Inhibition of p53 Transcriptional Activity by the S100B Calcium-binding ...

Download PDF
6 downloads 0 Views 183KB Size Report
Comment
May 14, 2001 - vector, SN3, were provided by Bert Vogelstein (Johns Hopkins Univer- sity, Baltimore, MD). These plasmids have been described previously.
THE JOURNAL OF BIOLOGICAL CHEMISTRY © 2001 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 276, No. 37, Issue of September 14, pp. 35037–35041, 2001 Printed in U.S.A.

Inhibition of p53 Transcriptional Activity by the S100B Calcium-binding Protein* Received for publication, May 14, 2001, and in revised form, July 9, 2001 Published, JBC Papers in Press, July 13, 2001, DOI 10.1074/jbc.M104379200

Jing Lin‡, Mellissa Blake‡, Chun Tang‡, Danna Zimmer§, Richard R. Rustandi‡, David J. Weber‡, and France Carrier‡¶ From the ‡Biochemistry and Molecular Biology Department, School of Medicine, University of Maryland, Baltimore, Maryland 21201 and the §Department of Pharmacology, University of South Alabama, Mobile, Alabama 36688

The levels of S100 Ca2ⴙ-binding proteins correlate with the progression of certain tumors, but their role, if any, in carcinogenesis is still poorly understood. S100B protein associates with both the p53 oligomerization domain (residues 325–355) and the extreme C terminus of the tumor suppressor p53 (residues 367–392). Consequently, S100B inhibits p53 tetramer formation and p53 phosphorylation mediated by protein kinase C, on p53 C-terminal end. In this report, we show that the S100B protein decreases p53 DNA binding and transcriptional activity. The effect of S100B is reflected in vivo by a reduced accumulation of p53, p21, and MDM2 protein levels in co-transfection assays and in response to bleomycin. The S100B can still interact with p53 in the absence of p53 extreme C-terminal end and reduce the expression of p53 downstream effector genes. These data indicate that S100B does not require p53 extreme C-terminal end to inhibit p53 activity. Collectively, these findings imply that elevated levels of S100B in tumors such as astrocytomas and gliomas could inhibit p53 functions and contribute to cancer progression.

The S100 proteins are dimeric Ca2⫹-binding proteins (⬃10 kDa/subunit) initially characterized by their solubility in 100% ammonium sulfate (S100) (1). Deregulation of Ca2⫹ homeostasis has been associated with different pathologies including neurodegenerative disorders, hypertension, and cancer (2). The S100 proteins are overexpressed in many tumor cells and have been used as a marker for the classification of tumors (3). A possible role for the S100 proteins in carcinogenesis has often been suspected, but their specific involvement is still ill defined. Evidence has indicated that S100B interacts with the tumor suppressor p53 (4). p53 plays a pivotal role in the maintenance and regulation of normal cellular functions, and its inactivation can affect cell cycle checkpoints, apoptosis, gene amplification, centrosome duplication, and ploidy (5). p53 interacts with a number of proteins to mediate its pleiotropic effects. The interactions of p53 with S100 calcium-binding proteins are of particular interest because like p53, the S100

* This work was supported by American Cancer Society Grant RPG00-040-01-CCG (to D. J. W. and F. C.) and by NIGMS, National Institutes of Health Grants RO1GM58888 (to D. J. W.) and 1RO1GM57827-01 (to F. C.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ¶ To whom correspondence should be addressed: Biochemistry and Molecular Biology Department, School of Medicine, 108 N. Greene St., Rm. 330, Univ. of Maryland, Baltimore, MD 21201-1503. Tel.: 410-7065105; Fax: 410-706-8297; E-mail: [email protected]. This paper is available on line at http://www.jbc.org

proteins affect cell cycle progression, are overexpressed in numerous tumor cells, and are associated with tumor progression (2). The S100B protein interacts with the p53 C-terminal end and inhibits both,p53 tetramerization and phosphorylation by PKC1 (4). Because these two events are known to be important for p53 activation (6), we wanted to determine the effect of S100B on p53 transcriptional activity in vivo. Our data indicate that overexpression of S100B can reduce p53 transcriptional activity by more than 50%. This effect is correlated with a decrease in p53 DNA binding activity and a reduction in the accumulation of MDM2 and p21 protein levels. Interaction of the S100B protein with p53 may thus impede p53 cellular functions. Such an interaction could especially be detrimental in astrocytomas and gliomas, where S100B levels are significantly elevated (7). EXPERIMENTAL PROCEDURES

