Stabilization and Activation of p53 by the Coactivator Protein TAFII31*

Regulation of the stability of p53 is key to its tumor-suppressing activities. mdm2 directly binds to the amino-terminal region of p53 and targets it for degradation through the ubiquitin-proteasome pathway. The coactivator protein TAFII31 binds to p53 at the amino-terminal region that is also required for interaction with mdm2. In this report, we demonstrate that expression of TAFII31 inhibits mdm2-mediated ubiquitination of p53 and increases p53 levels. TAFII31-mediated p53 stabilization results in activation of p53-mediated transcriptional activity and leads to p53-dependent growth arrest in fibroblasts. UV-induced stabilization of p53 coincides with an increase in p53-associated TAFII31 and a corresponding decrease in mdm2-p53 interaction. Non-p53 binding mutant of TAFII31 fails to stabilize p53. Our results suggest that direct interaction of TAFII31 and p53 not only mediates p53 transcriptional activation but also protects p53 from mdm2-mediated degradation, thereby resulting in activation of p53 functions.

Regulation of the stability of p53 is key to its tumorsuppressing activities. mdm2 directly binds to the amino-terminal region of p53 and targets it for degradation through the ubiquitin-proteasome pathway. The coactivator protein TAF II 31 binds to p53 at the amino-terminal region that is also required for interaction with mdm2. In this report, we demonstrate that expression of TAF II 31 inhibits mdm2-mediated ubiquitination of p53 and increases p53 levels. TAF II 31-mediated p53 stabilization results in activation of p53-mediated transcriptional activity and leads to p53-dependent growth arrest in fibroblasts. UV-induced stabilization of p53 coincides with an increase in p53-associated TAF II 31 and a corresponding decrease in mdm2-p53 interaction. Non-p53 binding mutant of TAF II 31 fails to stabilize p53. Our results suggest that direct interaction of TAF II 31 and p53 not only mediates p53 transcriptional activation but also protects p53 from mdm2-mediated degradation, thereby resulting in activation of p53 functions.
Mutation of the p53 tumor suppressor gene is frequently associated with different forms of human cancers. The ability of p53 protein to suppress tumor growth is attributed by two major biological processes induced by p53: cell cycle arrest and apoptosis (reviewed in Refs. [1][2][3]. Regulation of p53 protein stability plays a pivotal role in modulating p53 functions (4,5). Under normal conditions, the half-life of p53 is limited to minutes. Cellular stress or DNA damage leads to a rapid stabilization of p53 protein and activation of p53-mediated checkpoint functions. Several proteins are known to affect p53 stability through protein-protein interactions, including HPV16 E6 (6), WT-1 (7), E1B 55K/E4orf6 (8), SV40 large T antigen (9,10), JNK (11), HIF (12), p19ARF (13), p300 (14), and the cellular oncoprotein mdm2 (15,16). Whereas mdm2, JNK, and E6 have been implicated in targeting p53 for degradation, WT1, E1B 55K, and SV40T antigen stabilize p53 leading to increases in p53 levels. mdm2 functions as a pivotal p53 regulator and inhibits p53 functions by dual mechanisms. Interaction of mdm2 with p53 at its transactivation domain directly inhibits p53-mediated transcriptional activity (17,18). Binding of mdm2 to p53 also targets p53 for degradation through ubiquitin-mediated pathways (19,20). As one of p53 targets, expression of mdm2 is transcriptionally regulated by p53 (21), resulting in a feedback loop that regulates the level and activity of p53 (21). The biological implication of this regulatory loop is evident in that amplification of mdm2 gene has been found in about 30% of human soft tissue sarcomas and in other human cancers (22,23), suggesting that negative regulation of p53 by mdm2 is important in the development of those tumors. Loss of mdm2, on the other hand, results in an up-regulation of p53 activity and subsequent abrogation in cell cycle control and lethality in mice (24,25).
