14-3-3 Proteins recognize a histone code at histone H3 and are required for transcriptional activation
2007; Springer Nature; Volume: 27; Issue: 1 Linguagem: Inglês
10.1038/sj.emboj.7601954
ISSN1460-2075
AutoresStefan Winter, Elisabeth Simboeck, Wolfgang Fischle, Gordin Zupkovitz, Ilse Dohnal, Karl Mechtler, Gustav Ammerer, Christian Seiser,
Tópico(s)Histone Deacetylase Inhibitors Research
ResumoArticle6 December 2007free access 14-3-3 Proteins recognize a histone code at histone H3 and are required for transcriptional activation Stefan Winter Stefan Winter Max F Perutz Laboratories, Vienna Biocenter, Medical University of Vienna, Vienna, Austria Search for more papers by this author Elisabeth Simboeck Elisabeth Simboeck Max F Perutz Laboratories, Vienna Biocenter, Medical University of Vienna, Vienna, Austria Search for more papers by this author Wolfgang Fischle Wolfgang Fischle Laboratory of Chromatin Biochemistry, Max Planck Institute for Biophysical Chemistry, Goettingen, Germany Search for more papers by this author Gordin Zupkovitz Gordin Zupkovitz Max F Perutz Laboratories, Vienna Biocenter, Medical University of Vienna, Vienna, Austria Search for more papers by this author Ilse Dohnal Ilse Dohnal Christian Doppler Laboratory for Proteome Analysis, Vienna, Austria Search for more papers by this author Karl Mechtler Karl Mechtler Research Institute of Molecular Pathology, Vienna Biocenter, Vienna, Austria Search for more papers by this author Gustav Ammerer Gustav Ammerer Christian Doppler Laboratory for Proteome Analysis, Vienna, Austria Search for more papers by this author Christian Seiser Corresponding Author Christian Seiser Max F Perutz Laboratories, Vienna Biocenter, Medical University of Vienna, Vienna, Austria Search for more papers by this author Stefan Winter Stefan Winter Max F Perutz Laboratories, Vienna Biocenter, Medical University of Vienna, Vienna, Austria Search for more papers by this author Elisabeth Simboeck Elisabeth Simboeck Max F Perutz Laboratories, Vienna Biocenter, Medical University of Vienna, Vienna, Austria Search for more papers by this author Wolfgang Fischle Wolfgang Fischle Laboratory of Chromatin Biochemistry, Max Planck Institute for Biophysical Chemistry, Goettingen, Germany Search for more papers by this author Gordin Zupkovitz Gordin Zupkovitz Max F Perutz Laboratories, Vienna Biocenter, Medical University of Vienna, Vienna, Austria Search for more papers by this author Ilse Dohnal Ilse Dohnal Christian Doppler Laboratory for Proteome Analysis, Vienna, Austria Search for more papers by this author Karl Mechtler Karl Mechtler Research Institute of Molecular Pathology, Vienna Biocenter, Vienna, Austria Search for more papers by this author Gustav Ammerer Gustav Ammerer Christian Doppler Laboratory for Proteome Analysis, Vienna, Austria Search for more papers by this author Christian Seiser Corresponding Author Christian Seiser Max F Perutz Laboratories, Vienna Biocenter, Medical University of Vienna, Vienna, Austria Search for more papers by this author Author Information Stefan Winter1, Elisabeth Simboeck1, Wolfgang Fischle2, Gordin Zupkovitz1, Ilse Dohnal3, Karl Mechtler4, Gustav Ammerer3 and Christian Seiser 1 1Max F Perutz Laboratories, Vienna Biocenter, Medical University of Vienna, Vienna, Austria 2Laboratory of Chromatin Biochemistry, Max Planck Institute for Biophysical Chemistry, Goettingen, Germany 3Christian Doppler Laboratory for Proteome Analysis, Vienna, Austria 4Research Institute of Molecular Pathology, Vienna Biocenter, Vienna, Austria *Corresponding author. Max F Perutz Laboratories, Vienna Biocenter, Medical University of Vienna, Dr Bohr-Gasse 9/2, Vienna 1030, Austria. Tel.: +431 4277 61770; Fax: +431 