Voltar
Artigo Acesso aberto Revisado por pares
BIBTEX RIS

The Protein Kinase CK2 Site (Ser111/112) Enhances Recognition of the Simian Virus 40 Large T-antigen Nuclear Localization Sequence by Importin

1997; Elsevier BV; Volume: 272; Issue: 27 Linguagem: Inglês

10.1074/jbc.272.27.17191

ISSN

1083-351X

Autores

Stefan Hübner, Chong-Yun Xiao, David A. Jans,

Tópico(s)

HIV/AIDS drug development and treatment

Resumo

The mechanism by which phosphorylation regulates nuclear localization sequence (NLS)-dependent nuclear protein import is largely unclear. Whereas nuclear accumulation of SV40 large tumor antigen (T-ag) fusion proteins is completely dependent on the T-ag NLS (amino acids 126–132), the rate of nuclear import is increased 50-fold by amino acid residues 111–125 and in particular a site for the protein kinase CK2 (CK2) at serine 111/112. Because the first step of nuclear protein import involves the binding of the NLS by an NLS-receptor complex such as the importin 58/97 heterodimer, we established a novel enzyme-linked immunosorbent assay to test whether NLS recognition is influenced by amino acids amino-terminal to the NLS and the CK2 site. We found that recognition of the T-ag NLS by importin 58/97 was enhanced 10-fold in the presence of amino acid residues 111–125 and strongly dependent on importin 97. A T-ag fusion protein in which the spacer between the CK2 site and the NLS was decreased showed 30% reduced binding by importin 58/97. Maximal nuclear accumulation of this protein was reduced by more than 50%, indicating the physiological importance of the correctly positioned CK2 site. Phosphorylation by CK2 increased the T-ag NLS binding affinity for importin 58/97 by a further 40%. We conclude that flanking sequences and in particular phosphorylation at the CK2 site are mechanistically important in NLS recognition and represent the basis of their enhancement of T-ag nuclear import. This study thus represents the first elucidation of the mechanistic basis of the regulation of nuclear protein import through phosphorylation within a phosphorylation-regulated NLS. The mechanism by which phosphorylation regulates nuclear localization sequence (NLS)-dependent nuclear protein import is largely unclear. Whereas nuclear accumulation of SV40 large tumor antigen (T-ag) fusion proteins is completely dependent on the T-ag NLS (amino acids 126–132), the rate of nuclear import is increased 50-fold by amino acid residues 111–125 and in particular a site for the protein kinase CK2 (CK2) at serine 111/112. Because the first step of nuclear protein import involves the binding of the NLS by an NLS-receptor complex such as the importin 58/97 heterodimer, we established a novel enzyme-linked immunosorbent assay to test whether NLS recognition is influenced by amino acids amino-terminal to the NLS and the CK2 site. We found that recognition of the T-ag NLS by importin 58/97 was enhanced 10-fold in the presence of amino acid residues 111–125 and strongly dependent on importin 97. A T-ag fusion protein in which the spacer between the CK2 site and the NLS was decreased showed 30% reduced binding by importin 58/97. Maximal nuclear accumulation of this protein was reduced by more than 50%, indicating the physiological importance of the correctly positioned CK2 site. Phosphorylation by CK2 increased the T-ag NLS binding affinity for importin 58/97 by a further 40%. We conclude that flanking sequences and in particular phosphorylation at the CK2 site are mechanistically important in NLS recognition and represent the basis of their enhancement of T-ag nuclear import. This study thus represents the first elucidation of the mechanistic basis of the regulation of nuclear protein import through phosphorylation within a phosphorylation-regulated NLS. Nuclear protein transport is dependent on specific targeting signals called nuclear localization sequences (NLSs), 1The abbreviations used are: NLS, nuclear localization sequence; T-ag, SV40 large tumor antigen; prNLS, phosphorylation-regulated NLS; NPC, nuclear pore complex; GST, glutathione S-transferase; CLSM, confocal laser scanning microscopy; ELISA, enzyme-linked immunosorbent assay; β-Gal, β-galactosidase; BSA, bovine serum albumin. defined as the sequences sufficient and necessary for nuclear targeting (1Kalderon D. Richardson W.D. Markham A.F. Smith A.E. Nature. 