Cytoplasmic Localization of Tristetraprolin Involves 14-3-3-dependent and -independent Mechanisms
2002; Elsevier BV; Volume: 277; Issue: 20 Linguagem: Inglês
10.1074/jbc.m110465200
ISSN1083-351X
AutoresBarbra A. Johnson, Justine R. Stehn, Michael B. Yaffe, T. Keith Blackwell,
Tópico(s)Microbial Natural Products and Biosynthesis
ResumoThe immediate early gene tristetraprolin (TTP) is induced transiently in many cell types by numerous extracellular stimuli. TTP encodes a zinc finger protein that can bind and destabilize mRNAs that encode tumor necrosis factor-alpha (TNFα) and other cytokines. We hypothesize that TTP also has a broader role in growth factor-responsive pathways. In support of this model, we have previously determined that TTP induces apoptosis through the mitochondrial pathway, analogously to certain oncogenes and other immediate-early genes, and that TTP sensitizes cells to the pro-apoptotic signals of TNFα. In this study, we show that TTP and the related proteins TIS11b and TIS11d bind specifically to 14-3-3 proteins and that individual 14-3-3 isoforms preferentially bind to different phosphorylated TTP species. 14-3-3 binding does not appear to inhibit or promote induction of apoptosis by TTP but is one of multiple mechanisms that localize TTP to the cytoplasm. Our results provide the first example of 14-3-3 interacting functionally with an RNA binding protein and binding in vivo to a Type II 14-3-3 binding site. They also suggest that 14-3-3 binding is part of a complex network of stimuli and interactions that regulate TTP function. The immediate early gene tristetraprolin (TTP) is induced transiently in many cell types by numerous extracellular stimuli. TTP encodes a zinc finger protein that can bind and destabilize mRNAs that encode tumor necrosis factor-alpha (TNFα) and other cytokines. We hypothesize that TTP also has a broader role in growth factor-responsive pathways. In support of this model, we have previously determined that TTP induces apoptosis through the mitochondrial pathway, analogously to certain oncogenes and other immediate-early genes, and that TTP sensitizes cells to the pro-apoptotic signals of TNFα. In this study, we show that TTP and the related proteins TIS11b and TIS11d bind specifically to 14-3-3 proteins and that individual 14-3-3 isoforms preferentially bind to different phosphorylated TTP species. 14-3-3 binding does not appear to inhibit or promote induction of apoptosis by TTP but is one of multiple mechanisms that localize TTP to the cytoplasm. Our results provide the first example of 14-3-3 interacting functionally with an RNA binding protein and binding in vivo to a Type II 14-3-3 binding site. They also suggest that 14-3-3 binding is part of a complex network of stimuli and interactions that regulate TTP function. The immediate-early protein tristetraprolin (TTP 1The abbreviations used are: TTPtristetraprolinTNFαtumor necrosis factor-alphaAREAU-rich elementGFPgreen fluorescence proteinGSTglutathione S-transferaseMAPKmitogen-activated protein kinaseNESnuclear export signalNLSnuclear localization sequenceX-gal5-bromo-4-chloro-3-indolyl-β-d-galactopyranosideHAhemagglutininnTTPantiserum raised against the N terminus of TTPcTTPantiserum raised against the C terminus of TTPDMEMDulbecco's modified Eagle's medium ; also Nup475 and TIS11) is expressed transiently during responses to many extracellular stimuli, including TNFα (1Carballo E. Lai W.S. Blackshear P.J. Science. 