The role of amphiregulin in exemestane-resistant breast cancer cells: evidence of an autocrine loop

Abstract

Exemestane-resistant breast cancer cell lines (i.e., ExeR), derived from MCF-7 cells expressing a high level of aromatase (MCF-7aro), were generated in our laboratory. The epidermal growth factor (EGF)-like protein amphiregulin (AREG) was highly expressed in ExeR cells based on cDNA microarray analysis. The high levels of AREG mRNA in ExeR cell lines were confirmed by real-time reverse transcription-PCR. The high levels of AREG protein in ExeR cell lysates and culture media were confirmed by Western blot analysis and ELISA, respectively. Furthermore, our Western blot analysis showed that whereas no AREG was detected in the DMSO control, overnight treatment of parental MCF-7aro cells with 1 micromol/L exemestane strongly induced the expression of AREG. This induction was totally blocked by 100 nmol/L of pure antiestrogen ICI 182,780, implying estrogen receptor (ER) dependence of exemestane-induced AREG expression. MCF-7aro cells were not able to proliferate in hormone-free medium, but were able to proliferate in conditioned medium from ExeR cells, similar to the treatment of recombinant human AREG. Small interference RNA targeting AREG inhibited ExeR proliferation, confirming that AREG is truly functioning as a growth factor of ExeR cells. The specific inhibitors to ER (ICI 182,780), EGF receptor (EGFR; AG1478), and mitogen-activated protein kinase (MAPK; U0126) all showed dose-dependent suppression of the proliferation of ExeR cells, indicating the involvement of the ER, EGFR, and MAPK pathways. Based on these findings, we propose a possible mechanism that underlies exemestane resistance: exemestane induces AREG in an ER-dependent manner. AREG then activates the EGFR pathway and leads to the activation of the MAPK pathway that drives cell proliferation.

Figure 1

High levels of AREG expression in ExeR cells. AREG expression levels in ExeR cells were compared with those in LTEDaro cells by real-time RT-PCR (A; results normalized to β-actin), Western blot (B), and ELISA of culture media (C). Columns, mean; bars, SD. **, P < 0.01, compared with LTEDaro.

Figure 2

ER dependence of exemestane-induced AREG expression. A, MCF-7aro cells were cultured in hormone-free, phenol red—free medium supplemented with 1 nmol/L testosterone. Then cells were treated with 100 nmol/L ICI or DMSO for 7 h. B, MCF-7aro cells were cultured in hormone-free, phenol red—free medium for 24 h, followed by drug treatment overnight [testosterone (T) at 1 nmol/L, exemestane (Exem)at 1 μmol/L, and ICI at 100 nmol/L]. Cells were then lysed and Western blot analysis was done. C, aromatase activity was measured by the tritiated water release method and expressed as counts per minute (cpm). Exemestane at 1 μmol/L and ICI at 100 nmol/L were used. DMSO was used in control (Ctl). Placental microsome was used as source of aromatase.

Figure 3

MCF-7aro proliferation assay. Proliferation of MCF-7aro cells was examined with the addition of conditioned media (CM) from ExeR cells (A) and recombinant human AREG (rhAREG; B). Columns, mean; bars, SD. *, P < 0.05; **, P < 0.01, compared with 0 μL conditioned medium or 0 ng/mL recombinant human AREG.

Figure 4

Block of ExeR cell proliferation with AREG siRNA. All experiments were done on both LTEDaro cells and ExeR cells. The effective repression of AREG expression was examined by real-time RT-PCR (A) and ELISA (B) of AREG in culture media. C, effect of siRNA on cell proliferation. Columns, mean; bars, SD. *, P < 0.05; **, P < 0.01, compared with ExeR control.

Figure 5

Suppression of the proliferation of ExeR cells by inhibitors of ER, EGFR, or MAPK. Proliferation assay was done on ExeR cells treated with pure antiestrogen ICI (A), specific EGFR inhibitor AG1478 (B), or MAPK kinase inhibitor U0126 (C). Columns, mean; bars, SD. **, P < 0.01, compared with control (0 nmol/L or 0 μmol/L).

Figure 6

Proposed mechanism of AREG-mediated exemestane resistance.