1984; 311: 33-38Crossref PubMed Scopus (908) Google Scholar, 2Kalderon D. Roberts B.L. Richardson W.D. Smith A.E. Cell. 1984; 3: 499-509Abstract Full Text PDF Scopus (1867) Google Scholar). They are typically short sequences of a single series of basic residues (resembling the NLS of the SV40 large tumor antigen (T-ag)) or of two clusters of basic residues interrupted by a 10–12-amino acid spacer (bipartite NLSs), and they appear to be functional in various protein contexts, suggesting that NLS function is largely independent of secondary/tertiary structure (see Refs. 3Roberts B.L. Richardson W.D. Smith A.E. Cell. 1987; 50: 465-475Abstract Full Text PDF PubMed Scopus (138) Google Scholar, 4Jans D.A. Hübner S. Physiol. Rev. 1996; 76: 651-685Crossref PubMed Scopus (388) Google Scholar, 5Jans D.A. Biochem. J. 1995; 311: 705-716Crossref PubMed Scopus (179) Google Scholar). NLS-dependent nuclear protein import can be divided into two steps, the first of which is energy-independent and involves recognition and targeting of the NLS-bearing protein to the nuclear pore complex (NPC) by a heterodimeric protein complex. The NLS is specifically recognized by the smaller subunit of the complex or NLS receptor (6Adam S.A. Gerace L. Cell. 1991; 66: 837-847Abstract Full Text PDF PubMed Scopus (318) Google Scholar), known variously as importin 58 (7Imamoto N. Shimamoto T. Takao T. Tachibana T. Kose S. Matsubae M. Sekimoto T. Shimonishi Y. Yoneda Y. EMBO J. 1995; 14: 3617-3626Crossref PubMed Scopus (271) Google Scholar), importin α (8Görlich D. Prehn S. Laskey R.A. Hartmann E. Cell. 1994; 79: 767-778Abstract Full Text PDF PubMed Scopus (601) Google Scholar), hSRP1/NP-1 (9Weis K. Mattaj I.W. Lamond A.I. Science. 1995; 268: 1049-1053Crossref PubMed Scopus (307) Google Scholar), or karyopherin α (10Moroianu J. Blobel G. Radu A. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 2008-2011Crossref PubMed Scopus (251) Google Scholar). The larger protein subunit, importin 97 (11Imamoto N. Shimamoto T. Kose S. Takao T. Tachibana T. Matsubae M. Sekimoto T. Shimonishi Y. Yoneda Y. FEBS Lett. 1995; 368: 415-419Crossref PubMed Scopus (156) Google Scholar), importin β (12Görlich D. Kostka S. Kraft R. Dingwall C. Laskey R.A. Hartmann E. Prehn S. Curr. Biol. 1995; 5: 383-392Abstract Full Text Full Text PDF PubMed Scopus (417) Google Scholar), karyopherin β (13Radu A. Blobel G. Moore M.S. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 1769-1773Crossref PubMed Scopus (383) Google Scholar), p97 (14Chi N.C. Adam E.J.H. Adam S.A. J. Cell Biol. 1995; 130: 265-274Crossref PubMed Scopus (247) Google Scholar), or Kap95p (15Enenkel C. Blobel G. Rexach M. J. Biol. Chem. 1995; 270: 16499-16502Abstract Full Text Full Text PDF PubMed Scopus (187) Google Scholar), binds importin 58 specifically but cannot bind the NLS. Its role is to target the importin-NLS carrying protein complex to the NPC through its affinity for NPC components such as nucleoporins (13Radu A. Blobel G. Moore M.S. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 1769-1773Crossref PubMed Scopus (383) Google Scholar,16Krämer D.M. Strambio de Castillia C. Blobel G. Rout M.P. J. Biol. Chem. 1995; 270: 19017-19021Abstract Full Text Full Text PDF PubMed Scopus (114) Google Scholar, 17Radu A. Moore M.S. Blobel G. Cell. 1995; 81: 215-222Abstract Full Text PDF PubMed Scopus (389) Google Scholar, 18Moroianu J. Hijikata M. Blobel G. Radu A. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 6532-6536Crossref PubMed Scopus (249) Google Scholar). The second, energy-dependent step of translocation of the import substrate through the NPC into the nucleus (19Newmeyer D.D. Forbes D.J. Cell. 1988; 52: 641-653Abstract Full Text PDF PubMed Scopus (371) Google Scholar, 20Richardson W.D. Mills A.D. Dilworth S.M. Laskey R.A. Dingwall C. Cell. 1988; 52: 655-664Abstract Full Text PDF PubMed Scopus (373) Google Scholar) requires the GTPase Ran/TC4 (21Melchior F. Paschal B. Evans E. Gerace L. J. Cell Biol. 1993; 123: 1649-1659Crossref PubMed Scopus (472) Google Scholar, 22Moore M.S. Blobel G. Nature. 1993; 365: 661-663Crossref PubMed Scopus (641) Google Scholar) and the Ran-interacting factor p10/NTF2 (23Moore M.S. Blobel G. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 10212-10216Crossref PubMed Scopus (291) Google Scholar, 24Paschal B.M. Gerace L. J. Cell Biol. 1995; 129: 925-937Crossref PubMed Scopus (342) Google Scholar). In addition to the NLS, phosphorylation in the vicinity of NLSs has been shown to play a role in regulating nuclear protein import through modulation of NLS function in either a positive or negative fashion (4Jans D.A. Hübner S. Physiol. Rev. 1996; 76: 651-685Crossref PubMed Scopus (388) Google Scholar,5Jans D.A. Biochem. J. 1995; 311: 705-716Crossref PubMed Scopus (179) Google Scholar). The regulatory modules able to confer regulated nuclear protein import on heterologous proteins, called phosphorylation-regulated NLSs (prNLSs) (4Jans D.A. Hübner S. Physiol. Rev. 1996; 76: 651-685Crossref PubMed Scopus (388) Google Scholar, 5Jans D.A. Biochem. J. 1995; 311: 705-716Crossref PubMed Scopus (179) Google Scholar), have been identified for a number of nuclear proteins (25Jans D.A. Ackermann M.J. Bischoff J.R. Beach D.H. Peters R. J. Cell Biol. 1991; 115: 1203-1212Crossref PubMed Scopus (183) Google Scholar, 26Jans D.A. Moll T. Nasmyth K. Jans P. J. Biol. Chem. 1995; 270: 17064-17067Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar, 27Moll T. Tebb G. Surana U. Robitsch H. Nasmyth K. Cell. 1991; 66: 743-758Abstract Full Text PDF PubMed Scopus (414) Google Scholar, 28Mosialos G. Hamer P. Capobianco A.J. Laursen R.A. Gilmore T.D. Mol. Cell Biol. 1991; 11: 5867-5877Crossref PubMed Scopus (95) Google Scholar, 29Norris J.L. Manley J.L. Genes & Dev. 1992; 6: 1654-1667Crossref PubMed Scopus (67) Google Scholar). Despite clear evidence that prNLSs regulate nuclear protein import (4Jans D.A. Hübner S. Physiol. Rev. 1996; 76: 651-685Crossref PubMed Scopus (388) Google Scholar, 5Jans D.A. Biochem. J. 1995; 311: 705-716Crossref PubMed Scopus (179) Google Scholar), the precise mechanism of the action of phosphorylation in terms of regulating NLS function is largely unclear, especially with respect to prNLSs where phosphorylation enhances transport. Although nuclear import in the case of T-ag is completely dependent on the T-ag NLS (amino acids 126–132) (30Rihs H.-P. Peters R. EMBO J. 1989; 8: 1479-1484Crossref PubMed Scopus (124) Google Scholar), measurements of nuclear import kinetics at the single cell level have shown that the sequences amino-terminal to the NLS (amino acid residues 111–125) enhance the rate of nuclear transport about 50-fold (31Rihs H.-P. Jans D.A. Fan H. Peters R. EMBO J. 1991; 10: 633-639Crossref PubMed Scopus (302) Google Scholar, 32Jans D.A. Jans P. Oncogene. 1994; 9: 2961-2968PubMed Google Scholar). The sequence primarily responsible for this effect has been demonstrated to be the protein kinase CK2 (previously casein kinase II) site at serine 112 (31Rihs H.-P. Jans D.A. Fan H. Peters R. EMBO J. 1991; 10: 633-639Crossref PubMed Scopus (302) Google Scholar, 32Jans D.A. Jans P. Oncogene. 1994; 9: 2961-2968PubMed Google Scholar), but the mechanism of action of phosphorylation at the site is unknown. The cyclin-dependent kinase site at threonine 124 flanking the NLS also regulates T-ag nuclear import through negative modulation of the maximum level of nuclear accumulation (25Jans D.A. Ackermann M.J. Bischoff J.R. Beach D.H. Peters R. J. Cell Biol. 1991; 115: 1203-1212Crossref PubMed Scopus (183) Google Scholar), and the prNLS or regulatory module for T-ag nuclear import is thus called the "CcN motif," where "C" denotes the CK2 site, "c" denotes the cyclin-dependent kinase site, and "N" denotes the NLS (25Jans D.A. Ackermann M.J. Bischoff J.R. Beach D.H. Peters R. J. Cell Biol. 1991; 115: 1203-1212Crossref PubMed Scopus (183) Google Scholar, 32Jans D.A. Jans P. Oncogene. 1994; 9: 2961-2968PubMed Google Scholar). Because the CK2 phosphorylation event enhancing T-ag transport clearly occurs in the cytoplasm prior to nuclear import (32Jans D.A. Jans P. Oncogene. 