Cells, Plasmids and Chloramphenicol Acetyltransferase (CAT) Assays—The human large cell lung carcinoma cells H1299 and the human breast cancer cell line MCF-7 were grown in RPMI (Life Technologies, Rockville, MD) containing 10% fetal bovine serum. For CAT assay, MCF-7 transfection, and mobility shift assay, a fixed amount of plasmid (10 ␮g) was transfected by the LipofectAMINE technique (Life Technologies). The pCMV3 vector, without insert, was used to normalize the amount of plasmid transfected. The pG13CAT and the p53 expression vector, SN3, were provided by Bert Vogelstein (Johns Hopkins University, Baltimore, MD). These plasmids have been described previously (8). The p53⌬30 plasmid (expressing amino acid residues 1–363 of p53) was constructed by conventional polymerase chain reaction amplification of the 1–1092 sequence of the human p53 cDNA and subcloned in the HindIII-XhoI sites of the pCMV3 expression vector. The p53⌬66 plasmid (expressing amino acid residues 1–327 of p53) and the KEEK plasmid (containing four point mutations; residues 341 (F3 K), 344 (L3 E), 348 (L3 E), and 355 (A3 K) in the p53 oligomerization domain) were provided by Karen Vousden (NCI, National Institutes of Health, Frederick, MD). The Gadd153 expression vector, pCMV153, was provided by Al Fornace (National Institutes of Health, Bethesda, MD) and has been described previously (9). The S100B expression vector, pCMV5-␤Sense, was constructed by polymerase chain reaction cloning of the open reading frame region from human cDNA clone for S100B (10) into the EcoRI and HindIII sites of the pCMV5 vector. The cells were plated at 5 ⫻ 105 cells/100-mm dish the day before transfection. A mixture of lipids (LipofectAMINE) and plasmid DNA was then added to the cells according to the manufacturer’s procedure. Fresh media was added to the cells 5 h after transfection, and the cells were harvested 48 h later. A ␤-galactosidase expression vector (PCH110, Amersham Pharmacia Biotech) was also transfected in all the samples and used as an internal control to normalize transfection efficiency. The cellular proteins were extracted by three rounds of freeze-thaw of the cellular pellets in 25 mM Tris-HCl, pH 8.0. The protein extracts were analyzed for their content of ␤-galactosidase and the CAT activity was measured. The CAT assay was determined according to established protocol (11). 1 The abbreviations used are: PKC, protein kinase C; CAT, chloramphenicol acetyltransferase; DTT, dithiothreitol.

