Apoptosis

, Volume 18, Issue 1, pp 110–120

p53 acetylation enhances Taxol-induced apoptosis in human cancer cells

Authors

  • Jae Hyeong Kim
    • Research Institute National Cancer Center
    • Department of Life ScienceResearch Center for Woman’s Disease, Sookmyung Woman’s University
  • Eun-Kyung Yoon
    • Research Institute National Cancer Center
    • Division of Life and Phamaceutical Sciences, The graduate SchoolEwha Womans University
  • Hye-Jin Chung
    • Research Institute National Cancer Center
  • Seong-Yeol Park
    • Research Institute National Cancer Center
  • Kyeong-Man Hong
    • Research Institute National Cancer Center
  • Chang-Hun Lee
    • Research Institute National Cancer Center
  • Yeon-Su Lee
    • Research Institute National Cancer Center
  • Kyungho Choi
    • Research Institute National Cancer Center
    • Department of Biochemistry and Molecular BiologySeoul National University college of Medicine
  • Young Yang
    • Department of Life ScienceResearch Center for Woman’s Disease, Sookmyung Woman’s University
    • Research Institute National Cancer Center
  • In-Hoo Kim
    • Research Institute National Cancer Center
Original Paper

DOI: 10.1007/s10495-012-0772-8

Cite this article as:
Kim, J.H., Yoon, E., Chung, H. et al. Apoptosis (2013) 18: 110. doi:10.1007/s10495-012-0772-8

Abstract

Microtubule inhibitors (MTIs) such as Taxol have been used for treating various malignant tumors. Although MTIs have been known to induce cell death through mitotic arrest, other mechanisms can operate in MTI-induced cell death. Especially, the role of p53 in this process has been controversial for a long time. Here we investigated the function of p53 in Taxol-induced apoptosis using p53 wild type and p53 null cancer cell lines. p53 was upregulated upon Taxol treatment in p53 wild type cells and deletion of p53 diminished Taxol-induced apoptosis. p53 target proteins including MDM2, p21, BAX, and β-isoform of PUMA were also upregulated by Taxol in p53 wild type cells. Conversely, when the wild type p53 was re-introduced into two different p53 null cancer cell lines, Taxol-induced apoptosis was enhanced. Among post-translational modifications that affect p53 stability and function, p53 acetylation, rather than phosphorylation, increased significantly in Taxol-treated cells. When acetylation was enhanced by anti-Sirt1 siRNA or an HDAC inhibitor, Taxol-induced apoptosis was enhanced, which was not observed in p53 null cells. When an acetylation-defective mutant of p53 was re-introduced to p53 null cells, apoptosis was partially reduced compared to the re-introduction of the wild type p53. Thus, p53 plays a pro-apoptotic role in Taxol-induced apoptosis and acetylation of p53 contributes to this pro-apoptotic function in response to Taxol in several human cancer cell lines, suggesting that enhancing acetylation of p53 could have potential implication for increasing the sensitivity of cancer cells to Taxol.

Keywords

Taxolp53ApoptosisAcetylationSirt1HDAC inhibitor

Introduction

Microtubule inhibitors (MTIs) such as Taxol are broadly used as the first- or second-line chemotherapeutics to control various malignancies. MTIs bind to tubulin and interfere with the normal dynamic activity of microtubules, such as mitotic spindle assembly. Derangement of spindle assembly elicits the activation of spindle checkpoint which leads to mitotic arrest [1, 2]. Such mitotic arrest usually brings about apoptotic death of cancer cells [36]. Several mechanisms underlying this cell death process has been proposed. For example, upon mitotic arrest, accumulated cyclin B/CDK1 complex phosphorylates the anti-apoptotic molecule, Mcl1, which is then degraded. Downregulation of Mcl1 leads to activation of pro-apoptotic Caspase-9, which induces eventual cell death [7, 8]. On the other hand, it has been suggested that Taxol also stimulates the secretion of apoptosis-inducing cytokines, such as tumor necrosis factor-α, in a manner independent of its microtubule stabilizing ability [9].