1998; 281: 1001-1005Crossref PubMed Google Scholar). TTP and the related proteins TIS11b and TIS11d (TTP/TIS11 proteins) consist of two conserved Cys-X8-Cys-X5-Cys-X3-His (CCCH) zinc fingers, along with similarly sized but divergent N- and C-terminal regions. Several lines of evidence indicate that TTP binds and destabilizes cytokine mRNAs, through binding to an AU-rich element (ARE) located within their 3′-untranslated regions. This ARE is targeted by conserved signaling pathways, which regulate the localization, stability, and translation of these mRNAs (2Chen C.Y. Shyu A.B. Trends Biochem. Sci. 1995; 20: 465-470Abstract Full Text PDF PubMed Scopus (1688) Google Scholar, 3Kotlyarov A. Neininger A. Schubert C. Eckert R. Birchmeier C. Volk H.D. Gaestel M. Nat. Cell Biol. 1999; 1: 94-97Crossref PubMed Scopus (687) Google Scholar, 4Kontoyiannis D. Pasparakis M. Pizarro T.T. Cominelli F. Kollias G. Immunity. 1999; 10: 387-398Abstract Full Text Full Text PDF PubMed Scopus (1109) Google Scholar, 5Winzen R. Kracht M. Ritter B. Wilhelm A. Chen C.Y. Shyu A.B. Muller M. Gaestel M. Resch K. Holtmann H. EMBO J. 1999; 18: 4969-4980Crossref PubMed Scopus (713) Google Scholar, 6Dumitru C.D. Ceci J.D. Tsatsanis C. Kontoyiannis D. Stamatakis K. Lin J.H. Patriotis C. Jenkins N.A. Copeland N.G. Kollias G. Tsichlis P.N. Cell. 2000; 103: 1071-1083Abstract Full Text Full Text PDF PubMed Scopus (705) Google Scholar, 7Rutault K. Hazzalin C.A. Mahadevan L.C. J. Biol. Chem. 2001; 276: 6666-6674Abstract Full Text Full Text PDF PubMed Scopus (165) Google Scholar, 8Vasudevan S. Peltz S.W. Mol Cell. 2001; 7: 1191-1200Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar, 9Kontoyiannis D. Kotlyarov A. Carballo E. Alexopoulou L. Blackshear P.J. Gaestel M. Davis R. Flavell R. Kollias G. EMBO J. 2001; 20: 3760-3770Crossref PubMed Scopus (219) Google Scholar). TTP−/− mice develop a widespread inflammatory syndrome that is mediated largely by TNFα and is associated with elevated levels and half-life of the TNFα mRNA (10Taylor G.A. Carballo E. Lee D.M. Lai W.S. Thompson M.J. Patel D.D. Schenkman D.I. Gilkeson G.S. Broxmeyer H.E. Haynes B.F. Blackshear P.J. Immunity. 1996; 4: 445-454Abstract Full Text Full Text PDF PubMed Scopus (655) Google Scholar, 11Carballo E. Gilkeson G.S. Blackshear P.J. J. Clin. Invest. 1997; 100: 986-995Crossref PubMed Scopus (124) Google Scholar, 12Carballo E. Lai W.S. Blackshear P.J. Blood. 2000; 95: 1891-1899Crossref PubMed Google Scholar, 13Carballo E. Blackshear P.J. Blood. 2001; 98: 2389-2395Crossref PubMed Scopus (101) Google Scholar). A protein complex that contains TTP binds to the TNFα ARE (14Mahtani K.R. Brook M. Dean J.L. Sully G. Saklatvala J. Clark A.R. Mol. Cell. Biol. 2001; 21: 6461-6469Crossref PubMed Scopus (397) Google Scholar) and in transfection assays each TTP/TIS11 protein can bind and destabilize cytokine mRNAs that have related AREs (1Carballo E. Lai W.S. Blackshear P.J. Science. 1998; 281: 1001-1005Crossref PubMed Google Scholar, 12Carballo E. Lai W.S. Blackshear P.J. Blood. 2000; 95: 1891-1899Crossref PubMed Google Scholar, 15Lai W.S. Carballo E. Strum J.R. Kennington E.A. Phillips R.S. Blackshear P.J. Mol. Cell. Biol. 1999; 19: 4311-4323Crossref PubMed Scopus (638) Google Scholar, 16Lai W.S. Carballo E. Thorn J.M. Kennington E.A. Blackshear P.J. J. Biol. Chem. 2000; 275: 17827-17837Abstract Full Text Full Text PDF PubMed Scopus (299) Google Scholar, 17Ming X.F. Stoecklin G., Lu, M. Looser R. Moroni C. Mol. Cell. Biol. 2001; 21: 5778-5789Crossref PubMed Scopus (160) Google Scholar). These findings suggest that TTP limits expression of TNFα and other cytokines through a feedback mechanism, by destabilizing their mRNAs, and that this is a shared function of TTP/TIS11 proteins. Some other CCCH zinc finger proteins appear to regulate translation of their target genes (18Guedes S. Priess J. Development. 1997; 124: 731-739PubMed Google Scholar, 19Tabara H. Hill R.J. Mello C.C. Priess J.R. Kohara Y. Development. 