Steroid Receptor Coactivator-1 and Its Family Members Differentially ...

Steroid Receptor Coactivator-1 and Its Family Members Differentially Regulate Transactivation by the Tumor Suppressor Protein p53

Soo-Kyung Lee, Han-Jong Kim, Jung Woo Kim, and Jae Woon Lee Center for Ligand and Transcription (J.W.L, H.-J.K) Hormone Research Center (J.W.L) Department of Biology (S.-K.L) Chonnam National University Kwangju 500–757, Korea Department of Biochemistry (J.W.K) Paichai University Daejeon 302–735, Korea

The tumor suppressor protein p53 exerts its cell cycle-regulatory effects through its ability to function as a sequence-specific DNA-binding transcription factor. Herein, we show that p53 physically interacts with specific subregions of steroid receptor coactivator-1 (SRC-1) and its family members, p/CIP (p300/CBP interacting protein), xSRC-3, and AIB1 (amplified in breast cancer), originally isolated as transcription coactivators of nuclear receptors, as demonstrated by the yeast and mammalian two-hybrid tests as well as glutathione S-transferase pull-down assays. Interestingly, cotransfection of HeLa cells with SRC-1- or p/CIP expression vector potentiated the p53-mediated transactivation, whereas AIB1 and xSRC-3 were repressive. All of these SRC-1 members, however, similarly stimulated transactivation mediated by nuclear receptors and AP-1, as previously described. These results suggest that SRC-1 and its family members may differentially modulate the p53 transactivation in vivo. (Molecular Endocrinology 13: 1924–1933, 1999)

INTRODUCTION Mutations within the p53 gene represent one of the most common genetic aberrations in tumorigenesis (reviewed in Ref. 1). The wild-type p53 negatively regulates cell growth and division, whereas the mutant forms are unable to suppress or control cell cycle progression. The tumor suppressor function of p53 is thought to result from its ability to act as a cell cycle checkpoint protein, thus halting the cell cycle in the G1 0888-8809/99/$3.00/0 Molecular Endocrinology Copyright © 1999 by The Endocrine Society

phase if and when DNA damage occurs to a normal cell. Considerable evidence has accumulated for regulation of transcription as one of the primary mechanisms of the p53 action, in which p53 binds to a specific motif on gene promoters and thus transactivates the genes required to suppress cellular transformation (2, 3). Two repeats of a 10-bp motif, PuGPuCATGPyCPy, in which the G and A at positions 2 and 5 are critical determinants in p53-DNA binding, has been reported (2). Mutant p53 proteins are reported to show a dominant-negative effect by forming oligomeric complexes with the wild-type p53 before DNA binding, which results in a change in conformation and subsequently a loss of affinity for DNA. However, recent reports also suggest that, at least in some cases, the mutant forms can even promote the growth of the parental tumor cell and therefore exhibit an oncogenic gain of function of their own (4–6). Transcription coactivators bridge transcription factors and the components of the basal transcriptional apparatus (reviewed in Ref. 7). Functionally conserved proteins, CREB binding protein (CBP) and p300, have been shown to be essential for the activation of transcription by a large number of regulated transcription factors, including CREB, nuclear factor-kB (NFkB), basic helix-loop-helix (bHLH) factors, signal transducers and activators of transcription (STATs), AP-1, ternary complex factor (TCF)/serum response factor (SRF), nuclear receptors, and p53 (8). In particular, the nuclear receptor superfamily is a group of ligand-dependent transcriptional regulatory proteins that function by binding to specific DNA sequences named hormone response elements in promoters of target genes (reviewed in Ref. 9). Transcriptional regulation by nuclear receptors depends primarily upon a ligand-dependent activation function, AF2, with its core located 1924

Coactivation of p53 by SRC-1

in the C terminus and predicted to undergo an allosteric change upon ligand binding. Consistent with this, CBP and p300 have been found to interact directly with nuclear receptors in a ligand- and AF2dependent manner (10–13). In addition, a series of factors that exhibit ligand- and AF2-dependent binding to nuclear receptors have been identified both biochemically and by expression cloning. Among these, a group of highly related proteins have been shown to form a complex with CBP and p300 and enhance transcriptional activation by several nuclear receptors; i.e. steroid receptor coactivator-1 (SRC-1) (12, 14), xSRC-3 (15), AIB1 (amplified in breast cancer) (16), TIF2 (transcription intermediary factor 2) (17), RAC3 (18), ACTR (19), TRAM-1 (20), and p/CIP (p300/ CBP-interacting protein) (21). In particular, AIB1 was cloned as a gene whose expression and copy number were significantly elevated in human breast and ovarian cancers (16). We and others have recently shown that SRC-1 and p/CIP also mediate transactivation by other transcription factors including AP-1 (22), SRF (23), NFkB (24), CREB, and STATs (21). Based on this rather broad spectrum of action, we proposed that SRC-1 should be renamed transcription integrator, like CBP and p300 (22). Interestingly, SRC-1 (25) and its homolog ACTR (19), along with CBP and p300 (26, 27), were recently shown to contain histone acetyltransferase activities themselves and associate with yet another histone acetyltransferase protein p/CAF (28). In contrast, it was shown that SMRT (29) and N-CoR (30), nuclear receptor corepressors, form complexes with Sin3 and histone deacetylase proteins (31, 32). From these results, it was suggested that cofactors also exploit chromatin remodeling for transcriptional regulation, through histone acetylation-deacetylation. In light of the fact that SRC-1 is capable of functional interaction with CBP and p300, which in turn coactivate p53 (33, 34), we tested whether SRC-1 is also involved with the p53-mediated transactivation. In this report, we showed that SRC-1 specifically bound to p53 and coactivated the p53-mediated transactivation, either alone

