Actinomycin D (Act D) is a general transcriptional inhibitor that is approved for the treatment of sarcomas, and Wilms, germ cell and trophoblastic tumors. Little is known about the molecular mechanisms that dictate the sensitivity of cancer cells to Act D. In this study, we investigated the effects of Act D on heat shock proteins (HSPs) and the expression and roles of HSP27 in Act D-induced cancer cell apoptosis. We show that Act D upregulates HSP27 and HSP70 expression in cancer cells, whereas it inhibits HSP90 expression. The upregulation of HSP27 by Act D is not attributable to changes in HSP27 transcription or HSP27 synthesis. HSP27 knockdown leads to an increase in Act D-induced caspase 3 and caspase 7 cleavage, and sensitizes rhabdosarcoma cells and breast cancer cells to Act D-induced apoptosis. We conclude that upregulation of HSP27 represents an adaptive response that compromises the anticancer activity of Act D.
terminal deoxynucleotidyl transferase dUTP nick end labeling
Actinomycin D (dactinomycin; Act D) is a general transcriptional inhibitor with anticancer activity, and is approved for the treatment of sarcomas, and Wilms, germ cell and trophoblastic tumors [1, 2]. Act D induces cytotoxicity by intercalating its phenoxazone ring into the minor groove of DNA, thus blocking the binding of RNA polymerase II and subsequently inhibiting RNA synthesis . As a transcriptional inhibitor, Act D reportedly inhibits expression of the prosurvival gene Mcl-1 and acts synergistically with ABT-737, the prototype BH3-mimetic BCL-2 antagonist, to induce small cell lung cancer, non-small cell lung cancer, and pancreatic cancer cell apoptosis [4, 5]. In addition, a previous study demonstrated that Act D could abrogate the heat shock protein (HSP)90 inhibitor 17-allylamino-17-demethoxygeldanamycin (17-AAG)-mediated increases in the levels of all HSPs and enhance 17-AAG-induced cell death in multiple myeloma cell lines . Low doses of Act D also induce ribosomal stress and the sequestration of MDM2 by free ribosomal proteins, leading to the accumulation of p53 . However, Act D induces p53-independent cell death and prolongs survival in chronic lymphocytic leukemia . These studies indicate a potential for Act D as a therapeutic agent for a variety of cancers, either as monotherapy or in combination with other drugs. Although Act D has been used as a cancer chemotherapeutic agent for more than five decades, the mechanisms underlying Act D resistance are largely unknown.
HSPs are known as molecular chaperones that regulate cellular homeostasis and promote cell survival under stress conditions. Mammalian HSPs have been classified into six families according to their molecular size: HSP100, HSP90, HSP70, HSP60, HSP40, and small HSPs (15–30 kDa). HSP90 regulates the stability of a diverse set of critical client proteins, including protein kinases (e.g. epidermal growth factor receptor, insulin-like growth factor receptor, and Akt) and steroid hormone receptors. Elevated expression or activation of some HSP90 client proteins has been implicated in tumor progression and drug resistance. HSP90 inhibitors have shown promising antitumor activities in preclinical studies. Like HSP90, the small HSP HSP27 (also known as HSPB1) has its client proteins, such as histone deacetylase 6, STAT2, and procaspase-3 . HSP27 also prevents cell death through multiple mechanisms. Cell apoptosis can be inhibited by HSP27 through prevention of the formation of the apoptosome and the subsequent activation of caspases [10, 11]. Moreover, HSP27 prevents cell death by interfering with cytochrome c release . Upon oxidative stress, HSP27 maintains glutathione to decrease the accumulation of reactive oxygen species, thereby inhibiting cell death [13, 14]. Finally, HSP27 interacts with Akt to control apoptosis by regulating Akt activation . Whereas hyperthermia reportedly protects Chinese hamster ovary cells against the cytotoxicity of Act D , heat shock induces doxorubicin resistance but not colchicine, 5-fluorouracil, cisplatin, Act D and methotrexate resistance in breast cancer cells . Ectopic expression of HSP27 in Chinese hamster ovary cells increases survival following treatment with daunorubicin, colchicine, vincristine, Act D, hydrogen peroxide, and sodium arsenite . Overexpression of HSP27 in breast, endometrial or gastric cancer has been associated with metastasis, poor prognosis, and resistance to chemotherapy or radiation therapy [19, 20]. Conversely, suppression of HSP27 leads to long-term dormancy in human breast cancer, owing to inhibition of tumor angiogenesis . Moreover, HSP27 silencing can inhibit breast cancer bone metastasis . In addition, HSP27 expression can be induced by anticancer agents, such as 17-AAG and doxorubicin [6, 23]. Blockade of HSP27 may sensitize cancer cells to these agents. Finally, overexpression of HSP27 contributes to androgen-independent progression in prostate cancer .
