- Email: [email protected]
Altered thioredoxin subcellular localization and redox status in MCF-7 cells following 1,25-dihydroxyvitamin D3 treatment
Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 57–64
Altered thioredoxin subcellular localization and redox status in MCF-7 cells following 1,25-dihydroxyvitamin D3 treatment Belinda M. Byrne, JoEllen Welsh ∗ Department of Biological Sciences, 214 Galvin Life Sciences Building, University of Notre Dame, Notre Dame, IN 46556, USA
Abstract 1,25-Dihydroxyvitamin D3 (1,25D) induces apoptosis in MCF-7 cells via the intrinsic pathway involving bax translocation to mitochondria, cytochrome c release and reactive oxygen species (ROS) generation. Vitamin D up-regulated protein 1 (VDUP1), an apoptotic regulatory gene induced by 1,25D in HL-60 cells, is a negative regulator of thioredoxin (Trx1), a redox protein which neutralizes ROS and protects against oxidative stress induced apoptosis. Due to the involvement of oxidative stress in 1,25D mediated apoptosis, we analyzed whether VDUP1 or Trx1 are altered by 1,25D in MCF-7 cells. In contrast to HL-60 cells, VDUP1 mRNA was not up-regulated by 1,25D in MCF-7 cells, indicating that transcriptional up-regulation of this gene is not required for 1,25D mediated apoptosis. 1,25D did not affect the expression or activity of Trx1 in MCF-7 cells, however, Trx1 activity was higher in MCF-7 cells selected for resistance to 1,25D mediated apoptosis. In untreated MCF-7 cells, Trx1 was present only in the cytosol, and the majority was in the oxidized state. In 1,25D treated MCF-7 cells, Trx1 was present in both cytosol and nucleus, and the nuclear Trx1 pool was in the reduced state. Nuclear localization of Trx1 in 1,25D treated MCF-7 cells was confirmed by immunofluorescent microscopy. Although redox status is known to alter the ability of Trx1 to bind apoptosis signal regulating kinase 1 (ASK1), no changes in ASK1 transcript or protein levels were observed in 1,25D treated MCF-7 cells. Collectively, these studies indicate that although VDUP1 and ASK1 are not altered by 1,25D, changes in redox status and sub-cellular distribution of Trx1 occurs during 1,25D mediated apoptosis of MCF-7 cells. © 2005 Elsevier Ltd. All rights reserved. Keywords: Vitamin D3 ; VDUP1; Thioredoxin; Breast cancer; Apoptosis
1. Introduction 1,25-Dihydroxyvitamin D3 (1,25D), the biologically active form of Vitamin D3 , is a fat soluble steroid hormone that exerts anti-proliferative and pro-apoptotic effects on breast cancer cells. The ability of 1,25D to inhibit growth has consistently been observed in a variety of breast cancer cell lines, including those derived from human tumors, both in vitro and in vivo [1]. 1,25D exerts its effects through binding the Vitamin D receptor (VDR), a ligand dependent transcription factor. The presence of VDR in breast tumors has been associated with slower growing tumors or less aggressive disease [1] and epidemiological studies have indicated an inverse correlation between breast cancer incidence and progression and Vitamin D status [2–4]. In addition, VDR agonists can potentiate the effects of certain cytotoxic compounds, ∗
Corresponding author. Tel.: +1 574 631 3371; fax: +1 574 631 7413. E-mail address: [email protected] (J. Welsh).
0960-0760/$ – see front matter © 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.jsbmb.2005.06.023
such as tumor necrosis factor ␣ (TNF␣) [5], doxorubicin [6], and the phorbol ester phorbol 12-myristate 13-acetate (TPA) [7]. Our lab initially demonstrated that 1,25D induces morphological, biochemical, and molecular features of apoptosis in MCF-7 human breast cancer cells [8]. 1,25D mediated apoptosis in MCF-7 cells involves a novel caspase independent pathway that triggers mitochondrial disruption [9,10]. To facilitate investigation into the mechanisms of 1,25D induced apoptosis, we generated a 1,25D resistant variant from MCF7 cell line—the MCF-7DR cell line [11]. MCF-7DR cells express the VDR but are selectively resistant to 1,25D, indicating that expression of the VDR is not sufficient for the induction of growth arrest and apoptosis by 1,25D [11]. In the studies reported here, we used MCF-7 and MCF-7DR cells to investigate the role of redox signaling during 1,25D induced apoptosis. Induction of apoptosis in MCF-7 cells by 1,25D is characterized by down regulation of Bcl-2, translocation of Bax to
58
B.M. Byrne, J. Welsh / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 57–64
Fig. 1. Thioredoxin redox cycling. The components of the thioredoxin system include thioredoxin (Trx1), thioredoxin reductase 1 (TrxR1), and NADPH. Trx1 becomes oxidized while reducing free radicals and/or maintaining proteins in their reduced state. NADPH and TrxR1 function to recycle Trx1 back to the reduced (active) form.
mitochondria, redistribution of cytochrome c and generation of reactive oxygen species (ROS) [9]. Time course studies indicated that ROS generation is an early event in 1,25D mediated apoptosis [9]. Neither cytochrome c release, ROS generation nor Bcl-2/Bax alteration occurs in MCF-7DR cells after treatment with 1,25D. MCF-7DR cells, however, are sensitive to TNF␣ mediated apoptosis, indicating that these cells express a functional apoptotic pathway. Since the cytotoxic action of TNF␣ also involves ROS generation and mitochondrial disruption, these data suggest that the MCF-7DR cells have uncoupled VDR signaling, but not TNF␣ signaling, from ROS generation. Vitamin D up-regulated protein 1 (VDUP1) was originally identified as a novel 1,25D induced gene in HL-60 cells [12]. Little is known about the function of VDUP1, although overexpression of VDUP1 induces oxidative stress, possibly via interaction with the redox protein thioredoxin 1 (Trx1) [13]. Trx1 binds and neutralizes oxygen free radicals, including ROS, and thus protects cells against oxidative stress. Binding of ROS oxidizes Trx1, which can be converted back to its reduced form by thioredoxin reductase (Trx1R) in the presence of NADPH (Fig. 1). In the reduced form, Trx1 also prevents apoptosis by binding and inhibiting apoptosis signaling kinase 1 (ASK1). VDUP1 binds directly to the redox active site of Trx1 and inhibits its ability to neutralize ROS as well as its ability to inhibit ASK1 [14]. We therefore hypothesized that accumulation of ROS in 1,25D treated MCF-7 cells could be secondary to up-regulation of VDUP1, inhibition of Trx1 and activation of ASK1. To test this hypothesis, we assessed components of this redox system in relation to ROS generation in 1,25D treated MCF-7 and MCF-7DR cells. Our studies provide evidence that 1,25D alters Trx1 redox state and subcellular localization in MCF-7 cells but not in MCF-7DR cells.
