Enniatin Exerts p53-Dependent Cytostatic and p53-Independent

Response of experimental malignant melanoma models to the pan-Aurora kinase inhibitor VE-465. Christine Pirker , Daniela Lötsch , Sabine Spiegl-Krein...
0 downloads 0 Views 604KB Size
Download PDF
Chem. Res. Toxicol. 2007, 20, 465-473

465

Enniatin Exerts p53-Dependent Cytostatic and p53-Independent Cytotoxic Activities against Human Cancer Cells Rita Dornetshuber,† Petra Heffeter,‡ Majid-Reza Kamyar,† Thomas Peterbauer,† Walter Berger,*,‡ and Rosa Lemmens-Gruber† Department of Pharmacology and Toxicology, UniVersity of Vienna, Althanstrasse 14, and Institute of Cancer Research, Department of Medicine I, Medical UniVersity of Vienna, Borschkegasse 8a, 1090 Vienna, Austria ReceiVed October 2, 2006

Worldwide, multiple Fusarium mycotoxins occur as contaminants of cereals with important impacts on human and animal health. The aim of this study was to investigate the effects of the widespread Fusarium secondary metabolite enniatin (ENN), a cyclic hexadepsipeptide, on human cell growth and survival. While short-term exposure (up to 8 h) to ENN at nanomolar concentrations slightly but significantly stimulated cell proliferation, it showed profound apoptosis-inducing effects especially against various human cancer cell types at low micromolar concentrations (already after 24 h of treatment). Several cellular changes indicative for programmed cell death such as cell shrinkage, chromatin condensation, DNA fragmentation, and apoptotic body formation were observed. Correspondingly, the cleavage of poly(ADP-ribosyl)polymerase and the activation of multiple caspases accompanied a distinct loss of mitochondrial membrane potential. To investigate the impact of apoptosis- and cell cycle-regulating proteins on ENN activity, HCT116 cells with homozygously disrupted p53, p21, or bax genes were analyzed. In vitality assays, no significant influences of these proteins on the anticancer activity of ENN were detectable. In contrast, 3H-thymidine incorporation revealed a significantly more efficient block of DNA synthesis in p53 wild-type as compared to knock-out cells. Accordingly, fluorescence-activated cell sorting analysis demonstrated a stronger ENN-induced cell cycle arrest in the G0/G1 phase. Profound ENN-mediated induction of p53 and the p53-downstream cell cycle inhibitor p21 were detectable in p53 wild-type cells by Western blotting. P53-independent p21 induction was also detectable at higher ENN concentrations in p53 (-/-) cells. In contrast, bax activation by ENN was independent of the cellular p53 status. In summary, our results suggest that short-term exposure to very low ENN concentrations, for example, via food intake, might have tumor-promoting functions based on growth stimulation. In contrast, elevated ENN concentrations exert profound p53-dependent cytostatic and p53-independent cytotoxic activities especially against human cancer cells, suggesting a potential quality of ENN as an anticancer drug. Introduction Fusarium strains belong to the most frequent cereal contaminants. Because of their ability to grow on corn during both field growth and storage, mycotoxin contamination of food and feed is a serious problem for human and animal health (1). While the toxicological impacts of several Fusarium mycotoxins including deoxynivalenol, zearalenone, and fumonisins have been extensively investigated (2), data about the secondary metabolite enniatin (ENN),1 produced by various strains of Fusarium, are limited. However, studies of conventional and organic grain-based products purchased from Italian and Finnish markets showed that ENN B and B1 as well as deoxynivalenol are the most predominant mycotoxins and were present in 97% of the samples analyzed (3). This implies that ENNs are more * To whom correspondence should be addressed. Tel: +43-1-427765173. Fax: +43-1-4277-65169. E-mail: [email protected]. † University of Vienna. ‡ Medical University of Vienna. 1 Abbreviations: BEA, beauvericin; DAPI, 4′,6-diamidino-2-phenylindole; DMSO, dimethyl sulfoxide; ENN, enniatin; FACS, fluorescenceactivated cell sorting; FITC, fluorescein isothiocyanate; JC-1, 5,5′,6,6′tetrachloro-1,1′,3,3′-tetraethyl-benzimidazolylcarbocyanine iodide; LDH, lactate dehydrogenase; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; PARP, poly(ADP-ribosyl)polymerase; PBS, phosphatebuffered saline; PI, probidium iodide; Rh123, rhodamine 123.

widely distributed than anticipated. Thus, their impacts on human cells and tissues urgently need to be investigated. Structurally, ENN belongs to the group of cyclic hexadepsipeptides like beauvericin (BEA) or destruxins (4, 5). It is reported to have antibiotic and insecticidal activity, and it is a well-known inhibitor of acyl-CoA:cholesterol acyl transferase (6). Besides its antibiotic activity, it inhibits enzymes like acyl-CoA: cholesterol acyl transferase (ACAT) (6) and 3′,5′-cyclo-nucleotide phosphodiesterase and binds to calmodulin (7). The ionophoric properties of ENN are based on the ability to incorporate into the cell membrane and to form cation selective pores with a high affinity for K+, Mg2+, Ca2+, and Na+ (8). Nevertheless, data about cellular activities are currently rare as compared to other mycotoxins like the structurally related BEA, which is also produced by the genus Fusarium. BEA has been reported to induce cell death in several cancer cell lines through apoptosis via the mitochondrial pathway resulting in an increase of caspase-3 activity and formation of apoptotic bodies (9). Experiments in human non-small cell lung cancer A549 cells suggested that among other factors also pro- and anti-apoptotic bcl-2 family proteins participate in the regulation of BEAinduced apoptotis (10). Very recently, one study suggested cytotoxic activity of ENN in the low micromolar range against human fibroblast-like and Hep G2 cells (11).

10.1021/tx600259t CCC: $37.00 © 2007 American Chemical Society Published on Web 02/28/2007

466 Chem. Res. Toxicol., Vol. 20, No. 3, 2007

Because of the widespread contamination of human and animal food with ENN, this study aimed to clarify whether ENN has an impact on human cell growth and survival and to analyze the underlying molecular mechanisms. Moreover, it was investigated whether ENN might in principle have a potential to be used as an anticancer compound.

