Regulation of the human Vitamin D3 receptor promoter in breast cancer cells is mediated through Sp1 sites

Molecular and Cellular Endocrinology 230 (2005) 59–68

Regulation of the human Vitamin D3 receptor promoter in breast cancer cells is mediated through Sp1 sites Jennifer A. Wietzkea,b , Erin C. Warda , John Schneidera , JoEllen Welsha,∗ a

Department of Biological Sciences, University of Notre Dame, 214 Galvin Life Science Building, Notre Dame, IN 46556, USA b Department of Medicine, Division of Endocrinology, Northwestern University, Tarry 15th Floor, Room 762, 303 E. Chicago Ave., Chicago, IL 60611, USA Received 3 August 2004; received in revised form 28 October 2004; accepted 1 November 2004

Abstract 1,25-Dihydroxyvitamin D3 (1,25(OH)2 D3 ), the active form of Vitamin D, mediates gene transcription through the Vitamin D receptor (VDR), a nuclear receptor expressed in multiple normal and transformed cell types. In mammary epithelial cells, including those derived from breast cancers, 1,25(OH)2 D3 induces growth arrest and/or apoptosis through VDR dependent mechanisms, and VDR agonists represent potential therapeutic agents for hyperproliferative diseases, including cancer. Since target cell sensitivity to 1,25(OH)2 D3 and its analogs reflects VDR expression, understanding the transcriptional regulation of the VDR gene is fundamental to development of VDR agonists as therapeutic agents. The studies reported here focused on molecular characterization of the promoter region upstream of exon 1c in the human VDR gene. In transient transfection assays, luciferase reporter constructs containing −800 to +31 of the VDR gene exhibit basal promoter activity in T47D breast cancer cells which is enhanced by 1,25(OH)2 D3 , estrogen and the phytoestrogen resveratrol. Deletion constructs and site-directed mutagenesis were used to map three distinct GC-rich Sp1 consensus sites that independently mediate the effects of estrogen, resveratrol, and 1,25(OH)2 D3 on VDR promoter activity. Up-regulation of the VDR promoter by 1,25(OH)2 D3 was mapped to an Sp1 site 261 bp upstream of exon 1c, estrogen responsiveness to a proximal Sp1 site beginning at −50, and resveratrol regulation to a distal Sp1 site beginning at −381. Studies with estrogen receptor (ER) subtype specific ligands suggest that the effect of estrogen on VDR promoter is dependent on both ER␣ and ER␤, whereas the effect of resveratrol is dependent only on ER␣. In summary, these studies demonstrate transcriptional regulation of the exon 1c VDR promoter in breast cancer cells, and identify three distinct GC-rich, Sp1 consensus sites that differentially confer responsiveness to estrogen, resveratrol and 1,25(OH)2 D3 . © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Vitamin D3 ; Vitamin D3 receptor (VDR); Sp-1; Estrogen; Resveratrol; Breast cancer

1. Introduction 1,25-Dihydroxyvitamin D3 (1,25(OH)2 D3 ), the biologically active form of Vitamin D3 , exerts negative growth regulatory effects on breast cancer cells in vitro and in vivo (VanWeelden et al., 1998; Zinser et al., 2003). Synthetic Vitamin D3 analogs, which retain growth inhibitory effects with minimal calcemic side effects, represent potential therapeutics for hyperproliferative disorders (Mørk Hansen et al., 2001). The growth suppressive actions of 1,25(OH)2 D3 ∗

Corresponding author. Tel.: +1 574 631 3371; fax: +1 574 631 7413. E-mail address: [email protected] (J. Welsh).

0303-7207/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.mce.2004.11.001

and its synthetic analogs are mediated through the Vitamin D receptor (VDR), a member of the nuclear receptor superfamily which acts as a ligand dependent transcription factor (Zinser et al., 2003). Though it is clear that cellular sensitivity to 1,25(OH)2 D3 is proportional to expression of the VDR, surprisingly few studies have described the human VDR promoter region or examined its molecular regulation. Miyamoto et al. (1997) and Crofts et al. (1998) provided the first evidence for multiple VDR transcripts originating from putative promoters upstream of exons 1f, 1e, 1a and 1d. Tissueand tumor-specific expression of distinct VDR transcripts has been described (Crofts et al., 1998; Correa et al., 2002), although factors that regulate VDR promoter usage remain

60

J.A. Wietzke et al. / Molecular and Cellular Endocrinology 230 (2005) 59–68

to be characterized. The VDR promoter upstream of exon 1f has been reported to contain sequences responsive to a number of transcription factors, including WT1, Cdx-2, and CREB (Yamamoto et al., 1999; Huening et al., 2002; Lee and Pelletier, 2001). However, while exon 1f and exon 1a promoter sequences fused to reporter genes display constitutive activity in a variety of cell lines (Miyamoto et al., 1997; Crofts et al., 1998), regulation of these promoters in the context of cellular signaling pathways has yet to be demonstrated. In contrast, a region centered around exon 1c (proximal to the exon 1a and 1f promoter regions) was shown to be retinoic acid responsive in osteoblasts (Miyamoto et al., 1997). Our studies have focused on the 5 -flanking region of exon 1c, which has been shown to drive reporter gene activity in MCF7 and T47D breast cancer cells (Byrne et al., 2000; Wietzke and Welsh, 2003). Furthermore, in contrast to the VDR promoter regions upstream of exons 1f, 1e, 1a or 1d, the promoter upstream of exon 1c displays hormonal regulation in intact cells (Byrne et al., 2000; Wietzke and Welsh, 2003). Thus, 1,25(OH)2 D3 , retinoic acid, estrogen and phytoestrogens (such as resveratrol and genistein) up-regulate exon 1c promoter activity in breast cancer cells. The sequence of the exon 1c hVDR promoter contains no consensus Vitamin D or estrogen response elements, however, several GC-rich, Sp1 consensus sites are present. In the studies described here, we demonstrate a functional role for these Sp1 binding sites in hormonal regulation of the human VDR exon 1c promoter. Three distinct GC-rich Sp1 consensus sites in the exon 1c promoter were identified that independently confer responsiveness to 1,25(OH)2 D3 , estrogen and resveratrol in breast cancer cells. These data support the hypothesis that modulation of Sp1 transcription factor activity contributes to molecular regulation of the human VDR promoter.

