- Email: [email protected]
Localization of a mutant p53 response element on the Tissue Inhibitor of Metalloproteinase-3 promoter: Mutant p53 activities are distinct from wild-type
Cancer Letters 240 (2006) 48–59 www.elsevier.com/locate/canlet
Localization of a mutant p53 response element on the Tissue Inhibitor of Metalloproteinase-3 promoter: Mutant p53 activities are distinct from wild-type Shana Thomas, David Reisman* Department of Biological Sciences, University of South Carolina, Columbia, SC 29208, USA Received 11 July 2005; received in revised form 30 August 2005; accepted 31 August 2005
Abstract Missense mutations in the p53 gene have been observed in greater than 60% of all human tumors. Recent evidence indicates that some mutations in p53 arise as the cancer progresses from a benign tumor to a metastatic tumor and that these mutations in p53 actively contribute to the process of cancer progression. Previously, we reported that the expression of the gene encoding the tissue inhibitor of metalloproteinase-3 (TIMP-3) is repressed in cells expressing codons 248 and 281 mutant p53 alleles. The ability of tumor-derived p53 mutants to inhibit TIMP-3 expression provides a novel mechanism for understanding how p53 mutations might contribute to tumorigenesis. Since mutant p53 is often expressed at elevated levels in a variety of cancers, the generation of cells in a tumor carrying certain mutations in p53 would cause inappropriately reduced expression of TIMP-3 and lead to elevated matrix metalloproteinase activity. We present the results of experiments that begin to determine the mechanism by which mutant p53 represses TIMP-3 gene expression. By generating deletion derivatives of the TIMP-3 promoter and testing them for expression and by performing DNA protein binding assays on the regions determined to be required for repression, we have identified elements that are essential for mutant p53-mediated transcriptional repression. These elements respond specifically to mutant but not wild type p53. While mutant p53 itself does not bind to the TIMP-3 promoter, we provide evidence for the presence of DNA binding proteins whose activity is enhanced in the presence of mutant p53. q 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Gene expression; p53 tumor-suppressor; Oncogenic mutants; Metalloproteinase
1. Introduction The normal or wild type p53 gene is a tumor suppressor gene, which encodes a protein that regulates a cell cycle checkpoint in response to * Corresponding author. Tel.: C1 803 777 8108; fax: C1 803 777 4002. E-mail address: [email protected] (D. Reisman).
DNA damage, cell stress or the aberrant expression of some oncogenes (for reviews see Refs. [1,2]). Depending on the cell type and extent of DNA damage, expression of wild-type p53 induces either a cell cycle arrest or apoptosis [3]. Missense mutations in the p53 gene, which inactivate its growth suppressing activities, have been observed in greater than 60% of all human tumors [4]. Interestingly, in a number of tumor types,
0304-3835/$ - see front matter q 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2005.08.027
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
such as colorectal and ovarian, the frequency of p53 mutations appears to increase as the cancer progresses from benign to metastatic tumors, suggesting that mutations in p53 may be contributing to the process of cancer progression. In agreement with this notion, a number of studies indicate that tumor-derived mutant forms of p53, while losing many of their DNAdamage checkpoint functions, serve as active transforming genes [4,5]. It has been shown, for example, that introduction of mutant p53 into p53 negative immortalized cells can increase their growth rates in culture and transform them into cells that are tumorigenic in vivo [5,6]. Most recently, mice expressing p53 missense mutations, in an otherwise p53-null background, develop metastatic tumors not seen in p53-null mice [7,8]. These results clearly demonstrate that activities expressed by some mutant p53 alleles contribute to tumorigenesis, and can therefore, be considered oncogenic (for recent reviews see Refs. [9,10]). Supporting the hypothesis that some p53 mutant alleles can contribute novel activities to cells, p53 mutants have been described that have altered DNA binding specificities [11–13] and many reports have described p53 mutant alleles that can alter the transcription of genes that are not normal targets of wild-type p53 [14–17]. Genes that have been reported to be modulated in their expression by mutant p53 include the human multi-drug resistance gene (MDR1; [18]), c-myc [19], tissue inhibitor of metalloproteinase-3 [14], insulin-like growth factor II [20], parathyroid hormone-related protein [21], and CD95FAS/ APO-1 [22]. In addition to affecting gene expression, mutant p53 can express other novel activities that include binding to DNA elements of the nuclear matrix [23] and disrupting spindle formation checkpoints [24]. Finally, the ability of mutant p53 to complex with the family of p53 homologues, p73abg and p63 and interfere with their activities, is likely to further contribute to the transforming potential of mutant p53 [25–28]. Previously, we reported that the gene encoding the tissue inhibitor of metalloproteinase-3 (TIMP-3) is repressed in cells expressing either wild type p53 or the codons 248 (R248W) and 281 (D281G) mutant p53 alleles. We propose that although wild-type p53 can repress TIMP-3 expression, it is mutant p53 that is
49
relevant to tumor progression since wild type p53 is expressed at exceedingly low levels in the absence of DNA damage. Whereas mutant p53 is expressed at very high levels in tumor cells and thus is much more likely to have an effect on TIMP-3 expression. TIMP-3 in fact has multiple physiological functions and in addition to inhibiting the matrix metalloproteinases has a newly established role in the inflammatory response and appears to be a regulator of tumor-necrosis factor alpha [29]. Wild type p53 expression may potentially have a role in this pathway in normal cells, but this remains to be explored. The finding that some tumor-derived p53 mutants can inhibit TIMP-3 expression therefore provides evidence for a novel function of mutant p53 and how these mutations could contribute to tumor progression. Since mutant p53 is often expressed at elevated levels in a variety of cancers, the generation of cells in a non-invasive tumor carrying a mutation in p53 would, in addition to losing crucial wild-type p53 mediated cell cycle checkpoints, cause inappropriate reduced expression of TIMP-3. As TIMP-3 has been shown to inhibit the activity of matrix metalloproteinases [30], repressed TIMP-3 expression would be predicted to result in elevated matrix metalloproteinase activity that, in turn, could lead to increased tissue degradation and an increased likelihood of metastasis. Interestingly, the loss of TIMP-3 expression has been observed in many cancers and its loss has been associated with increased metastatic potential [31–35]. While much is known about the biochemical and gene regulatory activities of the wild-type p53 protein, relatively little is known about mutant p53 and its contribution to the transformed phenotype. Here, we present the results of experiments, where we begin to determine the mechanism by which mutant p53 represses TIMP-3 gene expression. By generating a series of deletion derivatives of the TIMP-3 promoter and testing them for p53 responsiveness as well as by performing DNA protein binding assays, we have identified elements that are essential for mutant p53-mediated transcriptional repression. While mutant p53 does not bind to the TIMP-3 promoter, we provide evidence for the presence of DNA binding proteins that likely contribute to TIMP-3 repression.
50
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
2. Materials and methods 2.1. Cell culture 10(3) cells were kindly provided by Dr Arnold Levine (Princeton University). These cells are murine fibroblasts that lack p53 expression. 10(3) cells were maintained in Dulbecco’s modified Eagle’s media (DMEM) and supplemented with 10% fetal bovine serum (FBS), 2 mM glutamine, 100 m/ml penicillin and 100 mg/ml streptomycin. CMV248 Cl.2 cells are murine fibroblasts that contain a colon carcinoma-derived p53 mutation (codon 248 Arg to Trp; R248W). These cells constitutively express mutant p53 through the control of the cytomegalovirus (CMV) promoter. CMV248 Cl.2 cells were maintained in DMEM and supplemented with 10% FBS, 2 mM glutamine, 50 mg/ml G418, 100 m/ml penicillin, and 100 mg/ml streptomycin. Human colon carcinoma cell lines containing the R248W p53 mutation Colo 320 HSR, Colo 320 DM, and SW 837 were grown in RPMI 1640 media with 10% FBS, 2 mM glutamine plus penicillin and streptomycin. 2.2. Western blot analysis Protein extracts were prepared from 10(3), SVT2, HT-29, and H1299 cells. The cells were washed with PBS, harvested and lysed in RIPA buffer (1 mM Tris–HCl pH 7.4, 150 mM NaCl, 1% sodium deoxycholate, 0.1% SDS, and 1 mM PMSF). The cell lysate (25–50 mg) was run on a 12% denaturing polyacrylamide gel in Tris/Glycine/SDS running buffer along with an appropriate standard. The extracts were then transferred to a nitrocellulose membrane. The membranes were blocked with 5% non-fat dry milk for 1 h, washed with TBST (1 M Tris–HCl pH 8, 5 M NaCl, and 0.5 ml/L Tween-20), and incubated with primary antibody PAb421 for 1 h. The membrane was washed again with TBST and incubated with its respective secondary antibody conjugated to horse-radish peroxidase. The signals were visualized after exposing X-ray films using ECL Western Blotting Detecting Reagents (Amersham). 2.3. Tetracycline inducible p53 expression The human non-small cell lung carcinoma-derived cell, line H1299, expressing mutant p53 isoforms
from a tetracycline-withdrawal inducible promoter, was kindly provided by Dr Xinbin Chen (University of Alabama). The cells were maintained in DMEM media containing 10% FBS, 2 mM glutamine, 250 mg/ml G418, 2 mg/ml puromycin, 2 mg/ml of the tetracycline derivative, doxycycline, plus penicillin and streptomycin. Under these conditions expression of p53 was low or undetectable. To induce p53 expression, cells were washed four times with complete media lacking doxycycline. The cells were then grown for additional lengths of time (3, 6, 12, 24 h) in complete media lacking doxycycline prior to harvesting for analysis of p53 expression by Western blotting (as previously described). To assay for TIMP-3 promoter activity, cells growing in the presence of doxycycline were transfected with the TIMP-3 luciferase reporter vector. Twenty-four hours after transfection, the cells were washed to remove doxycycline and after an additional 24 h of growth in the absence of doxycycline, extracts were prepared and equal amounts (protein) were assayed for luciferase activity. An internal control expressing renilla from the HSV-TK promoter (TK-renilla) was included in all transfections and used to normalize the transfection efficiencies. Transfection experiments were repeated four times. 2.4. Luciferase reporter constructs The murine TIMP-3 promoter [36] was kindly provided by Dr Nancy Colburn (National Cancer Institute). TIMP-3 promoter deletion constructs were generated by PCR. Primers were designed from the 2900 bp full length promoter sequence and used to amplify regions of the promoter that were deleted from the 5 0 -end. These deletion products were cloned into the pGL3 expression vector (Promega) upstream of the luciferase reporter gene. 2.5. Transient transfections and reporter gene assays Transfections were performed using Transfast Transfection Reagent (Promega) in a 1:1 ratio of total DNA to Transfast. Exponentially growing 10(3), CMV248 Cl.2, Colo 320, Colo 320 DM, and SW 837 cells were plated at a density of 5!104 cells/well in a 24 well plate and grown to 80% confluence. The 10(3) cells were cotransfected with 0–1.0 mg/well mutant
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
p53 (CMV248) and 0.25 mg/well of the TIMP-3 deletion constructs. The CMV248 Cl.2, Colo 320, DM 320, and SW 837 cells were transfected with 0.25 mg/well of the TIMP-3 deletion constructs. Twenty-four hours post transfection, cells were harvested and luciferase activity detected using a Luminometer. Lysates were normalized by measuring TK-renilla activity as an internal control. Relative fold repression was calculated by dividing the activities of the TIMP-3 constructs in the presence of mutant p53 by the activities of the constructs in the absence of mutant p53. All transfections were performed in triplicate.
