Estrogen-Responsive Genes Overlap with Triiodothyronine-Responsive Genes in a Breast Carcinoma Cell Line
It has been well established that estrogen plays an important role in the progression and treatment of breast cancer. However, the role of triiodothyronine (T3) remains controversial. We have previously shown its capacity to stimulate the development of positive estrogen receptor breast carcinoma, induce the expression of genes (PR, TGF-alpha) normally stimulated by estradiol (E2), and suppress genes (TGF-beta) normally inhibited by E2. Since T3 regulates growth hormones, metabolism, and differentiation, it is important to verify its action on other genes normally induced by E2. Therefore, we used DNA microarrays to compare gene expression patterns in MCF-7 breast adenocarcinoma cells treated with E2 and T3. Several genes were modulated by both E2 and T3 in MCF-7 cells (Student’s t-test, ). Specifically, we found eight genes that were differentially expressed after treatment with both E2 and T3, including amphiregulin, fibulin 1, claudin 6, pericentriolar material 1, premature ovarian failure 1B, factor for adipocyte differentiation-104, sterile alpha motif domain containing 9, and likely ortholog of rat vacuole membrane protein 1 (fold change > 2.0, pFDR < 0.05). We confirmed our microarray results by real-time PCR. Our findings reveal that certain genes in MCF-7 cells can be regulated by both E2 and T3.
Most breast cancer risk factors are associated with prolonged exposure of the mammary gland to high levels of estrogen. The biological effects of estrogen are predominantly mediated by two estrogen receptors (ER) that bind to estrogen response elements (EREs) in the promoter region of target genes . Although the involvement of thyroid hormones (TH) in the development and differentiation of normal breast tissue has been established [2–4] and epidemiological and experimental studies have associated TH pathologies with an increased risk of breast cancer, the role of TH remains controversial [5–14]. Vorherr  described an increase in the survival of hyperthyroid patients with breast cancer, whereas we have identified a biological link between breast cancer in postmenopausal women and subclinical hyperthyroidism . Most, if not all, major triiodothyronine (T3) actions are mediated by specific high affinity nuclear receptors (thyroid receptor, TR), which are encoded by the two genes THRA and THRB that are also ligand-regulated transcription factors that act via DNA response elements . Recent results have revealed substantial changes in the expression profile of thyroid hormone receptors, suggesting that their deregulation may be involved in breast cancer development .
Thyroid receptor is present in both MCF-7 and MDA-MB-231 breast cancer cell lines . We previously demonstrated that T3 mimics the effects of estradiol (E2) in the ER-positive breast cancer cell line MCF-7, stimulating growth, modulating mRNA transcription of growth factors, and inducing the expression and activity of E2-inducible proteins. In addition, these T3 effects were antagonized by the simultaneous addition of tamoxifen (TAM), which is a competitive inhibitor of E2 that binds to ER. However, we did not observe similar effects in the ER-negative MDA-MB-231 breast cancer cell line, in spite of high amounts of TR. This suggests that in MCF-7 cells both ligands share a common signaling pathway via ER, since the sequence similarity of these hormone receptors allows interactions of TR with EREs or ER with TREs . These results are consistent with those of Zhou-Li et al. (1992), but contradictory to Dinda et al. [20, 21], who found no evidence that T3 competitively displaces E2 from ER. Recently, Hall et al.  showed that both E2 and T3 promoted proliferation in MCF-7 and T47 cell lines, which was suppressed by coadministration of the ER antagonist fulvestrant (ICI 182780), and T3 induced activation of ERE-mediated gene expression (ER, ER, and PR) in MCF-7 cells. They also demonstrated that T3 enhanced the effect of E2 on cell proliferation in a dose-dependent manner.
We have demonstrated that tamoxifen inhibits transforming growth factor alpha (TGFα) gene expression in human breast carcinoma samples treated in vitro with T3 . These results suggest that T3 may play a role in breast cancer development and progression by regulating proliferation, gene expression, and activity of E2-inducible proteins such as progesterone receptor (PR) and TGFα and indicate an interaction between E2 and T3 signaling systems.