2.2. hVDR promoter vectors and transient transfections T47D or SUM159PT cells (2 × 105 per well) were plated in six-well dishes, incubated overnight and transfected with FuGENE 6 reagent (Roche, Indianapolis, IN) with either the pRLnull promoterless empty vector (Promega, Madison, WI) or the pRLnull vector containing the intact exon 1c hVDR promoter region (pRL800) or various mutated or truncated sequences upstream of renilla luciferase. The pRL800 vector was created by insertion of 800 bp of the 5 -flanking region of exon 1c of the hVDR gene into the pRLnull vector (Fig. 1A; Byrne et al., 2000). Truncated versions of pRL800 were created and designated pRL620, pRL250 and pRL100 (Fig. 1A). The basal transcriptional activity of the truncated vectors in T47D cells was not significantly different from that of the intact pRL800. The mutated versions of pRL800 were created by site-directed mutagenesis as described below. In all experiments, transfection efficiency was monitored by cotransfection of pGL3, an sv40-driven firefly luciferase vector (Promega). Cells were treated with hormones or ligands at the concentrations indicated in figure legends for 18 h prior to

2. Materials and methods 2.1. Cells and culture T47D human breast cancer cells, obtained from ATCC, were routinely cultured in RPMI media (Sigma, St. Louis, MO) supplemented with 5% fetal bovine serum (FBS) (Sigma). SUM159PT cells were obtained from Dr. Steve Ethier at the University of Michigan Breast Cancer Cell/Tissue Bank. For the experiments described here, T47D cells were plated in RPMI media containing 1% FBS, and SUM159PT cells were plated in Hams F12 media with 5% charcoal stripped serum. Treatments included one or more of the following: ethanol vehicle control, estrogen (17␤estradiol, Sigma), resveratrol (Sigma), the synthetic estrogen receptor (ER) ligands DPN and PPT (Tocris, Ellisville, MO), 1,25(OH)2 D3 and the synthetic Vitamin D3 analogs CB1093 and EB1089 (Leo Pharmaceuticals, Ballerup, Denmark). Final concentrations for individual agents were optimized in previous experiments (Wietzke and Welsh, 2003) and are indicated in figure legends.

Fig. 1. Details of truncated and mutated hVDR promoter constructs. (A) Top schematic shows the 5 -flanking region of exon 1c of the human VDR gene, detailing locations of potential regulatory sites. Below are details of the luciferase reporter vectors containing full length (pRL800) and truncated (pRL620, pRL250 and pRL100) versions of the exon 1c VDR promoter region. (B) Mutagenesis of Sp1 sites in pRL800. The sequence of the original pRL800 vector (top) is aligned with the sequences of each mutated vector derived from pRL800. Bold type indicates which base pairs have been changed within the predicted Sp1 sites.

J.A. Wietzke et al. / Molecular and Cellular Endocrinology 230 (2005) 59–68

harvest and analysis with the Dual Luciferase kit (Promega) on a Wallac Victor 2 microplate reader. For each experiment, triplicate wells for transfection and treatment were analyzed and hVDR promoter luciferase activity was normalized to pGL3 luciferase activity and reported relative to activity of pRLnull, which was set to 1. Data are thus reported as mean ± standard error of relative luciferase units (RLU). Experiments were repeated at least three times and representative data are presented. 2.3. Site-directed mutagenesis The pRL800 vector was mutated in four separate reactions directed at four different Sp1 consensus binding sites positioned at −50 bp (site A), −262 bp (site C), −284 bp (site D) and −379 bp (site E) upstream from the exon 1c transcriptional start site of the hVDR gene (Fig. 1A). The Quick Change site-directed mutagenesis Kit (Stratagene, La Jolla, CA) was used according to manufacturer’s instructions to mutate 4 bp at each site, changing GC-rich regions to AT-rich regions (Fig. 1B). DNA sequencing of the mutated inserts was conducted with the Big Dye Terminator v3 kit (Applied Biosystems, Foster City, CA) with assistance from Dr. Neil Lobo (University of Notre Dame, Notre Dame, IN). Colonies that contained the desired mutations while the rest of the sequence remained intact were selected for the experiments described here. The mutated vectors were designated pRL800mutA, pRL800mutC, pRL800mutD, and pRL800mutE, with A–E designations corresponding to the Sp1 site within the 800 bp sequence that had been modified. Transient transfections with these vectors were carried out in T47D cells as described above. Vectors containing mutated Sp1 sites exhibited basal activity in T47D cells similar to that of the intact pRL800 vector. 2.4. Electrophoretic mobility shift assays Nuclear extracts were prepared from T47D cells treated with estrogen (1 nM) or vehicle control for up to 24 h using the NE-PER nuclear extraction reagents from Pierce (Rockford, IL). A 29 bp oligonucleotide probe corresponding to the estrogen responsive Sp1 sequence (site A) in the exon 1c promoter was biotin labeled with the Biotin 3 End DNA Labeling kit (Pierce). Nuclear extracts were pre-incubated (10 min) with binding buffer in the presence or absence of unlabeled oligonucleotide followed by 40 min incubation with biotinylated probe using the Light Shift Chemiluminescent EMSA kit from Pierce. Reaction mixes were loaded onto pre-run 6% polyacrylamide gels, electrophoresed for an additional 35 min, transferred to nylon membranes, crosslinked and developed. 2.5. Data analysis Statistical evaluation was by one-way analysis of variance (ANOVA), followed by multiple comparison tests (Dun-

61

nett or Tukey as appropriate) with Graph Pad Prism software version 3.00 for Windows (GraphPad Software, San Diego, CA, USA; www.graphpad.com). Differences between means were considered significant if p-values less than 0.05 were obtained.

