Mechanisms of acquired resistance to endocrine therapy in hormone-dependent breast cancer cells

Abstract

Acquired resistance is a major problem limiting the clinical benefit of endocrine therapy. To investigate the mechanisms involved, two in vitro models were developed from MCF-7 cells. Long-term culture of MCF-7 cells in estrogen deprived medium (LTED) mimics aromatase inhibition in patients. Continued exposure of MCF-7 to tamoxifen represents a model of acquired resistance to antiestrogens (TAM-R). Long-term estrogen deprivation results in sustained activation of the ERK MAP kinase and the PI3 kinase/mTOR pathways. Using a novel Ras inhibitor, farnesylthiosalicylic acid (FTS), to achieve dual inhibition of the pathways, we found that the mTOR pathway plays the primary role in mediation of proliferation of LTED cells. In contrast to the LTED model, there is no sustained activation of ERK MAPK but enhanced responsiveness to rapid stimulation induced by E(2) and TAM in TAM-R cells. An increased amount of ERalpha formed complexes with EGFR and c-Src in TAM-R cells, which apparently resulted from extra-nuclear redistribution of ERalpha. Blockade of c-Src activity drove ERalpha back to the nucleus and reduced ERalpha-EGFR interaction. Prolonged blockade of c-Src activity restored sensitivity of TAM-R cells to tamoxifen. Our results suggest that different mechanisms are involved in acquired endocrine resistance and the necessity for individualized treatment of recurrent diseases.

Fig. 1. Effect of FTS on serum-stimulated activation of MAP kinase and mTOR

MCF-7 and LTED cells grown in 60 mm dishes were treated with FTS for 24 h. Cells were then harvested and cell lysate prepared. (A) Activity of ERK1/2 MAP kinase was measured by incubation of immunoprecipitated ERK1/2 with Elk1 for 30 min followed by Western blot with antibody against phosphorylated Elk1. (B) Activity of mTOR was examined by Western blot analysis of phosphorylated p70 S6 kinase at Thr³⁸⁹ and PHAS-I at Ser⁶⁵ using specific antibodies and quantitated by densitometry scanning.

Fig. 2. Effect of FTS on IGF-1 induced phosphorylation of Akt (Ser473) and p70 S6K (Thr³⁸⁹ and Thr²²⁹)

Sub-confluent LTED cells grown in 60 mm dishes were serum starved for 24 h, pretreated with FTS (100 µM) or LY 294002 (LY, 20 µM) for 10, 30 or 60 min before addition of IGF-1 (20 ng/ml for 10 min). Cells were then harvested and cell lysate prepared. Phosphorylated and total kinases were detected by Western analysis using specific antibodies.

Fig. 3. Growth response of wild type MCF-7 cells (A) and LTED cells (B) to rapamycin alone and in combination with U0126

Sixty thousand cells were plated into each well of 6-well plates in their culture media. Two days later, the cells were treated in triplicate wells for five days with rapamycin alone at indicated concentrations or in combination with U0126 for 5 days before counting the cell number. The results are expressed as percent inhibition.

Fig. 4. Altered acute responses of MCF-7 cells to tamoxifen and estradiol after long term tamoxifen exposure

MCF-7 cells cultured for 0–12 months with 10⁻⁷ M tamoxifen (TAM) or ethanol were plated in 6-well plates at a density of 60,000 cells/well in IMEM containing 5% FBS. The cells were then treated with TAM (10⁻⁷ M) (A) or E₂ (10⁻¹⁰ M) (B) for 5 days and cell number was counted. For TAM pre-exposed cells, tamoxifen was continuously present in the medium before the 5-day treatment with fresh medium containing either TAM or E₂. The Reference line represents average response of untreated or ethanol treated MCF-7 cells under the same condition.

Fig. 5. Estradiol and tamoxifen induced rapid activation of ERK MAPK

Control ells (ethanol treated) and tamoxifen resistant cells (TAM-R) were cultured for 96 h in phenol red-free medium containing 1% DCC-FBS and then incubated with E2 (A) or tamoxifen (B) for various time before preparation of cell lysates and Western blot analysis. The numbers shown under the bands of phospho-MAPK are fold changes of phospho-MAPK normalized by total MAPK. Results shown in this figure are representative of three repeatable experiments.

Fig. 6. Kinase active c-Src facilitates ERα-EGFR interaction in TAM-R cells

(A) Cell lysates were prepared from TAM-R cells and matched controls that had been exposed to TAM or ethanol for more than four months. Interaction between EGF receptor and ERα was examined by immunoprecipitation with antibody against EGFR followed by immunoblotting for ERα. (B) Co-immunoprecipitaion of c-Src by antibodies against EGFR and ERα. (C) TAM-R cells were treated with or without PP2 (5 µM) for 48 h before preparation of cell lysate. ERα-EGFR interaction was examined as mentioned above.

Fig. 7. Altered subcellular distribution of ERα in TAM-R cells

(A) Cells cultured on coverslips for 2 days were fixed and incubated with anti-ERα antibody overnight at 4°C. After thorough wash with PBS, the cells were incubated with Alexa Fluor 488 labeled secondary antibody (green) for 1 h. Actin was stained with Alexa Fluor 546 labeled phalloidin (red). Confocal images were captured under Zeiss LSM 510 confocal microscope (40× objective). (B) Protein levels of ERα in subcellular fractions. Nuclei, cytoplasm, and plasma membrane were isolated as described in Material and Methods. An equal amount of proteins from different fractions was subjected to SDS-PAGE, transferred to nitrocellulose membrane and Western blot analysis for ERα. 5’-Nucleotidase (5′NT), GW182, and nuclear transport factor 2 (NTF2) were used as markers for plasma membrane, cytosol, and nucleus, respectively. Numbers under the ERα bands represent fold changes of ERα normalized by the marker of each fraction. PP2 treatment: 5 µM, 48 hours.

Fig. 8. Growth responses of wild type MCF-7 cells, TAM-R cells, and PP2 treated TAM-R cells to acute treatment of tamoxifen

PP2 (5 µM) treatment lasted for eight months. Cell numbers were counted as described in the Materials and Methods. The result was expressed as percent reduction compared with the vehicle control of each cell type.