Hormonal Regulation of Metastasis-Associated Protein 3 Transcription in Breast Cancer Cells (original) (raw)

Journal Article

,

1Department of Pathology and Laboratory Medicine, Winship Cancer Institute, Emory University, Atlanta, Georgia 30322

Search for other works by this author on:

,

1Department of Pathology and Laboratory Medicine, Winship Cancer Institute, Emory University, Atlanta, Georgia 30322

Search for other works by this author on:

,

1Department of Pathology and Laboratory Medicine, Winship Cancer Institute, Emory University, Atlanta, Georgia 30322

Search for other works by this author on:

1Department of Pathology and Laboratory Medicine, Winship Cancer Institute, Emory University, Atlanta, Georgia 30322

*Address all correspondence and requests for reprints to: Paul A. Wade, Department of Pathology, Emory University, Whitehead Building Room 142, 615 Michael Street, Atlanta, Georgia 30322.

Search for other works by this author on:

Published:

01 December 2004

• PDF
• Split View
• • Article contents
  • Figures & tables
  • Video
  • Audio
  • Supplementary Data
• Cite

Cite

Naoyuki Fujita, Masahiro Kajita, Panya Taysavang, Paul A. Wade, Hormonal Regulation of Metastasis-Associated Protein 3 Transcription in Breast Cancer Cells, Molecular Endocrinology, Volume 18, Issue 12, 1 December 2004, Pages 2937–2949, https://doi.org/10.1210/me.2004-0258
Close

Navbar Search Filter Mobile Enter search term Search

Abstract

Metastasis-associated protein 3 (MTA3) is a cell type-specific subunit of the Mi-2/NuRD transcriptional corepressor complex. In breast cancer cells, MTA3 and the Mi-2/NuRD complex mediate repression of Snail, a transcription factor that promotes epithelial to mesenchymal transitions. Thus, MTA3 functions to maintain a differentiated, epithelial status in breast cancer. Interestingly, in mammary epithelial cells, MTA3 biosynthesis requires both functional estrogen receptor (ER) and estradiol. Here we have investigated the molecular basis for estrogen and ER-dependent expression of MTA3 in breast cancer cells. Molecular dissection of the MTA3 promoter using transient transfection assays identified a composite element required for high-level transcription consisting of an SP1 site in close proximity to a consensus estrogen response element half-site. Depletion of either SP1 or ER-α by RNA interference led to loss of MTA3 transcript in multiple breast cancer cell lines, indicating a requirement for both transcription factors in expression of endogenous MTA3. The MTA3 gene thus joins a growing list of loci regulated by both SP1 and ER.

THE VERTEBRATE MI-2/NuRD complex is a multisubunit protein assembly containing both histone deacetylase and chromatin remodeling ATPase activities (1–5). Current models predict that hypoacetylation of core histones is associated with inactive genes; thus, complexes containing histone deacetylase (HDAC) activity are generally thought to participate in transcriptional repression (6, 7). The Mi-2/NuRD complex, like the mammalian SWI/SNF complex, contains several subunits whose pattern of expression is heterogeneous in various cell and tissue types (8). Presumably, subunit heterogeneity provides these complexes with additional regulatory capacity and provides unique functional properties. One particularly interesting subunit of the Mi-2/NuRD complex is the metastasis-associated protein 3 (MTA3) protein (8, 9). MTA3 is a member of a small protein family in mammals, the founding member of which (MTA1) was isolated in a differential display screen designed to identify mRNAs up-regulated in metastatic breast cancer cells (9–11). All three members of this family, MTA1, MTA2, and MTA3, have been reported as subunits of Mi-2/NuRD (1, 2, 4, 9, 12, 13). Notably, MTA3 is expressed at high levels in an estrogen-dependent fashion in breast cancer cells, predicting that the MTA3-containing version of the Mi-2/NuRD complex functions downstream of the estrogen receptor (ER) in breast cancer cells (9, 14, 15). A requirement was demonstrated for this complex in repression of the transcription factor Snail, a master regulatory gene involved in epithelial to mesenchymal transitions (9, 14, 15). MTA3 thus participates in the genetic program downstream of ER. Importantly, MTA3 and the Mi-2/NuRD complex act to repress genes, providing a novel functionality to ER, transcriptional repression mediated through the biological action of a target gene.

ERs are required for normal development and function of female reproductive tissue in mammals (16). Ligand binding by these molecules converts the receptors into transcriptional activators (17, 18). The physiological effects of estrogen action are mediated by direct transcriptional activation of genes by ERs and by the biological actions of direct target genes. The latter category constitutes secondary, or indirect, biological effects representing an important subset of the biological outcomes mediated by estrogen action. The MTA3 subunit of the Mi-2/NuRD complex represents a transcriptional regulator integral to the maintenance of differentiated epithelial status of breast cancer cells (9, 14, 15). The observation that MTA3 biosynthesis is closely correlated with ER status and with the presence of hormone in breast cancer cell lines raises an important question. Is MTA3 a direct target of ER, or is transcription of the MTA3 locus driven by an indirect effect of estrogen action?

In this work, we have dissected the molecular basis for estrogen dependence of MTA3 expression in breast cancer cells. Transient transfection experiments were employed to identify the _cis_-acting regulatory DNA at the MTA3 locus driving high-level, estrogen-dependent transcription in breast cancer cells. A small sequence element was isolated, containing an SP1 site in close proximity to an estrogen response element (ERE) half-site. Molecular dissection of the requirements for these two transcription factors revealed synergy between SP1 and ER-α in driving high-level transcription of the MTA3 locus in breast cancer cells. Importantly, SP1 was absolutely required for productive binding of ER-α to MTA3 chromatin in human cells. The data support a model predicting that local chromatin architecture influences the ability of ER-α to selectively bind the MTA3 promoter in an SP1-dependent fashion.