4277 9617; E-mail: [email protected] The EMBO Journal (2008)27:88-99https://doi.org/10.1038/sj.emboj.7601954 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Interphase phosphorylation of S10 at histone H3 is linked to transcriptional activation of a specific subset of mammalian genes like HDAC1. Recently, 14-3-3 proteins have been described as detectors for this phosphorylated histone H3 form. Here, we report that 14-3-3 binding is modulated by combinatorial modifications of histone H3. S10 phosphorylation is necessary for an interaction, but additional H3K9 or H3K14 acetylation increases the affinity of 14-3-3 for histone H3. Histone H3 phosphoacetylation occurs concomitant with K9 methylation in vivo, suggesting that histone phosphorylation and acetylation can synergize to overcome repressive histone methylation. Chromatin immunoprecipitation experiments reveal recruitment of 14-3-3 proteins to the HDAC1 gene in an H3S10ph-dependent manner. Recruitment of 14-3-3 to the promoter is enhanced by additional histone H3 acetylation and correlates with dissociation of the repressive binding module HP1γ. Finally, siRNA-mediated loss of 14-3-3 proteins abolishes the transcriptional activation of HDAC1. Together our data indicate that 14-3-3 proteins are crucial mediators of histone phosphoacetylation signals. Introduction The unstructured N-terminal tails of histone proteins are targeted by various different post-translational modifications (PTMs) like acetylation, methylation, phosphorylation or ADP ribosylation. These PTMs are critical factors in the regulation of local and global chromatin function and confer distinct properties to regions of the genome that finally modulate chromatin-associated processes such as transcription. It was suggested that the specific modification make-up constitutes a ‘histone code’, which is recognized via a ‘decoding machinery’ comprised by modification-dependent, chromatin-associated polypeptides (Strahl and Allis, 2000; Fischle et al, 2003). One particular PTM, the phosphorylation of histone H3 at S10, emerges in two distinct phases of the cell cycle, with considerable differences in dynamics and abundance (Johansen and Johansen, 2006; McManus and Hendzel, 2006). Global mitosis-specific histone H3 phosphorylation is mediated by the Aurora B kinase (Hsu et al, 2000) and is required for the displacement of HP1 proteins (Mateescu et al, 2004; Fischle et al, 2005; Hirota et al, 2005). During interphase, H3S10 phosphorylation is targeted to only a minute fraction of nucleosomes and is tightly linked to acetylation of H3K9 and H3K14 (phosphoacetylation) (Mahadevan et al, 1991; Cheung et al, 2000b; Clayton and Mahadevan, 2003). Histone H3 phosphoacetylation can be mediated by kinases MSK1/2 that are downstream of the ERK (p42/44) or the p38 mitogen-activated protein (MAP) kinase pathways and has been correlated with transcriptional induction of the immediate-early (IE) genes c-fos and c-jun (Clayton et al, 2000; Cheung et al, 2000b; Thomson et al, 2001; Clayton and Mahadevan, 2003; Soloaga et al, 2003; Mahadevan et al, 2004). The concept of gene activation by histone phosphorylation was extended to a variety of mammalian genes (Strelkov and Davie, 2002; Clayton and Mahadevan, 2003; Vicent et al, 2006). For instance, we reported the regulation of the HDAC1 gene by cooperative histone H3 phosphorylation and acetylation (Hauser et al, 2002). In contrast to the rapid and transient phosphoacetylation associated with IE gene activation, transcriptional induction of HDAC1 requires stable phosphoacetylation, which is achieved by stimulation of MAP kinase pathways and fine tuned via histone acetylation in an autoregulatory loop (Bartl et al, 