However, there is still much uncertainty that remains to be elucidated for the mechanism of MTI-induced cell death. One of such uncertainty is the role of p53 in MTI-induced apoptosis. Although it has been known that p53 protein is upregulated by Taxol treatment, there was a lot of controversy on pro- or anti-apoptotic role of p53 in this process. When mouse embryonic fibroblasts (MEFs) generated from p53−/− mice were treated with Taxol, their apopotic cell death increased compared to that of MEFs from the wild type mice, which suggests that p53 protects cells from apoptosis [1012]. Consistently, functional inactivation of p53 protein by viral proteins such as HPV E6 and SV40 large T antigen enhanced Taxol-induced apoptosis in human fibroblasts [10]. On the other hand, pro-apoptotic function of p53 was also demonstrated in several studies. Disruption of p53 function by viral proteins in human ovarian cancer cells significantly decreased Taxol-induced apoptosis [13]. In addition, when p53−/− MEFs were transformed with E1A-Ras, they now became more resistant to Taxol-induced cell death than similarly transformed wild type MEFs [12]. Furthermore, treatment of human lung cancer cells with Taxol leads to the accumulation of p53 in the nucleus through microtubule-dependent trafficking, resulting in the activation of p53 target genes such as pro-apoptotic PUMA [14].

The level and function of p53 proteins are known to be regulated by post-translational modifications such as phosphorylation, acetylation, methylation and ubiquitinylation [15]. Among them, phosphorylation and acetylation have received much attention. Phosphorylation of p53 interferes with the interaction between p53 and Mdm2, an inhibitor of p53, and thereby stabilizes p53 proteins. Acetylation of p53 is also known to prevent p53 from binding to Mdm2 [16]. These modifications also enhance the ability of p53 to induce transcription of downstream genes involved in apoptosis such as BAX and PUMA [1719]. The role of post-translational modification of p53 in Taxol-induced cell death is not well-defined either, as is the case with the role of p53 proteins themselves. There is a few reports showing that MTI treatment stabilizes p53 protein by phosphorylation, which allowed the transcriptional activation of downstream p53 target genes [20, 21]. However, the role of acetylation in this process has not been demonstrated yet.

In this study, we investigated these issues using series of cancer cell lines, in which p53 gene is either intact (p53 wild type cells) or deleted (p53 null cells). We found that p53 null cells are more resistant to Taxol-induced cell death than p53 wild type cells. Acetylation of p53 increased upon Taxol treatment and accentuation of acetylation through inhibition of deacetylases enhanced Taxol-induced apoptosis. Furthermore, acetylation-defective mutant of p53 partially abolished pro-apoptotic effect of p53 in Taxol-treated cells. Thus, we now offer evidence of a novel role of p53 acetylation in Taxol-induced apoptosis in human cancer cells.

Materials and methods

Cell lines and reagents

HCT116 and MCF7 cells were grown in Dulbecco’s Modified Eagle’s Medium (Invitrogen, Carlsbad, USA) supplemented with 10 % fetal bovine serum (FBS) and 1 % penicillin–streptomycin (P/S) (Invitrogen). A549 and H1299 cells were grown in RPMI1640 medium (Invitrogen) supplemented with 10 % FBS and 1 % P/S. All cells were maintained in a 5 % CO2 atmosphere at 37 °C. MS-275 was purchased from BioVision (Mountain View, USA). Other reagents used in this study were purchased from Sigma-Aldrich (St. Louis, USA).

Cell survival assay with MTT

HCT116 p53 wild type or p53 null cells were incubated for 48 h with various doses of Taxol (2 nM, 5 nM, 10 nM, and 25 nM) and the cells were treated with MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) solution for 2 h. Mitochondrial dehydrogenase converted-formazans were dissolved with DMSO and the optical density (OD) of each sample was measured at 560 nm using an ELISA reader (Molecular Devices ELISA Reader, Sunnyvale, USA), and the results were analyzed with SoftMax Pro 5 software.

Apoptotic profiling

Harvested cells were washed once with PBS, fixed with ice-cold 70 % ethanol, and stored at 4 °C. Cells were stained in 1 ml PBS containing 50 μg/ml propidium iodide (PI) and 100 U RNase A for 30 min at 37 °C. In all experiments, at least 10,000 events were acquired using a FACS Calibur flow cytometer and the results were analyzed with Cellquest software (Becton–Dickinson Immunocytometry Systems, San Diego, CA).