3. Results 3.1. The 5 -flanking region of exon 1c in the hVDR gene is hormonally regulated in T47D cells The VDR mRNA is dynamically regulated by hormones and growth factors in both normal and transformed mammary epithelial cells (Krishnan and Feldman, 1991; Escaleira and Brentani, 1999; Lazzaro et al., 2000), but the mechanisms underlying this regulation are poorly understood. To determine whether hormones that are known to alter VDR expression correspondingly alter activity of the exon 1c VDR promoter, we utilized T47D cells, a human breast cancer cell line that expresses VDR and both ␣ and ␤ isoforms of the estrogen receptor (ER). As demonstrated in Fig. 2A, treatment with physiologically relevant concentrations of estrogen (1 nM) or the phytoestrogen resveratrol (4 nM) significantly enhanced activity of the pRL800 hVDR promoter construct in T47D cells. Similar results have been observed in MCF-7 cells, another breast cancer cell line that expresses both VDR and ER (Byrne et al., 2000; Wietzke and Welsh, 2003). However, neither estrogen (Byrne et al., 2000) or resveratrol (Fig. 2B) up-regulate the pRL800 promoter in ER negative SUM159PT cells. In T47D cells, the Vitamin D3 metabolite 1,25(OH)2 D3 , as well as two Vitamin D3 analogs, CB1093 and EB1089, also significantly up-regulated VDR promoter activity (Fig. 2C). Furthermore, co-incubation of T47D cells with E2 and 1,25(OH)2 D3 resulted in additive effects on VDR promoter activity (Fig. 2D), suggesting the possibility that these two agents act via independent mechanisms to activate the exon 1c hVDR promoter. 3.2. Activity and regulation of truncated hVDR promoter constructs in T47D cells The sequence of the exon 1c hVDR promoter region contains a number of potential transcription factor binding sites, including AP-1, AP-2 and Sp1 sites, but no consensus VDRE or ERE sequences (Byrne et al., 2000). To begin to define the hormonally responsive regions underlying VDR promoter up-regulation by estrogen, resveratrol and 1,25(OH)2 D3 , we repeated the experiments shown in Fig. 2 with shorter segments of the exon 1c promoter (shown schematically in Fig. 1A). Truncations of the 800 bp sequence were inserted into the pRLnull vector to yield several vectors including pRL100, containing just 100 bp of sequence upstream of exon 1c and the most proximal Sp1 site. In T47D cells transfected with pRL100, treatment with 1 nM estrogen for 18 h

62

J.A. Wietzke et al. / Molecular and Cellular Endocrinology 230 (2005) 59–68

up-regulated relative luciferase activity approximately 2-fold above that of control cells (Fig. 3A). Since the magnitude of this up-regulation is similar to that observed for the intact pRL800 vector in both T47D (Fig. 1) and MCF-7 cells (Byrne et al., 2000), these data suggested that the estrogen responsive region of the hVDR promoter is contained within the first 100 bp of the pRL800 sequence. In contrast, treatment with resveratrol or 1,25(OH)2 D3 at doses that up-regulated the pRL800 vector did not up-regulate pRL100 activity (Fig. 3A),

Fig. 2. Regulation of exon 1c VDR promoter activity by estrogens and Vitamin D compounds. Activity of the exon 1c VDR promoter (pRL800) was measured in: (A) T47D cells treated with ethanol (EtOH) vehicle, 1 nM estrogen (E2 ) or 4 nM resveratrol (RES); (B) SUM159PT cells treated with ethanol (EtOH) vehicle, 1 ␮M forskolin (FSK, as positive control) or 4 nM resveratrol (RES); (C) T47D cells treated with 100 nM 1,25(OH)2 D3 (D3 ) or the Vitamin D analogs CB1093 or EB1089 at 100 nM, or; (D) T47D cells treated with 1 nM E2 , 100 nM 1,25(OH)2 D3 or both E2 and 1,25(OH)2 D3 . Null: activity of pRLNull vector lacking the 800 bp exon 1c promoter insert. Data are expressed as relative luciferase units (RLU) and represent mean ± standard error of triplicate samples. Similar results were obtained in three independent experiments. * p < 0.05, treated vs. vehicle control, significance of other comparisons as indicated on figure.

Fig. 3. Activity of truncated hVDR promoter constructs in T47D cells following treatment with E2 , RES or 1,25(OH)2 D3 . (A) Activity of pRL800 and a truncated version containing 100 bp upstream of exon 1c (pRL100) was measured in T47D cells treated with ethanol (EtOH) vehicle control, 1 nM estrogen (E2 ), 4 nM resveratrol (RES) or 100 nM 1,25(OH)2 D3 (D3 ). (B) T47D cells were transfected with pRL800 or truncated versions containing 250 bp (pRL250) or 620 bp (pRL620) upstream of exon 1c and treated with vehicle control (−) or 4 nM resveratrol (RES). (C) T47D cells were transfected with pRL800 or truncated versions containing 250 bp (pRL250) or 620 bp (pRL620) upstream of exon 1c and treated with vehicle control (−) or 100 nM 1,25(OH)2 D3 (D3 ). Null: activity of pRLNull vector lacking the 800 bp exon 1c promoter insert. Data are expressed as relative luciferase units (RLU) and represent mean ± standard error of triplicate samples. Similar results were obtained in three independent experiments. * p < 0.05, treated vs. vehicle control, significance of other comparisons as indicated on figure. NS: not significant.