1925

or in synergy with p300. Surprisingly, we found that p/CIP acted as a strong coactivator of p53 while AIB1 and xSRC-3 were repressive. These results are in marked contrast to earlier reports (12, 14–21) in which these different SRC-1 family members similarly modulated transactivation by nuclear receptors and suggest that each of these different SRC-1 proteins may have a specific set of target transcription factors in vivo.

RESULTS Interactions of SRC-1 and p53 We have recently shown that the AP-1 components c-Jun and c-Fos (22), SRF (23), and the NFkB component p50 (24) functionally interact with specific subregions of SRC-1 (Fig. 1). In addition, we have also found that xSRC-3 interacts with p53 in the LexAbased yeast two-hybrid system (35) (results not shown). To localize the p53 interaction domain of SRC-1, we examined LexA proteins fused to a series of SRC-1 fragments we recently described (22–24) (Fig. 1). Consistent with an idea that p53 interacts with SRC-A, SRC-D, and SRC-E, coexpression of a B42 fusion to the full-length p53 further stimulated the LexA/SRC-A, LexA/SRC-D-, and LexA/SRC-E-mediated LacZ expression in the yeast two-hybrid tests (Fig. 2A). In contrast, the LacZ expressions mediated by LexA alone or LexA fusions to SRC-B or C were not further stimulated by coexpression of B42/p53. As previously described (8), a LexA fusion to the C-terminal subregion of CBP (i.e. LexA/CBP-E) efficiently interacted with B42/p53. As previously reported (12, 14), LexA/SRC-C also interacted with a B42 fusion to thyroid hormone receptor or retinod X receptor in a ligand-dependent manner (results not shown), indicating that the lack of interactions between B42/p53 and LexA/SRC-C is not due to altered expression of the SRC-C fragment in this system. Similar results were also obtained with a B42 fusion to p53DC (i.e. the p53

Fig. 1. Schematic Representation of the SRC-1 Constructs The full-length human SRC-1 (12, 14) and a series of five SRC-1 fragments are as depicted. The nuclear receptor-interacting-, CBP-p300-interacting-, bHLH/PAS-, serine-threonine-rich-, and glutamine-rich domains, along with the recently identified histone acetyltransferase domain (25), the NFkB component p50-binding domain (24), c-Jun/c-Fos-binding domain (22), and SRF-binding domain (23), are as indicated. The amino acid numbers for each construct are shown.

MOL ENDO · 1999 1926

Vol 13 No. 11

Fig. 2. Interactions of p53 with SRC-1 in Vivo A, The yeast two-hybrid tests were employed to map the p53-SRC-1-interation domains. The indicated B42- and LexA plasmids were transformed into a yeast strain containing an appropriate LacZ reporter gene (35). Open, hatched, and solid bars indicate coexpression of the constitutive transactivation domain B42 (35) alone, a B42 fusion to the full-length p53, and a B42 fusion to p53DC, respectively. The results are expressed as induction (n-fold) over the value obtained with B42/- and LexA/-, which was given an arbitrary value of 1. The data are representative of three similar experiments. B, The p53-SRC-1 interactions were probed in the mammalian two-hybrid tests (36). HeLa cells were transfected with LacZ expression vector and VP16/SRC1-expression vector along with a reporter gene Gal4-Luc reporter (36), as indicated. Gal4/N represent parental Gal4 expression vector and numbers indicate the amount of each expression vector used (in nanograms). The results are expressed as induction (n-fold) over the value obtained with VP16/SRC-1 alone, which was given an arbitrary value of 1. The data are representative of three similar experiments.

residues 1–290), except that the interaction with SRC-A was not observed (Fig. 2A). These interactions were further probed in the mammalian two-hybrid tests (36). In HeLa cells, a Gal4 fusion to the full-length p53 (Gal4/p53) was a powerful transactivator of a reporter construct controlled by upstream Gal4 binding, increasing the activation up to 200-fold depending on the amount of Gal4/p53 cotransfected (results not shown). To easily assess the fold-activation, however, we used only 50 ng of Gal4/p53, which increased the activation approximately 10-fold (Fig. 2A). Coexpression of a VP16 fusion protein to the full-length SRC-1 further augmented this Gal4/p53-stimulated transactivation in a dose-dependent manner, supporting the idea that SRC-1 associates with p53 in vivo. Similar results were also obtained with Gal4/p53DC, consistent with the yeast results. To further characterize these interactions in vitro, glutathione-S-transferase (GST) alone and GST fusion to p53 were expressed, purified, and tested for interaction with in vitro translated luciferase and various SRC-1 proteins. The radiolabeled SRC-1 interacted with GST/p53 and GST/p53DC, but not with GST alone (Fig. 3A). In contrast, the radiolabeled luciferase did not bind to any of the GST proteins, as expected. In agreement with the yeast two-hybrid results (Fig. 2A), only SRC-A, SRC-D, and SRC-E, among various SRC-1 fragments, specifically interacted with GST/ p53, but not with GST alone, whereas only SRC-D and SRC-E interacted with p53DC (Fig. 3A). CBP-E also interacted with GST/p53 and GST/p53DC, but not with GST alone, and GST/TR interacted with SRC-1 and