Materials and Methods Drugs. All compounds were supplied by Sigma-Aldrich GmbH (Vienna, Austria). ENN was a mixture of 3% A, 20% A1, 19% B, and 54% B1 homologues (ca. 97% purity). ENNs are poorly soluble in H2O; therefore, stock solutions were frequently prepared in dimethyl sulfoxide (DMSO) and stored at 4 °C. Cell Culture. Embryonic human fibroblasts WI-38 [from American Type Culture Collection (ATCC), Manassas, VA] and the human umbilical vein endothelial cells (HUVEC, donated by T. Mohr, Medical University Vienna) as well as the following human cancer cell lines were used in this study: the glioblastoma cell lines GBL1, GBL2, GBL3, GBL4, MGC, U373, and T98-G (12); the melanoma cell lines VM-8, VM-18, VM-22, VM-33, and VM-25 (13); the osteosarcoma cell lines US-2, OS-9, OS-10, and SAOS (14); the human lung cancer cell lines A549, A427, GLC4, and VL-8 (15); the epidermal carcinoma-derived cell line KB3-1 (generously donated by Dr. Shen, Bethesda, MD) (16); the promyelocytic leukemia cell line HL-60 (by Dr. Center, Kansas State University, KS) (17); and the breast adenocarcinoma cell line MDA-MB-231 (by Dr. Ross, University of Maryland, Greenbaum Cancer Centre, United States) (18); the colon carcinoma cell lines SW 480 and CaCo-2 cells were obtained from ATCC. The colon carcinoma cell model HCT116 and respective sublines with deleted p53, p21, or bax genes were generously donated by Dr. Vogelstein, John Hopkins University (Baltimore, MD) (19-21). Bcl-2-negative non-small cell lung cancer cells A549 were transfected with a bcl-2 pBabe/Puro expression construct or the respective vector control by electroporation and clones selected in puromycin. Expression was checked, and two respective cell clones (A549 /bcl-2 and bcl2/vc) were chosen for analysis. HCT116 cell lines and SAOS were grown in McCoy’s culture medium; all osteosarcoma cell lines were cultivated in IMDM; SW480, MGC, and Caco-2 cells were grown in MEME; WI-38 and T98-G cells were grown in DMEM; U373 cells were cultivated in MNP; and HUVEC were cultivated in M199 (Promega, Mannheim, Germany) supplemented with 50 U/mL heparin. All other cell lines were cultivated in RPMI 1640. All culture media were supplemented with 10% fetal calf serum (PAA, Linz, Austria). Cultures were periodically checked for Mycoplasma contamination. Cytotoxicity Assay. Human cancer cell lines were seeded (2 × 104 cells/mL) in 100 µL per well in 96 well plates and allowed to recover for 24 h. For cytotoxicity studies, concentrations of ENN ranging from 0.1 up to 10 µM were added in another 100 µL of growth medium, and cells were exposed for 72 h. The cytotoxic effect of ENN was determined by an 3-(4,5-dimethylthiazol-2-yl)2,5-diphenyltetrazolium bromide (MTT)-based vitality assay (EZ4U, Biomedica, Vienna, Austria). In brief, 100 µL of the supernatant was removed and replaced with 100 µL of EZ4U assay solution. After 1-2 h of incubation, the absorbance was measured by a microplate reader, set at 450 nm with 620 nm as reference to reduce unspecific background values. All experiments were performed at least twice in triplicate. DAPI Staining. KB-3-1 cells (1 × 105/well) were plated in six well plates and after 24 h of recovery were treated for another 24 h with 5 and 10 µM ENN. Afterward, cells were harvested, cytospins were prepared, and apoptosis was evaluated by staining with 4′,6-diamidino-2-phenylindole (DAPI) containing antifade solution (Vector Laboratories, Inc., Burlingame, CA). The nuclear morphology of cells was examined with a Leica DMRXA fluorescence microscope (Leica Mikroskopie and System, Wetzlar, Germany) equipped with appropriate epifluorescence filters and a COHU charge-coupled device camera. The rate of apoptosis was

Dornetshuber et al. determined as the percentage of apoptotic nuclei whereby at least 500 nuclei/experimental group were analyzed. Fluorescein Isothiocyanate (FITC)-Phalloidin Staining. After growing KB-3-1 and A549 cells in culture medium for 1 day on chamber slides (BD, Falcon Culture Slides, Bedford, United States), several ENN concentrations were added for another 24 h. ENNmediated changes of the filamentous actin structure were detected using FITC-phalloidin staining (Sigma Chemical Company, St. Louis, MO). DNA Fragmentation. HL-60 cells (1 × 106) were incubated for 24 h with 5 and 10 µM ENN. HL-60 cells treated with 1 µM staurosporin for 24 h served as positive control for apoptosis induction. All cells were centrifuged at 1100 rpm for 5 min, washed with phosphate-buffered saline (PBS), and centrifuged. The resulting pellet was resuspended in DNA lysis buffer (50 mM Tris, pH 8, 10 mM EDTA, and 0.5% sodium lauryl sarcosine), followed by the addition of 20 U of RNase solution (Sigma) and incubated at 37 °C for 1 h. Finally, 150 µg of proteinase K (Sigma) was added and incubated overnight. Isolated DNA was analyzed by electrophoresis (80 V, 1 h) in 2% agarose gels containing ethidium bromide (Sigma). Mitochondrial Membrane Potential Detection. KB-3-1 cells (5 × 106) were exposed to 1, 2.5, and 5 µM ENN for 48 h. Cell staining was performed as follows: Cell suspensions were incubated with 10 µL/mL of the fluorescent dye 5,5′,6,6′-tetrachloro-1,1′,3,3′tetraethyl-benzimidazolylcarbocyanine iodide (JC-1; Mitochondrial Membrane Potential detection Kit; Stratagene, La Jolla, CA) in medium for 10 min at 37 °C. At the end of the incubation period, cells were washed twice with cold PBS, resuspended in PBS, and analyzed by flow cytometry. In addition, mitochondrial membrane potentials were assessed by following subcellular localization of the mitochondrial dye rhodamine 123 (Rh123). KB-3-1 and CaCo-2 cells (data not shown) were loaded with 10 µg/mL Rh123 (Molecular Probes, Leiden, The Netherlands) for 20 min. As a cationic dye, Rh123 was retained in active mitochondria by the negative potential of the inner membrane. During depolarization the dye was released into the cytosol. Rh123 localization was analyzed using a confocal laser scanning microscope (Zeiss LSM 510) equipped with a 63×, 1.4 NA, oil immersion objective. The dye was excited with the 488 nm laser line of an argon laser. A 505-550 nm band-pass filter was used to collect the emitted fluorescence. ENN was locally applied on a cell via a pipet at concentrations of 1 and 10 µM. 3H-Thymidine Incorporation. KB-3-1 cells (4 × 104 per well) were seeded into a 96 well plate. After 24 h of recovery, cells were exposed for another 24 h to 0.5-10 µM ENN. After the exposure, the culture medium was replaced for 4 h by a 2 nM 3H-thymidine solution (diluted in full culture medium; radioactivity, 25 Ci/mM) and cells were incubated at 37 °C. Afterward, cells were washed three times with PBS and cell lysates were prepared. Radioactivity was determined as described (22). Experiments were carried out in triplicate, and results were expressed as counts per minute (cpm). Cell Cycle Analysis. KB-3-1 (5 × 105 cells/well) were plated into six well plates and allowed to adhere for 24 h. Afterwards, they were incubated for another 24 h with 2.5, 5, and 10 µM ENN. To analyze cell cycle progression, cells were collected, washed with PBS, recovered in 0.9% NaCl solution, transferred in drops into 70% alcohol, and stored at 4 °C. For analysis, cells were incubated in PBS with RNAse (10 µg/mL) for 30 min at 37 °C, treated with 5 µL of probidium iodide (PI) for 30 min at 4 °C, and analyzed by flow cytometry using fluorescence-activated cell sorting (FACS) Calibur (Becton Dickinson, Palo Alto, CA). Cell Quest Pro software (Becton Dickinson and Co., New York) was used to analyze the resulting DNA histograms. Western Blot Analyses. After 24 h of drug exposure, proteins were isolated, resolved by SDS/PAGE, and transferred onto a polyvinylidene difluoride membrane for Western blotting as described (23). Antibodies used were as follows: p53 monoclonal mouse DO-1 (Neomarkers, CA); p21Waf1 polyclonal rabbit C-19 (Santa Cruz Biotechnology, CA); bcl-2 polyclonal rabbit ∆C21 (Santa Cruz Biotechnology); Apoptosis Sampler kit: poly(ADP-