As a transcription factor, the p53 protein activates transcription of target genes by binding RNA polymerase II complex. Two of the TATA-binding polypeptide (TBP)-associated factors (TAFs) 1 within the general transcription factor TFIID complex, TAF II 31 and TAF II 70, have been shown to interact with the amino-terminal activation domain of p53 directly in vitro (26,27). It is generally believed that TBP mediates only basal transcription, whereas the TAF-TBP complex (TFIID) can mediate specific transcription factor-directed transactivation (28). Although the in vivo requirement for TAFs in p53-mediated transcription activation remains unclear, the fact that TAF II 31 and TAF II 70 support p53-mediated transcriptional activation in reconstituted transcription reactions suggests a critical role for TAFs in p53-mediated transactivation in vivo.
The observation that both TAF II 31 and mdm2 bind to the same amino-terminal region of p53 suggests that TAF II 31 may play an important role in regulating p53 protein stability and functions. To test this hypothesis, we have measured the effect of TAF II 31 on ubiquitination and stability of p53 both in vivo and in vitro using either purified protein components or whole cell extracts (11). Our results showed that TAF II 31 enhances p53 stability by inhibiting mdm2-mediated p53 ubiquitination and degradation. Expression of TAF II 31 also induces p53dependent transcriptional activation and cell cycle inhibition, suggesting that TAF II 31 plays an active role in regulating p53 stability and activity in cells. sion vector was constructed by cloning human TAF II 31 cDNA into the pMTN expression vector under the control of SV40 promoter. The HA tag was engineered at the amino terminus. Mutant TAF II 31 was generated by polymerase chain reaction and constructed into pMTN with HA tag. The mdm2 and p53 expression plasmids have been described elsewhere (11). The p53 reporter plasmids were described previously (29). Transfections were carried out using Fugene or N- [1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethyl ammonium methyl sulfate transfection reagent according to the manufacturer's protocol (Roche Molecular Biochemicals). In transient transfection experiments, cells were harvested 48 -72 h after transfection.
For immunoprecipitation, equal amount of lysates were immunoprecipitated using a monoclonal antibody against p53, pAb421, and protein A-agarose (Life Technologies, Inc.) for 16 -24 h. The complex was washed three times with lysis buffer and subjected to SDS-polyacrylamide gel electrophoresis analysis. The proteins were transferred to a Hybond-C membrane (Amersham Pharmacia Biotech), blotted by a rabbit polyclonal antibody against p53 (Santa Cruz), and developed using ECL (Amersham Pharmacia Biotech). Endogenous TAF II 31 was detected using a monoclonal antibody against TAF II 31. mdm2 was detected by monoclonal antibody 2A10.
For Northern analysis, equal quantities of total RNA isolated by RNAzol (Life Technologies, Inc.) at 10 -20 g/sample were subjected to electrophoresis on a denaturing 1% formaldehyde agarose gel. The RNA was transferred to a nylon membrane and hybridized with [ 32 P]dCTPlabeled probes using the Rapid Hybridization System (Amersham Pharmacia Biotech).
Pulse-Chase Labeling-10.1 cells growing in 10-cm plates were transfected with p53, mdm2, and TAF II 31. A protein labeling mixture containing [ 35 S]methionine (50 Ci; Amersham Pharmacia Biotech) was added to the plates 48 h after transfection and incubated for 10 min. The label was removed, and fresh media were added. The cells were harvested at 0, 1, and 2 h. The p53 protein was detected as described (11).
