC showed some nuclear staining in 20–50% of transfected cells. Interestingly, many of the deletion mutants were more nuclear in SW480 cells than in HCT116 cells, and it is not yet clear if this may relate to differences in the endogenous form of APC expressed by these cell lines. In SW480 cells, the APC-(1–1309) mutant displayed the strongest cytoplasmic staining, at a level comparable with that seen previously (12Henderson B.R. Nat. Cell Biol. 2000; 2: 653-660Crossref PubMed Scopus (413) Google Scholar). We recently reported an increase in the nuclear staining of wild-type APC, and of the mutant APC (1–1309), following treatment with the CRM1-specific export inhibitor, leptomycin B (12Henderson B.R. Nat. Cell Biol. 2000; 2: 653-660Crossref PubMed Scopus (413) Google Scholar). Unexpectedly, we found that all C-terminal truncated forms of APC displayed a similar LMB-responsive shift toward the nucleus in transfected cells (nuclear in 70–80% of LMB-treated cells) (Fig. 1 A). This suggests that disease-associated forms of APC retain the ability to enter and exit the nucleus. In particular, the APC-(1–1941) and APC-(1–1309) mutants displayed an equivalent shift to the nucleus after 4 h of LMB treatment. Unlike the APC-(1–1941) peptide, mutant APC-(1–1309) does not contain the central 20-amino acid repeat NESs (R3, R4, and R7) reported by Rosin-Arbesfeld et al. (14Rosin-Arbesfeld R. Townsley F. Bienz M. Nature. 2000; 406: 1009-1012Crossref PubMed Scopus (307) Google Scholar), suggesting that deletion of this central region does not impair nuclear export activity. An LMB response was observed even for the shortest N-terminal fragment, APC-(1–302) (Fig. 1, A and B), although the strong nuclear staining of this fragment may reflect passive diffusion through the nuclear pores. In concordance with the above data, a full-length APC NES1+2 mutant (12Henderson B.R. Nat. Cell Biol. 2000; 2: 653-660Crossref PubMed Scopus (413) Google Scholar) was more nuclear than wild-type APC in transfected cells, but nuclear staining of the APC NES1+2 mutant was not significantly enhanced by LMB treatment (Fig. 1, A andB). In both colon tumor cell lines tested, full-length APC shifted less well to the nucleus than did APC mutants when export was blocked (Fig. 1), and this may be due in part to cytoplasmic retention caused by C-terminal microtubule binding sequences (20Smith K.J. Levy D.B. Maupin P. Pollard T.D. Vogelstein B. Kinzler K.W. Cancer Res. 1994; 54: 3672-3675PubMed Google Scholar). We conclude that the N-terminal nuclear export sequences, common to all of the mutants tested, are primarily responsible for CRM1-dependent nuclear export of APC. To exclude the possibility that the transient expression assay data may not reflect the physiological trafficking of cellular APC, we examined the localization and shuttling of endogenous truncated APC-(1–1337) in SW480 cells. Immunofluorescence staining of cellular APC in these cells was inconclusive, largely because of the poor expression of the APC mutant protein. We therefore prepared fractionated extracts from SW480 cells and compared them by Western blot. As shown in Fig. 2, APC appeared as the expected ∼150-kDa band predominantly in the cytoplasm. This result concurs with that previously described by Smith et al. (26Smith K.J. Johnson K.A. Bryan T.M. Hill D.E. Markowitz S. Willson J.K.V. Paraskeva C. Petersen G.M. Hamilton S.R. Vogelstein B. Kinzler K.W. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 2846-2850Crossref PubMed Scopus (433) Google Scholar), and quantitative densitometry of the bands revealed an 18-fold higher level of APC in cytoplasm compared with nucleus. The truncated APC shifted to the nucleus after a 4-h LMB treatment, redistributing to roughly equivalent levels between nucleus and cytoplasm (cytoplasm/nucleus ratio = 1.3) (Fig. 2). Similar results were obtained with antibodies Ab1 and Ab7 (Oncogene Research), and they correlated extremely well with the localization of the APC-(1–1309) mutant construct in transfected SW480 cells (Fig.1 A). These findings contradict the claim of Rosin-Arbesfeldet al. (14Rosin-Arbesfeld R. Townsley F. Bienz M. Nature. 