SRC-C only in the presence of T3. The full-length AIB1 (16), p/CIP (20), and xSRC-3 (15), as well as the Cterminal fragments from these proteins, were also found to specifically interact with GST/p53 (Fig. 3B and results not shown). Overall, these results suggest that SRC-1 and its distinct family members directly associate with p53, and the interaction interfaces include the p53 residues 1–290 and, at least, the C-terminal subregions of SRC-1 containing the previously shown CBP-binding and histone acetyltransferase domains (12, 14–21). Cotransfection of SRC-1 Stimulates p53Mediated Transactivation To assess the functional consequences of these interactions, SRC-1 was cotransfected into HeLa cells along with a reporter construct p53RE-Luc that consists of a minimal SV40 early promoter and 17 upstream consensus p53 sites. Increasing amounts of cotransfected p53 efficiently enhanced the reporter gene expressions up to 200-fold, as expected (results not shown). SRC-1 enhanced the p53-dependent transactivation in an SRC-1 dose-dependent manner, with cotransfection of 200 ng of SRC-1 increasing the fold activation approximately 9-fold relative to the level with 10 ng of p53 alone (Fig. 4A). Consistent with the reports that CBP and p300 are transcription coactivators of p53 (33, 34), cotransfected p300 also had stimulatory effects on the reporter gene expressions, with cotransfection of 50 ng of p300 increasing the fold activation approximately 3.5-fold relative to the level


1927

Fig. 3. Interactions of p53 with SRC-1 in Vitro A, Luciferase, SRC-1, a series of five SRC-1 fragments, and CBP-E were labeled with [35S]methionine by in vitro translation and incubated with glutathione beads containing GST alone, GST/p53, GST/p53DC, or GST/TR, as indicated. Beads were washed, and specifically bound material was eluted with reduced glutathione and resolved by SDS-PAGE. 2/1 Indicates the absence and the presence of 0.1 mM T3, respectively. Approximately 20% of the labeled proteins used in the binding reactions were loaded as inputs. B, The C-terminal domains of distinct SRC-1 family members were labeled with [35S]methionine by in vitro translation and incubated with glutathione beads containing GST alone or GST/p53, as indicated. AIB1DN, p/CIPDN, and xSRC-3DN contained the AIB1 residues 1102–1420, the p/CIP residues 1109–1402, and the xSRC-3 residues 1093–1391, respectively. Approximately 20% of the labeled proteins used in the binding reactions were loaded as inputs.

with 10 ng of p53 alone. Coexpression of both p300 and SRC-1 further increased the reporter gene expressions above the levels observed with SRC-1 or p300 alone, suggesting that SRC-1 and p300 cooperate to coactivate the p53-mediated transactivation (Fig. 4A). Similar results were also obtained with reporter constructs that contain promoters of MDM2 (37, 38) and p21waf (39), previously characterized target genes of p53. In particular, synergistic action of SRC-1 and p300 was apparent with the p21waf promoter (Fig. 4B). In contrast, cotransfection of SRC-1 did not affect the LacZ reporter expression of the transfection indicator construct pRSV-b-gal (results not shown). Distinct SRC-1 Family Members Differentially Regulate the p53 Transactivation Recently, three distinct CBP/p300-containing coactivator complexes were described that differentially mediate transactivation by retinoic acid receptor, CREB, and STATs (40). In this study, coactivator complex mediating transactivation by retinoic acid receptor was demonstrated to require p/CIP and SRC-1, whereas two distinct complexes containing p/CIP regulated CREB and STATs. These results led us to test whether distinct SRC-1 family members show such differences in their potential regulation of the p53-dependent transactivation. Surprisingly,

p/CIP, originally described as a relatively weaker coactivator of nuclear receptors (21), was a very potent coactivator of p53, with cotransfection of 200 ng of p/CIP increasing the fold activation approximately 150-fold (Fig. 5A). Interestingly, xSRC-3 (23) and AIB1 (16) were not able to stimulate the p53dependent transactivation, despite the fact that these proteins readily interacted with p53 (Fig. 3B). These differential effects may stem from different expression levels of these SRC-1 proteins. In contrast to this notion, however, all of these SRC-1 family members were capable of stimulating the AP1-dependent transactivation to similar extents, as we recently described (22) (Fig. 5B). These results are consistent with an idea that coactivator complexes containing AIB1 may function with AP-1 and nuclear receptors but not with p53, whereas complexes containing SRC-1 and p/CIP work for AP-1, nuclear receptors, and p53. Alternatively, p53 may preferentially interact with SRC-1 and p/CIP but not with AIB1 within the context of macromolecular coactivator complex in vivo, although p53 is capable of interacting with all of these proteins on a one-to-one basis in vitro (Fig. 3B). Consistent with this hypothesis, p53-dependent inhibition of the AP-1 transactivation (34) was relieved with cotransfected SRC-1 and p/CIP, but not with AIB1 (Fig. 5C).

