An induction of microRNA, miR-7 through estrogen treatment in breast carcinoma

Background Estrogen plays an important role in the development of estrogen-dependent breast carcinoma. Recently, several studies demonstrated a possible involvement of several micro RNAs (miRNAs) in the development of resistance to endocrine therapy in breast cancer patients, but the correlation between estrogen actions and miRNA expression in breast carcinoma still remains largely unknown. Therefore, in this study, we examined the in vitro effects of estrogen upon miRNA expression profiles in breast carcinoma. Methods We first screened the miRNA expression profiles induced by 17β-Estradiol (E2) using RT2 miRNA PCR Array in the ER-positive breast carcinoma cell line MCF-7. We identified miR-7 as the important miRNA associated with estrogen actions in these cells and further examined the changes of estrogen-dependent EGFR expression by miR-7 in ER-positive or -negative breast carcinoma cell lines including MCF-7. We also evaluated the correlation between miR-7 and EGFR expression in breast carcinoma cells derived from 21 patients using laser capture microdissection combined with quantitative reverse transcriptase-PCR. Results Seventeen miRNAs were significantly induced by E2 treatment in the MCF-7 cell line. Among 17 miRNAs induced by estradiol treatment, only miR-7 expression was significantly decreased by subsequent ICI treatment. The expression of miR-7 was up-regulated 2.94-fold by E2 treatment. miR-7 was reported to suppress epidermal growth factor receptor (EGFR) expression in several human malignancies. Transfection of miR-7 significantly suppressed EGFR mRNA levels in MCF-7 cells. Depletion of E2 from cell culture media also increased the expression level of EGFR mRNA in MCF-7 and T-47D cells but not in ER-negative, MDA-MB-231 and SK-BR-3 cells. We also evaluated the status of miR-7 in breast carcinoma tissues, but the correlation between the status of miR-7 and EGFR in carcinoma cells isolated by laser capture microscopy was not detected. Conclusions These results suggest that miR-7 may play a role in the development of resistance to endocrine therapy in breast cancer patients through regulating EGFR expression of carcinoma cells.


Background
Breast cancer is one of the leading causes of cancer deaths in women. Estrogen stimulates breast cancer cell proliferation by binding to the estrogen receptor (ER) and the estrogen-ER complex interacts directly with estrogen response element (ERE), located in the promoter regions of the target genes to either activate or repress gene expression. Aromatase inhibitors are currently employed in clinical practice for treatment of postmenopausal ERpositive breast cancer patients, but the acquired endocrine resistance has become one of the major problems in the clinical management of these patients. Several studies have demonstrated that activation of other cell proliferation pathways involving epidermal growth factor receptor (EGFR) and/or insulin-like growth factor 1 receptor (IGF-1R) in breast cancer cells can be at least partly responsible for the subsequent development of resistance to endocrine therapy [1,2]. The molecular mechanisms of conversion from ER signals to other proliferative signals in breast cancer cells however remain largely unknown.
MicroRNAs (miRNAs) are small (20-24 nucleotides), non-coding RNA gene products, which post-transcriptionally modulate gene expression by negatively regulating the stability or transcriptional efficiency of their target mRNAs. miRNAs have demonstrated in playing key regulating roles in various cellular processes, including cell cycle, differentiation, development and metabolism [3][4][5]. Aberrant expressions of miRNAs have been reported to be associated with malignant phenotypes in various human tissues, and some miRNAs might also function as tumor suppressor genes or oncogenes [6]. Several miRNAs are specifically expressed in breast cancer tissues [7]. In particular, miR-206 was reported to be associated with estrogen signals by targeting ESR1 (ERα) of breast carcinoma cell lines [8], and miR-221/222 could regulate estrogenic signals through ERα, resulting in the development of antiestrogen resistant breast cancer [9,10]. Estradiol (E2)-regulated miRNAs were also expressed in breast carcinoma cell lines. Maillot et al. [11] reported that eight miRNAs were repressed by E2 and might be involved in the tumor response to hormonal therapy, but the expression of E2/ ER-inducible miRNAs were different between published reports [8][9][10][11]. In general, miRNAs were demonstrated to be regulated by estrogen and their functions still remain unclear.
Therefore, in this study, we first screened miRNAs induced by E2 treatment using RT-PCR array in the ERpositive breast cancer cell line MCF-7, among human cancer-related miRNA expression profiles. We then examined the possible correlation between EGFR mRNA expression and ER/E2-inducible miRNA which was screened out from the above analysis using breast carcinoma cell lines and tissues. Clinicopathological features of these cases were summarized in Table 1. For laser capture microscopy evaluation, a total of 20 cases of breast carcinoma frozen tissues embedded in OCT compound and 21 cases of 10% formalin-fixed and paraffin embedded (FFPE) tissue specimens were evaluated.