2. Materials and methods 2.1. Cell culture and growth assays Parental MCF-7 and the 1,25D resistant variant MCF-7DR cells [11] were cultured in ␣-MEM medium (Life Technolo-
gies Inc., Gaithersburg, MD) containing 25 mM HEPES and 5% FBS (Life Technologies). Cells were routinely plated at 5000 cells/cm2 and passaged every 3–4 days. For experiments, cells were plated in ␣-MEM containing 5% FBS plus antibiotics, and treatments with 100 nM 1,25D or vehicle control were initiated 24 h after plating. For long term experiments, cells were re-fed with fresh medium containing treatments after 72 h. In some experiments, parallel cultures were treated with TNF␣ (2 ng/ml) 2 days prior to harvest as a positive control for apoptosis. For assessment of growth, 2000 cells were plated per well in 24-well plates and treated 24 h later. After 96 or 120 h, cultures were fixed in 1% gluteraldehyde, stained with 0.1% crystal violet in dH2 O and dried overnight. Crystal violet was resuspended in 0.2% Triton X100 and absorbance at 590 nm was read on a Wallac Victor 2 plate reader. Previous studies have confirmed that absorbance of crystal violet is proportional to adherent cell number under the conditions used. 2.2. Western blotting Cells treated for 96 or 120 h were harvested by direct lysis in Laemlli buffer, separated on 5% (ASK1) or 15% (Trx1) SDS–PAGE and electrophoretically transferred to nitrocellulose paper. Blots were probed with mouse anti-human Trx1 (1:100) (Pharmingen, CA) followed by anti-mouse secondary (1:5000) or anti-human ASK1 (1:500) (Cell Signaling Technology Inc., MA) followed by anti-rabbit secondary (1:3000). Specific antibody binding was visualized by chemiluminescence using products from Pierce (Pierce Biotechnology, IL). 2.3. Immunocytochemistry Cells were grown for 2 days on Lab-Tek II chamber slides (Fisher Scientific) and treated with ethanol vehicle or 100 nM 1,25D for 120 h, or 2 ng/ml TNF␣ for 48 h. The cells were fixed in 4% formaldehyde in PBS, permeabilized in methanol, and blocked overnight with PBS/1% BSA. The slides were incubated with anti-Trx1 mouse monoclonal antibody (1:50, Pharmingen) for 2 h at 37 ◦ C followed by incubation with anti-mouse Alexa 488. Nuclei were counter-stained with 1 g/ml Hoechst for 15 min at room temperature. Fluorescence was detected with a Bio-Rad MRC 1024 scanning confocal (three channel/LaserSharp 2000, version 5.0 program) system, utilizing a krypton–argon (Kr–Ar) laser. 2.4. Assessment of thioredoxin redox status Nuclear and cytosolic fractions were isolated based on the procedure of Janssen and Sen [15], with modifications to allow redox measurement of Trx1 via the redox Western blot. Cells were pelleted by centrifugation and lysed in hypotonic lysis buffer (10 mM HEPES, pH 7.8, 10 mM KCl, 2 mM MgCl2 , 0.1 mM EDTA, 0.2 mM NaF, 0.2 mM Na3 VO4 ·6H2 O) with protease inhibitors (leupeptin, aprotinin, pepstatin, phenylmethylsulfonyl fluoride). For redox
B.M. Byrne, J. Welsh / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 57–64
Western blot, 50 mM iodoacetic acid (IAA) was added to the lysis buffer and the pH was adjusted to 7.8. The suspensions were incubated on ice for 5 min, and Nonidet P40 was added to a final concentration of 0.6%. Following centrifugation at 16,000 × g, the pellet (nuclei) and the supernatant (cytosol) were suspended in 6 M guanidine–HCl, 50 mM Tris, 3 mM EDTA, 0.5% Triton X100 (pH 8.3) supplemented with 50 mM IAA and incubated at 37 ◦ C for 30 min. After removal of excess IAA, Sephadex chromatography (MicroSpin G25 columns, Amersham-Pharmacia, Piscataway, NJ, USA), samples were subjected to native electrophoresis to separate reduced and oxidized forms of Trx1 [16,17]. 2.5. Thioredoxin activity assay Thioredoxin activity was measured with the insulin reducing assay [13]. Briefly, cells were lysed in saline buffer (100 mM Na2 HPO4 , 1 mM EDTA, pH 7.5) using a dounce homogeniser and incubated at 70 ◦ C for 10 min, then DTT activation buffer (50 mM HEPES, 1 mM EDTA, 1 mg/ml BSA, 2 mM DTT) was added for a 15 min incubation at 37 ◦ C. Reactions were initiated by addition of 150 mM HEPES, 1 mM EDTA, 1 mg/ml NADPH, and 2 mg/ml insulin in the presence of thioredoxin reductase 1 (American Diagnostica Inc., CT) or water (for control reactions). After incubation at 37 ◦ C for 15 min, the reactions were stopped by the addition of 0.4 mg/ml DTNB/6M guanidine–HCl in 0.2M Tris/HCl, pH 8.0. Absorbance was determined at 412 nm. 2.6. Quantitative real-time PCR Total RNA was isolated from MCF-7 cells (2 × 106 cells) with the RNeasy Protect Mini Kit (Qiagen). After concentration and purity was determined by spectrophotometry, total RNA was used for reverse transcription reactions using TaqMan Reverse Transcription Reagents (Applied Biosystems, CA). For each cell line, reactions were performed in duplicate, generating two 1.5 ng cDNA stocks/sample. Each of the duplicate cDNA stocks were independently analyzed in duplicate (60 ng of cDNA/well) by real time PCR using the SYBR Green PCR Master Mix (Applied Biosystems) and a primer set specific for human VDUP1 (Forward primer: AGATCAGGTCTAAGCAGCAGAACA, Reverse primer: TCAGATCTACCCAACTCATCTCAGA) or ASK1 (Forward primer: CCAGGGCCTCTTCCTGCTA, Reverse primer: TCTGCTCCTCCCAGAGGATTT). Transcript values were normalized against GAPDH RNA (Forward primer: CCACCCATGGCAAATTCC, Reverse primer: TGATGGGATTTCCATTGATGAC) and reported as normalized gene expression. 2.7. Flow cytometric determination of intracellular ROS Both adherent and floating cells were collected, pelleted, resuspended, and incubated in the presence of 4 M hydroethidine (HE, Molecular Probes) as previously
59
described [9]. Analysis of HE to ethidium was analyzed on an Epics XL Flow Cytometer (Coulter Corp., Miami, FL) equipped with an argon laser. HE was analyzed on FL3 using a 620 nm band pass filter. Data was analyzed using MULTIPLUS AV analysis software (Phoenix Flow Systems). 2.8. Statistical analyses Data are expressed as mean ± S.E. One-way ANOVA was used to assess statistical significance between means. Differences between means were considered significant if P-values less than 0.05 were obtained with the Dunnett’s method using GraphPad Prism software (GraphPad Software, San Diego, CA).
3. Results 3.1. Comparative effects of 1,25D and TNFα on morphology and ROS generation in MCF-7 and MCF-7DR cells The effects of 1,25D and TNF␣ on morphology of MCF-7 and MCF-7DR cultures are shown in Fig. 2A. MCF-7 cells exposed to ethanol vehicle grow in a typical epithelial cobblestone pattern. After 120 h treatment with 100 nM 1,25D, MCF-7 cultures show evidence of apoptosis, including retraction of cell-to-cell contacts, cytoplasmic condensation and loss of adherence. The MCF-7DR cells, which were selected for growth in the presence of 100 nM 1,25D, do not demonstrate any morphological evidence of apoptosis in response to 1,25D. However, both cell lines exhibit typical apoptotic morphology after treatment with 2 ng/ml TNF␣. In all cases, apoptotic morphology correlated with a reduction in the number of adherent cells as measured by crystal violet staining (Fig. 2B). Thus, cell density was significantly reduced by 1,25D in MCF-7 cultures and by TNF␣ in MCF-7 and MCF-7DR cultures. TNF␣ induced a comparable reduction in adherent cell number in both MCF-7 and MCF-7DR cell lines. Apoptotic morphology and reduced cell density also correlated with generation of reactive oxygen species (ROS) as measured by flow cytometry (Fig. 3). The percentage of cells with intracellular ROS accumulation was over 30% in 1,25D treated MCF-7 cultures, but negligible in 1,25D treated MCF-7DR cultures. TNF␣ induced ROS accumulation in both MCF-7 and MCF-7DR cultures. These data indicate that apoptosis triggered by either 1,25D or TNF␣ is associated with ROS generation in MCF-7 cells, and that loss of 1,25D sensitivity in MCF-7DR cells does not alter sensitivity to TNF␣. 3.2. 1,25D does not alter VDUP1 expression in MCF-7 or MCF-7DR cells Since VDUP1, a VDR target gene in HL-60 leukemic cells, has been linked to induction of oxidative stress
60
B.M. Byrne, J. Welsh / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 57–64
Fig. 2. Effects of 1,25D and TNF␣ on MCF-7 and MCF-7DR cells. (A) MCF-7 and MCF-7DR cells were treated with ethanol, 100 nM 1,25D (120 h) or 2 ng/ml TNF␣ (48 h) and morphology was imaged by phase contrast microscopy. (B) Cells were treated as in A and relative adherent cell density was determined by crystal violet growth assay. Data are mean ± S.E. * Statistically significant from control value, Anova.
[13,18], we assessed whether steady state mRNA expression of VDUP1 is induced by 1,25D during MCF-7 cell apoptosis. MCF-7 and MCF-7DR cells were treated for 3, 6, 12, 24 or 48 h with vehicle control or 100 nM 1,25D, and VDUP1 was assessed by quantitative real time PCR (Fig. 4). Although VDUP1 mRNA increased from
3 to 12 h after treatment, there were no significant differences in VDUP1 transcript levels between control and 1,25D treated cells at any time point in either MCF-7 (Fig. 4A) or MCF-7DR (Fig. 4B) cells. These data indicate that VDUP1 is not a VDR regulated gene in MCF-7 cells.
Fig. 3. 1,25D induces ROS production in MCF-7 but not MCF-7DR cells. MCF-7 and MCF-7DR cells were treated for 120 h with ethanol, 100 nM 1,25D (120 h) or 2 ng/ml TNF␣ (48 h) and ROS production was analyzed as conversion of hydroethidine to ethidium by flow cytometry. Shaded peaks indicate cells positive for ROS, percentages are shown in upper right corner.
B.M. Byrne, J. Welsh / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 57–64
61
Fig. 4. VDUP1 expression in 1,25D treated MCF-7 and MCF-7DR cells. MCF-7 (A) and MCF-7DR (B) cells were treated with ethanol or 1,25D for 3, 6, 12, 24 or 48 h and VDUP1 transcript levels were measured by real time PCR. Data are mean ± S.E.
