Histone deacetylase inhibitor trichostatin a potentiates doxorubicin-induced apoptosis by up-regulating PTEN expression

Authors


Abstract

BACKGROUND.

The tumor suppressor gene PTEN is a major negative regulator of the PI3K/Akt cellular survival pathway. Overexpression of PTEN by adenoviral transfection increases doxorubicin-induced apoptosis. Whereas doxorubicin-induced apoptosis can be potentiated by the histone deacetylase (HDAC) inhibitor Trichostatin A (TSA), the mechanisms underlying this process remain unclear. The aim of this work was to investigate whether changes in PTEN expression are involved in TSA/doxorubicin-induced apoptosis.

METHODS.

We treated 293 T cells with TSA and doxorubicin, detected apoptosis by using Hoechst 33342 staining, and examined changes of PTEN and Egr-1 expression by quantitative real-time polymerase chain reaction (PCR). Luciferase reporter assay was used to evaluate the promoter activity of PTEN and Western blot and enzyme-linked immunosorbent assay (ELISA) were used to confirm changes in the expression of PTEN. The chromatin immunoprecipitation (ChIP) assay was performed to estimate the acetylation level of PTEN promoter.

RESULTS.

Doxorubicin-induced apoptosis was enhanced by TSA, whereas small interfering RNA (siRNA) targeting PTEN inhibited TSA/doxorubicin-induced apoptosis. Also, TSA promoted Egr-1 expression, which is the main transcription factor of PTEN, and this resulted in up-regulation of PTEN expression, which consequently potentiated apoptosis. Moreover, histone acetyltransferase p300 was able to synergistically activate PTEN transcription with Egr-1, implicating the role of histone acetylation in the regulation of PTEN expression.

CONCLUSIONS.

TSA promoted doxorubicin-induced apoptosis through a mechanism that involved the stimulation of Egr-1 expression, acetylation of core histones at the PTEN promoter, and consequently induction of PTEN transcription. These findings provide a theoretical basis for the therapeutic application of combined treatment of TSA/doxorubicin for cancer. Cancer 2007. © 2007 American Cancer Society.

The tumor suppressor gene phosphatase and tensin homolog deleted on chromosome 10 (PTEN) has been mapped on 10q23, a chromosomal region that is frequently loss of heterozygosity (LOH) in various human tumors, including brain, bladder, prostate, and endometrial cancers.1, 2 PTEN is both a lipid phosphatase and a protein phosphatase.3, 4 Being a lipid phosphatase, PTEN can dephosphorylate phosphatidylinositol 3,4,5-triphosphate (PIP3), the product of phosphatidylinositol 3-kinases (PI3K), negatively regulates the ubiquitous PI3K pathway. As a protein phosphatase, PTEN is able to dephosphorylate the highly acidic peptide substrates. A recent study has provided evidence that a combined treatment with the overexpression of PTEN via adenoviral transfection (Ad.PTEN) together with doxorubicin brought about a promoted therapeutic effect on prostate cancer.5 As a cancer chemotherapeutic agent used in treatments of a variety of solid tumors, osteosarcomas, and soft-tissue sarcomas, doxorubicin can intercalate into DNA molecules and alter their helical torsion resulting in DNA damage.6–8

Histone deacetylase (HDAC) inhibitors such as Trichostatin A (TSA) have emerged as accessory therapeutic agents for multiple human cancers. These HDAC inhibitors can induce transcriptional activation of target genes through reducing HDAC activity and hence affecting acetylation status of core histones. Studies revealed that HDAC inhibitors can induce tumor growth arrest and promote differentiation or apoptosis in a variety of cell lines.9, 10 It was reported that treatment of cells with TSA before exposure to doxorubicin enhanced the sensitivity of apoptosis in K562 cells11 and in anaplastic thyroid carcinoma (ATC).12

