Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2008 Mar 26;3(3):e1859.
doi: 10.1371/journal.pone.0001859.

Genomics of signaling crosstalk of estrogen receptor alpha in breast cancer cells

Peter Dudek  1 , Didier Picard
Affiliations

Genomics of signaling crosstalk of estrogen receptor alpha in breast cancer cells

Peter Dudek et al. PLoS One. .

Abstract

Background: The estrogen receptor alpha (ERalpha) is a ligand-regulated transcription factor. However, a wide variety of other extracellular signals can activate ERalpha in the absence of estrogen. The impact of these alternate modes of activation on gene expression profiles has not been characterized.

Methodology/principal findings: We show that estrogen, growth factors and cAMP elicit surprisingly distinct ERalpha-dependent transcriptional responses in human MCF7 breast cancer cells. In response to growth factors and cAMP, ERalpha primarily activates and represses genes, respectively. The combined treatments with the anti-estrogen tamoxifen and cAMP or growth factors regulate yet other sets of genes. In many cases, tamoxifen is perverted to an agonist, potentially mimicking what is happening in certain tamoxifen-resistant breast tumors and emphasizing the importance of the cellular signaling environment. Using a computational analysis, we predicted that a Hox protein might be involved in mediating such combinatorial effects, and then confirmed it experimentally. Although both tamoxifen and cAMP block the proliferation of MCF7 cells, their combined application stimulates it, and this can be blocked with a dominant-negative Hox mutant.

Conclusions/significance: The activating signal dictates both target gene selection and regulation by ERalpha, and this has consequences on global gene expression patterns that may be relevant to understanding the progression of ERalpha-dependent carcinomas.

PubMed Disclaimer

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1
Figure 1. Analysis of gene expression profiles in response to different signals.
(A) Venn diagrams of gene sets regulated by E2/GFs/cAMP. Top and bottom panels: genes regulated by at least 2-fold (total of 220 genes) and 1.5-fold (922 genes), respectively. (B) Hierarchical clustering analysis with Eisen tree of 2-fold gene set. (C) Clustering analysis using the gene sets defined separately by a 1.5-fold cut-off for each treatment. Green and red colour bars represent repression and induction, respectively, compared to control (untreated). Black bars indicate no change.
Figure 2
Figure 2. Verification of microarray data by Q-PCR.
Top panel: dendrogram of fold expression ratios from the microarray data of the genes DNM3, BCL3, RGS16, TFF1 (pS2), RAP1GAP, EPHB3 and CCNG2. Yellow and cyan indicate induction and repression, respectively. Bottom panel: Q-PCR results for three conditions (E2, GFs and cAMP). Log-scale graph of the fold expression values calculated with the ΔΔCt method using DNM3 as internal control. Values shown are the means of triplicate samples ± standard deviation.
Figure 3
Figure 3. Ligand-independent ERα-mediated gene expression profiles.
(A) Log expression ratios of the ICI-affected GF-regulated (65 genes) and cAMP regulated (28 genes) gene sets at the top and bottom, respectively. (B) Hierarchical cluster analysis with Eisen tree of gene sets shown in panel A. Interesting clusters are numbered on the right. Asterisks denote genes with unique response that are discussed in the text. Colour code as in Figure 1.
Figure 4
Figure 4. Interplay between OHT and signaling crosstalk.
(A) Hierarchical cluster analysis with Eisen tree of genes affected by OHT in GF- and cAMP-treated samples. Gene set includes 20 genes that also respond in the same manner to OHT alone (marked with dotted lines). Some sets of genes with interesting behaviors are marked with solid lines and listed. (B) Venn diagrams illustrating overlap between gene sets obtained with OHT alone or in combination with cAMP or GFs. (C) Selected patterns of genes most strongly affected by the agonistic activity of OHT. Arrows mark rows of most relevant pairwise comparisons.
Figure 5
Figure 5. Computational modelling of transcription factor binding site modules in ERα-dependent/cAMP-regulated promoters.
Top panel: microarray data of those genes from this set for which promoter sequences (−1000 bp to +100 bp) were available (21 out of 28). Middle panel: list of ten highest scoring 2-element transcription factor binding site (2xTFBS) modules. Bottom panel: list of three optimized modules containing 3 transcription factor binding-sites (3xTFBS) each; hashed lines, partial module coverage; grey boxes, conserved 2xTFBS module transmitted to a 3xTFBS module; black box, conserved 1xTFBS transmitted to a 3xTFBS module; asterisks denote genes with no detectable 3xTFBS modules. + and − indicate strand orientation, and digits indicate distances in base pairs between transcription factor elements.
Figure 6
Figure 6. A Hox protein mediates cAMP-induced agonistic switch of OHT for MCF7 proliferation.
(A) MCF7 proliferation assay. Equal numbers of cells (2×105) were seeded onto plates and then treated over a period of 6 days with vehicle control (0.1% ethanol), 10 nM E2, 1 mM 8-Br-cAMP (cAMP), 1 µM OHT or combinations thereof. (B) Venn diagram showing overlap between the cAMP-regulated/ERα-dependent genes and genes OHT-regulated by cAMP (corrected for E2 and OHT behavior as described in the text). The overlapping 14 genes include the 11 genes used in the promoter analysis of Figure 5 (ARF4L, ARFIP1, CLPS, DPP6, GRIA3, KLK1, NDUFB2, P2RY13, RAFTLIN, RAG2, UBCE7IP5) and the unannotated transcripts/genes KIAA0406, KIAA0556 and Hs.541338 (see Supporting Information Tables S3 and S4). (C) Effect of dominant-negative transcription factor mutants on proliferation of MCF7 cells. Cells were transfected with expression vectors for dnPax5, dnCREB or dnHox, or empty expression vector (control). Data in panels A and C are representative of two independent experiments.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Katzenellenbogen BS, Norman MJ. Multihormonal regulation of the progesterone receptor in MCF-7 human breast cancer cells: interrelationships among insulin/insulin-like growth factor-I, serum, and estrogen. Endocrinology. 1990;126:891–898. - PubMed
    1. Picard D. SCOPE/IUPAC project on environmental implications of endocrine active substances: Molecular mechanisms of cross-talk between growth factors and nuclear receptor signaling. Pure Appl Chem. 2003;75:1743–1756.
    1. Ignar-Trowbridge DM, Nelson KG, Bidwell MC, Curtis SW, Washburn TF, et al. Coupling of dual signaling pathways: epidermal growth factor action involves the estrogen receptor. Proc Natl Acad Sci USA. 1992;89:4658–4662. - PMC - PubMed
    1. Ignar-Trowbridge DM, Teng CT, Ross KA, Parker MG, Korach KS, et al. Peptide growth factors elicit estrogen receptor-dependent transcriptional activation of an estrogen-responsive element. Mol Endocrinol. 1993;7:992–998. - PubMed
    1. Curtis SW, Washburn T, Sewall C, DiAugustine R, Lindzey J, et al. Physiological coupling of growth factor and steroid receptor signaling pathways: estrogen receptor knockout mice lack estrogen-like response to epidermal growth factor. Proc Natl Acad Sci USA. 1996;93:12626–12630. - PMC - PubMed

Publication types