p53 Latency
2001; Elsevier BV; Volume: 276; Issue: 19 Linguagem: Inglês
10.1074/jbc.m100482200
ISSN1083-351X
AutoresT. V. Yakovleva, Aladdin Pramanik, Takashi Kawasaki, Koichi Tan‐No, I. P. Gileva, Heléne Lindegren, Ülo Langel, Tomas J. Ekström, Rudolf Rigler, Lars Terenius, Georgy Bakalkin,
Tópico(s)RNA modifications and cancer
ResumoThe p53 transcription factor is either latent or activated through multi-site phosphorylation and acetylation of the negative regulatory region in its C-terminal domain (CTD). How CTD modifications activate p53 binding to target DNA sequences via its core domain is still unknown. It has been proposed that nonmodified CTD interacts either with the core domain or with DNA preventing binding of the core domain to DNA and that the fragments of the CTD regulatory region activate p53 by interfering with these interactions. We here characterized the sequence and target specificity of p53 activation by CTD fragments, interaction of activating peptides with p53 and target DNA, and interactions of "latent" p53 with DNA by a band shift assay and by fluorescence correlation spectroscopy. In addition to CTD fragments, several long basic peptides activated p53 and also transcription factor YY1. These peptides and CTD aggregated target DNA but apparently did not interact with p53. The potency to aggregate DNA correlated with the ability to activate p53, suggesting that p53 binds to target sequences upon interactions with tightly packed DNA in aggregates. Latent full-length p53 dissociated DNA aggregates via its core and CTD, and this effect was potentiated by GTP. Latent p53 also formed complexes via both its core and CTD with long nontarget DNA molecules. Such p53-DNA interactions may occur if latent p53 binding to DNA via CTD prevents the interaction of the core domain with target DNA sites but not with nonspecific DNA sequences. The p53 transcription factor is either latent or activated through multi-site phosphorylation and acetylation of the negative regulatory region in its C-terminal domain (CTD). How CTD modifications activate p53 binding to target DNA sequences via its core domain is still unknown. It has been proposed that nonmodified CTD interacts either with the core domain or with DNA preventing binding of the core domain to DNA and that the fragments of the CTD regulatory region activate p53 by interfering with these interactions. We here characterized the sequence and target specificity of p53 activation by CTD fragments, interaction of activating peptides with p53 and target DNA, and interactions of "latent" p53 with DNA by a band shift assay and by fluorescence correlation spectroscopy. In addition to CTD fragments, several long basic peptides activated p53 and also transcription factor YY1. These peptides and CTD aggregated target DNA but apparently did not interact with p53. The potency to aggregate DNA correlated with the ability to activate p53, suggesting that p53 binds to target sequences upon interactions with tightly packed DNA in aggregates. Latent full-length p53 dissociated DNA aggregates via its core and CTD, and this effect was potentiated by GTP. Latent p53 also formed complexes via both its core and CTD with long nontarget DNA molecules. Such p53-DNA interactions may occur if latent p53 binding to DNA via CTD prevents the interaction of the core domain with target DNA sites but not with nonspecific DNA sequences. C-terminal domain fluorescence correlation spectroscopy big dynorphin N-terminal domain rhodamine tetramethylrhodamine-big dynorphin single-stranded double-stranded double-stranded specific oligonucleotide double-stranded 5-carboxytetramethylrhodamine oligonucleotide with wild type p53 consensus binding sites double-stranded 5-carboxytetramethylrhodamine oligonucleotide mutant