C-terminal Recognition by 14-3-3 Proteins for Surface Expression of Membrane Receptors
2005; Elsevier BV; Volume: 280; Issue: 43 Linguagem: Inglês
10.1074/jbc.m507559200
ISSN1083-351X
AutoresBrian Coblitz, Sojin Shikano, Meng Wu, Sandra B. Gabelli, Lisa M. Cockrell, Matt Spieker, Yoshiro Hanyu, Haian Fu, L. Mario Amzel, Man Li,
Tópico(s)Microbial Natural Products and Biosynthesis
ResumoDiverse functions of 14-3-3 proteins are directly coupled to their ability to interact with targeted peptide substrates. RSX(pS/pT)XP and RXΦX(pS/pT)XP are two canonical consensus binding motifs for 14-3-3 proteins representing the two common binding modes, modes I and II, between 14-3-3 and internal peptides. Using a genetic selection, we have screened a random peptide library and identified a group of C-terminal motifs, termed SWTY, capable of overriding an endoplasmic reticulum localization signal and redirecting membrane proteins to cell surface. Here we report that the C-terminal SWTY motif, although different from mode I and II consensus, binds tightly to 14-3-3 proteins with a dissociation constant (KD) of 0.17 μm, comparable with that of internal canonical binding peptides. We show that all residues but proline in -SWTX-COOH are compatible for the interaction and surface expression. Because SWTY-like sequences have been found in native proteins, these results support a broad significance of 14-3-3 interaction with protein C termini. The C-terminal binding consensus, mode III, represents an expansion of the repertoire of 14-3-3-targeted sequences. Diverse functions of 14-3-3 proteins are directly coupled to their ability to interact with targeted peptide substrates. RSX(pS/pT)XP and RXΦX(pS/pT)XP are two canonical consensus binding motifs for 14-3-3 proteins representing the two common binding modes, modes I and II, between 14-3-3 and internal peptides. Using a genetic selection, we have screened a random peptide library and identified a group of C-terminal motifs, termed SWTY, capable of overriding an endoplasmic reticulum localization signal and redirecting membrane proteins to cell surface. Here we report that the C-terminal SWTY motif, although different from mode I and II consensus, binds tightly to 14-3-3 proteins with a dissociation constant (KD) of 0.17 μm, comparable with that of internal canonical binding peptides. We show that all residues but proline in -SWTX-COOH are compatible for the interaction and surface expression. Because SWTY-like sequences have been found in native proteins, these results support a broad significance of 14-3-3 interaction with protein C termini. The C-terminal binding consensus, mode III, represents an expansion of the repertoire of 14-3-3-targeted sequences. The 14-3-3 proteins were initially identified from brain for their abundance and acidic properties (1Moore B.W. Perez V.J. Physiological and Biochemical Aspects of Nervous Integration. Prentice-Hall, Englewood Cliffs, NJ1967: 343-359Google Scholar). They have been found in all eukaryotes and recognized for their diverse functional roles in biological processes including metabolism, cell signaling, intracellular trafficking, stress responses, cell cycle progression, and malignant transformation (see reviews in Refs. 2Fu H. Subramanian R.R. Masters S.C. Annu. Rev. Pharmacol. Toxicol. 2000; 40: 617-647Crossref PubMed Scopus (1328) Google Scholar, 3Yaffe M.B. FEBS Lett. 2002; 513: 53-57Crossref PubMed Scopus (557) Google Scholar, 4Dougherty M.K. Morrison D.K. J. Cell Sci. 2004; 117: 1875-1884Crossref PubMed Scopus (396) Google Scholar). Multiple isoforms of 14-3-3 are commonly found in a given species. Saccharomyces cerevisiae has two 14-3-3 isoforms, encoded by BMH1 and BMH2. Deletion of either does not cause growth defects, but the double deletion of both is lethal in most genetic backgrounds (5Roberts R.L. Mosch H.U. Fink G.R. Cell. 1997; 89: 1055-1065Abstract Full Text Full Text PDF PubMed Scopus (185) Google Scholar, 6van Heusden G.P. Griffiths D.J. Ford J.C. Chin A.W.T.F. Schrader P.A. Carr A.M. Steensma H.Y. Eur. J. Biochem. 