3.3. Effects of 1,25D on Trx1 expression, redox status and activity The lack of effect of 1,25D on VDUP1 prompted us to examine whether 1,25D alters expression or activity of Trx1, a redox protein that neutralizes intracellular ROS and protects cells against oxidative stress induced apoptosis [19]. In MCF7 cells, expression of Trx1, measured by quantitative real time PCR, was stable over 48 h in culture and was not affected by 100 nM 1,25D (Fig. 5A). Furthermore, Western blotting indicated that neither 1,25D nor TNF␣ altered Trx1 protein expression in MCF-7 or MCF-7DR cells (Fig. 5B). However, Trx1 expression was elevated in MCF-7DR cells compared to MCF-7 cells. Since Trx1 expression may not reflect its activity, we used an insulin reduction assay to assess Trx1 reducing activity in both cell lines under basal and 1,25D treated conditions. No significant difference was seen between control and 1,25D treated MCF-7 or MCF-7DR cells, but thioredoxin activity was higher in MCF-7DR cells (Fig. 5C), consistent with the higher Trx1 protein expression in this cell line. To confirm the Trx1 activity assay results, we used a redox Western blot technique to distinguish between oxidized and reduced pools of Trx1 in nuclear and cytoplasmic fractions of MCF-7 cells. In control MCF-7 cells, Trx1 was only detected in the cytosol, and although both oxidized and reduced forms of Trx1 were present, the majority was in the oxidized state (Fig. 6). In 1,25D treated MCF-7 cells, cytosolic Trx1 was markedly reduced compared to control cells, and only the
Fig. 5. Trx1 expression and function in MCF-7 and MCF-7DR cells. (A) MCF-7 cells were treated with ethanol or 1,25D for 3, 6, 12, 24 or 48 h and Trx1 mRNA was measured by real time PCR. Data are mean ± S.E. (B) MCF-7 cells were treated with ethanol or 1,25D for 120 h or TNF␣ for 48 h, and Trx1 protein was measured by Western blotting. (C) MCF-7 or MCF7DR cells were treated with ethanol or 1,25D for 120 h and Trx1 activity was measured by insulin reduction assay as described in Section 2. Statistical significance was assessed by Anova.
oxidized form was present. In contrast to control MCF-7 cells, in which Trx1 was only present in the cytosol, Trx1 was detected in the nucleus of 1,25D treated cells, and this pool of Trx1 was exclusively in the reduced form. TNF␣ treatment did not alter Trx1 redox state or localization in MCF-7 cells. These data indicated that 1,25D alters the redox status of Trx1, and suggested subcellular relocalization of the protein as well. 3.4. 1,25D induces nuclear translocation of Trx1 in MCF-7 cells Nuclear redistribution of Trx1 has previously been reported during radiation induced apoptosis [20]. In light of the results of the redox blot showing nuclear Trx1 in 1,25D treated but not vehicle treated MCF-7 cells, we used
62
B.M. Byrne, J. Welsh / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 57–64
Fig. 6. Effect of 1,25D on Trx1 redox status in subcellular fractions. Subcellular fractions from MCF-7 cells treated with ethanol or 1,25D for 120 h were separated on native polyacrylamide gels and probed for Trx1. Arrows indicate positions of oxidized (top) and reduced (bottom) forms of Trx1. Native marker is indicated by 14.
immunofluorescence to localize Trx1 in MCF-7 cells. In control cells (Fig. 7, top panels), faint diffuse staining of Trx1 was detected in the cytosol, and no co-localization of Trx1 with the nuclear Hoechst stain was observed in merged images. In MCF-7 cells treated with 100 nM 1,25D for 120 h (Fig. 7, bottom panels), intense, punctuate staining of Trx1 was detected in early apoptotic cells. Trx1 staining colocalized with condensed chromatin visualized with Hoechst staining in merged images. This altered staining in response to 1,25D was not observed in MCF-7DR cells (data not shown). 3.5. 1,25D does not alter ASK1 mRNA or protein expression in either MCF-7 or MCF-7DR cells Apoptosis signaling kinase 1 (ASK1) has been shown to be essential for some forms of oxidative stress induced apoptosis [21], and reduced forms of Trx1 bind to and inhibit ASK1 activity [14]. We therefore used real-time PCR and Western blotting to determine whether ASK1 expression was altered
by 1,25D. As shown in Fig. 8A, ASK1 mRNA expression in MCF-7 cells was not altered by time in culture or treatment with 1,25D. Furthermore, no changes in ASK1 protein levels were observed in response to 1,25D or TNF␣ treatment in MCF-7 or MCF-7DR cell line (Fig. 8B).
4. Discussion These studies have confirmed and extended previous work demonstrating a role for oxidative stress in 1,25D mediated apoptosis of breast cancer cells [6,9]. MCF-7 cell death triggered by either 1,25D or the cytokine TNF␣ was associated with accumulation of ROS, a characteristic of apoptosis mediated by the mitochondrial (intrinsic) pathway. This finding is consistent with work demonstrating changes in the Bcl2/Bax ratio, mitochondrial permeability transition and cytochrome c release in 1,25D treated MCF-7 cells [9,23]. These alterations subsequently activate executioner caspases, leading to
Fig. 7. Translocation of Trx1 to nucleus in 1,25D treated MCF-7 cells. MCF-7 cells were treated with ethanol or 1,25D for 120 h, fixed, permeabilized, and probed with antibody directed against Trx1, followed by anti-mouse secondary antibody conjugated to Alexa 488. Nuclei were counterstained with Hoechst dye and cells were visualized on a biorad fluorescent microscope.
B.M. Byrne, J. Welsh / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 57–64
Fig. 8. Expression of ASK1 mRNA and protein in 1,25D treated MCF-7 cells. (A) MCF-7 cells were treated with ethanol or 1,25D for 3, 6, 12, 24 or 48 h and ASK1 transcript levels were measured by real time PCR. Data are mean ± S.E. (B) MCF-7 cells were treated with ethanol or 1,25D for 120 h or TNF␣ for 48 h and ASK1 protein was assessed by Western blotting.
cytoplasmic and nuclear condensation and irreversible DNA fragmentation [8]. The major goal of the experiments described here was to determine the impact of 1,25D on proteins involved in cellular oxidative stress responses. We specifically examined VDUP1, a 1,25D target gene in HL-60 cells [12] and Trx1, a VDUP1 interacting protein that protects cells from oxidative stress [13]. VDUP1 has been shown to inhibit the cytoprotective activity of Trx1 [13], and Trx1 reductase, an enzyme that recycles Trx1, has been identified as a 1,25D target gene by microarray profiling [25]. These data suggested that VDUP1 or Trx1 might be functionally involved in triggering oxidative stress in response to 1,25D in MCF-7 cells. In contrast to HL-60 cells, however, our data indicate that VDUP1 is not a 1,25D regulated gene in MCF-7 cells, thus changes in VDUP1 gene expression are unlikely to contribute to 1,25D mediated ROS generation or apoptosis. Although there were no changes in Trx1 transcript or protein levels in 1,25D treated MCF-7 cells, Trx1 subcellular localization and redox status was altered by 1,25D. Increased nuclear staining of Trx1, and accumulation of the reduced form of Trx1 in nuclear fractions detected by redox Western blotting, was observed in MCF-7 cells following 1,25D treatment. Similar nuclear localization in the absence of changes in total protein expression has been reported in HeLa cells induced to undergo apoptosis by ionizing radiation [20]. These data suggest that nuclear translocation of Trx1 may be a hallmark of oxidative stress in general rather than a specific marker of 1,25D mediated apoptosis.
63
In addition to VDUP1 and Trx1, we assessed the effects of 1,25D on ASK1, a mitogen activated protein (MAP) kinase, which promotes apoptosis through the JNK/p38 pathways [26]. The rationale for investigation of ASK1 was based on reports that Trx1 binds to and inhibits ASK1 [14,27], an interaction that is disrupted during oxidative stress. We therefore considered that ROS generation during 1,25D mediated apoptosis might alter Trx1–ASK1 interaction, promoting apoptosis through the JNK/p38 pathways. Although evidence for 1,25D regulation of JNK/p38 signaling has been provided in MCF-7 cells [28] and in squamous carcinoma cells [29], we found no effect of 1,25D on either ASK1 gene expression or protein levels. However, further studies will be necessary to determine if 1,25D alters ASK1 activity or JNK/p38 signaling in relation to apoptosis induction in MCF7 cells. As a redox protein, Trx1 can exist in both oxidized and reduced forms, and its interactions with other proteins, including VDUP1 and ASK1, are dependent on its redox status [14,30]. In control MCF-7 cells, both oxidized and reduced forms of Trx1 were detected, but only in the cytoplasmic fraction. In 1,25D treated cells, Trx1 was present in the nucleus in the reduced form, whereas cytosolic Trx1 was predominantly in the oxidized form. Since the oxidized form of Trx1 is unable to bind and inhibit ASK1, these data warrant further study to measure ASK1 activity during 1,25D treated MCF-7 cell apoptosis. Although both 1,25D and TNF␣ induce apoptosis via ROS generation and mitochondrial disruption in MCF-7 cells, no changes in Trx1 redox state or subcellular localization were observed in TNF␣ treated MCF-7 cells. These data indicate that apoptosis induced by 1,25D may be mechanistically different from that of TNF␣. In support of this concept, the MCF-7DR cells, which were selected for resistance to 1,25D, retain sensitivity to TNF␣ mediated apoptosis. These studies also provide insight into potential mechanisms of 1,25D resistance. Although neither Trx1 expression nor activity was altered by 1,25D, there were higher levels of Trx1 expression and activity in MCF-7DR cells than in parental MCF-7 cells. Thus, Trx1 overexpression may contribute to the resistant phenotype of MCF-7DR cells by neutralizing free radicals to protect against oxidative stress induced by 1,25D. In support of this suggestion, ROS accumulation was negligible in MCF-7DR cells treated with 1,25D under conditions, when >30% of parental MCF-7 cells were ROS positive. Also consistent with this concept, ROS generation in response to TNF␣ was blunted in MCF-7DR cells compared to parental MCF-7 cells. In summary, these studies indicate that VDUP1 is not a 1,25D regulated gene in MCF-7 cells, but confirm that 1,25D mediated apoptosis in MCF-7 cells is associated with oxidative stress. In addition to ROS generation, changes in Trx1 subcellular localization and redox state are associated with 1,25D mediated apoptosis in MCF-7 cells. Furthermore, upregulation of cellular anti-oxidant defenses, such as Trx1 may contribute to 1,25D resistance in breast cancer cells.