2.6. Nuclear extracts and DNA-binding assays Nuclear extracts were obtained from 10(3), CMV248 Cl.2, Colo 320, Colo 320 DM, and SW 837 cells by washing plates (80% confluent) twice with 10 mL of phosphate buffered saline (PBS). Cells were then lysed on the plate using 1 mL of lysis buffer (20 mM Hepes, pH 7.6, 10 mM NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, 20% glycerol, 0.1% Triton X100, 1 mM DTT, 1 mM PMSF, 1 mg/mL leupeptin, 1 mg/mL pepstatin, and 1 mg/mL aprotinin). Cells were collected by scraping and pelleted by centrifugation for 5 min at 2000 rpm at 4 8C. The nuclei were gently rocked for 1 h at 4 8C and centrifuged at 10,000 rpm for 10 min. The supernatant was aliquoted, quick frozen in liquid nitrogen and stored at K70 8C. Four 25 base pair double-stranded DNA oligonucleotides spanning the K2900 to K2800 nucleotide region of the TIMP-3 promoter were constructed and endlabeled with g-32P-ATP using T4 polynucleotide kinase. Probes (20 fmole/reaction) were added to binding buffer (50 mM Tris–HCl pH 7.0, 0.1 M KCl 12.5 mM MgCl2, 1.0 mM EDTA, 20% glycerol, and 1 mM DTT) and 0–10 mg of nuclear extract. The reaction was mixed and incubated on ice for 15 min and then at 20 8C for 15 min. For supershift experiments, anti-p53 antibody PAb421 was added to the reaction and incubated at 4 8C for an additional 20 min. The DNA-protein complexes were subjected to electrophoresis on a 4% polyacrylamide gel in 0.5X Tris/Borate/EDTA buffer. The gels were dried and subjected to autoradiography.
51
3. Results 3.1. Induced expression of mutant p53 results in repression of TIMP-3 promoter activity In order to confirm that TIMP-3 expression is inhibited by mutant p53 expression and to test another tumor-derived mutant for this activity, we took advantage of a tetracycline-regulated expression system. This expression system makes use of the herpes simplex virus transactivator, VP16, fused to the Tn10-encoded tetracycline repressor (tetR). In the presence of 1.0 mg/ml of the tetracycline derivative, doxycycline, the fusion protein is inhibited from binding to operator sequences of the tetracycline operon. Upon removal of doxycycline from the media, maximal activation of the gene can be obtained within 12–18 h. A construct expressing the codon 249 p53 mutant (R249S) from the tet-responsive promoter was introduced into H1299, a p53-null human lung carcinoma cell line (generously provided by Dr Xinbin Chen, University of Alabama). These p53-null cells were used in order to eliminate the influence of any endogenous p53 expression on TIMP-3 expression. The H1299 cells carrying the mutant p53 expression construct were tested for their inducibility upon removal of doxycycline from the media. At various time points (3, 6, 12, and 24 h), cells were harvested and assayed for p53 expression by Western transfer analysis. As shown in Fig. 1A, mutant p53 expression was dramatically induced in these cell lines upon withdrawal of doxycycline. These cells were then assayed for repression of the TIMP-3 promoter by transient transfections with a TIMP-3 luciferase reporter vector. As seen in Fig. 1B, the expression of the TIMP-3 promoter was significantly repressed upon induced expression of the R249W mutant p53. These results confirm our earlier findings [14] that, in addition to the codon R248W mutation, expression of other mutant p53 alleles (such as those with mutations at codons 281 and 249), result in repression of the TIMP-3 promoter. To again confirm that the TIMP-3 promoter is repressed by mutant (R248W) p53, a transfection was performed using p53-null 10(3) cells. These 10(3) cells were cotransfected with 0.25 mg of the TIMP-3 luciferase construct and increasing
52
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
concentrations of the mutant p53 CMV-based expression vector (0–1.0 mg). After 24 h, the cells were harvested and the luciferase activity was determined and plotted in relative light units. Fig. 1C shows that the full length (2900 bp) TIMP-3 promoter is repressed by mutant p53 in a dose dependent manner. To determine the relative levels of p53 protein at each DNA concentration, a Western blot was performed using anti-p53 antibody PAb421 (Fig. 1C, inset). As the concentration of mutant p53 DNA increases from 0 mg (lane 1), 0.05 mg (lane 2), 0.1 mg (lane 3), 0.25 mg (lane 4), 0.5 mg (lane 5), to 1.0 mg (lane 6) the corresponding protein levels also increase (top panel). Actin was used as a loading control and is pictured in the bottom panel. 3.2. Localization of a mutant p53 response element in the TIMP-3 promoter
Fig. 1. Induction of mutant p53 from a doxycycline (Dox) repressed promoter leads to inhibition of expression of the TIMP-3 promoter. H1299 cells were grown in the presence of 2 mg/ml Dox or in the absence of Dox for 3, 6, 12, and 24 h. (A) At the times indicated, cells were harvested and extracts were assayed for levels of p53 protein by Western transfer. Extracts of the human colon carcinoma cell line HT-29 were used as a positive control for p53 protein. The arrow indicates the position of p53. (B) Twenty-four hours after removal of Dox, cells were transfected with the TIMP-3-luciferase reporter vector. After an additional twenty-four hours, cells were harvested and assayed for luciferase activity. Cells with Dox express low levels of p53 and approximately 5–6 fold elevated levels of TIMP-3. An internal control, TK-renilla, showed no changes in expression. The results are the average of four experiments. (C) 10(3) cells were cotransfected with 0.25 mg of
In order to identify which region or regions of the TIMP-3 promoter are required for the repression observed in response to mutant p53, we generated a series of deletion derivatives of the TIMP-3 promoter (Fig. 2) and tested them for their response to either wild-type or mutant p53 expression. Deletions that progressively removed an increasing amount of DNA from the 5 0 -end of the promoter were generated by PCR. The resulting deletions which removed 600, 1000, 2000 and 2500 bp were subcloned into the pGL3 luciferase reporter vector and transfected into 10(3) cells in the presence of a either a R248W or wtp53 CMV-based expression vector. After 24 h, cells were harvested and extracts prepared and tested for luciferase activity. As shown in Fig. 3A, repression of TIMP-3 promoter activity by mutant p53 was eliminated with the deletion of 600 bp closest to the 5 0 -end of the promoter. Interestingly, repression by wt 3 the TIMP-3 promoter construct and 0–1.0mg mutant p53 expression vector (CMV 248). Twenty-four hours post transfection, cells were harvested and luciferase activity was determined. (Inset) A Western Blot was performed on extracts taken from the transfected cells and probed with anti-p53 antibody PAb421. The blot depicts the increasing level of mutant p53 from 0 mg (lane 1), 0.05 mg (lane 2), 0.1 mg (lane 3), 0.25 mg (lane 4), 0.5 mg (lane 5), to 1.0 mg (lane 6). Extract from a p53-positive control cell line (SVT2) is pictured in lane 7. An anti-actin antibody was used to detect actin and is shown as a loading control.
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
53
Fig. 2. Illustration of the 2900 bp TIMP-3 promoter with putative regulatory sites. Also shown are the two series of deletions that were generated by PCR and cloned into the luciferase expression vector to determine activity.
p53 was unaffected by any of the deletions and was still observed after deletion of 2500 bp. Since only approximately 400 bp of the TIMP-3 promoter remains on this last deletion derivative and it remains responsive to wt p53 expression, we conclude that repression by the mutant and wild-type p53 proteins are occurring through distinct elements in the TIMP-3 promoter, and therefore via different mechanisms. While repression by wt p53 may be occurring through elements of general transcription complex, as seen with other promoters [37], the repression by mutant p53 is occurring via an as yet unknown mechanism. Neither the wild-type nor mutant mediated repression appears to occur by virtue of binding to a predicted p53 consensus recognition site in the TIMP promoter at approx—650. Although wt p53 can bind to this site in vitro [33; plus data not shown], it has no effect on expression of the promoter. To further investigate the p53 mediated repression of TIMP-3 and to determine whether the 600-bp site at the 5 0 end of the promoter is sufficient to render a heterologous promoter responsiveness to mutant p53, we cloned this element upstream of the TK promoter and cotransfected this construct into p53 null 10(3) cells along with mutant p53 (CMV 248). Repression of TIMP-3 was not seen with this construct (data not shown). In view of this result, we reasoned that additional cooperating and essential elements may exist on the TIMP-3 promoter itself. Therefore in experiments described below, we chose to reconstruct a responsive TIMP-3 promoter rather than the heterologous TK promoter. In order to locate, where the R248W-responsive element is located within the first 600 bp of the TIMP-3 promoter, we generated an additional series of deletions that progressively removed 100 bp from the intact TIMP promoter. These deletions were again
tested for their ability to be repressed by the presence of mutant p53. As shown in Fig. 3B, deletion of 100 bp from the 5 0 -end of the promoter resulted in the loss of repression by mutant p53. Since further deletions of 200, 300, 400, and 500 bp from the promoter also maintained little to no activity, it was concluded that repression of the TIMP-3 promoter by mutant p53 is mediated through 100 bp of DNA sequences located approximately K2800 to K2900 bp upstream of the transcription start site. To provide further evidence that the K2800 to K2900 bp region of the TIMP-3 promoter mediates transcriptional repression by mutant p53, the construct containing the TIMP-3 promoter lacking the first 100 bp was tested for its activity in tumor-derived cell lines expressing mutant p53. As shown in Fig. 3C, the relative activity of the K100 bp TIMP-3 promoter is significantly greater than that of the full length TIMP-3 promoter in tumor-derived cells naturally expressing mutant p53. Again these results indicate that a negative regulatory element is located within the first 100 bp of the promoter and that this element exerts its effect in cells expressing physiologically occurring levels of mutant p53 protein. 3.3. Nuclear factors bind to the mutant p53responsive element in the TIMP-3 promoter Having determined that a 100-bp element is required for mutant p53 mediated repression of TIMP-3; we next asked whether we could identify nuclear proteins that bind to this region of the promoter. If so, one or more of these factors might be responsible for carrying out the repression. In order to assay for DNA binding proteins, nuclear extracts were prepared from cells expressing the R248W
54
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
experiments, an oligonucleotide comprising the 100 bp 5 0 regulatory region was labeled with 32P and assayed for its ability to bind nuclear proteins by electrophoretic mobility shift assays. Next, the 100 bp region was divided into three overlapping fragments, labeled with 32P and used to perform mobility shift assays (data not shown). Upon detection of binding activity, smaller regions were generated until ultimately a series of approximately 25-bp double stranded oligonucleotides was synthesized (Fig. 4A), labeled with 32P and used as probes in electrophoretic mobility shift assays. (Fig. 4B). All four oligonucleotide probes demonstrated binding, but only TIMP(3)1 and TIMP(3)4 were found to bind specifically to nuclear proteins. Specificity was demonstrated by competition for binding by either a 10- or 50-fold molar excess of the homologous unlabeled oligonucleotide or by a 10- or 50-fold molar excess of a random unlabeled oligonucleotide. Regions TIMP3(2) and TIMP3(3) were found to bind numerous factors, but all were shown to be nonspecific (data not shown). These results indicate that nuclear factors specifically interact with two regions (1 and 4) within the DNA sequences that are required for mutant p53-mediated repression. 3.4. Two nuclear factors bind to the TIMP-3 promoter selectively in cells expressing mutant p53
Fig. 3. Wild-type and mutant p53 repress TIMP-3 promoter activity through separate regions of the promoter. (A) Cells were cotransfected with 0.25 mg mutant p53 (CMV248) or wt p53 expression vectors and 0.25 mg of TIMP-3 promoter deletions. Twenty-four hours post transfection, cells were harvested and luciferase activity was determined. The size if the deletions are indicated. (B) Cells were co-transfected with 0.25 mg of mutant p53 and 0.25 mg of TIMP-3 promoter deletions. (C) The full length TIMP-3 promoter or the K100 bp deletion was transfected into 10(3) cells (transiently expressing mutant p53) and cells expressing endogenous mutant p53 (CMV248 Cl.2, Colo 320, and SW 837). Values were normalized to TK-Renilla expression. All transfections were done in triplicate.