1995; 229: 45-53Crossref PubMed Scopus (140) Google Scholar, 7Gelperin D. Weigle J. Nelson K. Roseboom P. Irie K. Matsumoto K. Lemmon S. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 11539-11543Crossref PubMed Scopus (146) Google Scholar). Essential to 14-3-3 function, in most cases, is the ability to bind a short peptide substrate only upon phosphorylation. As the first protein module identified with preferential affinity to phosphorylated Ser/Thr substrates, 14-3-3 binding provides an inducible mechanism of translating enzymatic activity to protein-protein interaction, a molecular feature with broad significance in regulation. Structural studies and examination of the interactive sequences in human proto-oncogene Raf-1 and from random phosphopeptide libraries have established two binding consensuses, RSX(pS/pT)XP and RXΦX(pS/pT)XP, which are also known as mode I and II recognition (8Yaffe 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 (1348) Google Scholar, 9Muslin A.J. Tanner J.W. Allen P.M. Shaw A.S. Cell. 1996; 84: 889-897Abstract Full Text Full Text PDF PubMed Scopus (1188) Google Scholar). The two binding modes have provided molecular insights into the recognition of 14-3-3 binding to a large number of interacting proteins. The vast majority of binding sites identified so far lies internally and matches either mode I or II binding consensus. In addition to phosphorylation-mediated induction of 14-3-3 binding, there is evidence for additional mechanisms of interaction. For example, a tight association between 14-3-3 and multiple regions of plant plasma membrane H+-ATPase is induced by additional binding of a fungal toxin, fusicoccin, which leads to the formation of a ternary complex (10Wurtele M. Jelich-Ottmann C. Wittinghofer A. Oecking C. EMBO J. 2003; 22: 987-994Crossref PubMed Scopus (272) Google Scholar, 11Fuglsang A.T. Borch J. Bych K. Jahn T.P. Roepstorff P. Palmgren M.G. J. Biol. Chem. 2003; 278: 42266-42272Abstract Full Text Full Text PDF PubMed Scopus (85) Google Scholar). More recently, evidence from studying arylalkylamine N-acetyltransferase (AANAT), 2The abbreviations used are:AANATarylalkylamine N-acetyltransferaseERendoplasmic reticulumHAinfluenza A virus hemagglutininDansyl5-dimethylamino-1-naphthalenesulfonylPBSphosphate-buffered salineAbantibodyFAM5-(and 6)-carboxyfluoresceinFCMflow cytometry the penultimate enzyme in melatonin synthesis, has shown that AANAT has two 14-3-3 binding sites; one high affinity site resides internally, and one low affinity resides at the C terminus. The nonsaturated binary binding of 14-3-3 to these sites was suggested as a tuning mechanism for rhythmic enzyme activity that is coordinated with the daily cycle of melatonin production (12Ganguly S. Weller J.L. Ho A. Chemineau P. Malpaux B. Klein D.C. Proc. Natl. Acad. Sci. U. S. A. 2005; 102: 1222-1227Crossref PubMed Scopus (182) Google Scholar). Both of the above cases involve a low affinity interaction between 14-3-3 and the C terminus: QSYTV-COOH (plant H+-ATPase) and RRNSDR-COOH (AANAT). These sequences are distinct and could not have been predicted by the consensus, RSX(pS/pT)XP and RXΦX(pS/pT)XP, in mode I and II binding. The affinity (KD) for QSYpTV is 2.5 μm, strengthened to 0.027 μm upon ternary binding with fusicoccin (10Wurtele M. Jelich-Ottmann C. Wittinghofer A. Oecking C. EMBO