Annexin V staining

Cells were harvested and washed with ice-cold PBS. Approximately 2 × 105 cells were diluted in 100 μl cold binding buffer (10 mM HEPES pH 7.4, 140 mM NaCl, and 2.5 mM CaCl2) and stained with Annexin V-FITC and PI for 15 min at room temperature in the dark. After the staining, additional 500 μl of binding buffer was added, and flow cytometry analysis was conducted immediately.

Western blot analysis

The cells were lysed in the lysis buffer (20 mM Tris pH 7.4, 5 mM EDTA, 10 mM Na4P2O7, 100 mM NaF, 2 mM Na3VO4, 1 % NP-40, 0.2 % protease inhibitor cocktail, and 1 mM PMSF). Protein extracts were resuspended in loading buffer, boiled for 5 min and subjected to 8–12 % SDS–polyacrylamide gel electrophoresis. Separated proteins were transferred onto trans-blot nitrocellulose membranes (Schleicher & Schuell, Keene, USA), blocked in TBS-Tween 20 buffer containing 5 % skim milk and incubated with each of the following mouse monoclonal antibodies against phospho-p53-S15 (Cell Signaling, Danvers, USA), p53, Mdm2 (Santa Cruz Biotechnology, Santa Cruz, USA), β-actin, and α-tubulin (Sigma), as well as rabbit polyclonal antibodies against Sirt1 (kindly provided by Ja-Eun Kim, Kyunghee University, Seoul, Korea), phospho-p53-S20, acetylated-p53-K382, PUMA, PARP, Caspase-3 (Cell Signaling), BAX (Santa Cruz Biotechnology) and a rabbit monoclonal antibody against acetylated-p53-K373 (Epitomics, Inc., Burlingame, USA). Protein expression was detected by chemiluminescence using SuperSignal West Pico Chemiluminescent Substrate (Pierce Rockford, USA) after incubation with horseradish peroxidase-tagged secondary antibodies (Jackson Immuno Research Laboratories, Inc., West Grove, USA).

Blocking Sirt1 mRNA expression using siRNA

Sirt1 mRNA expression was knocked down by transfection of cells with a siRNA specific for SIRT1. The sense sequence of the siRNA oligo used was “ACUUUGCUGUAACCCUGUA” [22]. Short oligo transfection was performed using Lipofectamine (Invitrogen) according to the manufacturer’s recommended protocol.

Plasmid constructs and site-directed mutagenesis

The wild-type p53 cDNA insert was subcloned into the pcDNA3.1 expression vector (Invitrogen) using EcoRI and XhoI restriction enzymes (Fermentas, Glen Burnie, USA). Using the parental plasmid pcDNA3.1-p53, the acetylation-defective point mutations of eight lysine residues of p53 protein sequence (K120R, K164R, K370R, K372R, K373R, K381R, K382R and K386R) were created using a two-step process. First, to create seven point mutations, PCR was performed using the following mutation-containing primers: 5′-CGCGGGATGGCCATCTACAGGCAGTCA-CAG-3′ and 5′-AGTCCTCGAGTCAGTCTGAGTCAGGCCCTTCTGTCCTGAACATGAGTCTTCTATGGCG GGAGGTAGACTGACCCCTTCTGGACCTCAGGTGGCTG-3′. The PCR product was subcloned into pcDNA3.1-p53 using the NcoI and XhoI restriction enzymes. Second, the eighth point mutation was added using the QuickChange II Site-Directed Mutagenesis kit (Stratagene, Santa Clara, USA), according to the manufacturer’s instructions. The sense oligonucleotide sequence used for this mutation was 5′-CATTCTGGGACAGCCAGGTCTGTGACTTGCACG-3′. All the sequences were confirmed by automatic DNA sequencing. p53 null HCT116 cells or H1299 cells were transfected with either the wild-type p53 or its acetylation-defective mutant plasmid using the AMAXA Nucleofector (Lonza, Cologne, Germany) transfection method according to the manufacturer’s recommended protocol. 24 h after transfection, the medium was changed, and the cells were treated with Taxol for 24–48 h before harvesting.