mutant. These cells included the CMV248 Cl2 murine cell line that expresses mutant p53 from the CMV promoter as well as Colo 320 and SW 837 cells that are derived from human colon cancers and also express the R248W mutant. In a first series of
Since the TIMP-3 promoter is repressed when mutant p53 is expressed, we wanted to determine whether either of the two factors described above would bind to their target DNA in response to mutant p53 expression. To ask this question, we again carried out DNA binding assays, using binding sites, TIMP(3) 1 and TIMP(3)4, and nuclear extracts derived from R248W expressing cells and p53-null cells. As shown in Fig. 4C, while a slower migrating DNA-protein complex was observed in all cells tested, faster migrating complexes, shown to be specific (Fig. 4B), were only observed in cells expressing the R248W mutant. These two sites are candidates for sites that mediate repression of TIMP-3 in response to mutant p53 expression. To further examine the role of mutant p53 in this complex, a DNA binding assay was performed in the presence of an antibody (PAb421) previously demonstrated to shift protein-DNA complexes that contain p53 (data not shown).
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
55
Fig. 4. The mutant p53 responsive site on the TIMP-3 promoter binds two nuclear factors. (A) Four approximately 25 bp double-stranded oligonucleotides were synthesized spanning the p53 responsive region at the 5 0 end of the TIMP-3 promoter for use in binding and competition assays. The positions indicate the location of the sequence relative to the 5 0 -end of the cloned 2900-bp promoter. (B) Nuclear extracts from Colo 320 (R248W) cells were used to assay for protein binding to sites TIMP3(1), TIMP3(2), TIMP3(3), and TIMP3(4) using EMSA. Specific (SC) or nonspecific (NSC) competitors at (10- and 50-molar excess, C and C, respectively) were added to the reactions to test for specificity of binding. The specific complexes seen with TIMP3(1) and TIMP3(4) are indicated by arrows. (C) TIMP3(1) and TIMP3(4) were used to probe extracts from p53 null 10(3) cells (lanes 1 and 5), as well as R248W mutant p53 Colo 320 (lanes 2 and 6), Colo 320 DM (lanes 3 and 7), and SW 837 (lanes 4 and 8) cells. (D) EMSA was performed with Colo 320 (lanes 1–4) and SW 837 (lanes 5–8) nuclear extracts. These extracts were probed with binding sites TIMP3(1) and TIMP3(4) with and without the addition of anti-p53 antibody PAb421, where indicated.
56
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
4. Discussion
Fig. 5. Both cis-acting regions contribute to repression of TIMP-3 promoter activity. Concatermized regions TIMP3(1) and TIMP3(4) were cloned into the TIMP-3 promoter lacking the 100 bp p53 responsive element and tested for expression in transient transfection assays. SW 837 and CMV248 Cl.2 cells, and 10(3) cells transiently expressing mutant p53 were transfected with 0.25 mg/well of the full length (2900 bp), K100 (2800 bp), 2800 bp promoter with site TIMP3(1), or the 2800 bp promoter with site TIMP3(4).
As seen in Fig. 4D, when the antibody was added to either the Colo 320 or SW 837 nuclear extracts and probed with binding sites TIMP3(1) and TIMP3(4), no supershift was detected. Therefore, it was concluded that mutant p53 is not bound directly to this DNA–protein complex. 3.5. Each of the two novel factor binding sites contributes to repression of TIMP-3 expression In order to determine which or both of these binding sites are required for TIMP-3 repression, each site was concatomerized to yield three linked binding sites and cloned upstream of the 100-bp TIMP-3 deletion derivative that had lost its ability to be repressed by mutant p53. These reconstituted promoters controlling the luciferase reporter gene were cotransfected into p53-null 10(3) cells and into cells expressing mutant p53. As shown in Fig. 5, the inclusion of either 25-bp element led to a significant repression of promoter activity. This indicates that these elements serve to inhibit TIMP-3 expression. Since neither element reconstitutes complete mutant p53-mediated transcriptional repression, it is likely that both are required to be present to achieve the full response.
Our goal is to define the mechanism by which mutant p53 regulates gene expression and thus may contribute to the transformed phenotype. The data that we have presented indicate that although both wild type and tumor derived p53 mutants can repress the expression of TIMP-3, the sites on the TIMP-3 promoter through which these proteins act, are clearly distinct. Wild type p53 represses transcription through elements close to the transcription start site while mutant p53 represses transcription through elements located approximately 2800 bp upstream of the transcription start site. These data indicate that mutant p53, specifically the R248W p53 mutant, the most commonly occurring of the p53 mutations, represses transcription of TIMP-3 through two 25-bp elements near the 5 0 -end of the TIMP-3 promoter. Two nuclear factors were found to bind to these two elements and their binding activity appears to be present in cells expressing mutant p53, but not in cells lacking p53 expression. These studies therefore provide a framework by which we will be able to examine the activities of mutant p53. We are now beginning the purification of these DNA binding proteins. This will allow us to clone the genes for these factors and to explore their mechanism of action as well as the role that mutant p53 plays in the activity of these factors. TIMP-3 is one of four known tissue inhibitors of matrix metalloproteinases (MMPs) and is unique among them in its being secreted into the extracellular matrix [30,38]. The MMPs consist of a family of proteases that play a major role in the remodeling and turnover of the extracellular matrix as well as in angiogenesis and in the vascularization of tissues. The aberrant activity of MMPs, leading to a pathological degradation of the extracellular matrix, often appears to be an essential step in cancer progression and metastasis. The degradation of the extracellular matrix is necessary for tumor invasion and accordingly, the elevated activities of MMPs have in many cases been shown to be crucial for this process [39,40]. Appropriately, MMPs have been found on the surface or associated with a number of invasive and metastatic tumor cell types and the inhibition of MMP localization to the cell surface inhibits tumor invasiveness [41–46]
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
The ability of some tumor derived mutant forms of p53 to inhibit TIMP-3 expression thus provides a model for understanding how p53 mutations might contribute to tumor progression. Since mutant p53 is stabilized and accumulates to elevated levels in tumor cells [47], the generation of cells in a non-invasive tumor carrying a mutation in p53 could cause reduced expression of TIMP-3 and lead to elevated MMP activity. Interestingly, in some cancers, such as colorectal cancer, where p53 is found to be mutated and overexpressed in over 85% of cases, mutations in p53 have been shown to be relatively late events in tumor cell evolution and to be associated with increased metastatic potential [48]. Some mutant p53 alleles, such as codons 248, 249 and 281, may contribute to tumor progression by altering the expression of genes such as TIMP-3. The mechanism by which mutant p53 interfaces with the factors that we have identified remains to be characterized. Rather than directly interacting with the TIMP-3 promoter, the activity of the codon 248 p53 mutant appears to be mediated by a group of novel DNA binding proteins. Our current data indicate that mutant p53 may be responsible for activating, either transcriptionally or post-transcriptionally, one or more transcriptional repressors that inhibit TIMP-3 expression. One possibility is that mutant p53 interacts directly with the promoters of these genes and contributes to their elevated expression. Further work aimed at cloning the genes that encode these proteins and generating reagents to assay for their expression will allow us to address these problems and develop a much better understanding of the role of mutant p53 in cancer progression. 5. Summary Since reduced TIMP-3 expression has been demonstrated to contribute to tumor invasiveness and metastasis, the ability of tumor-derived p53 mutants to inhibit TIMP-3 expression provides a model for understanding how p53 mutations contribute to tumor progression. We have set out to determine how mutant p53 represses TIMP-3 gene expression and conclude from our findings that the activity of novel DNA binding proteins is enhanced in the presence of mutant p53.
57
Acknowledgements We thank Tricia Solito and Sondra Spiegl for excellent technical assistance and Drs Michael Felder and Richard Showman for comments on the manuscript. We also thank Dr Nancy Colburn for providing us with the TIMP-3 promoter construct, Dr Arnold Levine for the 10(3) cells, and Dr Xinbin Chen for the doxycycline-responsive cells lines. This work was supported by the Elsa U Pardee Foundation; NIH/BRIN P20 RR16461 and a Sloan Foundation Pre-Doctoral Fellowship (S.T.).