2000; 406: 1009-1012Crossref PubMed Scopus (307) Google Scholar) that APC is predominantly nuclear in SW480 cells and instead demonstrate that the endogenous mutated form of APC is an active nuclear shuttling protein. The above findings question the relative contribution of the APC central repeat NESs. We therefore utilized a sensitive in vivo transport assay (19Henderson B.R. Eleftheriou A. Exp. Cell Res. 2000; 256: 213-224Crossref PubMed Scopus (350) Google Scholar) to compare the individual export activities of the N-terminal NES1 and NES2 sequences (12Henderson B.R. Nat. Cell Biol. 2000; 2: 653-660Crossref PubMed Scopus (413) Google Scholar, 13Neufeld K.L. Nix D.A. Bogerd H. Kang Y. Beckerle M.C. Cullen B.R. White R.L. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 12085-12090Crossref PubMed Scopus (140) Google Scholar) and the R3, R4, and R7 NESs (14Rosin-Arbesfeld R. Townsley F. Bienz M. Nature. 2000; 406: 1009-1012Crossref PubMed Scopus (307) Google Scholar). As shown in Fig. 3, the two N-terminal NESs exhibited a similar level of activity (NES1, a maximal activity of 9+; NES2, a modest activity of 2+) in transfected SW480 colon cancer cells as was previously observed in other cell lines (12Henderson B.R. Nat. Cell Biol. 2000; 2: 653-660Crossref PubMed Scopus (413) Google Scholar). While the three APC 20-amino acid repeat sequences (R3, R4, and R7) also displayed some export activity (Fig. 3), even the strongest of these sequences (R3 motif) was weak by comparison to NES1 (see Fig. 3; and Table I for distribution profiles of each construct). Similar results were observed in NM39 melanoma cells (data not shown). The export activities observed were negated (NES1; 91% nuclear) or reduced (R3; 66% nuclear) by a 3-h leptomycin B treatment, consistent with export by the CRM1-dependent transport pathway (27Kudo N. Matsumori N. Taoka H. Fujiwara D. Schreiner E. Wolff B. Yoshida M. Horinouchi S. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 9112-9117Crossref PubMed Scopus (847) Google Scholar).Table IComparing the relative activity of different APC export sequencesNo drug3 h Act DNNCCnNNCCn%%%%%%Rev(1.4)-GFP86.313.30(792)85.714.30(787)Rev NES05446(666)04159(549)APC NES101882(649)01783(614)APC NES263370(528)22780(576)R318766(619)48313(601)R477230(478)51490(421)R766340(363)52480(403)3 h Act D + LMBNNCCn%%%APC NES19190(222)R366340(82)The data represent the proportion of transfected SW480 cells displaying nuclear (N), nuclear and cytoplasmic (NC), or only cytoplasmic (C) staining of the various GFP fusion proteins, in the presence or absence of actinomycin D (blocks nuclear import in this assay; see Ref. 19Henderson B.R. Eleftheriou A. Exp. Cell Res. 2000; 256: 213-224Crossref PubMed Scopus (350) Google Scholar). Values shown are the mean (<10% variation) of at least two independent experiments. n, number of transfected cells scored. Open table in a new tab The data represent the proportion of transfected SW480 cells displaying nuclear (N), nuclear and cytoplasmic (NC), or only cytoplasmic (C) staining of the various GFP fusion proteins, in the presence or absence of actinomycin D (blocks nuclear import in this assay; see Ref. 19Henderson B.R. Eleftheriou A. Exp. Cell Res. 2000; 256: 213-224Crossref PubMed Scopus (350) Google Scholar). Values shown are the mean ( 40% the proportion of cells with nuclear APC (Fig. 5). B56α also stimulated nuclear relocalization of the mutant APC-(1–1309) in transfected NIH 3T3 cells (Fig. 5 C) and SW480 cells (data not shown), revealing that regulation by this protein is independent of the two central NLSs in APC. Furthermore, when co-expressed with different APC fragments, B56α only enhanced the nuclear localization of sequences that contain the ARM domain (Fig. 6). Interestingly, the smallest B56-responsive sequence mapped here (amino acids 334–625) comprises just the first half of the ARM repeat region, equivalent to the first four armadillo repeats (28Peifer M. Berg S. Reynolds A.B. Cell. 1994; 76: 789-791Abstract Full Text PDF PubMed Scopus (546) Google Scholar) (Fig. 6). This finding significantly refines the APC sequence to which B56α can bind. Several of the B56α-regulated