35037

35038

S100B Inhibits p53 through Its Oligomerization Domain

The levels of acetylation were evaluated by PhosphorImager (Molecular Dynamics, Sunnyvale, CA) using the ImageQuant software. Where indicated cells were treated with bleomycin (ICN Pharmaceutical Inc., Costa Mesa, CA) 10 ␮g/ml for 4 h before harvesting. Co-transfection for Western blot analysis was performed in a 150-mm dish with 12 ␮g of plasmid. A ratio of three times (9 ␮g) S100B plasmid versus p53 (3 ␮g) was used. For in vitro translation, all inserts were subcloned in to pcDNA3.1 vector (Invitrogen, Carlsbad, CA). The p53 and ⌬30 inserts were subcloned in the HindIII-XhoI sites by polymerase chain reaction. The S100B was first cloned into pCMV-Myc vector (CLONTECH, Palo Alto, CA) into SalI-XhoI sites then subcloned in to pcDNA3.1 vector in to HindIII-XhoI sites. Mobility Shift Assay—The mobility shift assay was performed as described previously (12) with the exception that the running buffer contained 1 mM Ca2⫹ instead of EDTA, and the binding buffer was 50 mM Tris-HCl (pH 7.5),4 mM dithiotreitol,200 mM NaCl,2 mM CaCl2,10% glycerol. The human p53 protein extracts were obtained from Sf9 insect cells infected with a p53 baculovirus expression vector (13). The infected cells were lysed by three rounds of freeze-thaw in 10 mM Hepes (pH 7.9),10 mM KCl,1 mM phenylmethylsulfonyl fluoride,1.5 mM MgCl2,1.0 mM DTT,2 ␮g/ml leupeptine,2 ␮g/ml aprotinin. The lysate was clarified by centrifugation, and the supernatant was used for band shift analysis. Alternatively, 5 ␮l of in vitro translated p53 reaction or 15 ␮l of in vitro translated S100B (see below) were used. The p53 deletion mutant protein extracts were obtained by extraction of the cellular proteins after transient transfection of the H1299 cells as described above (CAT assays). The S100B recombinant protein was produced and purified as described previously (14). The protein extracts were incubated with a radiolabeled 30-mer double-stranded synthetic oligonucleotide containing the sequence of the p53 binding site found in the third intron (⫹1569 to ⫹1598) of the GADD45 gene (12). The assay was performed in the absence and presence of p53 antibody (pAb421, Oncogene Science) and in the presence of increasing amounts of recombinant S100B protein. For the p53 interaction with S100B, the p53 extracts were incubated with the S100B recombinant protein for 15 min at room temperature, then the DNA probe was added, and the reaction was allowed to proceed for an additional 15 min. For experiments performed in the presence of the p53 antibody, the S100B recombinant protein was first incubated for 5 min with p53 containing protein extracts, the p53 antibody was then added, and the reaction was continued for 5 min. After adding the DNA probe, the reaction was allowed to proceed an additional 15 min. The mixtures were then run on a 4.5% non-denaturant polyacrylamide gel. The gel was dried and exposed to x-ray film. Densitometry analyses were performed with a chilled charge-couple device video camera (Nucleo Vision, Gel Expert software) and competition was calculated as a ratio of the intensity in the supershifted band compared with the total intensity in a given lane. The ratio obtained for the band shifted by the p53Ab in the absence of S100B protein was considered 100%. Western Blot—The Western blot analyses were performed on 100 ␮g of MCF-7 protein extracts or on 100 ␮g of H1299 protein extracts. The MCF-7 cells were lysed in RIPA buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS). The H1299 cells were lysed in RIPA buffer containing 50 mM NaF, 5 mM EDTA, 1 mM DTT, and 10 ␮g/ml aprotinin. The proteins were run on a 12% polyacrylamide gel, transferred on nitrocellulose, and reacted with either p53 mouse monoclonal antibody (DO-1, Oncogene Research Products, Boston, MA) at 1:1000 dilution, p21 rabbit polyclonal antibody (PharMingen, San Diego, CA) at 1:1000, MDM2 mouse monoclonal antibody (Oncogene Research Products) at 1:100 dilution, actin mouse monoclonal antibody (Oncogene Research Products) at 1:5000 dilution to control for even protein loading, or S100B rabbit polyclonal antibody (Swant, Bellinzona, Switzerland) at 1:500 dilution. The blots were then reacted with their respective secondary antibodies conjugated to horseradish peroxidase and reacted with a chemiluminescence substrate (ECL, Amersham Pharmacia Biotech) as recommended by the manufacturer. In Vitro Translation and Pull Down Assay—In vitro translation was performed in rabbit reticulocyte according to the manufacturer recommendation (TNT Coupled Transcription/Translation Systems, Promega, Madison, WI). Circular plasmids containing a T7 promoter were used for RNA synthesis. [35S]methionine (Redivue L-[35S]methionine, AG1094, 370MBq/ml, 10mCi/ml) was from Amersham Pharmacia Biotech. In vitro translation was performed with the following plasmids: pcDNA-Myc, pcDNA3.1-Myc-S100B, pcDNA3.1-p53, and pcDNA3.1-p53⌬30. The pcDNA3.1-Myc-S100B translation mixture (20 ␮l) or the pcDNA3.1-Myc translation mixture (20 ␮l) was combined with either the pcDNA-p53 translation mixture (20 ␮l) or the pcDNA-p53⌬30 mixture (20 ␮l). Sixty ␮l

FIG. 1. S100B competes p53 DNA binding activity stimulated by pAb421. A, mobility shift assay was performed in the presence of increasing amounts (0 –100 ␮g, lanes 3– 6) of recombinant S100B protein and in the presence (⫹) or absence (⫺) of p53 protein extracts (20 ␮g) and p53 antibody pAb421 (1.4 ␮g) as described under “Experimental Procedures.” The indicated percentage (%) of competition was calculated by scanning analyses performed on a PhosphorImager. Bovine serum albumin (BSA) (100 ␮g) was added as a nonspecific competitor (lane 7). The positions of the free DNA, the constitutive binding, and the bands shifted by pAb421 are indicated. B, mobility shift assays performed as in A), except that in vitro translated p53 and S100B protein extracts were used. of buffer A (50 mM Tris-HCl, pH 7.5, 200 mM NaCl, 1 mM phenylmethylsulfonyl fluoride, 1 mM DTT, 2 mM CaCl2) and 2 ␮l of c-Myc monoclonal antibody (200 ␮g/ml) was added to each mixture. The reactions were mixed gently at 4 °C for 2 h. Then, 20 ␮l of protein A/G Plus-Agarose (Santa Cruz Biotechnology, Santa Cruz, CA) was added to each reaction and mixed gently at 4 °C overnight. The agarose beads were then washed three times with buffer A. Sample buffer (40 ␮l of 125 mM Tris-HCl, pH 6.8, 4% SDS, 4% ␤-mercaptoethanol, 20% glycerol, 0.1% Bromphenol Blue) was then added to the beads. An aliquot (2.5 ␮l) of each reaction prior to immunoprecipitation was also dissolved in the sample buffer. The samples were heated at 100 °C for 5 min and centrifuged at 14,000 rpm for 5 min. The supernatants were then loaded on a 12% SDSpolyacrylamide gel electrophoresis. The gel was fixed (50% methanol, 10% glacial acetic acid, 40% water) for 30 min and soaked in Amplify (Amersham Pharmacia Biotech) for 15–30 min with gentle mixing. The gel was dried under vacuum at 80 °C for 1 h and exposed to x-ray film at ⫺70 °C overnight. RESULTS

