C/EBPβ Participates in Regulating Transcription of the p53 Gene in Response to Mitogen Stimulation*

The tightly regulated expression of p53 contributes to genomic stability, and transcription of the p53 gene is induced prior to cells entering S phase, possibly as a mechanism to ensure a rapid p53 response in the event of DNA damage. We have previously described the cloning of an additional 1000 bp of upstream p53 sequences that we have demonstrated play a role in the regulated expression of p53. As described in an earlier report, we preliminarily identified that a member of the CAAT/enhancer-binding protein (C/EPB) family of transcription factors may play a role in regulating p53. Here we have demonstrated that a particular C/EBPβ isoform, C/EBPβ-2, efficiently binds to the p53 promoter and induces its expression in a fashion that reflects the pattern of p53 expression seen as cells are induced to enter S phase and is absent from cells that are defective in proper p53 regulation. We conclude from these findings that C/EBPβ-2 plays a central role in the regulating of p53 transcription during the transition into S phase.

The p53 tumor suppressor gene plays a central role in the cell in maintaining genomic stability by functioning as a sequencespecific transcription factor that regulates the expression of genes required for cell cycle arrest or apoptosis in response to DNA damage (1,2). Missense mutations in the p53 gene, which inactivate its DNA damage checkpoint and growth-suppressing activities, have been observed in Ͼ60% of all human tumors (2,3). In many instances where the gene is not mutated, the complexing of p53 with overexpressed cellular proteins, such as MDM2 or Parc (4 -9), has been shown to inactivate the protein.
Loss of activity of wild-type p53 may also occur through mechanisms that prevent its entry into the nucleus (8,10,11) or through the loss of required p53 cofactors (12,13). Finally, loss of p53 expression in some tumors has been shown to occur through inhibition of transcription of the p53 gene itself (14 -18). These findings indicate that loss of wild-type p53 expression is an essential event in the genesis of cancer.
Given that the p53 protein is a critical regulator of cell growth, its expression must be tightly regulated for normal cell division as well as for its ability to function as a tumor suppressor. p53 levels must be suppressed for normal cell division to occur (19), which explains why p53 is found at low levels in normal dividing cells (2). When cells are stressed or damaged, p53 is rapidly induced. Several mechanisms exist to regulate p53 levels and function within the cell.
It is becoming more widely accepted that the transcriptional regulation of the p53 gene plays an important role in regulating the overall level of p53 protein. This was recently illustrated by Takaoka et al. (20), who demonstrate that interferon-␣/␤, an essential component leading to apoptosis in response to viral infections, activates p53 transcription, thus clearly implicating p53 in the interferon pathway. Likewise, it has been known for many years that the level of p53 is increased at the transcriptional level after growth factor stimulation of resting murine (21) and human (22) lymphocytes and resting fibroblasts (23)(24)(25)(26). Induced expression of p53 prior to S phase may serve as a mechanism for providing a rapid response to DNA damage during S phase. Until recently, the molecular basis for this transcriptional regulation has not been investigated. In view of these observations, we have focused on defining the mechanisms controlling p53 expression in response to mitogen induction. In a previous report, we characterized an additional 1000 bp of upstream DNA sequences and identified a number of new positive and negative regulatory elements. One of these newly identified elements, located ϳ960 bp upstream from the transcription start site, was found to bind to a trans-acting factor in a cell cycle-regulated manner and found to be required for proper S phase expression (23). Here we have demonstrated that the transcription factor C/EBP␤ (CAAT/enhancer-binding protein-␤) binds to this regulatory site on the p53 promoter in response to mitogen stimulation and serves to increase p53 promoter expression as cells enter S phase. Because C/EBP␤ is a transcription factor that is critical for normal proliferation and differentiation of a number of cell types (27)(28)(29)(30) and has also been demonstrated to play a role in mitogen-stimulated induction of cell division (31), we have carried out a experiments to establish whether it also has a role in regulating p53.
