Vitamin D analogs are effective inhibitors of breast cancer cell growth, but many breast cancer cell lines show various degrees of resistance to the growth inhibitory effect of vitamin D. In this study, we investigated the mechanism of 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] resistance of the human breast epithelial cell line HBL100, which had been immortalized by Simian virus 40 (SV40) large T antigen. We determined the expression, DNA binding and transactivation activity of vitamin D3 receptor (VDR) in HBL100 and a vitamin D-sensitive ZR75-1 breast cancer cell line. Western blot analysis revealed a comparable expression of VDR gene in both cell lines. However, gel retardation assays demonstrated nuclear proteins from ZR75-1 cells but not from HBL100; cells expressed a 9-fold increase in the binding activity with a vitamin D response element (VDRE). Using a transient transfection assay, we showed that the VDRE was activated by 8-fold in ZR75-1. However, in HBL100 cells there was no activation observed in response to 1,25(OH)2D3. On the other hand, co-transfection of a VDR expression vector could restore 1,25(OH)2D3-induced VDRE transcription in HBL100 cells. Moreover, stable expression of VDR in HBL100 cells resulted in enhanced sensitivity of the cells to the growth inhibitory effect of 1,25(OH)2D3. Since CV-1 cells express very little endogenous VDR, the interactions of VDR and large T antigen were carried out in these cells. By transient co-transfection, we observed that expression of the large T antigen strongly inhibited 1,25(OH)2D3-induced VDRE transcriptional activity in a dose-dependent fashion in CV-1 cells. At 120 ng VDR concentration, the inhibition was completely reversed. Thus the loss of the growth inhibitory effect of vitamin D3 in HBL100 cells may be caused by the expression of the large T antigen in the cells, and provide further evidence that VDR is required for efficient growth inhibition by vitamin D3.
Vitamin D and analogs are known to inhibit the proliferation of cultured breast cancer cells and to cause regression of experimental mammary tumors in vivo, due to their ability to induce differentiation and/or apoptosis of breast epithelial cells (1–6). The molecular mechanism by which vitamin D and its analogs inhibit the growth of cancer cells is currently the subject of intensive research. There is evidence that the hormonally active form of vitamin D, 1,25-dihydroxyvitamin D3 [1,25(OH)2D3], can generate biological responses through its transcriptional regulation of genes involved in cell growth and differentiation. They include transforming growth factor β (7), epidermal growth factor receptor (8), c-myc (9), insulin-like growth factor-binding protein (10), cell cycle regulators p21 and p27 (11) and cell survival factor Bcl-2 (12). Transcriptional activity of 1,25(OH)2D3 is mainly mediated by vitamin D receptors (VDR) expressed in target cells (6). VDR belong to a large superfamily of steroid/thyroid receptors that function as ligand-dependent transcriptional factors by binding to a set of specific DNA sequences on target genes (13–15). Vitamin D response elements (VDRE) have been identified in several genes and consist of a direct repeat of the hexanucleotide sequences (GGGTGA) separated by a 3 nucleotide spacer (16,17). VDR bind to VDRE either as homodimers or as heterodimers with retinoid X receptor (RXR), i.e. VDR/RXR (18). Transcriptional activation by nuclear hormone receptors requires their interaction with co-activators that appear to provide a direct link of nuclear receptors to the core transcriptional machinery and to modulate chromatin structure (19–27). Several receptor co-activators have been identified so far. One of them, the CREB-binding protein (CBP), functions not only as nuclear receptor co-activator but also as a co-activator of many other transcriptional factors (19–27). Results from previous studies have demonstrated a general correlation between the levels of VDR present and the degree of differentiation of breast cancer cells (6,28,29). In addition, loss of VDR may be responsible for decreased anti-proliferative effect of vitamin D3 in cancer cells (6,28,29). Moreover, VDR polymorphism has been shown to be associated with increased incidence of a variety of diseases (30). Thus, VDR can mediate the biological action of vitamin D3 and may be involved in cancer development.