p53 consensus binding sites guanylyl-imidodiphosphate glutathione S-transferase high pressure liquid chromatography base pair(s) The p53 tumor suppressor protein is a latent transcription factor that is activated by various forms of cellular stress including DNA damage. Activation of p53 binding to target DNA sequences and, consequently, p53-dependent transcription is controlled by the negative regulatory region in the CTD of p53, which includes amino acids 361–382. Post-translational modifications of amino acid residues in this region through phosphorylation activate sequence-specific DNA binding. Acetylation of several C-terminal lysines by p300/CBP/PCAF, recruited through phosphorylation of the distant N-terminal serines, critically regulates the site-specific DNA binding function of p53 (for reviews see Refs. 1Albrechtsen N. Dornreiter I. Grosse F. Kim E. Wiesmüller L. Deppert W. Oncogene. 1999; 18: 7706-7717Crossref PubMed Scopus (151) Google Scholar, 2Meek D.W. Oncogene. 1999; 18: 7666-7675Crossref PubMed Scopus (208) Google Scholar, 3Oren M. J. Biol. Chem. 1999; 274: 36031-36034Abstract Full Text Full Text PDF PubMed Scopus (491) Google Scholar). p53 exists in vitro in a latent form that cannot bind to p53-responsive DNA elements (4Hupp T.R. Sparks A. Lane D.P. Cell. 1995; 83: 237-245Abstract Full Text PDF PubMed Scopus (448) Google Scholar). Deletion of the critical regulatory region in the 30-amino acid C-terminal domain (CTD)1 results in p53 activation for sequence-specific DNA binding. Furthermore, binding of the monoclonal antibody PAb421 to this region activates p53 sequence-specific DNA binding and triggers the transcriptional activity of p53 in vivo (4Hupp T.R. Sparks A. Lane D.P. Cell. 1995; 83: 237-245Abstract Full Text PDF PubMed Scopus (448) Google Scholar). The mechanism of the CTD-mediated inhibition of p53 sequence-specific DNA binding is not understood. It has been proposed that by intramolecular interaction with the p53 DNA-binding core, CTD allosterically locks the p53 molecule in a state latent for binding to DNA (4Hupp T.R. Sparks A. Lane D.P. Cell. 1995; 83: 237-245Abstract Full Text PDF PubMed Scopus (448) Google Scholar). In accordance with this model, CTD or CTD fragments including p53(361–382) can displace CTD from its binding site within the central domain, resulting in a change in p53 conformation and activation of p53 binding to DNA (4Hupp T.R. Sparks A. Lane D.P. Cell. 1995; 83: 237-245Abstract Full Text PDF PubMed Scopus (448) Google Scholar). The CTD fragments can bind to the isolated core domain in vitro (5Selivanova G. Ryabchenko L. Jansson E. Iotsova V. Wiman K.G. Mol. Cell. Biol. 1999; 19: 3395-3402Crossref PubMed Scopus (130) Google Scholar, 6Kim A.L. Raffo A.J. Brandt-Rauf P.W. Pincus M.R. Monaco R. Abarzua P. Fine R.L. J. Biol. Chem. 1999; 274: 34924-34931Abstract Full Text Full Text PDF PubMed Scopus (111) Google Scholar). Activation of latent p53 by peptides derived from CTD has been reported to be sequence-specific because neither mutant p53(361–382) peptides nor peptides derived from the N-terminal p53 domain activate p53 or demonstrate only weak activity (4Hupp T.R. Sparks A. Lane D.P. Cell. 1995; 83: 237-245Abstract Full Text PDF PubMed Scopus (448) Google Scholar, 5Selivanova G. Ryabchenko L. Jansson E. Iotsova V. Wiman K.G. Mol. Cell. Biol. 1999; 19: 3395-3402Crossref PubMed Scopus (130) Google Scholar, 7Selivanova G. Iotsova V. Okan I. Fritsche M. Storm M. Groner B. Crafstorm R.C. Wilman K.G. Nat. Med. 1997; 3: 632-638Crossref PubMed Scopus (318) Google Scholar). CTD also shows target specificity because it did not influence the interaction of SRF and GAL4-VP16 proteins with DNA (8Jayaraman L. Prives C. Cell. 1995; 81: 1021-1029Abstract Full Text PDF PubMed Scopus (353) Google Scholar). The overlap of the regulatory region in CTD, which is rich in basic amino acids, with the second DNA-binding site in the p53 molecule is the key feature of the steric model (9Bayle J.H. Elenbaas B. Levine A.J. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 5729-5733Crossref PubMed Scopus (115) Google Scholar, 10Anderson M.E. Woelker B. Reed M. Wang P. Tegtmeyer P. Mol. Cell. Biol. 1997; 17: 6255-6264Crossref PubMed Scopus (105) Google Scholar). This model postulates that p53 binds to genomic DNA via CTD, and this binding prevents the interaction of the core domain with target sequences in gene promoters. The steric model is supported by the fact that long nonspecific DNA molecules inhibit sequence-specific binding of the full-length p53 to DNA but shows no effect on p53 with deletion of the CTD regulatory sequence (9Bayle J.H. Elenbaas B. Levine A.J. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 5729-5733Crossref PubMed Scopus (115) Google Scholar, 10Anderson M.E. Woelker B. Reed M. Wang P. Tegtmeyer P. Mol. Cell. Biol. 1997; 17: 6255-6264Crossref PubMed Scopus (105) Google Scholar). To gain insight into the mechanism of p53 latency, the sequence and target specificity of p53 activation by CTD fragments and the interactions of activating peptides with p53 and DNA were characterized. A conventional band shift assay was used to characterize p53 binding to DNA, whereas fluorescence correlation spectroscopy (FCS) was applied to follow the interactions of activating peptides with p53 and DNA. The high sensitivity of FCS (Fig.1 and Refs. 11Eigen M. Rigler R. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 5740-5747Crossref PubMed Scopus (908) Google Scholar, 12Rigler R. J. Biotechnol. 1995; 41: 177-186Crossref PubMed Scopus (205) Google Scholar, 13Rigler R. Pramanik A. Jonasson P. Kratz G. Jansson O.T. Nygren P.-Å. Ståhl S. Ekberg K. Johansson B.-L. Uhlén S. Uhlén M. Jörnvall H. Wahren J. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 13318-13323Crossref PubMed Scopus (311) Google Scholar, 14Tjernberg L. Pramanik A. Björling S. Thyberg P. Thyberg J. Nordstedt C. Terenius L. Rigler R. Chem. Biol. 1999; 6: 53-62Abstract Full Text PDF PubMed Scopus (139) Google Scholar, 15Wohland T. Friedrich K. Hovius R. Vogel H. Biochemistry. 1999; 38: 8671-8681Crossref PubMed Scopus (117) Google Scholar) allows analysis of such interactions and their nature even if they are weak in a reaction mixture. DNA molecules unexpectedly formed aggregates in the presence of the CTD fragment and other activating peptides. Our findings suggest that p53 is activated for binding to target DNA sites when it interacts with juxtaposed DNA molecules in DNA aggregates via both its core and CTD. Thus, CTD, when it binds to DNA, may prevent the interaction of the core domain of the full-length p53 with target but not with nontarget DNA sequences. This hypothesis seems to unify the previously postulated steric and allosteric models. The plasmid encoding human wild type p53, GST-human wild type p53 fusion protein, the deletion fusion proteins GST-p53(1–100), GST-p53(99–307), GST-p53(320–393), and GST-p53Δ30, the mutant fusion proteins GST-p53His273 and GST-p53Trp248 have been described elsewhere (5Selivanova G. Ryabchenko L. Jansson E. Iotsova V. Wiman K.G. Mol. Cell. Biol. 1999; 19: 3395-3402Crossref PubMed Scopus (130) Google Scholar, 16Bakalkin G. Yakovleva T. Selivanova G. Magnusson K.P. Szekely L. Kiseleva E. Klein G. Terenius L. Wiman K.G. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 413-417Crossref PubMed Scopus (280) Google Scholar, 17Bakalkin G. Selivanova G. Yakovleva T. Kiseleva E. Kashuba E. Magnusson K.P. Szekely L. Klein G. Terenius L. Wiman K.G. Nucleic Acids Res. 1995; 23: 362-369Crossref PubMed Scopus (157) Google Scholar, 18Selivanova G. Iotsova V. Kiseleva E. Strom M. Bakalkin G. Grafstrom R.C. Wiman K.G. Nucleic Acids Res. 1996; 24: 3560-3567Crossref PubMed Scopus (65) Google Scholar). The PG and MG fragments with 13 copies of consensus and 15 copies of mutant sites, respectively, were obtained by digestion of the PG-CAT and MG-CAT plasmids with HindIII/EcoRI. The RD fragment was cut from the pRC/CMV-Dyn plasmid. Plasmid DNA and oligonucleotides were purified by the Qiagen kit (Hilden, Germany) and polyacrylamide-urea gel electrophoresis, respectively; their concentrations were determined from their absorbance at 260 nm, and their extinction coefficients were calculated from nucleotide composition. p53 proteins were produced in Escherichia coli, purified as outlined earlier, and quantitated by Coomassie Blue or silver staining following SDS-polyacrylamide gel electrophoresis with bovine serum albumin as a standard (17Bakalkin G. Selivanova G. Yakovleva T. Kiseleva E. Kashuba E. Magnusson K.P. Szekely L. Klein G. Terenius L. Wiman K.G. Nucleic Acids Res. 1995; 23: 362-369Crossref PubMed Scopus (157) Google Scholar, 18Selivanova G. Iotsova V. Kiseleva E. Strom M. Bakalkin G. Grafstrom R.C. Wiman K.G. Nucleic Acids Res. 1996; 24: 3560-3567Crossref PubMed Scopus (65) Google Scholar). The basic 32-amino acid fragment of human prodynorphin (prodynorphin 207–238 or big dynorphin (BD), YGGFLRRIRPKLKWDNQKRYGGFLRRQFKVVT), dynorphins A (BD1–17) and B (BD20–32) were obtained from Phoenix Pharmaceuticals Inc. (Mountain View, CA), and penta-l-lysine and poly-l-lysine were purchased from Bachem (Bubendorf, Switzerland). Eight CTD fragments (22-mers) spanning CTD residues 337–393 with a 14-residue overlap including p53(361–382) (GSRAHSSHLKSKKGQSTRHKK) were described earlier (18Selivanova G. Iotsova V. Kiseleva E. Strom M. Bakalkin G. Grafstrom R.C. Wiman K.G. Nucleic Acids Res. 1996; 24: 3560-3567Crossref PubMed Scopus (65) Google Scholar). Binding of p53 to specific sequences was performed as described earlier (4Hupp T.R. Sparks A. Lane D.P. Cell. 1995; 83: 237-245Abstract Full Text PDF PubMed Scopus (448) Google Scholar, 18Selivanova G. Iotsova V. Kiseleva E. Strom M. Bakalkin G. Grafstrom R.C. Wiman K.G. Nucleic Acids Res. 1996; 24: 3560-3567Crossref PubMed Scopus (65) Google Scholar) in the presence of 2 mm MgCl2. 1 nm32P-labeled 30-mer BC oligonucleotide containing five adjacent copies of p53 consensus pentamer binding site and GST-p53 protein (35 nm of monomer) were incubated in the absence or in the presence of 25 nm of nonspecific competitor poly(dI-dC). p53 proteins binding to long 32P-end-labeled DNA was described earlier (17Bakalkin G. Selivanova G. Yakovleva T. Kiseleva E. Kashuba E. Magnusson K.P. Szekely L. Klein G. Terenius L. Wiman K.G. Nucleic Acids Res. 1995; 23: 362-369Crossref PubMed Scopus (157) Google Scholar). AP-1, NF-κB, YY1, and protein Ku were assayed in nuclear extracts from SH-SY5Y cells and YY1 with purified protein as described earlier (19Bakalkin G. Yakovleva T. Terenius L. Biochem. Biophys. Res. Commun. 1996; 231: 135-139Crossref Scopus (11) Google Scholar, 20Bakalkin G. Yakovleva T. Melzig M. Terenius L. Neuroreport. 1997; 8: 2143-2148PubMed Google Scholar). BD with the C-terminal Cys-amide extension was synthesized and labeled with tetramethylrhodamine-5-iodoacetamide (Molecular Probes Europe BV) according to the manufacturer's instructions. The Rh-BD was isolated on an analytical Nucleosil 120–3 C18 reverse phase HPLC column (Macherey-Nagel) and lyophilized, and the powder was weighed and used for preparation of solutions. The identity of the purified labeled peptide was confirmed by plasma desorption mass spectrometry. The purity of the peptide was > 90%. 