Transfections and Reporter Gene Assays-The 0.7 kbp contained the 672-bp murine promoter fragment cloned into the HindIII site of the pGL3-basic luciferase vector (Promega). The 1.7 kbp p53 promoter contained a 1672-bp KpnI-HindIII fragment isolated from the murine genome and cloned into the KpnI and HindIII sites of the pGL3-basic luciferase vector (33). 5 ϫ 10 4 cells in 24-well plates were transfected (TransFast Reagent; Promega) with the pGL3-basic luciferase reporter construct and 50 ng of Renilla driven by the HSV-TK promoter (pRL-TK vector) as an internal control. Eighteen hours after transfection, cells were grown in serum-depleted medium (0.1% FBS) for 24 h followed by serum stimulation (15% FBS). To assay wild-type full-length 1.7-kbp p53 promoter activity in the presence of C/EBP␤ isoforms (34), exponentially growing Swiss3T3 and 6629 (C/EBP␤-null) cells were transfected with increasing 1.7-kbp p53 promoter pGL3-basic luciferase vector with or without co-transfection of 0.25 g of C/EBP␤-2 and 50 ng of Renilla. Twenty-four hours after transfection, the cells were harvested, lysed, and assayed for luciferase activity. The results were normalized to TK-Renilla expression. To assay the activity of the 1.7 kbp p53 promoter harboring a site-directed deletion or mutation within the Ϫ972/Ϫ953 site on the promoter, exponentially growing Swiss3T3 cells were transfected with an increasing concentration of the mutant 1.7-kbp promoter pGL3-basic luciferase vector with or without co-transfection of 0.25 g of C/EBP␤-2 and 50 ng of Renilla. Cells were lysed at the indicated time points and assayed for luciferase activity using equal amounts of protein, as determined by Bradford assays. Reporter gene activity was normalized to TK-Renilla activity. Data are shown as means ϩ S.E.
In Vitro Transcribed/Translated C/EBP␤ Isoforms-C/EBP␤-1, -2, and -3 cDNA were cloned into pcDNA3.1 expression vector containing a T7 RNA polymerase promoter in vitro transcribed/translated using the TNT T7 quick coupled transcription/translation protocol (Promega). Plasmid DNA template (0.5 g) was transcribed/translated either in the presence of [ 35 S]methionine (Ͼ1000 Ci/mmol) or with cold methionine. Non-programmed rabbit reticulocyte lysate was used as a negative control. Synthesized proteins were analyzed on 12% SDS-polyacrylamide gel and visualized after impregnating the gel with EN 3 HANCE and subjecting the dried gel to autoradiography for 2 h.
The binding reaction consisted of 10 fmol/reaction [␥-32 P]ATP end-labeled double-stranded oligonucleotide, 2 g of poly(dI-dC) in binding buffer TM.1 (50 mM Tris-HCl, pH 7.9, 0.1 M KCl, 12.5 mM MgCl 2 , 1.0 mM EDTA, 20% glycerol and 1 mM dithiothreitol), and incubation on ice for 15 min followed by room temperature incubation for 15 min. The products were separated on a 4% polyacrylamide gel at 4°C in 0.5ϫ TBE (0.045 M Tris borate, 1 mM EDTA). Gels were dried and subjected to autoradiography for 30 min-6 h. To test for specificity, EMSA was performed with unlabeled specific and nonspecific competitors ranging from 10 to 50-fold molar excess of the labeled probe.
Chromatin Immunoprecipitation (ChIP) Assay-ChIP assays were performed using reagents in the ChIP-IT kit (Active Motif) on formaldehyde-fixed chromatin isolated from SWISS 3T3 cells. After shearing of the chromatin by sonication, samples were precleared with protein-G-agarose and incubated with antibodies. Complexes were immunoprecipitated and the DNA released, purified, and amplified by 36 cycles of PCR. Oligonucleotides used to amplify the putative C/EBP site in the p53 promoter were forward (5Ј-AGCGCTGGAGAAT-TCCTAGAGG-3Ј) and reverse (5Ј-CGAGATACTTGGTAT-CGCAC-3Ј) and yielded a 468-bp product assayed by electrophoresis through a 3% agarose (MetaPhor, Cambrex, Inc.). Controls included anti-RNApolII (positive), mouse IgG (negative), and a non-reactive anti-LAP (negative) for immunoprecipitations and random oligonucleotides for PCR.

