Abstract
Zac1 is a novel seven–zinc finger protein which possesses the ability to bind specifically to GC-rich DNA elements. Zac1 not only promotes apoptosis and cell cycle arrest but also acts as a transcriptional cofactor for p53 and a number of nuclear receptors. Our previous study indicated that the enhancement of p53 activity by Zac1 is much more pronounced in HeLa cells compared with other cell lines tested. This phenomenon might be due to the coactivator effect of Zac1 on p53 and the ability of Zac1 to reverse E6 inhibition of p53. In the present study, we showed that Zac1 acted synergistically with either p53 or a histone deacetylase inhibitor, trichostatin A, to enhance p21WAF1/Cip1 promoter activity. We showed that Zac1 physically interacted with some nuclear receptor corepressors such as histone deacetylase 1 (HDAC1) and mSin3a, and the induction of p21WAF1/Cip1 gene and protein by Zac1 was suppressed by either overexpressing HDAC1 or its deacetylase-dead mutant. In addition, our data suggest that trichostatin A–induced p21WAF1/Cip1 protein expression might be mediated through a p53-independent and HDAC deacetylase–independent pathway. Taken together, our data suggest that Zac1 might be involved in regulating the p21WAF1/Cip1 gene and protein expression through its protein-protein interaction with p53 and HDAC1 in HeLa cells. (Mol Cancer Res 2008;6(7):1204–14)
- p21WAF1/Cip1
- Zac1
- p53
- TSA
- HDAC1
Introduction
Zac1 is a novel seven–zinc finger protein which regulates apoptosis and cell cycle arrest (1). Both Zac1 and p53 have been found to induce the expression of type I pituitary-adenylate-cyclase–activating polypeptide receptor (PACAP1-R) gene. Zac1 provides the first example, apart from p53, of a molecule which can concomitantly induce both cell cycle arrest and apoptosis, leading to the inhibition of the growth of tumor cells.
Human Zac1 is located on chromosome 6q24-q25, a region known to harbor a suppressor gene (2). Because loss of Zac1 expression has been reported in a number of tumor types including breast and ovary cancers, melanomas, astrocytomas, and renal cell carcinomas, it has been proposed that Zac1 acts as a tumor suppressor gene (2-6). In contrast, recent studies have unveiled another role for Zac1 in the etiology of transient neonatal diabetes mellitus. This is a rare form of childhood diabetes which usually resolves in the first 6 months of life but strongly predisposes individuals to adult-onset type 2 diabetes (7-10).
The biochemical property of Zac1, required for exerting an antiproliferative effect, involves its ability to bind to specific DNA sequences and to function as a transcription factor (11). Zac1 is reported to induce the expression of PACAP1-R, cytokeratin, peroxisome proliferator–activated receptor-γ genes through the binding of its seven–zinc finger motif to GC-rich DNA elements (1, 11, 12). Functionally, Zac1 not only exerts an antiproliferative effect but also acts as a transcriptional cofactor for p53 and a number of nuclear receptors (NR; refs. 13-15). The NR coregulating activity of Zac1 may result from its ability to bind directly to NRs and to interact with two classes of NR coactivators: the p160 and the CBP/p300 family of coactivators. Whether Zac1 acts as a NR coactivator or repressor depends on the type of NR, the cell line, or the reporter gene (15). Although Zac1 serves as a transcriptional coactivator for p53, it also acts as a transcriptional repressor for a viral oncoprotein, human papillomavirus E6.18, in HeLa cells (14). Thus, in order to determine the molecular mechanism by which Zac1 functions as a positive or negative regulator is important to accurately delineate the role that Zac1 plays within the gene networks that regulate cell proliferation, cell cycle arrest, and apoptosis.
p21WAF1/Cip1 (later referred to as p21) is a well-characterized cyclin-dependent kinase (cdk) inhibitor that belongs to the Cip/Kip family (16). It mainly inhibits the activity of cyclin/cdk2 complexes and negatively modulates cell cycle progression (17). Because p21 is a transcriptional target of p53, it plays a crucial role in mediating growth arrest when cells are exposed to DNA-damaging agents such as doxorubicin and γ-irradiation (16). Apart from p53, a variety of other factors including Sp1, p300/CBP, c-Jun, and E2F are known to activate p21 transcription (17-20). Additionally, various mechanisms exist to regulate the levels of p21 protein in a cell including transcriptional regulation, epigenetic silencing, mRNA stability, and ubiquitin-dependent and ubiquitin-independent degradation of the protein (17, 21, 22). Steady-state acetylation levels of core histones are the result of a balance between the opposing activities of histone acetyltransferases and histone deacetylases (HDAC; refs. 23, 24). Many different types of HDAC inhibitors (HDACi), ranging from complicated structures of bacterial or fungal origin such as trichostatin A (TSA; ref. 25) to simple compounds such as butyrate (24) have been discovered over the years. Recent reports have shown that the induction of p21 gene by HDACi is mediated through chromatin remodeling, following acetylation of histones H3 and H4 in the p21 promoter region and promoter-associated proteins, including HDAC1 (26-28).
In this study, we report that Zac1 modulates p21 promoter activity and protein expression through p53-dependent and p53-independent mechanisms. The induction of p21 protein by TSA treatment was mediated through a p53-independent and HDAC deacetylase–independent pathway in HeLa cells. In addition, Zac1 directly formed a complex with HDAC1 and its coactivator and transactivation activities were impaired by the coexpressing HDAC1, independent of its deacetylase activity. Thus, Zac1 seems to modulate p21 gene and protein expression through positive factors (such as p53) or negative factors (such as HDAC1).