The immediate early growth response-1 (Egr-1) gene encodes a zinc finger transcription factor that binds to a GC-rich consensus sequence GCG(T/G) GGGCG in promoters of a variety of target genes.13, 14 In response to stimuli such as growth factors, cytokines, hormones, phorbol esters, irradiation, and stresses the Egr-1 gene can be rapidly and transiently induced. It appears that Egr-1 acts as a tumor suppressor in certain cells by initiating apoptosis upon activation of PTEN15 and p53 expression.16 Moreover, it was shown that treatment of C2C12 cells with TSA induced Egr-1 expression.17

The regulation of PTEN has been extensively studied. It was reported that PTEN expression was regulated by p53,18 PPARγ,19 Egr-1,15 NF-κB,20, 21 and TNF-β.22 However, whether histone acetylation modification is involved in PTEN transcriptional regulation has not been elucidated. Whereas the HDAC inhibitor TSA can modulate cellular response to doxorubicin, the mechanism underlying this process is unclear. Moreover, a recent study showed that treatment of C2C12 cells with TSA induced the expression of Egr-1, which has been known to be the major transcription factor of PTEN.17 It is therefore intriguing to speculate that TSA may up-regulate PTEN expression by promoting histone acetylation at its promoter and/or by inducing Egr-1 expression. The aim of the present study was to clarify the functional relations between the action of HDAC inhibitor TSA and doxorubicin and to explore the mechanism by which TSA increases the apoptosis induced by doxorubicin. We discovered that TSA was able to promote the doxorubicin-induced apoptosis partly through up-regulating PTEN, during which Egr-1 played a central role. Furthermore, we demonstrated that the histone acetyltransferase p300 could cause the hyperacetylation of histones at the PTEN promoter to stimulate PTEN transcription through interacting with Egr-1.

MATERIALS AND METHODS

Plasmids

The human PTEN gene promoter luciferase reporter construct was obtained by subcloning the 1978 basepair (bp) genomic DNA region upstream of the human PTEN gene into the pGL-3basic-luc vector.15 The expression vectors containing the wildtype p300 (pCI-p300) and its HAT-deletion derivative (pCI-p300, HATΔ1472–1522) were generously provided by Dr. Joan Boyes (Institute of Cancer Research, UK).

Small interfering RNAs (siRNA) targeting PTEN (5′-GAC TTG AAG GCG TAT ACA G-3′),23 Egr-1 (5′-GCA AGT GGA TCT TGG TAT G-3′),24 and p300 (5′-TGA CAC AGG CAG GCT TGA C-3′)25 genes were synthesized according to published data. Oligonucleotides that represent these siRNAs were cloned into the pSliencer2.0-U6 vector (Ambion, Austin, Tex) between BamHI and HindIII restriction sites following the manufacturer's instructions.

Cell Culture

The 293T human embryonic kidney epithelial and A549 type II alveolar adenocarcinoma cell lines were purchased from the Institute of Cell Biology (Shanghai, China). Cells were cultured in DMEM medium supplemented with 10% FBS (fetal bovine serum), 100 U/mL penicillin, and 100 μg/mL streptomycin and kept in a humidified atmosphere of 5% CO2.

Transient Transfection and Luciferase Assay

For transient transfection, cells were seeded in 24-well or 6-well plates and cultured for 18 hours before being transfected using a standard calcium phosphate transfection method. Cells were then incubated at 37°C for 4–6 hours before changing fresh medium. After transfection, cells were cultured for 30 hours before harvested, then washed with phosphate-buffered saline (PBS) and lysed in 30 μL lysis buffer. Reporter gene expression was measured and quantified using a dual Luciferase Reporter Assay System (Promega, Madison, Wis). Relative luciferase activity was analyzed by using a Turner Designs TD20/20 Luminometer (Sunnyvale, Calif). Firefly luciferase activity was normalized to the activity of the Renilla luciferase control.26 Extracts from at least 3 independent transfection experiments were assayed in triplicate. The results are shown as means ± SD.

Detection of Apoptosis

Cells were seeded in 24-well plates and cultured for 18 hours. After transient transfection or treatments with TSA and doxorubicin, cells were cultured for 48 hours. Then cells were treated with 2% formaldehyde for 10 minutes at 37°C and washed with PBS. One mL of 1% Triton X-100 and 4 μg/mL Hoechst33342 (Sigma, St. Louis, Mo) in PBS were slowly added and the cells were incubated at room temperature for 15 minutes before replaced in 1 mL PBS. Cells with typical apoptotic nuclear morphology (nuclear shrinkage, condensation, and fragmentation) were identified and counted within randomly selected fields. The apoptosis rates were calculated based on at least 300 cells from 3 independent experiments and are shown as means ± SD.