Apoptosis Induction by Enniatins ribosyl)polymerase (PARP), caspase 7, cleaved caspase 7, caspase 9, and cleaved caspase 9 (Cell Signalling Technology, Beverly, MA) polyclonal rabbit; Pro-Apoptosis Bcl-2 Family Antibody Sampler kit: bax, bim, and puma polyclonal rabbit (Cell Signalling Technology); bcl-xL polyclonal rabbit (Oncogene, Cambridge, MA); β-actin monoclonal mouse AC-15 (Sigma); cyclin A polyclonal rabbit sc-751; cyclin B1 monoclonal mouse sc-245; and cyclin D1 polyclonal rabbit sc-246 (all from Santa Cruz Biotechnology). All primary antibodies were diluted 1:1000, except anti-cyclin A and anti-cyclin D1, which were diluted 1:200. Second, horseradish peroxidase-labeled antibodies from Santa Cruz Biotechnology were used at working dilutions of 1:10000. Detection of Caspase-3 and -7 Activities. Activities of caspase-3 and -7 were measured using the Caspase-Glo 3/7 Assay (Promega) according to the manufacturer’s instruction. In brief, KB-3-1 cells were treated with 10 µM ENN for 18 h or left untreated. CaspaseGlo 3/7 reagent was added directly to the cells in 96 well plates and incubated for 3 h before recording the luminescence. As a positive control, staurosporin at a concentration of 1 µM was used.

Results Effects of ENN on Viability and Proliferation of Human Tumor Cell Models. To investigate an impact of ENN on growth characteristics of human normal and cancer cell lines, concentration-response curves (0.1 and 10 µM) were established using a 72 h exposure time. Normal fibroblasts (WI-38) and endothelial cells (HUVEC) were compared to diverse tumor cell lines of different tissue origin including carcinomas, sarcomas, leukemias, and tumors of neuronal origin (Table 1). ENN exerted a profound cytostatic/cytotoxic effect in all cell lines analyzed except normal fibroblasts WI-38 and the glioblastoma cell line T98G where even the IC75 value was not reached at 10 µM ENN. HUVEC, the second investigated normal cell type, was significantly more sensitive as compared to WI-38 but still relatively resistant as compared to the majority of malignant cells. While the IC50 for most tumor cell models was 10 >10

IC25 (µM) meana

(SD

>10

4.84

endothelial cells 0.322 7.89

0.214

>10

VM-8 VM-18 VM-22 VM-33 VM-25

2.57 1.85 1.29 2.63 1.91

melanoma 0.94 3.19 0.19 2.67 0.14 1.75 0.12 9.65 0.15 2.72

0.85 0.08 0.15 0.13 0.11

4.11 >10 2.34 >10 >10

GBL1 GBL2 GBL3 GBL4 MGC U373 T98-G

1.87 1.81 2.02 1.89 1.99 4.2 >10

glioblastoma 0.33 2.65 0.06 2.29 0.12 2.55 0.29 2.33 0.21 3.04 0.134 4.88 >10

0.3 0.05 0.14 0.3 0.58 0.09

>10 3 >10 >10 >10 5.9 >10

U2-OS OS-9 OS-10 SAOS

1.27 1.29 1.23 1.73

osteosarcoma 0.32 1.77 0.14 3.55 0.832. 2.10 0.12 2.13

0.24 0.77 0.15 0.07

2.25 6.64 2.43 2.59

0.18 0.61 0.05 0.05

A549 A427 VL-8 GLC-4

3.42 1.25 2.79 2.09

human lung cancer 1.05 4.08 0.14 1.61 0.22 >10 1.26 2.4

1.04 0.14

1.67 0.11

1.53

6.2 2.05 >10 2.67

1.64

KB-3-1 MDA-MB-231 SW480 CaCo-2

1.63 1.02 2.51 1.38

diverse carcinoma 0.15 1.95 0.47 1.45 0.36 4.00 0.07 1.99

0.12 0.49 1.12 0.09

2.26 2.05 9.04 2.63

0.13 0.51 0.39 0.21

HL60

1.17

leukemia 0.25 1.74

0.2

2.53

0.06

HUVEC

0.5 0.12

0.08

0.31

a

IC75, IC50, and IC25 were calculated from whole dose-response curves. Values given are means ( SD from at least two independent experiments performed in triplicate.