1999; 126: 1-11Crossref PubMed Google Scholar, 20Schubert C.M. Lin R. de Vries C.J. Plasterk R.H. Priess J.R. Mol Cell. 2000; 5: 671-682Abstract Full Text Full Text PDF PubMed Scopus (190) Google Scholar, 21Tenenhaus C. Subramaniam K. Dunn M.A. Seydoux G. Genes Dev. 2001; 15: 1031-1040Crossref PubMed Scopus (69) Google Scholar), suggesting that many members of this protein family may regulate specific genes post-transcriptionally. tristetraprolin tumor necrosis factor-alpha AU-rich element green fluorescence protein glutathione S-transferase mitogen-activated protein kinase nuclear export signal nuclear localization sequence 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside hemagglutinin antiserum raised against the N terminus of TTP antiserum raised against the C terminus of TTP Dulbecco's modified Eagle's medium It appears likely that TTP has additional functions and may play a broader role during responses to extracellular stimuli. TTP expression is induced rapidly and directly in numerous cultured cell types, by a wide variety of growth factors and mitogens (22Lai W.S. Stumpo D.J. Blackshear P.J. J. Biol. Chem. 1990; 265: 16556-16563Abstract Full Text PDF PubMed Google Scholar, 23Varnum B.C. Lim R.W. Sukhatme V.P. Herschman H.R. Oncogene. 1989; 4: 119-120PubMed Google Scholar, 24DuBois R.N. Bishop P.R. Graves-Deal R. Coffey R.J. Cell Growth Differ. 1995; 6: 523-529PubMed Google Scholar). In mice, TTP is expressed in developing oocytes and regenerating small intestine and liver, in addition to hematopoietic tissues (10Taylor G.A. Carballo E. Lee D.M. Lai W.S. Thompson M.J. Patel D.D. Schenkman D.I. Gilkeson G.S. Broxmeyer H.E. Haynes B.F. Blackshear P.J. Immunity. 1996; 4: 445-454Abstract Full Text Full Text PDF PubMed Scopus (655) Google Scholar, 24DuBois R.N. Bishop P.R. Graves-Deal R. Coffey R.J. Cell Growth Differ. 1995; 6: 523-529PubMed Google Scholar, 25Heximer S.P. Forsdyke D.R. DNA Cell Biol. 1993; 12: 73-88Crossref PubMed Scopus (54) Google Scholar, 26te Kronnie G. Samallo J. Schipper H. Stroband H.W. Dev. Genes Evol. 2001; 211: 261-262Crossref PubMed Scopus (2) Google Scholar). Like some other immediate early proteins, in certain contexts TTP is also expressed during induction of apoptosis (27Haas C.A. Donath C. Kreutzberg G.W. Neuroscience. 1993; 53: 91-99Crossref PubMed Scopus (111) Google Scholar, 28Mesner P.W. Epting C.L. Hegarty J.L. Green S.H. J. Neurosci. 1995; 15: 7357-7366Crossref PubMed Google Scholar, 29Mittelstadt P.R. DeFranco A.L. J. Immunol. 1993; 150: 4822-4832PubMed Google Scholar, 30Harkin D.P. Bean J.M. Miklos D. Song Y.H. Truong V.B. Englert C. Christians F.C. Ellisen L.W. Maheswaran S. Oliner J.D. Haber D.A. Cell. 1999; 97: 575-586Abstract Full Text Full Text PDF PubMed Scopus (513) Google Scholar). Consistent with the model that TTP/TIS11 proteins influence growth, survival, or apoptotic signals, we have determined that their constitutive expression at modest levels induces apoptosis through the mitochondrial pathway (31Johnson B.A. Geha M. Blackwell T.K. Oncogene. 2000; 19: 1657-1664Crossref PubMed Scopus (68) Google Scholar). We have also shown that TTP has diverged functionally from the other two TTP/TIS11 proteins, in that TTP alone dramatically sensitizes cells to the apoptotic stimulus of TNFα (31Johnson B.A. Geha M. Blackwell T.K. Oncogene. 