Abstract
Background: Vitamin D analog, 1α-hydroxy-24-ethyl-cholecalciferol (1α(OH)D5), is a less toxic VDR agonist that suppresses proliferation of breast cancer cells in vitro and in vivo. The present study assessed 1α(OH)D5-mediated regulation of VDR, and its potential anti-estrogenic activity in BT-474 cells. Materials and Methods: The mRNA and protein expression of steroid receptors were determined using RT-PCR and Western blot analyses, respectively. Results: VDR mRNA was up-regulated (180% of control) by 1α(OH)D5 within seven hours, whereas the expression of VDR protein increased by two-fold in 24 hours. This increase was abolished in presence of either actinomycin D or cyclohexamide. Additionally, there was a four-fold decrease in ERα mRNA and 40% decrease in ERα protein after 28 and 48 hours following 1α(OH)D5 treatment, respectively. Down-regulation of some of the estrogen-inducible genes was observed. Conclusion: Although no VDR stabilization by 1α(OH)D5 was observed, there was an increased expression of the VDR followed by partial anti-estrogenic activity in hormone-responsive BT-474 cells.
- Breast cancer
- vitamin D analogs
- estrogen receptor
- BT-474 cells
- vitamin D receptor stability
The biologically active form of vitamin D3, 1α,25(OH)2D3 or calcitriol, exerts cellular effects mainly through vitamin D receptor (VDR). The VDR belongs to the steroid receptor superfamily of ligand-activated transcription factors (1, 2). In the classic model of vitamin D3 action, the lipophilic 1α,25(OH)2D3 crosses cellular membranes and enters the nucleus, where it binds to VDR to activate the receptor which can then heterodimerize with retinoid X receptor (RXR). This 1α,25(OH)2D3-VDR-RXR complex binds to various vitamin D response elements (VDREs) in the genome, thereby modulating expression of a variety of genes involved in calcium and phosphate metabolism, and cellular proliferation as well as differentiation (3, 4). Formation of 1α,25(OH)2D3-VDR complex has also been shown to induce VDR transcription in target epithelial cells (5). In addition, ligand binding was shown to stabilize the VDR by preventing its ubiquitin-mediated degradation (6). In rat osteosarcoma cells, ligand binding to VDR resulted in increased protein levels due to inhibition of proteosomal degradation (7, 8). The ligand-bound VDR levels increased rapidly and continued to rise for 24 hours in the rat osteosarcoma cells.
Due to the toxicity associated with the active metabolite of vitamin D3 in vivo, numerous analogs of vitamin D3 including 1α-hydroxy24-ethyl cholecalciferol (1α(OH)D5) have been developed as selective VDR agonists (9). Previous studies have shown chemopreventive efficacy of 1α(OH)D5 in experimental models. Both carcinogenesis models used in these studies were estrogen responsive and the tumors were estrogen receptor positive (10, 11). A few studies have suggested a cross talk between 1α,25(OH)2D3 and estrogen signaling pathway (12, 13). Treatment with 1α,25(OH)2D3 has been shown to down-regulate ERα mRNA level in breast cancer cells, preventing estrogen-induced cellular proliferation. While estrogen via ERα is known to induce proliferation, progesterone that down-regulates its own receptor is associated with cellular differentiation (14). Since growth inhibitory effects of 1α(OH)D5 are more pronounced in hormone-responsive breast cancer cells, it is likely that estrogen signaling is involved. In this report the effects of 1α(OH)D5 on the expression and stabilization of VDR and its influence on expression of ERα and estrogen inducible genes were evaluated in BT-474 breast cancer cells.
Materials and Methods
Vitamin D3 analog. 1α(OH)D5 was synthesized as described previously (15) and stored at -80°C as a stock solution of 10 mM 1α(OH)D5 in ethanol. Throughout the study, the cells were incubated with 1 μM final concentration of 1α(OH)D5, found to be the optimal dose in previous studies (15). The appropriate controls for each experiment consisted of treatment with the vehicle ethanol.