Breast carcinoma cell lines and culture conditions
The human breast carcinoma cell lines MCF-7 and SK-BR-3 were provided by the Cell Resource Center for Biomedical Research, Tohoku University (Sendai, Japan), and MDA-MB-231 and T-47D were commercially obtained from the American Type Cell Culture (Manassas, VA). All the cell lines were cultured in RPMI 1640 (Sigma-Aldrich, St Louis, MO) and supplemented with 10% fetal bovine serum (FBS; Nichirei Biosciences, Tokyo, Japan).

Analysis of miRNA in breast carcinoma cell lines
Total RNA was carefully extracted from breast carcinoma cell lines using the TRIzol method (Invitrogen, Carlsbad, CA). cDNA for mRNA-PCR was synthesized using a QuantiTect Reverse Transcription kit (QIAGEN, Hilden, Germany), and cDNA for miRNA-PCR was synthesized using a RT 2 miRNA First Strand Kit (QIAGEN).
Analysis of miRNAs in human breast cancer cases using laser capture microdissection (LCM) The tissue sections of 8 μm were used for LCM/qPCR. LCM was performed using mmi CellCut (MMI Molecular Machines and Industries, Flughofstrasse, Glattbrugg,

Effects of E2 depletion on cultured cells
For estrogen depletion studies, cells were cultured in phenol-red-free RPMI 1640 (Sigma-Aldrich) supplemented with 10% DCC-FBS. Estrogen and androgen concentrations in this medium were below the detection limits measured by LC-MS/MS. After five days, EGFR mRNA levels were evaluated using the real time PCR described above. MiR-7 expression levels in MCF-7 cultivated with normal FBS and DCC-FBS were also evaluated using the quantitative RT-PCR method described above.

Statistical analysis
Statistical analysis was performed using StatView 5.0J software (SAS Institute, Cary, NC, USA). qRT-PCR data were analyzed using analysis of variance followed by the post-hoc Bonferroni/Dunnet multiple comparison test. A p-value < 0.05 was considered to be statistically significant.

miRNA expression profiles in MCF-7 cell treated with E2
We examined the miRNA profiling in MCF-7 cells treated with 10 pM E2 or combination of 1 μM ICI and 10 pM E2 in order to identify the E2-ER regulated miRNAs. Those with expression ratios above 2.0-fold compared to control cells following 24 hours were summarized in Fig. 1 and Table 2. 17 miRNAs were significantly up-regulated following E2 treatment (Fig. 1, Table 2). Among these 17 miRNAs induced by E2, only mir-7 was significantly decreased in its expression by the treatment of ICI compound (Table2). Therefore, in this study, we focused on expression of miR-7 and further examined biological feature of miR-7 as an estrogen inducible miRNA in breast carcinoma cells.
Effects of miR-7 on EGFR mRNA expression in MCF-7 Analysis of potential target genes of miR-7 was evaluated using TargetScan, Microcosm Targets, and microRNA.org target prediction. Results indicated EGFR as one of the target genes of miR-7. Therefore, in this study, we focused on the correlation between EGFR mRNA and miR-7 expression. Results of miR-7 transfection assay demonstrated that EGFR mRNA was significantly decreased in MCF-7 transfected with miR-7 compared to those transfected with control scramble RNA (Fig. 2). Other EGFR family such as HER2, HER3, and HER4 were not potential target gene of miR-7 in three target prediction algorisms described above.