MOL ENDO · 1999 1928

Vol 13 No. 11

Fig. 4. Effects of SRC-1 and p300 Cotransfection on the p53-Mediated Transactivations HeLa cells were transfected with p53-, LacZ-, SRC-1-, or p300 expression vector along with a reporter gene p53RE-Luc (A), p21waf-Luc (B), or MDM2-Luc (C), as indicated. The results are expressed as induction (n-fold) over the value obtained with a reporter construct alone, which was given an arbitrary value of 1. The data are representative of three similar experiments.

AIB1 Inhibits the p53-Dependent Transactivation AIB1 was originally isolated based on its amplification and overexpression in human breast and ovarian cancers (16). Since mutations within the p53 gene and perturbed transactivation potential of the p53 protein represent one of the most common genetic aberrations in tumorigenesis, we further examined the effects of overexpressed AIB1 on the p53 transactivation potential. As shown in Fig. 6A, increasing amount of cotransfected AIB1 inhibited the p/CIP-stimulated p53 transactivation in a dose-dependent manner. With the reporter construct p21waf-Luc or MDM2-Luc, p/CIP showed potent stimulation of the p53-dependent transactivation, whereas AIB1 showed dose-dependent inhibitory actions (Fig. 6, B and C). Similarly, AIB1 and xSRC-3 also inhibited the Gal4/p53-directed activation of Gal4-Luc reporter gene expressions in a dose-dependent manner, whereas SRC-1 and p/CIP showed stimulatory actions (Fig. 6D). Consistent with the interaction results (Figs. 2 and 3), SRC-1 also

coactivated transactivation by Gal4/p53DC. We have also examined whether these results reflect differential p53 expression levels directed by each SRC-1 family member. As shown in Fig. 7, cotransfection of each SRC-1 family member didn’t significantly change the level of p53 expression. Similar results were also obtained with Gal4/p53 expressions (results not shown). From these results, we concluded that overexpressed AIB1 may seriously perturb the p53 transactivation potential in vivo.

DISCUSSION SRC-1 (12, 14) and its family member p/CIP (21), originally isolated as transcription coactivators of nuclear receptors, were recently shown to mediate transactivation by AP-1 (22), SRF (23), NFkB (24), CREB, and STATs (21). In this report, we added the tumor suppressor p53, as a new member to the list of transcrip-


1929

Fig. 5. Differential Modulation of the p53 Transactivation by Distinct SRC-1 Family Members HeLa cells were transfected with LacZ-, p53-, c-Fos-, SRC-1-, p/CIP-, xSRC-3-, or AIB1 expression vector along with a reporter gene p53RE-Luc (A) or TRE-Luc (B and C), as indicated. The results are expressed as induction (n-fold) over the value obtained with a reporter construct alone, which was given an arbitrary value of 1. The data are representative of three similar experiments.

tion factors that are regulated by SRC-1 (as summarized in Fig. 1). These results are not surprising, considering the recent reports (12, 14–21) in which SRC-1 and its family members were shown to functionally form a complex with CBP or its human homolog p300, which in turn binds and coactivates a wide spectrum of different transcription factors. These include nuclear receptors, CREB, NFkB, bHLH factors, STATs, TCF/SRF, AP-1, and p53 (reviewed in Ref. 8). Based on this broad spectrum of action, we have recently proposed to regroup SRC-1 into a class of proteins, such as CBP/p300, called integrator (22). Recently, p300 and CBP, despite their similarities, were shown to have distinct functions during retinoidinduced differentiation of embryonic carcinoma F9 cells (40). In addition, different classes of mammalian transcription factors were shown to functionally require distinct components of the CBP/p300 coactivator complex, based on their platform or assembly properties; retinoic acid receptor was demonstrated to require p/CIP and SRC-1, whereas CREB and STATs

required only p/CIP (41). In addition, these transcription factors were shown to require different histone acetyltransferase activities within the CBP/p300 complex to activate transcription (41). Overall, these results suggest that related but distinct CBP/p300-containing coactivator complexes exist in the cell, which exhibit a different specificity for each transcription factor. The results shown in Figs. 5 and 6 are consistent with this view, for which a few different coactivator complexes, each containing a distinct SRC-1 family member, can be proposed to exist. Coactivator complexes containing SRC-1 and/or p/CIP may mediate the nuclear receptor-, AP-1-, and p53-dependent transactivation, whereas a complex containing AIB1 may function with nuclear receptors and AP-1 but not with p53. Alternatively, p53 may preferentially interact with SRC-1 and p/CIP, but not AIB1, within the context of macromolecular coactivator complex in vivo, although these proteins are capable of interacting equally well with p53 in a one-to-one basis in vitro (Fig. 3B). Consistent with this hypothesis, p53-dependent