2000; 19: 1657-1664Crossref PubMed Scopus (68) Google Scholar,32Johnson, B. A., and Blackwell, T. K. (2002)Oncogene, in pressGoogle Scholar). This last finding suggests that TTP could contribute to the cellular decision between activation or apoptosis in response to TNFα. Although the isolated TTP zinc finger region can mediate its effects on TNFα mRNA stability in transfection assays (16Lai W.S. Carballo E. Thorn J.M. Kennington E.A. Blackshear P.J. J. Biol. Chem. 2000; 275: 17827-17837Abstract Full Text Full Text PDF PubMed Scopus (299) Google Scholar), we have observed that the TTP zinc finger region is incapable of inducing apoptosis or of sensitizing cells to TNFα-induced apoptosis (32Johnson, B. A., and Blackwell, T. K. (2002)Oncogene, in pressGoogle Scholar). Together with the zinc fingers, the TTP N- and C-terminal regions each contribute to induction of apoptosis, and the N-terminal region is specifically required to sensitize cells to TNFα (32Johnson, B. A., and Blackwell, T. K. (2002)Oncogene, in pressGoogle Scholar). In addition, although the isolated TTP zinc finger region is localized predominantly to the nucleus, both the N- and C-terminal regions of TTP promote its localization to the cytoplasm (32Johnson, B. A., and Blackwell, T. K. (2002)Oncogene, in pressGoogle Scholar). The importance of the TTP N- and C-terminal regions for these TTP activities makes it critical to identify proteins with which these TTP regions interact functionally. In this study, we have determined that 14-3-3 proteins bind to the TTP C-terminal region sequence-specifically and in a phosphorylation-dependent manner. This interaction appears to be conserved among all three TTP/TIS11 proteins. Mutagenesis analysis has identified a specific site in the TTP C terminus that is important for 14-3-3 binding. Binding to 14-3-3 through this site does not appear to be required for the apoptotic effects of TTP but is critical for one of at least three mechanisms that localize TTP to the cytoplasm. Our findings suggest that interactions with 14-3-3 are involved in phosphorylation-mediated signals that may regulate TTP functions in vivo. TTP and site-directed TTP mutants were introduced into the cytomegalovirus-based expression vector CS2+ (33Turner R. Tjian R. Science. 1989; 243: 1689-1694Crossref PubMed Scopus (425) Google Scholar) by PCR (Pfu, Stratagene), with a Kozak consensus and ATG added where appropriate. For two-hybrid analysis, TTP, and mutants indicated in Fig. 3 A were cloned by PCR using Pfu (Stratagene) into pC98, a pC97 derivative (34Vidal M. Bartel P.L. Fields S. The Yeast Two-hybrid System. Oxford University Press, Oxford1997: 109-147Google Scholar). To remove its 5′-untranslated region, the 14-3-3η prey coding sequence was re-cloned by PCR. TTP and TTP deletion mutants each were fused to green fluorescence protein (GFP) at their N terminus by restriction cloning into C2eGFP (CLONTECH). Analogous GFP fusions of TIS11b and TIS11d were made by PCR cloning in C1eGFP and C2eGFP, respectively. Transfections were carried out as described (31Johnson B.A. Geha M. Blackwell T.K. Oncogene. 2000; 19: 1657-1664Crossref PubMed Scopus (68) Google Scholar) using LipofectAMINE (Invitrogen), and 35-mm plates unless otherwise stated. DNA amounts were supplemented to 2 μg by addition of pBluescript. LipofectAMINE Plus and 1 μg of total DNA were used when higher efficiencies were desired. DNA amounts used for 10-cm plates for these two transfection methods were 10 and 5 μg, respectively. FuGENE (Roche Molecular Biochemicals) and 8 μg of total DNA were used to transfect 10-cm plates for Fig. 3 C. Cell death was assayed by co-transfection of a β-galactosidase reporter and examination of cell morphology after X-gal staining 24 h later (31Johnson B.A. Geha M. Blackwell T.K. Oncogene. 2000; 19: 1657-1664Crossref PubMed Scopus (68) Google Scholar). Under a variety of conditions, numbers of apoptotic cells identified by this method were reproducibly comparable to those detected by scoring Hoechst-stained pyknotic nuclei (not shown) (31Johnson B.A. Geha M. Blackwell T.K. Oncogene. 2000; 19: 1657-1664Crossref PubMed Scopus (68) Google Scholar). For serum-dependent relocalization assays, cells were washed twice in Dulbecco's modified Eagle's medium (DMEM) prior to transfection, and DMEM without serum was added 3 h later. 