J. 2003; 22: 987-994Crossref PubMed Scopus (272) Google Scholar). The affinity for RRNpSDR is considerably lower than that of the internal canonical binding sequence, and its KD value remains to be determined (12Ganguly S. Weller J.L. Ho A. Chemineau P. Malpaux B. Klein D.C. Proc. Natl. Acad. Sci. U. S. A. 2005; 102: 1222-1227Crossref PubMed Scopus (182) Google Scholar). This mode of interaction is sufficiently different and thus was recently named mode III (12Ganguly S. Weller J.L. Ho A. Chemineau P. Malpaux B. Klein D.C. Proc. Natl. Acad. Sci. U. S. A. 2005; 102: 1222-1227Crossref PubMed Scopus (182) Google Scholar). It is not well understood whether the interacting C terminus could function independently and whether the reduced affinity is caused by the lack of the conserved proline residue. Information concerning their binding affinity, sequence requirements, and possible selectivity to interact with different 14-3-3 isoforms remain unknown. arylalkylamine N-acetyltransferase endoplasmic reticulum influenza A virus hemagglutinin 5-dimethylamino-1-naphthalenesulfonyl phosphate-buffered saline antibody 5-(and 6)-carboxyfluorescein flow cytometry We have performed a genetic screen of random peptides to identify C-terminal signal motifs that override endoplasmic reticulum (ER) localization to confer protein expression on cell surface (13Shikano S. Coblitz B. Sun H. Li M. Nat. Cell Biol. 2005; DOI: 10.1038/ncbl1297PubMed Google Scholar). Among the identified motifs is a group of homologous C-terminal peptides that function through an interaction with 14-3-3. Here we report a series of studies to examine the interaction between 14-3-3 and synthetic C-terminal peptides and the ability of the interacting peptides to confer receptor surface expression. The resultant C-terminal peptide consensus expands the repertoire of 14-3-3 recognition motifs and facilitates further functional studies. Molecular Biology—The yeast expression vector for mammalian Kir2.1 was generated by cloning mouse Kir2.1 cDNA in pADNS vector at HindIII and NotI (13Shikano S. Coblitz B. Sun H. Li M. Nat. Cell Biol. 2005; DOI: 10.1038/ncbl1297PubMed Google Scholar). The PstI site was created at the C terminus of Kir2.1 by replacing the last residue with the PstI sequence (LQ). The DNA fragment encoding the C-terminal 36 amino acids of mouse Kir6.2 (LLDALTLASSRGPLRKRSVAVAKAKPKFSISPDSLS) was generated by annealing and extending the complementary oligonucleotides with PstI and NotI overhangs and then ligating to Kir2.1. The SacI site was engineered 9 amino acids downstream of the RKR signal (this caused mutation of PK to EL) to replace SISPDSLS with the DNA fragments encoding the X8 random peptide library. The X8 library DNA fragments were generated by annealing and extending the sense strand (5′-GAGCTCTTT(NNK)8TAGGCGGCCGCTACATACA, where N indicates any of A, T, C, or G and K indicates either of G or T) and antisense strand (5′-TGTATGTAGCGGCCGCCTA) oligonucleotides that were synthesized to randomly encode 8 amino acids preceded by Phe with SacI and NotI overhangs. To clone the CD8 fusion vectors, the cDNA encoding the full-length human CD8 sequence was isolated for fusion with the HA epitope and subcloned into the HindIII-BamHI sites of pCDNA3.1(+) (Invitrogen). The C-terminal 36-amino acid sequence of Kir6.2 was cloned from yeast Kir2.1 fusion vector by PCR with BamHI and EcoRI overhangs and ligated into CD8HA vector at BamHI and EcoRI. A SfiI site was engineered immediately after the SacI site in the 36-amino acid Kir6.2 sequence (LLDALTLASSRGPLRKRSVAVAKAKELFSISPDSLS), causing a substitution of ELF with ELGHLGQ. This SfiI site facilitated replacement of SISPDSLS with the DNA fragments flanking