1994; 9: 2961-2968PubMed Google Scholar), we hypothesized that phosphorylation may modulate the affinity of the interaction of the NLS receptor with the T-ag NLS. In the present study we address this possibility directly by using a novel ELISA-based binding assay to quantitate NLS recognition by importin subunits. We find that NLS binding by importin 58 was strongly dependent on importin 97. Significantly, sequences flanking the T-ag NLS enhance recognition by importin 58/97 by about 10-fold, whereas CK2 phosphorylation increases binding by a further 40%. This modulation of NLS recognition through CK2 phosphorylation presumably constitutes the basis of enhancement of the rate of nuclear import by the CK2 site in vivo. T-ag fusion proteins CcN-β-Gal, CN-β-Gal, N-β-Gal, Cc-β-Gal, cN-β-Gal1, cN-β-Gal2, and sh-cN-β-Gal containing T-ag amino acids fused amino-terminal to theEscherichia coli β-galactosidase enzyme sequence (amino acids 9–1023) have been described previously (25Jans D.A. Ackermann M.J. Bischoff J.R. Beach D.H. Peters R. J. Cell Biol. 1991; 115: 1203-1212Crossref PubMed Scopus (183) Google Scholar, 31Rihs H.-P. Jans D.A. Fan H. Peters R. EMBO J. 1991; 10: 633-639Crossref PubMed Scopus (302) Google Scholar, 32Jans D.A. Jans P. Oncogene. 1994; 9: 2961-2968PubMed Google Scholar) (for amino acid sequences see Table I). T-ag fusion protein sh-CcN-β-Gal (in which T-ag amino acids 116–121 are deleted) was derived through oligonucleotide site-directed mutagenesis (CLONTECHTransformer kit) of plasmid pPF10, which encodes fusion protein sh-cN-β-Gal (31Rihs H.-P. Jans D.A. Fan H. Peters R. EMBO J. 1991; 10: 633-639Crossref PubMed Scopus (302) Google Scholar). The peptides pep101Lys and pep101Thr (see Table I) were provided by Imre Pavo and Gabor Toth (Endocrine Unit, Szeged University Medical School, Szeged, Hungary), whereas pep11Lys was from Bachem. T-ag fusion proteins were expressed, purified, and labeled fluorescently as described previously (31Rihs H.-P. Jans D.A. Fan H. Peters R. EMBO J. 1991; 10: 633-639Crossref PubMed Scopus (302) Google Scholar). Mouse importin 58-glutathione S-transferase (GST) and 97-GST expression plasmids were expressed as described (7Imamoto N. Shimamoto T. Takao T. Tachibana T. Kose S. Matsubae M. Sekimoto T. Shimonishi Y. Yoneda Y. EMBO J. 1995; 14: 3617-3626Crossref PubMed Scopus (271) Google Scholar, 11Imamoto N. Shimamoto T. Kose S. Takao T. Tachibana T. Matsubae M. Sekimoto T. Shimonishi Y. Yoneda Y. FEBS Lett. 1995; 368: 415-419Crossref PubMed Scopus (156) Google Scholar).Table IBinding of importin 58/97 to T-ag fusion proteins and peptides as quantified using an ELISA-based binding assayFusion protein or peptideSV40 T-antigen sequence1-aAll proteins contain T-ag sequences fused amino-terminal to E. coli β-galactosidase (amino acids 9–1023). The single letter amino acid code is used, the NLS is double underlined, and the CK2 site is in bold letters. Small letters indicate deviations from the T-ag sequence.Presence of functionalB max relative to CcN-β-Gal1-bData represent the means ± S.E., with n in parentheses.K D relative to CcN-β-GalInitial nuclear import rate relative to CcN-β-Gal1-cThe initial rates of nuclear import are calculated from Refs. 31, 33 and this study (Fig. 2 B).CK2 siteNLS%fold%Fusion proteins CcN-β-GalSSDDEATADSQHSTPPKKKRKV132++100 (8)1100 CN-β-GalSSDDEATADaQHaaPPKKKRKV132++104 ± 1.2 (2)1.17 ± 0.24100 sh-CcN-β-GalSSDDEHSTPPKKKRKV132++70 ± 5.7 (4)1.7 ± 0.1042 sh-cN-β-GalSQHSTPPKKKRKV132−+53 ± 5.2 (5)2.4 ± 0.1810.5 cN-β-Gal1agDDEATADSQHSTPPKKKRKV132−+72 ± 1.6 (8)2.3 ± 0.2611.9 cN-β-Gal2SSnnqATADSQHSTPPKKKRKV132−+74 ± 4.0 (2)2.4 ± 0.5211.6 N-β-GalPKKKRKV132−+10 ± 1.8 (2)ND1-dND, not able to be determined.6.4 Cc-β-GalSSDDEATADSQHSTPPKtKRKV132+−1.9 ± 1.9 (2)ND NDPeptides pep101LysPgSDDEAaADaQHaaPPKKKRKV132++100 (2)ND ND pep11LysPKKKRKV132−+7.3 (1)ND ND pep101ThrPgSDDEAaADaQHaaPPKtKRKV132+−1.2 ± 1.2 (2)ND ND1-a All proteins contain T-ag sequences fused amino-terminal to E. coli β-galactosidase (amino acids 9–1023). The single letter amino acid code is used, the NLS is double underlined, and the CK2 site is in bold letters. Small letters indicate deviations from the T-ag sequence.1-b Data represent the means ± S.E., with n in parentheses.1-c The initial rates of nuclear import are calculated from Refs. 31Rihs H.-P. Jans D.A. Fan H. Peters R. EMBO J. 1991; 10: 633-639Crossref PubMed Scopus (302) Google Scholar, 33Xiao C.-Y. Hübner S. Elliot R.M. Caon A. Jans D.A. J. Biol. Chem. 1996; 271: 6451-6457Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar and this study (Fig. 2 B).1-d ND, not able to be determined. Open table in a new tab Cells of the HTC rat hepatoma tissue culture cell line (a derivative of Morris hepatoma 7288C) were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum (25Jans D.A. Ackermann M.J. Bischoff J.R. Beach D.H. Peters R. J. Cell Biol. 1991; 115: 1203-1212Crossref PubMed Scopus (183) Google Scholar, 31Rihs H.