1997; Schuettengruber et al, 2003). In addition, detector proteins that specifically recognize PTMs on histones play a key role in the regulation of chromatin-associated events. For example, the bromodomains of GCN5 and TFIID250 have been shown to specifically associate with acetylated histones, while chromodomain proteins exemplified by the heterochromatin protein 1 (HP1) interact with specific methylated forms (Lachner et al, 2003). Recently, 14-3-3 proteins, which are well-established phospho-serine adaptor molecules (Muslin et al, 1996; Yaffe et al, 1997), have been described as detectors for phosphorylated histone H3 (Macdonald et al, 2005). However, the role of this interaction in the context of transcription is unclear. Here, we report that interaction of 14-3-3ε and ζ with histone H3 is modulated by combinatorial PTM patterns. Binding of these proteins to phosphorylated H3S10 is stabilized by additional lysine acetylation. Phosphoacetylation of histone H3 at the HDAC1 promoter leads to the recruitment of 14-3-3 proteins concomitant with dissociation of HP1. As detector for specific active histone marks, we found that 14-3-3ζ is necessary for activation of the HDAC1 gene. Finally, we identify 14-3-3 as counterpart of the repressive binding module HP1, with reciprocal binding affinities for the modified histone H3 tail. Results Purification of 14-3-3 proteins as phosphoacetylated histone H3-binding factors Histone H3 phosphorylation and acetylation synergize in transcriptional activation of the late inducible HDAC1 gene (Hauser et al, 2002), implying that phosphoacetylation is a biologically relevant PTM pattern. Therefore, we asked whether the phosphoacetylation mark is recognized by specific cellular factors. To this end, differentially modified matrix-coupled histone H3 peptides (Table I) were incubated with nuclear extracts from HeLa cells that were either left untreated or treated with the p38–MAP kinase activator anisomycin and the HDAC inhibitor trichostatin A (TSA). This combinatorial treatment was previously shown to stimulate histone H3 phosphoacetylation and HDAC1 gene expression (Hauser et al, 2002). Two modification-dependent factors of approximately 30 and 27 kDa were found that specifically interacted with the S10phK14ac (ph/ac)–histone H3 peptide (Supplementary Figure S1). These proteins were identified as the epsilon (ε) and zeta (ζ) isoforms of the 14-3-3 protein family via mass spectrometry. The presence of the two isoforms in HeLa nuclear extracts was verified by immunoblotting with isoform-specific 14-3-3 antibodies (Supplementary Figure S1B). In in vitro binding assays, 14-3-3ζ extracted either from untreated or anisomycin/TSA treated HeLa cells bound equally well to S10phK14ac H3 peptide, suggesting that activation of the MAP kinase pathway or HDAC inhibition does not alter the affinity of 14-3-3 for the modified histone H3 tail (Supplementary Figure S1C). Although 14-3-3 proteins have been recently shown to interact with phosphorylated histone H3 (Macdonald et al, 2005), the importance of this interaction for gene regulation is not yet clarified. We therefore examined the role of 14-3-3 proteins as detectors for ph/ac histone H3 and studied their role in the activation of transcription. Table 1. Histone H3 peptides used in this study Peptide Sequence N–C um ARK STG GKA PRK QLC K14ac ARK STG GacKA PRK QLC S10ph ARK phSTG GKA PRK QLC K9ac/S10ph ARacK phSTG GKA PRK QLC S10ph/K14ac ARK phSTG GacKA PRK QLC K9me2/S10ph Arme2K phSTG GKA PRK QLC K9me2/S10ph/K14ac ARme2K phSTG GacKA PRK QLC H3(1–20) um ART KQT ARK STG GKA PRQ LC H3(1–20) S10ph ART KQT ARK phSTG GKA PRQ LC H3(1–20) K9ac/S10ph ART KQT ARacK phSTG GKA PRQ LC H3(1–20) S10ph/K14ac ART KQT ARK phSTG GacKA PRQ LC H3(1–20) K9me2/S10ph ART KQT AR me2K phSTG GKA PRQ LC H3(1–20) K9me2/S10ph/ K14ac ART QT AR me2K phSTG GacKA PRQ LC H3 um (25–38) ARK SAP ATG GVK KPC H3 S28ph (25–38) ARK phSAP ATG GVK KPC acK, acetylated lysine; me2K, dimethylated lysine; pS, phosphoserine. 