References [1] B. Vogelstein, D. Lane, A.J. Levine, Surfing the p53 network, Nature 408 (2000) 307–310. [2] N.E. Sharpless, R.A. DePinho, p53: good cop/bad cop, Cell 110 (2002) 9–12. [3] K.H. Vousden, X. Lu, Live or let die: the cell’s response to p53, Nat. Rev. Cancer 2 (2000) 594–604. [4] S.P. Hussain, C.C. Harris, Molecular epidemiology of human cancer: contribution of mutation spectra studies of tumor suppressor genes, Cancer Res. 58 (1998) 4023–4037. [5] D. Dittmer, S. Pati, G. Zambetti, S. Chu, A.K. Teresky, M. Moore, et al., Gain of function mutations in p53, Nat. Genet. 4 (1993) 42–46. [6] G. Liu, T.J. McDonnell, R. Montes de Oca Luna, M. Kapoor, B. Mims, A.K. El-Naggar, et al., High metastatic potential in mice inheriting a targeted p53 missense mutation, Proc. Natl Acad. Sci. USA 97 (2000) 4174–4179. [7] K.P. Olive, D.A. Tuveson, Z.C. Ruhe, B. Yin, N.A. Willis, R.T. Bronson, et al., Mutant p53 gain of function in two mouse models of Li-Fraumeni, Cell 119 (2004) 847–860. [8] G.A. Lang, T. Iwakuma, Y.A. Suh, G. Liu, V.A. Rao, J.M. Parant, et al., Gain of function of a p53 hot spot mutation in a mouse model of Li-Fraumeni, Cell 119 (2004) 861–872. [9] E. Kim, W. Deppert, Transcriptional activities of mutant p53: when mutations are more than a loss, J. Cell. Biochem. 93 (2004) 878–886. [10] K.H. Vousden, C. Prives, P53 and prognosis: new insights and further complexity, Cell 120 (2005) 7–10. [11] J. Gagnebin, H. Kovar, A.V. Kajava, A. Estreicher, P. Monnierand, R. Iggo, Use of transcription reporters with novel p53 binding sites to target tumor cells expressing endogenous or virally transduced p53 mutants with altered sequence-specificity, Oncogene 6 (1998) 685–690. [12] H. Koga, W. Deppert, Identification of genomic DNA sequences bound by mutant p53 protein (Gly245-Ser245), Oncogene 19 (2000) 4178–4183. [13] P. Friedlander, Y. Haupt, C. Prives, M. Oren, A mutant p53 that discriminates between p53-responsive genes cannot induce apoptosis, Mol. Cell. Biol. 16 (1996) 4961–4971.
58
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
[14] T. Loging, D. Reisman, Inhibition of the putative tumor suppressor gene TIMP-3 by an oncogenic tumor derived p53 mutant, Oncogene 18 (1999) 7608–7615. [15] L. Weisz, A. Zalcentstein, P. Stamolsky, Y. Cohen, N. Goldfinger, M. Oren, et al., Transactivation of the EGR1 gene contributes to mutant p53 gain of function, Cancer Res. 64 (2004) 8318–8327. [16] M.J. Scian, K.E. Stagliano, D. Deb, E.H. Carchman, A. Das, K. Valerie, et al., Tumor derived p53 mutants induce oncogenesis by transactivating growth-promoting genes, Oncogene 23 (2004) 4430–4443. [17] M.J. Scian, K.E. Stagliano, M.A. Ellis, S. Hassan, M. Bowman, M.F. Miles, et al., Modulation of gene expression by tumor derived p53 mutants, Cancer Res. 64 (2004) 7447–7454. [18] K.V. Chin, K. Ueda, I. Pastan, M.M. Gottesman, Modulation of activity of the promoter of the human MDR1 gene by ras and p53, Science 255 (1992) 459–462. [19] M.W. Frazier, X. He, J. Wang, Z. Gu, J.L. Cleveland, G.P. Zambetti, Activation of c-myc gene expression by tumorderived p53 mutants requires a discrete C-terminal domain, Mol. Cell. Biol. 18 (1998) 3735–3743. [20] K. Lee, A. Crowe, M. Barton, p53-mediated repression of alpha-fetoprotein gene expression by specific DNA binding, Mol. Cell. Biol. 19 (1999) 1279–1288. [21] J. Foley, J.J. Wysolmerski, A.E. Broadus, W.M. Philbrick, Parathyroid hormone-related protein gene expression in human squamous carcinoma cells is repressed by mutant isoforms of p53, Cancer Res. 56 (1996) 4056–4062. [22] A. Zalcenstein, P. Stamolsky, L. Weisz, M. Muller, D. Wallach, D. Goncharov, et al., Mutant p53 gain of function: repression of CD95(Fas/APO-1) gene expression by tumor associated p53 mutants, Oncogene 22 (2003) 5667–5676. [23] K. Will, G. Warnecke, L. Wiesmuller, W. Deppert, Specific interaction of mutant p53 with regions of matrix attachment region DNA elements (MARs) with a high potential for baseunpairing, Proc. Natl Acad. Sci. USA 95 (1998) 13681–13686. [24] A. Gualberto, K. Aldape, K. Kozakiewics, T.D. Tlsty, An oncogenic form of p53 confers a dominant, gain-of-function phenotype that disrupts spindle checkpoint control, Proc. Natl Acad. Sci. USA 95 (1998) 5166–5171. [25] C.J. DiComo, C. Gaiddon, C. Prives, p73 function is inhibited by tumor-derived p53 mutants in mammalian cells, Mol. Cell. Biol. 19 (1999) 1438–1449. [26] S. Strano, E. Munarriz, M. Rossi, B. Cristofanelli, Y. Shaul, L. Castagnoli, et al., Physical and functional interaction between p53 mutants and different isoforms of p73, J. Biol. Chem. 275 (2000) 29503–29512. [27] C. Gaiddon, M. Lokshin, J. Ahn, T. Zhang, C. Prives, A subset of tumor-derived mutant forms of p53 down-regulate p63 and p73 through a direct interaction with the p53 core domain, Mol. Cell. Biol. 21 (2001) 1874–1887. [28] S. Strano, G. Fontemaggi, A. Costanzo, M.G. Rizzo, O. Monti, A. Baccarini, et al., Physical interaction with human tumorderived p53 mutants inhibits p63 activities, J. Biol. Chem. 277 (2002) 18817–18826.
[29] F.F. Mohammed, D.S. Smookler, S.E. Taylor, B. Fingleton, Z. Kassiri, O.H. Sanchez, et al., Abnormal TNF activity in TimpK/K mice leads to chronic hepatic inflammation and failure of liver regeneration, Nat. Genet. 36 (2004) 969–977. [30] K. Brew, D. Dinakarpandian, H. Nagase, Tissue inhibitors of metalloproteinases: evolution, structure and function, Biochim. Biophys. Acta 1477 (2000) 267–283. [31] T. Andreu, T. Beckers, E. Thoenes, P. Hilgard, H. von Melchner, Gene trapping identifies inhibitors of oncogenic transformation. The TIMP3 and collagen type I alpha2 (COL1A2) are epidermal growth factor-regulated growth repressors, J. Biol. Chem. 273 (1998) 13848–13854. [32] J. Bian, C. Jacobs, Y. Wang, Y. Sun, Characterization of a putative p53 binding site in the promoter of the mouse tissue inhibitor of metalloproteinases-3 (TIMP-3) gene: TIMP-3 is not a p53 target gene, Carcinogenesis 17 (1996) 2559–2562. [33] W.D. Pennie, G.A. Hegamyer, M.R. Young, N.H. Colburn, Specific methylation events contribute to the transcriptional repression of the mouse tissue inhibitor of metalloproteinases-3 gene in neoplastic cells, Cell Growth. Differ. 10 (1999) 279–286. [34] S. Zochbauer-Muller, K.M. Fong, A.K. Virmani, J. Geradts, A.F. Gazdar, J.D. Minna, Aberrant promoter methylation of multiple genes in non-small cell lung cancers, Cancer Res. 61 (2001) 249–255. [35] E. Dulaimi, I. Ibanez de Caceres, R.G. Uzzo, T. Al-Saleem, R.E. Greenberg, T.J. Polascik, et al., Promoter hypermethylation profile of kidney cancer, Clin. Cancer Res. 10 (2004) 3972–3979. [36] Y. Sun, G. Hegamyer, H. Kim, K. Sithanandam, H. Li, R. Watts, et al., Molecular cloning of mouse tissue inhibitor of metalloproteinases-3 and its promoter. Specific lack of expression in neoplastic JB6 cells may reflect altered gene methylation, J. Biol. Chem. 270 (1995) 19312–19319. [37] W.S. El-Diery, Regulation of p53 downstream genes, Semin. Cancer Biol. 8 (1998) 345–357. [38] B. Anand-Apte, L. Bao, R. Smith, K. Iwata, B.R. Olsen, B. Zetter, et al., A review of tissue inhibitor of metalloproteinases-3 (TIMP-3) and experimental analysis of its effect on primary tumor growth, Biochem. Cell Biol. 74 (1996) 853–862. [39] L.M. Matrisian, Cancer biology: extracellular proteinases in malignancy, Curr. Biol. 9 (1999) 776–778. [40] L.M. Matrisian, G.W. Sledge, S. Mohla, Extracellular proteolysis and cancer: meeting summary and future directions, Cancer Res. 63 (2003) 6105–6109. [41] H. Sato, T. Takino, Y. Okada, J. Cao, A. Shinagawa, E. Yamamoto, et al., A matrix metalloproteinase expressed on the surface of invasive tumor cells, Nature 370 (1994) 61–65. [42] P.C. Brooks, S. Stromblad, L.C. Sanders, T.L. von Schalscha, R.T. Aimes, W.G. Stetler-Stevenson, et al., Localization of matrix metalloproteinase MMP-2 to the surface of invasive cells by interaction with integrin avb3, Cell 85 (1996) 683–693.
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59 [43] C.P. Blobel, Metalloproteinase-disintigrins: links to cell adhesion and cleavage of TNFa and notch, Cell 90 (1997) 589–592. [44] H. Ueno, H. Nakamura, M. Inoue, K. Imai, M. Noguchi, H. Sato, et al., Expression and tissue localization of membranetypes 1,2, and 3 matrix metalloproteinases in human invasive breast carcinomas, Cancer Res. 57 (1997) 2055–2060. [45] L.M. Coussens, B. Fingleton, L.M. Matrisian, Matrix metalloproteinase inhibitors and cancer: trials and tribulations, Science 295 (2002) 2387–2392.
59
[46] Q. Yu, I. Stamenkovic, Cell surface-localized matrix metalloproteinase-9 proteolytically activates TGF-b and promotes tumor invasion and angiogenesis, Genes Dev. 14 (2000) 163–176. [47] D.A. Freedman, A.J. Levine, Regulation of the p53 protein by the MDM2 oncoprotein, Cancer Res. 59 (1999) 1–7. [48] K.W. Kinzler, B. Vogelstein, Colorectal tumors in: K.W. Kinzler, B. Vogelstein (Eds.), The Genetic Basis of Human Cancers, McGraw-Hill, New York, 1998, pp. 565–587.