Results

p53 enhances Taxol-induced apoptosis in HCT116 colon cancer cells

To examine the role of p53 in Taxol-induced cytotoxicity, we utilized a human colon cancer cell line, HCT116. While the original HCT116 cells have the intact p53 gene and proteins (p53 wild type), p53 null HCT116 cells generated by homologous recombination of p53 gene loci do not express p53 proteins [23]. Therefore, this pair of cell lines provides an ideal system to study the role of p53 in cancer cell biology. First, we checked the p53 protein expression in response to Taxol by western blot analysis, since it was reported that Taxol upregulates p53 proteins in cancer cell lines. As expected, p53 protein levels increased by Taxol in p53 wild type cells, whereas p53 null cells did not express p53 proteins at all (Fig. 1a). Especially, kinetic analysis showed that p53 upregulation plateaued 24 h after the treatment. This time point is the one at which Taxol-induced mitotic arrest is released and apoptotic process begins as indicated by cyclin B expression and phosphorylation of histone H3 (Supplemental Fig. 1). Therefore, we decided to compare the apoptotic profiles of the two cells lines at 24 h time point using flow cytometry. When p53 wild type HCT116 cells were treated with Taxol, sub-G1 fraction, which represents apoptotic cell population, increased in a dose-dependent manner. In contrast, the increase in sub-G1 fraction by Taxol was significantly less in p53 null cells under the same conditions (Fig. 1b). Consistently, the actual survival of p53 null cells was higher than that of p53 wild type cells after Taxol-treatment (Fig. 1c). Thus, p53 expression is upregulated by Taxol-treatment and the absence of p53 prevents Taxol-induced cell death partially in this system, which suggests that p53 contributes to Taxol-induced apoptosis.
https://static-content.springer.com/image/art%3A10.1007%2Fs10495-012-0772-8/MediaObjects/10495_2012_772_Fig1_HTML.gif
Fig. 1

p53 enhances Taxol-induced apoptosis in HCT116 colon cancer cells. a p53 wild type (p53 +/+) or p53 null (p53−/−) HCT116 cells were treated with Taxol (25 nM) for indicated time periods and subject to western blot analysis with anti-p53 antibody. b 24 h After p53 +/+ or p53−/− HCT116 cells were treated with various doses of Taxol, harvested cells were stained with PI and analyzed by flow cytometry. c p53 +/+ or p53−/− HCT116 cells were incubated for 48 h with various doses of Taxol and the cells survival was measured by MTT assay. Data represent the percentage of relative cell survival compared with the vehicle control, which was calculated by the following formula: (OD of Taxol-treated sample/OD of vehicle-treated control sample) × 100. Data were obtained from the quadruplicate samples and presented as mean ± SEM. *p < 0.01; **p < 0.001 (student’s t test)

Next, we examined whether the upregulated p53 was functioning properly in p53 wild type cells by following the induction of downstream target genes of p53. In accordance with induction kinetics of p53, the expression of target proteins such as MDM2, p21, BAX, and β-isoform of PUMA were all upregulated after 24 h and maintained up to 72 h (Fig. 2). The same pattern of upregulation of both p53 and target proteins was also observed with the cells treated with nocodazole or vincristine, other microtubule inhibitors (Fig. 2; data not shown). This induction pattern was observed, suggesting that this phenomenon is universal for the microtubule inhibitors. Thus, p53 proteins upregulated by microtubule inhibitors are functionally intact, further supporting the idea that p53 would have a functional role in the cells exposed to microtubule inhibitors.
https://static-content.springer.com/image/art%3A10.1007%2Fs10495-012-0772-8/MediaObjects/10495_2012_772_Fig2_HTML.gif
Fig. 2

p53 target proteins are upregulated upon Taxol-treatment in p53 wild type HCT 116 cells. p53 wild type HCT116 cells were treated with Taxol (25 nM) or nocodazole (50 ng/ml) for indicated time periods and subject to western blot analysis with the various antibodies. Anti-β-actin antibody was used to prove an equal loading per each sample. C, Taxol-untreated, the vehicle-treated cells