Here we systematically examined the transcriptional effects of E2 and T3 in the MCF-7 breast adenocarcinoma cell line using DNA microarrays in order to better understand the actions of these two hormones. We identified their effects on the expression of a large number of genes by using a microarray platform containing 4,608 open reading frame expressed sequence tags (Orestes) . We demonstrated that the expression of eight genes was significantly altered by both E2 and T3 in MCF-7 cells (fold change > 2.0, positive false discovery rate (pFDR) < 0.05). Out of these eight genes, amphiregulin (AREG), fibulin 1 (FBLN1), and claudin 6 (CLDN6) were the most differentially expressed.
2. Materials and Methods
2.1. Cell Line
MCF-7 cells were grown for 14 days before harvesting in RPMI 1640 supplemented with L-glutamine (2.8 mM), insulin (8 mIU/mL), penicillin-streptomycin (100 U/mL), and 5% charcoal-stripped calf serum (FCS) and kept at 37°C in humidified 5% CO2 and air. The medium was changed every two days.
2.2. Cell Treatment Conditions
MCF-7 cells were propagated in 150 cm3 culture flasks until they reached 40% confluence. Before starting hormone treatments, the medium was replaced with phenol red free RPMI 1640 to eliminate all known sources of estrogen . After 24 h (day 0), the medium was changed and cells were treated in triplicate with 10−8 M E2 (Sigma-Aldrich, St. Louis, MO, USA, E8874), 10−8 M T3 (Sigma-Aldrich, T2752), and absolute ethanol (vehicle control) for 72 h. Medium was changed every 24 h. Cells were harvested at the times indicated and cell numbers were counted.
2.3. RNA Isolation and Reverse Transcription
Total RNA was extracted from cultured MCF-7 cells by the guanidinium thiocyanate method and analyzed by electrophoresis using 1% agarose gels. One microgram of total RNA was reverse-transcribed with SuperScript III First-Strand Synthesis System for RT-PCR (Invitrogen, Carlsbad, CA, USA, 18080-051).
2.4. DNA Microarrays
The microarray glass slide was generated in conjunction with the Laboratório de Biologia Computacional (LBC—Computational Biology Laboratory) at the Hospital do Câncer, São Paulo, Brazil, together with the Laboratório de Análise de Expressão Gênica (LGEA—Gene Expression Analysis Laboratory) at the Instituto Ludwig de Pesquisa sobre o Câncer, São Paulo. The slide contained 4,608 genes from the Human Cancer Genome Project Bank, Instituto Ludwig para a Pesquisa do Câncer, São Paulo. Microarray hybridization and analysis were performed as previously described .
2.5. Real-Time RT-PCR
Assay-on-Demand Gene Expression Product (Applied Biosystems, Foster City, CA, USA, 4331182), consisting of unlabeled PCR primers and a TaqMan MGB probe (FAM dye-labeled) optimized to work with the TaqMan Universal PCR Master Mix (P/N 4304437) in an ABI Prism 7700 system (Perkin Elmer Life Sciences, Boston, MA, USA), was employed to quantitatively measure AREG, FBLN1, CLDN6, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA expression. All assays were performed in triplicate. mRNA content was normalized to the GAPDH mRNA level and differences in expression were determined by the Ct method described in the ABI user’s manual (Applied Biosystems).
2.6. Statistical Analysis
“Permutation” Student’s t-test (10,000 permutations) was performed on microarray results without bootstrapping, with a positive false discovery rate (pFDR) less than 0.05, and fold change greater than 2.0.
The influence of T3 on the gene expression profile of MCF-7 cells was examined and compared to the effects of treatment with E2. RNA samples extracted from triplicate samples of MCF-7 cells after 24 h of E2 or T3 treatment were analyzed for E2- or T3-regulated gene expression by comparing to cells treated with vehicle control. Genes with (Student’s t-test) in paired group comparisons were considered as differentially expressed. We verified that 393 genes were differentially expressed after both treatments (up- or downregulated) (Figure 1). After applying a 2-fold change as a cut-off point and a pFDR < 0.05, MCF-7 cells treated with E2 displayed 39 genes that were differentially expressed compared to the control, whereas after T3 treatment only 25 genes were differentially expressed relative to control cells. Eight genes were commonly modulated after treatment with both E2 and T3 (Table 1), although the response to T3 was less pronounced than to E2. Out of these eight genes, AREG, FBLN1, and CLDN6 were strongly expressed after treatment with both T3 and E2. In order to confirm our microarray data, we quantified AREG, FBLN1, and CLDN6 mRNAs using quantitative RT-PCR. Treatment with both E2 and T3 at 10−8 M resulted in overexpression of AREG (26.3- and 13.8-fold more, resp.), FBLN1 (5.3- and 1.9-fold more, resp.), and CLDN6 (4.4- and 2.2-fold more, resp.) (Figure 2).