Localization of a mutant p53 response element on the Tissue Inhibitor of Metalloproteinase-3 promoter: Mutant p53 activities are distinct from wild-type Shana Thomas, David Reisman* Department of Biological Sciences, University of South Carolina, Columbia, SC 29208, USA Received 11 July 2005; received in revised form 30 August 2005; accepted 31 August 2005
Abstract Missense mutations in the p53 gene have been observed in greater than 60% of all human tumors. Recent evidence indicates that some mutations in p53 arise as the cancer progresses from a benign tumor to a metastatic tumor and that these mutations in p53 actively contribute to the process of cancer progression. Previously, we reported that the expression of the gene encoding the tissue inhibitor of metalloproteinase-3 (TIMP-3) is repressed in cells expressing codons 248 and 281 mutant p53 alleles. The ability of tumor-derived p53 mutants to inhibit TIMP-3 expression provides a novel mechanism for understanding how p53 mutations might contribute to tumorigenesis. Since mutant p53 is often expressed at elevated levels in a variety of cancers, the generation of cells in a tumor carrying certain mutations in p53 would cause inappropriately reduced expression of TIMP-3 and lead to elevated matrix metalloproteinase activity. We present the results of experiments that begin to determine the mechanism by which mutant p53 represses TIMP-3 gene expression. By generating deletion derivatives of the TIMP-3 promoter and testing them for expression and by performing DNA protein binding assays on the regions determined to be required for repression, we have identified elements that are essential for mutant p53-mediated transcriptional repression. These elements respond specifically to mutant but not wild type p53. While mutant p53 itself does not bind to the TIMP-3 promoter, we provide evidence for the presence of DNA binding proteins whose activity is enhanced in the presence of mutant p53. q 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Gene expression; p53 tumor-suppressor; Oncogenic mutants; Metalloproteinase
1. Introduction The normal or wild type p53 gene is a tumor suppressor gene, which encodes a protein that regulates a cell cycle checkpoint in response to * Corresponding author. Tel.: C1 803 777 8108; fax: C1 803 777 4002. E-mail address: [email protected] (D. Reisman).
DNA damage, cell stress or the aberrant expression of some oncogenes (for reviews see Refs. [1,2]). Depending on the cell type and extent of DNA damage, expression of wild-type p53 induces either a cell cycle arrest or apoptosis [3]. Missense mutations in the p53 gene, which inactivate its growth suppressing activities, have been observed in greater than 60% of all human tumors [4]. Interestingly, in a number of tumor types,
0304-3835/$ - see front matter q 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2005.08.027
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
such as colorectal and ovarian, the frequency of p53 mutations appears to increase as the cancer progresses from benign to metastatic tumors, suggesting that mutations in p53 may be contributing to the process of cancer progression. In agreement with this notion, a number of studies indicate that tumor-derived mutant forms of p53, while losing many of their DNAdamage checkpoint functions, serve as active transforming genes [4,5]. It has been shown, for example, that introduction of mutant p53 into p53 negative immortalized cells can increase their growth rates in culture and transform them into cells that are tumorigenic in vivo [5,6]. Most recently, mice expressing p53 missense mutations, in an otherwise p53-null background, develop metastatic tumors not seen in p53-null mice [7,8]. These results clearly demonstrate that activities expressed by some mutant p53 alleles contribute to tumorigenesis, and can therefore, be considered oncogenic (for recent reviews see Refs. [9,10]). Supporting the hypothesis that some p53 mutant alleles can contribute novel activities to cells, p53 mutants have been described that have altered DNA binding specificities [11–13] and many reports have described p53 mutant alleles that can alter the transcription of genes that are not normal targets of wild-type p53 [14–17]. Genes that have been reported to be modulated in their expression by mutant p53 include the human multi-drug resistance gene (MDR1; [18]), c-myc [19], tissue inhibitor of metalloproteinase-3 [14], insulin-like growth factor II [20], parathyroid hormone-related protein [21], and CD95FAS/ APO-1 [22]. In addition to affecting gene expression, mutant p53 can express other novel activities that include binding to DNA elements of the nuclear matrix [23] and disrupting spindle formation checkpoints [24]. Finally, the ability of mutant p53 to complex with the family of p53 homologues, p73abg and p63 and interfere with their activities, is likely to further contribute to the transforming potential of mutant p53 [25–28]. Previously, we reported that the gene encoding the tissue inhibitor of metalloproteinase-3 (TIMP-3) is repressed in cells expressing either wild type p53 or the codons 248 (R248W) and 281 (D281G) mutant p53 alleles. We propose that although wild-type p53 can repress TIMP-3 expression, it is mutant p53 that is
49
relevant to tumor progression since wild type p53 is expressed at exceedingly low levels in the absence of DNA damage. Whereas mutant p53 is expressed at very high levels in tumor cells and thus is much more likely to have an effect on TIMP-3 expression. TIMP-3 in fact has multiple physiological functions and in addition to inhibiting the matrix metalloproteinases has a newly established role in the inflammatory response and appears to be a regulator of tumor-necrosis factor alpha [29]. Wild type p53 expression may potentially have a role in this pathway in normal cells, but this remains to be explored. The finding that some tumor-derived p53 mutants can inhibit TIMP-3 expression therefore provides evidence for a novel function of mutant p53 and how these mutations could contribute to tumor progression. Since mutant p53 is often expressed at elevated levels in a variety of cancers, the generation of cells in a non-invasive tumor carrying a mutation in p53 would, in addition to losing crucial wild-type p53 mediated cell cycle checkpoints, cause inappropriate reduced expression of TIMP-3. As TIMP-3 has been shown to inhibit the activity of matrix metalloproteinases [30], repressed TIMP-3 expression would be predicted to result in elevated matrix metalloproteinase activity that, in turn, could lead to increased tissue degradation and an increased likelihood of metastasis. Interestingly, the loss of TIMP-3 expression has been observed in many cancers and its loss has been associated with increased metastatic potential [31–35]. While much is known about the biochemical and gene regulatory activities of the wild-type p53 protein, relatively little is known about mutant p53 and its contribution to the transformed phenotype. Here, we present the results of experiments, where we begin to determine the mechanism by which mutant p53 represses TIMP-3 gene expression. By generating a series of deletion derivatives of the TIMP-3 promoter and testing them for p53 responsiveness as well as by performing DNA protein binding assays, we have identified elements that are essential for mutant p53-mediated transcriptional repression. While mutant p53 does not bind to the TIMP-3 promoter, we provide evidence for the presence of DNA binding proteins that likely contribute to TIMP-3 repression.
50
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
2. Materials and methods 2.1. Cell culture 10(3) cells were kindly provided by Dr Arnold Levine (Princeton University). These cells are murine fibroblasts that lack p53 expression. 10(3) cells were maintained in Dulbecco’s modified Eagle’s media (DMEM) and supplemented with 10% fetal bovine serum (FBS), 2 mM glutamine, 100 m/ml penicillin and 100 mg/ml streptomycin. CMV248 Cl.2 cells are murine fibroblasts that contain a colon carcinoma-derived p53 mutation (codon 248 Arg to Trp; R248W). These cells constitutively express mutant p53 through the control of the cytomegalovirus (CMV) promoter. CMV248 Cl.2 cells were maintained in DMEM and supplemented with 10% FBS, 2 mM glutamine, 50 mg/ml G418, 100 m/ml penicillin, and 100 mg/ml streptomycin. Human colon carcinoma cell lines containing the R248W p53 mutation Colo 320 HSR, Colo 320 DM, and SW 837 were grown in RPMI 1640 media with 10% FBS, 2 mM glutamine plus penicillin and streptomycin. 2.2. Western blot analysis Protein extracts were prepared from 10(3), SVT2, HT-29, and H1299 cells. The cells were washed with PBS, harvested and lysed in RIPA buffer (1 mM Tris–HCl pH 7.4, 150 mM NaCl, 1% sodium deoxycholate, 0.1% SDS, and 1 mM PMSF). The cell lysate (25–50 mg) was run on a 12% denaturing polyacrylamide gel in Tris/Glycine/SDS running buffer along with an appropriate standard. The extracts were then transferred to a nitrocellulose membrane. The membranes were blocked with 5% non-fat dry milk for 1 h, washed with TBST (1 M Tris–HCl pH 8, 5 M NaCl, and 0.5 ml/L Tween-20), and incubated with primary antibody PAb421 for 1 h. The membrane was washed again with TBST and incubated with its respective secondary antibody conjugated to horse-radish peroxidase. The signals were visualized after exposing X-ray films using ECL Western Blotting Detecting Reagents (Amersham). 2.3. Tetracycline inducible p53 expression The human non-small cell lung carcinoma-derived cell, line H1299, expressing mutant p53 isoforms
from a tetracycline-withdrawal inducible promoter, was kindly provided by Dr Xinbin Chen (University of Alabama). The cells were maintained in DMEM media containing 10% FBS, 2 mM glutamine, 250 mg/ml G418, 2 mg/ml puromycin, 2 mg/ml of the tetracycline derivative, doxycycline, plus penicillin and streptomycin. Under these conditions expression of p53 was low or undetectable. To induce p53 expression, cells were washed four times with complete media lacking doxycycline. The cells were then grown for additional lengths of time (3, 6, 12, 24 h) in complete media lacking doxycycline prior to harvesting for analysis of p53 expression by Western blotting (as previously described). To assay for TIMP-3 promoter activity, cells growing in the presence of doxycycline were transfected with the TIMP-3 luciferase reporter vector. Twenty-four hours after transfection, the cells were washed to remove doxycycline and after an additional 24 h of growth in the absence of doxycycline, extracts were prepared and equal amounts (protein) were assayed for luciferase activity. An internal control expressing renilla from the HSV-TK promoter (TK-renilla) was included in all transfections and used to normalize the transfection efficiencies. Transfection experiments were repeated four times. 2.4. Luciferase reporter constructs The murine TIMP-3 promoter [36] was kindly provided by Dr Nancy Colburn (National Cancer Institute). TIMP-3 promoter deletion constructs were generated by PCR. Primers were designed from the 2900 bp full length promoter sequence and used to amplify regions of the promoter that were deleted from the 5 0 -end. These deletion products were cloned into the pGL3 expression vector (Promega) upstream of the luciferase reporter gene. 2.5. Transient transfections and reporter gene assays Transfections were performed using Transfast Transfection Reagent (Promega) in a 1:1 ratio of total DNA to Transfast. Exponentially growing 10(3), CMV248 Cl.2, Colo 320, Colo 320 DM, and SW 837 cells were plated at a density of 5!104 cells/well in a 24 well plate and grown to 80% confluence. The 10(3) cells were cotransfected with 0–1.0 mg/well mutant
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
p53 (CMV248) and 0.25 mg/well of the TIMP-3 deletion constructs. The CMV248 Cl.2, Colo 320, DM 320, and SW 837 cells were transfected with 0.25 mg/well of the TIMP-3 deletion constructs. Twenty-four hours post transfection, cells were harvested and luciferase activity detected using a Luminometer. Lysates were normalized by measuring TK-renilla activity as an internal control. Relative fold repression was calculated by dividing the activities of the TIMP-3 constructs in the presence of mutant p53 by the activities of the constructs in the absence of mutant p53. All transfections were performed in triplicate.
