Arsenic induces functional re-expression of estrogen receptor α by demethylation of DNA in estrogen receptor-negative human breast cancer

Abstract

Estrogen receptor α (ERα) is a marker predictive for response of breast cancers to endocrine therapy. About 30% of breast cancers, however, are hormone- independent because of lack of ERα expression. New strategies are needed for re-expression of ERα and sensitization of ER-negative breast cancer cells to selective ER modulators. The present report shows that arsenic trioxide induces reactivated ERα, providing a target for therapy with ER antagonists. Exposure of ER-negative breast cancer cells to arsenic trioxide leads to re-expression of ERα mRNA and functional ERα protein in in vitro and in vivo. Luciferase reporter gene assays and 3-(4,5-dimethylthiazol-2-yl)- 5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) assays show that, upon exposure to arsenic trioxide, formerly unresponsive, ER-negative MDA-MB-231 breast cancer cells become responsive to ER antagonists, 4-hydroxytamoxifen and ICI 182,780. Furthermore, methylation- specific PCR and bisulfite-sequencing PCR assays show that arsenic trioxide induces partial demethylation of the ERα promoter. A methyl donor, S-adenosylmethionine (SAM), reduces the degree of arsenic trioxide-induced re-expression of ERα and demethylation. Moreover, Western blot and ChIP assays show that arsenic trioxide represses expression of DNMT1 and DNMT3a along with partial dissociation of DNMT1 from the ERα promoter. Thus, arsenic trioxide exhibits a previously undefined function which induces re-expression ERα in ER-negative breast cancer cells through demethylation of the ERα promoter. These findings could provide important information regarding the application of therapeutic agents targeting epigenetic changes in breast cancers and potential implication of arsenic trioxide as a new drug for the treatment of ER-negative human breast cancer.

Figure 1. ERα mRNA and protein re-expression induced by arsenic trioxide in MDA-MB-231 cells (A) and Hs578T cells (B).

ERα protein re-expression induced by AZA (2.5 μmol/L for 4 days) in MDA-MB-231 cells (C). RT-PCR and Western blotting analysis showed the re-expression of ER mRNA and protein after exposure of cells to arsenic trioxide (1, 2, 4 μmol/L for 6 days), whereas SAM (200 μmol/L for 6 days) reduced the degree of re-expression. The ER-positive prototype, MCF-7, was used as a positive control; untreated ER-negative MDA-MB-231 cells were used as a negative control. GAPDH and β-Actin provided a control for the amount of intact RNA and protein used in the reactions. ## P<0.01 compared with control cells. # P<0.05 compared with control cells. *P<0.05 compared with cells exposed to 1 μM arsenic trioxide.

Figure 2. Re-expression of functional ERα in ER-negative MDA-MB-231 and Hs578T cells by arsenic trioxide.

(A) Effect of arsenic trioxide (2 μmol/L for 6 days), E2 (10 nmol/L for 24 hr), OHT (10 μmol/L for 24 hr), and ICI (2 μmol/L for 24 hr) on cell growth was determined by MTS. MCF-7 and MDA-MB-231 cells were exposed to E2, OHT or ICI for 24 hr before assay. In addition, MDA-MB-231 cells were exposed to arsenic trioxide (2 μmol/L for 6 days), as described, along with E2, OHT or ICI for 24 hr before MTS assay. This experiment was repeated twice with similar results. (B) The transcriptional activities of re-expressed ERα was examined in MDA-MB-231 cells by luciferase reporter gene assay. Cells were transiently co-transfected with 0.5 μg of pERE-TATALuc+, 0.2 μg of rERa/pCI, and 0.1 μg of phRL-tk as an positive control; untreated MDA-MB-231 cells were transiently co-transfected with 0.5 μg of pERE-TATALuc+ and 0.1 μg of phRL-tk as a negative control. MDA-MB-231 cells were exposed to arsenic trioxide alone (2 μmol/L for 6 days), or in combination with E2 (10 nmol/L) or ICI (2 μmol/L) for 24 hr. The cell lysates were analyzed by use of the Dual-Luciferase Reporter Assay System kit. Ethanol solvent was used as control, and transcriptional activity was presented as fold of control. Values are presented as the means ± SD of three independent experiments. (C) Expression of ERα and its target genes (pS2 and GREB1) in MDA-MB-231 cells. Before RT-PCR was performed, cells untreated or pretreated with arsenic trioxide were exposed to SAM (200 μmol/L for 6 days), OHT (10 μmol/L for 24 hr), ICI (2 μmol/L for 24 hr), or vehicle in fresh estrogen-free medium. This experiment was repeated twice with similar results. (D) Expression of ERα target genes pS2 and GREB1) in Hs578T cells. Before RT-PCR was performed, cells untreated or pretreated with arsenic trioxide were exposed to SAM (200 μmol/L for 6 days), AZA (2.5 μmol/L for 4 days), or vehicle in fresh estrogen-free medium. This experiment was repeated twice with similar results.