S100B Decreases p53 DNA Binding Activity—Early studies (15) have shown that modulation of the p53 C-terminal end by either an antibody directed against this domain or truncation of p53 C-terminal end was sufficient to activate p53 DNA binding activity. Because S100B interacts with both the p53 C-terminal end (residues 367–392) (4, 16) and its oligomerization domain (residues 325–355) (17), we wanted to evaluate its effect on p53 DNA binding activity. The data presented in Fig. 1A indicate that, as reported before (18), the p53-infected Sf9 protein extracts contain proteins that can bind constitutively to p53 DNA binding sites (lane 2). Addition of pAb421 leads to the formation of a new complex (15) of slower mobility than the protein extract alone (lanes 3–7). Addition of increasing amounts of S100B protein progressively decreases the intensity of this complex (lanes 4 – 6) as well as the constitutive binding. Densitometry analyses indicate that the p53 DNA binding activity can be reduced by 48% in the presence of 100 ␮g of S100B (lane 6) suggesting that S100B and the pAb421 compete for the same or an overlapping site on p53. A similar amount of bovine serum albumin (lane 7) did not affect the specific p53 DNA binding activity induced by p53 antibody but reduced the constitutive binding. To determine whether p53 DNA binding ac-

S100B Inhibits p53 through Its Oligomerization Domain

35039

FIG. 3. S100B inhibits p53 transcriptional activity. H1299 cells were transiently transfected with a p53 reporter gene construct (pG13CAT) in the presence (⫹) or absence (⫺) of p53 and S100B expression vectors. The relative CAT activity (% conversion) was normalized for background and determined by the ratio of acetylated over total substrate. Gadd153 was used as a negative control. Each value represents the average of four determinations performed in duplicate.

FIG. 2. S100B competes p53⌬30 DNA binding activity. A) Western blot analysis of H1299 cells transfected with p53 deletion mutants and p53KEEK. B) Mobility shift assay performed as in Fig. 1 with p53 deletions mutants (20 ␮g) in the presence (⫹) or absence (-) of S100B (60 ␮g) recombinant protein.

tivity could be reduced with smaller amounts of S100B protein we used in vitro translated protein extracts. The data presented in Fig. 1B indicate that no constitutive binding (lane 2) can be detected in these extracts and that the S100B extract had no binding activity on its own (lanes 3, 5). However, the addition of three times as much S100B than p53 extract (lane 6) competed p53 DNA binding activity by 37%. Assuming that the same amount of protein was made in each extract, this represents a 16:1 molar ratio of S100B/p53. This ratio is much less than what is expected for an in vivo setting where S100B molar concentration (100 pmol/mg, (19)) can be estimated at 1000 to 4000 times higher than p53 concentration (24.5–73.5 fmol/mg, (20)). Moreover, we could not detect p53 DNA binding activity in the absence of p53 antibody (lanes 2), and thus S100B had to compete with the antibody to decrease the binding activity. The data presented above indicate that the S100B inhibitory effect on p53 DNA binding activity is specific and has the potential to be significant in vivo. To determine whether S100B reduces p53 DNA binding activity by interacting with p53 C-terminal end or its oligomerization domain, we performed DNA band shift with H1299 protein extracts containing different p53 deletion mutants (Fig. 2A). The H1299 cells have been selected for their null p53 genotype (9), they lack endogenous expression of either wild type or mutant p53 protein. To facilitate the detection of p53 DNA binding activity in the absence of p53 antibody, we increased the concentration of DTT to 10 mM in the reaction buffer (21). The data presented in Fig. 2B indicate that as expected, deletion of p53 oligomerization domain (⌬66, lane 3) or insertion of a point mutation that prevents p53 oligomerization (KEEK, lane 4) (22) abolishes p53 DNA binding activity. However, as reported before (21), deletion of p53 C-terminal end (⌬30, lane 2) markedly increase p53 DNA binding activity (compare with lane 1). Addition of the S100B protein (lanes 5– 8) reduces the DNA binding activity of the p53 C-terminal deletion mutant (⌬30, lane 6) by 37%. This reduction is in the same range as what was observed with the full-length p53 protein in the presence of p53 antibody (Fig. 1). These data indicate that S100B reduction of p53 DNA binding activity does not require p53 extreme C-terminal end.