In breast cancer, several clinicopathological markers are frequently used alone or in combination to assess patient risk. For example, lymph-node stage, tumor size, and histologic grade are important elements of the major prognostic indices, whereas ER status is widely regarded as the primary predictor of response to hormonal (antiestrogen) therapy. Microarray data sets from large studies of breast carcinomas have revealed several underlying signatures associated with the primary physiology of the tumor that have important prognostic and predictive implications.
In previous studies, we have demonstrated that T3, in supraphysiological doses, is able to increase the progression of ER-positive breast cancer, enhancing the expression of genes normally stimulated by E2 and suppressing genes normally inhibited by E2 . Based on those results, we sought to identify genes that are influenced by both hormones in order to identify additional markers of progression.
Many genes were equally modulated by E2 and T3 in MCF-7 breast carcinoma cells (Student’s t-test, ) (Figure 1). Using more stringent criteria (2-fold cutoff and pFDR < 0.05), the number of genes modified by E2 and T3 decreased to eight. Out of these eight genes, AREG, FBLN1, and CLDN6 were strongly expressed by both T3 and E2 treatment. We validated our microarray results using real-time PCR on AREG, FBLN1, and CLDN6 expression (Figure 2).
Amphiregulin (AREG), the most differentially expressed gene, codes for a glycoprotein, that is, a member of the epidermal growth factor family (EGF), the members of which are ligands that can bind EGFR. AREG was discovered in concentrated conditioned medium from MCF-7 cells . Numerous studies have sought to characterize the transcriptional network associated with estrogen response using cell culture experiments. Kenney et al.  inserted pads with recombinant AREG in the breast of oophorectomized rats, which reestablished the premature development of the ductal mammary epithelium and prompted hyperplasia. Therefore, AREG seems to play an intermediary role in maturing the mammary gland and in stimulating the initiation of mammary oncogenesis. Martinez et al.  treated MCF-7 cells with 10−9 M E2, 10−9 M E2 with 10−6 M TAM (E2 + TAM), TAM alone, and vehicle control for 24 hours. They observed that adding TAM to the treatment with E2 decreased AREG mRNA expression by 38%, suggesting that E2 stimulates AREG expression via ER. Using microarrays, Frasor et al.  observed an upregulation of AREG in MCF-7 cells treated with E2, which was reversed by tamoxifen. When Vendrell et al.  treated MVLN cells, a breast carcinoma cell line derived from MCF-7 cells, for 4 days with E2, they observed that AREG was one of the differentially expressed genes using cDNA miniarrays. A recent microarray analysis identified E2-regulated genes in a model in which human breast tissue was transplanted into mice, which were then treated with estradiol . Interestingly, AREG was the most upregulated gene. Several studies have examined AREG expression in breast carcinomas by immunohistochemistry and found that AREG expression is higher in infiltrative breast carcinoma than in normal epithelium and is associated with regional lymph-node metastasis [33–35]. In addition, AREG upregulates the expression of genes associated with invasion . Modulation of the AREG gene by T3 has not been previously reported. While the increase in AREG expression in MCF-7 cells after T3 treatment that we found was lower than the increase after E2 treatment, it was highly significant. Thus, AREG appears to be an important target gene for E2 and T3 in MCF-7 breast cancer cells.
Fibulin 1 (FBLN1) is the prototype member of the fibulin family of ECM proteins and binds to many ECM proteins, including fibronectin, laminin-1, fibrinogen, aggrecan, and versican. Fibulin 1 may be involved in cell motility and anchorage-independent growth of tumor cells  and is overexpressed in breast cancer specimens and breast cancer cell lines [37–39]. FBLN1 was previously reported to be induced by E2 in MCF-7 cells  and in ER-positive ovarian cancer cell lines . We have shown for the first time that FBLN1 is upregulated in breast cancer cell lines treated with T3.