Here, we show that Act D upregulates HSP27 levels in cancer cells. The upregulation of HSP27 by Act D is not attributable to increased HSP27 transcription or protein synthesis. Instead, Act D prevents the degradation of HSP27. HSP27 knockdown potentiates Act D-induced caspase activation and apoptosis.
Act D decreases the level of HSP90 but increases the levels of HSP27 and HSP70
To detect the effect of Act D on HSP90, HSP70 and HSP27 expression, we treated RD rhabdosarcoma cells with Act D for 24 h, at concentrations ranging from 1 to 8 ng·mL−1, and then performed western blot analysis of HSP90, HSP70 and HSP27 levels. Treatment with Act D led to a decrease in HSP90 expression and an increase in HSP27 and HSP70 expression (Fig. 1A). MDA-MB-231 breast cancer cells and HepG2 hepatoma cells were less sensitive to Act D than RD cells. However, treatment of MDA-MB-231 cells with 2–8 ng·mL−1 Act D for 48 h also resulted in increased HSP27 expression, whereas it did not affect HSP70 and HSP90 expression (Fig. 1B). Treatment of MDA-MB-231 cells with higher doses of Act D (0.1–0.5 μg·mL−1) for 24 h resulted in an increase in HSP70 expression and a decrease in HSP90 expression (Fig. 1B). In addition, treatment of HepG2 cells with higher doses of Act D (0.1–0.5 μg·mL−1) for 24 h resulted in an increase in HSP27 and HSP70 expression and a decrease in HSP90 expression (Fig. 1C). Notably, the upregulation of HSP27 by Act D was dramatic in all three cell lines tested. Act D also increased HSP27 phosphorylation in a dose-dependent manner (Fig. 1). To determine whether HSP27 is induced by Act D at the transcriptional level, total RNA from nontreated or Act D-treated cells was subjected to real-time RT-PCR analysis. Treatment of MDA-MB-231 cells with 0.1 or 0.25 μg·mL−1 Act D had little effect on HSP27 transcription, whereas treatment with 0.5 and 0.75 μg·mL−1 Act D dramatically inhibited HSP27 transcription (Fig. 1D). For HepG2 cells, treatment with 0.5 μg·mL−1 and 0.75 μg·mL−1 Act D dramatically inhibited HSP27 transcription, whereas treatment with 0.25 μg·mL−1 Act D slightly inhibited HSP27 transcription (Fig. 1D). These data indicate that the increase in the level of HSP27 is not attributable to enhancement of HSP27 transcription.
The increase in the level of HSP27 results from the blockade of HSP27 degradation
To determine whether the increase in the level of HSP27 results from an increase in HSP27 protein synthesis, we treated HepG2 cells with cycloheximide (CHX), an effective inhibitor of protein biosynthesis in eukaryotes, with or without Act D. Interestingly, treatment with CHX slightly increased HSP27 levels (Fig. 2A). Combination of CHX with Act D did not abrogate Act D upregulation of HSP27 (Fig. 2B). These data suggest that it is less likely that the upregulation of HSP27 by Act D results from changes in HSP27 protein synthesis.
Another possibility is that treatment with Act D results in changes in the turnover of HSP27. Usually, a practical methodology for monitoring protein stability is treatment with CHX to inhibit new protein synthesis, followed by chasing protein levels over a time course. Because treatment with CHX did not lead to a decrease in the level of HSP27, we could not detect the effect of Act D on HSP27 turnover with this methodology. Instead, we detected the effect of Act D on the level of HSP27 in the absence or presence of the proteasome inhibitor MG132. Both Act D and MG132 induced an increase in the level of HSP27. In the presence of MG132, Act D did not increase the level of HSP27 (Fig. 2C). These data indicate that the increase in the level of HSP27 caused by Act D may result from blockade of HSP27 degradation.