2.6. Nuclear extracts and DNA-binding assays Nuclear extracts were obtained from 10(3), CMV248 Cl.2, Colo 320, Colo 320 DM, and SW 837 cells by washing plates (80% confluent) twice with 10 mL of phosphate buffered saline (PBS). Cells were then lysed on the plate using 1 mL of lysis buffer (20 mM Hepes, pH 7.6, 10 mM NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, 20% glycerol, 0.1% Triton X100, 1 mM DTT, 1 mM PMSF, 1 mg/mL leupeptin, 1 mg/mL pepstatin, and 1 mg/mL aprotinin). Cells were collected by scraping and pelleted by centrifugation for 5 min at 2000 rpm at 4 8C. The nuclei were gently rocked for 1 h at 4 8C and centrifuged at 10,000 rpm for 10 min. The supernatant was aliquoted, quick frozen in liquid nitrogen and stored at K70 8C. Four 25 base pair double-stranded DNA oligonucleotides spanning the K2900 to K2800 nucleotide region of the TIMP-3 promoter were constructed and endlabeled with g-32P-ATP using T4 polynucleotide kinase. Probes (20 fmole/reaction) were added to binding buffer (50 mM Tris–HCl pH 7.0, 0.1 M KCl 12.5 mM MgCl2, 1.0 mM EDTA, 20% glycerol, and 1 mM DTT) and 0–10 mg of nuclear extract. The reaction was mixed and incubated on ice for 15 min and then at 20 8C for 15 min. For supershift experiments, anti-p53 antibody PAb421 was added to the reaction and incubated at 4 8C for an additional 20 min. The DNA-protein complexes were subjected to electrophoresis on a 4% polyacrylamide gel in 0.5X Tris/Borate/EDTA buffer. The gels were dried and subjected to autoradiography.
51
3. Results 3.1. Induced expression of mutant p53 results in repression of TIMP-3 promoter activity In order to confirm that TIMP-3 expression is inhibited by mutant p53 expression and to test another tumor-derived mutant for this activity, we took advantage of a tetracycline-regulated expression system. This expression system makes use of the herpes simplex virus transactivator, VP16, fused to the Tn10-encoded tetracycline repressor (tetR). In the presence of 1.0 mg/ml of the tetracycline derivative, doxycycline, the fusion protein is inhibited from binding to operator sequences of the tetracycline operon. Upon removal of doxycycline from the media, maximal activation of the gene can be obtained within 12–18 h. A construct expressing the codon 249 p53 mutant (R249S) from the tet-responsive promoter was introduced into H1299, a p53-null human lung carcinoma cell line (generously provided by Dr Xinbin Chen, University of Alabama). These p53-null cells were used in order to eliminate the influence of any endogenous p53 expression on TIMP-3 expression. The H1299 cells carrying the mutant p53 expression construct were tested for their inducibility upon removal of doxycycline from the media. At various time points (3, 6, 12, and 24 h), cells were harvested and assayed for p53 expression by Western transfer analysis. As shown in Fig. 1A, mutant p53 expression was dramatically induced in these cell lines upon withdrawal of doxycycline. These cells were then assayed for repression of the TIMP-3 promoter by transient transfections with a TIMP-3 luciferase reporter vector. As seen in Fig. 1B, the expression of the TIMP-3 promoter was significantly repressed upon induced expression of the R249W mutant p53. These results confirm our earlier findings [14] that, in addition to the codon R248W mutation, expression of other mutant p53 alleles (such as those with mutations at codons 281 and 249), result in repression of the TIMP-3 promoter. To again confirm that the TIMP-3 promoter is repressed by mutant (R248W) p53, a transfection was performed using p53-null 10(3) cells. These 10(3) cells were cotransfected with 0.25 mg of the TIMP-3 luciferase construct and increasing
52
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
concentrations of the mutant p53 CMV-based expression vector (0–1.0 mg). After 24 h, the cells were harvested and the luciferase activity was determined and plotted in relative light units. Fig. 1C shows that the full length (2900 bp) TIMP-3 promoter is repressed by mutant p53 in a dose dependent manner. To determine the relative levels of p53 protein at each DNA concentration, a Western blot was performed using anti-p53 antibody PAb421 (Fig. 1C, inset). As the concentration of mutant p53 DNA increases from 0 mg (lane 1), 0.05 mg (lane 2), 0.1 mg (lane 3), 0.25 mg (lane 4), 0.5 mg (lane 5), to 1.0 mg (lane 6) the corresponding protein levels also increase (top panel). Actin was used as a loading control and is pictured in the bottom panel. 3.2. Localization of a mutant p53 response element in the TIMP-3 promoter
Fig. 1. Induction of mutant p53 from a doxycycline (Dox) repressed promoter leads to inhibition of expression of the TIMP-3 promoter. H1299 cells were grown in the presence of 2 mg/ml Dox or in the absence of Dox for 3, 6, 12, and 24 h. (A) At the times indicated, cells were harvested and extracts were assayed for levels of p53 protein by Western transfer. Extracts of the human colon carcinoma cell line HT-29 were used as a positive control for p53 protein. The arrow indicates the position of p53. (B) Twenty-four hours after removal of Dox, cells were transfected with the TIMP-3-luciferase reporter vector. After an additional twenty-four hours, cells were harvested and assayed for luciferase activity. Cells with Dox express low levels of p53 and approximately 5–6 fold elevated levels of TIMP-3. An internal control, TK-renilla, showed no changes in expression. The results are the average of four experiments. (C) 10(3) cells were cotransfected with 0.25 mg of
In order to identify which region or regions of the TIMP-3 promoter are required for the repression observed in response to mutant p53, we generated a series of deletion derivatives of the TIMP-3 promoter (Fig. 2) and tested them for their response to either wild-type or mutant p53 expression. Deletions that progressively removed an increasing amount of DNA from the 5 0 -end of the promoter were generated by PCR. The resulting deletions which removed 600, 1000, 2000 and 2500 bp were subcloned into the pGL3 luciferase reporter vector and transfected into 10(3) cells in the presence of a either a R248W or wtp53 CMV-based expression vector. After 24 h, cells were harvested and extracts prepared and tested for luciferase activity. As shown in Fig. 3A, repression of TIMP-3 promoter activity by mutant p53 was eliminated with the deletion of 600 bp closest to the 5 0 -end of the promoter. Interestingly, repression by wt 3 the TIMP-3 promoter construct and 0–1.0mg mutant p53 expression vector (CMV 248). Twenty-four hours post transfection, cells were harvested and luciferase activity was determined. (Inset) A Western Blot was performed on extracts taken from the transfected cells and probed with anti-p53 antibody PAb421. The blot depicts the increasing level of mutant p53 from 0 mg (lane 1), 0.05 mg (lane 2), 0.1 mg (lane 3), 0.25 mg (lane 4), 0.5 mg (lane 5), to 1.0 mg (lane 6). Extract from a p53-positive control cell line (SVT2) is pictured in lane 7. An anti-actin antibody was used to detect actin and is shown as a loading control.
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
53
Fig. 2. Illustration of the 2900 bp TIMP-3 promoter with putative regulatory sites. Also shown are the two series of deletions that were generated by PCR and cloned into the luciferase expression vector to determine activity.
p53 was unaffected by any of the deletions and was still observed after deletion of 2500 bp. Since only approximately 400 bp of the TIMP-3 promoter remains on this last deletion derivative and it remains responsive to wt p53 expression, we conclude that repression by the mutant and wild-type p53 proteins are occurring through distinct elements in the TIMP-3 promoter, and therefore via different mechanisms. While repression by wt p53 may be occurring through elements of general transcription complex, as seen with other promoters [37], the repression by mutant p53 is occurring via an as yet unknown mechanism. Neither the wild-type nor mutant mediated repression appears to occur by virtue of binding to a predicted p53 consensus recognition site in the TIMP promoter at approx—650. Although wt p53 can bind to this site in vitro [33; plus data not shown], it has no effect on expression of the promoter. To further investigate the p53 mediated repression of TIMP-3 and to determine whether the 600-bp site at the 5 0 end of the promoter is sufficient to render a heterologous promoter responsiveness to mutant p53, we cloned this element upstream of the TK promoter and cotransfected this construct into p53 null 10(3) cells along with mutant p53 (CMV 248). Repression of TIMP-3 was not seen with this construct (data not shown). In view of this result, we reasoned that additional cooperating and essential elements may exist on the TIMP-3 promoter itself. Therefore in experiments described below, we chose to reconstruct a responsive TIMP-3 promoter rather than the heterologous TK promoter. In order to locate, where the R248W-responsive element is located within the first 600 bp of the TIMP-3 promoter, we generated an additional series of deletions that progressively removed 100 bp from the intact TIMP promoter. These deletions were again
tested for their ability to be repressed by the presence of mutant p53. As shown in Fig. 3B, deletion of 100 bp from the 5 0 -end of the promoter resulted in the loss of repression by mutant p53. Since further deletions of 200, 300, 400, and 500 bp from the promoter also maintained little to no activity, it was concluded that repression of the TIMP-3 promoter by mutant p53 is mediated through 100 bp of DNA sequences located approximately K2800 to K2900 bp upstream of the transcription start site. To provide further evidence that the K2800 to K2900 bp region of the TIMP-3 promoter mediates transcriptional repression by mutant p53, the construct containing the TIMP-3 promoter lacking the first 100 bp was tested for its activity in tumor-derived cell lines expressing mutant p53. As shown in Fig. 3C, the relative activity of the K100 bp TIMP-3 promoter is significantly greater than that of the full length TIMP-3 promoter in tumor-derived cells naturally expressing mutant p53. Again these results indicate that a negative regulatory element is located within the first 100 bp of the promoter and that this element exerts its effect in cells expressing physiologically occurring levels of mutant p53 protein. 3.3. Nuclear factors bind to the mutant p53responsive element in the TIMP-3 promoter Having determined that a 100-bp element is required for mutant p53 mediated repression of TIMP-3; we next asked whether we could identify nuclear proteins that bind to this region of the promoter. If so, one or more of these factors might be responsible for carrying out the repression. In order to assay for DNA binding proteins, nuclear extracts were prepared from cells expressing the R248W
54
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
experiments, an oligonucleotide comprising the 100 bp 5 0 regulatory region was labeled with 32P and assayed for its ability to bind nuclear proteins by electrophoretic mobility shift assays. Next, the 100 bp region was divided into three overlapping fragments, labeled with 32P and used to perform mobility shift assays (data not shown). Upon detection of binding activity, smaller regions were generated until ultimately a series of approximately 25-bp double stranded oligonucleotides was synthesized (Fig. 4A), labeled with 32P and used as probes in electrophoretic mobility shift assays. (Fig. 4B). All four oligonucleotide probes demonstrated binding, but only TIMP(3)1 and TIMP(3)4 were found to bind specifically to nuclear proteins. Specificity was demonstrated by competition for binding by either a 10- or 50-fold molar excess of the homologous unlabeled oligonucleotide or by a 10- or 50-fold molar excess of a random unlabeled oligonucleotide. Regions TIMP3(2) and TIMP3(3) were found to bind numerous factors, but all were shown to be nonspecific (data not shown). These results indicate that nuclear factors specifically interact with two regions (1 and 4) within the DNA sequences that are required for mutant p53-mediated repression. 3.4. Two nuclear factors bind to the TIMP-3 promoter selectively in cells expressing mutant p53
Fig. 3. Wild-type and mutant p53 repress TIMP-3 promoter activity through separate regions of the promoter. (A) Cells were cotransfected with 0.25 mg mutant p53 (CMV248) or wt p53 expression vectors and 0.25 mg of TIMP-3 promoter deletions. Twenty-four hours post transfection, cells were harvested and luciferase activity was determined. The size if the deletions are indicated. (B) Cells were co-transfected with 0.25 mg of mutant p53 and 0.25 mg of TIMP-3 promoter deletions. (C) The full length TIMP-3 promoter or the K100 bp deletion was transfected into 10(3) cells (transiently expressing mutant p53) and cells expressing endogenous mutant p53 (CMV248 Cl.2, Colo 320, and SW 837). Values were normalized to TK-Renilla expression. All transfections were done in triplicate.