64
B.M. Byrne, J. Welsh / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 57–64
Acknowledgements This work was supported by Department of Defense Breast Cancer Research Program Grant DAMD17-03-10358. We thank Dr. Judy Narvaez for technical assistance on the flow cytometer.
[14]
[15] [16]
References [1] K.W. Colston, U. Berger, R.C. Coombes, Possible role for Vitamin D in controlling breast cancer cell proliferation, Lancet 1 (8631) (1989) 188–191. [2] J. Welsh, J.A. Wietzke, G.M. Zinser, B. Byrne, K. Smith, C.J. Narvaez, Vitamin D3 receptor as a target for breast cancer prevention, J. Nutr. 133 (Suppl. 7) (2003) 2425S–2433S. [3] E.M. John, G.G. Schwartz, D.M. Dreon, J. Koo, Vitamin D and breast cancer risk: the NHANES I Epidemiologic follow-up study, 1971–1975 to 1992. National Health and Nutrition Examination Survey, Cancer Epidemiol. Biomar. Prev. 8 (5) (1999) 399–406. [4] E. Mawer, B.J. Walls, A. Howell, M. Davies, W.A. Ratcliffe, N.J. Bundred, Serum 1,25-dihydroxyvitamin D may be related inversely to disease activity in breast cancer patients with bone metastases, J. Clin. Endocrinol. Metab. 82 (1997) 118–122. [5] D. Rocker, A. Ravid, U.A. Liberman, O. Garach-Jehoshua, R. Koren, 1,25-Dihydroxyvitamin D3 potentiates the cytotoxic effect of TNF on human breast cancer cells, Mol. Cell Endocrinol. 106 (1–2) (1994) 157–162. [6] A. Ravid, D. Rocker, A. Machlenkin, C. Rotem, A. Hochman, G. Kessler-Icekson, U.A. Liberman, R. Koren, 1,25-Dihydroxyvitamin D3 enhances the susceptibility of breast cancer cells to doxorubicininduced oxidative damage, Cancer Res. 59 (4) (1999) 862–867. [7] C.J. Narvaez, B.M. Byrne, S. Romu, M. Valrance, J. Welsh, Induction of apoptosis by 1,25-dihydroxyvitamin D3 in MCF-7 Vitamin D3 -resistant variant can be sensitized by TPA, J. Steroid Biochem. Mol. Biol. 84 (2–3) (2003) 199–209. [8] M. Simboli-Campbell, C.J. Narvaez, M. Tenniswood, J. Welsh, 1,25Dihydroxyvitamin D3 induces morphological and biochemical markers of apoptosis in MCF-7 breast cancer cells, J. Steroid Biochem. Mol. Biol. 58 (4) (1996) 367–376. [9] C.J. Narvaez, J. Welsh, Role of mitochondria and caspases in Vitamin D-mediated apoptosis of MCF-7 breast cancer cells, J. Biol. Chem. 276 (12) (2001) 9101–9107. [10] I.S. Mathiasen, I.N. Sergeev, L. Bastholm, F. Elling, A.W. Norman, M. Jaattela, Calcium and calpain as key mediators of apoptosis-like death induced by Vitamin D compounds in breast cancer cells, J. Biol. Chem. 277 (34) (2002) 30738–30745. [11] C.J. Narvaez, K. Vanweelden, I. Byrne, J. Welsh, Characterization of a Vitamin D3 -resistant MCF-7 cell line, Endocrinology 137 (2) (1996) 400–409. [12] K.S. Chen, H.F. DeLuca, Isolation and characterization of a novel cDNA from HL-60 cells treated with 1,25-dihydroxyvitamin D3 , Biochim. Biophys. Acta 1219 (1) (1994) 26–32. [13] E. Junn, S.H. Han, J.Y. Im, Y. Yang, E.W. Cho, H.D. Um, D.K. Kim, K.W. Lee, P.L. Han, S.G. Rhee, I. Choi, Vitamin D3 up-regulated
[17]
[18]
[19]
[20]
[21]
[23]
[25]
[26]
[27]
[28]
[29]
[30]
protein 1 mediates oxidative stress via suppressing the thioredoxin function, J. Immunol. 164 (12) (2000) 6287–6295. M. Saitoh, H. Nishitoh, M. Fujii, K. Takeda, K. Tobiume, Y. Sawada, M. Kawabata, K. Miyazono, H. Ichijo, Mammalian thioredoxin is a direct inhibitor of apoptosis signal regulating kinase 1 (ASK1), EMBO J. 17 (9) (1998) 2596–2606. Y.M. Janssen, C.K. Sen, Nuclear factor B activity in response to oxidants and antioxidants, Meth. Enzymol. 300 (1999) 363–374. W.H. Watson, D.P. Jones, Oxidation of nuclear thioredoxin during oxidative stress, FEBS Lett. 543 (2003) 144–147. M.R. Fernando, H. Nanri, S. Yoshitake, K. Nagata-Kuno, S. Minakami, Thioredoxin regenerates proteins inactivated by oxidative stress in endothelial cells, Eur. J. Biochem. 209 (3) (1992) 917–922. Y. Wang, G.W. De Keulenaer, R.T. Lee, Vitamin D3 up-regulated protein 1 is a stress-responsive gene that regulates cardiomyocyte viability through interaction with thioredoxin, J. Biol. Chem. 277 (29) (2002) 26496–26500. Y. Chen, J. Cai, T.J. Murphy, D.P. Jones, Overexpressed human mitochondrial thioredoxin confers resistance to oxidant-induced apoptosis in human osteosarcoma cells, J. Biol. Chem. 277 (36) (2002) 33242–33248. S. Karimpour, J. Lou, L.L. Lin, L.M. Rene, L. Lagunas, X. Ma, S. Karra, C.M. Bradbury, S. Markovina, P.C. Goswami, D.R. Spitz, K. Hirota, D.V. Kalvakolanu, J. Yodoi, D. Gius, Thioredoxin reductase regulates AP-1 activity as well as thioredoxin nuclear localization via active cysteines in response to ionizing radiation, Oncogene 21 (41) (2002) 6317–6327. K. Tobiume, A. Matsuzawa, T. Takahashi, H. Nishitoh, K. Morita, K. Takeda, O. Minowa, K. Miyazono, T. Noda, H. Ichijo, ASK1 is required for sustained activations of JNK/p38 MAP kinases and apoptosis, EMBO Rep. 2 (3) (2001) 222–228. I.S. Mathiasen, U. Lademann, M. Jaattela, Apoptosis induced by Vitamin D compounds in breast cancer cells is inhibited by Bcl-2 but does not involve known caspases or p53, Cancer Res. 59 (19) (1999) 4848–4856. S. Swami, N. Raghavachari, U.R. Muller, Y.P. Bao, D. Feldman, Vitamin D growth inhibition of breast cancer cells: gene expression patterns assessed by cDNA microarray, Breast Cancer Res. Treat. 80 (1) (2003) 49–62. H. Liu, H. Nishitoh, H. Ichijo, J.M. Kyriakis, Activation of apoptosis signal regulating kinase 1 (ASK1) by tumor necrosis factor receptorassociated factor 2 requires prior dissociation of the ASK1 inhibitor thioredoxin, Mol. Cell Biol. 20 (6) (2000) 2198–2208. Y. Liu, W. Min, Thioredoxin promotes ASK1 ubiquitination and degradation to inhibit ASK1-mediated apoptosis in a redox activityindependent manner, Circ. Res. 90 (12) (2002) 1259–1266. G.E. Weitsman, A. Ravid, U.A. Liberman, R. Koren, The role of p38 MAP kinase in the synergistic cytotoxic action of calcitriol and TNF␣ in human breast cancer cells, J. Steroid Biochem. Mol. Biol. 89–90 (2004) 361–364. T.F. McGuire, D.L. Trump, C.S. Johnson, Vitamin D3 -induced apoptosis of murine squamous cell carcinoma cells. Selective induction of caspase-dependent MEK cleavage and up-regulation of MEKK-1, J. Biol. Chem. 276 (28) (2001) 26365–26373. A. Nishiyama, H. Masutani, H. Nakamura, Y. Nishinaka, J. Yodoi, Redox regulation by thioredoxin and thioredoxin-binding proteins, IUBMB Life 52 (1–2) (2001) 29–33.