Although a functional VDR is necessary for the growth regulatory effect of vitamin D3, 1,25(OH)2D3 does not always show growth inhibitory effect on human breast cancer cell lines, despite their expression of VDR (6,31). Typically, both ER+ (MCF-7 and ZR75–1) and ER– breast cancer cell lines (MCF-10neo, BCA-4, SK-BR-3) show growth inhibitory response to vitamin D analogs. Yet some breast cancer cell lines have shown resistance to vitamin D analogs regardless of their steroid receptor status. These include MDA-MB-231, MDA-MB 436, MCF-7-D-resistant and HBL100 cells (6,32,33). The mechanisms by which resistance to vitamin D3 occurs are largely unknown. One possibility may be a non-functional state of VDR in certain cells.
In the present study we have focused on evaluating the mechanism of resistance to the active metabolite of vitamin D, 1,25(OH)2D3, in HBL100 cells. The human breast epithelial cell line HBL100, originally derived from the milk secretion of a nursing mother without detectable breast lesion, has been extensively characterized as immortal but not tumorigenic in nude mice due to the presence of Simian virus 40 (SV40) large T antigen (34). The SV40 large T antigen is known to contribute to virus-induced tumorigenesis by interacting with and altering the function of key cellular regulatory proteins, such as members of the retinoblastoma gene family (35), members of the CBP family of transcriptional co-activator proteins (36,37) and p53 tumor suppressor (38,39). Unlike many other breast cancer cell lines, HBL100 cells do not show a clear response to the growth inhibitory effect of 1,25(OH)2D3. In addition to HBL100 cells, we selected ER+, ZR75-1 breast cancer cells as a positive control to establish clear resistance to effects of 1,25-(OH)2D3 in HBL100 cells. Our results demonstrate that 1,25(OH)2D3 resistance in HBL100 cells may be attributed to the expression of SV40 large T antigen in these cells and provide evidence for the role of functional VDR in mediating growth inhibition by 1,25(OH)2D3.
Materials and methods
The breast epithelial cell lines HBL100 and ZR75-1 were obtained from the American Type Culture Collection (Rockville, MD). HBL100 cells were maintained in minimal essential medium with Earl's salts (MEME) medium supplemented with 10% fetal bovine serum (FBS). The ZR75-1 cells were maintained in RPMI 1640 supplemented with 10% FBS.
Growth inhibition assay
For growth inhibition studies, cells were seeded in 96-well plates at a density of 1000 cells/well, in 100 μl/well in a medium containing 10% steroid-stripped serum. After 24 h, cells were incubated with 1 and 10 nM concentrations of 1,25(OH)2D3. An aliquot of 50 μl of medium with fresh 1,25(OH)2D3 was added to each well after 3 days of culture. The viable cell number was determined after 7 days of cultures using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) proliferation/cytotoxicity assay kit (Promega, Madison, WI). The assay measures the capacity of cells to convert a tetrazolium salt to a blue formazan (40). Results obtained were confirmed by cell counting with a hemocytometer.
Western blot analysis
Cells were lysed in a buffer containing 10 mM Tris–HCl (pH 7.4), 150 mM NaCl, 5 mM EDTA, 1% Triton X-100 and 1 μg/ml each PMSF, aproteinin, leupeptin and pepstatin. Equal amounts of lysates were boiled in sodium dodecyl sulfate (SDS) sample buffer. Proteins were separated by SDS–polyacrylamide gel electrophoresis (SDS–PAGE) using NOVEX pre-cast 4–20% gels. After transfer, the nitrocellulose membrane was blocked with 1% milk TBST [50 mM Tris–HCl (pH 7.4), 150 mM NaCl, 0.1% Tween 20], and subsequently incubated for 2 h with anti-VDR polyclonal antibody (Biogenex, San Ramos, CA). The membrane was then washed with TBST and incubated for 1 h at room temperature in TBST containing horseradish peroxidase-linked anti-immunoglobulin. After three washes with TBST the protein was detected with an enhanced chemiluminescence.