RESULTS

The MTA3 Promoter in Breast Cancer Cells

High-level expression of MTA3 in breast cancer cell lines requires functional ER and estradiol (9). The slow decline in steady state levels of MTA3 protein after disruption of ER activity (9) suggested that either MTA3 protein has a long half-life, or that ER does not directly activate the MTA3 promoter. To differentiate between these possibilities, we have dissected the transcriptional regulatory elements present in the MTA3 promoter using a transient transfection strategy. As a prelude to these studies, it was imperative to determine the transcription start site at the MTA3 locus in breast cancer cell lines. The human MTA3 locus is located on chromosome 2p21. Analysis of the 5′ ends of cDNAs corresponding to MTA3 revealed two possible transcription start sites. One cDNA, KIAA1266 (19–21), includes a noncoding first exon not present in any other expressed sequence tag (EST) sequence displayed in public databases (see genome.ucsc.edu). In fact, the KIAA1266 mRNA sequence lacks an exon predicted to contain the translation initiation site as well as nine amino acids at the amino terminus of the predicted MTA3 protein (Fig. 1, A and B). Previous analysis of the MTA3 cDNA from HeLa and MCF7 cell lines indicated the presence of these amino acids in the mature mRNA from these cells (9). These conflicting observations raised the question of whether the upstream, noncoding exon was transcribed in breast cancer cell lines as opposed to brain, the tissue from which the KIAA1266 cDNA was isolated (19–21). To determine whether the noncoding exon was transcribed in breast cancer cells, RT-PCR was performed with primer sets capable of detecting the presence of this exon (Fig. 1A). We failed to detect any RNA species containing this noncoding exon in either MCF7 or in T47D (Fig. 1A). We concluded from this analysis that initiation of MTA3 transcription likely occurs at a site downstream of what appears to be a CNS-specific noncoding first exon. Multiple EST clones in the public databases have 5′ ends in the region upstream of the initiator methionine and downstream of the noncoding exon 1b (Fig. 1B). The most 5′ cDNA end, found in EST clone BU859486 (http://mgc.nci.nih.gov/) initiates at a cytosine residue that we designated as +1. Sequence analysis of the MTA3 locus immediately surrounding this putative transcription initiation site revealed the presence of multiple transcription factor binding sites, including several SP1 sites along with multiple ERE half-sites (Fig. 1, B and C).

Fig. 1.

The MTA3 Promoter in Breast Cancer Cells A, The cartoon on the left depicts the exon structure of a portion of the MTA3 locus. The positions of the noncoding exon from KIAA1266 (exon 1b) and the first protein coding exon (exon 1a) are indicated. The positions of PCR primers used in RT-PCR are also indicated in the cartoon. The right panel depicts RT-PCR performed as described in Materials and Methods using the indicated primer sets. B, The sequence of the MTA3 promoter is shown at the nucleotide level. SP1 binding sites, ERE half-sites, and the transcription start sites for KIAA1266 and BU859486 are indicated. The 5′ends of cDNAs in the public databases are indicated by a large font. The MTA3 CpG island is indicated by underlining. C, The locations of potential ERE half-sites and SP1 elements are depicted in cartoon fashion relative to the exons at the 5′ end of MTA3. Putative ERE half-sites are compared with other ERE sequences. The consensus sequence for an ERE half-site PuGGTCA is found in the MTA3 upstream region at positions −1232 to −1227 and at −134 to −129. The palindromic ERE sequences indicated as present at the human BCRP, c_-fos_, and Factor XII loci are examples of nonconsensus ERE sequences. No palindromic EREs, consensus or nonconsensus, were detected at MTA3.

We prepared a DNA construct fusing MTA3 sequence from −1332 to +228 to luciferase to determine _cis_-acting sequences necessary for high-level transcription in ER positive breast cancer cell lines. A plasmid containing sequences from intron 1 of the MTA3 locus served as a control for these experiments (Fig. 2A). First, the plasmids were transfected into the ER negative cell line MDA231 along with varying amounts of an ER-α expression plasmid. Luciferase expression driven by the control plasmid lacking any MTA3 genomic sequences or by the intronic sequences was insensitive to ER (Fig. 2A). In contrast, the DNA upstream of the MTA3 transcription start site, including the putative promoter, drove high-level expression of luciferase in an ER-dependent fashion (Fig. 2A). The dependence on ER for transcriptional activity of this putative promoter fragment was also analyzed in MCF7 cells. First, cells were grown for 3 d in media depleted of steroids, leading to declines in the level of ER (data not shown). Cells were then transfected with luciferase reporter plasmids, varying amounts of ER expression plasmid, and treated with 10 nm 17β-estradiol. High-level reporter expression occurred only in those cells transfected with the reporter plasmid containing MTA3 upstream sequence (Fig. 2B). To ascertain the dependence of this response on estradiol and ER-α, a dose response experiment was performed. The luciferase reporter construct was cotransfected with various quantities of ER-α expression vector, and cells were treated with vehicle or two different concentrations of 17β-estradiol. Transcriptional activation was substantially increased by estradiol at nanomolar concentration (Fig. 2C). We concluded from these experiments that _cis_-acting sequences located in the region upstream of exon 1a of MTA3 are sufficient to drive high-level expression in an ER and estradiol-dependent manner.

Fig. 2.

Transcription of DNA Upstream of the MTA3 Translation Start Site Is Estrogen Responsive A, Transient transfection assays were conducted using MDA231 cells. DNA constructs transfected included a mock luciferase plasmid containing no MTA3 sequence (diamonds in the graph), MTA3 sequence from −1332 to +228 fused to luciferase (squares in the graph), and MTA3 sequence from +257 to +1920 (triangles in the graph). Luciferase activity relative to a transfection control (Renilla luciferase driven by the Drosophila Actin 5C promoter) for each template is displayed as a function of amount of ER-α expression plasmid transfected. Each point on the graph represents the mean of three independent experiments; error bars indicate sd. Experimental values that differ significantly (by t test) from the mock control are indicated with asterisks. B, Transient transfection assays were performed as in Fig. 2A. The graph depicts normalized luciferase data for each transcription template displayed as a function of amount of ER-α expression plasmid transfected. Host cells used in this experiment were MCF7 grown in steroid-depleted media. After transfection, cells were treated with 10 nm 17β-estradiol. Each point on the graph represents the mean of three independent experiments; error bars indicate sd. Experimental values that differ significantly (by t test) from the mock control are indicated with asterisks. C, MCF7 cells were grown in steroid-depleted media and then transfected with ER-α expression vector and the MTA3 (−1332 to +228) luciferase plasmid. The graph depicts normalized luciferase activity as a function of amount of ER-α expression plasmid and dose of 17β-estradiol. Each bar on the graph represents the mean of three independent experiments; error bars indicate sd. Luciferase values of cells treated with estradiol that differ significantly (by t test) from buffer controls are indicated with an asterisk.