the proportion of nuclei demonstrating apoptotic features increased from 3% in the control to 34% at treatment with 5 µM and to 82.5% with 10 µM ENN for 24 h (Figure 2B). Additionally, changes in the microfilament system of KB-3-1 and A549 cells during treatment with ENN were analyzed by FITC-phalloidin staining. A549 cells are shown representatively in Figure 2C, as in this cell model ENN-induced changes can be better visualized microscopically due to the outspread cell shape and the distinct microfilamental system. While at 1 µM ENN stress fibers were distinctly enhanced, higher concentrations led to progressive loss of cell-cell contacts, accumulation of microfilaments at the cell borders, and cell shrinkage. DNA Fragmentation by ENN Treatment. One biochemical hallmark of apoptosis is degradation of DNA by endogenous DNases, which cut the internucleosomal regions into doublestranded DNA fragments of 180-200 base pairs (24). To further verify the apoptotic response of cells to ENN treatment, formation of nucleosomal DNA fragments was examined by agarose gel electrophoresis. For these experiments, the wellestablished HL-60 leukemia cell model was used and exposure to 1 µM staurosporin served as a positive control for apoptosis induction (Figure 2D). Treatment of HL-60 cells for 24 h induced nuclear DNA fragmentation discernible by a dosedependent formation of a DNA ladder. Apoptosis Induction by ENN Via the Mitochondrial Pathway. One molecular mechanism to induce programmed cell

468 Chem. Res. Toxicol., Vol. 20, No. 3, 2007

Figure 1. Concentration and time dependency of the impact of ENN on cell viability. KB-3-1 cells were treated with the indicated higher (A) and lower (B) concentrations of ENN. After the respective treatment durations, the drug solution was replaced by fresh culture medium. Vitality was measured by an MTT-based survival assay 72 h after first drug contact.

death is initiated by the so-called mitochondrial pathway based on a depolarization of the outer mitochondrial membrane followed by cytochrome c release into the cytoplasm, formation of the apoptosome, and activation of procaspase 9 (25). The lipophilic cation fluorescent dye JC-1 accumulates in the mitochondria (26) and can be used to signal the loss of mitochondrial membrane potential. In intact cells, the dye accumulates in mitochondria and exhibits red fluorescence, whereas in cells with collapsed mitochondrial membrane potential JC-1 remains in its monomeric, green-fluoresecent form in the cytoplasm. FACS analyses of KB-3-1 cells demonstrated that ENN treatment induced a concentrationdependent increase in mitochondrial membrane depolarization (Figure 3A). The number of cells with collapsed mitochondrial membrane potential (∆Ψmt) increased dose-dependently up to almost 30% at 5 µM ENN correlating well with the amount of cells with fragmented nuclei detected at this ENN concentration (compare Figure 2B). These results were also confirmed by a second method for investigating the mitochondrial membrane integrity, namely, the analysis of the subcellular localization of the mitochondrial fluorescent dye Rh123 by confocal laser scanning microscopy. About 1.5 min after ENN application, a gradual shift of Rh123-mediated fluorescence signals from mitochondria to the cytosol and nucleus was observed (Figure 3B), indicating a release of the dye due to mitochondrial membrane depolarisation. The rapid loss of mitochondrial membrane potentials occurs at comparable ENN concentrations as the ones leading to apoptotic cell death after a longer incubation period. To further investigate ENN-mediated apoptosis, the caspasemediated cleavage of PARP as well as caspases-9 and -7 was analyzed by Western blotting (Figure 3C). Cleavages of PARP

Dornetshuber et al.

Figure 2. Morphological changes and DNA cleavage after ENN treatment. (A) Induction of apoptosis was determined in KB-3-1 cells after treatment with 10 µM ENN for 24 h. Nuclei were stained with DAPI in untreated controls and drug-exposed cells. (B) DAPI-stained nuclei were classified (in interphase, mitotic, and apoptotic) and counted, and means/SD were calculated. At least 500 nuclei from three independent experiments were counted. (C) KB-3-1 cells were plated on chamber slides and treated with the indicated ENN concentrations. DNA and microfilaments were visualized by DAPI and FITC phalloidin staining, respectively. Photomicrographs were taken using 40× oil objectives. (D) Genomic DNA was prepared from HL-60 cells treated with 5 and 10 µM ENN as indicated for 24 h. As negative and positive controls, DNA was prepared from untreated cells (control) and cells treated with 1 µM staurosporin for 24 h, respectively.

and both caspases were detectable and increased in a concentration-dependent manner paralleled by a loss of the respective uncut molecules. Additionally, activation of caspase-3 and -7 activities was observed in KB-3-1 cells treated with 10 µM ENN, comparable to levels achieved with 1 µM staurosporin used as positive control (Figure 3D). Taken together, these results point out that ENN-treated cells undergo apoptosis activated via the mitochondrial pathway. Furthermore, no signs of necrosis either determined microscopically or measured by release of lactate dehydrogenase (LDH) were detectable after treatment with ENN for 24 h (data not shown). With regard to the impact of ENN treatment on the apoptosis-regulating bcl-2 protein family (27), the expression levels of the pro-apoptotic members bax, bim, and puma as well as the anti-apoptotic proteins bcl-2 and bcl-xL were investigated by Western blots (Figure 3C). Concerning the pro-apoptotic molecules, an increased expression of bax at lower ENN concentrations (0.52.5 µM) was observed after 24 h of ENN treatment, in contrast to puma and bim, which remained widely unchanged. In parallel, the levels of the apoptosis inhibitor bcl-xL were distinctly and dose-dependently downregulated while the impact on bcl-2 levels was comparably weak.

Apoptosis Induction by Enniatins

Chem. Res. Toxicol., Vol. 20, No. 3, 2007 469

Figure 3. Apoptosis induction by ENN via the mitochondrial pathway. (A) Loss of mitochondrial membrane potential was measured by JC-1 staining after 48 h of ENN treatment. Percentages of apoptotic (green fluorescent, FL-1) cells, located in the right lower quarter, are indicated at the left bottom. (B) KB-3-1 cells, loaded with 10 µg/mL Rh123, were exposed to 10 µM ENN. Photomicrographs shown were taken after 2 min. Bar, 10 µm (Rh123 accumulation in mitochondria pointed by arrows). (C) Caspase-induced cleavage of PARP, caspase-9, and -7 as well as influences on the expression levels of the indicated bcl-2 family members in KB-3-1 cells after 24 h of treatment with ENN were determined via Western blotting. β-Actin served as the loading control. (D) Caspase-3 and -7 activities in ENN-treated KB-3-1 cells were measured using the Caspase-Glo 3/7 Assay (Promega). Staurosporin (1 µM) was used as the positive control.