Effects of E2 on EGFR mRNA in breast carcinoma cells
The amounts of EGFR mRNA were significantly increased by the removal of E2 from the culture medium in MCF-7 cells. This EGFR induction was significantly decreased by an addition of 10 nM E2, but the effect was significantly decreased by 1μM ICI treatment in the E2depleted MCF-7 cells (Fig. 3A). The ICI treatment significantly increased the levels of EGFR mRNA expression in MCF-7 cells without E2 depletion (Fig. 3A). The miR-7 expression level was significantly decreased by the depletion of E2 in MCF-7 cells (Fig. 3B).
In T-47D cells, the expression levels of miR-7 (Fig. 4A) and EGFR mRNA (Fig. 4B) were increased by E2 (10nM) treatment and depletion of E2, respectively. However, the changes did not reach statistical significance. E2 depletion in cell culture medium did not influence the levels of EGFR mRNA in ER-negative MDA-MB-231 and SK-BR-3 cell lines (Fig. 4C, D).

Expression of miR-7 in breast cancer tissues
MiR-7 was detected in both fresh frozen and FFPE breast carcinoma tissues. No significant correlations were detected between the levels of miR-7 expression and ER (Fig. 5A) or PR (Fig. 5B) LIs in 21 ER-positive breast carcinoma cases. There were no significant correlations between miR-7 and ER (Fig. 5C) or EGFR (Fig. 5D) mRNA detected in carcinoma cells isolated by LCM from 20 ER-positive fresh frozen breast carcinoma tissues in this study.