RNA Isolation and Quantitative Real-Time Polymerase Chain Reaction (PCR)

Total RNA was extracted and reverse-transcribed to cDNA using the RNA extraction and RT Systems supplied by Promega. The resulting cDNA was diluted 5-fold with RNase-free water. Quantitative real-time PCR analysis was performed using an ABI Prism 7000 Sequence Detection System (Applied Biosystems, Foster City, Calif). SYBR Green (TOYOBO, Japan) was used as a double-stranded DNA-specific fluorescent dye. β-Actin was used as a housekeeping gene for standardizing PTEN and Egr-1 mRNA expression. Amplification primers were 5′-GCG TGC AGA TAA TGA C-3′ and 5′-GAT TTG ACG GCT CCT-3′ for PTEN gene; 5′-AGC CCT ACG AGC ACC TG-3′ and 5′-CGG TGG GTT GGT CAT G-3′ for Egr-1 gene; and 5′-GAC CCT CAG CTT TTA GGA ATC C-3′ and 5′-TGC CGT AGC AAC ACA GTG TCT-3′ for p300 gene. Data were analyzed by using the 2−ΔΔCt method.27 All the results represent means ± SD of 3 independent experiments.

Western Blot and Enzyme-Linked Immunosorbent Assay (ELISA)

Cells were treated with TSA, incubated for 48 hours, and then lysed in 50 μL luciferase lysis buffer. Cell lysates were separated by SDS-PAGE in 12% gels and then transferred to nitrocellulose membrane and subjected to Western blot analysis with rabbit polyclonal antibody against PTEN, and rabbit polyclonal antibody against β-actin (Santa Cruz Biotechnology, Santa Cruz, Calif). Cell lysates were coated on a polystyrene plate for ELISA using polyclonal antibodies against PTEN, β-actin, and p300 (Santa Cruz).

Chromatin Immunoprecipitation (ChIP)

ChIP assays were carried out using a kit supplied by Upstate Biotechnology (Lake Placid, NY) following the manufacturer's protocol. Cells were plated at a density of 1 × 105/mL in 24-well plate and cultured for 48 hours. After transient transfection or TSA and doxorubicin stimulation, cells were cross-linked with 2% formaldehyde for 10 minutes at 37°C, then lysed in SDS lysis buffer (1% SDS, 10 mM EDTA, 50 mM Tris, pH 8.1) with protease inhibitors. The sonicated lysates were processed using a ChIP assay kit, essentially as described by the manufacturer (Upstate Biotechnology). Antibodies against Ac-H3 (cat. no. 06-599) and Ac-H4 (cat. no. 06-866) were purchased from Upstate Biotechnology and antibody against p300 protein (sc-585) was purchased from Santa Cruz Biotechnology. Immunoprecipitated chromatin was analyzed by quantitative PCR (ABI Prism 7000 Sequence Detection System Instrument) using SYBR green dye with primers specific to sequences at the PTEN promoter, i.e., 5′-AAA CGA GCC GAG TTA CCG-3′ and 5′-GAC TGC ATT CGC TCT TTC CT-3′ for RE I; 5′-CGG GCG GTG ATG TGG C-3′ and 5′-GCC TCA CAG CGG CTC AAC TCT-3′ for RE II; and 5′-GTT GGT CTC TCC CCT TCT A-3′ and 5′-CGC AGC CGG GTA ATG-3′ for RE III.

Cellular Immunofluorescence

The 293T cells were seeded in 6-well plates and cultured for 18 hours, then transient transfected with p300 expression plasmid or control empty vector. The cells were fixed with 1% formaldehyde in culture medium for 10 minutes at 37°C and permeabilized with 0.2% Triton X-100 in PBS for 10 minutes at 4°C. The PTEN protein was detected with antibody against PTEN and visualized with an FITC-conjugated antirabbit IgG secondary antibody (Zhongshan, China). Photographs were taken with a Nikon microscope with a fluorescein isothiocyanate filter.

