Expression of the K303R estrogen receptor-alpha breast cancer mutation induces resistance to an aromatase inhibitor via addiction to the PI3K/Akt kinase pathway

Abstract

Aromatase inhibitors (AI) are rapidly becoming the first choice for hormonal treatment of estrogen receptor-alpha (ERalpha)-positive breast cancer in postmenopausal women. However, de novo and acquired resistance frequently occurs. We have previously identified a lysine to arginine transition at residue 303 (K303R) in ERalpha in premalignant breast lesions and invasive breast cancers, which confers estrogen hypersensitivity and resistance to tamoxifen treatment. Thus, we questioned whether resistance to AIs could arise in breast cancer cells expressing the ERalpha mutation. As preclinical models to directly test this possibility, we generated K303R-overexpressing MCF-7 cells stably transfected with an aromatase expression vector. Cells were stimulated with the aromatase substrate, androstenedione, with or without the AI anastrozole (Ana). We found that Ana decreased androstenedione-stimulated growth of wild-type cells, whereas K303R-expressing cells were resistant to the inhibitory effect of Ana on growth. We propose that a mechanism of resistance involves an increased binding between the mutant receptor and the p85alpha regulatory subunit of phosphatidylinositol-3-OH kinase (PI3K), leading to increased PI3K activity and activation of protein kinase B/Akt survival pathways. Inhibition of the selective "addiction" to the PI3K/Akt pathway reversed AI resistance associated with expression of the mutant receptor. Our findings suggest that the K303R ERalpha mutation might be a new predictive marker of response to AIs in mutation-positive breast tumors, and that targeting the PI3K/Akt pathway may be a useful strategy for treating patients with tumors resistant to hormone therapy.

Figure 1

Characterization of K303R ERα-expressing cells. A. Immunoblotting showing ERα and Rho GDIα (loading control) in MCF-7 parental (MCF-7) and YFP-K303R ERα-cells (MCF-7 K303R). B. Anchorage-independent growth assay in cells treated with vehicle or 17β-Estradiol (E₂ 10⁻¹², 10⁻¹¹, 10⁻¹⁰, 10⁻⁹M). Bars, SD. n.s.=nonsignificant, **P<0.005, ***P<0.0005 versus vehicle. C. WT ERα and K303R ERα–cells were injected into mice (n=5/group) supplemented with E₂. Tumor growth was plotted as the time for tumor tripling (months[mos.]). D. Immunoblotting showing aromatase (Arom), ERα and Rho GDIα. Numbers below blots represent aromatase activity. E. Anchorage-independent growth assay in cells treated with vehicle or 4-Androstene-3,17-dione (AD1, 10 and 25nM). Bars, SD. *P<0.05, **P<0.01, ***P<0.0005: K303R Arom 1-group versus the respective treatment in MCF-7 Arom 1-group. F. MCF-7 Arom 1 and K303R Arom 1-cells were injected into mice, supplemented with AD. Tumor growth was plotted as the percentage of change in mean tumor volume/animal (n=5) compared to day 0 of treatment ±S.D. *P<0.05, **P<0.01 K303R Arom 1-group versus MCF-7 Arom 1-group.

Figure 2

K303R ERα mutation confers resistance to anastrozole. A. MTT growth assays in cells treated with vehicle, E₂1nM, AD10nM and/or anastrozole (Ana100nM, 1µM, 10µM). Cell proliferation is expressed as fold change relative to vehicle-treated cells. Data are representative of three independent experiments, performed in quadruplicate. Columns, mean. Bars, SD. **P<0.005, n.s.=nonsignificant AD versus Ana+AD. B. Anchorage-independent growth assay in cells treated with vehicle, E₂1nM, AD10nM±Ana1µM. Bars, SD. *P=0.01, n.s.=nonsignificant AD versus Ana+AD. C-D. Anchorage-independent growth assay in MCF-7 Arom P (C) and CHO P (D) pools treated with vehicle, AD10nM±Ana1µM. Bars, SD. **P<0.01, n.s.=nonsignificant AD versus Ana+AD. Upper panels for C-D showed ERα and Rho GDIα. Numbers below blots represent aromatase activity.