26–28 h after transfection, cells were stimulated with 20% serum for the times stated before fixation and antibody staining. Cells were lysed in 1% Triton X-100 (or Nonidet P-40), 50 mm Tris, pH 8, 150 mm NaCl, 1 mm MgCl2, 1 mm dithiothreitol, 10% glycerol, Complete protease inhibitors (Roche Pharmaceuticals), 1 mm sodium vanadate, and 50 mm NaF (cell lysis buffer). Electrophoresis was performed on a minigel apparatus unless otherwise indicated. For Western blotting, 100 μg of protein was generally used per lane, unless otherwise stated. The following commercial monoclonal antibodies were used: Mouse anti-GFP (Zymed Laboratories Inc.), anti-14-3-3 (H-8, Santa Cruz Biotechnology), and anti-HA (12CA5, Roche Molecular Biochemicals). A polyclonal GFP antiserum was purchased from CLONTECH. The TTP peptide antibodies nTTP and cTTP (31Johnson B.A. Geha M. Blackwell T.K. Oncogene. 2000; 19: 1657-1664Crossref PubMed Scopus (68) Google Scholar, 32Johnson, B. A., and Blackwell, T. K. (2002)Oncogene, in pressGoogle Scholar) were typically used at 1/5000 dilution. Immunofluorescence was carried out as described previously (31Johnson B.A. Geha M. Blackwell T.K. Oncogene. 2000; 19: 1657-1664Crossref PubMed Scopus (68) Google Scholar), using Cy3- or fluorescein isothiocyanate-conjugated secondaries (Jackson). 200–300 cells per slide were typically counted to determine TTP localization. Yeast two-hybrid screening and interaction assays were performed as described previously (34Vidal M. Bartel P.L. Fields S. The Yeast Two-hybrid System. Oxford University Press, Oxford1997: 109-147Google Scholar). For analysis of binding in vitro to glutathione S-transferase (GST)-14-3-3 fusion proteins, each 10 cm plate of cells was lysed in 100 μl of lysis buffer. 100 μg of protein was saved as an input control. 0.5–1 mg of lysate protein was incubated with 50 μl of a 1:1 bead slurry of GST-14-3-3 isoforms that had been synthesized as described (35Yaffe M.B. Rittinger K. Volinia S. Caron P.R. Aitken A. Leffers H. Gamblin S.J. Smerdon S.J. Cantley L.C. Cell. 1997; 91: 961-971Abstract Full Text Full Text PDF PubMed Scopus (1353) Google Scholar, 36Rittinger K. Budman J., Xu, J. Volinia S. Cantley L.C. Smerdon S.J. Gamblin S.J. Yaffe M.B. Mol Cell. 1999; 4: 153-166Abstract Full Text Full Text PDF PubMed Scopus (422) Google Scholar), in 1 ml of lysis buffer at 4 °C with rocking for 1 h, or overnight (Fig. 3 C). Beads were washed three times with phosphate-buffered saline containing 1% Triton X-100, then boiled in SDS loading buffer for 10 min before electrophoresis and Western blotting. Incubations with calf intestinal phosphatase were performed at 1 unit/10 μg of total protein in lysis buffer (without phosphatase inhibitors) for 30 min at 30 °C. For immunoprecipitation, cells were lysed in 150 μl of lysis buffer, with 3 μl of cleared lysate saved as input. After addition of either 2 μg of rabbit HA antibody (Y-11, Santa Cruz Biotechnology) or 5 μl of cTTP, the remaining cleared lysate was rocked at 4 °C for 1 h. Samples were spun for 5 min to remove precipitates, then incubated for 1 h with 20 μl of protein A beads (Santa Cruz Biotechnology) that had been preincubated in bovine serum albumin. For monoclonal anti-HA immunoprecipitations, 10 μl of antibody-conjugated beads (F-7, Santa Cruz Biotechnology) was added directly to the lysate for 2 h. Beads were washed three times in lysis buffer and boiled in SDS loading buffer for 10 min before electrophoresis and Western blotting. To identify proteins that interact with TTP, we performed a yeast two-hybrid screen of a mouse mixed-stage embryonic cDNA library using a full-length TTP bait. After isolating a full-length 14-3-3η cDNA from this screen, we determined that 14-3-3η bound strongly to TTP but not to various control baits, suggesting that this interaction was specific (Fig. 1 A). Mammals encode seven closely related 14-3-3 isoforms, each of ∼31 kDa (37Yaffe M.B. Elia A.E. Curr. Opin. Cell Biol. 2001; 13: 131-138Crossref PubMed Scopus (290) Google Scholar, 38Muslin A.J. Xing H. Cell. Signal. 