S100B Inhibits p53 Transcriptional Activity—S100B has been shown to inhibit in vitro phosphorylation of p53 C-terminal end (14) and to disrupt p53 tetramers (4), two functions important for p53 transcriptional activity (6). The reduced p53 DNA binding activity observed in the presence of S100B (Figs. 1 and 2) suggests that in addition to inhibiting phosphorylation and preventing tetramer formation, S100B could affect p53 transcriptional activity. To verify this possibility we performed transient transfections with S100B and p53 expression vectors concomitantly with a p53 reporter gene construct (pG13-CAT) (8) in H1299 human large cell lung carcinoma cells. The data presented in Fig. 3 indicate that as reported before (8), cotransfection of a p53 expression vector with a p53 reporter gene construct results in transcriptional activation of the reporter gene (Fig. 3, lane 6). Under the conditions used, about 17% of the substrate was converted. Transfection of S100B alone did not affect the reporter gene (lane 4), but co-transfection of S100B with p53 reduced the transcriptional activity by more than 50% (lane 7). When the p53 expression vector was cotransfected with an expression vector for an unrelated protein (Gadd153, lane 8) no effect was observed on p53 transcriptional activity indicating that the effect of S100B on p53 transcriptional activity is specific. We also transfected the cells with p53 ⌬66 and KEEK constructs in the presence or absence of S100B. Even though p53 monomers have been reported to retain their transcriptional activity (23), we did not detect any transcriptional activity with the constructs used here (data not shown). This is probably due to the fact that we used a different reporter construct and a different cell line. Nevertheless, the data presented above indicate that S100B is likely to interact with p53 in vivo and decrease its transcriptional activity. S100B Reduces Accumulation of p53, MDM2, and p21 Protein Levels—To assess the significance of S100B interaction with p53, we measured the protein levels of p53 and two of its downstream effector genes, MDM2 and p21. We first co-transfected transiently the human large cell lung carcinoma cells H1299 with p53 and S100B expression vectors (Fig. 4A). To emphasis the effect of S100B and to mimic the overexpression of S100B observed in certain tumors (7), we used a ratio of three times more S100B plasmid than p53 DNA. The total amount of plasmid DNA was maintained constant with the empty vector pCMV3. No apoptosis was detected under the conditions used here (data not shown). As shown in Fig. 4A, p53 was not detected in the absence of p53 expression vector (lanes, 1, 2). Consequently, the levels of MDM2 and p21 proteins were low in the absence of p53 expression (lanes 1, 2). As expected, expression of p53 triggered expression of

35040

S100B Inhibits p53 through Its Oligomerization Domain

FIG. 4. S100B decreases p53, MDM2, and p21 proteins levels. A, Western blot analysis of H1299 cells transfected with either p53, or p53 ⌬30 with (⫹) or without (⫺) S100B. The positions of the p53, MDM2, p21, S100B, and actin proteins are indicated. B, Western blot analysis of MCF-7 cells transiently transfected (⫹) with S100B or pCMV (⫺) and treated or not treated with 10 mg/ml bleomycin for 4 h. The positions of the p53, MDM2, p21, S100B, and the actin proteins are indicated.