-P. Jans D.A. Fan H. Peters R. EMBO J. 1991; 10: 633-639Crossref PubMed Scopus (302) Google Scholar). Analysis of nuclear import kinetics at the single cell level using microinjected HTC cells in conjunction with confocal laser scanning microscopy (CLSM) (Bio-Rad MRC-600) was as described previously (33Xiao C.-Y. Hübner S. Elliot R.M. Caon A. Jans D.A. J. Biol. Chem. 1996; 271: 6451-6457Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar). Microinjected HTC cells were fused with polyethylene glycol about 1 h prior to microinjection to produce polykaryons (31Rihs H.-P. Jans D.A. Fan H. Peters R. EMBO J. 1991; 10: 633-639Crossref PubMed Scopus (302) Google Scholar). Quantitation of fluorescence using CLSM (see Refs. 34Jans D.A. Peters R. Zsigo J. Fahrenholz F. EMBO J. 1989; 8: 2481-2488Crossref PubMed Scopus (51) Google Scholar and 35Jans D.A. Peters R. Fahrenholz F. EMBO J. 1990; 9: 2693-2699Crossref PubMed Scopus (29) Google Scholar for other applications), image analysis of CLSM files using the NIH Image public domain software, and curve fitting have been described previously in detail (26Jans D.A. Moll T. Nasmyth K. Jans P. J. Biol. Chem. 1995; 270: 17064-17067Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar, 33Xiao C.-Y. Hübner S. Elliot R.M. Caon A. Jans D.A. J. Biol. Chem. 1996; 271: 6451-6457Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar). In vitrophosphorylation of T-ag fusion proteins by rat liver CK2 (Promega) was performed with 1 mg/ml fusion protein as described previously (33Xiao C.-Y. Hübner S. Elliot R.M. Caon A. Jans D.A. J. Biol. Chem. 1996; 271: 6451-6457Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar, 36Ackermann P. Osherhoff N. J. Cell Biol. 1989; 264: 11958-11965Google Scholar) in the absence or the presence of [γ-32P]ATP (3000 Ci/mmol). The stoichiometry of phosphorylation was quantitated as described (25Jans D.A. Ackermann M.J. Bischoff J.R. Beach D.H. Peters R. J. Cell Biol. 1991; 115: 1203-1212Crossref PubMed Scopus (183) Google Scholar). Subsequent to either spotting T-ag fusion proteins onto nitrocellulose membrane (dot blot) or electrophoresis of T-ag fusion proteins on a 7.5% SDS-polyacrylamide gel and transfer to a nitrocellulose membrane (ligand-Western blot), the membrane was blocked in intracellular buffer (IB; 110 mm KCl, 5 mm NaHC03, 5 mm MgCl2, 1 mm EGTA, 0.1 mm CaCl2, 20 mm Hepes, 1 mm dithiothreitol, 5 μg/ml leupeptin, pH 7.4) containing 5% BSA for 4 h at 22 °C. Hybridization was performed for 16 h at 4 °C in IB/1% BSA supplemented with importin 58-GST, importin 97-GST, or precomplexed importin 58 (with or without GST)/97-GST. Precomplexation of mouse importin subunits (molar ratio of 1:1) was carried out in IB for 15 min at 22 °C. The membrane was extensively washed, and bound mouse importin subunit(s) were detected and visualized using a GST-specific antibody (Pharmacia Biotech Inc.), alkaline phosphatase-conjugated second antibody (Sigma) and nitro blue tetrazolium/5-bromo-4-chloro-3-indolyl-1-phosphate (Promega). Fusion proteins (0.5 μg/well) or peptides (1 μg/well) in 50 mm NaHCO3, pH 9.8, were coated in triplicate wells of polystyrene microtiterplates (Nunc) for 16 h at 4 °C. After blocking in IB/5% BSA for 90 min at 22 °C, appropriate dilutions of precomplexed importin 58/97-GST in IB/1% BSA were then added to the microtiterplates and incubated for 16 h at 4 °C. After washing the wells with IB/1% BSA, bound mouse importin 58/97-GST complex was detected as described for the ligand-Western blot assay and visualized withp-nitrophenyl phosphate (Sigma). The change of absorbance at 405 nm was followed with time using a plate reader (Molecular Devices), with values corrected by subtracting both the absorbance at 0 min, and the absorbance in wells incubated without importin 58/97-GST complex. To correct for differences in coating, T-ag fusion proteins were subjected to a parallel β-galactosidase ELISA assay using β-galactosidase-specific antibodies. Measurements for importin binding were ultimately corrected for any differences in coating efficiencies