In Vitro and in Vivo Ubiquitination Assays-Baculovirus-expressed mdm2 and histidine-tagged p53 proteins were obtained as described (19). The TAF II 31 protein was produced by in vitro transcription and translation (Promega). Ubiquitination was performed using reticulocyte lysates depleted of mdm2 and JNK that provide the necessary components. The mdm2 and/or TAF II 31 (in 1-ng quantities) were incubated on ice with bacterially expressed p53 (1-5 g) bound to nickelnitrilotriacetic acid beads for 45 min. After extensive washes (four times with 1 ml of kinase buffer), the substrate-bound beads were equilibrated with 1ϫ ubiquitination buffer (50 mM Tris-HCl, pH8.0, 5 mM MgCl 2 , 0.5 mM dithiothreitol, 2 mM NaF, and 3 mM okadaic acid) and incubated in the same buffer supplemented with 2 mM ATP, 10 mM creatine phosphate, 0.02 unit creatine phosphokinase, 2 g of Ub HA, 1.5 mM ATP␥S (Sigma), and 33% RL (v/v) in a total volume of 30 l at 30°C for 5 min. The reaction was stopped by adding 0.5 ml of 8 M urea in sodium phosphate buffer (pH 6.3) with 0.1% of Nonidet P-40. The beads were washed three times with the stop buffer and once with PBS supplemented with 0.5% Triton X-100, and the protein moiety was eluted with Laemmli sample buffer at 100°C and subjected to Western blotting as described (11). For the in vivo ubiquitination assay, the cells were transfected with his-p53 (1 g), mdm2 (3 g), TAF II 31 (3 g), and HA-ubiquitin (1 g). The cells were treated with MG132 (10 M, Cal-Biochem) for 12 h at 48 h after transfection and lysed in kinase buffer. The p53-ubiquitin complex was precipitated by nickel-nitrilotriacetic acid beads and analyzed on SDS-polyacrylamide gel electrophoresis as described (19).
Reporter Assay-Saos-2 cells were transfected using the N-[1-(2,3dioleoyloxy) propyl]-N,N,N-trimethyl ammonium methyl sulphate liposomal transfection reagent (Roche Molecular Biochemicals) as follows. 100,000 cells were plated in each well of 6-well dishes and incubated overnight. Cells were fed with Dulbecco's modified Eagle's medium containing 10% FBS and incubated for an additional 3 h. N-[1-(2,3-dioleoyloxy) propyl]-N,N,N-trimethyl ammonium methyl sulphate/DNA mixtures containing appropriate amounts of reporter constructs and expression plasmids were prepared according to the manufacturer's instructions and incubated at room temperature for 15 min. Serum-free medium was then added to the mixtures and used to replace the medium in the wells. The dishes were incubated at 37°C for 3 h, after which the transfection mix was removed and replaced with Dulbecco's modified Eagle's medium containing 10% FBS. After 48 h, the 6-well plates were placed on ice and washed once with PBS. The cells were then lysed by scraping into 120 l of reporter buffer (Promega Luciferase Assay System), and samples were spun for 1 min at 14,000 rpm at 4°C. The total protein concentration was determined using a commercially available assay (Bio-Rad). Of each sample, 40 l were warmed to room temperature and mixed with luciferase assay substrate reconstituted with Luciferase Assay Buffer (Promega). Light emission was determined in a TD-20e luminometer (Turner).
Colony Forming Assay-10.1 and 12.1 cells were plated in 60 cm plates. The cells were transfected with 2 g of neo vector (pCMV-BM), p53, TAF II 31, and 5 g of salmon sperm DNA as carrier using the calcium phosphate precipitation procedure. Transfected cells were selected by Geneticin (50 g/ml, Life Technologies, Inc.) for 2-3 weeks. The colony was stained with Giemsa stain (Sigma).
Flow Cytometry Analysis-The flow cytometry were performed as described (30). Briefly, cells were transfected with 2 g of green fluorescence protein expression plasmid pEGFP-N (CLONTECH) and 10.0 g of either control DNA (empty vector), or p53, or TAF II 31 expressing plasmid. Transfected cells were harvested 48 -72 h later and washed twice in PBS. The cells were fixed in paraformaldehyde in PBS for 10 min and permeabilized in 70% ethanol for 16 h at 4°C. Cells were resuspended in PBS containing 5 g of propidium iodide and RNase A (5 g/ml). Flow cytometric analysis was carried out in a fluorescenceactivated cell sorting analyzer (Coulter). GFP positive cells were gated and analyzed for DNA content.