Figure 3

Mutant cells exhibited constitutive pAkt. A-F. Immunoblotting showing phospho-Akt (pAkt), Akt and Rho GDIα in cells treated with vehicle or AD 1, 10, 25 and 50nM (1h, A) or AD 10, 30, 60, 120min (10nM, B) or heregulin2ng/ml (H 5min, B); in xenograft tumors extracts (C, numbers below blots represent the average fold change in pAkt levels of K303R Arom 1-samples versus MCF-7 Arom 1-samples); in cells treated with vehicle, AD10nM±Ana1µM (1h, D); in cells treated with vehicle and ICI1µM for 1, 5, 10 and 30min (E) or 1, 3, 24 and 48h (F). Quantitative analysis is the fold difference in pAkt/Akt/Rho GDIα ratio relative to vehicle-treated MCF-7 Arom 1-cells.

Figure 4

Mutant cells showed increased PI3K/Akt activation. A. Lysates from CHO cells transiently transfected with WT or K303R ERα were immunoprecipitated (IP) with anti-p85 or anti-ERα and immunoblotted for ERα or p85. Input: whole-cell lysates. B, C. In vitro PI3-kinase activity of p85 immunoprecipitates from CHO cells transiently transfected with YFP-WT or YFP-K303R ERα, and CHO WT P and CHO K303R P pools (B) or MCF-7 Arom 1 and K303R Arom 1-cells treated with vehicle or LY294002 (LY10µM) for 45min (C). The enzyme activity was expressed as amounts of PIP₃ (picomoles) produced by cells. Columns, mean. Bars, SD. Values are means from triplicate readings ±S.D. in two experiments. *P<0.05, **P=0.001. Upper panels for H-I showed p85 expression in immunoprecipitates. L. Immunoblotting of pAkt, Akt, phospho-Ser167-ERα (pS167), YFP-ERα and Rho GDIα in CHO cells transiently transfected with YFP-WT or YFP-K303R ERα, or CHO WT P and CHO K303R P pools. E. CHO cells were transiently transfected with YFP-WT or YFP-K303R ERα. plus an ERE-luciferase reporter, and treated with vehicle, LY10µM or ICI1µM. Luciferase activities were normalized to β-galactosidase for relative luciferase units. Data are representative of three independent experiments, performed in triplicate. Columns, mean. Bars, SD. ***P=0.0008 versus vehicle. ERα and Rho GDIα expression was determinated by immunoblot (right panel).

Figure 5

Altered apoptotic response in mutant cells. A-C. Immunoblotting showing PARP and Rho GDIα in cells treated with vehicle, E₂1nM (24h), AD10nM (12, 24 and 48h, A) or AD10nM and AD+Ana1µM (24h, B); Bcl-2, Bax and Rho GDIα in cells treated with vehicle, AD10nM±Ana1µM (24h, C). D. Quantitative analysis is the fold difference in Bcl-2/Bax ratio relative to vehicle-treated MCF-7 Arom 1-cells. E. Elisa cell detection assay in cells treated with AD10nM±Ana1µM. Columns, mean. Bars, SD. *P<0.05, **P<0.005, n.s.=nonsignificant AD versus Ana+AD.

Figure 6

Inhibition of PI3K/Akt pathway reversed AI^R. A. Immunoblotting showing phospho-Akt and Akt in cells treated with vehicle, LY10µM or PI-103 (1, 5, 10µM) or Akti1/2 (1, 5, 10µM) for 30min. B-D. Anchorage-independent growth assay in cells treated with vehicle, AD10nM, Ana1µM ±LY10µM (B) or Akti1/21µM (C) or PD98059 (PD10µM, D) or ICI1µM (D). Bars, SD. ***P<0.0005 LY, Akti1/2 and ICI-treated cells versus control cells and cells treated with AD or Ana+AD. E. Cell death detection assay in cells treated with AD10nM ±Ana1µM±LY. Columns, mean. Bars, SD. *P<0.05 versus AD in MCF-7 Arom 1 and AD and Ana+AD in K303R Arom 1.