with SfiI and XhoI sites. Mutagenesis of the SWTY sequence was performed by insertion between SfiI and XhoI sites, following extension by Klenow enzyme of oligonucleotides including mutations. Transfection—The HEK293 cells were transiently transfected using FuGENE 6 (Roche Applied Sciences) and were analyzed at 24-36 h after transfection. Immunocytochemistry—HEK293 cells plated on polylysine-treated coverslips were transiently transfected. The cells were fixed in 4% paraformaldehyde, PBS for 15 min and permeabilized in 0.05% Triton X-100, PBS for 5 min, all at room temperature. Kir2.1 was detected with mouse anti-HA Ab (Santa Cruz Biotechnologies, Santa Cruz, CA), followed by Alexa Fluor 488-conjugated goat anti-mouse IgG (Molecular Probes, Eugene, OR). The ER marker protein, calreticulin, was detected by rabbit anti-calreticulin Ab (Affinity BioReagents, Golden, CO), followed by Alexa Fluo 568-conjugated goat anti-rabbit IgG (Molecular Probes). The coverslips were mounted on glass slides with Vectashield (Vector Labs, Burlingame, CA), and the images were captured using confocal microscopy. Spinning Nipkow disk confocal microscopy was performed with an UltraView LCI System (PerkinElmer Life Sciences) in conjunction with an Eclipse 200 inverted microscope (Nikon, Melville, NY). Imaging through a 60× oil immersion planar apochromatic lens (numerical aperture, 1.4) was accomplished via excitation using the 488- or 568-nm lines of a krypton/argon laser. Flow cytometry—The transfected HEK293 cells were harvested by incubation with 0.5 mm EDTA, PBS for 10 min at 37 °C and washed with Hanks' balanced salt solution supplemented with 5 mm HEPES (pH 7.3) and 2% fetal bovine serum (staining medium). All of the incubation and washing steps were performed in staining medium at 4 °C. Surface CD8 was detected with mouse anti-human CD8 monoclonal Ab (Santa Cruz Biotechnologies), followed by Alexa Fluor 488-conjugated goat anti-mouse IgG. The stained cells were examined for surface expression with FACSCalibur (BD Biosciences, San Jose, CA). Peptide Synthesis and Affinity Measurement—The Dansyl and 5-(and 6)-carboxyfluorescein (FAM)-labeled peptides, Dansyl-RGRSWpTY-COOH, and FAM-RGRSWpTY-COOH, as well as control sequences as RGRSWpTY-COOH, RGRSWTY-COOH, RGRSWpTE-COOH, RGRSWpTD-COOH, RGRSWpTP-COOH, and RGRSWpTG-COOH, were synthesized (New England Peptides, Boston, MA; Abgent, San Diego, CA). The peptide R18, with a sequence of PHCVPRDLSWLDLEANMCLP, was purchased from Biomol International (Plymouth Meeting, PA). The peptides were prepared as stock solutions by dissolving in water and, if necessary, with the addition of minimal amount of acetonitrile. The working solutions for the affinity detection were prepared in the 10 mm HEPES buffer (pH 7.3). For detection of both the fluorescence intensity and the fluorescence anisotropy, a multifunctional plate reader Safire2 (Tecan US, Research Triangle Park, NC) was used. For fluorescence intensity-based affinity detections with Dansyl-RGRSWpTY-COOH, the excitation was set at 335 nm and emission was set at 545 nm with 20-nm bandwidth for both excitation and emission. The integration time was set at 200 μs with 3 reads/well. The detection was performed on 96-well black microtiter plates with transparent bottoms in top measurement mode. For the fluorescence anisotropy-based affinity measurements with FAM-RGRSWpTY-COOH, the detection was performed with 384-well black microtiter plates in fluorescence polarization measurement mode. The excitation was set at 470 nm and emission was set at 525 nm with 20-nm bandwidth for emission. The time between move and flash was set at 100 ms with 3 reads/well. G-Factor was determined as 1.1 by calibration