RESULTS
Stabilization of p53 in Vivo by TAF II 31-As a first step to address the possible role of TAFs in regulating p53 stability, mammalian expression vectors encoding human TAF II 31 and p53 were transfected into 10.1 mouse Balb/c 3T3 fibroblast cells that lack endogenous p53 (31). Co-transfection of mdm2 and p53 decreased expression levels of p53 as a result of its destabilization, as previously reported (15,16) (Fig. 1A). Expression of TAF II 31 in 10.1 cells increased the levels of p53 protein by 2-fold (Fig. 1A). Furthermore, forced expression of TAF II 31 attenuated mdm2-mediated on p53 degradation in a dosedependent manner (Fig. 1A). TAF II 31 did not change mRNA levels of p53 (Fig. 1A), suggesting that the elevation of p53 levels by TAF II 31 was not due to increased transcription of the p53 gene. In a duplicate experiment, expression levels of TAF II 31 protein in the transfection experiment were confirmed by Western analysis (Fig. 1A). Pulse-chase labeling with [ 35 S]methionine revealed that the half-life of p53 was extended in cells expressing TAF II 31 (Fig. 1B), supporting the notion that induction of p53 levels by TAF II 31 is due to enhanced p53 stability. To determine whether TAF II 31 mediates its effect on p53 through mdm2 or through an independent mechanism, p53 and TAF II 31 were cotransfected into a p53/mdm2 double null cell line. The results demonstrated that p53 levels are not affected by the expression of TAF II 31 in p53/mdm2 null cells, indicating that stabilization of p53 by TAF II 31 is mdm2 dependent (Fig. 1C). Because binding of p53 by mdm2 is required for mdm2-mediated p53 degradation (15,16), these results suggest that TAF II 31 either out-competes mdm2 for binding to p53 or excludes mdm2 from the p53 complex, thus preventing mdm2-mediated p53 degradation.
TAF II 31 Blocks mdm2-mediated Ubiquitination of p53-It has been shown that mdm2 targets p53 for ubiquitination, and we demonstrated that TAF II 31 inhibits mdm2-mediated degradation of p53. To extend our finding that the effect of TAF II 31 on p53 is at the level of ubiquitination, an in vitro ubiquitination assay was used to determine the degree of p53 ubiquitination after expression of TAF II 31 (19). In this assay nickel bead-bound histidine-tagged p53 (his-p53) was incubated with TAF II 31 and the targeting protein mdm2. Subsequently, the ubiquitination reaction was carried out using reticulocyte lysates depleted of mdm2 and JNK (Fig. 2A). The addition of mdm2 to p53 in the in vitro ubiquitination reaction resulted in ubiquitinated p53. Although TAF II 31 alone did not appear to affect p53 ubiquitination, it inhibited mdm2-mediated p53 ubiquitination in a dose-dependent manner (Fig. 2A). These findings suggest that TAF II 31 increases p53 stability by inhibiting mdm2-mediated p53 ubiquitination.