Altered thioredoxin subcellular localization and redox status in MCF-7 cells following 1,25-dihydroxyvitamin D3 treatment Belinda M. Byrne, JoEllen Welsh ∗ Department of Biological Sciences, 214 Galvin Life Sciences Building, University of Notre Dame, Notre Dame, IN 46556, USA
Abstract 1,25-Dihydroxyvitamin D3 (1,25D) induces apoptosis in MCF-7 cells via the intrinsic pathway involving bax translocation to mitochondria, cytochrome c release and reactive oxygen species (ROS) generation. Vitamin D up-regulated protein 1 (VDUP1), an apoptotic regulatory gene induced by 1,25D in HL-60 cells, is a negative regulator of thioredoxin (Trx1), a redox protein which neutralizes ROS and protects against oxidative stress induced apoptosis. Due to the involvement of oxidative stress in 1,25D mediated apoptosis, we analyzed whether VDUP1 or Trx1 are altered by 1,25D in MCF-7 cells. In contrast to HL-60 cells, VDUP1 mRNA was not up-regulated by 1,25D in MCF-7 cells, indicating that transcriptional up-regulation of this gene is not required for 1,25D mediated apoptosis. 1,25D did not affect the expression or activity of Trx1 in MCF-7 cells, however, Trx1 activity was higher in MCF-7 cells selected for resistance to 1,25D mediated apoptosis. In untreated MCF-7 cells, Trx1 was present only in the cytosol, and the majority was in the oxidized state. In 1,25D treated MCF-7 cells, Trx1 was present in both cytosol and nucleus, and the nuclear Trx1 pool was in the reduced state. Nuclear localization of Trx1 in 1,25D treated MCF-7 cells was confirmed by immunofluorescent microscopy. Although redox status is known to alter the ability of Trx1 to bind apoptosis signal regulating kinase 1 (ASK1), no changes in ASK1 transcript or protein levels were observed in 1,25D treated MCF-7 cells. Collectively, these studies indicate that although VDUP1 and ASK1 are not altered by 1,25D, changes in redox status and sub-cellular distribution of Trx1 occurs during 1,25D mediated apoptosis of MCF-7 cells. © 2005 Elsevier Ltd. All rights reserved. Keywords: Vitamin D3 ; VDUP1; Thioredoxin; Breast cancer; Apoptosis
1. Introduction 1,25-Dihydroxyvitamin D3 (1,25D), the biologically active form of Vitamin D3 , is a fat soluble steroid hormone that exerts anti-proliferative and pro-apoptotic effects on breast cancer cells. The ability of 1,25D to inhibit growth has consistently been observed in a variety of breast cancer cell lines, including those derived from human tumors, both in vitro and in vivo [1]. 1,25D exerts its effects through binding the Vitamin D receptor (VDR), a ligand dependent transcription factor. The presence of VDR in breast tumors has been associated with slower growing tumors or less aggressive disease [1] and epidemiological studies have indicated an inverse correlation between breast cancer incidence and progression and Vitamin D status [2–4]. In addition, VDR agonists can potentiate the effects of certain cytotoxic compounds, ∗
Corresponding author. Tel.: +1 574 631 3371; fax: +1 574 631 7413. E-mail address: [email protected] (J. Welsh).
0960-0760/$ – see front matter © 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.jsbmb.2005.06.023
such as tumor necrosis factor ␣ (TNF␣) [5], doxorubicin [6], and the phorbol ester phorbol 12-myristate 13-acetate (TPA) [7]. Our lab initially demonstrated that 1,25D induces morphological, biochemical, and molecular features of apoptosis in MCF-7 human breast cancer cells [8]. 1,25D mediated apoptosis in MCF-7 cells involves a novel caspase independent pathway that triggers mitochondrial disruption [9,10]. To facilitate investigation into the mechanisms of 1,25D induced apoptosis, we generated a 1,25D resistant variant from MCF7 cell line—the MCF-7DR cell line [11]. MCF-7DR cells express the VDR but are selectively resistant to 1,25D, indicating that expression of the VDR is not sufficient for the induction of growth arrest and apoptosis by 1,25D [11]. In the studies reported here, we used MCF-7 and MCF-7DR cells to investigate the role of redox signaling during 1,25D induced apoptosis. Induction of apoptosis in MCF-7 cells by 1,25D is characterized by down regulation of Bcl-2, translocation of Bax to
58
B.M. Byrne, J. Welsh / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 57–64
Fig. 1. Thioredoxin redox cycling. The components of the thioredoxin system include thioredoxin (Trx1), thioredoxin reductase 1 (TrxR1), and NADPH. Trx1 becomes oxidized while reducing free radicals and/or maintaining proteins in their reduced state. NADPH and TrxR1 function to recycle Trx1 back to the reduced (active) form.
mitochondria, redistribution of cytochrome c and generation of reactive oxygen species (ROS) [9]. Time course studies indicated that ROS generation is an early event in 1,25D mediated apoptosis [9]. Neither cytochrome c release, ROS generation nor Bcl-2/Bax alteration occurs in MCF-7DR cells after treatment with 1,25D. MCF-7DR cells, however, are sensitive to TNF␣ mediated apoptosis, indicating that these cells express a functional apoptotic pathway. Since the cytotoxic action of TNF␣ also involves ROS generation and mitochondrial disruption, these data suggest that the MCF-7DR cells have uncoupled VDR signaling, but not TNF␣ signaling, from ROS generation. Vitamin D up-regulated protein 1 (VDUP1) was originally identified as a novel 1,25D induced gene in HL-60 cells [12]. Little is known about the function of VDUP1, although overexpression of VDUP1 induces oxidative stress, possibly via interaction with the redox protein thioredoxin 1 (Trx1) [13]. Trx1 binds and neutralizes oxygen free radicals, including ROS, and thus protects cells against oxidative stress. Binding of ROS oxidizes Trx1, which can be converted back to its reduced form by thioredoxin reductase (Trx1R) in the presence of NADPH (Fig. 1). In the reduced form, Trx1 also prevents apoptosis by binding and inhibiting apoptosis signaling kinase 1 (ASK1). VDUP1 binds directly to the redox active site of Trx1 and inhibits its ability to neutralize ROS as well as its ability to inhibit ASK1 [14]. We therefore hypothesized that accumulation of ROS in 1,25D treated MCF-7 cells could be secondary to up-regulation of VDUP1, inhibition of Trx1 and activation of ASK1. To test this hypothesis, we assessed components of this redox system in relation to ROS generation in 1,25D treated MCF-7 and MCF-7DR cells. Our studies provide evidence that 1,25D alters Trx1 redox state and subcellular localization in MCF-7 cells but not in MCF-7DR cells.