of 100 nm FAM-RGRSWpTY-COOH as 20 mP for florescence polarity, with buffer solution as the reference. For fluorescence intensity-based affinity detections with Dansyl-RGRSWpTY-COOH, the KD of Dansyl-RGRSWpTY-COOH with 14-3-3ζ was determined by the titration of 1.2 μm of Dansyl-RGRSWpTY-COOH in 100 μl of total volume with different concentrations of 14-3-3ζ, with one binding site fitting with Equation 1 using Origin 7.0, with y = F - F0, where F is the fluorescence intensity of Dansyl-RGRSWpTY-COOH with the addition of 14-3-3ζ, F0 is that of Dansyl-RGRSWpTY-COOH alone, and x is the concentration of 14-3-3ζ. Competitive experiments were designed for the KD measurements of unlabeled control peptides (14Harris B. Hillier B. Lim W. Biochemistry. 2001; 40: 5921-5930Crossref PubMed Scopus (107) Google Scholar). The titrations with different concentrations of Dansyl-RGRSWpTY-COOH were performed with 25 μm of control sequences and 2.65 μm 14-3-3ζ in 100 μl total volume. The fitting was done using Origin 7.0, with the (Eq. 2), (Eq. 3), (Eq. 4), with y = F - F0, where F is the fluorescence intensity after the addition of Dansyl-RGRSWpTY-COOH to the mixture of control sequences and 14-3-3ζ, F0 is that of Dansyl-RGRSWpTY-COOH alone in another well without 14-3-3ζ, and x is the concentration of Dansyl-RGRSWpTY. For fluorescence anisotropy-based affinity detections with FAM-RGRSWpTY-COOH, the KD value of FAM-RGRSWpTY-COOH with 14-3-3ζ was determined by the titration of 100 nm of FAM-RGRSWpTY-COOH in 100 μl total volume with different concentrations of 14-3-3ζ, with one binding site fitting with Equation 1 using Origin 7.0. In the fitting, y = r - r0, where r is the fluorescence anisotropy of FAM-RGRSWpTY-COOH with the addition of 14-3-3ζ, r0 is FAM-RGRSWpTY-COOH alone, and x is the concentration of 14-3-3ζ. For the KD measurements of unlabeled peptides, competitive anisotropy experiments were performed with the mixture of different concentrations of control sequences with 80 nm FAM-RGRSWpTY-COOH following the addition of 0.65 μm 14-3-3ζ in 100 μl total volume. The fitting was done using Origin 7.0, with Equation 4, and y = (r - r0)/(rmax - r0), where r is the fluorescence anisotropy of FAM-RGRSWpTY-COOH with the addition of 14-3-3ζ, r0 is that of FAM-RGRSWpTY-COOH alone, rmax is the maximum of fluorescence anisotropy of FAM-RGRSWpTY-COOH with the addition of 14-3-3ζ, D0 is the concentration of FAM-RGRSWpTY-COOH, P0 is the concentration of 14-3-3ζ, and x is the concentrations of control peptides. y=BxKDDansylSWpTY+X(Eq. 1) y=BxKDApp+x(Eq. 2) KDApp=(1+[Control]0KDControl)KDDansylSWpTY(Eq. 3) y=12[D0]{(KDFAM−SWpTYKDControlx+KDFAM−SWpTY+[D0]+[P0])−(KDFAM−SWpTYKDControlx+KDFAM−SWpTY+[D0]+[P0]2−4[D0][P0]}(Eq. 4) Immunoprecipitation and Immunoblot—For immunoprecipitation, transfected cells were washed with PBS once and lysed with lysis buffer (1% Nonidet P-40, 25 mm Tris, 150 mm NaCl, pH 7.50) with protease inhibitor cocktails for 20 min at 4 °C. After spinning for 20 min at 11,000 × g, the supernatant was mixed with protein A-conjugated agarose beads that were preincubated with 1 μg of anti-HA Ab (Santa Cruz Biotechnologies). After 5 h of incubation, the beads were washed five times with lysis buffer and then boiled with 2× sample buffer for SDS-PAGE analysis. The samples resolved in SDS-PAGE gels were transferred to nitrocellulose and blotted with anti-14-3-3 Ab (Zymed Laboratories Inc., South San Francisco, CA) or anti-HA Ab (Santa Cruz Biotechnologies), followed by horseradish peroxidase-conjugated secondary antibodies. The immunoblots were developed with the ECL system (Amersham Biosciences). Expression and Purification of Recombinant 14-3-3 Proteins in Escherichia