14-3-3 Proteins interact with histone H3 in a modification-dependent manner To verify that 14-3-3 proteins bind indeed to ph/ac histone H3, we performed GST pull-down assays with histones isolated from 3T3 mouse fibroblasts. To avoid mitotic S10 phosphorylation, we used resting cells, which show only low levels of ph/ac histone H3 and cells that were simultaneously treated with anisomycin and TSA to stimulate H3 phosphoacetylation (Figure 1A, panel i, lanes 1 and 2). After incubation with GST–14-3-3ζ or GST as control, bound histones were analyzed by immunoblotting. Probing of the blot with antibodies specific for S10phK14ac H3 revealed that this modification form could interact with GST–14-3-3ζ but not with GST (Figure 1A, panel i). The total amount of H3 associated with 14-3-3 proteins was increased upon TSA/anisomycin treatment as displayed with modification independent C-terminal H3 antibodies (Figure 1A, panel ii). A similar in vitro interaction was also found for 14-3-3ε (data not shown). These results indicate that 14-3-3ζ and ε bind to histone H3 in a modification-dependent manner. Figure 1.14-3-3 Binding to histone H3 is dependent on H3 phosphorylation and stabilized by additional acetylation. (A) Induction of phosphoacetylation increases histone H3 interaction with 14-3-3. Histones were isolated from resting 3T3 fibroblasts that were either left untreated (0) or stimulated for 1 h with anisomycin and TSA (A/T) and incubated with GST or GST–14-3-3ζ. Bound histones were analyzed by immunoblotting with antibodies against ph/ac histone H3 (panel i) and C-terminal histone H3 (H3 C-term) (panel ii). Loading of GST and GST–14-3-3 was monitored by Ponceau staining (panel iii). (B) In vitro modification of histone H3. Recombinant histone H3 was phosphorylated by MSK1 (lane2), acetylated by PCAF (lane 4) or phosphoacetylated with both enzymes (lane 3). Enzymes were omitted in control reactions (lane 1). The modification status was analyzed by sequential immunoblotting with the indicated antibodies. Corresponding modifications are denoted at the top. (C) Acetylation effects on the 14-3-3/histone H3 interaction are more dominant for the R23A28 mutant than for the A10R14 mutant. The indicated histone H3 mutants were in vitro modified as indicated and incubated with GST–14-3-3ζ proteins. Bound histone H3 proteins were analyzed by immunoblotting with C-terminal histone H3 antibodies. The panel shows one representative experiment for each mutant or WT histone H3. Download figure Download PowerPoint Additional acetylation stabilizes the interaction between S10 phosphorylated histone H3 and 14-3-3 proteins Histone proteins extracted from mammalian cells may carry in addition to phosphorylation and acetylation various other PTMs. To utilize a more defined set of modifications, we modified recombinant histone H3 in vitro. Phosphorylation or acetylation reactions were performed using MSK1 (Figure 1B, panel iii, lane 2) or the histone acetyltransferase PCAF, respectively (Figure 1B, panel ii, lane 4). Initial phosphorylation and subsequent acetylation reactions generated ph/ac histone H3 (Figure 1B, panel i, lane 3). As control, enzymes were omitted from reactions (Figure 1B, lane 1) and total amounts of H3 in the different modification reactions were visualized with the C-terminal H3 antibodies (Figure 1B, panel iv). The interaction of in vitro modified H3 with 14-3-3ζ was analyzed in GST pull-down assays. As expected, phosphorylation led to association with GST–14-3-3ζ (Figure 1C, panel i), whereas acetylation by PCAF alone did not mediate any binding (Supplementary Figure S2C, and data not shown). Strikingly, 14-3-3ζ binding was stronger for phosphoacetylated than for phosphorylated H3, indicating that in the context of S10 phosphorylation acetylation exerts a stabilizing effect (Figure 1C, panel i). Mass spectrometry analysis of MSK1-modified histone H3 revealed that not only S10 but also S28 was phosphorylated (Supplementary Figure S2D). 