5-Carboxytetramethylrhodamine (Rh′)-labeled ((+)-strands) and unlabeled ((−)-strands), HPLC-purified DNA 50-mer oligonucleotides were purchased from TIB-MOLBIOL (Berlin, Germany). The oligonucleotides represent the wild type and mutant BC 30-mer oligonucleotides, extended from 5′ and 3′ ends, and contain six adjacent consensus pentamer wild type or mutant (mutant nucleotides are shown in italic type) p53 binding sites (underlined; the sequences for the (+)-strands are shown): Rh′-5′-AGTCGTCGACCGGGCATGTCCGGGCATGTCCGGGCATGTCCCGTACTAGG-3′ (ss-Rh′-SO) and Rh′-5′-AGTCGTCGACCGCGTACTGTGGGCGATCGGCGACACGTCTCCGTACTAGG-3′ (ss-Rh′-NO), respectively. Concentrations of fluorescent single-stranded oligonucleotides were determined from their absorbance at 260 nm using the extinction coefficients calculated from nucleotide composition after subtraction of the absorbance of 5-carboxytetramethylrhodamine at this wavelength. Contribution of 5-carboxytetramethylrhodamine to the absorbance at 260 nm was calculated from the spectrum of this dye and from the absorbance of fluorescent oligonucleotides at 546 nm. Concentrations of fluorescent peptide and oligonucleotides, as well as the presence of free dye were verified by measuring the rhodamine absorbance at 546 nm and by FCS with 1 nm rhodamine standard solution prepared by weighing. At 1 nmconcentration of a standard rhodamine solution and solutions of fluorescent peptides and oligonucleotides, 0.2 fl confocal volume of measurement contains in average one molecule of fluorescent dye that was detected by FCS. Free dye in solutions did not exceed 10% of fluorescently labeled compounds. FCS was performed with confocal illumination of a volume element of 0.2 fl in a ConfoCor® instrument (Carl Zeiss-Evotec, Jena, Germany; Fig. 1) as described previously (21Rigler R. Mets Ü. Widengren J. Kask P. Eur. Biophys. J. 1993; 22: 169-175Crossref Scopus (878) Google Scholar). As focusing optics a Zeiss Neofluar 40× numerical aperture 1.2 objective for water immersion was used in an epiillumination setup. Separation of exciting from emitted radiation was achieved by dichroic (Omega 540 DRL PO2; Omega Optical, Blattleboro, VT) and bandpass (Omega 565 DR 50) filters. The Rh-BD and Rh′-oligonucleotides were excited with the 514.5-nm line of an argon laser. The intensity fluctuations were detected by an avalanche photodiode (SPCM 200; EG & G, Quebec, Canada) and processed with a digital correlator (ALV 5000, ALV, Langen, Germany). The intensity autocorrelation functionG(t) is an average of the product between the intensity and its time shifted version (Equation 1). G(t)=〉I(t)I(t+τ)=1T∫0TI(t)I(t+τ)dtorG(t)=〈I〉2+〈∂I(t)∂I(t+τ)〉Equation 1 Correlation of the observed fluorescence intensity fluctuations δI(t) with fluorescence intensity fluctuations at time t + τ yields the normalized intensity autocorrelation function G(τ).G(τ)=1+〈δI(t)δI(t+τ)〉〈I〉2Equation 2 where the brackets describe the time average and describes the mean fluorescence intensity (21Rigler R. Mets Ü. Widengren J. Kask P. Eur. Biophys. J. 1993; 22: 169-175Crossref Scopus (878) Google Scholar,22Ehrenberg M. Rigler R. Chem. Phys. 1974; 4: 390-401Crossref Scopus (329) Google Scholar). In our experiments fluorescence intensity fluctuations δI(t) occurring in a volume element of 0.2 fl with half-axes ω = 0.25 μm and z = 1.25 μm are correlated (21Rigler R. Mets Ü. Widengren J. Kask P. Eur. Biophys. J. 1993; 22: 169-175Crossref Scopus (878) Google Scholar, 22Ehrenberg M. Rigler R. Chem. Phys. 1974; 4: 390-401Crossref Scopus (329) Google Scholar), and for calculation of parameters of the autocorrelation function G(τ) the nonlinear least square minimization is used (23Marquardt D.W. J. Soc. Indust. Appl. Math. 1963; 11: 431-441Crossref Google Scholar). The intensity autocorrelation function for three-dimensional diffusion (13Rigler R. Pramanik A. Jonasson P. Kratz G. Jansson O.T. Nygren P.