2000; 12: 703-709Crossref PubMed Scopus (351) Google Scholar). 14-3-3 proteins bind to phosphorylated proteins, generally as dimers, and often bind to more than one site in the same protein (35Yaffe M.B. Rittinger K. Volinia S. Caron P.R. Aitken A. Leffers H. Gamblin S.J. Smerdon S.J. Cantley L.C. Cell. 1997; 91: 961-971Abstract Full Text Full Text PDF PubMed Scopus (1353) Google Scholar, 37Yaffe M.B. Elia A.E. Curr. Opin. Cell Biol. 2001; 13: 131-138Crossref PubMed Scopus (290) Google Scholar, 38Muslin A.J. Xing H. Cell. Signal. 2000; 12: 703-709Crossref PubMed Scopus (351) Google Scholar, 39Muslin A.J. Tanner J.W. Allen P.M. Shaw A.S. Cell. 1996; 84: 889-897Abstract Full Text Full Text PDF PubMed Scopus (1195) Google Scholar, 40Zeng Y. Piwnica-Worms H. Mol. Cell. Biol. 1999; 19: 7410-7419Crossref PubMed Scopus (134) Google Scholar). 14-3-3 binding influences the activity of several proteins, and localizes others to the cytoplasm in response to signals (37Yaffe M.B. Elia A.E. Curr. Opin. Cell Biol. 2001; 13: 131-138Crossref PubMed Scopus (290) Google Scholar, 38Muslin A.J. Xing H. Cell. Signal. 2000; 12: 703-709Crossref PubMed Scopus (351) Google Scholar). 14-3-3 proteins also act as signal-responsive anti-apoptotic factors by binding to phosphorylated forms of regulators such as the A20 protein, Forkhead-related transcription factors, the apoptosis-inducing kinase ASK1, and the BH3-only protein BAD. Because TTP is a phosphoprotein that induces apoptosis, and because its localization within the cell is influenced by extracellular stimuli and growth conditions (14Mahtani K.R. Brook M. Dean J.L. Sully G. Saklatvala J. Clark A.R. Mol. Cell. Biol. 2001; 21: 6461-6469Crossref PubMed Scopus (397) Google Scholar, 31Johnson B.A. Geha M. Blackwell T.K. Oncogene. 2000; 19: 1657-1664Crossref PubMed Scopus (68) Google Scholar, 41Taylor G.A. Thompson M.J. Lai W.S. Blackshear P.J. Mol. Endocrinol. 1996; 10: 140-146PubMed Google Scholar,42Taylor G.A. Thompson M.J. Lai W.S. Blackshear P.J. J. Biol. Chem. 1995; 270: 13341-13347Abstract Full Text Full Text PDF PubMed Scopus (113) Google Scholar), we chose to investigate the specificity and functional significance of TTP-14-3-3 binding. To test further the specificity of TTP-14-3-3 binding, we expressed TTP in 3T3 cells by transfection, and assayed whether it bound in vitro to a mixture of bacterially expressed 14-3-3 isoforms that were fused to GST. TTP bound robustly to the GST-14-3-3 protein mix but not to GST alone (Fig. 1 B, lanes 4–6). Similarly, fusion proteins in which TTP, TIS11b, and TIS11d were each linked at their N terminus to GFP bound comparably well to a GST-14-3-3η protein in vitro, whereas GFP alone did not bind, suggesting that binding to 14-3-3 is characteristic of all three TTP/TIS11 proteins (Fig. 1 C). TTP that was expressed in mammalian cells bound comparably well to fusion proteins that corresponded to each of the seven closely related mammalian 14-3-3 isoforms (Fig. 2, A and B; GST-14-3-3ε is not shown). These GST-linked 14-3-3 isoforms bound preferentially to distinct but overlapping sets of TTP species that appeared larger than the predicted TTP molecular mass of 34 kDa (Fig. 2 A, lanes 3–7). As reported previously (14Mahtani K.R. Brook M. Dean J.L. Sully G. Saklatvala J. Clark A.R. Mol. Cell. Biol. 2001; 21: 6461-6469Crossref PubMed Scopus (397) Google Scholar, 31Johnson B.A. Geha M. Blackwell T.K. Oncogene. 