To determine the effect of TAF II 31 on p53 ubiquitination in vivo, mdm2, TAF II 31, and his-p53 expression plasmids were cotransfected into Balb/c 3T3 fibroblasts in various combinations together with the HA-tagged ubiquitin expression vector. After treatment with proteasome inhibitor MG132, the transfected p53 was purified on nickel beads and detected by immunoblotting with pAb421 antibody to p53. The degree of polyubiquitinated p53 was detected using antibody against HA. Consistent with the in vitro data, expression of mdm2 increased p53 ubiquitination and coexpression of TAF II 31 blocked mdm2-mediated ubiquitination of p53 in vivo (Fig. 2B). These data indicate that TAF II 31 attenuates the effect of mdm2 on p53 ubiquitination. TAF II 31 and p53 Interaction Is Required for p53 Stabilization-To test whether TAF II 31-p53 interaction is required for the ability of TAF II 31 to stabilize p53, we generated mutant TAF II 31 truncated at either the amino or carboxyl terminus. The TAF-C expresses amino acids 130 -264 of TAF II 31 and TAF-N produces amino acids 1-141 of the protein (Fig. 3A). Co-transfecting mutant TAF II 31 with p53 in mammalian cells showed that TAF-C coimmunoprecipitated with p53, and no interaction was detectable between TAF-N and p53 in the same assay (Fig. 3B). To explore the effect of TAF II 31 mutants on p53 levels and stability, we first examined the ability of these mutants to inhibit mdm2-mediated p53 degradation. Expression of mdm2 resulted in p53 degradation (Figs. 1A and 3C). Co-transfection of TAF-C significantly attenuated the ability of mdm2 to degrade p53 (Fig. 3C), although it is not as effective as the full-length protein (Fig. 3C). On the other hand, coexpression of TAF-N has no substantial effect on p53 levels (Fig. 3C), suggesting that TAF II 31-p53 interaction is required for the inhibitory function TAF II 31 inhibitory function on mdm2-mediated p53 degradation. To further confirm this observation, we investigated the effect of mutant TAF II 31 on mdm2-induced p53 ubiquitination (Fig. 3D). Consistent with p53 stability data (Fig. 3C), expression of TAF-C, but not TAF-N, resulted in a dose-dependent inhibition of mdm2-mediated p53 ubiquitination. Similarly, the anti-ubiquitination activity of TAF-C appears to be weaker than its full-length counterpart (Fig. 3D), suggesting that other mechanisms are also involved in stabilizing p53 by TAF II 31 besides direct competition with mdm2 binding. The transfected cells were harvested 48 h after transfection. Half of the material was processed to determine the level of p53 protein by immunoprecipitation with pAb421 and immunobloting using polyclonal antisera to p53. The other half was used for RNA preparation to assess p53 mRNA levels via Northern blots. In a separate transfection, cells were analyzed for TAF II 31 expression by Western blots using monoclonal antibody against HA epitope. B, increase of p53 half-life by TAF II 31. 10.1 cells were transfected with p53, TAF II 31, and mdm2. The cells were pulse-labeled with [ 35 S]methionine for 10 min and chased at the time indicated. P53 was immunoprecipitated by pAb421. C, stabilization of p53 by TAF II 31 is mdm2-dependent. Immortalized fibroblast cells from p53/mdm2 double knockout mice were cotransfected with p53 expressing plasmid (1 g) and TAF II 31 (2-10 g). The levels of p53 were detected by immunoprecipitation-Western blotting.

FIG. 2. TAF II 31 inhibits mdm2-induced p53 ubiquitination. A, TAF II 31 blocks mdm2-mediated p53 ubiquitination in vitro.
Recombinant his p53 proteins, purified mdm2 (11), and the TAF II 31 protein produced using an in vitro transcription and translation kit (Promega) were added to a ubiquitination reaction containing reticulocyte lysates predepleted of mdm2 and supplemented with bacterially produced HAtagged ubiquitin, ATP and ATP␥S, as described (19,39). The degree of ubiquitination was measured using immunoblots probed with anti-HA antibody. The p53-ubiquitin is represented as smears of more slowly migrating proteins in the upper part of the membrane. B, inhibition of p53 ubiquitination by TAF II 31 in vivo. 10.1 cells were transfected his-p53 (1 g), HA-ubiquitin (1 g), mdm2 (3 g), and TAF II 31 (3 g) in various combinations. The p53 complex was captured on nickel beads, and p53-ubiquitin was detected by Western blots against HA. Levels of p53 were determined using antibody against p53 on a Western blot.