Results
Zac1 Enhances p21 Promoter Activity and Protein Expression through p53-Dependent and p53-Independent Pathways
Our previous studies have shown that Zac1 is a transcriptional coactivator for p53 through its direct interaction with this molecule (14). They also suggest that Zac1 might be involved in the regulation of the expression of the p53 target gene, p21 (14). Therefore, we first examined the functional interaction between Zac1 and p53 in our p21 promoter reporter assay using a p21-LUC reporter gene containing at least two copies of a p53-responsive element fused to the luciferase gene (Fig. 1A ). The effect on the p21 promoter reporter by exogenous p53 was dose-dependent (Fig. 1B, closed circles). The effect between Zac1 and p53 is shown in Fig. 1B, where the effect on the p21 promoter reporter by Zac1 is shown to be increased (from 45-fold to 188-fold) by the addition of exogenous mouse p53 from 1 to 10 ng. Above 10 ng, the addition of exogenous p53 decreased reporter gene expression from 188-fold to 23-fold (Fig. 1B, open circles). The expression of the endogenous p21 protein was also affected by addition of Zac1 with exogenous p53, as shown when 5 to 50 ng of HA.p53 was transfected into HeLa cells (Fig. 1C, compare lanes 7-12). Either exogenously transfected p53 or Zac1 slightly induced endogenous p21 protein expression in HeLa cells as determined by Western blot analysis (Fig. 1C, compare lanes 1-6 and lane 7). The coactivator effect of Zac1 on p53-induced p21 promoter activity and expression of the exogenous Zac1 construct was inhibited when high concentrations of exogenous p53 were expressed in HeLa cells (>10 ng for promoter activity or 50 ng for protein expression; Fig. 1B and C compare lanes 7-12 labeled as HA.Zac1).
The functional interaction between Zac1 and p53 in the induction of p21 gene and protein expression. A. p21 promoter reporter was used for p21 gene expression in this study. B. HeLa cells were transiently transfected with 0.3 μg of p21-LUC reporter gene with the indicated amount of pSG5.HA.p53 in the absence (•) or presence (○) of pSG5.HA.Zac1 (0.15 μg). C. Cell extracts collected as indicated were subjected to Western blot analysis with anti-p21, anti-p53, anti-HA, and anti-HuR antibodies (as control antibody). D. HeLa/vector and HeLa/p53 shRNA cells were cultured in a 100 mm Petri dish and were analyzed by Western blot analysis to monitor the levels of endogenous p21, p53, and HuR proteins. E. HeLa/vector and HeLa/p53 shRNA cells were transiently transfected with 0.25 μg of p21-LUC reporter genes; where indicated, expression vectors for pSG5.HA and 0.15 μg of pSG5.HA.Zac1. Luciferase activities of the transfected cell extracts were determined. RLU, relative light units. Numbers above the symbols (B and E), activity relative to an index of 1 for these samples in which no factor was added in HeLa cells. These data (B and E) are the average of three experiments (mean ± SD; n = 3). Results (C and D) are representative of two independent experiments.
To rule out the functional role of p53, we constructed stable short hairpin human p53 (nucleotide 775) RNAs (p53 shRNA) to repress the endogenous p53 protein expression in HeLa cells (HeLa/p53 shRNA; ref. 29). The efficiency of shRNA repression on p53 protein was examined by Western blotting. Endogenous p53 protein expression was barely detectable in the HeLa/p53 shRNA cells compared with parental HeLa (HeLa/vector) cells (Fig. 1D, compared with HuR internal/loading control). We further examined the influence of p53 on p21 promoter reporter activity (Fig. 1E compare histograms 1 and 2, closed columns). An enhanced effect of Zac1 on p21 promoter activity was observed in both the HeLa and HeLa/p53 shRNA cells but at 20-fold and 7-fold, respectively (Fig. 1E, compare histograms 1 and 2, open columns), suggesting that Zac1 might regulate p21 transcriptional activity through a p53-independent pathway in HeLa cells.
The Induction of p21 Protein by TSA Is Mediated through p53-Dependent and p53-Independent Pathways in HeLa Cells
The inhibition of HDAC deacetylase activity by TSA results in an increase or a decrease in the expression of a limited number (2-5%) of genes (30, 31). Recent reports show the induction of p21 gene by HDACi through the Sp1 element in p21 promoter (26, 27). Therefore, we first treated HeLa cells with various concentrations of TSA and then monitored the expression of p21, p53, and cyclin D1 proteins in these cells. We identified 1 μmol/L of TSA as an appropriate concentration for inducing changes in the expression of these three proteins in HeLa cells (Fig. 2A, lane 7 ). Even the IC50 value of TSA is ∼0.1 to 0.3 μmol/L for all HDACs, we chose the 1 μmol/L TSA to observe the apparent induction of p21 proteins in HeLa cells. Next, we analyzed the expression level of endogenous p21 protein in response to 1 μmol/L of TSA at various time points in HeLa cells (Fig. 2B). The p21 protein became detectable ∼6.5 hours after TSA treatment, and this was accompanied by a slight induction of the p53 protein. We treated HeLa/vector and HeLa/p53 shRNA cells with 1 μmol/L of TSA to dissect out the functional role of p53 and observed a dramatic induction of p21 protein in response to TSA in both the HeLa and HeLa/p53 shRNA cells (Fig. 2C, compare lanes 2 and 4, labeled as p21).
