Novel insight into the protective role of androgen receptor in ER-positive breast cancer

Abstract

Introduction: Androgen receptor (AR) role in breast cancer appears to be clinically relevant and disease context specific. In estrogen receptor (ER) α-positive primary breast cancers, AR positivity correlates with low tumor grade and a better clinical outcome. These clinical-pathological findings mirror androgen capability to counteract ERα-dependent proliferation in both normal and tumor mammary epithelium. Tumor microenvironment is a key factor in cancer development and progression since the physical and hormonal paracrine exchanges with the epithelial compartment promote tumor cell proliferation and metastasis. This research project reports additional molecular mechanisms in the dynamic interplay between AR and ER-α signaling pathway in breast cancer cells and showed AR expression and role in the stroma of ER-positive breast cancers. Material and methods: MCF-7, T47D, SKBR3 breast cancer cells and cancer-associated fibroblats (CAFs) from biopsies of primary breast tumors (n=3); cell proliferation assay, transient transfection, quantitative Real Time PCR (qRT-PCR), western blotting (WB), immunoprecipitation assay (IP), chromatin immunoprecipitation assay (ChIP), RNA silencing, tunel assay, DNA affinity precipitation assay (DAPA), immunofluorescence analysis (IF), immunocytochemical staining (IHC), wound-healing scratch assay, F-actin staining assay, Boyden-chamber transmigration assay, matrigel-based invasion assay, cytokine array, zymography assay, mammosphere forming efficiency assay. Data were analyzed by ANOVA. Results: This study demonstrated an androgen-dependent mechanism through which ligand-activated AR decreased estradiol-induced cyclin D1 protein, mRNA and gene promoter activity in MCF-7 cells. This mechanism involved the competition of AR and ERα for the steroid receptor coactivator AIB1, a limiting factor in the functional coupling of ERα with the cyclin D1 promoter. Indeed, AIB1 overexpression was able to reverse the down-regulatory effects exerted by AR. Co-immunoprecipitation studies showed that AIB1 preferentially interacted with ERα or AR in relation to their intracellular levels. In addition, ChIP analysis evidenced that androgen administration decreased E2-induced recruitment of AIB1 at the AP-1 site on the cyclin D1 promoter gene. Moreover, this research project showed an increased expression of the pro-apoptotic protein BAD following androgen treatment while the levels of anti-apoptotic protein Bcl-2 as well as of the pro-apoptotic proteins BID and BAX remained unchanged. As consequence, the Bcl-2/BAD ratio was reduced, shifting the delicate balance between inhibitors and inducers of cell death. Androgen stimulation increased also BAD levels into the nuclear compartment in ERα/AR-positive MCF-7 as well as in ERα negative/AR-positive SKBR3 cells. The androgen-regulated intracellular localization of BAD involved an AR/BAD physical interaction, suggesting a nuclear role for BAD upon androgen stimulation. Indeed, androgens induced both AR and BAD recruitment at the AP-1 and the ARE sites within the cyclin D1 promoter region, contributing to explain the anti-proliferative effect of androgens in breast cancer cells. Finally, the study demonstrated AR expression and functionally activation upon androgen treatment in primary human breast CAFs. Androgen-activated AR affected CAFs secretory phenotype as evidenced by cytokine array performed on CAFs conditioned medium (CM). Co-culture experiments showed that CM from androgen-treated CAFs was less effective in stimulating MCF-7 and T47D cells motility and invasion compared to CM from untreated ones, indicating that androgens, via AR, may influence CAFs secretion of paracrine soluble factors involved in tumor cell motility and invasiveness sustainment. Conclusions: Taken together all these data showed that ligand activated AR exerts a protective role in E2-dependent breast cancer development and progression by inhibiting CD1 expression through the squelching of the steroid receptor co-activator AIB1 and the overexpression of the pro-apoptotic protein BAD. Furthermore, the study highlighted the protective role of AR in the tumor microenvironment, since activated-stromal AR affects the paracrine factors secreted by CAFs reducing ER-positive breast cancer cell migration and invasiveness. Thus, these findings reinforce the possibility to couple the androgen-based therapy with therapies targeting other important pathways in ERα-positive breast cancer patient treatment.