Preparation of nuclear protein extracts and gel retardation assay
Nuclear extracts were prepared essentially according to the method described previously (41). Briefly, following the vitamin D treatment HBL100 and ZR75-1 cells were washed with ice-cold phosphate-buffered saline and collected using a rubber policeman. The cells were pelleted by low-speed centrifugation and resuspended in 1 ml buffer A (10 mM Tris–HCl, pH 7.4, 3 mM CaCl2 and 2 mM MgCl2) containing 1% NP40 and homogenized using ice-cold Downce homogenizer. Nuclei were collected by low-speed centrifugation and washed with buffer B [10 mM HEPES–KOH (pH 7.9), 1.5 mM MgCl2, 10 mM KCl and 0.5 mM dithiothreitol (DTT)]. Nuclear proteins were extracted with 200 μl of high salt buffer [20 mM HEPES–KOH (pH 7.9), 25% glycerol, 420 mM NaCl, 1.5 mM MgCl2, 0.2 mM EDTA and 0.5 mM DTT]. All buffers contained protease inhibitors PMSF (100 μg/ml), leupeptin (1 μg/ml) and aproteinin (1 μg/ml). To study VDRE binding, nuclear extracts were analyzed by gel retardation assay using 32P-labeled VDRE as a probe as described previously (42). For comparison, in vitro synthesized VDR and RXRα proteins (42) were used. The VDRE used in the experiments was derived from the osteocalcin gene promoter (TGGGTGAATGAGGACATTACTGACC) (17). Labeled DNA probe was purified by gel electrophoresis and used for the gel retardation assay.
Transient transfection assay
The reporter VDRE-tk-CAT was obtained by inserting one copy of VDRE oligonucleotide into the BamHI site of the pBLCAT2 as described previously (43). For transient transfection assay in HBL100 and ZR75-1 cells, 5×105 cells were seeded in six-well culture plates. A modified calcium phosphate precipitation procedure was used for transfection as described previously (43). Briefly, 250 ng of VDRE-tk-CAT reporter plasmid, 250 ng of β-galactosidase (β-gal) expression vector (pcH110; Pharmacia, Piscataway, NJ) and various amounts of VDR and/or RXRα expression vector were mixed with carrier DNA (pBluescript) to 2.5 μg of total DNA per well. For CV-1 cells, 1×105 cells/well were seeded in 24-well plates. For transfection, 100 ng of reporter plasmid, 100 ng of β-gal expression vector and expression vectors for VDR, RXRα and large T antigen were mixed with carrier DNA to 1 μg of total DNA/well. CAT activity was normalized for transfection efficiency by the corresponding β-gal activity. Large T antigen expression vector was constructed by cloning the large T antigen cDNA into a eukaryotic expression vector pECE (43).
To construct VDR expression vector, cDNA for the VDR gene was cloned into the pRc/CMV expression vector (Invitrogen, San Diego, CA) as described previously (42). The resulting recombinant construct was then stably transfected into HBL100 cells by the calcium phosphate precipitation method and screened with G418 (0.5 mg/ml) (Gibco BRL, Grand Island, NY).
We investigated the growth inhibitory effect of 1,25(OH)2D3 on HBL100 cells by the MTT assay. For comparison, ER+, ZR75-1 breast cancer cells, which are vitamin D3 responsive cells, were used. 1,25(OH)2D3 did not show any growth inhibitory effect at either 1 or 10 nM concentration of 1,25(OH)2D3 on HBL100 cells. In contrast, ~30 and 60% inhibition was observed in ZR75-1 cells with 1 or 10 nM of vitamin D3, respectively (Figure 1). Thus, HBL100 cells are resistant to the growth inhibitory effect of 1,25(OH)2D3.
Since the effect of vitamin D3 is mainly mediated by VDR, we examined whether the loss of 1,25(OH)2D3 effect in HBL100 cells was due to a low expression level of VDR. Cellular proteins were prepared from both HBL100 and ZR75-1 cells and examined for VDR expression by western blot analysis. The results demonstrated that the relative levels of VDR expression in the absence or presence of 10 nM 1,25(OH)2D3 were comparable in both cell lines (Figure 2).