A p53-dependent and p53-independent induction of p21 protein expression by TSA in HeLa cells. A and B. HeLa cells were incubated with the indicated amounts of TSA for 12 h (A) and 1 μmol/L of TSA for the indicated period of time (B) was analyzed by Western blotting to monitor the levels of endogenous p21, p53, cyclin D1, or HuR proteins. C and D. HeLa/vector and HeLa/p53 shRNA cells were cultured in the absence or presence (+) of 1 μmol/L TSA treatment for 16 h and were analyzed by Western blot analysis (C) and RT-PCR (D) to monitor the levels of endogenous p21, p53, and HuR proteins (C) and endogenous p21, p53 and GAPDH genes (D). A-D. Results are representative of two independent experiments. E. HeLa/vector (closed columns) and HeLa/p53 shRNA (open columns) cells were transiently transfected with p21-LUC reporter gene (0.3 μg) with 1 μmol/L of TSA for 16 h. Luciferase activities of the transfected cell extracts were determined. Numbers above the columns, activity relative to an index of 1 for these samples in which no factor was added. These data are the average of three experiments (columns, mean; bars, SD; n = 3).
We further examined the effect of TSA on the transcription regulation of p21 gene by reverse transcriptase-PCR (RT-PCR). In the RT-PCR analysis, we observed a similar TSA effect on p21 gene induction in both the HeLa and HeLa/p53 shRNA cells (Fig. 2D, compare lanes 2 and 4, labeled as p21), but with higher expression levels of p21 in parental HeLa cells than in p53-silenced HeLa cells. Hence, TSA-induced p21 gene expression might be mediated through both p53-dependent and p53-independent pathways. The lower steady state levels of the p53 gene remained detectable in HeLa/p53 shRNA cells, suggesting that stable human p53 shRNA effectively degraded p53 mRNA in the cells (Fig. 2D, compare lanes 1 and 3, labeled as p53). Unexpectedly, p53 mRNA expression was repressed by TSA treatment in both the HeLa and HeLa/p53 shRNA cells (Fig. 2D, compare lanes 1 and lanes 2 or lanes 3 and 4, labeled as p53). The enhanced effect of TSA on p21 promoter activity, which was ∼7-fold or 8-fold in the HeLa and HeLa/p53 shRNA cells (Fig. 2E, compare histograms 1 and 2), also supported the idea that TSA might regulate p21 transcription through a p53-independent pathway.
The Potential Effect of Synergy on p21 Reporter Activity, not p21 Protein, by Zac1 and TSA in HeLa Cells
In line with our findings in Figs. 1 and 2, we further examined the functional interaction of Zac1 and TSA on p21 gene and protein expression. First, the inductive effect of TSA on p21 promoter reporter activity was enhanced by increasing the incubation time for TSA treatment (Fig. 3A, closed circles, maximal 8-fold ). Next, transiently transfected Zac1 enhanced promoter activity by 22-fold and increased the TSA-induced maximal response by 62-fold (Fig. 3A, open circles at 0 and 13 hours of TSA treatment, respectively). Our findings suggested that Zac1 acted synergistically with TSA on p21 promoter reporter activity in HeLa cells because the effect of the two together was 2.2-fold greater than the sum of their individual effects. In addition, we monitored both effects on the endogenous p21 and p53 protein status (Fig. 3B). As shown, Zac1 alone had the ability to induce endogenous p21 protein expression (Fig. 3B, compare lanes 1 and 5). The endogenous p21 protein expression was dramatically increased 4 hours after TSA treatment whether or not Zac1 was transfected in HeLa cells (Fig. 3B, compare lanes 1-4 and lanes 5-8). The endogenous p53 protein was slightly increased in a parallel experiment (Fig. 3B). Moreover, we found that TSA also enhanced the exogenous Zac1 protein level in HeLa cells (Fig. 3B, see lanes 5-8, labeled as HA.Zac1) and that the amount of Zac1 could contribute to the effect on reporter activity. As shown in Fig. 3C, p21 reporter activity was enhanced by the increase in the amount of exogenously transfected Zac1 DNA (from 8-fold to 28-fold) and the TSA effect on Zac1-induced reporter activity (from 59-fold to 86-fold) in HeLa cells. However, the minimal TSA effect on Zac1-induced reporter activity (59-fold at 50 ng Zac1, open column) was stronger than the maximal transfected Zac1 dose effect (28-fold at 500 ng, closed column) in this analysis. The TSA-induced Zac1 protein level might not be the primary factor for the TSA effect on Zac1-induced reporter activity and other possible effects on Zac1 function by TSA are investigated in later sections.
The functional interaction between Zac1 and TSA in the induction of p21 gene and protein expression. A. HeLa cells were transiently transfected with the p21-LUC reporter gene (0.3 μg) and 0.15 μg of pSG5HA (•) or pSG5HA.Zac1 (○) with 1 μmol/L of TSA for the indicated periods of time. Luciferase activities of the transfected cell extracts were determined. Numbers above the circles, activity relative to an index of 1 for these samples in which no factor was added. B. Cell extracts collected as indicated were subjected to Western blot analysis with anti-p21, anti-p53, anti-HA, and anti-HuR antibodies. Results are representative of two independent experiments. C. HeLa cells were transiently transfected with 0.3 μg of p21-LUC reporter gene with the indicated amounts of pSG5.HA.Zac1 in the absence (•) or presence (○) of 1 μmol/L of TSA treatment for 16 h. These data (A and C) are the average of three experiments (points, mean; bars, SD; n = 3).