14-3-3 Proteins were previously shown to interact not only with H3S10ph but also with H3S28ph peptides (Macdonald et al, 2005). Furthermore, acetylation of the neighboring K23 residue was reported (Daujat et al, 2002). Therefore, we performed binding assays with different mutated histone H3 proteins. The efficiency of phosphorylation by MSK1 and acetylation by PCAF was monitored by immunoblot analysis and kinase assays with γ-32P-ATP (Supplementary Figures S2A and B). Mutation of either S10 in combination with K14 or S28 and K23 led to 55–60% reduction in 32P incorporation, while the quadruple mutant displayed about 20% residual phosphorylation (Supplementary Figure S2B). According to the mass spectrometry analysis, histones H3 becomes phosphorylated at T45 and S57 in the absence of both serines. Surprisingly, a K9R mutation significantly increased 32P incorporation by MSK1 and therefore these mutants were omitted from further analysis. Pull-down assays with histone H3 bearing K23R/S28A double mutations (R23/A28) revealed an increased interaction of 14-3-3ζ with the phosphoacetylated than to the phosphorylated form (Figure 1C, panel ii). Mass spectrometry analysis confirmed that this mutant was predominantly phosphorylated on S10. In contrast, 14-3-3ζ binding to the S10A/K14R mutant (A10/R14) was similar in the presence and absence of acetylation (Figure 1C, panel iii), and loss of both serines in the quadruple mutant resulted in only weak interaction with 14-3-3ζ. Taken together, these data indicate that the binding affinity of 14-3-3ζ for S10-phosphorylated histone H3 is significantly enhanced by additional lysine acetylation. Since combinatorial binding of 14-3-3 to target factors has not been reported, we decided to investigate this effect in more detail. Interphase phosphorylation of histone H3 occurs in the context of additional PTMs The efficiency of phosphorylation-mediated binding of 14-3-3 to target proteins is strongly dependent on the amino-acid environment around the phosphorylated residue (Yaffe et al, 1997; Uchida et al, 2006). As a basis for studying the impact of additional modifications on the interaction with 14-3-3, we determined PTM patterns present on S10-phosphorylated N-terminal tails of histone H3 via a mass spectrometry approach. As already mentioned, interphase phosphorylation of histone H3 affects only a small sub-fraction of all nucleosomes (Barratt et al, 1994). Our mass spectrometry analysis clearly indicates the presence of various different modifications like lysine methylation and acetylation in addition to H3S10ph (Table II). In agreement with previously published data (Dyson et al, 2005), we observed some residual histone H3 phosphorylation in samples derived from resting and untreated cells. Table 2. PTMs on the S10-phosphorylated tryptic histone H3 peptide K9STGGKAPR17 Condition Peptide sequence MH+ LTQ-FT LTQ Resting R.KpSTGGKAPR.K 1093.5 0/3 1/1 R.me1KpSTGGKAPR.K 1107.556 1/3 0/1 R.me2KpSTGGKAPR.K 1065.545 2/3 0/1 R.me3KpSTGGKAPR.K 1079.561 2/3 1/1 R.KpSTGGacKAPR.K 1079.524 1/3 0/1 R.acKpSTGGme3KAPR.K 1065.545 1/3 0/1 R.me3KpSTGGacKAPR.K sAn R.KpSTGGKAPR.K 1093.540 1/3 1/1 