HSP27 knockdown potentiates the inhibition of cell survival induced by Act D
As HSP27 can promote cell survival under stress conditions, we investigated whether downregulation of HSP27 by small interfering RNA (siRNA) could sensitize cancer cells to Act D. In RD cells, treatment with Act D or HSP27 siRNA (siHSP27) slightly inhibited cell viability. HSP27 knockdown significantly enhanced the inhibitory effect of Act D on cell viability (Fig. 3A,B). The efficacy of RNA interference was confirmed by western blotting (Fig. 3C). HSP27 knockdown also significantly enhanced the inhibitory effect of Act D on MDA-MB-231 cell growth (Fig. 3D,E). The efficacy of RNA interference was confirmed by western blotting (Fig. 3F). In addition, a clonogenic assay on MDA-MB-231 cells demonstrated that HSP27 knockdown significantly enhanced the inhibitory effect of Act D on cell survival (Fig. 4). These data suggest that HSP27 knockdown potentiates the inhibition of cell survival by Act D.
HSP27 knockdown potentiates Act D-induced caspase cleavage and apoptosis
HSP27 can inhibit apoptosis by inhibiting caspase activation [18, 19]. To determine whether HSP27 knockdown potentiates Act D-induced apoptosis, the effect of Act D on control siRNA (siCtrl)-transfected or siHSP27-transfected RD cells was examined by Hoechst 33342 staining and terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay. Whereas HSP27 knockdown alone did not induce apoptosis, HSP27 knockdown led to a significant increase in apoptosis in Act D-treated RD cells (Figs 5A,B and 6A,B). The Act D-induced necrosis was not stimulated by HSP27 knockdown (data not shown). To determine whether HSP27 downregulation potentiated the activation of caspase by Act D, the effects of HSP27 knockdown on the cleavage of two executor caspases, caspase 3 and caspase 7, were examined. In Act D-treated RD cells, the cleavage of caspase 3 and caspase 7 was more robust in siHSP27-transfected cells than in siCtrl-transfected cells (Fig. 6C). Similar effects were detected in MDA-MB-231 cells (Figs 7 and 8). These data indicate that HSP27 deletion can potentiate the activation of caspase 3 and caspase 7 by Act D. Additionally, the stimulatory effects of HSP27 knockdown on Act D-induced apoptosis could be abrogated by caspase inhibitor (Figs 6A,B and 8A,B). These data demonstrate that HSP27 knockdown potentiates Act D-induced apoptosis through caspase activation.
Chemotherapy is one of the major treatment for human tumors. However, a major problem in the treatment of most tumors is the high percentage of recurrence and chemoresistance. Thus, it is very important to determine the regulatory events in chemosensitivity or chemoresistance. Increased multidrug resistance efflux pump activity is one of the main mechanisms of tumor chemoresistance. Overexpression of p-glycoprotein, multidrug resistance-associated protein and breast cancer resistance protein, either before or after chemotherapy, may lead to reduced drug accumulation in cancer cells, thereby attenuating cellular toxicity . In addition, acquired or inducible chemoresistance may develop in response to chemotherapy. Acquired resistance may result from drug-induced overexpression of antiapoptotic proteins or activation of prosurvival pathways. Nuclear factor-κB, Akt and ERBB2 activation suppress apoptosis in tumor cells in response to chemotherapy . The adaptive response to chemotherapeutic drugs eventually helps cancer cells to withstand high doses of these drugs.
Exposure of cells to elevated temperatures or other environmental stresses induces the expression of a set of proteins called HSPs that render cells resistant to these stresses [27, 28]. HSPs can protect cells from stress-induced apoptosis. Among them, HSP27 has been shown to protect mammalian cells exposed to a variety of stress stimuli, including heat shock, oxidative stress, and chemotherapeutic agents, probably by inhibiting the mitochondrial pathway of apoptosis [12, 30, 31]. Overexpression of HSP27 in tumor cells leads to resistance to therapeutic drugs. Previous studies have suggested that the expression of HSP27 is associated with poor prognosis in gastric, liver and prostate carcinoma and osteosarcomas . HSP27 has also been implicated in resistance to chemotherapy in breast cancer and leukemia, and is associated with the acquisition of drug-resistant phenotypes [32-34]. HSP27 expression can be induced by the HSP90 inhibitor 17-AAG, the proteasome inhibitor bortezomib, and doxorubicin, a chemotherapeutic drug that creates dsDNA breaks by inhibiting topoisomerase II [6, 23, 35, 36]. Also, HSP27 expression increases after androgen ablation in hormone-refractory prostate cancer . Blockade of HSP27 by HSP27-specific siRNA can sensitize cancer cells to these chemotherapeutic agents [6, 17, 29, 30]. HSP27 correlates with increased survival in response to cytotoxic stimuli; the cytoprotective effects of HSP27 may result from its role as a molecular chaperone or from direct interference with caspase activation, modulation of oxidative stress, and regulation of the cytoskeleton [30, 37].