Activation of p53-mediated Transcriptional Activation by TAF II 31-Stabilization of p53 in cells is expected to result in activation of p53 downstream pathways, including activation of p53-mediated transcription activity, inhibition of cell cycle progression, and/or activation of p53-mediated apoptosis. To elucidate the biological implications of p53 stabilization mediated by TAF II 31, we first analyzed p53 transcriptional activity using a luciferase reporter containing p53 binding sites from the human p21/WAF-1 gene (29). Transfection of TAF II 31 did not alter the basal transcriptional activity of the reporter containing p53 binding site (Fig. 4A), whereas cotransfection of p53 and TAF II 31 super-activated the reporter activities in a dosedependent manner (Fig. 4A). The synergetic activation of the p53 reporter gene by TAF II 31 may be attributed to its dual effects on p53 stability and its function as a p53 coactivator. The effect of TAF II 31 mutants on p53-mediated transcription activation was also tested in similar assays. In contrast to the wild type protein, both mutants inhibited p53-activated transcription (Fig. 4B). Furthermore, these mutants inhibit wild type TAF II 31-mediated p53 transcription activation (Fig.  4B). Because TAF II 31 functions as a bridge to link p53 to the transcriptional machinery (26,27), our data suggest that expression of TAF mutants disrupts the formation of p53 transcription complex and thus blocks p53-mediated transcriptional activity.
Forced Expression of TAF II 31 Induces p53-dependent Growth Suppression-To further explore the consequences of activation of p53-mediated transcription by TAF II 31, we used colony formation assay to monitor whether expression of TAF II 31 could affect cell growth (Fig. 4C and Table I). Transfection of p53 into either 12.1 cells (p53 wild type) or 10.1 cells (p53-null) resulted in a 8-fold reduction in plating efficiency, as a result of p53-mediated growth arrest in both cell types. Transfection of TAF II 31 into 12.1 cells resulted in a 7-fold decrease in colony formation. This effect was not seen in 10.1 cells. These data suggest that the ability of TAF II 31 to reduce the number of colonies formed is mediated by p53. fluorescence-activated cell sorter analysis of cell transiently transfected with TAF II 31 further confirmed this finding. Expression of TAF II 31 causes accumulation of cells in G 1 and reduction of cells in S phase of the cell cycle in a p53-dependent manner (Fig. 4D). These results demonstrate that expression of TAF II 31 leads to activation of p53-mediated growth arrest.
Association of TAF II 31 with p53 in Cells after UV Treatment-Because forced expression of TAF II 31 leads to stabilization of p53 and subsequent growth suppression, the association of p53 and TAF II 31 in cells was then explored. Analysis of the p53-TAF II 31 complex was carried out in Swiss3T3 cells before and after DNA damage (UV irradiation). Immunoprecipitation using monoclonal antibodies against p53 and immunoblot with antibodies to TAF II 31 identified TAF II 31 bound to p53 after UV exposure. The amount of TAF II 31 protein associated with p53 was proportional to the levels of p53, whereas the levels of mdm2 associated with p53 were inversely correlated with those of TAF II 31 (Fig. 5). TAF II 31 was in a complex with p53 in untreated cells, because TAF II 31 was immunoprecipitated by anti-p53 antibody at time 0 where low levels of p53 is expressed. This band is specific, because it is not present in 10.1 cells which lack p53 (data not shown). The increasing TAF II 31 associated with p53 after UV was not due to an increase in TAF II 31 levels as a result of UV exposure, because the total TAF II 31 protein did not change before and after UV treatment. These results suggest that TAF II 31 contributes to the stabilization of p53 after DNA damage.

FIG. 3. TAF II 31-p53 interaction is required for stabilization of p53.