MDM2 and p21 (lane 3) but co-expression with the S100B protein reduced markedly the accumulation of p53, MDM2, and p21 protein levels (lane 4). As reported previously (21), expression of p53 C-terminal deletion mutant (⌬30) resulted in increased levels of p53 downstream effector genes (lane 5). Co-expression of S100B with p53⌬30 reduced to near basal levels the expression of p53⌬30, MDM2, and p21 protein levels (lane 6). This inhibitory effect of S100B is dose-dependent because a more subtle effect was observed with a one to one ratio of S100B versus p53 plasmid (data not shown). Expression of S100B alone (lane 2) in the absence of p53 did not affect significantly the basal levels of MDM2 or p21. The S100B protein is not expressed endogenously in H1299 cells (Fig. 4A, lane 1). Our data suggest that reduction of p53, MDM2, and p21 levels is dependent on S100B interaction with p53 but does not require the p53 extreme C-terminal end. The effect of S100B on p53 mutants lacking the oligomerization domain (⌬66) or having point mutations in the oligomerization domain (KEEK) could not be verified in this assay since none of these mutants induced p53 downstream effector genes (data not shown). These results are in good agreement with our data presented in Fig. 2, where S100B did not require p53 C-terminal end to reduce p53 DNA binding activity. We then asked whether S100B could also interact with the endogenous p53 protein and affect the expression of MDM2 and p21. The human breast cancer cell line MCF-7, which has a wild type p53 genotype (24), was transiently transfected with S100B. In this experiment we used three times less S100B plasmid than in Fig. 4A. Forty eight hours after transfection the cells were either left untreated or treated with the x-ray mimic agent bleomycin. The data presented in Fig. 4B indicate that the basal levels of p53, MDM2 and p21 can be induced by exposure to bleomycin (lane 2). Endogenous levels of S100B are undetectable in MCF-7 cells (lane 1), whereas over-expression of S100B protein reduces the induced levels of p53 and almost completely blocked MDM2 and p21 accumulation (lane 4). Even loading was verified by incubating the blots with an anti-actin antibody and indicates that over-expression of S100B does not affect the overall levels of protein (compare lane 1 with lane 3 and lane 2 with lane 4). These data suggest that S100B can inhibit the endogenous expression of p53 downstream effector genes in vivo. The p53 Extreme C-terminal End Is Not Required for Inter-

FIG. 5. S100B interacts with p53 in the absence of the extreme C-terminal end. In vitro translation was performed as described under “Experimental Procedures.” Immunoprecipitation was performed with c-Myc monoclonal antibody in the presence (⫹) or absence of (⫺) of c-Myc-S100B and p53 or ⌬30. The positions of the different proteins are indicated.

action with S100B—The data presented in Figs. 2 and 4 suggest that S100B could interact with p53 in the absence of the p53 extreme C-terminal end and inhibit its transcriptional activity. To verify this possibility we performed co-immunoprecipitation with in vitro translated proteins. To increase the antibody specificity for the S100B protein, we fused a c-Myc epitope tag to the N terminus of S100B. The data presented in Fig. 5 indicate that in vitro translation of each product generated a protein of expected size (lanes 1–3). The in vitro translated proteins were mixed and incubated as indicated in Fig. 5. Immunoprecipitation of the mixture with a c-Myc antibody pulled down, as expected, the c-Myc-S100B protein (lanes 4 – 6). The antibody also pulled down the p53 full-length protein (lane 5) as well as the p53⌬30 mutant (lane 6) in the presence of c-Myc-S100B protein. No significant amounts of proteins were pulled down in the absence of c-Myc-S100B (lanes 7, 8). A negative control, performed with an unrelated antibody (A18 polyclonal), did not pull down any of the in vitro translated protein (data not shown). These data indicate that S100B can interact with p53 even in the absence of its extreme C-terminal end. DISCUSSION

In this report, we present evidence that S100B can reduce p53 DNA binding and transcriptional activity. S100B interacts with the p53 C-terminal end and inhibits both p53 tetramerization and phosphorylation in vitro (4, 14, 16). These two events are known to be important for p53 transcriptional activation (6) and suggest that in vivo S100B could interfere with p53 functions. Our data (Fig. 1) indicate that recombinant S100B protein reduces p53 DNA binding activity stimulated by an antibody (pAb421) directed against the residues 371–380 of the p53 C-terminal end. These data suggested that S100B competes with the p53 antibody pAb421 binding site. This possibility is supported by a recent NMR study (16) showing a direct interaction between S100B and a p53 peptide (residues 367–388) derived from the p53 C-terminal end, which overlaps with the pAb421 epitope (residues 371–380). However, in contrast to pAb421, S100B does not stimulate p53 DNA binding activity (Fig. 2). This would suggest that the S100B interaction with the p53 C-terminal end is different from the interaction with the pAb421. In fact, S100B forms dimers (16) and also interacts with the p53 oligomerization domain (17). The occupation of both sites, oligomerization and C-terminal end, by