Discussion
Many investigators have reported the regulation of miR-NAs in estrogen actions in breast carcinoma cells [7][8][9][10][11]. However, these reported results are not consistent. Maillot et al. [11] reported that eight miRNAs (miR-181a, miR-26a, miR-181b, miR-26b, miR-200c, miR-21, miR-23b and miR-27b) were down-regulated by E2 in MCF-7 cells but these miRNAs were not identified among E2-repressed miRNAs in our present study. We identified miR-21 and miR-27b as E2-induced miRNAs and Bhat-Naksharti et al. [12] also reported an increment of miR-21 by E2 treatment. These discrepancies regarding the correlation between estrogen actions and miRNA profiles in breast carcinoma cells among reported studies may be due to the different cell culture conditions including the treatment time and the dose of E2. It was reported that E2-induced increment of miRNAs started to be detected only after 18 hours of E2 treatment because of the stability of the mature miRNA [11]. Therefore, we evaluated alterations of miRNA by E2 as reported by Maillot et al [11]. In our present study, among 17 miRNAs induced by E2 treatment, ICI treatment significantly decreased the expression of miR-7 alone. The miRNAs induced by estrogen treatment may be inhibited by the co-treatment with ER antagonist, but several studies demonstrated that some miRNAs were induced by ICI treatment alone in endometrial cells and breast carcinoma cells including MCF-7 cells [13][14][15]. Further studies including specific inhibition assays using siRNA of ERα are required to clarify an induction of miRNAs through the ERα. MiR-7 belongs to intronic miRNA and is present in an intron of hnRNP K (heterogeneous nuclear ribonucleic protein K) in both insects and mammals [16,17]. The amounts of hnRNP K expression were also reported to be relatively higher in ER + /PR + primary breast tumors compared to ER -/PRtissues [18]. In addition, ERE is located in the 5' flanking region of the hnRNP K gene [18]. These results suggested that an up-regulation of miR-7 by E2 was due to the splicing of hnRNP K, which was also increased by E2 treatment in MCF-7 cells [18]. In this study, we also examined the expression of miR-7 in T-47D cells. The miR-7 expression was increased by E2 treatment in T-47D cells but the change did not reach statistical significance (p=0.0508). ERβ expression was detected predominantly in T-47D cells but not in MCF-7 cells (data not presented). Therefore, the induction of miR-7 by E2 treatment through the hnRNP K is considered to be influenced by the expression patterns of ER subtypes in the cells. In our present study, we focused on the change in EGFR expression because miR-7 was reported to inhibit EGFR expression and subsequently suppress cell proliferation in several human carcinoma cell lines [19][20][21]. E2-depletion from culture media of MCF-7 cells suppressed miR-7 expression and increased EGFR expression. In addition, EGFR mRNA expression was also suppressed by miR-7 transfection in MCF-7 cells. These results all indicated that the signals regulated E2-induced miR-7 expression through EGFR, and the inhibition or suppression of ER-mediated signaling suppressed miR-7 and subsequently increased EGFR mRNA expression in ER-positive breast carcinoma cells. The EGFR overexpression by E2 depletion may be considered within the spectrum of the survival mechanisms to avoid cell death in carcinoma cells. This finding may also account partly for an emergence of alternative proliferative pathways for the survival of carcinoma cells under endocrine treatment. However, the detailed mechanisms regarding this 'switching theory' in breast carcinoma cells treated with endocrine therapy remain largely unknown. The results of our present study also suggest that miR-7 may be involved in these cell survival mechanisms of breast carcinoma cells under estrogen depletion, but further investigations are warranted.
An overexpression of growth factor receptors such as EGFR has been also reported as one of the mechanisms of resistance to endocrine therapy in ER-positive breast cancer patients [1,2,22]. Many clinical trials studied the efficacy of adding an inhibitor of growth factor receptors such as gefitinib to endocrine therapy such as tamoxifen or aromatase inhibitors [23][24][25][26]. MiR-7 was also reported to regulate IGF-1R expression in tongue squamous cell carcinoma cells [27]. In breast carcinoma cells, an interaction between ERα and IGF-1R has been recognized to enhance proliferative activity [28][29][30]. In addition, E2 treatment was also known to increase the expression of IGF-1R protein in MCF-7 cells [31]. In our present study, IGF-1R mRNA was relatively low in MCF-7 cells cultivated in E2-depleted condition (data not presented). Therefore, miR-7-mediated actions are considered to be related to EGFR than IGF-1R in MCF-7 cells. However, the correlation between miR-7 and ER or EGFR was not detected in our in vitro analysis of the tumor samples of breast cancer patients, but none of them received endocrine therapy before surgery. Therefore, further examinations including the alterations of the miR-7 expression in those who received neoadjuvant endocrine therapy are required to clarify the roles of miR-7 in ER and EGFR signaling.

Conclusions
Aromatase inhibitors are commonly used as hormone therapy in postmenopausal estrogen-sensitive breast cancer patients. It is true that both plasma and intratumoral estrogen concentration was decreased by the treatment of aromatase inhibitor in postmenopausal ER-positive patients [32]. Acquired resistance to aromatase inhibitor has become one of the major problems in the clinical management of these patients. It is possible that the development of endocrine resistance is attributed to the up-regulation of EGFR caused by estrogen depletion which represents an attempt to rescue cell growth by switching to alternative pathway. The results of our study suggest that miR-7 may play central roles in the development of resistance to endocrine therapy in breast cancer patients through regulating EGFR expression of cancer cells. Abbreviations DCC-FBS: 10% dextran-coated charcoal fetal bovine serum; E2: Estradiol; EGFR: epidermal growth factor receptor; ER: estrogen receptor; ERE: estrogen response element; FBS: fetal bovine serum; FFPE: 10% formalin-fixed and paraffin embedded tissue; hnRNP K: heterogeneous nuclear ribonucleic protein K; ICI: ICI 182, 780; IGF-1R: insulin-like growth factor 1 receptor; LCM: laser capture microdissection; miRNA: microRNA; MRP1: multidrug resistance protein 1; PCR: polymerase chain reaction; RT-PCR: reverse-transcriptase polymerase chain reaction.