quantified in the β-galactosidase-based ELISA assay to enable a true estimate of bound importin. Initial experiments to investigate binding of the importin 58/97 subunits to the T-ag NLS (Table I for T-ag sequences of the fusion proteins) using the dot blot technique indicated poor binding of importin 58 to the T-ag NLS in the absence of importin 97, which alone did not bind the T-ag NLS at all (Fig.1 A). This was consistent with the results of others that the importin 58/97 complex constitutes the high affinity NLS receptor (37Rexach M. Blobel G. Cell. 1995; 83: 683-692Abstract Full Text PDF PubMed Scopus (665) Google Scholar, 38Görlich D. Pante N. Kutay U. Aebi U. Bischoff F.R. EMBO J. 1996; 15: 5584-5594Crossref PubMed Scopus (533) Google Scholar). Interestingly, importin 58/97 only showed strong binding to the T-ag NLS in the presence, compared with in the absence, of T-ag amino acids 111–125 flanking the NLS (amino acids 126–132) (Fig. 1 A). To evaluate this further, the ligand-Western blot technique was used to assess binding of the importin 58/97 complex to various T-ag fusion proteins differing in the T-ag sequences amino-terminal to the NLS. As in the dot blot assay, the T-ag fusion proteins Cc-β-Gal and N-β-Gal bound importin 58/97 poorly compared with CcN-β-Gal (Fig. 1 B). The sh-cN-β-Gal protein bound importin 58/97 to intermediate levels, implying that the reduced binding relative to CcN-β-Gal may have been due to the absence of the CK2 site. The T-ag fusion protein derivative sh-CcN-β-Gal, containing a deletion of amino acids 116–121 ("short" version of CcN-β-Gal), thereby reducing the distance between the CK2 site and the NLS from 10 to 4 residues (Table I), did not bind importin 58/97 complex as well as CcN-β-Gal (see also below), implying that positioning of the CK2 site relative to the NLS was crucial to efficient binding by the importin 58/97 complex. The functional significance of the altered position of the CK2 site in sh-CcN-β-Gal was assessed in nuclear import studies in microinjected cells of the HTC rat hepatoma line using CLSM. The sh-CcN-β-Gal fusion protein exhibited maximal nuclear accumulation less than half that of CcN-β-Gal (Fig. 2) but higher than that of a protein lacking a functional CK2 site (31Rihs H.-P. Jans D.A. Fan H. Peters R. EMBO J. 1991; 10: 633-639Crossref PubMed Scopus (302) Google Scholar, 32Jans D.A. Jans P. Oncogene. 1994; 9: 2961-2968PubMed Google Scholar), indicating the functional importance of the correct position of the CK2 site relative to the NLS for nuclear import. It also exhibited a reduced initial rate of nuclear import relative to the wild type fusion protein, the reduced nuclear accumulation of sh-CcN-β-Gal relative to the wild type fusion protein thus correlating with the decreased binding by importin 58/97 (Fig. 1 B and see below). That reduced transport was not attributable to reduced efficiency of phosphorylation at the CK2 site was demonstrated by phosphorylation experiments with purified CK2, where sh-CcN-β-Gal was phosphorylated to an identical extent to CcN-β-Gal (stoichiometry of phosphorylation of 2.04 and 2.02 mol Pi/mol tetramer for sh-CcN-β-Gal and CcN-β-Gal, respectively). To define the role of the T-ag amino-terminal sequences in NLS binding by importin 58/97 in quantitative terms, a novel ELISA-based binding assay was established (see "Experimental Procedures" for details) whereby T-ag fusion proteins coated onto microtiterplates were incubated with increasing amounts of importin 58 and 97 or importin 58/97 complex. Negligible association of importin 97 with CcN-β-Gal was observed, while only low efficiency binding was evident when importin 58 alone was used compared with that when the importin 58/97 complex was used (not shown), consistent with dot blot analysis (Fig. 1 A). The apparent dissociation constant (K D ) of CcN-β-Gal for importin 58 alone was 10 times that for the importin 58/97 complex. The role of the different regions of the CcN motif in recognition of the T-ag NLS by importin 58/97 was assessed using the CN-β-Gal, sh-CcN-β-Gal, cN-β-Gal1, cN-β-Gal2, sh-cN-β-Gal, and N-β-Gal T-ag fusion protein derivatives (Fig.3 A and Table I). The amino-terminal sequences flanking the NLS enhanced binding by importin 58/97 whereby binding to wild type protein (CcN-β-Gal) was about 10-fold higher than that to the fusion protein containing the T-ag NLS alone (N-β-Gal) (Table I). Cc-β-Gal bound negligible amounts of importin 58/97, indicating the specificity of the assay (Table I). The sh-cN-Gal protein, which lacks the CK2 site, bound importin 58/97 to levels about 50% that of CcN-β-Gal, implying that the CK2 site was necessary for efficient importin binding. The cN-β-Gal1 protein, containing nonphosphorylatable residues in place of the CK2 site serines 111 and 112 (32Jans D.A. Jans P. Oncogene. 