Mapping of EREs at MTA3

To determine the functional EREs in the MTA3 promoter as well as other _cis_-acting regulatory DNA, a series of promoter deletion constructs were prepared. These constructs were cotransfected along with ER-α expression plasmid into MCF7 cells previously grown in steroid-free media for 3 d. After transfection, estradiol was added to the media at a final concentration of 10 nm. Progressive deletion of MTA3 sequence from the 5′ end of the reporter construct defined two regions containing regulatory DNA (Fig. 3A). Basal promoter activity was driven by a small DNA element within the CpG island encompassing exon 1a (Fig. 3A and data not shown). A second regulatory sequence was defined within the interval −573 to −3 (Fig. 3A). This sequence contains two ERE half-sites as well as one SP1 site (Fig. 3A). Sequence at the far 5′ end of the putative promoter failed to drive luciferase expression in the absence of sequence 3′ of −553 (Fig. 3A).

Fig. 3.

Identification of a Functional ERE Half-Site at the MTA3 Promoter A, MCF7 cells grown for 3 d in steroid-depleted media were transfected with ER-α expression plasmid and the indicated MTA3 promoter deletion luciferase plasmids. The mock plasmid pGL2 contains no MTA3 sequence. After transfection, cells were treated with 10 nm 17β-estradiol. The bar graph depicts normalized (to Drosophila Actin 5C Renilla luciferase) luciferase activity for each deletion construct. Each bar on the graph represents the mean of three independent experiments; error bars indicate sd. Luciferase values of promoter deletion mutants that differ significantly (by t test) from wild type are indicated with an asterisk. B, The ERE half-site at approximately −130 was mutated as indicated in the figure. Wild-type and ERE-mutant versions of the MTA3 full-length promoter luciferase plasmid were transfected and assayed for luciferase activity as described in panel A. The bar graph depicts luciferase activity as a function of dose of ER-α expression plasmid for the control luciferase vector without any MTA3 sequence (pGL2 mock), the wild-type MTA3 promoter luciferase fusion (wt), and the reporter in which the ERE half-site is mutated (mtERE). Each bar on the graph represents the mean of three independent experiments; error bars indicate sd. Luciferase values of promoter deletion mutants that differ significantly (by t test) from wild type are indicated with an asterisk.

To determine the contribution of ER and the potential role of the ERE half-sites to reporter activity, a point mutant version of the reporter plasmid was generated. The consensus ERE half-site at −134 was altered from TGACC to AAGCTT in the context of the full-length MTA3 promoter plasmid. Mutation of this single ERE half-site resulted in markedly decreased levels of luciferase activity and a failure of ER cotransfection to activate transcription to the same levels as observed with the wild-type reporter (Fig. 3B). The effect was most profound at low levels of ER-α expression. We concluded from these experiments that the ERE half-site situated at −134 plays a crucial role in driving estrogen-dependent expression of the luciferase reporters. However, we also note that the reporter plasmid containing the mutant ERE half-site was activated by exogenously expressed ER-α. This result suggests a potential role for ER-α activation through one of the SP1 sites on the reporter as previously demonstrated for multiple genes in MCF7 (22).

A Role for SP1 in MTA3 Expression

Interestingly, this ERE half-site is located in close proximity to an SP1 element, approximately 65 bp away (Fig. 1B). Because cooperation of ER-α and SP1 has been reported at multiple ER-dependent loci (23–29), we further investigated the role of this SP1 site in expression of MTA3. A series of promoter deletion mutants were cotransfected into Drosophila S2 cells, which lack SP1 expression, along with an SP1 expression plasmid. The results of these experiments were very similar to those in MCF7 using ER-α. Again, two putative regulatory intervals were defined in the deletion analysis. Basal activity was driven by sequence containing a single SP1 site located within the MTA3 CpG island (Fig. 4A). High-level transcriptional activity required the presence of the SP1 site at −211, the site located in close proximity to the ERE half-site. Fine structure mapping of the SP1 response was pursued by construction of a mutant promoter plasmid where the GC box SP1 binding site at −211 was mutated. In the context of the full-length promoter construct, mutation of this SP1 site resulted in loss of roughly 2/3 of the SP1 response (Fig. 4B). A minimal promoter construct (−234 to +18) was sufficient to drive high-level transcription in an SP1-dependent manner (Fig. 4B). We concluded that a small sequence element encompassing the MTA3 transcription start and an SP1 site near an ERE half-site constituted a dominant regulatory sequence driving high-level transcription of the MTA3 reporter plasmids.

Fig. 4.

Identification of Functionally Important SP1 Elements at the MTA3 Promoter A, Drosophila S2 cells, which lack endogenous SP1 expression, were cotransfected with the indicated promoter deletion constructs and an SP1 expression plasmid or mock expression vector. Normalized luciferase activity for each promoter deletion is depicted in the bar graph. The control luciferase vector (pGL2) contains no MTA3 sequence. The bar graph depicts normalized (to Drosophila Actin 5C Renilla luciferase) luciferase activity for each deletion construct. Each bar on the graph represents the mean of three independent experiments; error bars indicate sd. Luciferase values of promoter deletion mutants that differ significantly (by t test) from wild type are indicated with an asterisk. B, Drosophila S2 cells were cotransfected with the indicated promoter luciferase plasmids and either SP1 expression or mock expression plasmids. The bar graph depicts normalized luciferase activity for each transfection experiment. Each bar on the graph represents the mean of three independent experiments; e_rror bars_ indicate sd. Luciferase values of promoter mutants that differ significantly (by t test) from wild type are indicated with an asterisk.

A number of recent reports have described cooperation between SP1 and ER at promoters containing ERE half-sites within close proximity to SP1 sites (23–29). To investigate the potential cooperation between SP1 and ER-α in MTA3 expression, we employed multiple systems. First, MCF7 breast cancer cells were grown for 3 d in steroid-free media. Under these conditions, basal levels of ER-α expression decline dramatically (data not shown). Using these ER-depleted cells, full-length MTA3 reporter plasmid was cotransfected with an expression plasmid for ER-α. Cells were treated with 10 nm 17β-estradiol for 48 h, and then luciferase assays were performed. In this system, high-level transcription was clearly dependent on ER-α (Fig. 5A). The MTA3 minimal promoter element also demonstrated a clear dependence on ER-α for high-level transcription (Fig. 5A). Mutation of either the SP1 site or the ERE half-site impaired the response of the minimal promoter (Fig. 5A). Next, Drosophila S2 cells were used to confirm these results and to independently analyze the contributions of SP1 to MTA3 expression. Cotransfection of the MTA3 minimal promoter with an expression plasmid for ER-α (cells were treated with 50 nm 17β-estradiol) resulted in very modest levels of expression (Fig. 5B). Addition of SP1 to these transfections resulted in high-level luciferase expression (Fig. 5B), consistent with cooperative activation of transcription by SP1 and ER-α.