mutant. These cells included the CMV248 Cl2 murine cell line that expresses mutant p53 from the CMV promoter as well as Colo 320 and SW 837 cells that are derived from human colon cancers and also express the R248W mutant. In a first series of
Since the TIMP-3 promoter is repressed when mutant p53 is expressed, we wanted to determine whether either of the two factors described above would bind to their target DNA in response to mutant p53 expression. To ask this question, we again carried out DNA binding assays, using binding sites, TIMP(3) 1 and TIMP(3)4, and nuclear extracts derived from R248W expressing cells and p53-null cells. As shown in Fig. 4C, while a slower migrating DNA-protein complex was observed in all cells tested, faster migrating complexes, shown to be specific (Fig. 4B), were only observed in cells expressing the R248W mutant. These two sites are candidates for sites that mediate repression of TIMP-3 in response to mutant p53 expression. To further examine the role of mutant p53 in this complex, a DNA binding assay was performed in the presence of an antibody (PAb421) previously demonstrated to shift protein-DNA complexes that contain p53 (data not shown).
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
55
Fig. 4. The mutant p53 responsive site on the TIMP-3 promoter binds two nuclear factors. (A) Four approximately 25 bp double-stranded oligonucleotides were synthesized spanning the p53 responsive region at the 5 0 end of the TIMP-3 promoter for use in binding and competition assays. The positions indicate the location of the sequence relative to the 5 0 -end of the cloned 2900-bp promoter. (B) Nuclear extracts from Colo 320 (R248W) cells were used to assay for protein binding to sites TIMP3(1), TIMP3(2), TIMP3(3), and TIMP3(4) using EMSA. Specific (SC) or nonspecific (NSC) competitors at (10- and 50-molar excess, C and C, respectively) were added to the reactions to test for specificity of binding. The specific complexes seen with TIMP3(1) and TIMP3(4) are indicated by arrows. (C) TIMP3(1) and TIMP3(4) were used to probe extracts from p53 null 10(3) cells (lanes 1 and 5), as well as R248W mutant p53 Colo 320 (lanes 2 and 6), Colo 320 DM (lanes 3 and 7), and SW 837 (lanes 4 and 8) cells. (D) EMSA was performed with Colo 320 (lanes 1–4) and SW 837 (lanes 5–8) nuclear extracts. These extracts were probed with binding sites TIMP3(1) and TIMP3(4) with and without the addition of anti-p53 antibody PAb421, where indicated.
56
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
4. Discussion
Fig. 5. Both cis-acting regions contribute to repression of TIMP-3 promoter activity. Concatermized regions TIMP3(1) and TIMP3(4) were cloned into the TIMP-3 promoter lacking the 100 bp p53 responsive element and tested for expression in transient transfection assays. SW 837 and CMV248 Cl.2 cells, and 10(3) cells transiently expressing mutant p53 were transfected with 0.25 mg/well of the full length (2900 bp), K100 (2800 bp), 2800 bp promoter with site TIMP3(1), or the 2800 bp promoter with site TIMP3(4).
As seen in Fig. 4D, when the antibody was added to either the Colo 320 or SW 837 nuclear extracts and probed with binding sites TIMP3(1) and TIMP3(4), no supershift was detected. Therefore, it was concluded that mutant p53 is not bound directly to this DNA–protein complex. 3.5. Each of the two novel factor binding sites contributes to repression of TIMP-3 expression In order to determine which or both of these binding sites are required for TIMP-3 repression, each site was concatomerized to yield three linked binding sites and cloned upstream of the 100-bp TIMP-3 deletion derivative that had lost its ability to be repressed by mutant p53. These reconstituted promoters controlling the luciferase reporter gene were cotransfected into p53-null 10(3) cells and into cells expressing mutant p53. As shown in Fig. 5, the inclusion of either 25-bp element led to a significant repression of promoter activity. This indicates that these elements serve to inhibit TIMP-3 expression. Since neither element reconstitutes complete mutant p53-mediated transcriptional repression, it is likely that both are required to be present to achieve the full response.
Our goal is to define the mechanism by which mutant p53 regulates gene expression and thus may contribute to the transformed phenotype. The data that we have presented indicate that although both wild type and tumor derived p53 mutants can repress the expression of TIMP-3, the sites on the TIMP-3 promoter through which these proteins act, are clearly distinct. Wild type p53 represses transcription through elements close to the transcription start site while mutant p53 represses transcription through elements located approximately 2800 bp upstream of the transcription start site. These data indicate that mutant p53, specifically the R248W p53 mutant, the most commonly occurring of the p53 mutations, represses transcription of TIMP-3 through two 25-bp elements near the 5 0 -end of the TIMP-3 promoter. Two nuclear factors were found to bind to these two elements and their binding activity appears to be present in cells expressing mutant p53, but not in cells lacking p53 expression. These studies therefore provide a framework by which we will be able to examine the activities of mutant p53. We are now beginning the purification of these DNA binding proteins. This will allow us to clone the genes for these factors and to explore their mechanism of action as well as the role that mutant p53 plays in the activity of these factors. TIMP-3 is one of four known tissue inhibitors of matrix metalloproteinases (MMPs) and is unique among them in its being secreted into the extracellular matrix [30,38]. The MMPs consist of a family of proteases that play a major role in the remodeling and turnover of the extracellular matrix as well as in angiogenesis and in the vascularization of tissues. The aberrant activity of MMPs, leading to a pathological degradation of the extracellular matrix, often appears to be an essential step in cancer progression and metastasis. The degradation of the extracellular matrix is necessary for tumor invasion and accordingly, the elevated activities of MMPs have in many cases been shown to be crucial for this process [39,40]. Appropriately, MMPs have been found on the surface or associated with a number of invasive and metastatic tumor cell types and the inhibition of MMP localization to the cell surface inhibits tumor invasiveness [41–46]
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
The ability of some tumor derived mutant forms of p53 to inhibit TIMP-3 expression thus provides a model for understanding how p53 mutations might contribute to tumor progression. Since mutant p53 is stabilized and accumulates to elevated levels in tumor cells [47], the generation of cells in a non-invasive tumor carrying a mutation in p53 could cause reduced expression of TIMP-3 and lead to elevated MMP activity. Interestingly, in some cancers, such as colorectal cancer, where p53 is found to be mutated and overexpressed in over 85% of cases, mutations in p53 have been shown to be relatively late events in tumor cell evolution and to be associated with increased metastatic potential [48]. Some mutant p53 alleles, such as codons 248, 249 and 281, may contribute to tumor progression by altering the expression of genes such as TIMP-3. The mechanism by which mutant p53 interfaces with the factors that we have identified remains to be characterized. Rather than directly interacting with the TIMP-3 promoter, the activity of the codon 248 p53 mutant appears to be mediated by a group of novel DNA binding proteins. Our current data indicate that mutant p53 may be responsible for activating, either transcriptionally or post-transcriptionally, one or more transcriptional repressors that inhibit TIMP-3 expression. One possibility is that mutant p53 interacts directly with the promoters of these genes and contributes to their elevated expression. Further work aimed at cloning the genes that encode these proteins and generating reagents to assay for their expression will allow us to address these problems and develop a much better understanding of the role of mutant p53 in cancer progression. 5. Summary Since reduced TIMP-3 expression has been demonstrated to contribute to tumor invasiveness and metastasis, the ability of tumor-derived p53 mutants to inhibit TIMP-3 expression provides a model for understanding how p53 mutations contribute to tumor progression. We have set out to determine how mutant p53 represses TIMP-3 gene expression and conclude from our findings that the activity of novel DNA binding proteins is enhanced in the presence of mutant p53.
57
Acknowledgements We thank Tricia Solito and Sondra Spiegl for excellent technical assistance and Drs Michael Felder and Richard Showman for comments on the manuscript. We also thank Dr Nancy Colburn for providing us with the TIMP-3 promoter construct, Dr Arnold Levine for the 10(3) cells, and Dr Xinbin Chen for the doxycycline-responsive cells lines. This work was supported by the Elsa U Pardee Foundation; NIH/BRIN P20 RR16461 and a Sloan Foundation Pre-Doctoral Fellowship (S.T.).