RESULTS
In Vitro Translated C/EBP␤ Binds to the p53 Promoter-Having previously shown (23) that a nuclear factor that binds to a site on the p53 promoter at Ϫ972/Ϫ953 (Fig. 1A) contributes to cell cycle-regulated transcription of p53, we carried out a data base search for transcription factors that may bind the p53 promoter at this site. To identify the factor(s) that may be binding to this element, we entered the 20-bp sequence into the Genomatix MatInspector transcription factor data base. As described previously (23), three potential candidates (C/EBP␤, RBP-J (CBF1), and Ikaros-2) were identified as potential candidates, but only oligonucleotides specific for C/EBP␤ were able to block binding of this factor to the promoter. In view of these findings, we further examined the role of C/EBP␤ in regulating p53 transcription.
C/EBP␤ is critical for the normal growth and differentiation of various cell types (27,28,35). Three protein isoforms of C/EBP␤ are formed by alternative translation of three in-frame initiation sites on C/EBP␤ mRNA (34,36,37) (Fig. 1B). C/EBP␤-1 is the full-length form of the protein (38 kDa) that contains an intact N-terminal transactivation domain and C-terminal DNA-binding domain. C/EBP␤-2 (35 kDa) differs from C/EBP␤-1 by only 21 amino acids at the N terminus; however, the N-terminal transactivation domain is still functional. Both C/EBP␤-1 and C/EBP␤-2 are transactivators, although only recently have studies addressed their functional differences. C/EBP␤-1 is the only isoform detected in normal human mammary tissue, although in breast cancer cells, C/EBP␤-1 is absent (34,38). C/EBP␤-3 (21 kDa) completely lacks the N-terminal transactivation domain and is thought to repress transcription by complexing with C/EBP␤-1 or -2 and inhibiting their ability to transactivate target genes (34,36,37).
To confirm whether the factor binding to the Ϫ972/Ϫ953 cis-acting regulatory element may be one of the C/EBP␤ isoforms, we in vitro transcribed/translated the three isoforms of C/EBP␤ from pcDNA3.1 C/EBP␤ cDNAs encoding C/EBP␤-1, -2, and -3. The in vitro synthesized proteins were used in EMSA to assay binding to the Ϫ972/Ϫ953 site ( Fig. 2A). Although equivalent amounts of the individual proteins, as measured by Western transfer, were introduced into the DNA-binding reactions, C/EBP␤-2 demonstrated greater binding activity, as compared with either C/EBP␤-1 or -3. Anti-C/EBP␤ antibodies are specific for the C-terminal DNA-binding domain, which is present and functional in all three C/EBP␤ isoforms. In the presence of anti-C/EBP␤ antibody, all three isoforms supershifted, and as above, the supershifted complex containing C/EBP␤-2 was much more prominent than either of the other C/EBP␤ isoforms ( Fig. 2A). The intensity of in vitro synthesized C/EBP␤-2 binding as compared with the other two isoforms indicates that C/EBP␤-2 interacts more effectively to the p53 promoter than the other two isoforms of C/EPB␤.
Endogenous Nuclear C/EBP␤ Binds to the p53 Promoter-Having determined that in vitro synthesized C/EBP␤ can bind the p53 promoter, we proceeded to test for binding by endogenous C/EBP␤ from exponentially growing Swiss3T3 cells. Anti-C/EBP␤ antibody, specific for the C-terminal DNA-binding domain, when included in the DNA-binding assays resulted in a supershift of the bound complex (Fig. 2B). A C/EBP␤neutralizing peptide, which blocks the ability of the C/EBP␤ antibody to bind, prevented the supershift and demonstrated the specificity of the anti-C/EBP␤ antibody. Also, a supershift was not seen in the presence of anti-p53 antibody when used as a negative control. Furthermore, chromatin immunoprecipitation with anti-C/EBP␤ antibody demonstrated the association of C/EBP␤ with the p53 promoter (Fig.  2C). These results indicate that endogenous C/EBP␤ in Swiss3T3 nuclear extracts binds to the Ϫ972/ Ϫ953 site on the p53 promoter. Because the anti-C/EBP␤ antibody recognizes all three isoforms of C/EBP␤, these experiments cannot distinguish which form is binding. As described below, our results led us to conclude that it is likely to be the C/EBP␤-2 isoform.