APC sequences that we tested contain no CRM1-dependent nuclear export signals, indicating that B56α does not function simply by blocking APC nuclear export.Figure 6Mapping the minimal B56α-responsive sequence in APC.A, different APC sequences were tested for B56α-dependent nuclear localization. APC constructs were transfected alone or with pHA-B56α into NIH 3T3 cells and scored for nuclear staining by fluorescence microscopy. As indicated, the minimal B56-responsive sequence comprised amino acids 334–625, which contained the first four ARM repeats. B, the proportion of cells showing at least partial nuclear APC staining is graphed as the means ± S.E. from two experiments. The relative increase in nuclear staining is summarized in the top right panel: +, 10–15% increase; ++, 15–30% increase; +++, 30–50% increase.View Large Image Figure ViewerDownload Hi-res image Download (PPT) The APC-B56α interaction was previously implicated in down-regulation of β-catenin (17Seeling J.M. Miller J.R. Gil R. Moon R.T. White R. Virshup D.M. Science. 1999; 283: 2089-2091Crossref PubMed Scopus (364) Google Scholar), by a mechanism thought to involve the B56α-dependent dephosphorylation and activation of the kinase GSK-3β (17Seeling J.M. Miller J.R. Gil R. Moon R.T. White R. Virshup D.M. Science. 1999; 283: 2089-2091Crossref PubMed Scopus (364) Google Scholar). We therefore asked whether B56α-dependent modulation of APC nuclear localization required its phosphatase activity and/or was an indirect consequence of its activation of GSK-3β. When overexpressed with APC, ectopic GSK-3β induced only a modest increase in nuclear localization (TableIII), comparable with that elicited by ASEF (Fig. 5). Furthermore, the effect of B56α on nuclear localization of APC (1–1309) was not at all perturbed (and in fact increased slightly) by an 18-h treatment with the protein phosphatase 2A inhibitor, okadaic acid (Table III). The effectiveness of the drug was confirmed by its ability to induce p21 expression as previously reported (Ref. 32Yan Y. Mumby M.C. J. Biol. Chem. 1999; 274: 31917-31924Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar and data not shown). We conclude that the influence of B56α on APC nuclear transport is independent of PP2A phosphatase function.Table IIIEffect of okadaic acid, a PP2A inhibitor, on B56α-mediated APC nuclear importConstructSubcellular distribution (% cells)NNCCn%%%APC-(1–1309)022 ± 278 ± 2(347)APC-(1–1309) + B56α063 ± 337 ± 4(332)APC-(1–1309) + B56α/okadaic acid069 ± 231 ± 2(162)APC-(1–1309) + GSK-3β032 ± 268 ± 2(297)pCMV-APC-(1–1309) was transfected into NIH 3T3 cells, alone or with pHA-B56α or pHA-GSK-3β. After 48 h, cells were immunostained for APC and the HA tag of each co-factor (as in Fig. 5), and the localization pattern of APC was determined by fluorescence microscopy. Okadaic acid (Sigma) was made up fresh and added to cells 18 h prior to harvest, at a final concentration of 20 nm as previously reported (17Seeling J.M. Miller J.R. Gil R. Moon R.T. White R. Virshup D.M. Science. 1999; 283: 2089-2091Crossref PubMed Scopus (364) Google Scholar). Okadaic acid is a potent inhibitor of protein phosphatase 2A, but it had no effect on B56-mediated nuclear localization of APC in 3T3 cells (above) or SW480 cells (not shown). Co-transfection of the kinase GSK-3β induced a modest increase in APC nuclear staining. Values shown are mean ± standard error from two experiments. n, total number of transfected cells scored. Open table in a new tab pCMV-APC-(1–1309) was transfected into NIH 3T3 cells, alone or with pHA-B56α or pHA-GSK-3β. After 48 h, cells were immunostained for APC and the HA tag of each co-factor (as in Fig. 5), and the localization pattern of APC was determined by fluorescence microscopy. Okadaic acid (Sigma) was made up fresh and added to cells 18 h prior to harvest, at a final concentration of 20 nm as previously reported (17Seeling J.M. Miller J.R. Gil R. Moon R.T. White R. Virshup D.M. Science. 