Enhanced p53 acetylation increases Taxol-induced apoptosis

Since p53 protein upregulation is mostly mediated by protein stabilization by post-translational modifications such as phosphorylation and acetylation [15], we checked the phosphorylations and acetylations of p53 proteins in Taxol-treated HCT116 cells (p53 wild type). Interestingly, although we could not detect a significant increase in phosphorylations of serine 15 and 20 in p53, acetylations of lysine 373 and 382 were dramatically increased 24 h after Taxol-treatment. These effects were also observed in nocodazole-treated cells. Hence, while the role of phosphorylation p53 is not clear in this system, acetylation of p53 correlated well with p53 protein upregulation. This suggests that acetylation may have an important role in p53 stabilization in microtubule inhibitor-induced cell death of HCT116 cells (Fig. 3).
https://static-content.springer.com/image/art%3A10.1007%2Fs10495-012-0772-8/MediaObjects/10495_2012_772_Fig3_HTML.gif
Fig. 3

p53 is acetylated following Taxol treatment in p53 wild type HCT116 cells. p53 wild type HCT116 cells were treated with Taxol (25 nM) or nocodazole (50 ng/ml) for indicated time periods and subject to western blot analysis with the various antibodies. Anti-β-actin antibody was used to prove an equal loading per each sample. C, Taxol-untreated, the vehicle-treated cells

In order to further examine the role of acetylation in pro-apoptotic function of p53 in HCT116 cells, we tried to increase the acetylation by inhibiting deacetylation of p53. Acetylated p53 has been known to be deacetylated by deacetylases like Sirt1 and HDAC1 [2426]. When p53 wild type HCT116 cells were transfected with anti-Sirt1 siRNA, acetylated p53 greatly increased upon Taxol-treatment, which confirmed that Sirt1 is responsible for deactylation of p53 (Fig. 4a). Consistently, the increase of p53 target proteins such as p21, BAX, and β-isoform of PUMA following Taxol-treatment was further enhanced in the presence of anti-Sirt1 siRNA. These effects were not seen in p53 null HCT116 cells. Next, we checked if increased acetylation would enhance Taxol-induced cell death through the apoptosis profiling by flow cytometry. Although Sirt1 knockdown increased sub-G1 fraction to a certain degree in untreated control p53 wild type cells, sub-G1 fraction of anti-Sirt1 siRNA-treated cells markedly increased compared to that of control siRNA-treated cells in the presence of Taxol (Fig. 4b). This enhancement was nearly abolished in p53 null cells. These effects were statistically significant when analyzed out of four independent experiments (Supplemental Fig. 2a). When the apoptosis was examined with another readout, annexin-V positivity on the cell surface, there was a significant increase of early (PIAnnexin-V+) and late apoptotic populations (PI+Annexin-V+) after anti-Sirt1 siRNA transfection in the presence of Taxol in p53 wild type cells. Again, this effect was significantly diminished in p53 null cells, which suggests that most of the effect of anti-Sirt1 siRNA on Taxol-induced apoptosis is mediated through p53 (Fig. 4c). Furthermore, the cleaved fragment of Caspase 3, a downstream indicator of successful apoptosis, was further increased by Sirt1 knockdown in the presence of Taxol (Fig. 4a). Thus, the enhancement of acetylation of p53 by anti-Sirt1 siRNA increased p53-dependent apoptosis in the presence of Taxol.
https://static-content.springer.com/image/art%3A10.1007%2Fs10495-012-0772-8/MediaObjects/10495_2012_772_Fig4_HTML.gif
Fig. 4

Increased p53 acetylation by Sirt1 knockdown enhances Taxol-induced apoptosis. ac p53 wild type (p53 +/+) or p53 null (p53−/−) HCT116 cells tranfected with either anti-Sirt1 siRNA or control siRNA were treated with the vehicle alone (c) or with 3 nM of Taxol (T) for 24 h (b,c) or 48 h (a). Then the harvested cells were subjected to western blot analysis using various antibodies (a), or stained with PI (b) or along with Annexin-V-FITC (c) and then analyzed by flow cytometry (b, c)