S100B Inhibits p53 through Its Oligomerization Domain S100B could thus lead to disruption of the p53 tetramer formation and reduction of p53 DNA binding activity (Figs. 1 and 2). A transition of the p53 latent form to the active DNA binding form mediated by pAb421 has been shown to occur for p53 tetramers (25). By inhibiting p53 tetramer formation and occupying the C-terminal end, S100B may interfere with the p53 activation and reduce DNA binding activity (Fig. 1). This hypothesis is supported by the more than 50% reduction of p53 transcriptional activity measured in the presence of S100B (Fig. 3). However, the reduction of p53 DNA binding activity observed in the absence of p53 extreme C-terminal end (Fig. 2) suggests that interaction with the p53 C-terminal end alone is not sufficient to mediate the S100B inhibitory effect. Our data indicate that the p53 extreme C-terminal end is not required for interaction with S100B (Fig. 5) or reduction of p53 downstream effector genes accumulation (Fig. 4A). It is possible that in vivo, modifications of p53 C-terminal end hinders S100B binding. The extreme p53 C-terminal end (residues 367–392) contains several phosphorylation and acetylation sites important for p53 functions. A recent analysis by NMR (16) has indicated that S100B sterically blocks two important PKC phosphorylation sites (Ser-376 and Thr-377). This study (16), performed on non-phosphorylated peptide, could either explain the inhibitory effect of S100B on p53 phosphorylation by PKC (4, 14) or indicate that phosphorylation of these sites may change the interaction. A recent report (26) has indicated that p53 mutation at these PKC phosphorylation sites (S376A and T377A) reduces p53 transcriptional activity by ⬃38 and 51% respectively. This is very similar to what we observed here when we transfected p53 with S100B (Fig. 3). Moreover, mutation at Ser-378, another PKC site on p53, did not affect p53 transcriptional activity (26). This Ser is well exposed and not blocked by S100B (16). The level of p53 C-terminal end phosphorylation could thus influence the interaction with S100B and explain the partial inhibitory effect encountered in vivo. Alternatively, interaction of other regulatory proteins such as XP-B, XP-D, or Ref-1 (21) with p53 C-terminal end may reduce S100B interaction with p53 and explain why only a partial inhibitory effect is observed. As mentioned earlier, S100B inhibits p53 phosphorylation by PKC and formation of p53 oligomers. These two activities may actually be link because phosphorylation of p53 by another kinase, hCHK1, requires that p53 be oligomeric (27). Moreover, co-transfection of p53 with antisense hCHK1 reduces p53 levels (27). This is reminiscent of what we are observing with S100B (Fig. 4). By preventing oligomerization of p53, S100B could reduce phosphorylation and stabilization of p53 by not only PKC but also by other kinases such as hCHK1 (27). This possibility seems very likely because S100B also reduces the accumulation of p53 ⌬30, which lacks the PKC phosphorylation site (Fig. 4A). Regardless of the mechanism(s) involved, our findings indicate that S100B prevents p53 from regulating its downstream effector genes (Fig. 4) and could consequently alter p53 cellular functions. Our data seem to oppose a recent report (28) describing the cooperation of S100B with p53-mediated cellular functions. The study presented by Baudier’s team (28) used a temperature sensitive mutant (p53Val135) transfected in rodent cells. In such a system, a mixture of p53 wild type and mutant conformation exists. Since the p53 mutation is localized at amino acid 135 and that S100B interacts with p53 at the C-terminal end between amino acid residues 319 –393, it seems probable that S100B can not discriminate between the two p53 forms. The effects observed by this group could simply be due to the release of wild type p53 molecules from p53 wild typemutant heterodimers and not be a direct effect of S100B on p53 activity. The cell line used here, H1299, has a null genotype for