A, expression of TAF II 31 and mutants. Cells transfected with indicated plasmids or vector alone (control) were lysed. TAF II 31 proteins were detected using anti-HA antibody on a Western blot. The TAF-C expresses amino acids 130 -264 of TAF II 31 and TAF-N produces amino acids 1-141. B, association of TAF II 31with p53. Cells were transiently transfected with indicated plasmids or vector alone (control). The p53 protein was immunoprecipitated using monoclonal pAb421 antibodies from 1 mg of protein lysates. A polyclonal p53 antibody was used for p53 detection. Associated TAF II 31proteins were detected using a HA antibody. C, expression of carboxyl-terminal half of TAF II 31 inhibits mdm2-induced p53 degradation The DNA expressing p53 (1 g), mdm2 (3 g), TAF II 31 (3 g), TAF-C (3 g), TAF-N (3 g), and vector control (3 g) were transfected in various combinations into 10.1 cells. The levels of p53 were detected by immunoprecipitation-Western blots and mdm2, and TAF II 31 and mutants were detected by Western blots. D, effects of TAF II 31 mutants on mdm2-mediated p53 ubiquitination. 10.1 cells were transfected his p53 (1 g), HA-ubiquitin (1 g), mdm2 (3 g), TAF II 31 (3 g), TAF-N (3 g), and TAF-C (1 and 3 g) in various combinations. The ubiquitinated p53 was detected by Western blotting using an antibody against HA after precipitation with nickel beads. Aliquots of the cell lysates were examined for mdm2 and p53 expression by Western blots.

DISCUSSION
We have demonstrated in this study that expression of the coactivator protein TAF II 31 contributes to the regulation of p53 stability through mdm2-mediated p53 ubiquitination pathway. Stabilization of p53 by TAF II 31 leads to functional activation of p53-mediated transcriptional activation and induction of cell cycle inhibition, suggesting that TAF II 31 contributes positively to p53 functions in cells. Although most of the experiments in this study are performed in transient transfection assays which result in overexpression of proteins, the fact that there is a positive correlation between endogenous p53 stabilization and p53-TAF II 31 association after DNA damage (Fig. 5) suggests that TAF II 31 plays an important role in the regulation of p53 activity in vivo. This is consistent with the observation that expression of TAF II 31 induces growth arrest mediated by endogenous p53 (Fig. 4, C and D, and Table I).
The mdm2 protein regulates both the activity and levels of p53 through a negative feedback loop. The TAF II 31-p53 interaction, on the other hand, seems to form a positive loop to activate p53 and its downstream pathways. It appears that TAF II 31 can directly inhibit mdm2-p53 interaction and stabilize p53 as suggested in the in vitro ubiquitination assay ( Fig.  2A). It is further supported by the fact that p53-TAF II 31 interaction is required for TAF II 31-mediated p53 stabilization (Fig.  3). As a transcription coactivator, TAF II 31 may recruit p53 to the transcription complex (26,27), which leads to transcriptional activation of p53 as seen in Fig. 4A. The recruitment of p53 to the transcription complex by TAF II 31 may also protect p53 from the mdm2-mediated degradation. These properties of TAF II 31 reinforce its function as coactivator of p53. Although the mechanistic basis of regulation of p53 function by endogenous TAF II 31 requires additional investigation, the finding that the effect of TAF II 31 on p53 requires protein-protein interactions suggests that TAF II 31 may directly compete with mdm2 for p53 binding, and a delicate balance between mdm2 and TAF II 31 determines the levels and stability of p53.
TAF II 31 is part of the TFIID complex, containing TBP and many other TAFs. It has been shown that TAF-TAF interactions are highly conserved to allow interchange between homologues of TAFs from different species in transcription reactions in vitro (26,32). Sequence comparison of Drosophila dTAF II 40 and human TAF II 31 as well as dTAF II 60 and TAF II 70 showed that the amino-terminal region is highly conserved (data not shown), suggesting that this region may be used for TAF-TAF interactions. Consistent with this notion, it has been shown that the conserved amino-terminal domain of TAF II 70 interacts with TBP, TAF II 250, and TAF II 31 (33). The fact that the carboxyl half of TAF II 31 contains the p53 binding domain further supports this possibility (Fig. 3). Thus, TAF II 31 functions as a bridge between p53 and the general transcription machinery using its carboxyl terminus to interact with p53 while binding to TFIID using its amino-terminal domain.