Reporter Gene Assays Indicate C/EBP␤-2 Enhances p53 Promoter Activity-To characterize the role of C/EBP␤ in regulating p53 promoter activity, several approaches were taken. In one approach, we assayed the effect of the different C/EBP␤ isoforms on the expression p53 promoter in transient transfection assays. Exponentially growing Swiss3T3 cells were co-transfected with the 1.7-kbp p53 promoter and either C/EBP␤-1, -2, or -3 (Fig. 3). When C/EBP␤-1 was co-transfected with the 1.7kbp p53 promoter, it acted as a weak positive regulator and led to an enhancement of p53 promoter activity by ϳ2.5-fold. The expression of C/EBP␤-3, a putative transcriptional repressor, resulted in a slightly reduced p53 promoter activity. Of all three isoforms, co-transfection of C/EBP␤-2 with the 1.7-kbp promoter had the most dramatic effect on promoter activity. When compared with cells only transfected with the 1.7-kbp p53 promoter, co-transfection of C/EBP␤-2 resulted in an elevation of p53 promoter activity by ϳ15-fold (Fig. 3). Interestingly, increasing the amount of reporter plasmid in the transfection resulted in a gradual elimination of activation by C/EBP␤-2, presumably because of the presence of a vast excess of introduced target sequences that could not be bound by a limiting amount of input C/EBP␤-2.
To determine whether the enhanced p53 promoter activity required the C/EBP recognition site at Ϫ972/Ϫ953, we tested the activity of a 1.7-kbp p53 promoter that harbored either a deletion or a mutation within this site. Prior to assaying the promoter activity, the Ϫ972/Ϫ953-deleted and -mutated sites were tested for their ability to bind C/EBP␤ from Swiss3T3 nuclear extracts by EMSA. Both the mutated and the deleted No supershift was detected in the presence of anti-p53 antibody or neutralizing peptide. Exposure of the autoradiogram was carried out for 3 h. C, ChIP assays demonstrating C/EBP␤ associates with the p53 promoter. Lane a, non-precipitated input DNA-amplified with p53-specific primers; lanes b and c, incubation with a non-reactive anti-LAP antibody and PCR with random and p53-specific primers, respectively; lanes d and e, incubation with anti-C/EBP␤ antibody and PCR with random and p53-specific primers, respectively. sites eliminated binding of C/EBP␤ (data not shown) (23), and as a result, C/EBP␤-2 expression had no significant effect on either the mutated or deleted p53 promoter (Fig. 4). These results are in stark contrast to the 15-fold increase seen with the co-transfection of the wild-type full-length 1.7-kbp p53 promoter and C/EBP␤-2 (Fig. 3), indicating that the Ϫ972/Ϫ953 regulatory element on the p53 promoter is critical for p53 promoter activation by the C/EBP␤-2 protein.
Reporter Gene Assays Demonstrate that p53 Promoter Activity Is Reduced in C/EBP␤-null Cells-As another measure of the role of C/EBP␤ in contributing to p53 promoter activity, we asked whether cells devoid of C/EBP␤ expression were deficient in p53 promoter activity. Mouse embryo fibroblast cells prepared from wild-type C/EBP␤ (ϩ/ϩ) and C/EBP␤-null (Ϫ/Ϫ) mice (32) were used in reporter gene assays to further characterize the role of C/EBP␤ in regulating p53 promoter activity. Transfection of the 1.7-kbp p53 promoter into 6362 (C/EBP␤ ϩ/ϩ ) cells resulted in a 5-7-fold higher promoter activity as compared with the 1.7-kbp promoter activity in 6629 (C/EBP␤-null) cells (Fig. 5A). Likewise, p53 expression was undetectable in C/EBP␤-null cells, even in response to serum treatment and cell cycle induction (Fig. 5C). These results further support C/EBP␤ being a positive regulator of p53 expression.
Given that 6629 (C/EBP␤-null) cells have reduced p53 promoter activity in comparison to their wild-type counterpart 6632 (C/EBP␤ ϩ/ϩ ), we assayed the p53 promoter activity in the presence of ectopic C/EBP␤-2 expression in the 6629 C/EBP␤null cells. The cells were co-transfected with increasing amounts of 1.7-kbp p53 promoter and constant amounts of either C/EBP␤-1, -2, or -3 (Fig. 5B). In the presence of C/EBP␤-1 and -3, promoter activity was not significantly affected in comparison to the 1.7-kbp promoter activity in the absence of C/EBP␤ isoforms (Fig. 5B). Co-transfection of C/EBP␤-2 resulted in a Ͼ4-fold induction in p53 promoter activity relative to cells only transfected with the 1.7-kbp pro-moter. Taken together, these results indicate that C/EBP expression contributes to enhanced p53 promoter activity by binding to the regulatory site at position Ϫ972/Ϫ953.