R.me1KpSTGGKAPR.K 1107.556 2/3 1/1 R.me2KpSTGGKAPR.K 1065.545 3/3 1/1 R.me3KpSTGGKAPR.K 1079.561 3/3 1/1 R.KpSTGGacKAPR.K 1079.525 2/3 1/1 R.me1KpSTGGacKAPR.K 1093.540 1/3 0/1 R.me2KpSTGGKacAPR.K 1051.530 1/3 1/1 R.acKpSTGGme3KAPR.K R.me3KpSTGGacKAPR.K 1065.5 0/3 1/1 R.me3KpSTGGme3KAPR.K SAn/TSA R.KpSTGGKAPR.K 1093.540 1/3 1/1 R.me1KpSTGGKAPR.K 1107.556 1/3 0/1 R.me2KpSTGGKAPR.K 1065.545 2/3 1/1 R.me3KpSTGGKAPR.K 1079.561 3/3 1/1 R.KpSTGGacKAPR.K 1079.524 2/3 0/1 R.acKpSTGGacKAPR.K 1065.509 2/3 1/1 R.me1KpSTGGacKAPR.K 1093.540 3/3 1/1 R.me2KpSTGGacKAPR.K 1051.530 2/3 1/1 R.acKpSTGGme3KAPR.K 1065.545 1/3 1/1 R.me3KpSTGGacKAPR.K acK, acetylated lysine; LTQ-FT/LTQ, the value indicates the frequency of peptide recovery either on the LTQ-FTICR hybrid instrument or the LTQ mass spectrometer (e.g., 2/3 meaning two times out of three experiments); me1K, monomethylated lysine; me2K, dimethylated lysine; me3K, trimethylated lysine; MH+, mono-protonated mass; pS, phosphorylated serine; PTM, post-translational modification; sAn, anisomycin; TSA, trichostatin A. Vertical lines indicate one peptide species where modifications were not unequivocally assigned to a particular position. Anisomycin treatment increased the S10ph histone H3 pool, as well as the complexity of modification patterns (Table II). Besides single phosphorylated histone H3 and H3K9me1/2/3/S10ph forms, we identified a phosphoacetylated species with the acetyl moiety at position 14 (S10phK14ac). In addition, we identified a triple modified H3 peptide with the modification status K9me2S10phK14ac. This form is particular interesting as it carries an active and a repressive modification in addition to H3S10ph. Additional treatment with TSA led to further changes in the phospho-form composition and gave also rise to a K9K14 diacetylated ph/ac form (K9acS10phK14ac). While TSA treatment had no effect on overall S10 phosphorylation (Supplementary Figure S3A, panel i, lanes 5 and 6 and Supplementary Figure S3B), the abundance of the S10phK14ac epitope was almost doubled (Supplementary Figure S3A, panel i, compare lanes 2 and 3). These results suggest that p38 MAP kinase activation leads to the formation of several different phospho-histone H3 forms and the composition of this pool is altered by additional TSA treatment. Taken together, interphase H3S10 phosphorylation occurs as single modification, but frequently coincides with lysine methylation and acetylation on the same histone H3 tail. Additional histone modifications affect the 14-3-3/histone H3 interaction To investigate the impact of lysine acetylation and methylation on the interaction between 14-3-3 and histone H3, we performed in vitro peptide pull-down assays. This experimental setup also allowed us to use a homogenously modified system for the interaction studies. Differentially modified histone H3 peptides were synthesized on the basis of the mass spectrometry results (Table I). Since we are interested in the role of H3S10ph during transcriptional activation, we focused on modifications that are known to prevalently reside in euchromatin and excluded H3K9me3, the archetype of heterochromatic histone modifications. Equal amounts of the differentially modified immobilized H3 peptides were incubated with in vitro translated (IVT) 14-3-3ζ protein. Phosphorylation of H3S10 was required for significant interaction with 14-3-3 (Figure 2A, lane 2), whereas only slight background signals were observed for the unmodified or the H3K14ac peptide (Figure 2A, lane 1, and data not shown). Figure 2.Modulation of the histone H3/14-3-3 interaction by additional modifications. (A) Histone H3/14-3-3 interaction is modulated by additional lysine acetylation. IVT 35S-methionine-labeled 