Although TAF II 31 enhances p53-mediated transcription activation, the mutant TAF II 31 lacking either amino or carboxyl p53 transcriptional activity. All activities were normalized to reporter alone. C, TAF II 31 suppresses colony formation in a p53-dependent manner. 10.1 (no p53) and 12.1 (wild type p53) cells were transfected with neo vector, p53, TAF II 31 (2 g each). The Geneticin-resistant colonies were counted 2-3 weeks after transfection and stained with the Giemsa stain. D, effect of p53 and TAF II 31 expression on cell cycle progression. 10.1 and 12.1 cells were transiently transfected with control vector, p53, and TAF II 31 DNA, as indicated. A GFP expressing plasmid was cotransfected to mark the transfected cells. The cells were subjected to fluorescence-activated cell sorter analysis, and GFP positive cells were analyzed for DNA content. Cell cycle distribution for each fluorescence-activated cell sorter was presented. The experiments were repeated, and data represent one set of the experiments.
FIG. 4. TAF II 31 activates p53-mediated transcriptional activation and growth arrest pathway. Saos-2 cells were transfected with 2 g of the p53 reporter vector, 1 g of pCMV or pCMV-p53wt, and increasing amounts of TAF II 31 expressing vector (0.1, 0.5, and 2 g). A corresponding appropriate amount of empty vector was included to maintain a total of 2 g in all transfections. Cells were maintained at 37°C and assayed to determine luciferase units/g protein. The indicated values are the average of two independent experiments that were performed in duplicate. A, expression of TAF II 31 has no effect on basal transcription and enhances transcription via the p53 response element in the p21 promoter. Fold activation was calculated as the fold increase in luciferase units/g protein relative to reporter alone for the effect on basal transcription. The effect on p53-activated transcription was calculated as the fold increase relative to reporter and TAF II 31 plasmids in the presence and absence of p53 expression plasmid. B, effect of TAF II 31 mutant on p53 activated transcription and TAF II 31-mediated terminus cannot form proper transcription complex and inhibits p53 transcription activity (Fig. 4). The dominant negative effects of TAF II 31 mutants on p53 transactivation suggest that TAF II 31 plays an important role in mediating p53 transcriptional activity in vivo. Recent studies showed that various TAF complexes (with or without TBP) mediate distinct functions, such as transcription activation (34 -36) and histone acetylation (37,38). The findings presented in this report that TAF II 31 can stabilize p53 further expand the growing list of functions mediated by the coactivator proteins.
It appears that the p53 stability is regulated by multiple factors in cells. Besides mdm2 and TAF II 31, many other proteins have been shown to regulate p53 stability, such as WT-1 (7), JNK (11), HIF (12), p19ARF (13), and p300 (14). How TAF II 31 interacts with these proteins in stabilizing p53 is unknown. It is possible that the balance between these factors determines p53 levels. The emerging model suggests that the contributions of different factors to p53 stability may depend on cell context, cell cycle, and cell stressing signals. One example is the finding that JNK appears to regulate p53 ubiquitination in nonstressed cells at the G0/G1 phase of the cell cycle, whereas mdm2 controls the stability of p53 at S and G 2 /M (11).
FIG. 5. Association of TAF II 31 with p53 in Swiss 3T3. Swiss 3T3 cells were treated with UV at 50J/m 2 . The cells were harvested at the indicated time points. Lysates were immunoprecipitated by pAb421, and the membrane was blotted with anti-p53, anti-mdm2, or anti-TAF II 31 antibodies. Purified TAF II 31 served as a positive control (pos. cont.). Amount of TAF II 31 in untreated and UV-treated cells were analyzed using whole cell extracts (WCE), and 100 g quantities of lysates were subjected to Western blotting analysis using the anti-TAF II 31 antibody. IP, immunoprecipitation; IB, immunoblot.