To analyze whether VDR expressed in HBL100 cells is functional, we investigated VDRE-binding activity of nuclear proteins prepared from untreated or vitamin D3-treated HBL100 and ZR75-1 cells. For comparison, in vitro synthesized VDR and RXRα proteins were used. Nuclear proteins from ZR75-1 cells formed strong complexes with the VDRE; the binding was further enhanced by 9-fold when ZR75-1 cells were treated with 1,25(OH)2D3. In contrast, nuclear proteins of untreated HBL100 cells did not show any detectable binding complex on VDRE and only a weak complex was formed when HBL100 cells were treated with 1,25(OH)2D3 (Figure 3). The VDRE transactivation activity in both HBL100 and ZR75-1 cell lines was determined by the use of CAT reporter gene containing VDRE (VDRE-tk-CAT). When the reporter gene was transfected into HBL100 cells, transcription of the reporter gene was barely induced when cells were treated with either 1 nM or 10 nM of 1,25(OH)2D3. However, ~10-fold increase of reporter activity was observed in ZR75-1 cells when the cells were treated with 1 or 10 nM of 1,25(OH)2D3 (Figure 4). The results indicating that VDR could not bind and transactivate VDRE reporter in HBL100 cells (Figures 3 and 4) suggest that the loss of these activities may be due to low expression levels of VDR or RXRα. To confirm this hypothesis, we co-transfected VDR and/or RXRα expression vectors together with the reporter gene into HBL100 and ZR75-1 cells. As shown in Figure 5, co-transfection of 100 ng of RXRα expression vector did not show any effect on the reporter activity in HBL100 cells either in the absence or presence of 1,25(OH)2D3. In contrast, co-transfection of 100 ng of VDR expression vector strongly induced the reporter activity by 7- to 8-fold in response to 1,25(OH)2D3. The addition of RXRα did not show any enhancing effect on co-transfected VDR activity. These data suggest that loss of VDRE transcriptional activity in HBL100 cells is probably due to a low level of functional VDR. When the effect of transfected VDR or RXRα was evaluated in ZR75-1 cells, we did not observe any effect of RXRα and only a slight induction of 1,25(OH)2D3-induced reporter activity (<2-fold) was observed in the presence of VDR (Figure 5).
The above data suggest that loss of the growth inhibitory effect of 1,25(OH)2D3 in HBL100 cells may be due to an impaired VDR activity in these cells, and that exogenous transfected VDR may restore 1,25(OH)2D3 activity. We therefore stably transfected VDR into HBL100 cells. A stable clone (HBL100T) that expressed exogenous VDR was analyzed for the growth inhibitory effect of 1,25(OH)2D3. As shown in Figure 6, at 1 and 10 nM, 1,25(OH)2D3 induced ~35 and 40% growth inhibition in HBL100 and VDR cells, respectively, whereas it did not show any inhibitory effect on the growth of the parental HBL100 cells under the same conditions. Thus, exogenous transfected VDR could restore the 1,25(OH)2D3 sensitivity in the HBL100 cells.
One of the unique characteristics of HBL100 cells is the presence of SV40 large T antigen (34), which is known to contribute to tumorigenesis by binding and altering the function of several important cellular regulators (35–39). The interaction of large T antigen with CBP (36,37) is of particular interest, since CBP is a co-activator of several nuclear receptors (20,21). Interaction with CBP may affect receptor transactivation activity. We, therefore, investigated the possible mechanism of 1,25(OH)2D3 resistance of HBL100 cells by studying the effect of SV40 large T antigen on transcriptional activity of VDR. We selected CV-1 cells for these experiments. The reason is that CV-1 cells lack VDR. Use of HBL100 cells would be ideal; however, these cells do contain non-functional VDR and may interfere with the interpretation of the results. Therefore, CV-1 cells were transfected with the VDRE-tk-CAT together with VDR and RXRα expression vectors. Treatment of the cells with 1,25(OH)2D3 strongly induced the reporter transcriptional activity (Figure 7). However, when large T antigen expression vector was co-transfected, the 1,25(OH)2D3-induced reporter activity was inhibited in a concentration-dependent manner (Figure 7A). Thus, large T antigen can negatively regulate 1,25(OH)2D3 on VDRE. Co-transfection of 50 ng large T antigen almost completely abolished the induced reporter CAT activity. Interestingly, co-transfection of additional VDR could reverse the inhibitory effect of large T antigen (Figure 7B). These results are consistent with the stable transfection results shown in Figure 6, which indicates that overexpression of VDR in HBL100 cells could restore growth inhibition by 1,25(OH)2D3.