1999; 283: 2089-2091Crossref PubMed Scopus (364) Google Scholar). Okadaic acid is a potent inhibitor of protein phosphatase 2A, but it had no effect on B56-mediated nuclear localization of APC in 3T3 cells (above) or SW480 cells (not shown). Co-transfection of the kinase GSK-3β induced a modest increase in APC nuclear staining. Values shown are mean ± standard error from two experiments. n, total number of transfected cells scored. The APC tumor suppressor interacts with several proteins of diverse function; however, the influence of APC binding partners on its subcellular localization has not been investigated. Recently, we and others showed that APC can shuttle in and out of the nucleus, and multiple transport signals were identified that contribute to APC nuclear import (16Zhang F. White R.L. Neufeld K.L. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 12577-12582Crossref PubMed Scopus (92) Google Scholar) and export (12Henderson B.R. Nat. Cell Biol. 2000; 2: 653-660Crossref PubMed Scopus (413) Google Scholar, 13Neufeld K.L. Nix D.A. Bogerd H. Kang Y. Beckerle M.C. Cullen B.R. White R.L. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 12085-12090Crossref PubMed Scopus (140) Google Scholar, 14Rosin-Arbesfeld R. Townsley F. Bienz M. Nature. 2000; 406: 1009-1012Crossref PubMed Scopus (307) Google Scholar). Several of these are deleted in colon cancer, prompting earlier speculation that APC nuclear transport is impaired in cancer cells (14Rosin-Arbesfeld R. Townsley F. Bienz M. Nature. 2000; 406: 1009-1012Crossref PubMed Scopus (307) Google Scholar). Here, we show in fact that endogenous and transiently expressed APC cancer mutants retain the ability to shuttle between nucleus and cytoplasm. We report the first quantitative comparison of multiple nuclear export signals from the same protein, revealing that the strongest APC export signal (NES1) is located near the N terminus, thereby explaining why different APC truncation mutants can exit the nucleus. More important, we found that the N-terminal ARM repeat domain facilitates APC nuclear import and that ARM-mediated transport was stimulated by association with B56α, a subunit of PP2A. These findings provide important clarification of the functional role of different APC transport signals and identify the ARM domain as a new transport element in APC. Our findings do not support the recent proposal that APC nuclear export is abolished by C-terminal truncating mutations in the mutation cluster region (MCR) (14Rosin-Arbesfeld R. Townsley F. Bienz M. Nature. 2000; 406: 1009-1012Crossref PubMed Scopus (307) Google Scholar). We compared the same NESs tested by Rosin-Arbesfeldet al. (14Rosin-Arbesfeld R. Townsley F. Bienz M. Nature. 2000; 406: 1009-1012Crossref PubMed Scopus (307) Google Scholar) and found that they exhibit only a modest activity relative to the N-terminal NES1 (Fig. 3), and their deletion did not impair APC nuclear export activity. Importantly, our cell fractionation and Western blot data show that endogenous mutant APC-(1–1337) is mostly cytoplasmic in SW480 cells but shifts to the nucleus when nuclear export is blocked. The concordance between transfection data and experiments with the endogenous cellular APC demonstrate that disease-associated APC mutants retain nuclear shuttling activity, although it is important to note that the functional consequences of their shuttling remain to be determined. The ability of the human APC armadillo repeat region (known as the ARM domain and spanning amino acids 453–767) to mediate nuclear localization is consistent with the previous observation that theDrosophila e-APC ARM sequence displayed nuclear staining when fused to the green fluorescent protein (14Rosin-Arbesfeld R. Townsley F. Bienz M. Nature. 