EMSA Demonstrates C/EBP␤ Binds the Promoter in a Cell
Cycle-dependent Manner-Because we previously demonstrated differential binding of trans-acting factors within the Ϫ972/Ϫ953 element on the p53 promoter in a cell cycle-regulated manner (Fig. 6A) (23) that contributes to the regulation of p53 gene expression during the cell cycle and found that C/EBP␤ binds to this critical regulatory element and enhanced p53 promoter activity in exponentially growing cells, we examined the binding pattern of C/EBP␤ in arrested cells stimulated to re-enter the cell cycle.
To assay for C/EBP␤ binding the p53 promoter during the cell cycle, we tested nuclear extracts from arrested and serumtreated Swiss3T3 cells for the presence of C/EBP␤ by EMSA both in the presence and absence of anti-C/EBP antibodies. Upon serum-depletion, there was a decrease in C/EBP␤ binding to the promoter as denoted by the decrease in intensity of the complex seen at 0 h. By 3 h, post-serum stimulation binding of C/EBP␤ increased substantially and coincided with the  increased endogenous p53 mRNA levels and an increase in p53 promoter activity at 3 h of post-serum stimulation (Fig. 6D) (23).
Having established that C/EBP␤ is capable of differentially binding the p53 promoter during the cell cycle, we tested for the presence of C/EBP␤ protein during the cell cycle to determine whether changes in binding activity are reflected by changes in the abundance of C/EBP␤ protein. Swiss3T3 cells were serum-depleted for 24 h followed by serum stimulation prior to harvesting. Western blot analysis of the serum-treated cells detected the presence of the C/EBP␤ protein with protein levels at their highest in exponentially growing cells of 0, 3, and 8 h of post-serum stimulation (Fig.  6C). At 18 h of post-serum stimulation, the levels of C/EBP␤ begin to decrease to some extent. Although increased binding of the protein to the p53 promoter is seen at 3 h of post-serum stimulation, there is no apparent change in the level of C/EBP␤ during this period. These results indicate that C/EBP␤ protein is present throughout the cell cycle and that post-translational modification of the protein or interactions with additional regulatory partners at 3 h of postserum stimulation may be responsible for its enhanced binding to the promoter.
Cells Defective in Regulated p53 Expression Lack C/EBP␤ DNA-binding Activity-Swiss 3T3 cells as well as other cell lines expressing wild-type p53, such as NIH 3T3 and Balb 3T3, have been demonstrated to undergo a regulated p53 response upon growth arrest and induced re-entry into the cell cycle (23,25). The overall pattern observed is that the level of p53 mRNA significantly reduces by 18 -24 h of post-serum depletion and then increases in response to mitogen treatment. First detecta-ble by 3-6 h post-mitogen treatment, the level of p53 RNA generally is maximal by 18 h, after which it tends to return to the level observed in exponentially growing cells. As shown above, the appearance of C/EBP␤ DNA-binding activity generally parallels this response. To determine whether cells that do not undergo the normal p53 cell cycle response may have a detectable alteration in C/EBP␤ expression or activity, we assayed a series of murine breast carcinoma cell lines for their response to cell cycle arrest and mitogen stimulation. All cells were grown exponentially, transferred to medium containing 0.1% FBS for 18 h, and then supplemented with 15% FBS. Samples were taken from exponentially growing cells, serum-depleted cells, and cells stimulated with serum for 18 h and assayed for the level of p53 and C/EBP mRNA and for C/EBP␤ DNA-binding activity. As shown in Fig. 7, upper panel, four of the cell lines tested, NuMuMg, FSK-3, TM40-A, and TM-3, all show an elevation in p53 expression in response to treatment with serum. Two cell lines, HC11 and 4T1, show no increase in p53. HC11 cells express very low levels of p53 mRNA that do not change during the course of the experiment, whereas 4T1 cells express p53 mRNA that decreases upon growth arrest but fails to increase by 18 h post-serum treatment.