Statistical Analysis

A Student test was used to calculate the statistical significance of the experimental results. The significance level was set as *P < .05 and **P < .01.

RESULTS

PTEN Participated in TSA/Doxorubicin-Induced Apoptosis

It was reported that the doxorubicin-induced apoptosis can be enhanced by TSA,11, 12 and this has been confirmed in our experimental system. As shown inFigure 1A, no significant apoptosis was observed in control cells. Treatments of TSA and doxorubicin alone induced apoptosis rates of approximately 8% and 20%, respectively. Meanwhile, a combination of both TSA and doxorubicin resulted in an apoptosis rate of as high as 76.9% (Fig. 1A). In the next experiments we endeavored to explore some of the molecular mechanisms underlying this phenomenon.

Figure 1.

PTEN played important roles in the TSA/doxorubicin-induced apoptosis. 293T cells were treated with 250 nM Trichostatin A (TSA) or 200 nM doxorubicin (DOXO) for 48 hours and the apoptosis was examined microscopically after Hoechst33342 staining. (a) The apoptosis rates were calculated based on at least 300 cells from 3 independent experiments and are shown as means ± SD (*P < .05, **P < .01). (b) 293T cells were transiently transfected for 48 hours with siRNA targeting PTEN gene (PTENsiRNA) or control vector, treated with TSA and doxorubicin (DOXO), and the apoptosis rate was determined as above. (c) 293T cells were treated with TSA and doxorubicin (DOXO), transiently transfect with PTENsiRNA or control vector for 48 hours, and the PTEN mRNA level was measured by quantitative real-time polymerase chain reaction (PCR) (**P < .01).

We first focused on the tumor suppressor gene phosphatase and tensin homolog deleted on chromosome 10 (PTEN) because it was reported that overexpression of PTEN promoted the therapeutic effect of doxorubicin.5 To assess whether PTEN plays a role in the TSA-induced apoptosis we constructed an siRNA plasmid (PTENsiRNA) targeting the PTEN gene and transiently transfected it into 293T cells. We identified that knockdown of PTEN by siRNA partially inhibited the apoptosis induced by TSA and doxorubicin (Fig. 1B). We also found that PTEN expression was up-regulated by TSA and doxorubicin, whereas siRNA targeting PTEN could reduce PTEN expression (Fig. 1C). These findings implied that the effect of TSA in enhancing the doxorubicin-induced apoptosis was associated, at least in part, with the expression of PTEN.

PTEN Participated in TSA/Doxorubicin-Induced Apoptosis Through Up-regulation of Egr-1

We then intended to investigate whether any genes upstream of PTEN are involved in this process. The first gene we came across was the immediate early growth response-1 (Egr-1) gene because it can induce apoptosis by stimulating p53 and PTEN expression, and there has been evidence that TSA induces Egr-1 expression.17

To determine whether Egr-1 participates in the TSA/doxorubicin-induced apoptosis, we constructed an siRNA plasmid targeting the Egr-1 (Egr-1siRNA) and transiently transfected the 293T cells. We identified that TSA and doxorubicin alone induced apoptosis rates of 8% and 24%, respectively, whereas knockdown of Egr-1 by siRNA reduced the rates to 3% and 14%, respectively. Egr-1siRNA reduced the TSA/doxorubicin-induced apoptosis by approximately 50% (Fig. 2A). Subsequently, to find out whether Egr-1 is the target of TSA and doxorubicin action, we transfected 293T cells with Egr-1siRNA and treated the cells with TSA and doxorubicin. As shown in Figure 2B, the Egr-1 mRNA level was elevated by 3.5-fold and 1.8-fold when the cells were treated with TSA and doxorubicin, respectively, whereas a nearly 8-fold increase was seen when the cells were treated simultaneously with both TSA and doxorubicin. Meanwhile, transfection of the Egr-1siRNA expression vector markedly down-regulated Egr-1 expression (Fig. 2B) and PTEN expression (Fig. 2C), even when treated with TSA and doxorubicin. We presume that the expression of PTEN was decreased because the transcription factor Egr-1 was knocked down by Egr-1siRNA. These results suggested that both PTEN and Egr-1 participated in the TSA/doxorubicin-induced apoptosis.

