The transcription factor snail mediates epithelial to mesenchymal transitions by repression of estrogen receptor-alpha

Abstract

The estrogen receptor (ER)-alpha (ESR1) is a key regulatory molecule in mammary epithelial cell development. Loss of ER-alpha in breast cancer is correlated with poor prognosis, increased recurrence after treatment, and an elevated incidence of metastasis. A proposed molecular pathway by which ER-alpha acts to constrain invasive growth in breast cancer cells involves direct, ER-alpha-dependent expression of metastasis-associated protein 3, a cell-type-specific component of the Mi-2/NuRD chromatin remodeling complex. MTA3 in turn represses expression of Snail, a transcription factor linked to epithelial to mesenchymal transition and cancer metastasis. To elucidate its role(s) in epithelial to mesenchymal transition (EMT), we expressed Snail in the noninvasive, ER-alpha-positive MCF-7 cell line. Snail expression led to decreased cell-cell adhesion and increased cell invasiveness. Furthermore, we observed loss of ER-alpha expression at both the RNA and protein level that was accompanied by direct interaction of Snail with regulatory DNA sequences at the ESR1 locus. A consequence of loss of ER-alpha function in this system was the increased abundance of key components of the TGF-beta signaling pathway. Thus, cross-talk among ER-alpha, Snail, and the TGF-beta pathway appears to control critical phenotypic properties of breast cancer cells.

Figure 1. An inverse relationship exists between Snail and ER-α in breast cancer cell lines

A) Immunoblots using whole cell lysates from two ER-α positive cell lines (MCF-7 and T-47D) and an ER-α negative cell line (MDA-MB-231) were performed using antibodies against ER-α, Occludin, Snail, and β-Actin. B) MCF-7 cells were targeted with siRNA against ER-α or GFP (normalization control), and RNA isolated after 48 hours. RT-PCR was performed using primers directed against ER-α and Snail. Data shown are an average of three biological experiments.

Figure 2. Snail expression decreases cell-cell adhesion properties of MCF-7 cells

MCF-7 cells infected with control or Snail-adenovirus for 3 days were trypsinized, counted, and 10,000 cells were resuspended in DMEM/F-12 medium lacking both FBS and phenol-red and placed over a bed of Agar in a 96-well plate. After a period of 24 hours, they were observed under a 5X lens of a Zeiss imaging microscope. Data are representative of three independent biological experiments.

Figure 3. Snail increases invasiveness of MCF-7 cells through matrigel

MCF-7 cells infected with control or Snail-adenovirus for 3 days were trypsinized, counted, and 20,000 cells were resuspended in DMEM/F-12 medium lacking FBS and pipetted carefully over a membrane coated with Matrigel in the upper well of a Boyden chamber. The lower chamber contained the same media supplemented with 10% FBS as the chemoattractant. After a period of 24 hours, the membranes were removed and the cells that had invaded through the matrigel to the other side were stained with Diff-Quik and counted under the microscope. Data shown are an average of three biological replicates. A t-distribution analysis was performed on the samples within each set (single asterisk * indicates p-value≤0.05, ** indicates p-value ≤ 0.01). A two-sample t-test comparing Snail to control-treated samples showed a p-value of 0.01.

Figure 4. Snail expression causes a decrease in ER-α, while causing an increase in TGF-β family RNA levels in MCF-7 cells

MCF-7 cells were infected with Snail or control-adenovirus over a period of 4 days and RNA expression detected via RT-PCR. RNA fold change was calculated after normalizing to 18S RNA for (A) ER-α (ESR1) and Occludin (OCLN)(B) Snail (SNAI1), TGF-β2 and TGF-βRII (note log scale), (C) pS2 and (D) Slug (SNAI2). Data shown are an average of three biological replicates. A t-test was performed on the samples within each set (single asterisk * indicates p-value≤0.05, ** indicates p-value ≤ 0.01). MCF-7 cells were infected with Snail or control-adenovirus for a period of 2 days and total protein was isolated. Equal amounts were loaded and analyzed via immunoblotting (E) using antibodies against ER-α, Occludin, and Snail. Actin (ACTB) was used as a loading control. Data are representative of at least three biological replicates. MCF-7 cells were infected with Snail or control-adenovirus for a period of 2 days and the cellular distribution of ER-α (F) and occludin (G) were visualized as described in materials and methods. Data are representative of at least three biological replicates. Arrowheads indicate adenovirus-infected cells.

Figure 5. Snail associates with the ER-α locus

(A) Schematic of the ER-α locus showing location of the primer sets used in ChIP analysis. The boxes represent the exons, while the hash marks represent clusters of putative Snail binding sites. (B) MCF-7 cells were infected with Snail or control adenovirus and ChIP performed as described in materials and methods. Amplification of input and DNA precipitated with either a control IgG or anti-Snail antibody is shown for the indicated primer sets. The data represent fold-change of material immunoprecipitated with the Snail antibody versus with IgG, and normalized to Snail binding at the beta-Actin intron, which has no known Snail binding sites.

Figure 6. Snail elicits changes in histone H3K9 modifications at the ER-α promoter

(A) Schematic representation of the primer sets used in chromatin IP covering ∼9.3Kb of ER-α promoter region. The putative Snail binding sites are depicted as well. (B) Chromatin IP (ChIP) was performed using antibodies against acetylated and trimethylated histone H3K9, with Rabbit IgG as a negative control. A survey of ChIP DNA across the ER-α locus reveals a peak of histone H3K9 acetylation over the region encompassed by primer set ‘M’ in MCF-7 cells (top left panel). This peak is also present in MCF-7 cells treated with control (bottom left) or Snail (bottom right) adenovirus. However, the peak of acetylation in Snail-infected cells is decreased about two-fold relative to control-treated cells. In MDA-MB-231 cells (top right), on the other hand, there is virtually no histone H3K9 acetylation; instead a peak of histone H3K9 trimethylation is seen around primer set ‘M’. Data are representative of three independent biological replicates. The data are normalized as indicated in the materials and methods.

Figure 7. A model for interactions between ER-α, Snail and the TGF-β pathways in breast cancer

TGF-β and ER-α pathways are mutually exclusive in breast cancer cells, and Snail helps the switch from ER-α to TGF-β regulation. Features of this model are described in more detail in the text.