We next assayed whether C/EBP␤ activity is present in these cells, and as shown in Fig. 7, lower panel, although the cells undergoing an induction in p53 expression upon serum treatment exhibited C/EBP␤ DNA-binding activity, the two lines, HC11 and 4T1, which did not express p53 in response to mitogen treatment, also did not harbor any detectable C/EBP␤ DNA-binding activity (Fig. 7). In view of these results, we conclude that C/EBP␤ is essential for the proper regulation of p53 transcription during the transition from growth arrest to entry into the cell cycle.

DISCUSSION
Although it has been known for quite some time that the p53 gene is induced upon mitogenic stimulation of murine fibroblasts (26) and human lymphocytes (22), the molecular mechanism responsible for this regulation has remained unexplored. Because elevated levels of p53 protein have been shown to lead to either growth arrest or apoptosis in response to DNA damage, it might seem anomalous that transcription of the p53 gene induced upon induction with mitogens, with a peak in transcription prior to the onset of DNA synthesis. This type of response has been suggested to be important for a rapid p53induced arrest in DNA synthesis in response to DNA damage at a time when cells are synthesizing DNA and thus would be most susceptible to DNA-damaging events. In fact, Mosner et al. (25) have demonstrated an exceptionally rapid accumulation of active p53 protein in response to DNA damage in synchronized cell populations in mid-S phase. Increased synthesis of p53 during this phase of the cell cycle would be predicted to provide p53 protein poised to act and serve as a type of "rapid response" system for preventing the replication of damaged DNA. Therefore, a description of the mechanisms responsible for regulating p53 transcription will have implications with respect to our understanding of the normal response to DNA damage as well as to how perturbations in this response could contribute to oncogenesis.
Endogenous levels of p53 mRNA and p53 promoter activity are significantly reduced in cells that are serum-depleted for 24 h and begin to increase at 3 h of post-serum stimulation. During different phases of the response, we have shown differential binding of a trans-acting factor to the Ϫ972/Ϫ953 region of the p53 promoter. Binding was reduced after cells were serum-depleted for 24 h and increased at 3 h of post-serum stimulation. This pattern of binding parallels the expression pattern seen previously with the promoter and the endogenous p53 mRNA levels (23).
Previous work has indicated that binding to the Ϫ972/ Ϫ953 positive cis-acting element is necessary for maximal p53 promoter activity in exponentially growing cells as well as for the appropriate response to mitogen treatment (23). A data base search to identify candidate transcription factors that may bind the p53 promoter within the Ϫ972/Ϫ953 region as well as some preliminary DNA-binding assays revealed one factor (C/EBP␤) as a likely candidate. C/EBP␤ is a CCAAT enhancer-binding protein and is critical for normal proliferation and differentiation of mammary epithelial cells (29,30,34,38) as well as numerous other cell types (27,28). It has also been shown to transactivate the c-fos serum response element upon activation of the Ras-dependent signaling pathway in response to mitogenic stimulation (31). C/EBP␤ is found to be constitutively expressed in the liver, intestine, lung, adipose tissue, spleen, and kidney (39 -43) and plays an important role in adipocyte (40,44), hepatocyte (45), and mammary gland differentiation (29), as well as in ovulation (32,46). C/EBP␤ knock-out mice (43) have alteration in glucose homeostasis (47), immunological defects (48), develop lymphoproliferative disorders (49,50) and defects in female reproduction and infertility (32). The participation of C/EBP␤ in so many diverse pathways suggests that multiple levels of regulation contribute to its various biological functions.
Because C/EBP␤ exists as three different isoforms, the binding of the different C/EBP␤ isoforms to the p53 promoter was assayed to determine the activity of each isoform on the p53 promoter. In vitro synthesized C/EBP␤-2 demonstrated greater binding to the p53 promoter when compared with the binding activity of C/EBP␤-1 and -3. Binding of C/EBP␤ to the p53 promoter was also demonstrated in nuclear extracts from exponentially growing Swiss3T3 cells.
C/EBP␤ not only binds to the promoter in exponentially growing cells but also contributes to the differential binding pattern seen in nuclear extracts from serum-treated Swiss3T3 cells within the Ϫ972/Ϫ953 p53 regulatory element. Binding of C/EBP␤ to the p53 promoter is reduced upon serum depletion and increases substantially at 3 h of post-serum stimulation. C/EBP␤ binding the promoter at 3 h of post-serum stimulation coincides with the increase observed in p53 mRNA levels (23), indicating that binding of C/EBP␤ to the promoter likely plays a role in regulating p53 promoter activity during the cell cycle. This conclusion is further supported by experiments showing that cells defective in the regulation of p53 expression, such as the murine breast cancer lines HC11 and 4T1, lack detectable C/EBP␤ activity (Fig. 7).