2. Materials and methods 2.1. Cell culture and growth assays Parental MCF-7 and the 1,25D resistant variant MCF-7DR cells [11] were cultured in ␣-MEM medium (Life Technolo-
gies Inc., Gaithersburg, MD) containing 25 mM HEPES and 5% FBS (Life Technologies). Cells were routinely plated at 5000 cells/cm2 and passaged every 3–4 days. For experiments, cells were plated in ␣-MEM containing 5% FBS plus antibiotics, and treatments with 100 nM 1,25D or vehicle control were initiated 24 h after plating. For long term experiments, cells were re-fed with fresh medium containing treatments after 72 h. In some experiments, parallel cultures were treated with TNF␣ (2 ng/ml) 2 days prior to harvest as a positive control for apoptosis. For assessment of growth, 2000 cells were plated per well in 24-well plates and treated 24 h later. After 96 or 120 h, cultures were fixed in 1% gluteraldehyde, stained with 0.1% crystal violet in dH2 O and dried overnight. Crystal violet was resuspended in 0.2% Triton X100 and absorbance at 590 nm was read on a Wallac Victor 2 plate reader. Previous studies have confirmed that absorbance of crystal violet is proportional to adherent cell number under the conditions used. 2.2. Western blotting Cells treated for 96 or 120 h were harvested by direct lysis in Laemlli buffer, separated on 5% (ASK1) or 15% (Trx1) SDS–PAGE and electrophoretically transferred to nitrocellulose paper. Blots were probed with mouse anti-human Trx1 (1:100) (Pharmingen, CA) followed by anti-mouse secondary (1:5000) or anti-human ASK1 (1:500) (Cell Signaling Technology Inc., MA) followed by anti-rabbit secondary (1:3000). Specific antibody binding was visualized by chemiluminescence using products from Pierce (Pierce Biotechnology, IL). 2.3. Immunocytochemistry Cells were grown for 2 days on Lab-Tek II chamber slides (Fisher Scientific) and treated with ethanol vehicle or 100 nM 1,25D for 120 h, or 2 ng/ml TNF␣ for 48 h. The cells were fixed in 4% formaldehyde in PBS, permeabilized in methanol, and blocked overnight with PBS/1% BSA. The slides were incubated with anti-Trx1 mouse monoclonal antibody (1:50, Pharmingen) for 2 h at 37 ◦ C followed by incubation with anti-mouse Alexa 488. Nuclei were counter-stained with 1 g/ml Hoechst for 15 min at room temperature. Fluorescence was detected with a Bio-Rad MRC 1024 scanning confocal (three channel/LaserSharp 2000, version 5.0 program) system, utilizing a krypton–argon (Kr–Ar) laser. 2.4. Assessment of thioredoxin redox status Nuclear and cytosolic fractions were isolated based on the procedure of Janssen and Sen [15], with modifications to allow redox measurement of Trx1 via the redox Western blot. Cells were pelleted by centrifugation and lysed in hypotonic lysis buffer (10 mM HEPES, pH 7.8, 10 mM KCl, 2 mM MgCl2 , 0.1 mM EDTA, 0.2 mM NaF, 0.2 mM Na3 VO4 ·6H2 O) with protease inhibitors (leupeptin, aprotinin, pepstatin, phenylmethylsulfonyl fluoride). For redox
B.M. Byrne, J. Welsh / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 57–64
Western blot, 50 mM iodoacetic acid (IAA) was added to the lysis buffer and the pH was adjusted to 7.8. The suspensions were incubated on ice for 5 min, and Nonidet P40 was added to a final concentration of 0.6%. Following centrifugation at 16,000 × g, the pellet (nuclei) and the supernatant (cytosol) were suspended in 6 M guanidine–HCl, 50 mM Tris, 3 mM EDTA, 0.5% Triton X100 (pH 8.3) supplemented with 50 mM IAA and incubated at 37 ◦ C for 30 min. After removal of excess IAA, Sephadex chromatography (MicroSpin G25 columns, Amersham-Pharmacia, Piscataway, NJ, USA), samples were subjected to native electrophoresis to separate reduced and oxidized forms of Trx1 [16,17]. 2.5. Thioredoxin activity assay Thioredoxin activity was measured with the insulin reducing assay [13]. Briefly, cells were lysed in saline buffer (100 mM Na2 HPO4 , 1 mM EDTA, pH 7.5) using a dounce homogeniser and incubated at 70 ◦ C for 10 min, then DTT activation buffer (50 mM HEPES, 1 mM EDTA, 1 mg/ml BSA, 2 mM DTT) was added for a 15 min incubation at 37 ◦ C. Reactions were initiated by addition of 150 mM HEPES, 1 mM EDTA, 1 mg/ml NADPH, and 2 mg/ml insulin in the presence of thioredoxin reductase 1 (American Diagnostica Inc., CT) or water (for control reactions). After incubation at 37 ◦ C for 15 min, the reactions were stopped by the addition of 0.4 mg/ml DTNB/6M guanidine–HCl in 0.2M Tris/HCl, pH 8.0. Absorbance was determined at 412 nm. 2.6. Quantitative real-time PCR Total RNA was isolated from MCF-7 cells (2 × 106 cells) with the RNeasy Protect Mini Kit (Qiagen). After concentration and purity was determined by spectrophotometry, total RNA was used for reverse transcription reactions using TaqMan Reverse Transcription Reagents (Applied Biosystems, CA). For each cell line, reactions were performed in duplicate, generating two 1.5 ng cDNA stocks/sample. Each of the duplicate cDNA stocks were independently analyzed in duplicate (60 ng of cDNA/well) by real time PCR using the SYBR Green PCR Master Mix (Applied Biosystems) and a primer set specific for human VDUP1 (Forward primer: AGATCAGGTCTAAGCAGCAGAACA, Reverse primer: TCAGATCTACCCAACTCATCTCAGA) or ASK1 (Forward primer: CCAGGGCCTCTTCCTGCTA, Reverse primer: TCTGCTCCTCCCAGAGGATTT). Transcript values were normalized against GAPDH RNA (Forward primer: CCACCCATGGCAAATTCC, Reverse primer: TGATGGGATTTCCATTGATGAC) and reported as normalized gene expression. 2.7. Flow cytometric determination of intracellular ROS Both adherent and floating cells were collected, pelleted, resuspended, and incubated in the presence of 4 M hydroethidine (HE, Molecular Probes) as previously
59
described [9]. Analysis of HE to ethidium was analyzed on an Epics XL Flow Cytometer (Coulter Corp., Miami, FL) equipped with an argon laser. HE was analyzed on FL3 using a 620 nm band pass filter. Data was analyzed using MULTIPLUS AV analysis software (Phoenix Flow Systems). 2.8. Statistical analyses Data are expressed as mean ± S.E. One-way ANOVA was used to assess statistical significance between means. Differences between means were considered significant if P-values less than 0.05 were obtained with the Dunnett’s method using GraphPad Prism software (GraphPad Software, San Diego, CA).
3. Results 3.1. Comparative effects of 1,25D and TNFα on morphology and ROS generation in MCF-7 and MCF-7DR cells The effects of 1,25D and TNF␣ on morphology of MCF-7 and MCF-7DR cultures are shown in Fig. 2A. MCF-7 cells exposed to ethanol vehicle grow in a typical epithelial cobblestone pattern. After 120 h treatment with 100 nM 1,25D, MCF-7 cultures show evidence of apoptosis, including retraction of cell-to-cell contacts, cytoplasmic condensation and loss of adherence. The MCF-7DR cells, which were selected for growth in the presence of 100 nM 1,25D, do not demonstrate any morphological evidence of apoptosis in response to 1,25D. However, both cell lines exhibit typical apoptotic morphology after treatment with 2 ng/ml TNF␣. In all cases, apoptotic morphology correlated with a reduction in the number of adherent cells as measured by crystal violet staining (Fig. 2B). Thus, cell density was significantly reduced by 1,25D in MCF-7 cultures and by TNF␣ in MCF-7 and MCF-7DR cultures. TNF␣ induced a comparable reduction in adherent cell number in both MCF-7 and MCF-7DR cell lines. Apoptotic morphology and reduced cell density also correlated with generation of reactive oxygen species (ROS) as measured by flow cytometry (Fig. 3). The percentage of cells with intracellular ROS accumulation was over 30% in 1,25D treated MCF-7 cultures, but negligible in 1,25D treated MCF-7DR cultures. TNF␣ induced ROS accumulation in both MCF-7 and MCF-7DR cultures. These data indicate that apoptosis triggered by either 1,25D or TNF␣ is associated with ROS generation in MCF-7 cells, and that loss of 1,25D sensitivity in MCF-7DR cells does not alter sensitivity to TNF␣. 3.2. 1,25D does not alter VDUP1 expression in MCF-7 or MCF-7DR cells Since VDUP1, a VDR target gene in HL-60 leukemic cells, has been linked to induction of oxidative stress
60
B.M. Byrne, J. Welsh / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 57–64
Fig. 2. Effects of 1,25D and TNF␣ on MCF-7 and MCF-7DR cells. (A) MCF-7 and MCF-7DR cells were treated with ethanol, 100 nM 1,25D (120 h) or 2 ng/ml TNF␣ (48 h) and morphology was imaged by phase contrast microscopy. (B) Cells were treated as in A and relative adherent cell density was determined by crystal violet growth assay. Data are mean ± S.E. * Statistically significant from control value, Anova.
[13,18], we assessed whether steady state mRNA expression of VDUP1 is induced by 1,25D during MCF-7 cell apoptosis. MCF-7 and MCF-7DR cells were treated for 3, 6, 12, 24 or 48 h with vehicle control or 100 nM 1,25D, and VDUP1 was assessed by quantitative real time PCR (Fig. 4). Although VDUP1 mRNA increased from
3 to 12 h after treatment, there were no significant differences in VDUP1 transcript levels between control and 1,25D treated cells at any time point in either MCF-7 (Fig. 4A) or MCF-7DR (Fig. 4B) cells. These data indicate that VDUP1 is not a VDR regulated gene in MCF-7 cells.
Fig. 3. 1,25D induces ROS production in MCF-7 but not MCF-7DR cells. MCF-7 and MCF-7DR cells were treated for 120 h with ethanol, 100 nM 1,25D (120 h) or 2 ng/ml TNF␣ (48 h) and ROS production was analyzed as conversion of hydroethidine to ethidium by flow cytometry. Shaded peaks indicate cells positive for ROS, percentages are shown in upper right corner.
B.M. Byrne, J. Welsh / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 57–64
61
Fig. 4. VDUP1 expression in 1,25D treated MCF-7 and MCF-7DR cells. MCF-7 (A) and MCF-7DR (B) cells were treated with ethanol or 1,25D for 3, 6, 12, 24 or 48 h and VDUP1 transcript levels were measured by real time PCR. Data are mean ± S.E.