Breast cancer cells. The human breast cancer cell line BT-474 was purchased from American Type Culture Collection (ATCC, Rockville, MD, USA) and maintained in Minimum Essential Medium with Earle's salts containing 5% heat inactivated fetal bovine serum at 37°C in 5% CO2. All experiments on these cells were performed in the regular culture conditions and all cell culture chemicals were purchased from Invitrogen Inc. (Carlsbad, CA, USA).
Regulation of VDR transcription in breast cancer cells. BT-474 cells were treated with 1α(OH)D5 for various time points to extract RNA for expression studies. Each experiment was run in duplicate and results were adjusted to the housekeeping gene GAPDH. In order to determine if the change in VDR level was due to increased transcription or increased stabilization of VDR mRNA, the control and treated BT-474 cells were incubated with culture media containing RNA synthesis inhibitor, actinomycin D (ActD, 2 μg/mL). Total RNA was extracted from these cells and subjected to semi-quantitative and real time RT-PCR for assessment of VDR levels using gene specific primers. The comparison of changes in VDR levels between ActD-treated and untreated samples were used to assess the proportion of change due to de novo transcription or stabilization of the receptor.
Regulation of VDR translation in breast cancer cells. BT-474 cells were treated with 1α(OH)D5 for different time points and total proteins were extracted from control and treated cells. Western blots were used for determination of VDR expression. In order to determine if the change in VDR level was due to stabilization or de novo synthesis of the receptor, the protein synthesis inhibitor cycloheximide (CHX, 2 μg/mL) was used in the culture media in a second set of samples. The control and 1α(OH)D5-treated cells were lysed to extract proteins and subjected to Western blot analysis for VDR levels. Each experiment was run in duplicate and results were adjusted to the housekeeping gene β-actin.
Antiestrogenic action of 1α(OH)D5. To explore potential anti-estrogenic activities of 1α(OH)D5 in hormone-responsive breast cancer BT-474 cells, steady state mRNA levels for ERα were determined using semi-quantitative RT-PCR in control and treated samples taken at different time points. Each experiment was run in duplicates and results were adjusted to the housekeeping gene GAPDH. Changes in ERα expression were determined using Western blot analysis of protein extracts from control and treated BT-474 cells. Anti-estrogenic effects of 1α(OH)D5 were determined by using estrogen-inducible genes as markers of estrogenic activity. The control and treated BT-474 cells were subjected to RT-PCR to assess mRNA levels of PR, pS2 and cathepsin D, which are transcriptionally up-regulated by ERα. Moreover, the expression of genes that are down-regulated by estrogen-mediated signaling, E-cadherin and caspase-2, was also studied.
RNA extraction and RT-PCR. Total RNA was extracted using Clontech's Nucleospin RNA II kit (Clontech, BD Biosciences, Palo Alto, CA, USA) according to the manufacturer's instructions. To study effects of 1α(OH)D5 on expression of target genes, RT-PCR was performed with 0.5 μg of total RNA from control and treated samples using the Superscript™ One-Step RT-PCR system with Platinum® Taq (Invitrogen Corporation) according to the manufacturer's instructions. The reaction mix was subjected to RT-PCR using the Perkin Elmer DNA Thermal Cycler 480 (Perkin-Elmer Corporation, CT, USA). The real-time RT-PCR was performed using SYBR Green PCR Master Mix and RT Taqman system (Applied Biosystems, Foster City, CA, USA) according to the instruction manual. Primer Express software (Applied Biosystems) was used to design primers for real time PCR. Each sample was run in duplicates and adjusted to corresponding control before being averaged. The data were expressed as a percentage of controls.
Protein extraction and Western blot analysis. Cells were grown in 100 mm cell culture dishes and proteins were extracted from control and treated plates at the end of specific time points. Total proteins from cells were extracted using a protein lysis buffer containing HEPES (pH 7.9, 20 mM), NaCl (400 mM), Nonidet P-40 (0.1%), glycerol (10%), Na vanadate (1 mM) and Na fluoride (1 mM) with a cocktail of protease inhibitors. The primary antibodies were purchased from Neomarkers (Lab Vision Corp., Fremont, CA, USA). Protein separation was performed using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The proteins were visualized by electro-chemiluminescence (ECL, Amersham Biosciences Corp., Piscataway, NJ, USA) on the photographic film (Kodak X-Omat film; Fisher Scientific, Pittsburgh, PA,USA). The protein bands were scanned and band density was determined using Kodak 1D Image Analysis Software version 3.5 (Kodak Digital Science Imaging, Eastman Kodak Company, New Haven, CT, USA). Each expression study was repeated twice and the data were normalized for housekeeping gene β-actin and plotted using XY scatter plot.