Physical Interaction between Zac1 and HDAC1
To elucidate the mechanism by which TSA regulates Zac1 functions through NR corepressors, we showed by glutathione S-transferase (GST) pull-down and coimmunoprecipitation analysis that Zac1 bound to HDAC1, mSin3a, and SMRTα, but not to HDAC4 (data not shown). We further examined whether the endogenous p53 protein was required for the physical interaction between endogenous HDAC1 and Zac1 proteins in HeLa cells. Our results showed that there were similar coimmunoprecipitation patterns for Zac1 and HDAC1 proteins in the HeLa and HeLa/p53 shRNA cells (Fig. 4A ). Hence, Zac1 might form a complex with HDAC1 or HDAC1/p53 in HeLa cells. The detailed GST pull-down analysis showed that the fragment of HDAC1 in codons 1 to 321 was the primary sequence for Zac1 interactions (Fig. 4B). In addition, various fragments of Zac1 (codons 1-704, 1-520, 1-380, 1-220, and 221-520) were used to identify the HDAC1 binding domain of Zac1, with all immunoprecipitates being collected using an anti-myc antibody (for HDAC1) and probed using an anti-HA antibody (for Zac1). As shown, the HDAC1-binding domains of Zac1 were mainly located in the regions containing amino acids 1 to 220 and 221 to 520 (Fig. 4C, bottom, lanes 5 and 6).
The physical interaction between Zac1 and HDAC1 in vitro and in vivo. A. HeLa/vector and HeLa/p53 shRNA cells were treated with 1 μmol/L of TSA at the indicated times. Cell extracts (5% aliquots) were subjected to immunoprecipitation with anti-Zac1 (G-18) or anti-HDAC1 (N-19) antibodies. They were then immunoblotted with anti-hZac1 (Bioman), anti-HDAC1 (H-51), or anti-p53 (DO-1) antibodies. B. Full-length HDAC1 or the indicated HDAC1 fragments were translated in vitro and incubated with bead-bound GST or GST-Zac11-704. Bound proteins were eluted, separated by SDS-PAGE, and visualized by autoradiography. For comparison, each panel shows 10% of the input protein used in the binding reaction (left). C. COS7 cells were transfected with HDAC1.myc (5 μg) in the presence of various indicated HA.Zac1 fragments (5 μg). Cell extracts were subjected to immunoprecipitation with anti-HA antibody and then immunoblotted with anti-myc antibody (bottom) or immunoblotted with the respective anti-myc and anti-HA antibodies for 5% of the cell extracts (input controls, top and middle) by immunoprecipitation. We observed a similar binding pattern in two independent experiments.
HDAC1 Suppresses the Inductive Effect on p21 Promoter Activity of Zac1, p53, or TSA but Fails to Down-Regulate the Induction of the p21 Protein by TSA in HeLa cells
Many studies show that the regulation of the p21 promoter by HDAC1 may be mediated through the competition of p53 with the Sp1 complex or by the recruitment of other transcription factors to the p21 promoter element (32, 33). Because of the physical interaction between Zac1 and HDAC1, and the functional interaction between TSA and Zac1 or TSA and HDAC1, we examined the potential effect on p21 promoters with these factors (Fig. 5A ). First, our results indicate that Zac1 enhanced p21 promoter activity in a p53-dependent manner (Fig. 5A, compare histograms 1 and 2, closed columns), and possibly acted synergistically with the exogenous p53 by 1.8-fold (Fig. 5A, compare histograms 1-4, closed columns). The p21 promoter with two intact p53-binding sites showed improved Zac1 enhancement compared with a single p53-responsive element, whereas Zac1 failed to enhance the activity for reporters lacking the p53-responsive element (data not shown).
HDAC1 suppresses the induction of p21 promoter activity by Zac1, p53, or the combination of both in TSA-treated and -untreated HeLa cells, although it fails to suppress TSA-induced p21 protein expression. A. HeLa cells were transiently transfected with 0.25 μg of p21-LUC reporter gene; where indicated, expression vectors for 0.005 μg of pSG5.HA, pSG5.HA.53, or 0.15 μg of pSG5.HA.Zac1 and/or HDAC1 (0.25 μg) in the absence (closed columns) or presence (open columns) of 1 μmol/L of TSA treatment for 16 h. Luciferase activities of the transfected cell extracts were determined. Numbers above the columns, activity relative to an index of 1 for these samples in which no factor was added. These data are the average of three experiments (columns, mean; bars, SD; n = 3). B and C. HeLa cells in 60 mm culture dishes were transiently cotransfected with 0.01 μg of pSG5.HA.53, 0.25 μg of pSG5.HA.Zac1, or 0.25 μg of HDAC1 in presence of vehicle (B) or 1 μmol/L of TSA (C) treatment for 16 h. Cell extracts were subjected to Western blot analysis with anti-p21, anti-p53, anti-HA, and anti-HuR antibodies. Results (B and C) are representative of two independent experiments.
Second, our results suggest that TSA might act synergistically on the p21 promoter reporter with either Zac1 (by 1.2-fold) or p53 (by 3.1-fold) alone, but not in combination (0.83-fold; Fig. 5A, compare histograms 1-4, open columns). However, HDAC1 failed to completely abolish TSA-induced promoter activity when exogenous Zac1 or p53 were added to cells (Fig. 5A, compare histograms 5-8, open columns) because TSA is a pan-HDACi. The induction of endogenous p21 protein expression by Zac1 or p53 was suppressed by the coexpression of exogenous HDAC1 (Fig. 5B). In spite of the repression effect on the p21 promoter by HDAC1, cotransfected HDAC1 failed to suppress the induction of endogenous p21 protein expression in the TSA-treated HeLa cells (Fig. 5C), regardless of whether Zac1, p53, or both were present in the cells.