-Å. Ståhl S. Ekberg K. Johansson B.-L. Uhlén S. Uhlén M. Jörnvall H. Wahren J. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 13318-13323Crossref PubMed Scopus (311) Google Scholar) of the unbound Rh-BD or Rh′-SO and bound Rh-BD or Rh′-SO is given by the following equation.G(τ)=1+1N(1−y)11+ττDF11+ωz2ττDF1/2+y11+ττDB11+ωZ2ττDB1/2Equation 3 N is the number of molecules, τDB = ω2/4DB is the diffusion time for bound Rh-BD or Rh′-SO; τDF = ω2/4DF is the diffusion time for unbound Rh-BD or Rh′-SO; y is the fraction of bound Rh-BD or Rh′-SO diffusing with τDB; (1 − y) is the fraction of unbound Rh-BD or Rh′-SO diffusing with τDF. Fluorescent probes Rh-BD, Rh′-SO, or Rh′-NO were incubated at 2 nm concentrations in the absence or the presence of p53 proteins, peptides, or other compounds in binding buffer (20 mm HEPES, pH 7.4, 50 mm KCl, 0.1 mm EDTA, 2 mm MgCl2, 1 mm dithiothreitol, 20% glycerol, and 0.05% Triton X-100) used for p53-DNA binding studies at 20 °C, and droplets (10 μl) of the samples were analyzed by FCS for 1 up to 3 min. Bovine serum albumin (1 μg/μl) was included when Rh′-SO and Rh′-NO oligonucleotides were used as fluorescent probes. Measurements were performed in triplicate 2–5, 10–20, and 30–40 min after the initiation of the reaction. Practically identical results were obtained at these time points. Anti-YY1 C20 antibody with blocking peptide and anti-p50 antibody were obtained from Santa Cruz Biotechnology (San Diego, CA), anti-p53 PAb421 antibody was from Oncogene Science Inc. (Unlondale, NY). Sequence specificity of peptide effects on p53 binding to DNA was studied in a band shift assay. As previously described (18Selivanova G. Iotsova V. Kiseleva E. Strom M. Bakalkin G. Grafstrom R.C. Wiman K.G. Nucleic Acids Res. 1996; 24: 3560-3567Crossref PubMed Scopus (65) Google Scholar), the GST-wild type p53 protein remained latent in the absence of any DNA competitor (Fig. 2, lane 1). p53 binding to the consensus p53-binding site, BC, was activated by PAb421 antibody (lane 2), and by the p53(361–382) fragment (lane 3). p53(361–382) is a basic peptide, which suggests that other basic peptides may also activate p53. Three model basic peptides with irregular alternation of basic residues, the 32-amino acid fragment of prodynorphin (human prodynorphin 207–238 or BD) and two of its constituent fragments, dynorphin A (17 amino acids) and B (13 amino acids), as well as penta-l-lysine and poly-l-lysine with a size up to 5 kDa, were tested for effects on p53 binding to DNA. Both BD and poly-l-lysine strongly activated p53 (lanes 4, 15, and16), whereas dynorphins A and B produced practically no activation (lanes 5 and 6). No activation was observed with penta-l-lysine at ∼30-fold higher molar concentration (lanes 12 and 13) or a mixture of dynorphins A and B at a concentration equimolar with BD (lane 7). No stable complexes were seen when radiolabeled DNA was incubated with these peptides alone (lane 11 and data not shown). Complex formation activated by BD was inhibited by unlabeled BC oligonucleotide (lane 9) but not by unlabeled oligonucleotide MN lacking p53 binding site (lane 10; Ref.18Selivanova G. Iotsova V. Kiseleva E. Strom M. Bakalkin G. Grafstrom R.C. Wiman K.G. Nucleic Acids Res. 1996; 24: 3560-3567Crossref PubMed Scopus (65) Google Scholar), demonstrating that the peptide activated sequence-specific interaction with DNA. p53-labeled BC complex formation was activated by basic peptides in the absence of any competitor DNA (lanes 3and 4) or in the presence of poly(dI-dC) (lanes 8, 10, 15, and 16). To test whether p53(361–382) and BD can target transcription factors other than p53, their effects on the DNA binding activity of AP-1, NF-κB, YY1, and protein Ku were studied in nuclear