J.A. Wietzke et al. / Molecular and Cellular Endocrinology 230 (2005) 59–68

suggesting that activation of the hVDR promoter by resveratrol or 1,25(OH)2 D3 requires sequences that lie outside the first 100 bp, and is therefore likely to be mechanistically distinct from that of estrogen. To determine if longer constructs conferred transcriptional up-regulation in response to resveratrol or 1,25(OH)2 D3 , T47D cells were transfected with the pRL250 and pRL620 constructs and treated with 4 nM resveratrol or 100 nM 1,25(OH)2 D3 . With either treatment, significant upregulation was observed with the pRL620 vector but not with the pRL250 vector (Fig. 3B and C). Thus, the resveratrol and 1,25(OH)2 D3 responsive sequences of the hVDR promoter appear to be contained within the region between −250 and −620 bp upstream of exon 1c of hVDR. 3.3. Role of GC-rich motifs in VDR promoter regulation Although many possible transcription factor binding sites are localized within the regulatory regions of the hVDR promoter, a common feature of the regions inducible by all three agents is the presence of GC-rich Sp1 consensus sites. We therefore tested the hypothesis that one or more of the Sp1 sites present in the pRL800 sequence is necessary for upregulation of the VDR promoter. To test this hypothesis, sitedirected mutagenesis was utilized to mutate four Sp-1 sites in pRL800. One of these sites lies within the first 100 bp region (at −50), the other three (at −262, −285, and −381) are between −250 and −620 bp upstream of the start site (Fig. 1B). These sites were chosen for site-directed mutagenesis based on the truncation experiments showing that estrogen, resveratrol and 1,25(OH)2 D3 responsive sites lie in these regions of the pRL800 sequence. Sequencing of the resulting mutant 800 bp inserts confirmed that each vector contained four altered base pairs within the designated Sp-1 site, with no changes in the remaining insert sequence. The mutated vectors and the original pRL800 vector were independently transfected into T47D cells, and cells were treated with estrogen, resveratrol or 1,25(OH)2 D3 at the doses shown to up-regulate activity of pRL800. As shown in Fig. 4, up-regulation of luciferase activity by estrogen for pRL800mutC, pRL800mutD and pRL800mutE was comparable to that of the intact pRL800 vector, suggesting that disruption of the Sp1 sites located at −262, −285 and −381 did not impair estrogen responsiveness. In contrast, the pRL800mutA vector was not up-regulated by estrogen, suggesting that the Sp1 site located at −50 bp upstream of the exon 1c start site is required for estrogen mediated upregulation of VDR promoter activity. In support of this suggestion, electromobility shift assays demonstrated that estrogen treatment enhanced binding of T47D cell nuclear extracts to the Sp-1 site A sequence within 4 h (Fig. 4D). Similar studies were carried out with the mutated hVDR promoter vectors in T47D cells treated with resveratrol (Fig. 5). Resveratrol up-regulated the luciferase activity of the pRL800mutA, pRL800mutC, and pRL800mutD vectors but not that of the pRL800mutE vector. These data indicate that

63

resveratrol responsiveness of the hVDR promoter requires an intact Sp1 site located 381 bp upstream of exon 1c. The 1,25(OH)2 D3 responsive region of the hVDR promoter was identified by similar approaches in T47D cells transfected with the mutated vectors and treated with 100 nM 1,25(OH)2 D3 . 1,25(OH)2 D3 up-regulated the luciferase activity of pRL800mutA, pRL800mutD, and pRL800mutE vectors, but not the pRL800mutC vector (Fig. 6). These data indicate that the Sp1 site situated 262 bp upstream of exon 1c is required for 1,25(OH)2 D3 mediated up-regulation of hVDR promoter activity. 3.4. Role of ER subtypes in hVDR promoter regulation The data presented in Figs. 4 and 5 indicated that upregulation of the hVDR promoter by estrogen and resveratrol required two distinct Sp1 sites, and we have previously demonstrated that hVDR promoter activation by estrogen and resveratrol also requires functional ER (Byrne et al., 2000; Wietzke and Welsh, 2003). Since ligand binding to ER␣ and ER␤ can activate multiple genes via Sp1 sites (Dong et al., 1999; Bruning et al., 2003; Salvatori et al., 2003; Petz et al., 2004), we hypothesized that one mechanism by which estrogen and resveratrol could alter hVDR promoter via distinct Sp1 sites might be via differential activation of ER␣ or ER␤. As T47D cells express both ER␣ and ER␤ subtypes, either one or both ER subtypes may be involved in hVDR regulation by estrogen and resveratrol. To provide insight into this issue, we used cell permeable ER subtype specific ligands, PPT and DPN, as tools to probe the role of ER subtypes in hVDR regulation (Kraichely et al., 2000; Meyers et al., 2001). PPT binds to ER␣, but not ER␤, and selectively blocks transactivation through ER␣, whereas DPN binds to and blocks ER␤ activity. Thus, these ER specific ligands were used as competitive inhibitors to probe the requirement for ER␣ and ER␤ in hVDR promoter regulation by estrogen and resveratrol in T47D cells. Cells transfected with pRL800 were treated with estrogen and resveratrol in the presence or absence of PPT or DPN at 1 nM, a dose previously shown to modulate ER␣ or ER␤ specific effects, respectively, in T47D cells (Kraichely et al., 2000; Meyers et al., 2001). As shown in Fig. 7A, PPT and DPN alone had no effect on hVDR promoter activity, but both ligands blocked the up-regulation of VDR promoter activity by estrogen. In contrast, PPT, but not DPN, blocked up-regulation of VDR promoter activity by resveratrol (Fig. 7B). These data suggest that the effects of estrogen on hVDR promoter activity may be mediated via both ER␣ and ER␤, whereas the effect of resveratrol requires ER␣ but not ER␤.