MOL ENDO · 1999 1930

Vol 13 No. 11

Fig. 6. AIB1, as an Inhibitor of the p53 Transactivation HeLa cells were transfected with LacZ-, p53-, c-Fos-, SRC-1-, p/CIP-, xSRC-3-, Gal4/p53-, Gal4/p53DC-, or AIB1 expression vector along with a reporter gene p53RE-Luc (A), p21waf-Luc (B), MDM2-Luc (C), or Gal4-Luc (D), as indicated. The results are expressed as induction (n-fold) over the value obtained with a reporter construct alone, which was given an arbitrary value of 1. The data are representative of three similar experiments.

inhibition of the AP-1 transactivation (34) was relieved with cotransfected SRC-1 and p/CIP, but not with AIB1 (Fig. 5C). However, it is interesting to note that p/CIP showed a much stronger coactivation with the AP1 transactivation when p53 was coexpressed (compare the results in panels B and C of Fig. 5). Thus, the p53-mediated repression of the AP-1 transactivation may also involve formation of a novel transcription inhibitory complex between AP-1 and p53, which is subjected to derepression and further coactivation by p/CIP, but not by AIB1. The SRC-1 family members were recently proposed to group into three subclasses based on their sequence homology; i.e. SRC-1/ NCoA-1 (17, 21, 42), SRC-2/TIF2/GRIP1/NCoA-2 (17, 42), and SRC-3/p/CIP/ACTR/AIB1/xSRC-3/Rac3 (15, 16, 18, 19, 21). In particular, ACTR (19), Rac3 (18), p/CIP (21), and AIB1 (16), which form an SRC-3 subfamily along with xSRC-3 (15), have been considered to derive from the same gene. However, ACTR and xSRC-3 are known to have multiple splicing isotypes (15, 19) that have not been extensively characterized, such as xSRC-3a shown in Fig. 8. Similarly, AIB1 and

Fig. 7. Western Analysis of p53 Expression HeLa cells were transfected with p53 alone or together with SRC-1-, p/CIP-, or AIB1 expression vector, and the resulting nuclear extracts were subjected to Western analyses for the p53 expression levels. The data are representative of two similar experiments. Similarly, the expression level of Gal4/p53 was not also affected by coexpression of each SRC-1 family member (results not shown).


1931

Fig. 8. Multiple Isotypes of SRC-3 Subfamily Members The nuclear receptor-interacting-, CBP-p300-interacting-, bHLH/PAS-, and glutamine-rich domains are as indicated. The relative positions of the glutamine-rich domains as well as specific insertions and deletions are as indicated. The nucleotide sequences of xSRC-3a (only a partial clone containing amino acids 252-1221 has been isolated) are identical to those of xSRC-3 (15) except in-frame insertion of unrelated 51 amino acids.

p/CIP may also represent different splicing isotypes from the same gene, although AIB1 is a human gene and p/CIP is a mouse gene. Therefore, we believe that this unexpected, differential regulation of the p53 transactivation by each SRC-3 family member may represent isotype-specific effects, although we can not exclude the possibility of species-specific differences. Overexpression of a component within an integrator complex could potentially perturb signal integration by the complex and affect multiple signal transduction pathways. Recently, AIB1 was identified as a gene amplified and overexpressed in human breast and ovarian cancers (16). Since AIB1 is capable of potentially regulating multiple transcription factors (21–24), such as CBP and p300 (8), it will be important to identify specific target transcription factors of overexpressed AIB1 that may lead to tumorigenesis. In particular, we have shown that the overexpressed AIB1 severely impaired p53 transactivation (Fig. 6). This may involve destabilizing the dynamic equilibrium established among distinct coactivator complexes in vivo, in which overexpressed AIB1 shifts the equilibrium to formation of the putative p53-inhibitory complexes containing AIB1. Alternatively, overexpressed AIB1 may lead to formation of a totally different macromolecular complex, which is also inhibitory to the p53 transactivation. Given the importance of p53 in tumorigenesis in general (1), p53 could be an excellent target factor of overexpressed AIB1 in tumorigenesis

processes in vivo, even though the involvement of other factors cannot be excluded. SRC-1 (25) and its homolog ACTR (19), along with CBP and p300 (26, 27), were recently shown to contain histone acetyltransferase activities themselves and associate with yet another histone acetyltransferase protein p/CAF (28). In contrast, it was shown that SMRT (29) and N-CoR (30), nuclear receptor corepressors, form complexes with Sin3 and histone deacetylase proteins (31, 32). Thus, it was suggested that chromatin remodeling by cofactors contributes to transcription factor-mediated transcriptional regulation, through histone acetylation-deacetylation. In addition, p300 was recently shown to be involved with an acetylation-mediated change in the function of a nonhistone-regulatory protein; i.e. p300 acetylated p53 both in vivo and in vitro, which stimulated its sequence-specific DNA-binding activity, possibly as a result of an acetylation-induced conformational change (43). SRC-1 synergized with p300 to stimulate the p53 transactivation (Fig. 4). This synergy is believed to reflect a cooperative recruitment of two different coactivator molecules (i.e. SRC-1 and CBPp300) by p53. These two distinct histone acetyltransferases could either modify selective sites on the histone tails or act in a concerted fashion to control different aspects of transcriptional activation. However, SRC-1 may not affect the acetylation-induced stimulation of the p53 DNA binding, like p300 (43), because SRC-1 efficiently coactivated transactivation