Figure 2.

PTEN participated in the Trichostatin A (TSA)/doxorubicin-induced apoptosis through up-regulation of Egr-1. (a) 293T cells were transiently transfected for 48 hours with an siRNA targeting the Egr-1 gene (Egr-1siRNA), or control vector, treated with TSA and doxorubicin (DOXO), and the apoptosis rate was determined (**P < .01). (b) 293T cells were treated with TSA and doxorubicin (DOXO), transiently transfected with Egr-1siRNA or control vector for 48 hours, and the Egr-1 mRNA level was measured by quantitative real-time polymerase chain reaction (PCR) (*P < .05, **P < .01). (c) The level of PTEN mRNA transcription was estimated by quantitative real-time PCR (*P < .05, **P < .01).

TSA Induced Histone Hyperacetylation and Up-regulated PTEN Expression

We have shown above that TSA increased the doxorubicin-induced apoptosis through regulation of PTEN and Egr-1 expression. As an HDAC inhibitor, TSA is capable of stimulating transcription of many genes by altering histone acetylation status at the gene's promoters. We then asked whether TSA can directly stimulate PTEN transcription, or indirectly via Egr-1 regulation. To assess this, we first treated 293T cells with TSA and found that it increased PTEN promoter activity (Fig. 3A, left). This implied that histone acetylation might be involved in PTEN regulation. Also, we treated 293T cells and another tumor cell, A549, with TSA, and as shown in Figure 3A (middle, right), the PTEN mRNA level was increased by 8-fold in 293T cells and by 2-fold in A549 cells, respectively, further suggesting that TSA could activate PTEN transcription. Furthermore, our Western blot and ELISA (Fig. 3B) data revealed that TSA could increase PTEN protein expression.

Figure 3.

Trichostatin A (TSA) inhibited the activities of histone deacetylase (HDAC), enhanced histone acetylation at PTEN promoter regions, and stimulated PTEN expression. 293T cells were transiently transfected for 30 hours with PTEN promoter-luciferase and Renilla-luc reporter constructs, and then treated with the HDAC inhibitor TSA. (a, left) Relative luciferase activity was normalized to Renilla luciferase control. 293T and A549 cells were treated with 125 nM TSA for 30 hours and the PTEN expression was estimated by (a, middle, right) quantitative real-time polymerase chain reaction (PCR) and (b) Western blot or ELISA analysis (*P < .05). (c) 293T cells were cotransfected with the 6 human HDACs (HDAC1-6) and PTEN promoter-luciferase reporter plasmid for 30 hours and the relative luciferase activity was measured (*P < .05, **P < .01). (d) A diagram showing the positions of Egr-1 binding sites at the PTEN promoter. The regions amplified by the PCR primers in chromatin immunoprecipitation (ChIP) assays are indicated (RE I, II, III) (upper). 293T cells were treated with TSA for 30 hours and the acetylation levels of histone H3 (left) and histone H4 (right) were examined by ChIP and quantitative real-time PCR (d, lower) (*P < .05).

We then cotransfected 293T cells with PTEN promoter-luciferase reporter plasmid with the 6 human HDACs (HDAC1-6), and as can be seen in Figure 3C, among the 6 HDACs tested, 5 (HDAC1, 3, 4, 5, and 6) were able to inhibit the PTEN promoter activity, to variable extents, providing further hints that TSA may act on PTEN through inhibiting these HDACs. However, the straight evidence for the action of TSA on the PTEN gene regulation came from our chromatin immunoprecipitation (ChIP) experiments using antibodies against Ac-H3 and Ac-H4. For a quantitative estimation of the precipitated DNA, coordinates of primers relative to the translation start site were designed as follows: RE I (−1703/−1477, distant domain); RE II (−920/−784, major Egr-1 binding domain), and RE III (−413/−156, proximal Egr-1 binding domain) (Fig. 3D, upper). The results showed that TSA treatment of 293T cells increased the acetylation level of histone H3 at the Egr-1 binding domains of the PTEN promoter, with the highest increase up to 55-fold at the RE III region (Fig. 3D, lower left), whereas the histone H4 acetylation level was also significantly increased at the major Egr-1 binding domain (RE II), but it was decreased at RE III (Fig. 3D, lower right). These results clearly indicated that TSA induced PTEN transcription by inhibiting HDAC activities that facilitated histone hyperacetylation.