4. Discussion Multiple promoters have been identified in the hVDR gene, including a TATA-containing promoter upstream of exon 1c that displays both basal and inducible transcriptional

64

J.A. Wietzke et al. / Molecular and Cellular Endocrinology 230 (2005) 59–68

Fig. 4. Effect of estrogen on VDR promoter vectors containing mutated Sp1 sites. T47D cells were transfected with pRL800 or (A) pRL800mutA, (B) pRL800mutC, (C) pRL800mutD or pRL800mutE and treated with vehicle control or 1 nM estrogen (E2 ) for 18 h. Null: activity of pRLNull vector lacking the 800 bp exon 1c promoter insert. Data are mean ± standard error of triplicate samples and are representative of three independent experiments. (D) Nuclear extracts from T47D cells treated with 1 nM estrogen (E2 ) or vehicle control for 4 h were incubated with biotinylated oligonucleotide corresponding to the Sp1 consensus sequence of site A [GGAGGGG]. Gel shift assays were performed as described in Section 2. Lane 1: free probe (no nuclear extract); lane 2: vehicle treated nuclear extract; lane 3: estrogen treated nuclear extract; lane 4: estrogen treated nuclear extract plus excess unlabeled oligo “A”. Arrows indicate positions of two shifted bands, both of which were enhanced in estrogen-treated cells and competed with cold oligo “A”. Data are representative of three experimental replicates.

activity in breast cancer cells (Byrne et al., 2000; Wietzke and Welsh, 2003). The major focus of the studies reported here was to provide insight into the mechanisms by which the exon 1c hVDR promoter is regulated. Our previous studies demonstrated that hormones (including estrogen), phytoestrogens (including resveratrol and genistein) and growth factors (including IGF and EGF) up-regulate exon 1c promoter activity in breast cancer cells (Byrne et al., 2000; Welsh et al., 2002; Wietzke and Welsh, 2003) and here we demonstrate that 1,25(OH)2 D3 and several Vitamin D3 analogs also up-regulate VDR promoter activity in T47D cells. While all of these agents have previously been shown to modulate VDR mRNA and/or protein expression in various cell lines (Davoodi et al., 1995; Escaleira and Brentani, 1999; Lazzaro et al., 2000; Wietzke and Welsh, 2003), ours are the first studies to implicate regulation at the level of the VDR promoter. We therefore exploited this model system to investigate po-

tential mechanisms by which transcription of the hVDR gene through this promoter is regulated. The major conclusion from our studies is that GC-rich, Sp1 consensus binding sites dictate responsiveness of the exon 1c hVDR promoter to 1,25(OH)2 D3 and estrogens. This conclusion is based on data generated with truncated and mutated versions of the VDR promoter sequence analyzed in T47D breast cancer cells which express both ER and VDR. Although signaling through both of these nuclear receptors classically involves binding of homo- or heterodimers to direct repeat sequences comprising specific hormone response elements, there are no obvious ER or VDR consensus binding sites in the exon 1c promoter sequence. Furthermore, our previous studies (Wietzke and Welsh, 2003) indicated that hVDR promoter regulation by resveratrol did not correlate with transcriptional activation through classical estrogen response elements (EREs). These observations

J.A. Wietzke et al. / Molecular and Cellular Endocrinology 230 (2005) 59–68

65

Fig. 6. Effect of 1,25D3 on VDR promoter vectors containing mutated Sp1 sites. T47D cells transfected with pRL800 or with (A) pRL800mutA or pRL800mutC or (B), pRL800mutD or pRL800mutE and treated in triplicate with 100 nM 1,25(OH2 D3 (D3 ). Null: activity of pRLNull vector lacking the 800 bp exon 1c promoter insert. Data are mean ± standard error of triplicate samples and are representative of three independent experiments.

Fig. 5. Effect of resveratrol on VDR promoter vectors containing mutated Sp1 sites. T47D cells were transfected with pRL800 or (A) pRL800mutA or pRL800mutC, (B) pRL800mutD and (C) pRL800mutE and treated with vehicle control or 4 nM resveratrol (RES) for 18 h. Null: activity of pRLNull vector lacking the 800 bp exon 1c promoter insert. Data are mean ± standard error of triplicate samples and are representative of three independent experiments.

prompted us to investigate the role of alternative signaling mechanisms in hVDR promoter regulation. Since both ER and VDR can activate transcription of target genes through interactions with the Sp1 transcription factor (Dong et al., 1999; Inoue et al., 1999; Bruning et al., 2003; Salvatori et al., 2003; Huang et al., 2004; Petz et al., 2004), we focused on GC-rich regions upstream of exon 1c. Through reporter gene assays with truncated and mutated versions of these GCrich regions, we identified three Sp1 consensus binding sites involved in regulation of the VDR promoter. Of particular interest, mutations in discrete Sp1 binding sites abolished responsiveness of this promoter to 1,25(OH)2 D3 , estrogen and the phytoestrogen, resveratrol. These observations suggest that additional potential transcription binding sites present in this promoter, such as AP-1 and AP-2, are not required for regulation of hVDR promoter activity by 1,25(OH)2 D3 , estrogen or resveratrol. However, the hVDR promoter is also