2000; 19: 1657-1664Crossref PubMed Scopus (68) Google Scholar, 43Carballo E. Cao H. Lai W.S. Kennington E.A. Campbell D. Blackshear P.J. J. Biol. Chem. 2001; 276: 42580-42587Abstract Full Text Full Text PDF PubMed Scopus (170) Google Scholar), treatment of transfected cell lysates with phosphatase converted these TTP species to a less heterogeneous group of faster-migrating forms, indicating that they represented different phosphorylated forms of TTP (Fig. 2 B, lanes 1 and 2). Interaction with 14-3-3 proteins was completely abrogated by dephosphorylation of TTP in these cell lysates (Fig. 2 B, lanes 3–10), indicating that binding of 14-3-3 proteins to TTP is phosphorylation-dependent. Phosphorylation-dependent 14-3-3 binding generally involves a conserved phosphoserine or phosphothreonine residue flanked by basic, aromatic, and aliphatic amino acids, along with additional serine and threonine residues (44van Hemert M.J. Steensma H.Y. van Heusden G.P. Bioessays. 2001; 23: 936-946Crossref PubMed Scopus (473) Google Scholar). Combinatorial screening using phosphoserine-oriented peptide libraries have identified two optimal classes of 14-3-3 binding consensus motifs, in which an arginine or lysine residue is preferred at either the −3 (Type I) or −4 (Type II) position relative to the phosphorylated residue, with nearby amino acids also being important (35Yaffe M.B. Rittinger K. Volinia S. Caron P.R. Aitken A. Leffers H. Gamblin S.J. Smerdon S.J. Cantley L.C. Cell. 1997; 91: 961-971Abstract Full Text Full Text PDF PubMed Scopus (1353) Google Scholar). Of these two consensus motifs, only Type I sites have been definitively identified previously as 14-3-3 targets in vivo. These screens have also indicated that individual 14-3-3 isoforms differ only subtly in their binding specificities. To identify TTP sequences that are required for 14-3-3 binding, we first analyzed a series of TTP deletion mutants (Fig. 3 A). In the yeast two-hybrid assay, the region of TTP located C-terminal to the zinc fingers was both necessary and sufficient for 14-3-3 binding (Fig. 3 B). This conclusion was supported by analysis of binding in vitro between a mixture of GST-fused 14-3-3 proteins, and GFP-tagged TTP mutants that had been expressed in 293T cells (Fig. 3 C). GFP-TTP and GFP-TTP(Zn-C) bound robustly to the GST-14-3-3 mix in this in vitro assay, even though they were expressed at the lowest relative levels (Fig. 3 C,lanes 2 and 6). GFP-TTP(C) also bound significantly to the GST-14-3-3 mix, but GFP and the other TTP mutants did not (Fig. 3 C, lanes 1, 3,4, and 5). Within the TTP C-terminal region, we identified four sequence elements that match the previously identified 14-3-3 binding consensus motifs with varying degrees of success (Fig. 3 D). Two of these elements are loosely conserved among all three TTP/TIS11 proteins (at TTP Ser-178 and Ser-206), and two are present only in TTP. To investigate whether these TTP elements are required for 14-3-3 binding, within each one we substituted alanine for the amino acid which 14-3-3 binding consensuses predict should be phosphorylated (Fig. 3 D). Binding of TTP to a 14-3-3 protein mixture was significantly reduced only by the TTP S178A mutation (Fig. 3 E, lane 6), which disrupts a predicted type II 14-3-3 binding site (35Yaffe M.B. Rittinger K. Volinia S. Caron P.R. Aitken A. Leffers H. Gamblin S.J. Smerdon S.J. Cantley L.C. Cell. 1997; 91: 961-971Abstract Full Text Full Text PDF PubMed Scopus (1353) Google Scholar). Because 14-3-3 often binds to pairs of sites in the same protein (38Muslin A.J. Xing H. Cell. Signal. 