35041

p53 (12), therefore the only p53 present in our system was the wild type form (full-length or deletions) that we transfected. Moreover, as mentioned above, the interaction of S100B with p53 has been shown to disrupt p53 oligomerization and phosphorylation (4). Because those two events are known to be important for p53 activation (6), one may anticipate that interaction with the S100B protein would interfere with p53 cellular functions. The data presented here and the observations mentioned above (16, 26) support the idea that S100B inhibits p53 activity. Moreover, a recent report (29) has indicated that S100A4, another member of the S100 protein family decreased p53 DNA binding and transcriptional activity. In summary, our data indicate that S100B inhibits p53 functions. The p53 extreme C-terminal end is not required for this effect. Since S100B has been shown to interact with p53 oligomerization domain and prevent the formation of oligomers (4), this mechanism is a likely explanation for the inhibitory effect observed here. However, we can not rule out the possibility that other p53 domains are involved. S100B is a very abundant protein expressed at high levels in certain tumors (7), its inhibitory effect on the low abundance p53 tumor suppressor protein may contribute to cancer progression. Acknowledgment—We thank Dr. Steven Hirschfeld for a careful reading of this manuscript and insightful discussion. REFERENCES 1. Moore, B. W. (1965) Biochem. Biophys. Res. Commun. 19, 739 –744 2. Schafer, B. W., and Heizmann, C. W. (1996) Trends Biochem. Sci. 21, 134 –140 3. Pedrocchi, M., Schafer, B. W., Mueller, H., Eppenberger, U., and Heizmann, C. W. (1994) Int. J. Cancer 57, 684 – 690 4. Baudier, J., Delphin, C., Grunwald, D., Khochbin, S., and Lawrence, J. J. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 11627–11631 5. Ko, L. J., and Prives, C. (1996) Genes Dev. 10, 1054 –1072 6. Giaccia, A. J., and Kastan, M. B. (1998) Genes Dev. 12, 2973–2983 7. Castets, F., Griffin, W. S., Marks, A., and Van Eldik, L. J. (1997) Brain Res. Mol. Brain Res. 46, 208 –216 8. Kern, S. E., Pietenpol, J. A., Thiagalingam, S., Seymour, A., Kinzler, K. W., and Vogelstein, B. (1992) Science 256, 827– 830 9. Zhan, Q., Lord, K. A., Alamo, I., Jr., Hollander, M. C., Carrier, F., Ron, D., Kohn, K. W., Hoffman, B., Liebermann, D. A., and Fornace, A. J., Jr. (1994) Mol. Cell. Biol. 14, 2361–2371 10. Van Eldik, L. J., Staecker, J. L., and Winningham-Major, F. (1988) J. Biol. Chem. 263, 7830 –7837 11. Gorman, C. M., Moffat, L. F., and Howard, B. H. (1982) Mol. Cell. Biol. 2, 1044 –1051 12. Kastan, M. B., Zhan, Q., el-Deiry, W. S., Carrier, F., Jacks, T., Walsh, W. V., Plunkett, B. S., Vogelstein, B., and Fornace, A. J., Jr. (1992) Cell 71, 587–597 13. Pietenpol, J. A., Tokino, T., Thiagalingam, S., el-Deiry, W. S., Kinzler, K. W., and Vogelstein, B. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 1998 –2002 14. Wilder, P. T., Rustandi, R. R., Drohat, A. C., and Weber, D. J. (1998) Protein Sci. 7, 794 –798 15. Hupp, T. R., Meek, D. W., Midgley, C. A., and Lane, D. P. (1992) Cell 71, 875– 886 16. Rustandi, R. R., Baldisseri, D. M., and Weber, D. J. (2000) Nat. Struct. Biol. 7, 570 –574 17. Delphin, C., Ronjat, M., Deloulme, J. C., Garin, G., Debussche, L., Higashimoto, Y., Sakaguchi, K., and Baudier, J. (1999) J. Biol. Chem. 274, 10539 –10544 18. Sheikh, M. S., Carrier, F., Johnson, A. C., Ogdon, S. E., and Fornace, A. J., Jr. (1997) Oncogene 15, 1095–1101 19. Fan, K. (1982) Brain Res. 237, 498 –503 20. Hassapoglidou, S., Diamandis, E. P., and Sutherland, D. J. (1993) Oncogene 8, 1501–1509 21. Jayaraman, L., Murthy, K. G., Zhu, C., Curran, T., Xanthoudakis, S., and Prives, C. (1997) Genes Dev. 11, 558 –570 22. Marston, N. J., Jenkins, J. R., and Vousden, K. H. (1995) Oncogene 10, 1709 –1715 23. Tarunina, M., and Jenkins, J. R. (1993) Oncogene 8, 3165–3173 24. O’Connor, P. M., Jackman, J., Bae, I., Myers, T. G., Fan, S., Mutoh, M., Scudiero, D. A., Monks, A., Sausville, E. A., Weinstein, J. N., Friend, S., Fornace, A. J., Jr., and Kohn, K. W. (1997) Cancer Res. 57, 4285– 4300 25. Waterman, J. L., Shenk, J. L., and Halazonetis, T. D. (1995) EMBO J. 14, 512–519 26. Youmell, M., Park, S. J., Basu, S., and Price, B. D. (1998) Biochem. Biophys. Res. Commun. 245, 514 –518 27. Shieh, S. Y., Ahn, J., Tamai, K., Taya, Y., and Prives, C. (2000) Genes Dev. 14, 289 –300 28. Scotto, C., Deloulme, J. C., Rousseau, D., Chambaz, E., and Baudier, J. (1998) Mol. Cell. Biol. 18, 4272– 4281 29. Grigorian, M., Andresen, S., Tulchinsky, E., Kriajevska, M., Carlberg, C., Kruse, C., Cohn, M., Ambartsumian, N., Christensen, A., Selivanova, G., and Lukanidin, E. (2001) J. Biol. Chem. 5, 5