Figure 3. Arsenic trioxide induces functional re-expression of ERα in vivo.

(A) Expression of ERα mRNA and its target genes (pS2 and GREB1) in MDA-MB-231 tumors. Arsenic trioxide (2 mg/kg BW) was administered i.p. in 100 μl of sterile saline for 4 weeks. The ER-positive prototype, MCF-7, was used as a positive control. C1–C5: control group, A1–A5: arsenic trioxide-treated group. (B) Expression of ERα protein in MDA-MB-231 tumors. The relative expression level of ERα protein is represented in a scatter plot. (C) Representative immunohistochemistry for ERα in MDA-MB-231 tumors using an anti- ERα polyclonal antibody.

Figure 4. Methylation analysis of the ERα promoter by MSP in MDA-MB-231 cells.

(A) A 2.5 kb genomic sequence of the ERα promoter, analyzed by the online program, www.urogene.org/methprimer, revealed the presence of a high GC percentage CpG islands (blue) between relative positions 2000 and 2500. Position 2000 indicates the transcription start site (TSS, arrow). A region of high CpG (red vertical bars) density was chosen for MSP analysis. This region includes 28 CpG dinucleotides. (B), Methylation analysis of the ERα promoter by MSP in breast cancer cell lines. Sensitivity of the utilized MSP primers was determined by a dilution series of methylated DNA with unmethylated DNA. At least 1% of methylated DNA (∼0.1 ng) can be detected with the ER MSP primers. In each set, M primer pairs anneal only to sequences that are methylated before bisulfite treatment, whereas the U primer pairs anneal only to sequences that are unmethylated.

Figure 5. Methylation analysis of the ERα promoter by BSP in breast cancer cells.

(A)Shown is a representative bisulfite sequence analysis for the high CpG region of the ERα promoter, which contains 28 CpG dinucleotides. For MDA-MB-231 and MCF-7 cell lines (10 replicates each), methylated CpGs are designated by closed circles; unmethylated CpGs are designated by open circles. (B)Statistical chart of methylation analysis; the results shown represent the means ± S.E. of three independent experiments. Significant differences from the controls as determined by Student's t test are indicated by asterisks (P <0.05).

Figure 6. Methylation-related proteins mediate arsenic trioxide-induced re-expression of ERα.

(A) Immunoblot analysis of DNMTs and MBD2 protein. Equal amounts of protein (80 μg) from whole-cell lysates of the control MCF-7 and MDA-MB-231 cells, treated as in Fig. 1, were separated by SDS-PAGE and subjected to Western blot analysis with specific antibodies against DNMT1, DNMT3a, DNMT3b and MBD2. Equivalence of protein loading was demonstrated by immunoblotting with anti-GAPDH antibody. (B) Relative protein levels (mean± SD) of DNMTs (n = 3). (C) Relative protein levels (mean± SD) of MBD2 (n = 3). (D) ChIP analysis of association of DNMT1 and ERα promoter. Cross-linked chromatin prepared from ER-positive MCF-7 and ER-negative MDA-MB-231 human breast cancer cells was immunoprecipitated with antibodies DNMT1. MDA-MB-231 cells treated as in Fig. 1. The immunoprecipitates were subjected to PCR analysis. Aliquots of chromatin taken before immunoprecipitation were used as input controls (n = 3). # P<0.05 compared with control cells. *P<0.05 compared with cells exposed to 1μM arsenic trioxide.