extracts of SH-SY5Y cells. Only the DNA binding activity of YY1, a multifunctional transcription regulator, was activated by the two peptides (data not shown). Experiments were repeated with affinity purified YY1, and DNA binding was strongly (10–50-fold) activated by p53(361–382) and BD (Fig.3 A, lines 1–6) with half-maximum effects observed at 5–10 nmconcentrations (data not shown). Activated YY1 retained its sequence specificity because an oligonucleotide with wild type but not mutant binding site inhibited binding (lanes 7–9). C20 anti-YY1 antibodies supershifted the protein-DNA complex activated by p53(361–382), showing that YY1 is present in the complex (lane 13), whereas other antibodies or C20 antibodies preincubated with a blocking peptide (lanes 11 and 12) failed to inhibit YY1 activation. To determine whether the basic segment within the p53 protein is able to activate YY1, the effects of the full-length p53 and p53 domains on the YY1 DNA binding activity were compared. GST-wild type p53 at a low concentration (0.7 nm) substantially (20–50-fold) stimulated specific DNA binding of purified YY1 (Fig. 3 B,lane 6). GST-CTD, GST-p53His273, and GST-p53Trp248 fusion proteins also enhanced DNA binding activity of YY1 (lanes 8,10, and 11). The lower band apparently represented the truncated YY1-DNA complex. Myoglobin, casein, RNase, polynucleotide kinase, bovine serum albumin, GST-retinoblastoma, and GST-EBNA5 fusion proteins used as controls at 15 nmconcentration failed to modify YY1-DNA binding (data not shown). GST protein alone, GST-p53-N-terminal domain (NTD), and core domains and the GST-p53 deletion mutant lacking the 30 C-terminal amino acids failed to activate YY1 (lanes 2–5). Thus, activation is associated with CTD. To map the sequence of activation in closer details, eight partly overlapping 22-mer peptides spanning most of the C-terminal domain (residues 337–393) were examined. Only the most basic CTD fragment, p53(361–382) retained the ability to strongly activate YY1 (Fig. 3 C, lane 6). The basic CTD segment in latent p53 appears to be available for intermolecular interaction with YY1 or YY1-target oligonucleotides but, paradoxically, does not activate p53 itself (Fig. 2, lane 1). Basic peptides may activate p53 by interacting with the p53 protein, with target DNA or with both molecules. Multiple peptide-protein and peptide-DNA interactions including formation of complexes of weakly interacting molecules at their nanomolar concentrations cannot be readily analyzed by conventional biochemical methods. FCS, a new highly sensitive biophysical technique, allows the analysis of weak interactions and intermolecular complex formation without preceding separation of a fluorescently labeled probe from labeled complexes (Fig. 1 and Refs. 11Eigen M. Rigler R. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 5740-5747Crossref PubMed Scopus (908) Google Scholar, 12Rigler R. J. Biotechnol. 1995; 41: 177-186Crossref PubMed Scopus (205) Google Scholar, 13Rigler R. Pramanik A. Jonasson P. Kratz G. Jansson O.T. Nygren P.-Å. Ståhl S. Ekberg K. Johansson B.-L. Uhlén S. Uhlén M. Jörnvall H. Wahren J. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 13318-13323Crossref PubMed Scopus (311) Google Scholar, 14Tjernberg L. Pramanik A. Björling S. Thyberg P. Thyberg J. Nordstedt C. Terenius L. Rigler R. Chem. Biol. 1999; 6: 53-62Abstract Full Text PDF PubMed Scopus (139) Google Scholar, 15Wohland T. Friedrich K. Hovius R. Vogel H. Biochemistry. 