Results
The modulation of gene expression by 1α,25(OH)2D3 is mediated mainly by binding to and stabilizing the VDR. Ligand binding to VDR increases levels of both VDR mRNA and protein. In order to understand the extent to which 1α(OH)D5 is capable of regulating the VDR, the transcription and expression of VDR were studied using gene specific RT-PCR and Western blot analysis for protein expression studies in BT-474 cells.
Induction of VDR mRNA upon 1α(OH)D5 treatment: Steady state levels of VDR mRNA were found to be up-regulated by 1α(OH)D5 treatment of BT-474 cells within a few hours of treatment compared to control (Figure 1A). By five to seven hours, the VDR mRNA levels reached 180% of the control, followed by a gradual decline to baseline by 24 hours of treatment. Real-time RT-PCR showed that mRNA levels of VDR peaked at six hours following treatment with 1α(OH)D5 with a four-fold increase in the levels of VDR mRNA (Figure 1B). The VDR mRNA levels started to decline after eight hours, and almost all of the induced VDR mRNA returned to baseline within 20 hours.
VDR mRNA levels upon 1α(OH)D5 treatment in the presence of RNA synthesis inhibitor actinomycin D: In order to determine whether the increase in VDR mRNA levels was due to an increase in transcription or merely a reduction in degradation, the RNA synthesis inhibitor actinomycin D was used. Prevention of degradation of VDR mRNA would lead to accumulation in the treated cells. However, the results showed that in the presence of ActD, the levels of VDR mRNA from treated BT-474 cells were similar to control, indicating that the increased VDR mRNA in actinomycin D-negative samples was a result of an increase in transcription of the VDR (Figure 1C and D).
VDR expression after 1α(OH)D5 treatment: In order to determine the time course of VDR protein levels, BT-474 cells were treated for various time points with 1α(OH)D5. Proteins were extracted and separated to determine expression of VDR levels using Western blots. Within a short period of six hours, the VDR levels were increased in treated samples by 40% as compared to the control (Figure 2A). There was a 2.5-fold increase in the VDR at 24 hours of 1α(OH)D5 treatment as compared to the control in the BT-474 cells (Figure 2B).
VDR expression upon 1α(OH)D5 treatment in the presence of protein synthesis blocker cycloheximide: Since the increased VDR expression is not specifically indicative of an increase in protein synthesis or a decrease in degradation, protein synthesis inhibitor CHX was used to inhibit new protein synthesis. Both the 1α(OH)D5 treated and control BT-474 cells were incubated with CHX (2 μg/mL) and samples were extracted at various time points to determine the levels of VDR by Western blots. Results indicated that there was no significant difference between the VDR levels of control and treated cells at any time point (Figure 2C and 2D). At 16 hours, the VDR level from treated cells was approximately 15% higher than control. However, by 24 hours there was no difference in VDR levels between control and the treated cells that had been incubated with CHX. This suggests that only a small proportion of the increase in VDR expression may be a result of stabilization of the receptor.
VDR expression upon 1α(OH)D5 treatment in the presence of RNA synthesis blocker actinomycin D: In order to determine whether new RNA synthesis is necessary for increased VDR expression, the BT-474 cells were treated with 1α(OH)D5 in the presence of ActD and sampled at different intervals. The Western blot analyses showed that upon treatment with ActD, no significant increase in expression of VDR was detected in BT-474 cells (Data not shown). These results indicated that not only translation, but also transcription was necessary for VDR expression to increase in BT-474 cells.
Effects of 1α(OH)D5 treatment on expression of estrogen-inducible genes: Breast cancer cell lines that showed growth inhibition upon treatment with 1α(OH)D5 in previous experiments, were not only VDR positive but also ERα positive and estrogen responsive (16). Therefore, the effect of 1α(OH)D5 on estrogenic activity was determined by measuring the expression of estrogen-inducible genes (ERα, PR, pS2 and cathepsin D) as markers of estrogenic activity. In addition, expression of E-cadherin and caspase-2 was assessed. Both E-cadherin and caspase-2 expression have been reported to increase in response to estrogen withdrawal (Data not shown).