As the first antibiotic used to treat cancer, Act D can be effectively used to treat gestational trophoblastic neoplasia, rhabdomyosarcoma, Ewing's sarcoma, and malignant hydatidiform mole. Although Act D has been used as a chemotherapeutic agent in the clinic for more than five decades, little is known about the adaptive response of tumor cells to this drug. As a transcription inhibitor, Act D is supposed to inhibit the expression of many genes, including oncogenes. Interestingly, the current study demonstrates that Act D paradoxically upregulates HSP27 expression, although it inhibits HSP27 transcription and HSP90 expression. We have also excluded the possibility that Act D upregulates HSP27 by increasing HSP27 synthesis. Thus, it is most likely that Act D prevents HSP27 degradation. Indeed, a recent study has demonstrated that HSP27 can be degraded by SMURF2, an E3 ubiquitin ligase . Our results contrast with those of another study, which demonstrated that Act D inhibits HSP70 and HSP27 expression in multiple myeloma cells , whereas HSP90 expression was inhibited by Act D in both studies. It is unknown whether Act D inhibits HSP27 expression in myeloma cells but not in other cells. Given that HSP90 is an important chaperone in stabilizing oncoproteins, the inhibition of HSP90 expression by Act D may contribute, at least in part, to the inhibitory effects of Act D on cancer cells. HSP90 inhibitors can induce HSP27 expression as a result of increased heat shock factor transcriptional activity. However, it is less likely that Act D induces HSP27 expression by inhibiting HSP90, because low doses of Act D can induce HSP27 expression even though HSP90 expression is not inhibited in MDA-MB-231 cells.
Given that HSP27 is an antiapoptotic protein, upregulation of HSP27 by Act D can be viewed as an adaptive response. The current study demonstrates that downregulation of HSP27 can potentiate Act D-induced apoptosis by caspase activation. Binding of HSP27 to caspase 3 prodomain inhibits caspase 3 proteolytic cleavage . Thus, HSP27 is a molecular target to enhance Act D sensitivity. We can speculate that small molecule inhibitors of HSP27 may be of use in treating cancer or overcoming chemoresistance. However, the small molecule inhibitors of HSP27 are still not available, and warrant further development.
Act D, CHX, Hoechst 33342 and the caspase inhibitor z-VAD-FMK were purchased from the Beyotime Institute of Biotechnology (Jiangsu, China). MG132 was purchased from Sigma (St. Louis, MO, USA). Antibody against HSP27 was purchased from Signalway Antibody (Maryland, MD, USA). Antibody against HSP70 was purchased from Epitomics (Burlingame, CA, USA). Antibody against HSP90 was purchased from Beyotime Institute of Biotechnology. Antibodies against caspase 3 and caspase 7 were from Cell Signaling Technology (Boston, MA, USA).
RD, MDA-MB-231 and HepG2 cells were purchased from Cell Lines Bank, Chinese Academy of Science (Shanghai, China). Cells were grown in tissue culture flasks at 37 °C in a humidified atmosphere of 5% CO2, and were maintained as monolayer cultures in DMEM supplemented with 10% fetal bovine serum, 100 U·mL−1 penicillin, and 100 μg·mL−1 streptomycin.