1994; 9: 2961-2968PubMed Google Scholar), and cN-β-Gal2, carrying a mutated CK2 recognition site (substitution of Asp-Asp-Glu115 by Asn-Asn-Gln) but retaining serines 111 and 112, also bound importin to a reduced level, supporting the idea of the CK2 site being directly involved in importin recognition. This was further supported by the fact that the sh-CcN-β-Gal protein, the "short" version of the CcN motif containing a functional CK2 site, bound higher levels of importin 58/97 than sh-cN-Gal (deleted for amino acids 111–119, including the CK2 site) (Fig. 3 A); binding, however, was reduced relative to wild type, implying that although the CK2 site is essential for high affinity binding by importin, its correct positioning relative to the NLS is a critical factor. The lack of a correctly positioned CK2 site in sh-CcN-β-Gal increased theK D by almost 2-fold relative to CcN-β-Gal (Fig.3 A and Table I), while the complete lack of a functional CK2 site (sh-cN-Gal, cN-β-Gal1, and cN-β-Gal2) increased the K D even further. Substitution of serines 120 and 123 together with threonine 124 (the site of phosphorylation by the cyclin-dependent kinasecdc2, which reduces the maximal level of T-ag fusion protein nuclear accumulation) (25Jans D.A. Ackermann M.J. Bischoff J.R. Beach D.H. Peters R. J. Cell Biol. 1991; 115: 1203-1212Crossref PubMed Scopus (183) Google Scholar) by alanine residues (the CN-β-Gal protein) had no significant effect on importin recognition relative to CcN-β-Gal (Table I), further underlining the central importance of the CK2 phosphorylation site in importin binding. Taking the results from Table I (the rightmost data columns), the K D relative to wild type for the respective fusion proteins correlated very well (regression coefficient of 0.99) with their initial rates of nuclear import in vivo determined either previously (31Rihs H.-P. Jans D.A. Fan H. Peters R. EMBO J. 1991; 10: 633-639Crossref PubMed Scopus (302) Google Scholar, 32Jans D.A. Jans P. Oncogene. 1994; 9: 2961-2968PubMed Google Scholar) or in this study, implying that the apparent dissociation constant is critical in determining the initial rate of nuclear import. Peptides comprising T-ag amino acids 112–132 (pep101Lys), T-ag amino acids 112–132 with a nonfunctional NLS (pep101Thr), and T-ag amino acids 126–132 (pep11Lys) (Table I) were also tested. Consistent with the studies using the T-ag fusion proteins, binding of the importin 58/97 complex to pep101Lys (K D of 17 nm) was significantly higher than that to pep11Lys, underlining the importance of the flanking regions to T-ag NLS recognition by importin 58/97. Peptide pep101Thr showed negligible binding (Fig. 3 Band Table I), again confirming the specificity of the ELISA-based binding assay. Our previous work had demonstrated that nuclear accumulation of CcN-β-Gal is much more rapid than that of cN-β-Gal1, cN-β-Gal2, and sh-cN-β-Gal, all of which lack a functional CK2 site (Refs. 31Rihs H.-P. Jans D.A. Fan H. Peters R. EMBO J. 1991; 10: 633-639Crossref PubMed Scopus (302) Google Scholar and 32Jans D.A. Jans P. Oncogene. 1994; 9: 2961-2968PubMed Google Scholar and Table I). To test whether phosphorylation at the CK2 site influences binding of importin 58/97 to the T-ag NLS, CcN-β-Gal was phosphorylated with CK2 to a stoichiometry of 1.0–1.5 mol Pi/mol tetrameric T-ag fusion protein, and importin 58/97 binding was quantified using the ELISA-based binding assay. The clear result was that phosphorylation at the CK2 site increased binding of importin 58/97, with about a 40% decrease in the value of the K D compared with unphosphorylated CcN-β-Gal (Table II). Differences between non- and prephosphorylated T-ag proteins were much more marked (up to 2.5-fold differences in the amount of