14-3-3ζ was incubated with differentially modified gel-coupled histone H3 peptides. Bound proteins were analyzed by SDS–PAGE and fluorography. The panel shows one representative experiment. The signal intensity for each band was quantified and is depicted as summary of five independent measurements (mean±s.d.). Values were normalized relative to H3S10ph peptide-bound fraction (lane 2). Additional acetylation increased the association with 14-3-3 proteins (lane 4, *P=0.001, t-test). A similar effect of H3K14 acetylation was observed for the H3K9me2/S10ph peptide (lane 7, **P<0.001, t-test). (B) Nuclear 14-3-3 proteins preferentially bind to the S10phK14ac histone H3 peptide. Nuclear extracts were incubated with unmodified, S10ph or S10phK14ac H3 peptides. An aliquot of the nuclear extracts was used as input control. Bound proteins were analyzed on immunoblots with 14-3-3ζ antibodies. (C) Additional K14 acetylation increases the competitor potential of the S10ph histone H3 peptide. Binding reactions were performed as described for panel A (lanes 1–3). In addition, binding reactions on the ph/ac peptide were performed in the presence of a 20-fold molar excess of unmodified, H3S10ph or H3S10phK14ac free competitor peptides. Bound 14-3-3ζ proteins were analyzed by SDS–PAGE and fluorography. Each signal was normalized to the non-competed H3S10phK14ac peptide binding (lane 3) and is depicted as histogram showing the average of three independent experiments (mean±s.d.). Download figure Download PowerPoint Acetylation of H3K9 caused a moderate but reproducible reinforcement of the interaction (Figure 2A, lane 3). Remarkably, additional acetylation of H3K14 strongly increased the interaction, supporting the results from the GST pull-down experiments (Figure 2A, lane 4 and Figure 1C). To rule out the possibility that the increased binding of 14-3-3ζ to the H3S10phK14ac peptide is a unique property of IVT or recombinant proteins, we confirmed this effect with endogenous 14-3-3 present in HeLa nuclear extracts (Figure 2B). In contrast, binding to the H3K9me2S10ph peptide was slightly reduced compared with the S10ph peptide (Figure 2A, compare lanes 2 and 5). Our mass spectrometry analysis revealed the presence of a K9me2S10phK14ac histone H3 form (Table II). Binding studies with the corresponding triple modified peptide (K9me2S10phK14ac) demonstrated that additional acetylation of K14 also increased binding of 14-3-3ζ to the phosphomethylated H3 peptide (Figure 2A, lane 6). To further confirm the stabilizing effect exerted by H3K14 acetylation, we performed competition assays using the H3S10phK14ac peptide as bait and free unmodified, H3S10ph and H3S10phK14ac peptides as competitors (Figure 2C, lanes 4–6). Addition of the H3S10ph peptide reduced binding of 14-3-3ζ to the H3S10phK14ac peptide to approximately 65% compared with non-competed binding, whereas the unmodified peptide had no effect (Figure 2C, lanes 4 and 5). Importantly, the H3S10phK14ac peptide was found to be a much more potent competitor with an average reduction of binding to about 15% of non-competed assays (Figure 2C, lane 6). These data also demonstrate that the binding of 14-3-3ζ to the phosphoacetylated histone H3 tail is highly dynamic and reversible. Together, our data suggest that significant binding of 14-3-3 to the histone H3 tail requires initial phosphorylation but is susceptible to the presence of additional PTMs. To quantify the binding of 14-3-3 to various combinations of H3 modifications, we determined dissociation constants of the interactions. Therefore, we performed fluorescence polarization measurements using recombinant 14-3-3 in combination with fluorescinated, differentially modified