2000; 406: 1009-1012Crossref PubMed Scopus (307) Google Scholar). By testing for a possible “piggy-back” mechanism of import, we identified the ARM-binding protein B56α as a nuclear import chaperone of APC. Our data do not exclude the possibility, however, that the seven ARM repeats of APC can independently promote nuclear import by directly associating with the nuclear pore complex, as was previously suggested for the much larger ARM repeat domain of β-catenin (29Funayama N. Fagotto F. McCrea P. Gumbiner B.M. J. Cell Biol. 1995; 128: 959-968Crossref PubMed Scopus (498) Google Scholar, 33Fagotto F. Gluck U. Gumbiner B.M. Curr. Biol. 1997; 8: 181-190Abstract Full Text Full Text PDF Google Scholar). B56α is a regulatory subunit of PP2A (34McCright B. Rivers A.M. Audlin S. Virshup D.M. J. Biol. Chem. 1996; 271: 22081-22089Abstract Full Text Full Text PDF PubMed Scopus (326) Google Scholar) and was previously claimed to target phosphatase activity to the cytoplasm (34McCright B. Rivers A.M. Audlin S. Virshup D.M. J. Biol. Chem. 1996; 271: 22081-22089Abstract Full Text Full Text PDF PubMed Scopus (326) Google Scholar). In contrast, we found that ectopic B56α localized to the nucleus in ∼85% of transfected 3T3 cells (Fig. 5 and data not shown). Unlike its effect on β-catenin stability (17Seeling J.M. Miller J.R. Gil R. Moon R.T. White R. Virshup D.M. Science. 1999; 283: 2089-2091Crossref PubMed Scopus (364) Google Scholar), B56α-mediated nuclear localization of APC was unaffected by the PP2A inhibitor, okadaic acid, and thus is unlikely to result from the dephosphorylation of APC or associated proteins. Also, while it is possible that B56α can reduce APC nuclear export, this does not account for its chaperone effect on NES-deficient APC fragments. We therefore propose that B56α functions as an APC chaperone to enhance the nuclear import or retention of APC. We mapped the B56-responsive sequence to the first four ARM repeats of APC, and showed that a sequence lacking ARM repeats was not regulated by B56α (Fig. 6). Furthermore, whereas co-transfection of APC and B56α strongly shifted the distribution of APC toward the nucleus, the B56α-staining pattern was unaffected (data not shown). These observations, when considered together, are consistent with B56α binding to the APC ARM domain and carrying APC into the nucleus. Specific nuclear functions of APC have not yet been identified. We speculate that APC/B56α complexes direct the PP2A heterotrimer to dephosphorylate certain proteins in the nucleus. PP2A regulates a diverse range of cellular processes including cell cycle progression, chromosome segregation, transcription factor activity, apoptosis, and nucleotide excision repair of DNA (35Janssens V. Goris J. Biochem. J. 2001; 353: 417-439Crossref PubMed Scopus (1521) Google Scholar, 36Ariza R.R. Keyse S.M. Moggs J.G. Wood R.D. Nucleic Acids Res. 1996; 24: 433-440Crossref PubMed Google Scholar). APC has known involvement in some of these processes, due to its C-terminal interaction with DNA (11Deka J. Herter P. Sprenger-Haussels M. Koosh S. Franz D. Muller K-M. Kuhnen C. Hoffman I. Muller O. Oncogene. 1999; 18: 5654-5661Crossref PubMed Scopus (38) Google Scholar), and with the proteins DLG (9Matsumine A. Ogai A. Senda T. Okamura N. Satoh K. Baeg G.H. Kawahara T. Kobayashi S. Okada M. Toyoshima K. Akiyama T. Science. 1996; 272: 1020-1023Crossref PubMed Scopus (407) Google Scholar) and EB1 (37Su L-K. Burrell M. Hill D.E. Gyuris J. Brent R. Wiltshire R. Trent J. Vogelstein B. Kinzler K.W. Cancer Res. 1995; 55: 2972-2977PubMed Google Scholar), which mediate its roles in cell growth inhibition (10Ishidate T. Matsumine A. Toyoshima K. Akiyama T. Oncogene. 2000; 19: 365-372Crossref PubMed Scopus (165) Google Scholar) and chromosome segregation (7Fodde R. Kuipers J. Rosenberg C. Smits R. Kielman M. Gaspar C. van Es J.H. Breukel C. Wiegant J. Giles R.H. Clevers H. Nat. Cell Biol. 2001; 3: 433-438Crossref PubMed Scopus (586) Google Scholar, 8Kaplan K.B. Burds A.A. Swedlow J.R. Bekir S.S. Sorger P.K. Nathke I.S. Nat. Cell Biol. 2001; 3: 429-432Crossref PubMed Scopus (468) Google Scholar), respectively. These interactions are lost in colon cancer due to truncation of the APC C terminus (1Kinzler K.W. Vogelstein B. Cell. 1996; 87: 159-170Abstract Full Text Full Text PDF PubMed Scopus (4264) Google Scholar). In future experiments we plan to define the mechanism of ARM-mediated APC nuclear translocation in more detail and to test the prediction that APC directs B56/PP2A-dependent dephosphorylation of nuclear proteins. We thank Drs. Bert Vogelstein and Kenneth Kinzler for pCMV-APC, Tetsu Akiyama for pHA-ASEF, Jim Woodgett for pHA-GSK3β and David Virshup for pHA-B56α plasmids, and Dr. M. Yoshida for leptomycin B.

Referência(s)