3.3. Effects of 1,25D on Trx1 expression, redox status and activity The lack of effect of 1,25D on VDUP1 prompted us to examine whether 1,25D alters expression or activity of Trx1, a redox protein that neutralizes intracellular ROS and protects cells against oxidative stress induced apoptosis [19]. In MCF7 cells, expression of Trx1, measured by quantitative real time PCR, was stable over 48 h in culture and was not affected by 100 nM 1,25D (Fig. 5A). Furthermore, Western blotting indicated that neither 1,25D nor TNF␣ altered Trx1 protein expression in MCF-7 or MCF-7DR cells (Fig. 5B). However, Trx1 expression was elevated in MCF-7DR cells compared to MCF-7 cells. Since Trx1 expression may not reflect its activity, we used an insulin reduction assay to assess Trx1 reducing activity in both cell lines under basal and 1,25D treated conditions. No significant difference was seen between control and 1,25D treated MCF-7 or MCF-7DR cells, but thioredoxin activity was higher in MCF-7DR cells (Fig. 5C), consistent with the higher Trx1 protein expression in this cell line. To confirm the Trx1 activity assay results, we used a redox Western blot technique to distinguish between oxidized and reduced pools of Trx1 in nuclear and cytoplasmic fractions of MCF-7 cells. In control MCF-7 cells, Trx1 was only detected in the cytosol, and although both oxidized and reduced forms of Trx1 were present, the majority was in the oxidized state (Fig. 6). In 1,25D treated MCF-7 cells, cytosolic Trx1 was markedly reduced compared to control cells, and only the
Fig. 5. Trx1 expression and function in MCF-7 and MCF-7DR cells. (A) MCF-7 cells were treated with ethanol or 1,25D for 3, 6, 12, 24 or 48 h and Trx1 mRNA was measured by real time PCR. Data are mean ± S.E. (B) MCF-7 cells were treated with ethanol or 1,25D for 120 h or TNF␣ for 48 h, and Trx1 protein was measured by Western blotting. (C) MCF-7 or MCF7DR cells were treated with ethanol or 1,25D for 120 h and Trx1 activity was measured by insulin reduction assay as described in Section 2. Statistical significance was assessed by Anova.
oxidized form was present. In contrast to control MCF-7 cells, in which Trx1 was only present in the cytosol, Trx1 was detected in the nucleus of 1,25D treated cells, and this pool of Trx1 was exclusively in the reduced form. TNF␣ treatment did not alter Trx1 redox state or localization in MCF-7 cells. These data indicated that 1,25D alters the redox status of Trx1, and suggested subcellular relocalization of the protein as well. 3.4. 1,25D induces nuclear translocation of Trx1 in MCF-7 cells Nuclear redistribution of Trx1 has previously been reported during radiation induced apoptosis [20]. In light of the results of the redox blot showing nuclear Trx1 in 1,25D treated but not vehicle treated MCF-7 cells, we used
62
B.M. Byrne, J. Welsh / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 57–64
Fig. 6. Effect of 1,25D on Trx1 redox status in subcellular fractions. Subcellular fractions from MCF-7 cells treated with ethanol or 1,25D for 120 h were separated on native polyacrylamide gels and probed for Trx1. Arrows indicate positions of oxidized (top) and reduced (bottom) forms of Trx1. Native marker is indicated by 14.
immunofluorescence to localize Trx1 in MCF-7 cells. In control cells (Fig. 7, top panels), faint diffuse staining of Trx1 was detected in the cytosol, and no co-localization of Trx1 with the nuclear Hoechst stain was observed in merged images. In MCF-7 cells treated with 100 nM 1,25D for 120 h (Fig. 7, bottom panels), intense, punctuate staining of Trx1 was detected in early apoptotic cells. Trx1 staining colocalized with condensed chromatin visualized with Hoechst staining in merged images. This altered staining in response to 1,25D was not observed in MCF-7DR cells (data not shown). 3.5. 1,25D does not alter ASK1 mRNA or protein expression in either MCF-7 or MCF-7DR cells Apoptosis signaling kinase 1 (ASK1) has been shown to be essential for some forms of oxidative stress induced apoptosis [21], and reduced forms of Trx1 bind to and inhibit ASK1 activity [14]. We therefore used real-time PCR and Western blotting to determine whether ASK1 expression was altered
by 1,25D. As shown in Fig. 8A, ASK1 mRNA expression in MCF-7 cells was not altered by time in culture or treatment with 1,25D. Furthermore, no changes in ASK1 protein levels were observed in response to 1,25D or TNF␣ treatment in MCF-7 or MCF-7DR cell line (Fig. 8B).
4. Discussion These studies have confirmed and extended previous work demonstrating a role for oxidative stress in 1,25D mediated apoptosis of breast cancer cells [6,9]. MCF-7 cell death triggered by either 1,25D or the cytokine TNF␣ was associated with accumulation of ROS, a characteristic of apoptosis mediated by the mitochondrial (intrinsic) pathway. This finding is consistent with work demonstrating changes in the Bcl2/Bax ratio, mitochondrial permeability transition and cytochrome c release in 1,25D treated MCF-7 cells [9,23]. These alterations subsequently activate executioner caspases, leading to
Fig. 7. Translocation of Trx1 to nucleus in 1,25D treated MCF-7 cells. MCF-7 cells were treated with ethanol or 1,25D for 120 h, fixed, permeabilized, and probed with antibody directed against Trx1, followed by anti-mouse secondary antibody conjugated to Alexa 488. Nuclei were counterstained with Hoechst dye and cells were visualized on a biorad fluorescent microscope.
B.M. Byrne, J. Welsh / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 57–64
Fig. 8. Expression of ASK1 mRNA and protein in 1,25D treated MCF-7 cells. (A) MCF-7 cells were treated with ethanol or 1,25D for 3, 6, 12, 24 or 48 h and ASK1 transcript levels were measured by real time PCR. Data are mean ± S.E. (B) MCF-7 cells were treated with ethanol or 1,25D for 120 h or TNF␣ for 48 h and ASK1 protein was assessed by Western blotting.
cytoplasmic and nuclear condensation and irreversible DNA fragmentation [8]. The major goal of the experiments described here was to determine the impact of 1,25D on proteins involved in cellular oxidative stress responses. We specifically examined VDUP1, a 1,25D target gene in HL-60 cells [12] and Trx1, a VDUP1 interacting protein that protects cells from oxidative stress [13]. VDUP1 has been shown to inhibit the cytoprotective activity of Trx1 [13], and Trx1 reductase, an enzyme that recycles Trx1, has been identified as a 1,25D target gene by microarray profiling [25]. These data suggested that VDUP1 or Trx1 might be functionally involved in triggering oxidative stress in response to 1,25D in MCF-7 cells. In contrast to HL-60 cells, however, our data indicate that VDUP1 is not a 1,25D regulated gene in MCF-7 cells, thus changes in VDUP1 gene expression are unlikely to contribute to 1,25D mediated ROS generation or apoptosis. Although there were no changes in Trx1 transcript or protein levels in 1,25D treated MCF-7 cells, Trx1 subcellular localization and redox status was altered by 1,25D. Increased nuclear staining of Trx1, and accumulation of the reduced form of Trx1 in nuclear fractions detected by redox Western blotting, was observed in MCF-7 cells following 1,25D treatment. Similar nuclear localization in the absence of changes in total protein expression has been reported in HeLa cells induced to undergo apoptosis by ionizing radiation [20]. These data suggest that nuclear translocation of Trx1 may be a hallmark of oxidative stress in general rather than a specific marker of 1,25D mediated apoptosis.
63
In addition to VDUP1 and Trx1, we assessed the effects of 1,25D on ASK1, a mitogen activated protein (MAP) kinase, which promotes apoptosis through the JNK/p38 pathways [26]. The rationale for investigation of ASK1 was based on reports that Trx1 binds to and inhibits ASK1 [14,27], an interaction that is disrupted during oxidative stress. We therefore considered that ROS generation during 1,25D mediated apoptosis might alter Trx1–ASK1 interaction, promoting apoptosis through the JNK/p38 pathways. Although evidence for 1,25D regulation of JNK/p38 signaling has been provided in MCF-7 cells [28] and in squamous carcinoma cells [29], we found no effect of 1,25D on either ASK1 gene expression or protein levels. However, further studies will be necessary to determine if 1,25D alters ASK1 activity or JNK/p38 signaling in relation to apoptosis induction in MCF7 cells. As a redox protein, Trx1 can exist in both oxidized and reduced forms, and its interactions with other proteins, including VDUP1 and ASK1, are dependent on its redox status [14,30]. In control MCF-7 cells, both oxidized and reduced forms of Trx1 were detected, but only in the cytoplasmic fraction. In 1,25D treated cells, Trx1 was present in the nucleus in the reduced form, whereas cytosolic Trx1 was predominantly in the oxidized form. Since the oxidized form of Trx1 is unable to bind and inhibit ASK1, these data warrant further study to measure ASK1 activity during 1,25D treated MCF-7 cell apoptosis. Although both 1,25D and TNF␣ induce apoptosis via ROS generation and mitochondrial disruption in MCF-7 cells, no changes in Trx1 redox state or subcellular localization were observed in TNF␣ treated MCF-7 cells. These data indicate that apoptosis induced by 1,25D may be mechanistically different from that of TNF␣. In support of this concept, the MCF-7DR cells, which were selected for resistance to 1,25D, retain sensitivity to TNF␣ mediated apoptosis. These studies also provide insight into potential mechanisms of 1,25D resistance. Although neither Trx1 expression nor activity was altered by 1,25D, there were higher levels of Trx1 expression and activity in MCF-7DR cells than in parental MCF-7 cells. Thus, Trx1 overexpression may contribute to the resistant phenotype of MCF-7DR cells by neutralizing free radicals to protect against oxidative stress induced by 1,25D. In support of this suggestion, ROS accumulation was negligible in MCF-7DR cells treated with 1,25D under conditions, when >30% of parental MCF-7 cells were ROS positive. Also consistent with this concept, ROS generation in response to TNF␣ was blunted in MCF-7DR cells compared to parental MCF-7 cells. In summary, these studies indicate that VDUP1 is not a 1,25D regulated gene in MCF-7 cells, but confirm that 1,25D mediated apoptosis in MCF-7 cells is associated with oxidative stress. In addition to ROS generation, changes in Trx1 subcellular localization and redox state are associated with 1,25D mediated apoptosis in MCF-7 cells. Furthermore, upregulation of cellular anti-oxidant defenses, such as Trx1 may contribute to 1,25D resistance in breast cancer cells.