Histone Acetyltransferase p300 Promoted PTEN Expression

Because the status of histone acetylation is balanced by a coordinated function of both histone acetyltransferases (HATs) and HDACs, we were then curious to know if any HATs participate in this process. Transcriptional coactivator p300 was the first HAT we were interested in because it can regulate the activities of many transcription factors and it has a physical interaction with Egr-1.28 To determine whether p300 plays a role in PTEN regulation, we transiently transfected 293T cell with p300 expression plasmid and the ectopic expression of p300 was confirmed by both quantitative real-time PCR (Fig. 4B, left) and ELISA (Fig. 4B, right). The transfection results demonstrated that both PTEN promoter activity (Fig. 4A, left) and endogenous mRNA level (Fig. 4A, right) were increased by the overexpression of p300 in a dose-dependent manner. Meanwhile, the protein level of PTEN was also enhanced accordingly after p300 transfection (Fig. 4C). To further verify the function of p300, we constructed an siRNA targeting p300 gene (p300siRNA), transiently transfected 293T cells with p300 and p300siRNA plasmid, followed by ChIP assays with antibodies against Ac-H3, Ac-H4, and p300. As shown in Figure 5A, acetylation levels of histone H4 at all the PTEN promoter domains were elevated, especially at the RE III region, when transfected with p300 plasmid (Fig. 5A, upper left). Conversely, the histone acetylation levels at these sites were reduced when transfected with p300siRNA (Fig. 5A, upper left). Moreover, our ChIP assays also showed that transfection of p300 plasmid increased the presence of p300 protein at the PTEN promoter, especially at the RE II and RE III regions (Fig. 5A, lower), suggesting that p300 was associated with the PTEN promoter and increased histone H4 acetylation, but it had little effect on histone H3 (data not shown).

Figure 4.

The histone acetyltransferase p300 promoted PTEN expression. (a, left) 293T cells were transiently transfected with p300 expression vector and the PTEN promoter activity was examined by luciferase reporter assay (**P < .01). The PTEN expression was assayed by using (a, right) quantitative real-time polymerase chain reaction (PCR) (*P < .05, **P < .01) and (c) cellular immunofluorescence. (b) The ectopic expression of p300 was confirmed by quantitative real-time PCR and enzyme-linked immunosorbent assay (ELISA) (*P < .05, **P < .01).

Figure 5.

p300 was associated with PTEN promoter and synergistically activated PTEN promoter with Egr-1. 293T cells were transiently transfected with p300 expression vector or siRNA targeting the p300 gene (p300siRNA) for 48 hours. (a, upper) Chromatin immunoprecipitation (ChIP) was performed using antibodies against Ac-H4 and p300, quantified by real-time polymerase chain reaction (PCR) and reverse-transcriptase PCR (RT-PCR) (*P < .05). (a, lower) The expression of p300 was detected by real-time RT-PCR. 293T cells were cotransfected with p300, Egr-1 or p300siRNA, p300 delat HAT vectors for 30 hours. (b, c, d) Relative luciferase activity was determined as described above (*P < .05, **P < .01).