Our previous chromatin immunoprecipitation data showed that Zac1 might be recruited to the p21 promoter through the Sp1 sites (13), similar to HDAC1 for this promoter (34). We wished to determine whether HDAC1 was involved in the regulation of the p21 gene via p53, Zac1, or both combinations. Using the HeLa/vector and HeLa/p53 shRNA nuclear extracts, we used a DNA affinity precipitation assay, a method which specifically detects the dynamic interactions of DNA/protein using specific antibodies. Our data show that HDAC1 was detectable in both the p53-responsive element site 2 within the p21 promoter and the conserved Zac1 G4C4 type responsive element complexes (Fig. 6 ). The levels of HDAC1 proteins in these two binding complexes were down-regulated when cells expressed higher levels of p53 protein (Fig. 6, compare with lanes 5 and 7 or lanes 9 and 11). Our data further indicate that the levels of p53 protein in the Zac1 responsive element complex were down-regulated in HeLa cells by TSA treatment (Fig. 6, compare with lanes 9 and 10), implying that modifying p53 protein with TSA is able to disrupt p53 interactions with Zac1 or other proteins. Our chromatin immunoprecipitation results showed that HDAC1 and Zac1 might be recruited through the p53 protein onto the p53 responsive element within the p21 promoter, whether or not TSA was added to HeLa cells (data not shown). However, our DNA affinity precipitation assay and chromatin immunoprecipitation data also indicate that the protein complex in the p21 promoter depends on the TSA treatment time and the responsive elements being analyzed (data not shown). This issue needs further investigation.
Zac1 complexes with HDAC1 and this is independent of the endogenous p53 in HeLa cells. HeLa/vector and HeLa/p53 shRNA cells were treated with 1 μmol/L of TSA for 3 h. Nuclear extracts (30% aliquots) were subjected to incubation with biotin-labeled p53 responsive element (p53 RE) or biotin-labeled Zac1 responsive element (Zac1 RE) and streptavidin-HPR agarose. They were then immunoblotted with anti-p53 (DO-1), anti-HDAC1 (H-51), and anti-Zac1 antibodies. Results are representative of two independent experiments.
HDAC1 Suppresses Zac1-Induced p21 Promoter Activity, not HDAC Deacetylase Activity, through Its Direct Interaction with Zac1
Although wild-type HDAC1 failed to suppress TSA-induced endogenous p21 protein expression in HeLa cells (Fig. 5C), HDAC1 suppressed at least 90% of the TSA-induced p21 promoter activity, when either Zac1 or p53 was transfected into HeLa cells (Fig. 5A, compare open columns). Therefore, we further examined if the deacetylase activity of HDAC1 was involved in the repressive effect by using a catalytically inactive HDAC1 mutant (H141A; ref. 35). Our studies confirmed that this mutated HDAC1 retained its ability to interact with Zac1 in the GST pull-down analysis as did wild-type HDAC1 (Fig. 7A ). Although the deacetylase activity of HDAC1 was required for its inhibitory effect on the basal p21 promoter activity, we observed a 3.3-fold increase in p21 promoter reporter activity when cells were transfected with H141A mutant HDAC1 (Fig. 7B, histogram 1, compare wild-type and H141A mutant HDAC1s). The mutant HDAC1 still suppressed Zac1-induced p21 promoter activities, but not p53-induced activity, without or with TSA treatment (Fig. 7B, compare histograms 2 and 6 and histograms 1 and 3). Either the wild-type or H141A mutant suppressed TSA-induced promoter activities by 40% to 50% (Fig. 7B, histogram 5), but failed to suppress the induction of the endogenous p21 protein expression in the TSA-treated HeLa cells (Fig. 7C, compare lanes 3, 6, and 9). The differential suppression on vehicle and TSA-induced p21 promoter activity in wild-type and H141A mutant suggested that the enzymatic activity of HDAC1 might be involved in the synergistic effect of TSA on Zac1 (by 2.6-fold) or p53 (by 2.6-fold; Fig. 7B).
HDAC1 can still suppress Zac1-induced p21 promoter activity in the absence of its deacetylase activity. A. Wild-type HDAC1 and H141A mutant were translated in vitro and incubated with bead-bound GST, GST-Zac11-704, GST-Zac11-520, or GST-Zac11-220. Bound proteins were eluted, separated by SDS-PAGE, and visualized by autoradiography. For comparison, each panel shows 10% of the input protein used in the binding reaction (left). B. HeLa cells were transiently transfected with 0.25 μg of p21-LUC reporter gene; where indicated, expression vectors for 0.4 μg of pcDNA3.flag or pcDNA3.HDAC1 (wild-type or H141A mutant).flag and pSG5.HA.p53 (0.005 μg), pSG5.HA.Zac1 (0.15 μg), or the combination of both in the absence (columns 1-4) or presence of 1 μmol/L of TSA for 16 h (columns 5-8). Luciferase activities of the transfected cell extracts were determined. Numbers above the columns, activity relative to an index of 1 for these samples in which no factor was added. These data are the average of three experiments (columns, mean; bars, SD; n = 3). C. HeLa cells grown in 60 mm culture dishes were transiently transfected with 1 μg of pcDNA3.flag or pcDNA3.HDAC1 (wild-type or H141A mutant).flag and treated with 1 μmol/L of TSA for 0, 6, and 16 h. Cell extracts were subjected to Western blot analysis with anti-p21, anti-flag, and anti-HuR antibodies. Results (A and C) are representative of two independent experiments.