Effects of ENN on DNA Synthesis and Cell Cycle Distribution. To clarify the effects of ENN treatment on DNA synthesis of KB-3-1 cells, 3H-thymidine incorporation assays were used (Figure 4A). ENN treatment for 24 h significantly reduced DNA synthesis at concentrations g2.5 µM. At lower doses, an increase of DNA replication was observed, correlating with an enhanced number of viable cells in these experimental groups as detected by MTT-based assays (compare Figure 1B). To investigate whether ENN has an impact on cell cycle distribution, flow cytometric analyses were performed and histograms were gated on live cells (Figure 4B). ENN treatment in KB-3-1 cells for 24 h led to a shift of cell cycle distribution, getting visible as increased proportion of cells in the S phase at lower ENN concentrations (2.5 µM) and the G2/M phase at higher dose levels (5 and 10 µM), while cells in the G0/G1 phase were diminished. To further characterize these alterations in cell cycle progression, changes in the cyclin expression patterns were analyzed by Western blotting. Figure 4C shows that ENN induced a continuous decrease in cyclin D1 in a dose-dependent manner correlating with a loss of cells in G0/G1. While cyclin A was upregulated at lower but reduced at higher cytotoxic concentrations, cyclin B1 was only slightly reduced at low ENN

doses but levels remained relatively stable up to 5 µM. At 10 µM ENN, levels of both cyclin B1 and D1 were strongly diminished. Impact of Apoptosis- and Cell Cycle-Regulating Proteins p53, p21, Bax, and Bcl-2 on ENN-Induced Cytotoxicity. As a next step, we investigated the mechanisms underlying ENNinduced cell death with a focus on apoptosis- and cell cycleregulating proteins. For this purpose, HCT116 cells with disrupted p53, p21, or bax genes by targeted homologous recombination were used (19-21). Moreover, intrinsically, bcl2-negative A549 cells stably overexpressing bcl-2 by gene transfection were compared to vector control cells (Figure 5A). In vitality assays, no significant influences of these proteins on cell survival were detected, resulting at a 72 h drug exposure for all HCT116 and A549 subclones in IC50 values at the low micromolar range (Table 2). In contrast, 3H-thymidine incorporation revealed a moderately but significantly (two-way ANOVA, p > 0.001) more efficient block of DNA synthesis by ENN in p53 wild-type as compared to knock-out cells (Figure 5B). Differing to KB-3-1 cells (compare Figure 4), ENN-treated HCT116 cells tended to accumulate in the G0/G1 phase of the cell cycle and demonstrated generally a more profound apoptosis

470 Chem. Res. Toxicol., Vol. 20, No. 3, 2007

Dornetshuber et al.

Figure 4. Impact of ENN on DNA synthesis and cell cycle distribution. (A) DNA synthesis of KB-3-1 cells was determined by 3H-thymidine incorporation after 24 h of ENN exposure at the indicated concentrations. Data are expressed as means ( SD of triplicates derived from one of three experiments delivering comparable results. (B) PI staining and FACS analyses were performed on KB-3-1 cells after 24 h of ENN exposure at the indicated drug concentrations. Percentages of cells in the G0/G1, S, and G2/M phases of the cell cycle as well as apoptotic cells are indicated. (C) The impact of ENN at the indicated doses on the expression pattern on cyclins in KB-3-1 cells was determined by Western blotting analysis.

induction detectable already at 0.5 and 1 µM ENN. In accordance with the DNA synthesis data (Figure 5B), PI staining and FACS analysis revealed a more potent cell cycle arrest in the G0/G1 phase following ENN treatment in the p53 wild-type as compared to the knock-out cell line (Figure 5C). Profound induction of both p53 and the p53-downstream cyclin-dependent kinase inhibitor p21 by intermediate ENN concentrations was detectable in p53 wild-type cells by Western blot analysis, while this response was reduced at cytotoxic ENN concentrations g5 µM (Figure 5D). At these cytotoxic doses, p21 induction was also detectable in p53 (-/-) cells. Bax activation by ENN was already obvious at 0.5 µM ENN, correlating with beginning apoptosis induction (Figure 5C), but was independent of the cellular p53 status. These results suggest that the cytostatic activities are significantly influenced by the cellular p53 status whereas the cytotoxic effects of ENN are mediated by a p53independent mechanism.

Discussion ENN, a secondary metabolite of the genus Fusarium, is a frequent and worldwide contaminant of grain-derived food and

feed (4). However, data about the impact of ENN on human and animal health are rare. Today, it is widely unknown whether enhanced ENN intake has consequences on human health, but very recent data suggested that the toxic potential of ENN might have been underestimated (11). The cytotoxic effects of ENN, observed in this study, occur at micromolar concentrations, which are unlikely to be reached due to food consumption. Nevertheless, we found that short-term treatment (up to about 8 h) with low ENN concentrations (0.1-1 µM) induced a significantly increased proliferation of tumor cell line cells (Figure 1B), which might be based on a compensatory mechanism to ENN-induced stress signals. Thus, it may be possible that frequent, transient exposure to low ENN concentrations could have a tumor-promoting function as has been reported for fumonisin-induced hepatocarcinogenesis (28). Considering these facts and its lipophilic character, which allows bioaccumulation in animal and human tissues, ENN contamination could represent a potential health hazard. In how far this holds true has to be investigated in further toxicological and epidemiological studies. Structurally, ENN is related to the cyclic hexadepsipeptide BEA, which has already been reported to have cytotoxic