We confirmed the above observations by inhibiting another p53 deacetylase, HDAC1, with an HDAC inhibitor, MS275 [27, 28]. When p53 wild type cells were treated with MS275 in the presence of Taxol, the increase in p53 acetylation by Taxol was enhanced (Fig. 5a). MS275 also increased p53 downstream target proteins in the presence of Taxol, just as anti-Sirt1 siRNA. These effects were not observed in p53 null cells. Thus, MS275 functioned efficiently as an acetylation enhancer for p53. Functionally, MS275-treated p53 wild type cells underwent more profound apoptosis in the presence of Taxol as measured with both aptoptosis profiling and Annexin-V staining, whereas p53 null cells did not show such profound effect by MS275 treatment (Fig. 5b, c). These effects were statistically significant (Supplemental Fig. 2b). Again, the cleavage of Caspase-3 and cleavage of PARP, another indicator of apoptosis, were also enhanced by MS275 in the presence of Taxol in p53 wild type cells, which was not evident in p53 null cells (Fig. 5a). Finally, this apoptosis-enhancing effect by MS275 was not merely confined to HCT116 cells, because MS275 also enhanced Taxol-induced cell death in two other cancer cell lines: MCF7, a human breast cancer cell line, and A549, a human lung cancer cell line (Supplemental Fig. 3). Collectively, the enhancement of acetylation of p53 by two different methods clearly demonstrated the functional role of p53 acetylation in Taxol-induced cell death.
https://static-content.springer.com/image/art%3A10.1007%2Fs10495-012-0772-8/MediaObjects/10495_2012_772_Fig5_HTML.gif
Fig. 5

Increased p53 acetylation by an HDAC inhibitor enhances Taxol-induced apoptosis. ac, p53 wild type (p53 +/+) or p53 null (p53−/−) HCT116 cells were treated with the vehicle alone (control) or with MS275 (1 μM) in the presence or absence of 3 nM of Taxol for 24 h (b,c) or 48 h (a). Then the harvested cells were subjected to western blot analysis using various antibodies (a), or stained with PI (b) or along with Annexin-V-FITC (c) and then analyzed by flow cytometry (b, c)

Defective acetylation of p53 diminishes p53-mediated apoptosis in the presence of Taxol

Although we showed that the enhancement of p53 acetylation increased Taxol-induced apoptosis, a more definitive experiment to prove the functional role of p53 acetylation in this process is the loss-of-function experiment, in which we can examine if blocking of p53 acetylation would now inhibit Taxol-induced cell death. In order to inhibit acetylation of p53 in a specific way, we generated an acetylation-defective mutant of p53. In this mutant, eight lysine residues of p53 reported to be acetylated were all replaced with arginines using point mutagenesis to prevent acetylations inside the cell [19]. We then, performed a reconstitution experiment, in which the wild type p53 expression plasmid or the mutant expression plasmid was transfected into p53 null HCT116 cells. Even in the absence of Taxol, re-introduction of the wild type p53 proteins into p53 null cells induced significant cell death as measured with MTT assay. In contrast, the acetylation-defective mutant induced cell death less efficiently than the wild type p53, suggesting that acetylation of p53 is partly responsible for pure p53-mediated cell death, as previously reported [19]. In a situation that Taxol induces cell death, the wild type p53-tranfected cells showed more profound cell death, whereas the mutant p53-transfected cells showed partial inhibition of Taxol-induced cell death (Fig. 6a). Thus, in the presence of Taxol, acetylation of p53 was necessary for full pro-apoptotic effect of p53.
https://static-content.springer.com/image/art%3A10.1007%2Fs10495-012-0772-8/MediaObjects/10495_2012_772_Fig6_HTML.gif
Fig. 6