Enhancement of Zac1 Transactivation Activity by TSA in HeLa Cells
Our previous studies showed that the transactivation activity of mouse Zac1 can be down-regulated by its own NH2-terminal (amino acids 1-102) and COOH-terminal (amino acids 521-704) regions in HeLa cells (15). Here, our findings suggested that Zac1 might be regulated by TSA via an unidentified mechanism (Fig. 2) and could directly complex with HDAC1 and some corepressors (Fig. 4). Hence, we employed TSA to examine whether a NR corepressor system or its deacetylase activity is involved in regulating Zac1 transactivation activity. Gal4DBD-fused full-length Zac1 (pM.Zac1) was tested for its ability to activate a Gal4-responsive reporter gene in transiently transfected HeLa cells treated with 1 μmol/L of TSA for 16 hours (Fig. 8A, compare open circles and closed circles ). The Zac1 effect on Gal4 reporter activity was enhanced by ∼15-fold (at 500 ng) to 25-fold (at 100 ng) in the presence of TSA, and hence, was dependent on the transfected Zac1 dose (Fig. 8B). Both the wild-type and the deacetylase-dead mutant HDAC1 repressed the Zac1 effect on Gal4 reporter activity, whether or not TSA was added to HeLa cells (Fig. 8C), suggesting the importance of protein-protein interaction between HDAC1 and Zac1 or other interacting proteins (unknown at this point) and working through an alternative TSA mechanism other than the typical HDAC inhibition. The TSA effect on Zac1 transactivation activity was also observed in other cell lines, including CV-1 and human embryonic kidney 293 cells (data not shown).
The effect of TSA on the transactivation activity of Zac1 in HeLa cells. A. HeLa cells were transiently transfected with the indicated amounts of Gal4DBD vector (pM) or pM.Zac1 together with the GK1 reporter gene (0.5 μg), which encodes luciferase and is controlled by GAL4-responsive elements, in the absence (•) or presence (○) of 1 μmol/L of TSA treatment for 16 h. B. The induction fold of Zac1 transactivation activity in response to TSA treatment. C. HeLa cells were transiently transfected with 0.1 or 0.4μg of pM or pM.Zac1 indicated together with the GK1 reporter gene (0.4 μg) and 0.3μg of pcDNA3.flag, wild-type, or mutant pcDNA3.HDAC1 (HDAC1 wt or HDAC1 mt) in the absence (histograms 1-3) or presence (histograms 4-6) of 1 μmol/L of TSA for 16 h. Luciferase activities of the transfected cell extracts were determined. Dotted line, relative activity to an activity of 1 for these samples in which no HDAC1 and no TSA were added to each of cells. These data are the average of three experiments [points, mean; bars, SD; n = 3 (A); columns, mean; bars, SD; n = 3 (C)].
Discussion
The results presented here show that TSA induces Zac1 transactivation activity and acts synergistically with Zac1 in the in the transcriptional activation of the p21 promoter. Our previous studies indicated that Zac1 not only interacts directly with NR but also forms a complex with some NR coactivators containing histone acetyltransferases, such as CBP and p300 (15). Hence, the association of acetyltransferase (CBP) and deacetylase (HDAC1) with Zac1 might support its dual roles in both positively and negatively regulating specific gene expressions. In physiologic conditions, the positive or negative activity of Zac1 could be regulated by the concentrations or activities of related cellular components (such as acetyltransferases and deacetylases), which are primarily responsible for the changes in the promoter and cell type–specific activities of Zac1. YY1 (yin yang 1) is a sequence-specific DNA-binding protein which either enhances or represses transcription in a cell type–dependent and promoter-specific manner (36). The functions of YY1 are regulated by acetylation and deacetylation, through physical interaction with p300/CBP and HDAC1, HDAC2, or HDAC3 (37, 38). Thus, we postulated that Zac1 might function in a similar fashion to YY1, acting as a docking protein for acetylating and deacetylating enzymes onto its specific mammalian promoters.
Earlier studies have suggested that DNA methylation or histone modifications may be involved in the regulation of mRNA expression of Zac1 (4, 5). The combined treatment of the demethylating agent 5-aza-2′-deoxycyticidine and TSA increased by 2-fold to 3-fold of Zac1 mRNA expression in HOC cell lines (39). In this study, the simple increase in the amount of Zac1 DNA transfected could not elicit the same effect as TSA on Zac1's coactivator and transactivation functions in HeLa cells. Therefore, the functional role of TSA on Zac1-induced p21 promoter activity might, at least in part, mediate the enhancement of Zac1's transactivation activity or down-regulate c-Myc and cause the release of repression of c-Myc from the p21 promoter (27). By decreased exogenous Zac1 protein expression levels or other unknown mechanisms, the synergistic effect could be abolished by delivering high exogenous p53 into HeLa cells. Indeed, this high-dose p53 inhibition phenomenon is also observed in other cell lines (14). Although recent work by Rozenfeld-Granot and colleagues showed that a putative consensus p53 sequence, in which there is binding and activating by p53, is located upstream of the first coding exon (exon 8) of Zac1 (40), the contradictory findings for Zac1 expression by p53 remains to be investigated in the future.