Apoptosis Induction by Enniatins

Chem. Res. Toxicol., Vol. 20, No. 3, 2007 471

Figure 5. Impact of apoptosis- and cell cycle-regulating proteins bcl-2, p53, p21, and bax on ENN-induced cytotoxicity. (A) Expression of bcl-2 in A549 cells transfected with a bcl-2 (A549/bcl-2) and a vector control (A549/vc) construct was determined by Western blot analysis. (B) DNA synthesis of HCT116 p53 (+/+) and HCT116 p53 (-/-) cells after 24 h of ENN treatment at the indicated concentrations was determined by 3 H-thymidine incorporation. Data are expressed as means ( SD of triplicates. Differences between the two cell lines were statistically significant; p < 0.001 (two-way ANOVA, followed by Bonferroni test). (C) Changes of cell cycle distribution in HCT116 p53 (+/+) and HCT116 p53 (-/-) cells after 24 h of ENN treatment at the indicated concentrations were analyzed by PI staining and flow cytometry analyses. (D) Impact of p53 on the expression pattern on apoptosis- and cell cycle-regulating proteins p53, p21, and bax, after 24 h of ENN treatment at the indicated concentrations, was determined by Western blot analysis.

activities capable of inducing apoptosis in human cancer cells (9, 10, 29). Here, we demonstrate that ENN at the low micromolar level exerts concentration- and time-dependent cytostatic and cytotoxic activities against diverse human cancer cell lines derived from various solid tumors and hematological malignancies. ENN-mediated cytotoxic effects against tumor cell lines were potent and almost independent of the histological origin including melanomas, sarcomas, diverse carcinomas, and the leukaemia cells HL60. In contrast, normal fibroblasts (WI38) were remarkably insensitive and also endothelial cells (HUVEC) were comparably resistant to ENN-induced cytotoxicity. As both normal and tumor cells were treated under

proliferating conditions, ENN might target a growth-independent mechanism hyperactivated in malignant cells. Additionally, significantly reduced ENN sensitivity also occurred in one case of melanoma, glioblastoma, and non-small cell lung cancer each, also suggesting that tumor cells might exhibit occasionally molecular mechanisms protecting against ENN-induced cell death. Recently, ENN has been identified as a substrate for the Saccharomyces cereVisiae multidrug transporter Pdr5p (30), a pleiotropic efflux mechanism for various compounds including diverse anticancer drugs (31). Pdr5p expression reduced the fungistatic activity of ENN, and the efflux pump was shown to

472 Chem. Res. Toxicol., Vol. 20, No. 3, 2007

Dornetshuber et al.

Table 2. Impact of p53 and the Target Genes p21 and Bax as Well as Bcl-2 on the Cytotoxic Activity of ENN IC75 (µM) cell line

meana

(SD

IC50 (µM) meana

IC25 (µM)

(SD

meana

(SD

p53 (+/+) p53 (+/-) p53 (-/-) p21 (+/+) p21 (-/-) Bax (+/+) Bax (-/-)

1.62 1.7 1.76 3.32 3.11 3.65 4.02

HCT116 cell line 0.02 1.96 0.19 2.02 0.12 2.09 0.22 4.32 0.11 3.83 0.12 4.18 0.06 4.41

0.01 0.14 0.11 0.05 0.32 0.07 0.07

2.01 2.35 2.46 6.06 4.75 4.75 4.84

0.02 0.13 0.11 0.40 0.11 0.03 0.08

A549/vc A549/bcl-2

2.81 2.32

human lung cancer 1.02 3.55 1.2 1.07 3.02 1.03

4.42 4.13

1.43 1.27

a IC , IC , and IC were calculated from whole dose-response curves. 75 50 25 Values given are means ( SD from at least two independent experiments performed in triplicate.

be inhibited by this cyclic hexadepsipeptide drug. These data suggest that the relative insensitivity of some cell models like T98G glioblastoma cells might be a consequence of ATPbinding cassette transporter overexpression. The protein Pdr5p has been shown to share nucleotide triphosphatases activities as well as substrates and modulators with human ABCB1 (Pglycoprotein) (31) responsible for chemotherapy failure in diverse clinical settings (32). This would suggest that ABCB1 might lead to ENN insensitivity as has already been hypothesized by Hiraga and co-workers (30). However, we have shown in a previous study (33) that T98G cells lack ABCB1 expression. Accordingly, in ongoing experiments dealing with the impact of diverse drug resistance mechanisms on the anticancer activities of mycotoxins, we could not detect any interaction of ENN with human ABCB1 (manuscript in preparation). This suggests that in some tumor models other resistance mechanisms might limit the cytotoxic activities of ENN. On the basis of several characteristics of programmed cell death like cell shrinkage, chromatin condensation, apoptotic body formation, and DNA fragmentation and the complete lack of cellular LDH release indicative for necrotic cell death (data not shown), we conclude that the cytotoxic effects of ENN are due to the induction of apoptosis. Additionally, ENN treatment led to permeabilization of the outer mitochondrial membrane, an important step to trigger programmed cell death (34). These data were further corroborated by the finding that a 24 h treatment with ENN induced cleavage of PARP and several procaspases including caspase-7 and -9 in a concentrationdependent manner. Additionally, functional activation of caspase-3 and -7 was observed in KB-3-1 cells treated for 18 h with ENN. Taken together, these results point out apoptosis induction via the mitochondrial pathway in ENN-treated cells. Comparably, studies on the structurally related cyclic hexadepsipeptide BEA also revealed activation of apoptotic pathways by release of cytochrome c from the mitochondria, leading to increased caspase-3 activity and finally resulting in typical morphologic signs of apoptosis (9, 10). This suggests that cyclic hexadepsipeptides generally induce programmed cell death via the “internal pathway”. In addition to apoptosis induction, our studies showed reduction of DNA synthesis already at a minimally effective concentration of 2.5 µM ENN, accompanied by cell cycle arrest. To investigate the impact of ENN on these apoptosis- and cell cycle-regulating proteins, HCT116 cells with disrupted p53, p21, or bax genes (19-21) and intrinsically bcl-2-negative A549 cells transfected with bcl-2 were used. In vitality assays, no significant influences of any of these proteins on the cytotoxic activity of