References [1] B. Vogelstein, D. Lane, A.J. Levine, Surfing the p53 network, Nature 408 (2000) 307–310. [2] N.E. Sharpless, R.A. DePinho, p53: good cop/bad cop, Cell 110 (2002) 9–12. [3] K.H. Vousden, X. Lu, Live or let die: the cell’s response to p53, Nat. Rev. Cancer 2 (2000) 594–604. [4] S.P. Hussain, C.C. Harris, Molecular epidemiology of human cancer: contribution of mutation spectra studies of tumor suppressor genes, Cancer Res. 58 (1998) 4023–4037. [5] D. Dittmer, S. Pati, G. Zambetti, S. Chu, A.K. Teresky, M. Moore, et al., Gain of function mutations in p53, Nat. Genet. 4 (1993) 42–46. [6] G. Liu, T.J. McDonnell, R. Montes de Oca Luna, M. Kapoor, B. Mims, A.K. El-Naggar, et al., High metastatic potential in mice inheriting a targeted p53 missense mutation, Proc. Natl Acad. Sci. USA 97 (2000) 4174–4179. [7] K.P. Olive, D.A. Tuveson, Z.C. Ruhe, B. Yin, N.A. Willis, R.T. Bronson, et al., Mutant p53 gain of function in two mouse models of Li-Fraumeni, Cell 119 (2004) 847–860. [8] G.A. Lang, T. Iwakuma, Y.A. Suh, G. Liu, V.A. Rao, J.M. Parant, et al., Gain of function of a p53 hot spot mutation in a mouse model of Li-Fraumeni, Cell 119 (2004) 861–872. [9] E. Kim, W. Deppert, Transcriptional activities of mutant p53: when mutations are more than a loss, J. Cell. Biochem. 93 (2004) 878–886. [10] K.H. Vousden, C. Prives, P53 and prognosis: new insights and further complexity, Cell 120 (2005) 7–10. [11] J. Gagnebin, H. Kovar, A.V. Kajava, A. Estreicher, P. Monnierand, R. Iggo, Use of transcription reporters with novel p53 binding sites to target tumor cells expressing endogenous or virally transduced p53 mutants with altered sequence-specificity, Oncogene 6 (1998) 685–690. [12] H. Koga, W. Deppert, Identification of genomic DNA sequences bound by mutant p53 protein (Gly245-Ser245), Oncogene 19 (2000) 4178–4183. [13] P. Friedlander, Y. Haupt, C. Prives, M. Oren, A mutant p53 that discriminates between p53-responsive genes cannot induce apoptosis, Mol. Cell. Biol. 16 (1996) 4961–4971.
58
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59
[14] T. Loging, D. Reisman, Inhibition of the putative tumor suppressor gene TIMP-3 by an oncogenic tumor derived p53 mutant, Oncogene 18 (1999) 7608–7615. [15] L. Weisz, A. Zalcentstein, P. Stamolsky, Y. Cohen, N. Goldfinger, M. Oren, et al., Transactivation of the EGR1 gene contributes to mutant p53 gain of function, Cancer Res. 64 (2004) 8318–8327. [16] M.J. Scian, K.E. Stagliano, D. Deb, E.H. Carchman, A. Das, K. Valerie, et al., Tumor derived p53 mutants induce oncogenesis by transactivating growth-promoting genes, Oncogene 23 (2004) 4430–4443. [17] M.J. Scian, K.E. Stagliano, M.A. Ellis, S. Hassan, M. Bowman, M.F. Miles, et al., Modulation of gene expression by tumor derived p53 mutants, Cancer Res. 64 (2004) 7447–7454. [18] K.V. Chin, K. Ueda, I. Pastan, M.M. Gottesman, Modulation of activity of the promoter of the human MDR1 gene by ras and p53, Science 255 (1992) 459–462. [19] M.W. Frazier, X. He, J. Wang, Z. Gu, J.L. Cleveland, G.P. Zambetti, Activation of c-myc gene expression by tumorderived p53 mutants requires a discrete C-terminal domain, Mol. Cell. Biol. 18 (1998) 3735–3743. [20] K. Lee, A. Crowe, M. Barton, p53-mediated repression of alpha-fetoprotein gene expression by specific DNA binding, Mol. Cell. Biol. 19 (1999) 1279–1288. [21] J. Foley, J.J. Wysolmerski, A.E. Broadus, W.M. Philbrick, Parathyroid hormone-related protein gene expression in human squamous carcinoma cells is repressed by mutant isoforms of p53, Cancer Res. 56 (1996) 4056–4062. [22] A. Zalcenstein, P. Stamolsky, L. Weisz, M. Muller, D. Wallach, D. Goncharov, et al., Mutant p53 gain of function: repression of CD95(Fas/APO-1) gene expression by tumor associated p53 mutants, Oncogene 22 (2003) 5667–5676. [23] K. Will, G. Warnecke, L. Wiesmuller, W. Deppert, Specific interaction of mutant p53 with regions of matrix attachment region DNA elements (MARs) with a high potential for baseunpairing, Proc. Natl Acad. Sci. USA 95 (1998) 13681–13686. [24] A. Gualberto, K. Aldape, K. Kozakiewics, T.D. Tlsty, An oncogenic form of p53 confers a dominant, gain-of-function phenotype that disrupts spindle checkpoint control, Proc. Natl Acad. Sci. USA 95 (1998) 5166–5171. [25] C.J. DiComo, C. Gaiddon, C. Prives, p73 function is inhibited by tumor-derived p53 mutants in mammalian cells, Mol. Cell. Biol. 19 (1999) 1438–1449. [26] S. Strano, E. Munarriz, M. Rossi, B. Cristofanelli, Y. Shaul, L. Castagnoli, et al., Physical and functional interaction between p53 mutants and different isoforms of p73, J. Biol. Chem. 275 (2000) 29503–29512. [27] C. Gaiddon, M. Lokshin, J. Ahn, T. Zhang, C. Prives, A subset of tumor-derived mutant forms of p53 down-regulate p63 and p73 through a direct interaction with the p53 core domain, Mol. Cell. Biol. 21 (2001) 1874–1887. [28] S. Strano, G. Fontemaggi, A. Costanzo, M.G. Rizzo, O. Monti, A. Baccarini, et al., Physical interaction with human tumorderived p53 mutants inhibits p63 activities, J. Biol. Chem. 277 (2002) 18817–18826.
[29] F.F. Mohammed, D.S. Smookler, S.E. Taylor, B. Fingleton, Z. Kassiri, O.H. Sanchez, et al., Abnormal TNF activity in TimpK/K mice leads to chronic hepatic inflammation and failure of liver regeneration, Nat. Genet. 36 (2004) 969–977. [30] K. Brew, D. Dinakarpandian, H. Nagase, Tissue inhibitors of metalloproteinases: evolution, structure and function, Biochim. Biophys. Acta 1477 (2000) 267–283. [31] T. Andreu, T. Beckers, E. Thoenes, P. Hilgard, H. von Melchner, Gene trapping identifies inhibitors of oncogenic transformation. The TIMP3 and collagen type I alpha2 (COL1A2) are epidermal growth factor-regulated growth repressors, J. Biol. Chem. 273 (1998) 13848–13854. [32] J. Bian, C. Jacobs, Y. Wang, Y. Sun, Characterization of a putative p53 binding site in the promoter of the mouse tissue inhibitor of metalloproteinases-3 (TIMP-3) gene: TIMP-3 is not a p53 target gene, Carcinogenesis 17 (1996) 2559–2562. [33] W.D. Pennie, G.A. Hegamyer, M.R. Young, N.H. Colburn, Specific methylation events contribute to the transcriptional repression of the mouse tissue inhibitor of metalloproteinases-3 gene in neoplastic cells, Cell Growth. Differ. 10 (1999) 279–286. [34] S. Zochbauer-Muller, K.M. Fong, A.K. Virmani, J. Geradts, A.F. Gazdar, J.D. Minna, Aberrant promoter methylation of multiple genes in non-small cell lung cancers, Cancer Res. 61 (2001) 249–255. [35] E. Dulaimi, I. Ibanez de Caceres, R.G. Uzzo, T. Al-Saleem, R.E. Greenberg, T.J. Polascik, et al., Promoter hypermethylation profile of kidney cancer, Clin. Cancer Res. 10 (2004) 3972–3979. [36] Y. Sun, G. Hegamyer, H. Kim, K. Sithanandam, H. Li, R. Watts, et al., Molecular cloning of mouse tissue inhibitor of metalloproteinases-3 and its promoter. Specific lack of expression in neoplastic JB6 cells may reflect altered gene methylation, J. Biol. Chem. 270 (1995) 19312–19319. [37] W.S. El-Diery, Regulation of p53 downstream genes, Semin. Cancer Biol. 8 (1998) 345–357. [38] B. Anand-Apte, L. Bao, R. Smith, K. Iwata, B.R. Olsen, B. Zetter, et al., A review of tissue inhibitor of metalloproteinases-3 (TIMP-3) and experimental analysis of its effect on primary tumor growth, Biochem. Cell Biol. 74 (1996) 853–862. [39] L.M. Matrisian, Cancer biology: extracellular proteinases in malignancy, Curr. Biol. 9 (1999) 776–778. [40] L.M. Matrisian, G.W. Sledge, S. Mohla, Extracellular proteolysis and cancer: meeting summary and future directions, Cancer Res. 63 (2003) 6105–6109. [41] H. Sato, T. Takino, Y. Okada, J. Cao, A. Shinagawa, E. Yamamoto, et al., A matrix metalloproteinase expressed on the surface of invasive tumor cells, Nature 370 (1994) 61–65. [42] P.C. Brooks, S. Stromblad, L.C. Sanders, T.L. von Schalscha, R.T. Aimes, W.G. Stetler-Stevenson, et al., Localization of matrix metalloproteinase MMP-2 to the surface of invasive cells by interaction with integrin avb3, Cell 85 (1996) 683–693.
S. Thomas, D. Reisman / Cancer Letters 240 (2006) 48–59 [43] C.P. Blobel, Metalloproteinase-disintigrins: links to cell adhesion and cleavage of TNFa and notch, Cell 90 (1997) 589–592. [44] H. Ueno, H. Nakamura, M. Inoue, K. Imai, M. Noguchi, H. Sato, et al., Expression and tissue localization of membranetypes 1,2, and 3 matrix metalloproteinases in human invasive breast carcinomas, Cancer Res. 57 (1997) 2055–2060. [45] L.M. Coussens, B. Fingleton, L.M. Matrisian, Matrix metalloproteinase inhibitors and cancer: trials and tribulations, Science 295 (2002) 2387–2392.
59
[46] Q. Yu, I. Stamenkovic, Cell surface-localized matrix metalloproteinase-9 proteolytically activates TGF-b and promotes tumor invasion and angiogenesis, Genes Dev. 14 (2000) 163–176. [47] D.A. Freedman, A.J. Levine, Regulation of the p53 protein by the MDM2 oncoprotein, Cancer Res. 59 (1999) 1–7. [48] K.W. Kinzler, B. Vogelstein, Colorectal tumors in: K.W. Kinzler, B. Vogelstein (Eds.), The Genetic Basis of Human Cancers, McGraw-Hill, New York, 1998, pp. 565–587.