HDAC1 and its enzymatic dead mutant not only suppressed Zac1's p53-dependent coactivator functions, but also suppressed TSA-induced enhancement of Zac1's transactivation. Recent studies show that HDAC1 might be recruited to the proximal region of the p21 promoter by Sp1 or, alternatively, to the distal region of the p21 promoter by other transcriptional regulators such as NRs and c-Jun (19, 41-44). Induction of p53 in response to DNA-damaging agents could result in the formation of p53 and Sp1 complexes and, simultaneously, dissociate HDAC1 from the COOH terminus of Sp1 (20, 28, 33, 45). Therefore, our findings on the physical interaction between Zac1 and HDAC1 suggest that Zac1 might act on the p21 promoter not only through the p53-dependent pathway, as a p53 coactivator or by competition of HDAC1 from HDAC1-p53 complex (35), but also through the Sp1-dependent pathway (13), as an Sp1 coactivator or in competition with HDAC1 from the HDAC1-Sp1 complex (data not shown).
Recent studies show various mechanisms to regulate the levels of p21 protein in a cell. These include transcriptional regulation, epigenetic silencing, mRNA stability, and ubiquitin-dependent and ubiquitin-independent degradation of the protein (26-28). The p21 gene and protein seem to be key targets for induction by TSA, even though the identity of the deacetylase(s) involved in regulating the p21 gene and protein expression have to be determined. Accordingly, our results indicate that the induction of p21 protein expression by TSA in HeLa cells might be mediated through p53-independent and HDAC1-independent pathways, and TSA-induced p21 protein expression could override the synergistic effect caused by Zac1 and p53. In this case, the involvement of the p21 gene and protein might cause changes in the acetylation status of histone and nonhistone proteins in TSA-treated HeLa cells. Current epigenetic analysis also supports this hypothesis, and it has been shown that TSA increases the protein stability of p53 and the Zac1 rat orthologue, Lot1, respectively (35, 46). In addition to the regulation of many HDACs by subcellular localization (47), we also observed that Zac1 colocalized with HDAC1 in the nucleus and their subcellular localization was redistributed by TSA in specific cell lines (data not shown). In this study, posttranslational modification was involved in the TSA-induced p21 stabilization and the potential mechanism needs to be clarified.
In this study, our work (a working model in Fig. 9 ) shows the regulatory effect of Zac1 on p21 gene and protein expression through p53-dependent and p53-independent pathways and protein-protein interactions with, at least, p53 and HDAC1 at the transcriptional level. In this article, the 1 μmol/L concentration is higher than the IC50 value of TSA (0.1-0.3 μmol/L for all HDACs). In contrast with the p21 protein induction, the TSA effects, even in the treatment with 1 μmol/L high concentration, on p53 protein induction and DNA damage were obscure in comparison with the actinomycin D treatment in the HeLa cells (data not shown). The various TSA concentrations might be involved in the HDAC enzyme activity and/or DNA damage and response. Thus, it will be of great interest in the future to examine which mechanism is responsible for the induction of the p21 gene and protein by various TSA concentrations in HeLa cells.
Model for the p53-dependent and p53-independent pathways for p21 gene and protein expression by Zac1. The enhancement of p21 gene expression by Zac1 may be mediated through a p53-dependent (I) and p53-independent (II) pathway in HeLa cells. The effect of TSA on Zac1 functions enhances not only its transactivation activity, but also its p53-dependent coactivator functions. Wild-type and deacetylase-dead mutant (H141A) HDAC1s both suppress the enhancement effect by Zac1 or Zac1/p53 complex on p21 gene expression, whereas only wild-type HDAC1 is able to suppress the transcription function of p53 on p21 gene. The TSA-induced p21 protein expression cannot be suppressed by the overexpression of exogenous wild-type HDAC1, suggesting that the TSA effect might be involved in the stability of p21 protein via the unidentified posttranslational modification mechanism(s). ⊕, activation; ⊝, inhibition; ○, no effect.
Materials and Methods
Plasmids
Various Zac1 coding regions were generated by PCR and subcloned into the EcoRI and XhoI sites of the pSG5.HA vector, which has promoters for expression in vitro and in mammalian cells and provides a NH2-terminal HA-tag for the expressed protein (48), or into the EcoRI and SalI sites of the pM vector (Clontech), a vector for expression of Gal4-DBD fusion proteins from a constitutive SV40 early promoter. Plasmid DNA encoding the pcDNA3.1.HDAC1.myc (49) was a gift from Dr. M.A. Lazar (University of Pennsylvania). pcDNA3.HDAC1.flag (wild-type and H141A mutant) were gifts from Dr. T.P. Yao (Duke University), and the pSG5.HA.p53 expression vector was described previously (14, 35). Reporter genes containing p21-LUC and GK1 were described previously (14). Bacterial expression vectors for generating GST fusion proteins of Zac11-520 and Zac11-220 were constructed by inserting the appropriate PCR fragments into pGEX-4T1 vector (GE HealthCare) at the EcoRI-XhoI sites. GST-Zac11-704 was described previously (15).