2000; 12: 703-709Crossref PubMed Scopus (351) Google Scholar), we similarly analyzed pairs of these Ala substitutions in all possible combinations, but none diminished binding compared with the corresponding single amino acid mutations (not shown). To determine whether TTP and 14-3-3 proteins interact specifically in vivo, we assayed for binding between TTP mutants and 14-3-3 proteins that were co-expressed in the cell line 293T, which produces anti-apoptotic adenovirus products and is relatively resistant to TTP-induced apoptosis (31Johnson B.A. Geha M. Blackwell T.K. Oncogene. 2000; 19: 1657-1664Crossref PubMed Scopus (68) Google Scholar). Supporting the findings shown in Fig. 3(B and C), in this assay both TTP and the TTP(Zn-C) mutant could be specifically co-immunoprecipitated along with HA-tagged 14-3-3β by the HA antibody (Fig. 4 A, lanes 1–4 and 7). The S178A substitution significantly reduced co-immunoprecipitation of 14-3-3β and TTP, and eliminated detectable binding of HA-14-3-3β to TTP(Zn-C) (Fig. 4 A, lanes 4, 5, 7, and 8). In parallel transfections, TTP was comparably co-immunoprecipitated by other HA-tagged 14-3-3 isoforms (not shown). To assay for in vivo binding between 14-3-3 and other TTP/TIS11 proteins, we co-expressed HA-14-3-3β along with GFP-tagged TTP, TIS11b, and TIS11d. Like GFP-TTP, GFP-TIS11b and GFP-TIS11d each co-immunoprecipitated with HA-14-3-3β (Fig. 4 B, lanes 6–9). Specific co-immunoprecipitation of GFP-TTP and HA-14-3-3β was reduced by the S178A mutation and eliminated by deletion of the TTP C-terminal region (Fig. 4 B, lanes 1–5). We assayed for binding between transfected TTP and endogenous 14-3-3 by using a TTP antibody to co-immunoprecipitate 14-3-3 (Fig. 4 C). When TTP was present, co-immunoprecipitation of endogenous 14-3-3 was elevated significantly over background (Fig. 4 C, lanes 1–3). This binding was abrogated by the S178A mutation (Fig. 4 C, lane 4), indicating that Ser-178 is important for binding to endogenous 14-3-3 proteins. 14-3-3 proteins inhibit apoptosis by binding to various pro-apoptotic proteins (37Yaffe M.B. Elia A.E. Curr. Opin. Cell Biol. 2001; 13: 131-138Crossref PubMed Scopus (290) Google Scholar, 38Muslin A.J. Xing H. Cell. Signal. 2000; 12: 703-709Crossref PubMed Scopus (351) Google Scholar), raising the question of whether TTP might cause apoptosis, in part, by titrating 14-3-3 proteins away from these anti-apoptotic interactions. To address this question, we first investigated whether the S178A substitution influenced the ability of TTP to induce apoptosis. TTP S178A induced apoptosis comparably to TTP over a range of input DNA concentrations (Fig. 5 B), despite its significantly reduced binding to 14-3-3 proteins (Figs. 3 E and 4). TTP and TTP S178A were also similarly capable of sensitizing cells to the apoptotic stimulus of TNFα (not shown). In addition, although TTP(Zn-C) S178A did not bind detectably to 14-3-3 (Fig. 4 A), it induced apoptosis comparably to TTP(Zn-C) (Fig. 5,A and B). The S178A substitution did not influence the levels in which either TTP or TTP(Zn-C) was expressed in these transfections (Fig. 4 A). Finally, simultaneous overexpression of 14-3-3 proteins did not attenuate TTP-induced apoptosis or differentially affect induction of apoptosis by either TTP or TTP S178A (not shown). These findings suggest that induction of apoptosis by TTP neither requires 14-3-3 binding at Ser-178 nor is mediated by TTP sequestering cellular pools of 14-3-3. Precedents set by various proteins suggest that 14-3-3 binding might influence how TTP is localized within the cell (37Yaffe M.B. Elia A.E. Curr. Opin. Cell Biol. 2001; 13: 131-138Crossref PubMed Scopus (290) Google Scholar, 38Muslin A.J. Xing H. Cell. Signal. 2000; 12: 703-709Crossref PubMed Scopus (351) Google Scholar). To test this model, we first investigated how 14-3-3 co-expression influences localization of TTP in HeLa cells. We used immunofluorescence to examine the subcellular localization o
Referência(s)