Effects of reconstitution of the wild type p53 and the acetylation-defective mutant p53 in p53 null cells in the presence of Taxol. a p53 null HCT116 cells tranfected with either the mock vector (mock), or the wild type p53 plasmid (p53 wt), or the acetylation-defective mutant p53 plasmid (p53 mut), were treated with the vehicle alone (control) or 10 nM of Taxol for 48 h. Cell survivals were measured by MTT assay and presented as OD values of MTT assay. Data were obtained from the six independent samples and presented as mean ± SEM. **p < 0.001 (Student’s t test). bc H1299 cells transfected as in a, were treated with the vehicle alone (control) or 10 nM of Taxol for 24 h (b) or 48 h (c). The harvested cells were stained with PI or along with Annexin-V-FITC, and then analyzed by flow cytometry (b). Or, the cells were immunoblotted with the indicated antibodies (c)

Next, in order to see if this phenomenon could be recapitulated in other cancer cells, we adopted another p53 null cancer cell line, H1299, which is derived from human lung cancer. We performed the same reconstitution experiment for H1299 cells and the apoptosis profile was analyzed by flow cytometry (Fig. 6b). Unlike p53 null HCT116 cells, H1299 cells were less vulnerable to the pure p53-mediated cell death as demonstrated in Taxol-untreated control cells. However, in the presence of Taxol, the wild type p53-transfected cells showed significantly increased cell death compared with the empty vector-transfected cells. In contrast, the p53 mutant-transfected cells showed less apoptosis than the wild type p53-transfected cells (Fig. 6b, top and middle). This phenomenon was also demonstrated by Annexin-V staining (Fig. 6b, bottom). Although we could not determine the acetylation status of the re-introduced proteins in p53 null HCT116 cells due to the profound cell death upon the wild type p53 transfection, we were able to analyze the acetylation status of the expressed p53 proteins in H1299 cells because we could obtain enough number of cells for western blot analysis (Fig. 6c). As expected, the re-introduced wild type p53 proteins showed enhanced acetylation upon Taxol-treatment, whereas the mutant p53 proteins did not show any acetylation either in the absence or presence of Taxol. In line with this observation, the expression of the downstream target proteins by the wild type p53 expression was enhanced by Taxol-treatment. Again, the mutant p53 transfection showed less prominent enhancement of the expression of these proteins upon Taxol-treatment. Also, the increased cleavage form of both Caspase-3 and PARP by the wild type p53 transfection was not observed by the mutant p53 transfection. Overall, the specific inhibition of acetylation of p53 in this reconstitution system partially inhibited p53-mediated cell death in the presence of Taxol in two different cancer cell lines, which suggests that acetylation of p53 contributes to the pro-apoptotic function of p53 in Taxol-induced cell death.

Discussion

Although it has been well-demonstrated that microtubule inhibitors induce cell death through mitotic arrest, other mechanisms can operate in MTI-induced cell death. Especially, since p53 is a well-known tumor suppressor and plays a role in regulating cell cycle arrest in various stages of cell cycle, the question whether p53 affects MTI-induced cell death can be easily raised. However, the role of p53 in this process has long been controversial [1013]. Here, we showed that p53 plays a pro-apoptotic role in Taxol-induced cell death for human cancer cells using several cancer cell lines. Although cell death mechanisms independent of p53 clearly operate in Taxol-induced cell death as shown in Fig. 1c, the deletion of p53 provided a moderate protection against Taxol-induced cell death in HCT116 colon cancer cells. In line with this observation, overexpression of the wild type p53 in p53 null HCT116 cells potentiated Taxol-induced cell death (Fig. 6a). This effect was also reproduced in another p53 null cancer cell line, H1299, which showed more prominent apoptosis by Taxol when the wild type p53 plasmid was transfected (Fig. 6b). Thus, it looks clear that p53 acts as an apoptosis enhancer for Taxol-induced cell death at least in some cancer cells.