Fig. 5.

SP1 and ER-α Synergy at the MTA3 Promoter in MCF7. A, The indicated wild-type and mutant versions of MTA3 promoter luciferase reporters were transfected into MCF7 cells grown in steroid-depleted media. Cells were cotransfected with various amounts of ER-α expression plasmid as indicated in the figure and treated with 10 nm 17β-estradiol. The bar graph depicts normalized luciferase activity for each condition. Each bar on the graph represents the mean of three independent experiments; error bars indicate sd. For each MTA3 promoter construct, luciferase values in the presence of exogenous, transfected ER-α that differ significantly (by t test) from the same construct without exogenous, transfected ER-α are indicated with an asterisk. B, The MTA3 minimal promoter fragment and a mutant version lacking the ERE half-site were cotransfected into Drosophila S2 cells with varying amounts of SP1 and ER-α expression plasmids as indicated in the figure. The bar graph depicts normalized luciferase activity for each condition. Each bar on the graph represents the mean of three independent experiments; error bars indicate sd. For each level of Sp1 plasmid used, normalized luciferase values obtained in the presence of cotransfected ER-α expression plasmid that differ significantly (by t test) from the value in the absence of cotransfected ER-α are indicated with an asterisk.

To ascertain the involvement of SP1 and ER-α in expression of MTA3 in the absence of exogenous expression, RNA interference (RNAi) was employed. When the MCF7 cells are grown in conventional media, ER-α is abundant. Levels of ER-α and SP1 were manipulated using RNAi, and cells were transfected with the full-length MTA3 promoter reporter plasmid. Green fluorescent protein (GFP) small interfering RNA (siRNA) served as a negative control for nonspecific effects of transfection with double-stranded RNA. Luciferase siRNA served as a positive control for knockdown of reporter activity. Depletion of either endogenous ER-α or SP1 led to dramatic declines in reporter gene activity (Fig. 6A). In steroid-depleted media, transcription from the MTA3 promoter-luciferase fusion is dependent on exogenous overexpression of ER-α. Knockdown of ER-α led to a loss of luciferase expression that was similar to either knockdown of luciferase itself or to failure to express ER-α at all. Even in the presence of excess ER-α after overexpression, depletion of SP1 also led to decreased luciferase expression (Fig. 6B). Thus, both SP1 and ER-α are critical to high-level transcription from the MTA3 promoter.

Fig. 6.

Depletion of ER-α or SP1 by RNAi Impairs the Activity of the MTA3 Promoter A, MCF7 cells grown in normal media were treated with the indicated siRNAs (GFP; GL2, firefly luciferase) and transfected with the full-length MTA3 promoter luciferase plasmid. The graph depicts normalized luciferase activity for each condition. Each bar on the graph represents the mean of three independent experiments; error bars indicate sd. Normalized luciferase values differing significantly (by t test) from the value obtained with buffer alone are indicated with an asterisk. B, MCF7 cells were grown in steroid-depleted media for 3 d, transfected with ER-α expression plasmid and full-length MTA3 promoter luciferase plasmid, and treated with 10 nm 17β-estradiol and the indicated siRNAs (GFP; GL2, firefly luciferase). The bar graph depicts the normalized luciferase activity for each condition. Each bar on the graph represents the mean of three independent experiments; error bars indicate sd. Normalized luciferase values differing significantly (by t test) from the value obtained with buffer alone are indicated with an asterisk.

Transient expression assays provide powerful tools for elucidation of _cis_-acting regulatory DNA and identification of _trans_-acting factors involved in gene regulation. However, these systems have well-documented limitations. Thus, RNA interference and chromatin immunoprecipitation (ChIP) were employed to ascertain the potential roles of endogenous SP1 and ER-α in regulation of the endogenous MTA3 locus. ER-α siRNA led to dramatic declines in the steady-state levels of both ER-α and MTA3 proteins (Fig. 7A). Likewise, SP1 siRNA led to dramatic declines in both SP1 and MTA3 protein levels (Fig. 7B). Interestingly, SP1 knockdown also led to declines in MTA2 protein levels (Fig. 7B), consistent with a role for SP1 in the regulation of mouse MTA2 (30). ChIP experiments were performed to determine whether ER-α and SP1 are present at the endogenous MTA3 locus and whether perturbation of either factor by RNAi led to alterations in the behavior of the other. Both ER-α and SP1 were detected by ChIP at the MTA3 minimal promoter (Fig. 7C). Depletion of ER-α by RNAi had no effect on SP1 presence at MTA3 (Fig. 7C). In contrast, depletion of SP1 by RNAi dramatically impaired the ability of ER-α to associate with the endogenous MTA3 promoter.

Fig. 7.

Depletion of SP1 and ER-α by RNAi Led to Loss of Endogenous MTA3 A, MCF7 and T47D cells were grown in normal media and treated with siRNAs specific for either GFP (negative control) or ER-α. The figure depicts immunoblots of whole-cell lysates derived from each condition. Antibodies used for protein detection are indicated on the left side of the figure. B, MCF7 and T47D were grown and treated with siRNA as described in panel A. The immunoblot of whole-cell lysates depicts abundance of the indicated proteins after the various treatments. C, ChIP assay was performed at the endogenous MTA3 locus as described in Materials and Methods. Before preparation of chromatin, cells were treated with the indicated siRNAs. Chromatin was precipitated with SP1 or ER-α antibodies and bound DNA detected by PCR using the primer sets depicted in the figure. The two primer sets indicated in the figure are approximately 2 kb apart.