importin 58/97 complex bound) at lower importin 58/97 concentrations (data not shown); it should also be stressed that the quantitative differences here almost certainly underestimate the effect of CK2 phosphorylation in terms of increasing the affinity for importin 58/97 because CK2 prephosphorylated CcN-β-Gal tetramers represent a mixed population of unphosphorylated and phosphorylated CK2 sites as a result of the stoichiometry of phosphorylation (i.e. 2.5 mol of CK2 sites per tetramer are not phosphorylated). This almost certainly results in underestimation of the K D for importin association of CK2-phosphorylated protein.Table IIEffect of CK2 prephosphorylation of T-ag fusion proteins on binding to importin 58/97 as quantified using an ELISA-based binding assayFusion protein2-aThe stoichiometry of phosphorylation for CcN-β-Gal was between 1.1 and 1.5, for Cc-β-Gal it was between 1.4 and 1.5, and for N-β-Gal and cN-β-Gal1 it was 0 mol Pi/mol tetrameric fusion protein.Presence of CK2 siteB maxrelative to unphosphorylated fusion protein2-bData represent the means ± S.E., with n in parentheses.K D relative to unphosphorylated fusion protein2-bData represent the means ± S.E., with n in parentheses.%foldCcN-β-Gal+107 ± 1.9 (4)0.72 ± 0.015cN-β-Gal1− 85 (1)ND2-cND, not able to be determined.N-β-Gal−104 (1)NDCc-β-Gal+316 ± 2.1 (2)ND2-a The stoichiometry of phosphorylation for CcN-β-Gal was between 1.1 and 1.5, for Cc-β-Gal it was between 1.4 and 1.5, and for N-β-Gal and cN-β-Gal1 it was 0 mol Pi/mol tetrameric fusion protein.2-b Data represent the means ± S.E., with n in parentheses.2-c ND, not able to be determined. Open table in a new tab To confirm that the effects on importin binding did not result from trace amounts of CK2 in the protein samples in the case of prephosphorylated protein, control experiments were carried out using sh-cN-β-Gal, N-β-Gal, and Cc-β-Gal preincubated with CK2. Phosphorylation of sh-cN-β-Gal and N-β-Gal was undetectable, as expected (Table II). There was no increase in the maximal binding of the CK2-preincubated proteins sh-cN-β-Gal and N-β-Gal compared with the nonpreincubated T-ag fusion proteins (Table II). We also found no increased binding of importin 58/97 to CK2 preincubated β-Gal (not shown). Significantly, a 3-fold increase in binding of importin 58/97 to CK2 prephosphorylated Cc-β-Gal was observed relative to nonphosphorylated Cc-β-Gal (Table II). This enhancing effect of CK2 site phosphorylation on binding by importin 58/97 even in the absence of a functional NLS is a further indication that the CK2 site probably participates directly in binding to importin 58. In conclusion, this study establishes that the mechanistic basis of CK2 site-mediated enhancement of T-ag nuclear import is through the CK2 site and CK2 site phosphorylation increasing the affinity of interaction of the NLS receptor importin 58/97 with the T-ag NLS. As supported by the correlation between the K D and initial nuclear import rate in vivo (see above), the higher affinity of binding presumably results in more rapid kinetics of association of the NLS receptor with the T-ag transport substrate and faster docking at the NPC ultimately leading to an accelerated rate of transport, which has been observed both in vivo and in vitro. (25Jans D.A. Ackermann M.J. Bischoff J.R. Beach D.H. Peters R. J. Cell Biol. 1991; 115: 1203-1212Crossref PubMed Scopus (183) Google Scholar, 31Rihs H.-P. Jans D.A. Fan H. Peters R. EMBO J. 1991; 10: 633-639Crossref PubMed Scopus (302) Google Scholar, 32Jans D.A. Jans P. Oncogene. 1994; 9: 2961-2968PubMed Google Scholar, 33Xiao C.-Y. Hübner S. Elliot R.M. Caon A. Jans D.A. J. Biol. Chem. 1996; 271: 6451-6457Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar). This study thus represents the first elucidation of the mechanistic basis of the regulation of nuclear protein import through phosphorylation within a prNLS. We thank Prof. Yoshihiro Yoneda (Osaka University Medical School, Osaka) for providing the mouse importin 58 and 97 expression plasmids, Drs. Imre Pavo and Gabor Toth (Endocrine Unit, University of Szeged) for peptides pep101Lys and pep101Thr, the Clive and Vera Ramaciotti Foundation for its support of the work, and Patricia Jans and Lyndall J. Briggs for skilled technical assistance.

Referência(s)