MOL ENDO · 1999 1932

by chimeric p53 proteins fused to Gal4, a heterologous, constitutive DNA- binding domain (i.e. Gal4/p53 and Gal4/p53DC, Fig. 6D). In particular, p53DC lacks the C-terminal acetylation sites by CBP-p300 (43). In summary, we have shown that SRC-1 and its family members interact with p53 and differentially modulate p53-mediated transactivation. In particular, our results suggest an interesting hypothesis for a novel tumorigenesis mechanism, in which the overexpressed AIB1 perturbs transactivation potential by p53. MATERIALS AND METHODS Plasmids LexA-, B42-, T7-, or GST vectors to express fragments of SRC-1 (A–E as depicted in Fig. 1) were as previously described (22–24). PCR-amplified fragments of CBP-E (i.e. the CBP residues 1867–2441), the full-length p53, and p53DC were subcloned into EcoRI–XhoI restriction sites of the B42 fusion vector pJG4–5 (35), the GST fusion vector pGEX4T (Pharmacia Biotech, Piscataway, NJ), the Gal4 fusion vector pCMXGal4/N (36), or the CMV/T7 vector pcDNA3 (Invitrogen, San Diego, CA). The expression vectors for p53, c-Fos, AIB1 (human gene in pcDNA3, kind gift of Paul Meltzer at NIH), p/CIP (mouse gene in pcMX, kind gift of Chris Glass at University of California at San Diego), xSRC-3, p300, SRC-1 (human SRC-1a in pCR3.1, kind gift from Ming Tsai at Baylor College of Medicine, Houston, TX) and VP16/SRC-1, the transfection indicator construct pRSV-b-gal, the AP-1-responsive reporter construct TRE-Luc, the Gal4-responsive reporter construct Gal4-Luc, and the T3-responsive reporter construct TREpal-Luc were as previously described (22–24). PCR fragment encoding AIB1DN (the AIB1 residues 1102–1420), p/CIPDN (the p/CIP residues 1109–1402), and xSRC-3DN (the xSRC-3 residues 1093–1391) were subcloned into EcoRI–XhoI sites of the CMV/T7 vector pcDNA3 (Invitrogen). Reporter constructs p53RE-Luc, MDM2-Luc, and p21waf-Luc were kind gifts of Moshe Oren (Weisman Institute of Science, Rehovot, Israel) and Leonard Freedman (Cornell University, New York, NY). Yeast Two-Hybrid Tests For the yeast two-hybrid tests, plasmids encoding LexA fusions and B42 fusions were cotransformed into Saccharomyces cerevisiae EGY48 strain containing the LacZ reporter plasmid, SH/18–34 (35). Plate and liquid assays of b-gal expression were carried out as described (35). Similar results were obtained in more than two similar experiments. GST Pull-Down Assays The GST fusions or GST alone was expressed in Escherichia coli, bound to glutathione-Sepahrose-4B beads (Pharmacia Biotech), and incubated with labeled proteins expressed by in vitro translation by using the TNT-coupled transcriptiontranslation system, with conditions as described by the manufacturer (Promega Corp., Madison, WI). Specifically bound proteins were eluted from beads with 40 mM reduced glutathione in 50 mM Tris (pH 8.0) and analyzed by SDS-PAGE and autoradiography as described previously (35). Cell Culture and Transfections HeLa cells were grown in 24-well plates with medium supplemented with 10% FCS for 24 h and transfected with 100

Vol 13 No. 11

ng of LacZ expression vector pRSV-b-gal and 100 ng of each reporter gene, along with increasing amounts of expression vectors for SRC-1, VP16/SRC-1, p53, c-Fos, Gal4/p53, Gal4/ p53DC, AIB1, xSRC-3, p/CIP, or p300 as indicated. After 12 h, cells were washed and re-fed with DMEM containing 10% FCS. Cells were harvested 36 h later, luciferase activity was assayed as described (35), and the results were normalized to the LacZ expression. Similar results were obtained in more than three similar experiments.

Acknowledgments We thank Dr. Moshe Oren for p53RE-Luc and MDM2-Luc plasmids and Dr. Leonard Freedman for p21waf-Luc plasmid. We also thank Chris Glass, Paul Meltzer, and Ming Tsai for plasmids. This work was exclusively supported by the National Creative Research Initiatives sponsored by the Korean Ministry of Science and Technology.