up-regulated by forskolin, dexamethasone and certain growth factors, and further studies may well provide evidence that AP-1 or AP-2 sites are involved in mediating effects of these agents (Byrne et al., 2000). In T47D cells, we observed that the exon 1c promoter region of the VDR is up-regulated by 1,25(OH)2 D3 as well as two synthetic Vitamin D3 analogs, EB1089 and CB1093 (Leo Pharmaceuticals, Ballerup, Denmark). Since we have previously reported no effect of 1,25(OH)2 D3 on this VDR promoter region in MCF-7 breast cancer cells (Byrne et al., 2000), there appear to be cell type specific differences in promoter regulation by Vitamin D3 compounds. We have observed that 1,25(OH)2 D3 up-regulates the exon 1c promoter in other VDR target cells such as those derived from bone and kidney (Wietzke and Welsh, unpublished data). Up-regulation of the VDR promoter by Vitamin D3 compounds is consistent with ligand-induced up-regulation of hVDR mRNA reported in vivo and in vitro (Mahonen and Maenpaa, 1994; Lazzaro et al., 2000; Soldati et al., 2004). However, conformational changes induced by 1,25(OH)2 D3 binding are known to stabilize the VDR protein from degradation, indicating that VDR is also regulated at the posttranslational level (Wiese et al., 1992; Santiso-Mere et al., 1993; Davoodi et al., 1995). While it is likely that VDR regulation by 1,25(OH)2 D3 is complex and is mediated via multiple mechanisms, our studies are the first to provide evidence that up-regulation of the VDR promoter may be involved. Our data further indicate that a GC-rich motif

66

J.A. Wietzke et al. / Molecular and Cellular Endocrinology 230 (2005) 59–68

Fig. 7. Effect of E2 or RES on VDR promoter activity in the presence of ER subtype specific ligands. (A) T47D cells transfected with the pRL800 vector were treated with vehicle control or 1 nM estrogen (E2 ) in the presence or absence of 1 nM of the ER specific ligands PPT or DPN. (B) T47D cells transfected with the pRL800 vector were treated with vehicle control (−) or 4 nM resveratrol (RES) in the presence or absence of 1 nM of the ER specific ligands PPT or DPN. Data are mean ± standard error of triplicate samples and are representative of three independent experiments. * p < 0.05 relative to vehicle control treated cells.

[–GGTCGGGGTC–], a consensus Sp1 binding site that lies 262 bp upstream of exon 1c, is required for 1,25(OH)2 D3 mediated up-regulation of hVDR promoter activity. While further studies will be necessary to determine whether 1,25(OH)2 D3 specifically activates Sp1 binding to this site, this observation is consistent with previous reports that transactivation of the p27Kip1 gene by 1,25(OH)2 D3 requires Sp1 and is mediated through VDR binding to GC-rich promoter regions (Inoue et al., 1999; Huang et al., 2004). Using similar approaches, responsiveness of the VDR promoter to estrogen and resveratrol was also linked to GC-rich Sp1 binding sites. Consistent with estrogen regulation of the VDR shown here, previous reports have demonstrated that liganded ER interacting with Sp1 mediates estrogen regulation of numerous target genes, including bcl-2 (Dong et al., 1999), the progesterone receptor (Petz et al., 2004), the LDL receptor (Bruning et al., 2003) and the EGF receptor (Salvatori et al., 2003). Resveratrol has previously been shown to activate classical ERE signaling through ER in breast cancer cells (Bowers et al., 2000; Mueller et al., 2004), but ours is the first data to demonstrate that resveratrol may also signal through GC-rich motifs. Interestingly,

estrogen and resveratrol up-regulated the exon 1c VDR promoter through two similar but spatially distinct GC-rich regions. A consensus Sp1 binding site 50 bp upstream of exon 1c [–GGAGGGG–] was essential for estrogen regulation, and enhanced binding to this sequence was observed in EMSA assays with nuclear extracts from estrogen treated cells. In contrast, resveratrol activation of the VDR promoter was linked to a more distal GC-rich region [–GGGAGG–] located 381 bp upstream of exon 1c. These findings indicate that estrogen and resveratrol likely impact on VDR promoter activity through subtly distinct signaling pathways involving Sp1. We have previously shown that up-regulation of the VDR promoter by either estrogen or resveratrol requires the presence of functional ER and is blocked by the anti-estrogen tamoxifen (Byrne et al., 2000; Wietzke and Welsh, 2003). However, T47D cells express both ER␣ and ER␤ (Vladusic et al., 2000), and both isoforms have been shown to interact with the C-terminal domain of the Sp1 protein (Saville et al., 2000). Furthermore, ER␣ and ER␤ are differentially activated by estrogenic compounds (Jones et al., 1999; Pearce and Jordan, 2004), and the relative amounts of each isoform within cells can influence Sp1 activity (Salvatori et al., 2003). Thus, one possible way in which estrogen and resveratrol could differentially activate GC-rich motifs in the VDR promoter is via interaction with distinct ER subtypes. Data from our studies with ER subtype specific ligands PPT (ER␣ selective) and DPN (ER␤ selective) support this concept, since estrogen and resveratrol activation of the VDR promoter were differentially blocked in the presence of DPN and PPT. Estrogen activation of the VDR promoter was blocked by either DPN or PPT, suggesting that ER␣ and ER␤ may both be necessary for estrogen regulation. In contrast, resveratrol activation of the VDR promoter was blocked by PPT but not DPN, suggesting that only ER␣ is required for resveratrol regulation. Although the ER ligands utilized are highly specific for ER␣ versus ER␤, studies to confirm the relative importance of ER isoforms and to identify the specific complexes that bind to the GC-rich motifs in the VDR promoter will be necessary to further characterize the regulation of this promoter. In summary, these studies have confirmed and extended previous work on hormonal regulation of the exon 1c VDR promoter region in breast cancer cells. Using truncated and mutated reporter gene constructs, we have identified three distinct GC-rich motifs, which correspond to consensus binding sites for the Sp1 transcription factor, that confer responsiveness of this promoter to 1,25(OH)2 D3 and estrogens in breast cancer cells. While the relevance of VDR promoter regulation in breast cancer remains to be determined, published work demonstrating differential VDR promoter usage in human parathyroid adenomas (Correa et al., 2002), activation of the VDR promoter by the Wilms’ tumor product, WT1 (Lee and Pelletier, 2001) and reduced VDR expression in transformed cells (Escaleira and Brentani, 1999; Agadir et al., 1999) points to deregulation of VDR signaling during cancer progression. Collectively, these studies provide the foundation for follow-up experiments to clarify the molec-