Cross Talk between MTA Family Members

A role for MTA1 in transcriptional regulation of the MTA3 locus has been recently proposed. In this report, breast cancer cell lines stably expressing MTA1 were compromised in their capacity to drive expression of MTA3 reporter plasmids (31). This study also reported that MTA3 transcription was driven by ER-α, although a different _cis_-acting element was identified. A key difference between the work reported by Mishra et al. (31) and that reported here lies in the reporter constructs used. Mishra et al. (31) used constructs containing the MTA3 transcription start site associated with the noncoding exon 1b. These plasmids lack downstream sequences including the MTA3 minimal promoter elements identified in this work. To independently ascertain what role, if any, MTA1 might play in MTA3 expression driven by the promoter sequences defined in this work, an adenovirus capable of high-level transient expression of MTA1 was constructed. Infection of MCF7 and T47D breast cancer cell lines with MTA1 adenovirus led to dramatic decreases in MTA3 protein as well as in MTA2 (Fig. 8A). Protein levels of other members of the Mi-2/NuRD complex, such as Mi-2, methyl CpG binding domain protein 3 (MBD3), and HDAC1, were unaffected. Thus, transient overexpression of MTA1 led to a shift in the composition of the cellular pool of MTA family members from predominantly MTA2 and MTA3 to predominantly MTA1. Because the steady-state levels of Mi-2, MBD3, and HDAC1/2 were not affected by this treatment, we predict that the Mi-2/NuRD complex has been shifted from predominantly the MTA2 and MTA3 containing versions to predominantly the MTA1 containing version. Despite the dramatic impact on protein levels of MTA2 and MTA3, overexpression of MTA1 had no effect on the endogenous mRNA levels for either of these factors (Fig. 8B). In addition, overexpression of MTA1 had no effect on luciferase expression driven by the full-length MTA3 promoter reporter plasmid (Fig. 8C). We concluded that transient expression of MTA1 has little, if any, repressive effect on the _ci_s-acting MTA3 regulatory sequences defined in this work.

Fig. 8.

Transient Expression of MTA1 Leads to Loss of MTA2 and MTA3 Protein But Not mRNA A, The immunoblot depicts the level of the indicated proteins in whole cell lysates after infection of MCF7 and T47D cells with either control (mock) or MTA1 expressing adenoviruses. Lysates were examined for alterations in the levels of the proteins indicated on the left, including the Mi-2/NuRD complex subunits MTA1, MTA2, MTA3, Mi-2, MBD3, HDAC1, and HDAC2. B, RNA from cells infected with MTA1 or control adenoviruses was analyzed by RT-PCR. The ethidium-stained gel indicates approximate levels of mRNA. Quantitative PCR was also performed as described in Materials and Methods. Relative mRNA levels for MTA2 and MTA3 are depicted in the bar graph. C, MCF7 cells were grown in normal media (left panel) or in steroid-depleted media (right panel). Cells were transfected with the full-length MTA3 promoter luciferase plasmid (cells grown in steroid-depleted media were cotransfected with ER-α expression plasmid). Transfected cells were then infected with either control or MTA1 adenoviruses as indicated in the figure. Normalized luciferase values for each condition are indicated in the figure. Each bar represents the mean of three independent experiments; error bars indicate sd.

DISCUSSION

The MTA3-containing version of the Mi-2/NuRD complex represents an abundant corepressor complex in breast cancer cells (9). This enzyme participates in regulation of gene expression through modulation of local chromatin architecture. Because MTA3 biosynthesis in breast cancer cell lines is entirely dependent on the action of ER, the pattern of genes regulated by MTA3 and its parent complex constitutes a subset of the genetic program downstream of ER in mammary epithelial cells. Characterization of the molecular mechanisms by which the ER regulates the transcription of MTA3 is thus of critical interest. The finding that ER-α directly binds to and regulates the promoter of MTA3 provides novel functionality to the estradiol-ER interaction. The documented involvement of MTA3 in gene repression supports the notion that ligand binding by ER results not only in gene activation, but also in specific gene repression mediated through the Mi-2/NuRD complex. MTA3 is known to participate in transcriptional repression of Snail (9), a gene implicated in epithelial to mesenchymal transitions (32–34). The indirect repression of Snail by ligand-bound ER provides a molecular mechanism whereby estradiol contributes to maintenance of the differentiated epithelial state in mammary epithelia.

Here, we have defined the minimal promoter for the MTA3 locus in MCF7 cells. This _cis_-acting regulatory element contains a binding site for the SP1 transcription factor in proximity to an ERE half-site. SP1 function was essential for the recruitment of ER-α to the endogenous MTA3 promoter chromatin. Thus, MTA3 joins a number of genes characterized by an SP1 site in close proximity to an ERE half-site, including Cathepsin D (23), Progesterone receptor (27, 28), and many others (24–26, 29). At the MTA3 promoter, the transcriptional response to SP1 appears to raise the basal activity of the promoter when assayed in a heterologous system (Drosophila S2 cells). ER addition had an additive effect. In human breast cancer cells, the relative contributions of SP1 and ER to transcriptional activity of the MTA3 minimal promoter were not strictly additive. Rather, synergistic activation resulted from the combination of both factors. These data suggest that the combination of SP1 and ER-α in a mammalian cell results in the recruitment of a specialized set of transcriptional coactivators not present in the Drosophila cell.

The MTA3 promoter offers an interesting example of a promoter driven by an SP1 element in proximity to an ERE half-site. The spacing between SP1 sites and ERE half-sites at such promoters is variable among the many genes for which such regulation has been documented (35–37). At MTA3, there are 73 bp between the center of the SP1 element and the center of the ERE half-site. This distance would place these two elements in immediate proximity on the surface of a single nucleosome (38), with the SP1 element being on one DNA turn around the histone octamer and the ERE half-site being on the other should these two sites occupy a single nucleosome in vivo. Both sites would be phased so that the major groove of each binding site would have approximately the same orientation to the histone octamer as well. Nuclear hormone receptors are known to bind productively to nucleosomal DNA, so long as the major groove containing the recognition sequence is facing away from the histone octamer surface (39–41). We predict that binding of SP1 to nucleosomal DNA increases the binding affinity of ER-α to a low-affinity binding element. The increased binding energy could result from direct protein-protein interactions between SP1 and ER-α, as have been observed (26). Alternatively, binding of SP1 could serve to orient the DNA helix wrapped around a nucleosome such that the major groove of the ERE half-site faces away from the histone octamer surface, making it available for binding by the receptor. However SP1 facilitates transcriptional activation by the ER, the MTA3 promoter provides an example of a locus where local chromatin architecture plays an important role in facilitating proper hormonal regulation of gene expression.