Received April 7, 1999. Revision received June 25, 1999. Accepted July 15, 1999. Address requests for reprints to: Jae Woon Lee, Ph.D., Center for Ligand and Transcription, Chonnam National University, Kwangju 500–757, Korea. E-mail: jlee@chonnam. chonnam.ac.kr.

REFERENCES 1. Levine AJ 1997 p53, The cellular gatekeeper for growth and division. Cell 88:323–331 2. El-Deiry WS, Kern SE, Pietenpol JA, Kinzler KW, Vogelstein B 1992 Definition of a consensus binding site for p53. Nat Genet 1:45–49 3. Funk WD, Pak DT, Karas RH, Wright WE, Shay JW 1992 A transcriptionally active DNA-binding site for human p53 protein complexes. Mol Cell Biol 12:2866–2871 4. Hsiao M, Low J, Dorn E, Ku D, Pattengale P, Yeargin J, Haas M 1994 Gain-of-function mutations of the p53 gene induce lymphohematopoietic metastatic potential and tissue invasiveness. Am J Pathol 145:702–714 5. Iwamoto KS, Mizuno T, Ito T, Tsuyama N, Kyoizumi S, Seyama T 1996 Gain-of-function p53 mutations enhance alteration of the T-cell receptor following X-irradiation, independently of the cell cycle and cell survival. Cancer Res 56:3862–3865 6. Gualberto A, Aldape K, Kozakiewicz K, Tlsty TD 1998 An oncogenic form of p53 confers a dominant, gain-of-function phenotype that disrupts spindle checkpoint control. Proc Natl Acad Sci USA 95:5166–5171 7. Horwitz KB, Jackson TA, Bain DL, Richer JK, Takimoto GS, Tung L 1996 Nuclear receptor coactivators and corepressors. Mol Endocrinol 10:1167–1177 8. Goldman PS, Tran VK, Goodman RH 1997 The multifunctional role of the co-activator CBP in transcriptional regulation. Recent Prog Horm Res 52:103–119 9. Mangelsdorf DJ, Thummel C, Beato M, Herrlich P, Schutz G, Umesono K, Blumberg B, Kastner P, Mark M, Chambon P, Evans RM 1995 The nuclear receptor superfamily: the second decade. Cell 83:835–839 10. Chakravarti D, LaMorte VJ, Nelson MC, Nakajima T, Juguilon H, Montminy M, Evans RM 1996 Role of CBP/ p300 in nuclear receptor signaling. Nature 383:99–103 11. Hanstein B, Eckner R, DiRenzo J, Halachmi S, Liu H, Searcy B, Brown M 1996 p300 is a component of an estrogen receptor coactivator complex. Proc Natl Acad Sci USA 93:11540–11545


12. Kamei Y, Xu L, Heizel T, Torchia J, Kurokawa R, Gloss B, Lin SC, Heyman RA, Rose DW, Glass CK, Rosenfeld MG 1996 A CBP integrator complex mediates transcriptional activation and AP-1 inhibition by nuclear receptors. Cell 85:403–414 13. Yao T–P, Ku G, Zhou N, Scully R, Livingston DM 1996 The nuclear hormone receptor coactivator SRC-1 is a specific target of p300. Proc Natl Acad Sci USA 93: 10626–10631 14. Onate SA, Tsai SY, Tsai MJ, O’Malley BM 1995 Sequence and characterization of a coactivator for the steroid hormone receptor superfamily. Science 270: 1354–1357 15. Kim H–J, Lee S–K, Na S-Y, Choi H–S, Lee JW 1998 Molecular cloning of xSRC-3, a novel transcription coactivator from Xenopus, that is related to AIB1, p/CIP, and TIF2. Mol Endocrinol 12:1038–1047 16. Anzick SL, Kononen J, Walker RL, Azorsa DO, Tanner MM, Guan XY, Sauter G, Kallioniemi OP, Trent JM, Meltzer PS 1997 AIB1, a steroid receptor coactivator amplified in breast and ovarian cancer. Science 277:965–968 17. Voegel JJ, Heine MJS, Zechel C, Chambon P, Gronemeyer H 1996 TIF2, a 160 kDa transcriptional mediator for the ligand-dependent activation function AF-2 of nuclear receptors. EMBO J 15:3667–3675 18. Li H, Gomes PJ, Chen JD 1997 RAC3, a steroid/nuclear receptor-associated coactivator that is related to SRC-1 and TIF2. Proc Natl Acad Sci USA 94:8479–8484 19. Chen H, Lin RJ, Schiltz RL, Chakravarti D, Nash A, Nagy L, Privalsky ML, Nakatani Y, Evans RM 1997 Nuclear receptor coactivator ACTR is a novel histone acetyltransferase and forms a multimeric activation complex with P/CAF and CBP/p300. Cell 90:569–580 20. Takeshita A, Gardona GR, Koibuchi N, Suen C–S, Chin WW 1997 TRAM-1, A novel 160-kDa thyroid hormone receptor activator molecule, exhibits distinct properties from steroid receptor coactivator-1. J Biol Chem 272: 27629–27634 21. Torchia J, Rose DW, Inostroza J, Kamei Y, Westin S, Glass CK, Rosenfeld MG 1997 The transcriptional coactivator p/CIP binds CBP and mediates nuclear receptor function. Nature 387:677–684 22. Lee S-K, Kim H-J, Na S-Y, Kim T-S, Choi H-S, Im S-Y, Lee JW 1998 Steroid receptor coactivator-1 coactivates activating protein-1-mediated transactivations through interaction with the c-Jun and c-Fos subunits. J Biol Chem 273:16651–16654 23. Kim H-J, Kim JH, Lee JW 1998 Steroid receptor coactivator-1 interacts with serum response factor and coactivates serum response element-mediated transactivations. J Biol Chem 273:28564–28567 24. Na S-Y, Lee S-K, Han S-J, Choi H-S, Im S-Y, Lee JW 1998 Steroid receptor coactivator-1 interacts with the p50 subunit and coactivates nuclear factor kB-mediated transactivations. J Biol Chem 273:10831–10834 25. Spencer TE, Jenster G, Burcin MM, Allis CD, Zhou J, Mizzen CA, McKenna NJ, Onate SA, Tsai SY, Tsai MJ, O’Malley BW 1997 Steroid receptor coactivator-1 is a histone acetyltransferase. Nature 389:194–198 26. Bannister AJ, Kouzarides T 1996 The CBP coactivator is a histone acetyltransferase. Nature 384:641–643 27. Ogryzko VV, Schiltz RL, Russanova V, Howard BH, Nakatani Y 1996 The transcription coactivators p300 and CBP are histone acetyltransferase. Cell 87:953–959