J.A. Wietzke et al. / Molecular and Cellular Endocrinology 230 (2005) 59–68

ular mechanisms underlying VDR promoter usage and cell specific regulation by signaling pathways as a function of transformation.

Acknowledgements This work was supported by research grants to J. Welsh from the National Institutes of Health (CA69700) and the US Army Medical Research and Materiel Command (DAMD17-02-1-0208).

References Agadir, A., Lazzaro, G., Zheng, Y., Zhang, X.K., Mehta, R., 1999. Resistance of HBL100 human breast epithelial cells to Vitamin D action. Carcinogenesis 20, 577–582. Bowers, J.L., Tyulmenkov, V.V., Jernigan, S.C., Klinge, C.M., 2000. Resveratrol acts as a mixed agonist/antagonist for estrogen receptors alpha and beta. Endocrinology 141, 3657–3667. Bruning, J.C., Lingohr, P., Gillette, J., Hanstein, B., Avci, H., Krone, W., Muller-Wieland, D., Kotzka, J., 2003. Estrogen receptor-alpha and Sp1 interact in the induction of the low density lipoprotein-receptor. J. Steroid Biochem. Mol. Biol. 86, 113–121. Byrne, I., Flanagan, L., Tenniswood, M., Welsh, J.E., 2000. Identification of a hormone responsive promoter immediately upstream of exon 1c in the human Vitamin D receptor gene. Endocrinology 141, 2829–2836. Correa, P., Akerstrom, G., Westin, G., 2002. Exclusive underexpression of Vitamin D receptor exon 1f transcripts in tumors of primary hyperparathyroidism. Eur. J. Endocrinol. 147, 671–675. Crofts, L., Hancock, M., Morrison, N., Eisman, J., 1998. Multiple promoters direct the tissue specific expression of novel N-terminal variant human Vitamin D receptor gene transcripts. Proc. Natl. Acad. Sci. U.S.A. 95, 10529–10534. Davoodi, F., Brenner, R.V., Evans, S.R., Schumaker, L.M., Shabahang, M., Nauta, R.J., Buras, R.R., 1995. Modulation of Vitamin D receptor and estrogen receptor by 1,25(OH)2-Vitamin D3 in T-47D human breast cancer cells. J. Steroid Biochem. Mol. Biol. 54, 147–153. Dong, L., Wang, W., Wang, F., Stoner, M., Reed, J., Harigan, M., Samudio, I., Kladde, M., Vyhlidal, C., Safe, S., 1999. Mechanisms of transcriptional activitaion of bcl-2 gene expression by 17 beta-estradiol in breast cancer cells. J. Biol. Chem. 274, 32099–32107. Escaleira, M.T., Brentani, M.M., 1999. Vitamin D3 receptor (VDR) expression in HC-11 mammary cells: regulation by growth modulatory agents, differentiation and Ha-ras transformation. Breast Cancer Res. Treat. 54, 123–133. Huang, Y.C., Chen, J.Y., Hung, W.C., 2004. Vitamin D3 receptor/Sp1 complex is required for the induction of p27Kip1 expression by Vitamin D3 . Oncogene 23, 4856–4861. Huening, M., Yehia, G., Molina, C.A., Christakos, S., 2002. Evidence for a regulatory role of inducible cAMP early repressor in protein kinase a-mediated enhancement of Vitamin D receptor expression and modulation of hormone action. Mol. Endocrinol. 16, 2052–2064. Inoue, T., Kamiyama, J., Sakai, T., 1999. Sp1 and NF-Y synergistically mediate the effect of Vitamin D(3) in the p27(kip1) gene promoter that lacks Vitamin D response elements. J. Biol. Chem. 274, 32309–32317. Jones, P., Parrott, E., White, I., 1999. Activation of transcription by estrogen receptor alpha and beta is cell type- and promoter-dependent. J. Biol. Chem. 274, 32008–32014. Kraichely, D.M., Sun, J., Katzenellenbogen, J.A., Katzenellenbogen, B.S., 2000. Conformational changes and coactivator recruitment by novel ligands for estrogen receptor-alpha and estrogen receptorbeta: correlations with biological character and distinct differences