1999; 38: 8671-8681Crossref PubMed Scopus (117) Google Scholar). In the first set of experiments, interactions of the tetramethyl rhodamine-labeled basic peptide, BD (Rh-BD) as a fluorescent probe with the p53 protein and target DNA were studied by FCS. BD, but not p53(361–382), was chosen to distinguish the interactions relevant for the p53 activation from those in which p53(361–382) may be also involved as a fragment of the p53 molecule. Rh-BD was incubated alone or with either GST-p53 or target oligonucleotide ds-SO. Aliquots of the incubation mixture were analyzed by FCS after 2–5, 10–20, and 30–40 min. The fluctuations (changes) of the fluorescence intensity of fluorophore-labeled BD molecules, excited by a focused laser beam, were registered (Figs. 1 and 4 A). These fluctuations reflected a change in the number of Rh-BD molecules in the volume of measurement and were expressed in kHz. At 2 nm concentration of fluorophore, an average of two fluorescent molecules are present in the volume. The fluorescence intensity is increased when new fluorescent molecules diffuse into the volume and decreased when some molecules diffuse out of the volume. From the intensity autocorrelation function (EquationsEquation 1, Equation 2, Equation 3) of the fluorescence intensity fluctuations, the diffusion time of molecules (τD) through the confocal volume is determined, which allows calculations of the molecular weight. Formation of complexes of fluorophore with other molecules in the reaction medium or aggregation of the fluorophore molecules result in an increase of molecular weight of fluorescent complexes and, consequently, their diffusion time (τD). From the correlation function,G(τ) (Equation 3) diffusion times and proportions (weight factors, y) of different fluorescent components in the reaction mixture were evaluated. Fluorescence intensity fluctuations and autocorrelation functions of 2 nm Rh-BD alone and in the presence of GST-p53 fusion protein and DNA are presented in Fig. 4. Rh-BD (control) exhibited typical fluctuations (Fig. 4 A) and a diffusion time (τD) of 0.175 ms (Fig. 4 C, curve 1). GST-p53 in the concentration range from 6 to 37 nm(Fig. 4 C, curve 2, and data not shown) did not have any effect on the diffusion time of Rh-BD. Addition of 1 nm double-stranded specific (ds-SO; Fig. 4 B) or double-stranded nonspecific (data not shown) oligonucleotide containing wild type or mutant p53 consensus sequences, respectively, as well as 0.24 nm of plasmid DNA (data not shown), to the reaction mixture with Rh-BD resulted in an increase and a broadening of the fluctuation peaks. The fluorescence intensity fluctuations exhibited several peaks with strongly increased heights (Fig. 4 B). These peaks, with up to five times higher intensity than the base line, are a clear representation of the Brownian motion of peptide-DNA aggregates containing several tetramethyl rhodamine-labeled BD molecules. The analysis of the autocorrelation of intensity fluctuations showed the presence of two fluorescent components in the reaction mixture characterized by diffusion times of τD1 = 0.175 ms and τD2 = 3899 ms, respectively, and corresponding weight factors (fractions) of y1 = 0.35 andy2 = 0.65 (Fig. 4 C, curve 3). The appearance of the second component with longer diffusion time reflected the formation of large Rh-BD-DNA aggregates. From the diffusion times 0.2 and 4 s, the hydrodynamic radii of equivalent spheres were calculated (14Tjernberg L. Pramanik A. Björling S. Thyberg P. Thyberg J. Nordstedt C. Terenius L. Rigler R. Chem. Biol. 1999; 6: 53-62Abstract Full Text PDF PubMed Scopus (139) Google Scholar) according to the Stoke-Einstein equation and were found to be 4 and 80 μm, respectively. The three-dimensional structure of aggregates is unclear, and, therefore, calculations assuming other shapes than a sphere were not performed. Practically identical results were obtained at different time points after the initiation of the reaction. Thus, at nanomolar concentrations of basic peptide, protein, and DNA, Rh-BD did not interact w
Referência(s)