1933

28. Yang X–J, Ogryzko VV, Nishikawa J, Howard BH, Nakatani Y 1996 A p300/CBP-associated factor that competes with the adenoviral oncoprotein E1A. Nature 382: 319–324 29. Horlein AJ, Naar AM, Heinzel T, Torchia J, Gloss B, Kurokawa R, Ryan A, Kamei Y, Soderstrom M, Glass CK, Rosenfeld MG 1995 Ligand-independent repression by the thyroid hormone receptor mediated by a nuclear receptor corepressor. Nature 377:397–404 30. Chen JD, Evans RM 1995 A transcriptional corepressor that interacts with nuclear hormone receptors. Nature 377:454–457 31. Heinzel T, Lavinsky RM, Mullen TM, Soderstrom M, Laherty CD, Torchia J, Yang WM, Brard G, Ngo SD, Davie JR, Seto E, Eisenman RN, Rose DW, Glass CK, Rosenfeld MG 1997 A complex containing N-CoR, mSin3 and histone deacetylase mediates transcriptional repression. Nature 387:43–48 32. Nagy L, Kao H–Y, Chakravarti D, Lin RJ, Hassig CA, Ayer DE, Schreiber SL, Evans RM 1997 Nuclear receptor repression mediated by a complex containing SMRT, mSin3A, and histone deacetylase. Cell 89:373–380 33. Lill NL, Grossman SR, Ginsberg D, DeCaprio J, Livingston DM 1997 Binding and modulation of p53 by p300/ CBP coactivators. Nature 387:823–827 34. Avantaggiati ML, Ogryzko V, Gardner K, Giordano A, Levine AS, Kelly K 1997 Recruitment of p300/CBP in p53-dependent signal pathways. Cell 89:1175–84 35. Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K (eds) 1995 Current Protocols in Molecular Biology. Greene Associates, New York 36. Forman BM, Umesono K, Chen J, Evans RM 1995 Unique response pathways are established by allosteric interactions among nuclear hormone receptors. Cell 81: 541–550 37. Barak Y, Juven T, Haffner R, Oren M 1993 mdm2 expression is induced by wild type p53 activity. EMBO J 12:461–468 38. Wu X, Bayle JH, Olsen D, Levine AJ 1993 The p53mdm-2 autoregulatory feedback loop. Genes Dev 7:1126–1132 39. El-Deiry WS, Tokino T, Velculescu VE, Levy DB, Parson R, Trent JM, Lin D, Mercer WE, Kinzler KW, Vogelstein B 1993 WAF1, a potential mediator of p53 tumor suppression. Cell 75:817–825 40. Kawasaki H, Eckner R, Yao T-P, Taira K, Chiu R, Livingston DM, Yokoyama KK 1998 Distinct roles of the coactivators p300 and CBP in retinoic-acid-induced F9-cell differentiation. Nature 393:284–289 41. Korzus E, Torchia J, Rose DW, Xu L, Kurokawa R, McInerney EM, Mullen TM, Glass CK, Rosenfeld MG 1998 Transcription factor-specific requirements for coactivators and their acetyltransferase functions. Science 279: 703–707 42. Hong H, Koli K, Garbedian MJ, Stallcup MR 1997 GRIP1, a transcriptional coactivator for the AF-2 transactivation domain of steroid, thyroid, retinoid, and vitamin D receptors. Mol Cell Biol 17:2735–2744 43. Gu W, Roeder RG 1997 Activation of p53 sequencespecific DNA binding by acetylation of the p53 C-terminal domain. Cell 90:595–606