67

among SRC coactivator family members. Endocrinology 141, 3534– 3545. Krishnan, A.V., Feldman, D., 1991. Stimulation of 1,25-dihydroxyvitamin D3 receptor gene expression in cultured cells by serum and growth factors. J. Bone Miner. Res. 6, 1099–1107. Lazzaro, G., Agadir, A., Qing, W., Poria, M., Mehta, R.R., Moriarty, R.M., Das Gupta, T.K., Zhang, X.K., Mehta, R.G., 2000. Induction of differentiation by 1alpha-hydroxyvitamin D(5) in T47D human breast cancer cells and its interaction with Vitamin D receptors. Eur. J. Cancer 36, 780–786. Lee, T.H., Pelletier, J., 2001. Functional characterization of WT1 binding sites within the human Vitamin D receptor gene promoter. Physiol. Genomics 7, 187–200. Mahonen, A., Maenpaa, P.H., 1994. Steroid hormone modulation of Vitamin D receptor levels in human MG-63 osteosarcoma cells. Biochem. Biophys. Res. Commun. 205, 1179–1186. Meyers, M.J., Sun, J., Carlson, K.E., Marriner, G.A., Katzenellenbogen, B.S., Katzenellenbogen, J.A., 2001. Estrogen receptor-beta potencyselective ligands: structure-activity relationship studies of diarylpropionitriles and their acetylene and polar analogues. J. Med. Chem. 44, 4230–4251. Miyamoto, K., Kesterson, R., Yamamoto, H., Taketani, Y., Nishiwaki, E., Tatsumi, S., Inoue, Y., Morita, K., Takeda, E., Pike, J.W., 1997. Structural organization of the human Vitamin D receptor chromosomal gene and its promoter. Mol. Endocrinol. 11, 1165–1179. Mørk Hansen, C., Binderup, L., Hamberg, K., Carlberg, C., 2001. Vitamin D and cancer: effects of 1,25(OH)2D3 and its analogues on growth control and tumorigenesis. Front. Biosci. 6, 820–848. Mueller, S.O., Simon, S., Chae, K., Metzler, M., Korach, K.S., 2004. Phytoestrogens and their human metabolites show distinct agonistic and antagonistic properties on estrogen receptor alpha (ERalpha) and ERbeta in human cells. Toxicol. Sci. 80, 14–25. Pearce, S.T., Jordan, V.C., 2004. The biological role of estrogen receptors alpha and beta in cancer. Crit. Rev. Oncol. Hematol. 50, 3–22. Petz, L.N., Ziegler, Y.S., Schultz, J.R., Kim, H., Kemper, J.K., Nardulli, A.M., 2004. Differential regulation of the human progesterone receptor gene through an estrogen response element half site and Sp1 sites. J. Steroid Biochem. Mol. Biol. 88, 113–122. Salvatori, L., Pallante, P., Ravenna, L., Chinzari, P., Frati, L., Russo, M.A., Petrangeli, E., 2003. Oestrogens and selective oestrogen receptor (ER) modulators regulate EGF receptor gene expression through human ER alpha and beta subtypes via an Sp1 site. Oncogene 22, 4875– 4881. Santiso-Mere, D., Sone, T., Hilliard IV, G.M., Pike, J.W., McDonnell, D.P., 1993. Positive regulation of the Vitamin D receptor by its cognate ligand in heterologous expression systems. Mol. Endocrinol. 7, 833–839. Saville, B., Wormke, M., Wang, F., Nguyen, T., Enmark, E., Kuiper, G., Gustafsson, J.A., Safe, S., 2000. Ligand-, cell-, and estrogen receptor subtype (alpha/beta)-dependent activation at GC-rich (Sp1) promoter elements. J. Biol. Chem. 275, 5379–5387. Soldati, L., Adamo, D., Bianchin, C., Arcidiacono, T., Terranegra, A., Bianchi, M.L., Mora, S., Cusi, D., Vezzoli, G., Soldati, L., Adamo, D., Bianchin, C., Arcidiacono, T., Terranegra, A., Bianchi, M.L., Mora, S., Cusi, D., Vezzoli, G., 2004. Vitamin D receptor mRNA measured in leukocytes with the TaqMan fluorogenic detection system: effect of calcitriol administration. Clin. Chem. 50, 1315–1321. VanWeelden, K., Flanagan, L., Binderup, L., Tenniswood, M., Welsh, J.E., 1998. Apoptotic regression of MCF-7 xenografts in nude mice treated with the Vitamin D analog EB1089. Endocrinology 139, 2102–2110. Vladusic, E.A., Hornby, A.E., Guerra-Vladusic, F.K., Lakins, J., Lupu, R., 2000. Expression and regulation of estrogen receptor beta in human breast tumors and cell lines. Oncol. Rep. 7, 157–167. Welsh, J.E., Wietzke, J.A., Zinser, G.M., Smyczek, S., Romu, S., Tribble, E., Welsh, J.C., Byrne, B., Narvaez, C.J., 2002. Impact of the Vitamin D3 receptor on growth regulatory pathways in mammary gland and breast cancer. J. Steroid Biochem. Mol. Biol. 83, 85–92.

68

J.A. Wietzke et al. / Molecular and Cellular Endocrinology 230 (2005) 59–68

Wiese, R.J., Uhland-Smith, A., Ross, T.K., Prahl, J.M., DeLuca, H.F., 1992. Up-regulation of the Vitamin D receptor in response to 1,25dihydroxyvitamin D3 results from ligand-induced stabilization. J. Biol. Chem. 267, 20082–20086. Wietzke, J.A., Welsh, J.E., 2003. Phytoestrogen regulation of a Vitamin D3 receptor promoter and 1,25-dihydroxyvitamin D3 actions in human breast cancer cells. J. Steroid Biochem. Mol. Biol. 84, 149–157.

Yamamoto, H., Miyamoto, K., Li, B., Taketani, Y., Kitano, M., Inoue, Y., Morita, K., Pike, J.W., Takeda, E., 1999. The caudal-related homeodomain protein Cdx-2 regulates Vitamin D receptor gene expression in the small intestine. J. Bone Miner. Res. 14, 240–247. Zinser, G.M., McEleney, K., Welsh, J.E., 2003. Characterization of mammary tumor cell lines from wild type and Vitamin D receptor knockout mice. Mol. Cell. Endocrinol. 200, 67–80.