ENN were detectable. Nevertheless, ENN exposure significantly stimulated p53 as well as bax expression while that of the antiapoptotic bcl-2 family member bcl-xL was distinctly inhibited. In contrast, bim and puma as pro-apoptotic and bcl-2 as antiapoptotic proteins remained relatively unchanged. Moreover, the profound activation of the pro-apototic p53-downstream target bax was detectable in HCT116 cells at relatively low ENN concentrations (g0.5 µM) irrespective of the cellular p53 status suggesting a p53-independent activation of bax by ENN. Comparably, bax expression in many tissues has been shown not to be strictly dependent on p53 (35). These data indicate that ENN treatment might specifically influence certain bcl-2 family members facilitating activation of apoptosis. However, results obtained by vitality assays for the p53- and bax-knockout HCT116 cell lines as well as the bcl-2-overexpressing A549 model suggest that the ENN-mediated apoptosis induction is not generally dependent on either of these proteins. In contrast to vitality assays, 3H-thymidine incorporation revealed a significantly more efficient block of DNA synthesis by ENN in p53 wild-type as compared to knock-out cells, suggesting that the cytostatic in contrast to the cytotoxic activity of ENN is dependent at least in part on p53-mediated signals. Accordingly, treatment with ENN induced wild-type p53 expression already at low concentrations. Coinciding with the activation of the p53-downstream cyclin-dependent kinase inhibitor p21, stronger cell cycle arrest in the G0/G1 phase following ENN treatment was demonstrated in p53 wild-type cells. In contrast, p53 knock-out cells underwent stronger programmed cell death. This suggests that cell cycle blockade via p21 might rescue cells from ENN-induced apoptosis. It has been shown previously that preventing the activation of p21 might in some cases be necessary to drive cells toward p53mediated apoptosis (36). Consequently, the enhanced apoptosis rate in p53 (-/-) cells might be a consequence of a lack of p53-dependent p21 activation. Remarkably, at higher ENN concentrations, p21 induction was also detectable in p53 (-/-) cells. These results are consistent with a previous study reporting that the mycotoxin fumonisin B1 transcriptionally activates the p21 promoter in a p53-independent manner (37). Generally, the trigger for p53 activation by ENN is currently unknown. Preliminary data from our lab using comet assay and the radical scavenger N-acetyl cysteine (NAC) argue against a direct induction of DNA strand breaks and against production of reactive oxygen species in ENN-treated cells (data not shown). However, a direct activation of p53 expression by a rise of intracellular Ca2+ levels by ENN might be assumed, as has also been described for the calcium mobilizer calcimycin (10). In summary, we demonstrate that ENN exerts potent anticancer activity, which is widely independent of the cellular p53 and bcl-2 family status. From the pharmacological point of view, the cytostatic/cytotoxic effects of ENN at low micromolar concentrations against diverse cancer cell models and the comparable insensitivity of normal cells suggest that ENN might be useful as a chemotherapeutic drug. Acknowledgment. We thank the Austrian Science Fund (FWF) for supporting this work (Project P17089-B11) as well as R.D. and M.-R.K. Moreover, we are in debt to Pakiza Rawnduzi and Christian Balcarek for competent technical assistance as well as to Vera Bachinger for the skillful handling of cell culture. Furthermore, we thank Irene Herbacek for FACS analyses.

Apoptosis Induction by Enniatins

Chem. Res. Toxicol., Vol. 20, No. 3, 2007 473

References (1) Klyszejko, A., Kubus, Z., and Zakowska, Z. (2005) Mycological analysis of cereal samples and screening of Fusarium strains’ ability to form deoxynivalenole (DON) and zearalenone (ZEA) mycotoxinss A pilot study. Pol. J. Microbiol. 54 (Suppl.), 21-25. (2) Bennett, J. W., and Klich, M. (2003) Mycotoxins. Clin. Microbiol. ReV. 16, 497-516. (3) Jestoi, M., Rokka, M., Yli-Mattila, T., Parikka, P., Rizzo, A., and Peltonen, K. (2004) Presence and concentrations of the Fusariumrelated mycotoxins beauvericin, enniatins and moniliformin in finnish grain samples. Food Addit. Contam. 21, 794-802. (4) Logrieco, A., Rizzo, A., Ferracane, R., and Ritieni, A. (2002) Occurrence of beauvericin and enniatins in wheat affected by Fusarium aVenaceum head blight. Appl. EnViron. Microbiol. 68, 82-85. (5) Thrane, U., Adler, A., Clasen, P. E., Galvano, F., Langseth, W., Lew, H., Logrieco, A., Nielsen, K. F., and Ritieni, A. (2004) Diversity in metabolite production by Fusarium langsethiae, Fusarium poae, and Fusarium sporotrichioides. Int. J. Food Microbiol. 95, 257-266. (6) Tomoda, H., Nishida, H., Huang, X. H., Masuma, R., Kim, Y. K., and Omura, S. (1992) New cyclodepsipeptides, enniatins D, E and F produced by Fusarium sp. FO-1305. J. Antibiot. (Tokyo) 45, 12071215. (7) Mereish, K. A., Solow, R., Bunner, D. L., and Fajer, A. B. (1990) Interaction of cyclic peptides and depsipeptides with calmodulin. Pept. Res. 3, 233-237. (8) Kamyar, M., Rawnduzi, P., Studenik, C. R., Kouri, K., and LemmensGruber, R. (2004) Investigation of the electrophysiological properties of enniatins. Arch. Biochem. Biophys. 429, 215-223. (9) Jow, G. M., Chou, C. J., Chen, B. F., and Tsai, J. H. (2004) Beauvericin induces cytotoxic effects in human acute lymphoblastic leukemia cells through cytochrome c release, caspase 3 activation: the causative role of calcium. Cancer Lett. 216, 165-173. (10) Lin, H. I., Lee, Y. J., Chen, B. F., Tsai, M. C., Lu, J. L., Chou, C. J., and Jow, G. M. (2005) Involvement of Bcl-2 family, cytochrome c and caspase 3 in induction of apoptosis by beauvericin in human nonsmall cell lung cancer cells. Cancer Lett. 230, 248-259. (11) Ivanova, L., Skjerve, E., Eriksen, G. S., and Uhlig, S. (2006) Cytotoxicity of enniatins A, A1, B, B1, B2 and B3 from Fusarium aVenaceum. Toxicon 47, 868-876. (12) Berger, W., Spiegl-Kreinecker, S., Buchroithner, J., Elbling, L., Pirker, C., Fischer, J., and Micksche, M. (2001) Overexpression of the human major vault protein in astrocytic brain tumor cells. Int. J. Cancer 94, 377-382. (13) Berger, W., Hauptmann, E., Elbling, L., Vetterlein, M., Kokoschka, E. M., and Micksche, M. (1997) Possible role of the multidrug resistance-associated protein (MRP) in chemoresistance of human melanoma cells. Int. J. Cancer 71, 108-115. (14) Kubista, B., Trieb, K., Sevelda, F., Toma, C., Arrich, F., Heffeter, P., Elbling, L., Sutterluty, H., Scotlandi, K., Kotz, R., Micksche, M., and Berger, W. (2006) Anticancer effects of zoledronic acid against human osteosarcoma cells. J. Orthop. Res. 24, 1145-1152. (15) Berger, W., Micksche, M., and Elbling, L. (1997) Effects of multidrug resistance-related ATP-binding-cassette transporter proteins on the cytoskeletal activity of cytochalasins. Exp. Cell Res. 237, 307-317. (16) Shen, D. W., Cardarelli, C., Hwang, J., Cornwell, M., Richert, N., Ishii, S., Pastan, I., and Gottesman, M. M. (1986) Multiple drugresistant human KB carcinoma cells independently selected for highlevel resistance to colchicine, adriamycin, or vinblastine show changes in expression of specific proteins. J. Biol. Chem. 261, 7762-7770. (17) McGrath, T., and Center, M. S. (1988) Mechanisms of multidrug resistance in HL60 cells: Evidence that a surface membrane protein distinct from P-glycoprotein contributes to reduced cellular accumulation of drug. Cancer Res. 48, 3959-3963. (18) Doyle, L. A., Yang, W., Abruzzo, L. V., Krogmann, T., Gao, Y., Rishi, A. K., and Ross, D. D. (1998) A multidrug resistance transporter from human MCF-7 breast cancer cells. Proc. Natl. Acad. Sci. U.S.A. 95, 15665-15670. (19) Bunz, F., Fauth, C., Speicher, M. R., Dutriaux, A., Sedivy, J. M., Kinzler, K. W., Vogelstein, B., and Lengauer, C. (2002) Targeted