coli—Seven isoforms of 14-3-3 proteins were expressed as hexahistidine-tagged fusions from a T7-driven promoter and purified from E. coli strain BL21-SI (Invitrogen). Briefly, E. coli BL21-SI cells harboring a 14-3-3 expression vector were grown overnight in salt free LB medium at 30 °C, which was used to inoculate (1%) in a fresh salt free LB medium. When the cell density reached an A600 of 0.5, NaCl was added to the final concentration of 0.3 m. The expression of 14-3-3s was induced for 5 h before harvesting. Recombinant His-tagged 14-3-3 proteins were purified using nickel-chelating ion exchange affinity chromatography essentially as previously described (15Zhang L. Wang H. Liu D. Liddington R. Fu H. J. Biol. Chem. 1997; 272: 13717-13724Abstract Full Text Full Text PDF PubMed Scopus (132) Google Scholar). Before use, 14-3-3 proteins were subject to gel filtration chromatography using a PD-10 column (Amersham Biosciences) and eluted in HEPES buffer (10 mm, pH 7.3, 130 mm NaCl). Pull-down of 14-3-3 by Peptides—Purified His-14-3-3ζ proteins (wild type, K49E, and V176D) were used for the pull-down studies. Specifically, 10 μm synthetic biotin-conjugated peptides for FRGRSWpTY, FRGRSWTY, and FRGRSWAY (Abgent, San Diego, CA) were incubated with 13 μm purified 14-3-3 for 30 min at room temperature. The bound material was mixed with streptavidin-Sepharose beads and gently agitated for 1 h. After washing four times with 5% Nonidet P-40 in PBS-T, the beads were resuspended in SDS loading buffer, boiled, and separated by SDS-PAGE followed by Coomassie staining. SWTX-bound 14-3-3 Homology Modeling—The co-crystal structure of 14-3-3 bound to ARSHpSYPA (a mode I 14-3-3 binding motif) (16Rittinger 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 (421) Google Scholar) was chosen to model the SWTY motif configuration in the 14-3-3-binding pocket. The electrostatic surface of the 14-3-3 structure (1QJB) was calculated with the program Pymol (by Warren L. DeLano), as well as the ray tracing. Using the programs O (17Jones T.A. Zou J.Y. Cowan S.W. Kjeldgaard M. Acta Crystallogr. Sect. A. 1991; 47: 110-119Crossref PubMed Scopus (13011) Google Scholar, 18Jones T.A. Kjeldgaard M. Methods Enzymol. 1997; 277: 173-208Crossref PubMed Scopus (504) Google Scholar) and Quanta (Accelrys, Inc. San Diego, CA), we modeled the sequence of the SWTY motif, RGRSWpTY-COOH, using the mode I peptide amino acids as a template. A radius of 6 Å around the SWpTY peptide was used to manage the contacts. The position of the sequence was anchored by the location of the phosphorylated serine (or threonine) residues in the mode I and the SWTY motif. Homology modeling was repeated for the SWTX panel using the same method. Affinity Determination of C-terminal Peptides—A typical ER localization signal is capable of dominantly conferring a surface-expressed protein to ER (19Jackson M.R. Nilsson T. Peterson P.A. J. Cell Biol. 1993; 121: 317-333Crossref PubMed Scopus (314) Google Scholar). In a Kir2.1 potassium channel, FCYENE, a cytoplasmic signal, potentiates protein surface expression (20Ma D. Jan L.Y. Curr. Opin. Neurobiol. 2002; 12: 287-292Crossref PubMed Scopus (152) Google Scholar). Addition of the RKR ER localization signal to Kir2.1 prevents surface expression (21Shikano S. Li M. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 5783-5788Crossref PubMed Scopus (89) Google Scholar). To isolate sequences with ability of overriding the RKR signal and redirecting ER localized proteins to the cell surface, we have constructed a reporter in which constitutive surface expression of Kir2.1 potassium channel protein was abolished by the addition of an RKR ER localization signal (21Shikano S. Li M. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 5783-5788Crossref PubMed Scopus (89) Google Scholar, 22Zerangue N. Schwappach B. Jan Y.N. Jan L.Y. Neuron. 1999; 22: 537-548Abstract Full Text Full Text PDF PubMed Scopus (902) Google Scholar). Selection of a random 8-mer peptide C-terminal to the RKR in a yeast genetic system has permitted the isolation of sequences capable of reconferring surface expression by a growth selection (13Shikano S. Coblitz B. Sun H. Li M. Nat. Cell Biol. 2005; DOI: 10.1038/ncbl1297PubMed Google Scholar). In mammalian cells, Kir2.1-RKR was found in ER, co-localized with calreticulin (Fig. 1B, left panels). The weak signal on the surface may represent ER compartments in proximity to the cell surface and some Kir2.1 protein escaped from ER localization activity. In contrast, fusion of these sequences to the C terminus, such as GHGRRHSW of clone 95, confers a significant increase of surface expression (Fig. 1B, right panels). For the SWTY motif, the ratio of surface/total protein increased by 4-fold compared with that of the wild type (13Shikano S. Coblitz B. Sun H. Li M. Nat. Cell Biol. 2005; DOI: 10.1038/ncbl1297PubMed Google Scholar). We have shown that one group of such sequences achieves their effect via interaction with 14-3-3, and their sequences have several conserved features such as threonine or serine at the -2 position amenable to phosphorylation (Fig. 1C) (13Shikano S. Coblitz B. Sun H. Li M. Nat. Cell Biol. 2005; DOI: 10.1038/ncbl1297PubMed Google Scholar). The difference in peptide length was caused by the introduction of a stop codon in the random 8-mer peptide library (see "Experimental Procedures"). The binding between RGRSWpTY-COOH and 14-3-3 has been detected by immunoprecipitation and affinity pull-down using biotinylated synthetic peptide (13Shikano S. Coblitz B. Sun H. Li M. Nat. Cell Biol. 2005; DOI: 10.1038/ncbl1297PubMed Google Scholar). However, like many studies, because of the dual binding site of dimeric 14-3-3 and involvement of immobilization of peptide ligands on solid surfaces, these binding studies are not most suitable for accurate evaluation of interaction affinity (see "Discussion"). To demonstrate direct and specific interaction with 14-3-3, we developed a fluorescence-based solution binding assay more suitable for affinity quantification with minimal contribution of avidity. Based on the known structures of 14-3-3, it appears that the binding of the mode I consensus, which SWTY more closely resembles, is no longer selective at the -4 position N-terminal to the phosphorylated residue (16Rittinger 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 (421) Google Scholar). In addition, the shortest motif isolated from our screen has five residues (Fig. 1C, #135). We reasoned that an N-terminal attachment of a fluorophore to the -5 position should be sufficiently distant from affecting the affinity of a binding peptide. Binding of a peptide to 14-3-3 brings the fluorophore to the protein surface and likely geometrically restrains the tethered fluorophore, potentially causing changes of fluorescence intensity and/or polarization (measured as anisotropy). We synthesized and tested FAM-RGRSWpTY-COOH and 5-dimethylaminonaphthalene-1-sulfonyl Dansyl-RGRSWpTY-COOH (Fig. 2A). For the Dansyl-RGRSWpTY, binding to purified recombinant 14-3-3ζ potentiates the fluorescence at 545 nm by more than 30% (Fig. 2B). No significant fluorescence signal was detect for 14-3-3 protein alone, either wild type or point mutants (Fig. 3C and data not shown). For the FAM-RGRSWpTY peptide, we observed a more than 6-fold increase in fluorescence polarization (Fig. 2C). These parameters provide a means to monitor the 14-3-3 binding