Claudin 6 (CLDN6), a member of the Claudin family, is involved in the formation of the GAP junction . While Offner et al.  reported the expression of CLDN6 in breast cancer cell lines, its role in carcinogenesis remains controversial [44–46]. Wu et al.  noted that cells with a high level of expression of claudin 6 grew slowly and had a higher rate of death than control cells, suggesting that claudin 6 may function as a cancer suppressor. Its downregulation may contribute to the malignant progression of certain types of breast cancers . To our knowledge, this is the first study showing that E2 and T3 modulate CLDN6 expression in a breast cancer cell line.
Some of the other genes that we identified as being regulated by E2 and T3 are directly involved in cell proliferation, such as EMP1, IFNAR2, VMP, FLJ20073, MYC, and AREG, and some are involved in nucleotide binding and/or protein binding, such as PFKFB3, APPBP2, SSFA2, and NALP7. Other genes, such as CSTA, show altered expression in the tissue invasion process during breast carcinogenesis . Expression of BRAP, responsible for ubiquitination of the product of the tumor suppressor gene BRCA1, was increased in our study and in other studies  and FBLN1 and ADAM9 are associated with migration and tissue invasion [51, 52]. We also identified genes that had not been previously correlated with breast cancer, including TERF2IP, which is involved in telomere regulation , IGSF1, which is involved in intracellular adhesion and transcription and is a signal transduction regulator , NMT2, which acts in the protein myristoylation process , FAD104, which positively regulates fat cell differentiation , PCM1, involved in centromere amplification and genomic instability , VWF, involved in cellular adhesion , C1R, which is involved in immune response and complement system activation , and POF1B, which binds nonmuscle actin filaments .
This is the first report of T3 modulation of FBLN1, CLDN6, and AREG. The magnitude of the increase in expression of AREG, FBLN1, and CLDN6 in MCF-7 cells treated with T3 was less than when treated with E2 but highly significant. This may be because these genes have more binding consensus sequences for E2 than T3 receptors. Specifically, within 5,000 bp upstream of the transcriptional initiation site, AREG has 10 EREs and 8 TREs, FBLN1 has 6 TREs and 6 EREs, and CLDN1 has 12 EREs and 10 TREs.
We have shown that T3 treatment results in a gene expression pattern similar to E2 treatment, up- or downregulating a group of the same genes, suggesting that these two hormones can cause similar phenotypes. Our in vitro observation suggests a molecular mechanism by which thyroid hormone can be a relevant factor for breast cancer progression through the induction of genes involved in growth and invasion.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
The authors are grateful to the LIM-24 (FM-USP) and LGEA teams (Cancer Hospital) for their technical support. They also thank FAPESP for financial support (Grant no. 2002/09798-0). This paper has been proofread and edited by native English speakers with related background in BioMed proofreading.
B. K. Vonderhaar and A. E. Greco, “Lobulo-alveolar development of mouse mammary glands is regulated by thyroid hormones,” Endocrinology, vol. 104, no. 2, pp. 409–418, 1979.View at: Google Scholar
M. Lopez-Barahona, I. Fialka, J. M. Gonzalez-Sancho et al., “Thyroid hormone regulates stromelysin expression, protease secretion and the morphogenetic potential of normal polarized mammary epithelial cells,” EMBO Journal, vol. 14, no. 6, pp. 1145–1155, 1995.View at: Google Scholar
T. Yokoe, Y. Iino, H. Takei et al., “Relationship between thyroid-pituitary function and response to therapy in patients with recurrent breast cancer,” Anticancer Research, vol. 16, no. 4, pp. 2069–2072, 1996.View at: Google Scholar
T. Yokoe, Y. Iino, H. Takei et al., “Changes of cytokines and thyroid function in patients with recurrent breast cancer,” Anticancer Research, vol. 17, no. 1, pp. 695–699, 1997.View at: Google Scholar
M. B. Goldman, “Thyroid diseases and breast cancer,” Epidemiologic Reviews, vol. 12, pp. 16–28, 1990.View at: Google Scholar