MATERIALS AND METHODS

Cell Culture, Transfection, and Adenovirus infection

Cells were cultured in DMEM (Cellgro, Herndon, VA) with 10% (vol/vol) fetal bovine serum (Invitrogen Life Technologies, Gaithersburg, MD). Steroids were depleted from serum with dextran-charcoal (42). For estrogen depletion, cells were grown in phenol red-free RPMI with 10% dextran-charcoal-stripped serum and 10 μg/ml insulin (Invitrogen Life Technologies, Carlsbad, CA). Cells were transfected with Lipofectamine (Invitrogen Life Technologies) and Fugene 6 (Roche Molecular Biochemicals, Indianapolis, IN). Schneider cell line 2 (S2) derived from Drosophila embryos was cultured in serum-free HyQ SFX-INSECT medium (HyClone, Logan, UT). S2 cell line was transfected with Cellfectin (Invitrogen Life Technologies). Transfected cells were treated with 10 nm or 50 nm 17β-estradiol. ICI 182, 780 (TOCRIS), tamoxifen, and mifepristone (Sigma, St. Louis, MO) were obtained from commercial sources at the highest quality available and cells were grown in normal growth media with 10 μm, 5 μm, and 10 μm, respectively. Adenovirus carrying human MTA1 was prepared as described (43). The virus was purified by CsCl banding with 1011 to 1012 plaque-forming units, and introduced into cells. Luciferase activities were determined as described (9). Values are the means and sd of the results from three independent experiments.

Plasmids

Human _ER_α expression plasmid (pcDNA3 _ER_α) was graciously provided by Dr. Trevor Archer, National Institute of Environmental Health Sciences (Research Triangle Park, NC). The cDNA for _ER_α was subcloned into Drosophila expression vector pAc5.1/V5-His (Invitrogen Life Technologies). Sp1 expression plasmid pPacSp1 was kindly provided by Dr. J. T. Kadonaga, University of California-San Diego, San Diego, CA). The promoter region of human MTA3 was amplified from human genomic DNA and cloned into pGL2-Basic (Promega, Madison, WI). The sets of primers used were as follows:

Wild-type 5′-TGTTCCCTGGTTTACAAAGCATTT-3′

5′-GTCCGCCCGCCCGCCCGCGGA-3′

Intron1 5′-GTAGGCAGGCTCGGCCCGACC-3′

5′-CTGTAATTAAAGGAAGTACAGAAC-3′

Δ1 5′-TTTTGCCTCAGCTATGCACAGTTT-3′

5′-GTCCGCCCGCCCGCCCGCGGA-3′

Δ2 5′-TTATTATGCAGCAATGTCTGGGAG-3′

5′-GTCCGCCCGCCCGCCCGCGGA-3′

Δ3 5′-TTTAGGCACGTAACCAGTTAAGTG-3′

5′-GTCCGCCCGCCCGCCCGCGGA-3′

Δ4 5′-TGGACAGTTGGTTCTGCTCAG-3′

5′-GTCCGCCCGCCCGCCCGCGGA-3′

Δ5 5′-CGGCCCACTCGGCCTTGCTAG-3′

5′-GTCCGCCCGCCCGCCCGCGGA-3′

Δ6 5′-GGCGTACCATCCCCTTTACTT-3′

5′-GTCCGCCCGCCCGCCCGCGGA-3′

Δ7 5′-TGTTCCCTGGTTTACAAAGCATTT-3′

5′-CTGAGCAGAACCAACTGTCCA-3′

Δ8 5′-TTTTGCCTCAGCTATGCACAGTTT-3′

5′-CTGAGCAGAACCAACTGTCCA-3′

Δ9 5′-GAAGAGGTGCTGGGGAGAG-3′

5′-CTAGCAAGGCCGAGTGGGCCG-3′

The MTA3 promoter mutants were generated by point mutagenesis using the QuikChange Site-Directed Mutagenesis kit (Stratagene, La Jolla, CA).

Antibodies and Western Blotting

Antibodies against MTA3, MTA2, Mi2, and MBD3 were previously described (9). Other antibodies used were anti-MTA1, anti-HDAC1, anti-Sp1, and anti-ERα (Santa Cruz), and anti-β-actin (Chemicon, Temecula, CA). Western blot analyses were carried out as described previously (44).

RNA Extraction, cDNA Synthesis, and RT-PCR

Total RNA was extracted by a modified guanidium thiocyanate-phenol-chloroform extraction as described (9). cDNA was synthesized with Moloney murine leukemia virus reverse transcriptase (Invitrogen Life Technologies) using random hexamers as described (9). The cDNA was amplified using the following primers:

MTA3 aFF 5′-ATGGCGGCCAACATGTACCGG-3′

bFF 5′-TATTTGCATCCAAGAGTTTGC-3′

3RR 5′-GATGTTTCTGCTTATCGGTCA-3′

MTA3 (short form specific)

5′-CAAGCTTTCTTCCTTCATACTACA-3′

5′-AAAGAGAGGAAAAGAGAAAATGG-3′

MTA2 5′-TCAATGATATTCGCCAGGAT-3′

5′-AAGTCAGGCCCTTCTGAAAT-3′

MTA1 5′-CTCAAGTCCTACCTGGAGCG-3′

5′-TGGTACCGGTTTCCTACTCG-3′

β-actin 5′-CTCTTCCAGCCTTCCTTCCT-3′

5′-AGCACTGTGTTGGCGTACAG-3′

RT-PCR with real time quantitation was carried out using the iCycler system (Bio-Rad, Hercules, CA) as described (9). Briefly, 3 μg total RNA was reverse transcribed using random hexamers. PCR amplification reactions included SYBR green (Sigma) at a 1:10,000 dilution. Positive control reactions were performed to determine the linear range of detection and establish a standard curve for each transcript. Unknowns were amplified in triplicate and cDNA was diluted so that threshold values fell within the linear range of detection. Transcripts were then quantified from the corresponding standard curve. β-Actin served as an internal control.