To ascertain whether p300 synergistically activates PTEN promoter with Egr-1, we cotransfected 293T cells with p300 and Egr-1 plasmids and the results showed that p300 or Egr-1 alone increased PTEN promoter activity by nearly 3-fold (Fig. 5B), whereas a simultaneous transfection of both p300 and Egr-1 caused a sharp increase in activation of the promoter by over 17-fold (Fig. 5B). Next, we cotransfected p300siRNA, wildtype p300, and HAT-deletion p300 (p300 delat HAT) plasmids, and found that knockdown of p300 by siRNA inhibited PTEN promoter activity and prevented p300's synergistic activity with Egr-1 (Fig. 5C). To our surprise, the p300 delat HAT had roughly the same effect on PTEN promoter in contrast to the wildtype p300 (Fig. 5D), implying that the HAT domain of p300 was not essential for PTEN activation, though it was necessary for p300 and Egr-1 interaction (Fig. 5D). Together, these findings suggested that p300 participated in Egr-1-directed PTEN transcriptional activation.

TSA Enhanced the Synergistic Role of p300and Egr-1 in Activating PTEN

In the following experiments we tested the effects of the HDAC inhibitor on the action of p300 and Egr-1 in stimulating the PTEN gene. As shown in Figure 6A, p300 and Egr-1 were able to activate the PTEN promoter synergistically, whereas treatment with the HDAC inhibitor TSA increased this transcriptional activity by around 1000-fold (Fig. 6A, the top bar). Meanwhile, siRNAs targeting p300 and Egr-1 inhibited interaction of p300 with Egr-1, resulting in a reduction of PTEN expression, with or without TSA treatment (Fig. 6B). It is apparent that knockdown of Egr-1 by siRNA had a more intense effect than knockdown of p300 (Fig. 6B). Because there have been reports that Egr-1 can regulate p300 expression,29 and Egr-1 can interact with p300,28 we deduce that in our experimental system Egr-1 may also up-regulate p300, interact with p300, and facilitate the binding of p300 to the PTEN promoter to activate its transcription.

Figure 6.

The histone deacetylase (HDAC) inhibitor Trichostatin A (TSA) enhanced the synergistic effect of p300 and Egr-1 in activating PTEN transcription. (a) 293T cells were transiently transfected with p300 and Egr-1 expression vectors, treated with TSA for 30 hours, and the luciferase reporter activity was determined (*P < .05, **P < .01). (b) 293T cells were cotransfected with p300, Egr-1, p300siRNA, and Egr-1siRNA vectors, and the PTEN expression was examined by quantitative real-time polymerase chain reaction (PCR) (*P < .05, **P < .01).

Overall, data arising from this study support a model that depicts the function of TSA in potentiating doxorubicin-induced apoptosis. Briefly, TSA promotes apoptosis through up-regulation of PTEN in 2 ways. First, as an HDAC inhibitor TSA is able to enhance the histone acetylation of the PTEN promoter that activates PTEN transcription (Fig. 3). In addition, TSA may stimulate Egr-1 expression, which in turn up-regulates PTEN (Fig. 2B). Conversely, the histone acetyltransferase p300 participates in this regulatory process either by acetylating histone H4 at the PTEN promoter to stimulate its transcription (Fig. 5A, upper left) or by functioning synergistically with Egr-1 to activate PTEN (Fig. 5B).

DISCUSSION

Previous studies revealed that TSA could promote apoptosis induced by doxorubicin.11, 12 Also, it has been well established that TSA induces cell differentiation, cell-cycle arrest, and apoptosis in numerous transformed cell lines and cancer cells.30 As a histone deacetylase inhibitor, TSA is thought to exert its anticancer effects by causing the accumulation of acetylated histones, which consequently results in the alteration of transcription of target genes, such as transcription factors NF-κB, GATA1, E2F-1, and MyoD, or components of the transcriptional machinery.31–33 Furthermore, it was proven that key regulators of apoptotic signaling cascades, such as Rb and p53, were altered by HDAC inhibitors.34

The aim of this study was to explore whether the tumor suppressor gene PTEN participated in the TSA/doxorubicin-induced apoptosis. As shown in Figure 1B, siRNA targeting the PTEN gene inhibited TSA/doxorubicin-induced apoptosis, and the mRNA level of PTEN was up-regulated by TSA and doxorubicin treatments (Fig. 1C). Moreover, we found that TSA was able to activate PTEN gene expression at the promoter activity, mRNA, and protein levels (Fig. 3). To gain further insight into this phenomenon, we designed a series of ChIP primer pairs targeting the Egr-1 binding domains (RE II and RE III) and an irrelevant domain (RE I) at the PTEN promoter for ChIP experiments to detect the histone acetylation levels at these domains, considering that Egr-1 is the main transcription factor of PTEN. As can be seen in Figure 3D, there was little change in the acetylation levels of both histone H3 and histone H4 at the RE I domain when the cells were treated with TSA, suggesting that it was not a transcriptional active domain. It has been reported that there are 3 Egr-1 binding sites at the PTEN promoter region spanning −947 to −914 relative to the translation start site,15 and the RE II domain is located in this region. Data from our ChIP assays showed that both histone H3 and histone H4 were hyperacetylated at the RE II domain (Fig. 3D). Previous study has established that the PTEN promoter from −551 to −220 is a transcriptional active domain, and there are 2 Egr-1 binding sites within this domain.35 Thus, these results are suggestive of the TSA function in inducing histone hyperacetylation and facilitating Egr-1 binding to PTEN promoter to stimulate its expression. Apparently, PTEN may be 1 of the target genes functioning in the TSA/doxorubicin-induced apoptosis.

Egr-1 is the major transcription factor of PTEN: it functions as a convergence point for many signaling cascades and regulates the expression of many genes. Here we show that Egr-1 is an intermediate regulator involved in the TSA-induced PTEN activation. We also show that TSA treatment could up-regulate Egr-1 expression (Fig. 2B) and that transient transfection of Egr-1 expression plasmid promoted the PTEN transcription (Fig. 5B). In addition, we found that treatment of cells with TSA significantly increased the transcriptional activity of Egr-1 (Fig. 6B). These findings implied that TSA activated PTEN expression by up-regulating Egr-1 expression or promoting histone hyperacetylation of the PTEN promoter, thus facilitating Egr-1 binding to the promoter.

A next reasonable assumption was that there might be a histone acetylase involved in the PTEN transcription regulation, considering that histone acetylation is a coordinated reversible modification process. Our data demonstrated that the histone acetylase p300 played an important role in PTEN regulation, as it could activate the PTEN promoter and up-regulate PTEN expression at both the mRNA and protein levels (Fig. 4). As a transcriptional coactivator and an HAT, p300 participates in gene transcriptional regulation through multiple mechanisms. It may act as a protein bridge connecting different sequence-specific transcription factors to the transcription apparatus.36, 37 p300 may also provide a protein scaffold for the assembly of a multicomponent transcriptional regulatory complex.37 However, the key property of p300 is its HAT activity, being able to catalyze histone acetylation. We have presented ChIP evidence that p300 could be associated with the PTEN promoter to induce the hyperacetylation of histone H4 (Fig. 5A), although it had little effect on histone H3. As shown in Figure 5D, the HAT-deleted mutant of p300 did not significantly affect its function in promoting the expression of PTEN, but its synergistic function with Egr-1 was markedly reduced (Fig. 5D). It appears that the HAT activity of p300 was not indispensable nor the only mechanism for PTEN regulation. Our results show that knockdown of endogenous expression of p300 by specific siRNA reduced PTEN expression and inhibited Egr-1 transcription (Fig. 5C). Because there has been evidence that p300 can interact with Egr-1,28 we deduce that p300 may participate in PTEN transcriptional regulation mainly through its protein bridge or scaffold mechanism. This assumption is supported by the result that the HAT-deletion p300 also stimulated PTEN expression (Fig. 5D).

Altogether, the experimental data presented in this report suggest a molecular mechanism that TSA potentiates the doxorubicin-induced apoptosis by up-regulating PTEN expression, and the histone acetyltransferase p300 and the transcription factor Egr-1 participate in this process. This molecular model will provide useful information for the development of a combinatorial therapeutic strategy for cancer treatment involving HDAC inhibitors (such as TSA) and doxorubicin.

Acknowledgements

We thank Dr. Eileen Adamson (Burnham Institute, USA) and Dr. Joan Boyes (Institute of Cancer Research, UK) for providing plasmids.

Ancillary