64
B.M. Byrne, J. Welsh / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 57–64
Acknowledgements This work was supported by Department of Defense Breast Cancer Research Program Grant DAMD17-03-10358. We thank Dr. Judy Narvaez for technical assistance on the flow cytometer.
[14]
[15] [16]
References [1] K.W. Colston, U. Berger, R.C. Coombes, Possible role for Vitamin D in controlling breast cancer cell proliferation, Lancet 1 (8631) (1989) 188–191. [2] J. Welsh, J.A. Wietzke, G.M. Zinser, B. Byrne, K. Smith, C.J. Narvaez, Vitamin D3 receptor as a target for breast cancer prevention, J. Nutr. 133 (Suppl. 7) (2003) 2425S–2433S. [3] E.M. John, G.G. Schwartz, D.M. Dreon, J. Koo, Vitamin D and breast cancer risk: the NHANES I Epidemiologic follow-up study, 1971–1975 to 1992. National Health and Nutrition Examination Survey, Cancer Epidemiol. Biomar. Prev. 8 (5) (1999) 399–406. [4] E. Mawer, B.J. Walls, A. Howell, M. Davies, W.A. Ratcliffe, N.J. Bundred, Serum 1,25-dihydroxyvitamin D may be related inversely to disease activity in breast cancer patients with bone metastases, J. Clin. Endocrinol. Metab. 82 (1997) 118–122. [5] D. Rocker, A. Ravid, U.A. Liberman, O. Garach-Jehoshua, R. Koren, 1,25-Dihydroxyvitamin D3 potentiates the cytotoxic effect of TNF on human breast cancer cells, Mol. Cell Endocrinol. 106 (1–2) (1994) 157–162. [6] A. Ravid, D. Rocker, A. Machlenkin, C. Rotem, A. Hochman, G. Kessler-Icekson, U.A. Liberman, R. Koren, 1,25-Dihydroxyvitamin D3 enhances the susceptibility of breast cancer cells to doxorubicininduced oxidative damage, Cancer Res. 59 (4) (1999) 862–867. [7] C.J. Narvaez, B.M. Byrne, S. Romu, M. Valrance, J. Welsh, Induction of apoptosis by 1,25-dihydroxyvitamin D3 in MCF-7 Vitamin D3 -resistant variant can be sensitized by TPA, J. Steroid Biochem. Mol. Biol. 84 (2–3) (2003) 199–209. [8] M. Simboli-Campbell, C.J. Narvaez, M. Tenniswood, J. Welsh, 1,25Dihydroxyvitamin D3 induces morphological and biochemical markers of apoptosis in MCF-7 breast cancer cells, J. Steroid Biochem. Mol. Biol. 58 (4) (1996) 367–376. [9] C.J. Narvaez, J. Welsh, Role of mitochondria and caspases in Vitamin D-mediated apoptosis of MCF-7 breast cancer cells, J. Biol. Chem. 276 (12) (2001) 9101–9107. [10] I.S. Mathiasen, I.N. Sergeev, L. Bastholm, F. Elling, A.W. Norman, M. Jaattela, Calcium and calpain as key mediators of apoptosis-like death induced by Vitamin D compounds in breast cancer cells, J. Biol. Chem. 277 (34) (2002) 30738–30745. [11] C.J. Narvaez, K. Vanweelden, I. Byrne, J. Welsh, Characterization of a Vitamin D3 -resistant MCF-7 cell line, Endocrinology 137 (2) (1996) 400–409. [12] K.S. Chen, H.F. DeLuca, Isolation and characterization of a novel cDNA from HL-60 cells treated with 1,25-dihydroxyvitamin D3 , Biochim. Biophys. Acta 1219 (1) (1994) 26–32. [13] E. Junn, S.H. Han, J.Y. Im, Y. Yang, E.W. Cho, H.D. Um, D.K. Kim, K.W. Lee, P.L. Han, S.G. Rhee, I. Choi, Vitamin D3 up-regulated
[17]
[18]
[19]
[20]
[21]
[23]
[25]
[26]
[27]
[28]
[29]
[30]
protein 1 mediates oxidative stress via suppressing the thioredoxin function, J. Immunol. 164 (12) (2000) 6287–6295. M. Saitoh, H. Nishitoh, M. Fujii, K. Takeda, K. Tobiume, Y. Sawada, M. Kawabata, K. Miyazono, H. Ichijo, Mammalian thioredoxin is a direct inhibitor of apoptosis signal regulating kinase 1 (ASK1), EMBO J. 17 (9) (1998) 2596–2606. Y.M. Janssen, C.K. Sen, Nuclear factor B activity in response to oxidants and antioxidants, Meth. Enzymol. 300 (1999) 363–374. W.H. Watson, D.P. Jones, Oxidation of nuclear thioredoxin during oxidative stress, FEBS Lett. 543 (2003) 144–147. M.R. Fernando, H. Nanri, S. Yoshitake, K. Nagata-Kuno, S. Minakami, Thioredoxin regenerates proteins inactivated by oxidative stress in endothelial cells, Eur. J. Biochem. 209 (3) (1992) 917–922. Y. Wang, G.W. De Keulenaer, R.T. Lee, Vitamin D3 up-regulated protein 1 is a stress-responsive gene that regulates cardiomyocyte viability through interaction with thioredoxin, J. Biol. Chem. 277 (29) (2002) 26496–26500. Y. Chen, J. Cai, T.J. Murphy, D.P. Jones, Overexpressed human mitochondrial thioredoxin confers resistance to oxidant-induced apoptosis in human osteosarcoma cells, J. Biol. Chem. 277 (36) (2002) 33242–33248. S. Karimpour, J. Lou, L.L. Lin, L.M. Rene, L. Lagunas, X. Ma, S. Karra, C.M. Bradbury, S. Markovina, P.C. Goswami, D.R. Spitz, K. Hirota, D.V. Kalvakolanu, J. Yodoi, D. Gius, Thioredoxin reductase regulates AP-1 activity as well as thioredoxin nuclear localization via active cysteines in response to ionizing radiation, Oncogene 21 (41) (2002) 6317–6327. K. Tobiume, A. Matsuzawa, T. Takahashi, H. Nishitoh, K. Morita, K. Takeda, O. Minowa, K. Miyazono, T. Noda, H. Ichijo, ASK1 is required for sustained activations of JNK/p38 MAP kinases and apoptosis, EMBO Rep. 2 (3) (2001) 222–228. I.S. Mathiasen, U. Lademann, M. Jaattela, Apoptosis induced by Vitamin D compounds in breast cancer cells is inhibited by Bcl-2 but does not involve known caspases or p53, Cancer Res. 59 (19) (1999) 4848–4856. S. Swami, N. Raghavachari, U.R. Muller, Y.P. Bao, D. Feldman, Vitamin D growth inhibition of breast cancer cells: gene expression patterns assessed by cDNA microarray, Breast Cancer Res. Treat. 80 (1) (2003) 49–62. H. Liu, H. Nishitoh, H. Ichijo, J.M. Kyriakis, Activation of apoptosis signal regulating kinase 1 (ASK1) by tumor necrosis factor receptorassociated factor 2 requires prior dissociation of the ASK1 inhibitor thioredoxin, Mol. Cell Biol. 20 (6) (2000) 2198–2208. Y. Liu, W. Min, Thioredoxin promotes ASK1 ubiquitination and degradation to inhibit ASK1-mediated apoptosis in a redox activityindependent manner, Circ. Res. 90 (12) (2002) 1259–1266. G.E. Weitsman, A. Ravid, U.A. Liberman, R. Koren, The role of p38 MAP kinase in the synergistic cytotoxic action of calcitriol and TNF␣ in human breast cancer cells, J. Steroid Biochem. Mol. Biol. 89–90 (2004) 361–364. T.F. McGuire, D.L. Trump, C.S. Johnson, Vitamin D3 -induced apoptosis of murine squamous cell carcinoma cells. Selective induction of caspase-dependent MEK cleavage and up-regulation of MEKK-1, J. Biol. Chem. 276 (28) (2001) 26365–26373. A. Nishiyama, H. Masutani, H. Nakamura, Y. Nishinaka, J. Yodoi, Redox regulation by thioredoxin and thioredoxin-binding proteins, IUBMB Life 52 (1–2) (2001) 29–33.