ChIP Assay

Cells (5 × 105) were cross-linked (in PBS) with 1% formaldehyde. Crude cell lysates were sonicated as described (45). Immunoprecipitation was performed with specific and control antibodies (Upstate Biotechnology, Inc., Lake Placid, NY). The MTA3 promoter region was detected by PCR amplification with the following primers (5′-GAAGAGGTGCTGGGGAGAG-3′ and 5′-CTGGAAGTGGAGGGCCCTGGGGTCA-3′).

siRNA Knock-Down Experiments

siRNA duplexes (SmartPool) for MTA3 (M-022665-00), Sp1 (M-026959-00), ERα (M-003401-00), GL2 (D-001100-01-20), and GFP (D-001300-01020) were obtained from Dharmacon (Lafayette, CO). siRNAs were transfected using Lipofectamine 2000 in accordance with the manufacturer’s instructions (Invitrogen Life Technologies).

Acknowledgments

We gratefully acknowledge Dr. Trevor Archer (National Institute of Environmental Health Sciences) for the gift of an ER-α expression plasmid. We also thank Dr. Harriet Kinyamu (National Institute of Environmental Health Sciences) and Dr. Nathan Bowen (Emory University) for critical input during the preparation of this manuscript.

This work was supported by a grant from the National Institute of Diabetes and Digestive and Kidney Diseases (DK065961 to P.A.W.). We gratefully acknowledge financial support from the Wilbur and Hilda Glenn Family Foundation.

Abbreviations:

• ChIP,
  Chromatin immunoprecipitation;
• ER,
• ERE,
  estrogen response element;
• EST,
• GFP,
  green fluorescent protein;
• HDAC,
• MBD3,
  methyl CpG binding domain protein 3;
• MTA3,
  metastasis-associated protein 3;
• RNAi,
• siRNA,

1

Humphrey

GW

,

Wang

Y

,

Russanova

VR

,

Hirai

T

,

Qin

J

,

Nakatani

Y

, Howard BH.

2001

Stable histone deacetylase complexes distinguished by the presence of SANT domain proteins CoREST/kiaa0071 and Mta-L1.

J Biol Chem

276

:

6817

–

6824

2

Tong

JK

,

Hassig

CA

,

Schnitzler

GR

,

Kingston

RE

,

Schreiber

SL

1998

Chromatin deacetylation by an ATP-dependent nucleosome remodelling complex.

Nature

395

:

917

–

921

3

Wade

PA

,

Jones

PL

,

Vermaak

D

,

Wolffe

AP

1998

A multiple subunit Mi-2 histone deacetylase from Xenopus laevis cofractionates with an associated Snf2 superfamily ATPase.

Curr Biol

8

:

843

–

846

4

Xue

Y

,

Wong

J

,

Moreno

GT

,

Young

MK

,

Cote

J

,

Wang

W

1998

NURD, a novel complex with both ATP-dependent chromatin-remodeling and histone deacetylase activities.

Mol Cell

2

:

851

–

861

5

Zhang

Y

,

LeRoy

G

,

Seelig

HP

,

Lane

WS

,

Reinberg

D

1998

The dermatomyositis-specific autoantigen Mi2 is a component of a complex containing histone deacetylase and nucleosome remodeling activities.

Cell

95

:

279

–

289

6

Struhl

K

1998

Histone acetylation and transcriptional regulatory mechanisms.

Genes Dev

12

:

599

–

606

7

Grunstein

M

1997

Histone acetylation in chromatin structure and transcription.

Nature

389

:

349

–

352

8

Bowen

NJ

,

Fujita

N

,

Kajita

M

,

Wade

PA

2004

Mi-2/NuRD: multiple complexes for many purposes.

Biochim Biophys Acta

1677

:

52

–

57

9

Fujita

N

,

Jaye

DL

,

Kajita

M

,

Geigerman

C

,

Moreno

CS

,

Wade

PA

2003

MTA3, a Mi-2/NuRD Complex subunit, regulates an invasive growth pathway in breast cancer.

Cell

113

:

207

–

219

10

Pencil

SD

,

Toh

Y

,

Nicolson

GL

1993

Candidate metastasis-associated genes of the rat 13762NF mammary adenocarcinoma.

Breast Cancer Res Treat

25

:

165

–

174

11

Toh

Y

,

Pencil

SD

,

Nicolson

GL

1994

A novel candidate metastasis-associated gene, mta1, differentially expressed in highly metastatic mammary adenocarcinoma cell lines. cDNA cloning, expression, and protein analyses.

J Biol Chem

269

:

22958

–

22963

12

Wade

PA

,

Gegonne

A

,

Jones

PL

,

Ballestar

E

,

Aubry

F

,

Wolffe

AP

1999

Mi-2 complex couples DNA methylation to chromatin remodelling and histone deacetylation.

Nat Genet

23

:

62

–

66

13

Zhang

Y

,

Ng

HH

,

Erdjument-Bromage

H

,

Tempst

P

,

Bird

A

,

Reinberg

D

1999

Analysis of the NuRD subunits reveals a histone deacetylase core complex and a connection with DNA methylation.

Genes Dev

13

:

1924

–

1935

14

Fearon

ER

2003

Connecting estrogen receptor function, transcriptional repression, and E-cadherin expression in breast cancer.

Cancer Cell

3

:

307

–

310

15

Kumar

R

2003

Another tie that binds the MTA family to breast cancer.

Cell

113

:

142

–

143

16

Parker

MG

1998

Transcriptional activation by oestrogen receptors.

Biochem Soc Symp

63

:

45

–

50

17

McDonnell

DP

,

Norris

JD

2002

Connections and regulation of the human estrogen receptor.

Science

296

:

1642

–

1644

18

Nilsson

S

,

Makela

S

,

Treuter

E

,

Tujague

M

,

Thomsen

J

,

Andersson

G

,

Enmark

E

,

Pettersson,

K

,

Warner

M

,

Gustafsson

JA

2001

Mechanisms of estrogen action.

Physiol Rev

81

:

1535

–

1565

19

Kikuno

R

,

Nagase

T

,

Nakayama

M

,

Koga

H

,

Okazaki

N

,

Nakajima

D

,

Ohara

O

2004

HUGE: a database for human KIAA proteins, a 2004 update integrating HUGEppi and ROUGE.

Nucleic Acids Res

32

(database issue):

D502

–

D504

20

Kikuno

R

,

Nagase

T

,

Waki

M

,

Ohara

O

2002

HUGE: a database for human large proteins identified in the Kazusa cDNA sequencing project.

Nucleic Acids Res

30

:

166

–

168

21

Kikuno

R

,

Nagase

T

,

Suyama

M

,

Waki

M

,

Hirosawa

M

,

Ohara

O

2000

HUGE: a database for human large proteins identified in the Kazusa cDNA sequencing project.