In this study, we investigated the possible mechanism of 1,25(OH)2D3 resistance of HBL100 cells to the growth inhibitory effect of 1,25(OH)2D3. Our results demonstrated that VDR, although well expressed in HBL100 cells, could not effectively bind and transactivate the VDRE. In addition, we found that the loss of VDRE transactivation and growth inhibitory effect of 1,25(OH)2D3 could be restored by expression of an exogenous VDR. Furthermore, our data suggest that the loss of VDR function in HBL100 cells may be due to the presence of SV40 large T antigen.
Western blot analysis indicates that the expression level of the VDR gene in HBL100 cells was comparable to that in 1,25(OH)2D3-sensitive ZR75-1 cells. In addition, the level of VDR could be up-regulated by 1,25(OH)2D3 similar to that observed in ZR75-1 cells (Figure 2). However, VDR expressed in ZR75-1 cells effectively bound to and transactivated a VDRE, whereas the VDRE-binding and transactivation activities were impaired in HBL100 cells (Figures 3 and 4). Our observation that 1,25(OH)2D3 could induce VDR expression in HBL100 cells (Figure 2) suggests that certain aspects of VDR function are normal. Thus, it is likely that the growth inhibitory effect of VDR is specifically diminished in HBL100 cells due to the non-functional state of VDR in these cells. Previous studies have also demonstrated a partial vitamin D3 resistance in MCF-7 breast cancer (31) and leukemia cell lines (45). In both cases, cells selected for resistance to growth inhibition by 1,25(OH)2D3 express VDR. In the case of the leukemia cells, induction of 24-hydroxylase by 1,25(OH)2D3 is retained (45), whereas up-regulation of VDR by vitamin D3 occurs in vitamin D3-resistant MCF-7 cells (31). Thus, different mechanisms may be utilized by VDR to regulate genes involved in growth control and genes involved in other functions, and they appear to be dissociated. Such dissociated biological functions and the underlying regulatory mechanisms are supported by our recent identification of a novel vitamin D3 analog, 1α(OH)D5, that exerts prominent anti-proliferative effect against cancer cells, but has reduced effect on serum calcium levels (46). Such an analog may induce VDR conformation, which only allows regulation of genes involved in growth inhibition. Thus, our study provides another piece of evidence supporting the concept that the anti-proliferative effect of VDR can be separated from other VDR-mediated functions, such as the hypercalcemic effect. Similar dissociation of the anti-proliferative effect from other hormone-dependent functions also occurs for vitamin A receptor in which its anti-AP-1 activity can be separated from its receptor transactivation function (44).
Our observation that VDR expressed in HBL100 cells could not bind (Figure 3) and transactivate (Figure 4) the VDRE suggests that VDR function is altered in HBL100 cells. Efficient DNA binding and transactivation of VDR requires heterodimerization of VDR with RXRα (13–15). Based on our observation that co-transfection of VDR, but not RXR, in HBL100 cells could restore 1,25(OH)2D3 transactivation activity (Figure 5), the loss of VDRE transactivation and growth inhibition by 1,25(OH)2D3 is unlikely to be due to a low level of RXR, but likely to be due to a low level of functional VDR. However, the western blot analysis (Figure 2) shows a comparable VDR expression level in both HBL100 and 1,25(OH)2D3-sensitive ZR75-1 cell lines. This observation suggests that loss of VDR function may be attributed to either a mutation in VDR gene or expression of a VDR inhibitor in HBL100 cells. Such a mutation in VDR, however, may only affect its binding and transactivation of the VDRE as well as its regulation of genes involved in growth control, whereas the mutation does not alter its ability to up-regulate VDR expression.