in solution and the ability to compare measurements obtained with two different fluorescence probes (see "Experimental Procedures").FIGURE 3Characterization of SWTY motif binding to 14-3-3ζ. A, comparison of affinity for association between Dansyl-RGRSWpTY peptide and 14-3-3ζ (closed circles) or the mutant 14-3-3ζ (K49E) (open circles). The change of fluorescence intensity (ΔF, vertical axis) is plotted against concentration of 14-3-3ζ (horizontal axis). B, comparison of binding affinity among different 14-3-3 isoforms. The ΔF values for the 14-3-3 isoforms were individually normalized to their maximums to calculate the bound fraction (vertical axis) and plotted against concentration of all seven human 14-3-3 isoforms. C, competition of 14-3-3ζ binding by Dansyl-RGRSWpTY-COOH. 14-3-3ζ was incubated with Dansyl-RGRSWpTY peptide in the presence of increasing concentrations of competitor peptides, RGRSWpTY (closed circles) and RGRSWTY (open circles). The reduction of ΔF values was individually normalized to their maximums to identify the bound fraction. D, affinity-precipitation 14-3-3ζ proteins with synthetic peptides. Biotinylated SWTY, SWpTY, and SWAY peptides were individually incubated with the wild type and K49E and V176D mutants as indicated. Input 14-3-3 proteins are shown in lanes 1, 5, and 9, respectively. The bound materials were precipitated by streptavidin protein A-Sepharose beads and visualized by 12% SDS-PAGE followed by Coomassie staining. The molecular mass standards are as indicated on the left in kDa.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Fig. 2B shows the dose-dependent increase of fluorescence intensity for Dansyl-RGRSWpTY-COOH binding to the purified wild type 14-3-3ζ. When such an increase is fit to a 1:1 bimolecular binding reaction, it gives rise to a KD of 0.55 ± 0.12 μm between Dansyl-RGRSWpTY and 14-3-3ζ. Using FAM-RGRSWpTY-COOH peptide and fluorescence anisotropy, we obtained a similar KD value of 0.49 ± 0.14 μm (Fig. 2C). The KD values in solution are comparable with that of canonical peptide substrates (8Yaffe 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 (1348) Google Scholar). Although it is unknown whether and to what extent the conjugated fluorophores may affect the binding affinity, the similarity of dissociation constants for the two different fluorophore-conjugated peptides is consistent with the notion of little contribution of fluorophore chemistry to the KD values (see "Discussion"). C-terminal Peptide Binding Requires the Canonical Binding Groove in 14-3-3—Based on the structure of 14-3-3 with a bound canonical peptide, the Lys49 residue forms a critical part of the binding groove. A specific K49E mutation abolished the binding (15Zhang L. Wang H. Liu D. Liddington R. Fu H. J. Biol. Chem. 1997; 272: 13717-13724Abstract Full Text Full Text PDF PubMed Scopus (132) Google Scholar). Fig. 3A shows the change of fluorescence with increasing concentrations of purified 14-3-3ζ. Consistent with a specific association with the known 14-3-3 binding groove, an increase of fluorescence intensity was observed from wild type 14-3-3ζ but not for the mutant 14-3-3ζ (K49E). There are seven isoforms of 14-3-3 found in human with the potential for isoform-specific functions (23Aitken A. Baxter H. Dubois T. Clokie S. Mackie S. Mitchell K. Peden A. Zemlickova E. Biochem. Soc. Trans. 2002; 30: 351-360Crossref PubMed Scopus (146) Google Scholar). To examine any isoform selectivity for the SWpTY-mediated interaction with 14-3-3, we expressed and purified each of the seven isoforms from E. coli (see "Experimental Procedures"). Under similar experimental conditions, the seven isoforms exhibited similar bindin
Referência(s)