Nucleic Acids Res

28

:

331

–

332

22

Abdelrahim

M

,

Samudio

I

, Smith

3rd

R

,

Burghardt

R

,

Safe

S

2002

Small inhibitory RNA duplexes for Sp1 mRNA block basal and estrogen-induced gene expression and cell cycle progression in MCF-7 breast cancer cells.

J Biol Chem

277

:

28815

–

28822

23

Krishnan

V

,

Wang

X

,

Safe

S

1994

Estrogen receptor-Sp1 complexes mediate estrogen-induced cathepsin D gene expression in MCF-7 human breast cancer cells.

J Biol Chem

269

:

15912

–

15917

24

Porter

W

,

Wang

F

,

Wang

W

,

Duan

R

,

Safe

S

1996

Role of estrogen receptor/Sp1 complexes in estrogen-induced heat shock protein 27 gene expression.

Mol Endocrinol

10

:

1371

–

1378

25

Porter

W

,

Saville

B

,

Hoivik

D

,

Safe

S

1997

Functional synergy between the transcription factor Sp1 and the estrogen receptor.

Mol Endocrinol

11

:

1569

–

1580

26

Safe

S

2001

Transcriptional activation of genes by 17β-estradiol through estrogen receptor-Sp1 interactions.

Vitam Horm

62

:

231

–

252

27

Petz

LN

,

Nardulli

AM

2000

Sp1 binding sites and an estrogen response element half-site are involved in regulation of the human progesterone receptor A promoter.

Mol Endocrinol

14

:

972

–

985

28

Petz

LN

,

Ziegler

YS

,

Schultz

JR

,

Kim

H

,

Kemper

JK

,

Nardulli

AM

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

29

Rishi

AK

,

Shao

ZM

,

Baumann

RG

,

Li

XS

,

Sheikh

MS

,

Kimura

S

,

Bashirelahi

N

,

Fontana

JA

1995

Estradiol regulation of the human retinoic acid receptor α gene in human breast carcinoma cells is mediated via an imperfect half-palindromic estrogen response element and Sp1 motifs.

Cancer Res

55

:

4999

–

5006

30

Xia

L

,

Zhang

Y

2001

Sp1 and ETS family transcription factors regulate the mouse Mta2 gene expression.

Gene

268

:

77

–

85

31

Mishra

SK

,

Talukder

AH

,

Gururaj

AE

,

Yang

Z

,

Singh

RR

,

Mahoney

MG

,

Franci

C

,

Vadlamudi

RK

,

Kumar

R

2004

Upstream determinants of estrogen receptor-α regulation of MTA3 pathway.

J Biol Chem

279

:

32709

–

32715

32

Ip

YT

,

Gridley

T

2002

Cell movements during gastrulation: snail dependent and independent pathways.

Curr Opin Genet Dev

12

:

423

–

429

33

Carver

EA

,

Jiang

R

,

Lan

Y

,

Oram

KF

,

Gridley

T

2001

The mouse snail gene encodes a key regulator of the epithelial-mesenchymal transition.

Mol Cell Biol

21

:

8184

–

8188

34

Nieto

MA

2002

The snail superfamily of zinc-finger transcription factors.

Nat Rev Mol Cell Biol

3

:

155

–

166

35

Batistuzzo de Medeiros

SR

,

Krey

G

,

Hihi

AK

,

Wahli

W

1997

Functional interactions between the estrogen receptor and the transcription activator Sp1 regulate the estrogen-dependent transcriptional activity of the vitellogenin A1 io promoter.

J Biol Chem

272

:

18250

–

18260

36

Scholz

A

,

Truss

M

,

Beato

M

1998

Hormone-induced recruitment of Sp1 mediates estrogen activation of the rabbit uteroglobin gene in endometrial epithelium.

J Biol Chem

273

:

4360

–

4366

37

Vyhlidal

C

,

Li

X

,

Safe

S

2002

Estrogen regulation of transferrin gene expression in MCF-7 human breast cancer cells.

J Mol Endocrinol

29

:

305

–

317

38

Luger

K

,

Mader

AW

,

Richmond

RK

,

Sargent

DF

,

Richmond

TJ

1997

Crystal structure of the nucleosome core particle at 2.8 A resolution.

Nature

389

:

251

–

260

39

Wong

J

,

Li

Q

,

Levi

BZ

,

Shi

YB

,

Wolffe

AP

1997

Structural and functional features of a specific nucleosome containing a recognition element for the thyroid hormone receptor.

EMBO J

16

:

7130

–

7145

40

Li

Q

,

Wrange

O

1993

Translational positioning of a nucleosomal glucocorticoid response element modulates glucocorticoid receptor affinity.

Genes Dev

7

:

2471

–

2482

41

Li

Q

,

Wrange

O

1995

Accessibility of a glucocorticoid response element in a nucleosome depends on its rotational positioning.

Mol Cell Biol

15

:

4375

–

4384

42

Reddel

RR

,

Sutherland

RL

1984

Tamoxifen stimulation of human breast cancer cell proliferation in vitro: a possible model for tamoxifen tumour flare.

Eur J Cancer Clin Oncol

20

:

1419

–

1424

43

He

TC

,

Zhou

S

, da

Costa

LT

,

Yu

J

,

Kinzler

KW

,

Vogelstein

B

1998

A simplified system for generating recombinant adenoviruses.

Proc Natl Acad Sci USA

95

:

2509

–

2514

44

Fujita

N

,

Shimotake

N

,

Ohki

I

,

Chiba

T

,

Saya

H

,

Shirakawa

M

,

Nakao

M

2000

Mechanism of transcriptional regulation by methyl-CpG binding protein MBD1.

Mol Cell Biol

20

:

5107

–

5118

45

Fujita

N

,

Wade

PA

2004

Use of bifunctional cross-linking reagents in mapping genomic distribution of chromatin remodeling complexes.

Methods

33

:

81

–

85

Copyright © 2004 by The Endocrine Society

Citations

Views

Altmetric

Metrics

Total Views 792

506 Pageviews

286 PDF Downloads

Since 2/1/2017

Citations

47 Web of Science

×

Email alerts

More on this topic

Related articles in PubMed

Citing articles via

• Latest

• Most Read

• Most Cited

More from Oxford Academic