Binding of C/EBP␤ to the p53 promoter is transient. C/EBP␤-binding activity is greatest at 3 h of post-serum stimulation and decreases thereafter, even though p53 mRNA levels continue to increase up to 18 h of post-serum stimulation. C/EBP␤ may bind the promoter at 3 h of post-serum stimulation, and other, as yet unidentified, binding factors may bind and regulate p53 promoter activity at later stages of the cell cycle. Identification of additional contributing factors will shed FIGURE 6. EMSA of the region of the p53 promoter that spans ؊972/؊953 demonstrates C/EBP␤ DNA-binding activity is induced as cells are released from growth arrest. A, the nuclear extracts used were from Swiss3T3 cells that were either growing exponentially (Exp) or were serum-depleted for 24 h (0 h) and then serum-stimulated and harvested at 3, 8, 18, and 24 h of post-serum stimulation. Extracts were probed with the labeled Ϫ972/Ϫ953 regulatory site. **, nonspecific binding (C/EBP complex). B, nuclear extracts from serum-treated cells (explained above) were tested for the presence of C/EBP␤ by adding 4 l of 0.6 g/l of anti-C/EBP␤ in the binding reaction. C/EBP␤ supershifted complexes are indicated by an arrow. **, nonspecific binding. Exposure of autoradiograms was carried out for 3 and 2 h, respectively. C, levels of C/EBP protein was assayed by Western blot analysis on Swiss3T3 cells that were either exponentially growing (Exp) or serum-depleted for 24 h (0 h) and then serum-stimulated and harvested at 3, light on regulation of the p53 promoter during later stages of the cell cycle.
Of the three C/EBP␤ isoforms, C/EBP␤-2 not only exhibits the greatest binding activity on the p53 promoter, it likewise has the greatest effect on the promoter activity showing a 15-fold induction of the p53 promoter in co-transfection assays. These results further support that C/EBP␤-2 binds to and regulates promoter expression during the cell cycle. C/EBP␤-2 up-regulated the p53 promoter in C/EBP-null cells Ͼ4-fold, whereas C/EBP␤-1 and -3 had no effect on the promoter. These results demonstrate that C/EBP␤-2 has a role in regulating p53 promoter activity.
Taken together, these findings provide insight as to how p53 gene expression is regulated during the cell cycle; however, several avenues of research need to be pursued to further elucidate p53 regulation during the cell cycle. First, identification of additional regulatory factors binding the promoter will provide further insight as to how p53 promoter activity is regulated. ChIP assays will demonstrate in vivo binding of the identified regulatory factors to the p53 promoter. Also, once additional proteins binding the p53 promoter have been identified, protein-protein interactions with C/EBP␤ can be demonstrated via co-immunoprecipitations and/or yeast two-hybrid assays. Ultimately, experiments utilizing small interfering RNA to inhibit C/EBP␤ expression will be invaluable in elucidating the role C/EBP␤ and any other identified trans-acting factors play in the cell cycle regulation of p53 expression in both normal and transformed cells.
Additional work will also be required to fully understand the role of induced p53 expression as cells enter S phase. This response has been suggested to be important for a rapid p53-induced arrest in DNA synthesis in response to DNA damage at a time when cells are synthesizing DNA and thus would be most susceptible to DNA-damaging events, as demonstrated by an exceptionally rapid accumulation of active p53 protein in response to DNA damage in synchronized cells populations in mid-S phase (25). Increased synthesis of p53 during this phase of the cell cycle would be predicted to provide p53 protein that is available, and once modified, could rapidly prevent the replication of damaged DNA. Cells that do not express p53 appropriately because of aberrations in C/EBP␤ activity may result in genetic instability and ultimately cancer and could explain the inability of some murine breast cancer cells to express p53 upon entry into the cell cycle. Therefore, the knowledge gained from these and future results will have implications with respect to our understanding of the normal response to DNA damage as well as to how perturbations in this response contribute to oncogenesis.