H3 peptides (Figure 3). In concordance with the peptide pull-down assays, we detected strong interaction of 14-3-3 with the histone H3 peptide upon phosphorylation of S10, whereas only very weak interaction with the unmodified peptide was observed. The H3K9me2S10ph peptide was bound with comparable strength as the H3S10ph peptide, suggesting that K9me2 does not significantly impair the binding of 14-3-3. The affinity of 14-3-3 for the H3S10ph and the H3K9me2S10ph peptide was further enhanced by additional acetylation. This effect was slightly more pronounced for K14ac (Kd=49 μM for K9ac/S10ph and Kd=35 μM for S10ph/K14ac). The H3S28ph peptide was bound with much higher initial affinity than the H3S10ph peptide (Kd=30 μM), which may possibly be attributed to the proline at position 30 that can also be found in one of the high-affinity 14-3-3-binding motifs RSXSpXP, where a proline is located at position n+2 from the phosphorylated serine (Yaffe et al, 1997). Importantly, similar results were also obtained with 14-3-3ζ without GST-tag and 14-3-3ε (Figure 3B). Figure 3.Binding of 14-3-3 to an H3S10ph peptide is enhanced by additional lysine acetylation. (A) Lysine acetylation increases the affinity of 14-3-3 for the phosphorylated H3 peptide. Binding of 14-3-3ζ to the indicated H3 peptides was analyzed by fluorescence polarization measurements. The panel shows the average of at least three independent measurements (mean±s.d.). (B) Dissociation constants (Kd in μM) for the interaction of different 14-3-3 isoforms with the indicated histone H3 peptides determined by fluorescence polarization measurements. Values are average (mean±s.d.) of at least three independent measurements. Download figure Download PowerPoint In conclusion, our biochemical and biophysical studies indicate a function of the double and triple modified histone H3 forms in the recruitment of 14-3-3 proteins: interaction between histone H3 and 14-3-3 is mediated by S10 phosphorylation and acetylation of K9 or K14 significantly increases the affinity of 14-3-3 for the histone H3 tail. 14-3-3 Proteins are recruited to the promoter region of the HDAC1 gene by histone H3 phosphoacetylation Next, we sought to determine whether 14-3-3 proteins associate with the HDAC1 gene in a histone H3 phosphoacetylation-dependent manner. As shown previously (Hauser et al, 2002), HDAC1 expression in resting 3T3 fibroblasts was low but could be efficiently stimulated by combinatorial treatment with anisomycin and TSA (Figure 4A). In contrast, anisomycin alone did not induce HDAC1 transcription and TSA had an intermediate effect. A possible recruitment of 14-3-3 proteins to the HDAC1 promoter region was investigated by ChIP assays of resting and stimulated 3T3 fibroblasts (Figure 4B; Supplementary Figure S4B). Histone H3 phosphoacetylation was absent from the HDAC1 promoter region in resting cells, both in the presence and absence of TSA (Figure 4B, panel iv). Anisomycin treatment moderately elevated the levels of ph/ac histone H3, whereas additional treatment with TSA led to high levels of phosphoacetylation that are linked to transcriptional induction of the HDAC1 gene (Figure 4A and B; Hauser et al, 2002). Interestingly, the recruitment of 14-3-3ζ to the HDAC1 promoter was in strong correlation with the levels of ph/ac histone H3 (Figure 4B, panel v, lanes 1–4). In contrast, H3 phosphoacetylation and 14-3-3 recruitment were not observed at control genes such as β-actin (Figure 4B, panel v, lanes 5–8) or histone H4 (data not shown). As transcriptional activation can be accompanied by a reduction of nucleosome density within the proximal promoter region (Yuan et al, 2005), ChIP assays with C-terminal histone H
Referência(s)