Whether HBL100 cells express a mutated VDR remains to be elucidated. We have investigated the possibility that a VDR inhibitor is present in the cells. HBL100 cells are well known as immortal due to the presence of functional SV40 large T antigen (36). It is, therefore, likely that the large T antigen may also play a role in the development of 1,25(OH)2D3 resistance. A previous study has shown that stable expression of SV40 large T antigen in the brown adipocyte cell line is responsible for insulin resistance, without affecting insulin receptor expression (47). When we investigated the effect of large T antigen on VDR function, we observed that VDR-mediated VDRE transactivation was strongly inhibited by large T antigen in a concentration dependent manner (Figure 7A). Recent studies have demonstrated that the large T antigen may contribute to virus-induced tumorigenesis by targeting key cellular regulatory proteins, such as three members of the retinoblastoma family, pRb, p107 and p130, the tumor suppressor p53 and members of the CBP family of transcriptional adapter proteins (37–41). While interaction with pRb and p53 may render cells resistant to the anti-proliferative effect of 1,25(OH)2D3, interaction with CBP could directly affect VDR DNA binding and transactivation. CBP is known to act as a co-activator of a number of nuclear hormone receptors and is required for their efficient DNA binding and transactivation (36,37). Thus, it is conceivable that inhibition of VDR transactivation activity by large T antigen (Figure 7A) may be due to its interaction with CBP, resulting in decreased availability of CBP for VDR to bind and activate the VDRE. Alternatively, large T antigen may directly interact with VDR and inhibit its DNA-binding and transactivation activity. The fact that co-transfection of additional VDR could abolish the inhibitory effect of large T antigen (Figure 7B) suggests that the inhibitory effect of the large T antigen may be reversible or that the level of large T antigen was limiting. This is also consistent with the stable transfection results, which showed that overexpression of VDR could restore growth inhibition by 1,25(OH)2D3 (Figure 6). Interestingly, the inhibitory effect of the large T antigen could be also observed with retinoic acid receptor (data not shown), indicating that the large T antigen could be a general inhibitor of a number of nuclear receptors.
Our observation that stable expression of VDR restored 1,25(OH)2D3 sensitivity in HBL100 cells provides a possibility for overcoming the eventual viral block or possible VDR mutation by transfecting a functional VDR into HBL100 cells. The finding also provides a piece of direct evidence to support the concept that the growth inhibitory effect of 1,25(OH)2D3 is in part mediated by the VDR. It remains to be determined if the transfected VDR activates alternative pathways to induce growth inhibition through bypassing the cell cycle checkpoints blocked by the virus. Alternatively, it could be hypothesized that stable expression of VDR may be able to overcome the blocking capacity of the virus by simply increasing the expression levels of the virus-controlled cellular regulatory protein. Our observation that VDR and the large T antigen inhibit each other's activity suggests that VDR may exert its anti-proliferative effect through its interaction with the large T antigen or similar proteins, thereby enhancing the biological function of p53 and pRb.
Anissa Agadir is supported by a post-doctoral fellowship from the Breast Cancer Research Program (BCRP, University of California). This work was supported in part by US Army Medical Research Program grant DAMDA 4440 (X.-K.Z.) and Illinois Department of Public Health Breast Cancer Program (R.M.).
- transcription, genetic
- western blotting
- viral tumor antigens
- cell line
- nuclear proteins
- vitamin d3 receptor
- simian virus 40
- trans-activation (genetics)
- vitamin d response element
- vitamin d
- epithelial cells
- breast cancer cell
- expression vector
- restore 1 trial