Then, how can we explain the controversial results on pro- or anti-apoptotic role of p53 in Taxol-induced cell death listed in the literature [1012]. p53−/− mouse embryonic fibroblasts (MEFs) showed enhanced cell death upon Taxol-treatment compared with p53 +/+ MEFs [1012]. Also, human fibroblasts showed similar effect when p53 was functionally inactivated [10]. By contrast, human lung cancer cells showed transfected p53 wild type sensitizes Taxol-induced cell death [29]. Our results with colon cancer cells (HCT116) and lung cancer cells (H1299) are consistent with this observation. Thus, the observed differences may stem from differences in cell type: fibroblasts versus epithelial cells. Otherwise, may be derived from the malignancy of the cells: normal cells versus malignant cells. In support of the latter hypothesis, when transformed with E1A-Ras, p53−/− MEFs now showed reduced cell death upon exposure to Taxol, which is opposite of the phenotype of the untransformed p53−/− MEFs [12]. p53 originally has both anti- and pro-apoptotic property depending on the state of cells [1013]. p53 may reduce Taxol-induced cell death in normal cells by preventing cells from progressing into mitotic phase which is highly vulnerable to Taxol-induced apoptosis [10, 11]. However, when many of the cells have already progressed to mitotic phase, as in cancer cells, p53 may not be able to protect cells from Taxol-induced cell death any longer. Instead, pro-apoptotic arm of p53 may become dominant in response to Taxol to eliminate these hopeless cells now. These possibilities may help understand the valuable, but still unclear, issues of p53 response against various physiological and pathological stimuli, and needs to be studied in detail in the future.

For the role of acetylation of p53 in Taxol-induced cell death, we proved here that p53 acetylation significantly contributes to cell death in the presence of Taxol, employing both gain-of-function (enhancement of acetylation) and loss-of-function (blockade of acetylation) techniques. When acetylation was increased by either anti-Sirt1 siRNA or an HDAC inhibitor, the cells were more prone to cell death induced by Taxol. Conversely, when an acetylation-defective mutant of p53 was expressed in p53 null cells, apoptosis decreased in response to Taxol. However, in the reconstitution experiments, the acetylation-defective mutant transfection did not fully restore cell survival to the level of the mock-transfected cells. This implies that other modifications may have stabilized this mutant p53 or maintained its function partially. Although we observed distinct acetylation of p53 in response to Taxol-treatment, weak induction of p53 phosphorylation was also observed (Fig. 3). Also, phosphorylation of p53 by MTIs treatment was reported in previous studies [20, 21]. Thus, phosphorylation of p53 may contribute to Taxol-induced apoptosis to some degree. Or, other modifications such as methylation and ubiquitinylation may regulate p53 function in the presence of Taxol, although these possibilities have not been tested yet [3033].

Currently, HDAC inhibitors are actively investigated as possible anti-cancer agents. Since an HDAC inhibitor enhances apoptosis in the presence of Taxol in several cancer cells including colon cancer (HCT116), lung cancer (H1299 and A549), and breast cancer cells (MCF7) in this study, combination of HDAC inhibitors and Taxol may be a desirable strategy for enhancing the therapeutic efficacy of these drugs through additive effect. Such additive effects have been reported several times both in preclinical studies and in early clinical trials [26, 34, 35].

In summary, this study shows that p53 contributes to Taxol-induced cell death, and in particular, acetylation of p53 plays an important role in pro-apoptotic function of p53 for several cancer cell lines. Thus, our results will provide a piece of meaningful information for solving the old controversies on the role of p53 in Taxol-induced cell death.

Acknowledgments

We would like to thank Dr. Ja-Eun Kim (Kyunghee University, Seoul, Korea) for kindly sharing her anti-Sirt1 antibody, Dr. Xuan Liu (University of California, Riverside, USA) for providing a mutant p53 plasmid, and Tae Sik Kim for technical support on flow cytometry. We would like to give special thanks to Dr. Sun-Shin Kim, and Dr. Hye-Jin Yu for scientific discussion and Dr. J.S. Ram for critical reading during the manuscript preparation. This work was supported by grants for K.K. from National Cancer Center Korea (NCC-1110080).

Conflict of interest

The authors declare no conflicts of interest.

Supplementary material

10495_2012_772_MOESM1_ESM.tif (29.3 mb)
Supplementary material 1 (TIFF 29986 kb)
10495_2012_772_MOESM2_ESM.tif (209.3 mb)
Supplementary material 2 (TIFF 214362 kb)
10495_2012_772_MOESM3_ESM.tif (29.3 mb)
Supplementary material 3 (TIFF 30046 kb)

Copyright information

© Springer Science+Business Media New York 2012