(20) (21) (22)

(23)

(24) (25) (26)

(27) (28)

(29)

(30)

(31)

(32) (33)

(34) (35)

(36) (37)

inactivation of p53 in human cells does not result in aneuploidy. Cancer Res. 62, 1129-1133. Waldman, T., Kinzler, K. W., and Vogelstein, B. (1995) p21 is necessary for the p53-mediated G1 arrest in human cancer cells. Cancer Res. 55, 5187-5190. Zhang, L., Yu, J., Park, B. H., Kinzler, K. W., and Vogelstein, B. (2000) Role of BAX in the apoptotic response to anticancer agents. Science 290, 989-992. Berger, W., Elbling, L., Minai-Pour, M., Vetterlein, M., Pirker, R., Kokoschka, E. M., and Micksche, M. (1994) Intrinsic MDR-1 gene and P-glycoprotein expression in human melanoma cell lines. Int. J. Cancer 59, 717-723. Berger, W., Elbling, L., and Micksche, M. (2000) Expression of the major vault protein LRP in human non-small-cell lung cancer cells: Activation by short-term exposure to antineoplastic drugs. Int. J. Cancer 88, 293-300. Loo, D. T., and Rillema, J. R. (1998) Measurement of cell death. Methods Cell Biol. 57, 251-264. Broker, L. E., Kruyt, F. A., and Giaccone, G. (2005) Cell death independent of caspases: A review. Clin. Cancer Res. 11, 31553162. Polla, B. S., Kantengwa, S., Francois, D., Salvioli, S., Franceschi, C., Marsac, C., and Cossarizza, A. (1996) Mitochondria are selective targets for the protective effects of heat shock against oxidative injury. Proc. Natl. Acad. Sci. U.S.A. 93, 6458-6463. Cory, S., and Adams, J. M. (2002) The Bcl2 family: Regulators of the cellular life-or-death switch. Nat. ReV. Cancer 2, 647-656. Gelderblom, W. C., Abel, S., Smuts, C. M., Marnewick, J., Marasas, W. F., Lemmer, E. R., and Ramljak, D. (2001) Fumonisin-induced hepatocarcinogenesis: mechanisms related to cancer initiation and promotion. EnViron. Health Perspect. 109 (Suppl. 2), 291-300. Calo, L., Fornelli, F., Ramires, R., Nenna, S., Tursi, A., Caiaffa, M. F., and Macchia, L. (2004) Cytotoxic effects of the mycotoxin beauvericin to human cell lines of myeloid origin. Pharmacol. Res. 49, 73-77. Hiraga, K., Yamamoto, S., Fukuda, H., Hamanaka, N., and Oda, K. (2005) Enniatin has a new function as an inhibitor of Pdr5p, one of the ABC transporters in Saccharomyces cereVisiae. Biochem. Biophys. Res. Commun. 328, 1119-1125. Rogers, B., Decottignies, A., Kolaczkowski, M., Carvajal, E., Balzi, E., and Goffeau, A. (2001) The pleitropic drug ABC transporters from Saccharomyces cereVisiae. J. Mol. Microbiol. Biotechnol. 3, 207214. Gottesman, M. M., Fojo, T., and Bates, S. E. (2002) Multidrug resistance in cancer: Role of ATP-dependent transporters. Nat. ReV. Cancer 2, 48-58. Spiegl-Kreinecker, S., Buchroithner, J., Elbling, L., Steiner, E., Wurm, G., Bodenteich, A., Fischer, J., Micksche, M., and Berger, W. (2002) Expression and functional activity of the ABC-transporter proteins P-glycoprotein and multidrug-resistance protein 1 in human brain tumor cells and astrocytes. J. Neurooncol. 57, 27-36. Fadeel, B., and Orrenius, S. (2005) Apoptosis: A basic biological phenomenon with wide-ranging implications in human disease. J. Intern. Med. 258, 479-517. Bouvard, V., Zaitchouk, T., Vacher, M., Duthu, A., Canivet, M., Choisy-Rossi, C., Nieruchalski, M., and May, E. (2000) Tissue and cell-specific expression of the p53-target genes: Bax, fas, mdm2 and waf1/p21, before and following ionising irradiation in mice. Oncogene 19, 649-660. Yu, J., and Zhang, L. (2005) The transcriptional targets of p53 in apoptosis control. Biochem. Biophys. Res. Commun. 331, 851-858. Zhang, Y., Dickman, M. B., and Jones, C. (1999) The mycotoxin fumonisin B1 transcriptionally activates the p21 promoter through a cis-acting element containing two Sp1 binding sites. J. Biol. Chem. 274, 12367-12371.

TX600259T