Cell Culture and Transient Transfection Assays
HeLa, HeLa/vector, and HeLa/p53 shRNA cells were grown in DMEM supplemented with 10% charcoal/dextran-treated fetal bovine serum as described previously (29). Transient transfections and luciferase assays were done in 24-well culture dishes as described previously (50). Total DNA was adjusted to 1 μg by adding each of the corresponding empty vectors. The luciferase activity of the transfected cell extracts is presented as relative light units and expressed as the mean and SD from three transfected cultures (Promega). Because the expression of many control vectors was used to monitor transfection efficiency which might be affected greatly by Zac1, internal controls were not used. Instead, the reproducibility of the observed effects was determined in multiple independent transfection experiments.
Immunoprecipitation and Immunoblots
For the association of HA.Zac1 protein with myc.HDAC1 or myc.HDAC4, COS7 cells were grown in DMEM supplemented with 10% charcoal/dextran-treated fetal bovine serum and transfected with these expression vectors. After 48 h of transfection, cells were lysed in radioimmunoprecipitation assay buffer [100 mmol/L Tris-HCl (pH 8.0), 150 mmol/L NaCl, 0.1% SDS, and 1% Triton 100] at 4°C. Lysates were subjected to immunoprecipitation with antibody against HA tag (3F10; Roche) for 3 h, followed by adsorption to Sepharose-coupled protein A/G (Santa Cruz Biotechnology) for 3 h. Immunoprecipitates were separated by SDS-PAGE and analyzed by immunoblots. For determination of total protein levels, aliquots of cell lysates were subjected to direct immunoblots. Immunoblots were done as previously described using 5% to 10% of the extract from lysates for immunoprecipitation, using antibodies against Zac1 or HDAC1 (G-18 and N-19; Santa Cruz, Biotechnology) for endogenous assays and HA- or myc-tag (9E10; Invitrogen) for tagged proteins. Transfection of HeLa cells and immunoblots of transfected cell extracts or TSA-treated HeLa cell extracts were done as described above, using antibodies against p21, p53, HDAC1, HuR, Gal4DBD, and cyclin D1 (C19, DO-1, H-51, 3A2, RK51C, and M-20; Santa Cruz Biotechnology), flag tag (M2; Sigma), hZac1 (Bioman), and HA.
RT-PCR Analysis
Total RNA was extracted from growing HeLa/vector and HeLa/p53 shRNA cells, using a total RNA reagent (Bioman) according to the manufacturer's instructions. One microgram of total RNA was subjected to reverse transcriptase, using random hexamers for 60 min at 42°C (Promega). RT-PCR was done in the linear range (30 cycles) with primers specific for p21, p53, and GAPDH. Primers for the amplification of target genes were as follows: p21, upper, 5′-atgtcagaaccggctgg-3′; lower, 5′-ttagggcttcctcttgg-3′; p53, upper, 5′-atggaggagccgcagtc-3′; lower, 5′-ccatgcaggaactgttac-3′; GAPDH, upper, 5′-aacggatttggccgtattggg-3′; lower, 5′-gggatgaccttgcccacagcc-3′. Thermocycling conditions were as follows: 1 cycle at 95°C for 5 min and 30 cycles at 95°C for 40 s, 46°C for 40 s, and 70°C for 40 s. Amplification products were subjected to 1.2% agarose gel electrophoresis and stained with ethidium bromide.
Protein-Protein Interaction Assays
For GST pull-down assays, 35S-labeled proteins were produced with the TNT T7-coupled reticulocyte lysate system (Promega), and GST fusion proteins were produced in Escherichia coli BL21. Proteins were translated in the presence of 35S-mentinion in vitro, incubated with immobilized GST fusion proteins, eluted and analyzed by gel electrophoresis as previously described (50).
DNA Affinity Precipitation Assay
One oligonucleotide containing a biotin-labeled sense strand primer (5′-agactgggcatgtctgggca-3′) for the p53-responsive element site 2 within p21 promoter (−1508/−1492) or (5′-gtactaacaGGGGCCCCatttaatcat-3′) for the Zac1-binding site (G4C4 type) was used in the assays. HeLa nuclear extracts were prepared as previously described (51) and were incubated with DNA affinity precipitation assay binding buffer [10 mmol/L Tris-HCl (pH 7.5), 50 mmol/L KCl, and 1 mmol/L DTT]. Poly(dI-dC) competitor was incubated with the nuclear extracts, followed by incubation with biotin-labeled p53 or Zac1 double-stranded oligonucleotide at 4°C for 3 h on a rocking platform. After incubation, streptavidin-HPR agarose (Sigma) was added to the reaction and incubated at room temperature for 1 h on a rocking platform. The specific protein-DNA-agarose complex was washed and analyzed by SDS-PAGE. Detection was done by adding anti-p53 antibody (DO-1; Santa Cruz Biotechnology), anti-HDAC1 antibody (H-51; Santa Cruz Biotechnology), or anti-hZac1 antibody (Bioman) and immunoblot analysis.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Acknowledgments
We thank Dr. MA Lazar (University of Pennsylvania, Philadelphia, PA) for pcDNA3.1.HDAC1.myc expression vector, Dr. W. el-Diery (University of Pennsylvania) for pG13-LUC reporter gene, and Dr. T.P. Yao (Duke University, Durham, NC) for pcDNA3.HDAC1.flag (wild-type and H141A mutant) expression vectors.
Footnotes
Grant support: National Health Research Institute and National Science Council, Taiwan, Republic of China (NHRI-EX92-9224NC, NHRI-EX93-9224NC, NHRI-EX94-9224NC, and NSC 96-2320-B-016-004; S-M. Huang).
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Accepted April 10, 2008.
- Received March 5, 2008.
- Revision received April 8, 2008.
- American Association for Cancer Research