As measured by gen-specific RT-PCR, the ERα mRNA levels decreased in the BT-474 cells by 40% after 40 hours of 1α(OH)D5 treatment (Figure 3A and 3B). However, the down-regulation of ERα mRNA by 1α(OH)D5 treatment was abolished in presence of the RNA synthesis inhibitor ActD, suggesting that VDR-mediated transcription was necessary to reduce levels of ERα mRNA (Results not shown).
The ERα protein expression was evaluated after treatment with 1α(OH)D5 using Western blots. Results showed that ERα expression decreased by 40% in BT-474 cells 48 hours following 1α(OH)D5 treatment (Figure 3C and 3D).
In order to determine whether the reduction in ERα level would lead to reduced estrogenic activity, estrogen-inducible genes PR, pS2 and cathepsin D were measured as markers of estrogenic activity. Both PR as well as pS2 mRNA and protein levels were down-regulated by 1α(OH)D5 treatment. The decrease in PR mRNA was 50% by 48 hours, beginning at 30 hours following 1α(OH)D5 treatment, whereas the protein expression was reduced by 40% by day 4 after 1α(OH)D5 treatment. Similarly, reduced expression of both pS2 mRNA and protein were observed at 48 hours and four days after treatment with 1α(OH)D5 (Results not shown).
Discussion
The growth inhibitory actions of 1α,25(OH)2D3 are attributed to its ability to bind to VDR and modulate gene expression. In order to understand the extent to which 1α(OH)D5 is capable of regulating VDR, the transcription and protein expression of VDR were studied in 1α(OH)D5-treated BT-474 cells. The results showed that 1α(OH)D5 up-regulates VDR at both transcriptional and translational levels by about two-fold. The VDR levels increased two and half fold in 2 hours and eight fold by 18 hours, while no significant change in transcription of VDR upon 1α,25(OH)2D3 treatment was observed. Vitamin D3 analog KH1060 has been shown to possess more potent biological activity than 1α,25(OH)2D3; it increases VDR stability by reducing proteolytic degradation (17). KH1060-mediated VDR stabilization resulted in increased VDR half-life (15 hours) compared to 1α,25(OH)2D3-liganded VDR half-life of 8-10 hours (17).
Previously, 1α(OH)D5 was shown to induce transcription of VDR in T-47D breast cancer cells (18). In order to determine whether the increase in VDR mRNA levels was due to an increase in transcription or a reduction in degradation of VDR, the RNA synthesis inhibitor actinomycin D was used in this study. In presence of the inhibitor, the VDR mRNA levels were not significantly different from those of the control, indicating that the increase in levels was a result of an increase in de novo synthesis of the VDR. Similarly, blocking protein synthesis with use of CHX did not significantly alter the levels of VDR in treated samples as compared to controls. The lack of VDR stabilizing activity of 1α(OH)D5 could be due to its lower binding affinity to VDR compared to 1α,25(OH)2D3. Regardless of its reduced receptor binding capacity, 1α(OH)D5 has previously been shown to be effective in reducing cancer cell growth in vitro as well as in vivo. Thus, receptor affinity and stability are not the only factors that would determine biological activity (19, 20). Receptor conformation can affect DNA binding as well as heterodimerization with other factors that would lead to differential gene transactivation (21). However, no structure-function studies have yet been performed to determine if the VDR-1α(OH)D5 complex could assume a unique conformation that would alter VDR-mediated transcription. In addition to structural dynamics of ligand-receptor complex, levels of 1α,25(OH)2D3 have been known to play a role in its biphasic effects in various cell lines.