D. P. Rose and T. E. Davis, “Plasma thyroid-stimulating hormone and thyroxine concentrations in breast cancer,” Cancer, vol. 41, no. 2, pp. 666–669, 1978.View at: Google Scholar
O. Takatani, T. Okumoto, H. Kosano, M. Nishida, H. Hiraide, and S. Tamakuma, “Relationship between the levels of serum thyroid hormones or estrogen status and the risk of breast cancer genesis in Japanese women,” Cancer Research, vol. 49, no. 11, pp. 3109–3112, 1989.View at: Google Scholar
O. Turken, Y. NarIn, S. DemIrbas et al., “Breast cancer in association with thyroid disorders,” Breast Cancer Research, vol. 5, no. 5, pp. R110–R113, 2003.View at: Google Scholar
H. Vorherr, “Thyroid function in benign and malignant breast disease,” European Journal of Cancer and Clinical Oncology, vol. 23, no. 3, pp. 255–257, 1987.View at: Google Scholar
P. P. Saraiva, N. B. Figueiredo, C. R. Padovani, M. M. Brentani, and C. R. Nogueira, “Profile of thyroid hormones in breast cancer patients,” Brazilian Journal of Medical and Biological Research, vol. 38, no. 5, pp. 761–765, 2005.View at: Google Scholar
S. J. Conde, R. A. M. Luvizotto, M. T. Sibio, M. L. H. Katayama, M. M. Brentani, and C. R. Nogueira, “Tamoxifen inhibits transforming growth factor-α gene expression in human breast carcinoma samples treated with triiodothyronine,” Journal of Endocrinological Investigation, vol. 31, no. 12, pp. 1047–1051, 2008.View at: Google Scholar
Y. Berthois, N. Pourreau-Schneider, and P. Gandilhon, “Estradiol membrane binding sites on human breast cancer cell lines. Use of a fluorescent estradiol conjugate to demonstrate plasma membrane binding systems,” Journal of Steroid Biochemistry, vol. 25, no. 6, pp. 963–972, 1986.View at: Google Scholar
N. J. Kenney, G. H. Smith, I. G. Maroulakou et al., “Detection of amphiregulin and Cripto-1 in mammary tumors from transgenic mice,” Molecular Carcinogenesis, vol. 15, pp. 44–56, 1996.View at: Google Scholar
L. Martinez, J. A. Castilla, T. Gil et al., “Thyroid hormones in fibrocystic breast disease,” European Journal of Endocrinology, vol. 132, no. 6, pp. 673–676, 1995.View at: Google Scholar
J. Frasor, F. Stossi, J. M. Danes, B. Komm, C. R. Lyttle, and B. S. Katzenellenbogen, “Selective estrogen receptor modulators: discrimination of agonistic versus antagonistic activities by gene expression profiling in breast cancer cells,” Cancer Research, vol. 64, no. 4, pp. 1522–1533, 2004.View at: Publisher Site | Google Scholar
J. A. Vendrell, F. Magnino, E. Danis et al., “Estrogen regulation in human breast cancer cells of new downstream gene targets involved in estrogen metabolism, cell proliferation and cell transformation,” Journal of Molecular Endocrinology, vol. 32, no. 2, pp. 397–414, 2004.View at: Publisher Site | Google Scholar
C.-F. Qi, D. S. Liscia, N. Normanno et al., “Expression of transforming growth factor α, amphiregulin and cripto-1 in human breast carcinomas,” British Journal of Cancer, vol. 69, no. 5, pp. 903–910, 1994.View at: Google Scholar
L. Panico, D. 'Antonio A, G. Salvatore et al., “Differential immunohistochemical detection of transforming growth factor alpha, amphiregulin and CRIPTO in human normal and malignant breast tissues,” International Journal of Cancer, vol. 65, no. 1, pp. 51–56, 1996.View at: Google Scholar
A. Bardin, F. Moll, R. Margueron et al., “Transcriptional and posttranscriptional regulation of fibulin-1 by estrogens leads to differential induction of messenger ribonucleic acid variants in ovarian and breast cancer cells.,” Endocrinology, vol. 146, pp. 760–768, 2005.View at: Google Scholar
G. M. Clinton, C. Rougeot, J. Derancourt et al., “Estrogens increase the expression of fibulin-1, an extracellular matrix protein secreted by human ovarian cancer cells,” Proceedings of the National Academy of Sciences of the United States of America, vol. 93, no. 1, pp. 316–320, 1996.View at: Publisher Site | Google Scholar
K. Kishimoto, X. Yuan, H. Auchincloss Jr., A. H. Sharpe, D. A. Mandelbrot, and M. H. Sayegh, “Mechanism of action of donor-specific transfusion in inducing tolerance: role of donor MHC molecules, donor co-stimulatory molecules, and indirect antigen presentation,” Journal of the American Society of Nephrology, vol. 15, no. 9, pp. 2423–2428, 2004.View at: Publisher Site | Google Scholar