Total RNA was extracted with TRIzol reagent (Invitrogen, Carlsbad, CA, USA). First-strand cDNAs were synthesized with Moloney murine leukemia virus reverse transcriptase and oligo(dT) primers. HSP27 was amplified by real-time PCR with SYBR Green PCR amplification mix (total volume, 25 μL) and 300 nm primers. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was also amplified as a reference gene. The primer sequences for human HSP27 were 5′-CCAGAGCAGAGTCAGCCAGCAT-3′ (sense) and 5′-CGAAGGTGACTGGGATGGTGA-3′ (antisense). Relative quantification with the comparative threshold cycle (Ct) was performed with the Ct method. The amount of HSP27 normalized to the endogenous reference gene (GAPDH) is given by 2−ΔCt, where ΔCt is Ct (HSP27) – Ct(GAPDH).
The target sequence used for knockdown of HSP27 was 5′-ACGGUCAAGACCAAGGAUG dTdT-3′. siHSP27 and siCtrl were purchased from Ribobio (Guangzhou, China). The double-stranded siRNA duplex was dissolved in diethylpyrocarbonate-treated water. For transfection, 5000 cells were plated into 24-well plates and incubated for 2 days. LipofecTAMINE 2000 reagent (Invitrogen) was diluted in 50 μL of Opti-MEM I reduced serum medium, and incubated at room temperature for 5 min. In addition, siRNA duplex was diluted in 50 μL of Opti-MEM I reduced serum medium, and mixed with the prediluted LipofecTAMINE 2000. The mixture was incubated at room temperature for 20 min, and 50 nm siRNA was added to each well and incubated at 37 °C.
Cells were washed twice with NaCl/Pi and harvested with cold lysis buffer containing protease inhibitors. Cell lysates were collected from culture plates. Protein concentrations were determined with the bicinchoninic acid protein assay (Pierce Biotechnology, Rockford, IL, USA). Forty micrograms of total protein was boiled in 2× loading buffer (0.1 m Tris/HCl, pH 6.8, 4% SDS, 0.2% bromophenyl blue, 20% glycerol) for 10 min, and then loaded onto Tris/HCl–polyacrylamide gels and transferred electrophoretically to poly(vinylidene difluoride) membranes. Membranes were incubated with primary antibodies and appropriate horseradish peroxidase-linked secondary antibodies. Membranes were additionally probed with an antibody against actin to normalize loading of protein among samples. The secondary antibodies were detected with chemiluminescent agents (Pierce Biotechnology).
The cells were digested with trypsin/EDTA solution, and then stained with trypan blue. The number of living cells without trypan blue incorporation was counted with a hemocytometer.
Cell apoptosis was detected by Hoechst 33342 staining and TUNEL assay. For Hoechst 33342 staining, replicate cultures of cells were plated in tissue culture plates. On the next day, the cells were transfected with siCtrl or siHSP27. Twenty-four hours later, the cells were treated with or without Act D for 24 hours, and incubated with Hoechst 33342 solution at 20 °C for 10 min; apoptotic cells were then examined under a fluorescence microscope. Strong fluorescence and condensed or fragmented nuclei were observed in the nuclei of apoptotic cells, whereas weak fluorescence was observed in live cells. Quantification of apoptotic cells was performed by taking the images in random fields and counting at least 300 cells in four random fields in each well.
For TUNEL assay, replicate cultures of cells were plated in tissue culture plates. On the next day, the cells were transfected with siRNA. Twenty-four hours later, the cells were treated with or without Act D for 24 h, and incubated with TUNEL solution and 4′,6-diamidino-2-phenylindole (DAPI) solution at 37 °C for 60 min without light; apoptotic cells were then examined under a fluorescence microscope. Green fluorescence was observed in the nuclei of apoptotic cells. All cells showed blue fluorescence for DAPI. Quantification of all of the cells and apoptotic cells in the same fields was performed by taking the images in random fields and counting at least 300 cells in four random fields in each well.
The cells were seeded in six-well plates at 1500 cells per well. On the next day, the cells were transfected with siCtrl or siHSP27. Forty-eight hours later, the cells was treated with or without Act D. Six hours later, the cells were grown in the absence of drug, and allowed to form colonies for 14 days. The colonies were stained with 1% methylene blue and then counted. The experiments were performed in triplicate, and each experiment was reproduced.
One-way analysis of variance with least significant difference post hoc tests (spss 13.0 for Windows) was used to test for the differences in cell growth and apoptosis. All statistical tests were two-tailed, and the difference was considered to be statistically significant when P < 0.05.
We thank the National Natural Science Foundation of China (Nos. 30973435 and 30900554) for their support.