ER deregulation plays a critical role in breast cancer development and progression. Targeting ERα with selective estrogen receptor modulators (SERMs) has achieved significant reduction in breast cancer progression in 40-50% of the women with hormone-responsive tumors (22). The breast cancer cell lines responsive to vitamin D3 analog are not only VDR positive but are largely ERα positive and estrogen responsive. Hence, the effects of 1α(OH)D5 on ERα expression and estrogenic activity were determined. The most widely used markers of estrogenic activity in breast cancer cells include PR, pS2 and cathepsin D (23). Therefore the effects of 1α(OH)D5 on the expression of these estrogen-inducible genes in BT474 cells was determined. Generally, ERα levels increase upon exposure to estrogens (24) with subsequent increase in estrogenic activity. The down-regulation of ERα mRNA by 1α(OH)D5 treatment was inhibited in presence of the RNA synthesis inhibitor ActD. Similarly, ERα expression decreased following 1α(OH)D5 treatment and this decrease was abolished in presence of protein synthesis inhibitor CHX. In another study, 1α,25(OH)2D3 treatment reduced ERα mRNA level and estrogenic activity in MCF-7 cells (13). This decrease in mRNA was not abolished in presence of CHX. The down-regulation of ERα by 1α,25(OH)2D3 in MCF-7 cells was reported by Lee et al. (12), who concluded that a decrease in ERα transcription could be the result of possible negative VDRE in the ERα promoter. While no conclusive studies about the presence of VDRE in the ERα promoter have been performed, reports of anti-estrogenic activity of 1α,25(OH)2D3 have been published (25). When vitamin D3 analog, EB1089, was combined with anti-estrogen ICI-164-384, its antiproliferative ability was increased several fold (26). Likewise, 1α,25(OH)2D3 treatment in combination with tamoxifen in MCF-7 and ZR-75-1 cells showed additive effects in inhibiting growth of breast cancer cells.
The reduction in ERα levels translate to reduced estrogenic activity, which was determined by assessing expression of estrogen-inducible genes (PR, pS2 and cathepsin D). PR, at both mRNA and protein levels, was down-regulated by 1α(OH)D5 treatment. Another estrogen-inducible gene, pS2 is expressed in normal breast epithelium but its expression increases in breast cancer. About 68% of hormone responsive tumors overexpress pS2, which has been used as a biomarker of well-differentiated tumors (27) that would respond to endocrine therapy. In addition to PR, pS2 was also down-regulated by 1α(OH)D5 treatment in BT-474 cells (data not shown). Cathepsin D levels, on the other hand, showed no alteration upon 1α(OH)D5 treatment of BT-474 cells. Thus, 1α(OH)D5 exerted a partial anti-estrogenic activity in BT-474 breast cancer cells which could be related to its growth inhibitory action.
E-cadherin is a glycoprotein involved in cell to matrix interactions and is a potent suppressor of invasion of breast tumors. Loss of E-cadherin is prevalent in about 60-70% of primary breast cancer (28). Interestingly, estrogen-stimulated growth is known to reduce E-cadherin expression in breast cancer cells which can be reversed by the use of anti-estrogens (29). Since E-cadherin expression was increased in the presence of 1α(OH)D5 (data not shown), its effects seem similar to estrogen withdrawal. Similarly, caspase-2, which is partially up-regulated by withdrawal of estradiol, was up-regulated in response to 1α(OH)D5 treatment in BT-474 cells. This indicated that the apoptosis induced by 1α(OH)D5 treatment might be associated with estrogen withdrawal in addition to the direct effect of 1α(OH)D5 on the apoptosis machinery.
In conclusion, the current study shows that the 1α(OH)D5-induced increase in transcription and translation of VDR is due mainly to de novo synthesis, not receptor stabilization. 1α(OH)D5 also exhibited partial anti-estrogenic activity by reducing expression of ERα and estrogen-inducible genes such as PR and pS2 within 30-48 hours of treatment resulting in growth arrest at 72 hours. In addition, 1α(OH)D5 treatment in hormone-responsive BT-474 cells resulted in up-regulation of genes that have been reported to be suppressed by estrogens. These events are schematically presented in Figure 4. Thus, the growth inhibitory effects of 1α(OH)D5 appear to be associated with its ability to modify steroid receptor expression in BT-474 cells in addition to its direct VDR-mediated effects.
Acknowledgements
These studies were supported by generous funding from the Department of Defense (DAMD17-01-1-0272) and the National Cancer Institute (CA R01-82361).
- Received January 29, 2009.
- Revision received March 10, 2009.
- Accepted May 14, 2009.
- Copyright© 2009 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved