microRNA regulation of mammalian target of rapamycin expression and activity controls estrogen receptor function and RAD001 sensitivity

Abstract

Background: The AKT/mammalian target of rapamycin (mTOR) signaling pathway is regulated by 17α-estradiol (E2) signaling and mediates E2-induced proliferation and progesterone receptor (PgR) expression in breast cancer.

Methods and results: Here we use deep sequencing analysis of previously published data from The Cancer Genome Atlas to demonstrate that expression of a key component of mTOR signaling, rapamycin-insensitive companion of mTOR (Rictor), positively correlated with an estrogen receptor-α positive (ERα+) breast tumor signature. Through increased microRNA-155 (miR-155) expression in the ERα+ breast cancer cells we demonstrate repression of Rictor enhanced activation of mTOR complex 1 (mTORC1) signaling with both qPCR and western blot. miR-155-mediated mTOR signaling resulted in deregulated ERα signaling both in cultured cells in vitro and in xenografts in vivo in addition to repressed PgR expression and activity. Furthermore we observed that miR-155 enhanced mTORC1 signaling (observed through western blot for increased phosphorylation on mTOR S2448) and induced inhibition of mTORC2 signaling (evident through repressed Rictor and tuberous sclerosis 1 (TSC1) gene expression). mTORC1 induced deregulation of E2 signaling was confirmed using qPCR and the mTORC1-specific inhibitor RAD001. Co-treatment of MCF7 breast cancer cells stably overexpressing miR-155 with RAD001 and E2 restored E2-induced PgR gene expression. RAD001 treatment of SCID/CB17 mice inhibited E2-induced tumorigenesis of the MCF7 miR-155 overexpressing cell line. Finally we demonstrated a strong positive correlation between Rictor and PgR expression and a negative correlation with Raptor expression in Luminal B breast cancer samples, a breast cancer histological subtype known for having an altered ERα-signaling pathway.

Conclusions: miRNA mediated alterations in mTOR and ERα signaling establishes a new mechanism for altered estrogen responses independent of growth factor stimulation.

Figure 1

mTORC1 activation is Associated with an ER Negative Breast Tumor Phenotype. Deep sequencing data obtained from TCGA data portal was analyzed with respect to ERα (either ERα positive or negative) status for expression of (A) mTOR associated gene sets (mTOR, TSC1, TSC2, Rheb, Rictor, Raptor, Rheb) and (B) miR-155 host gene (HG) expression. Heat mat depicts high expression (red, +1) and low expression (green, −1). Platform analyzed was TCGA breast invasive carcinoma gene expression IlluminaHiSeq n = 1032. Analysis was sorted based on ERα gene Signature where positive for ER = orange and negative for ER = blue. (C) MCF-7-vector and –miR-155 cells were harvested for qPCR analysis for mTOR associated genes Rictor, TSC1, Deptor, Rheb, Raptor, and p70s6 kinase. Ct values were normalized to β-actin and MCF-7-vector cells designated as 1. * p < 0.05 for n ≥ 3. (D) MCF-7-vector and –miR-155 cells were harvested for western blot analysis for total and phospho-mTOR (S2448) and mTOR associated proteins Rictor, Raptor, TSC1, and p-70 s6 kinase. Expression normalized to RHO-GDIα, n ≥ 3. (E) MCF-7-vector and miR-155 cells were harvested for western blot analysis for phosphorylation of downstream mTORC1 associated proteins p-eIF4B (S422), p-S6 ribosomal protein (S235/236), p-p70s6 kinase (Thr389), p-eEF2K (S366). Expression was normalized to RHO-GDIα, n = 2. (F) qPCR of MCF-7-vector and miR-155 cells for downstream mTORC2 regulated gene PKCα, normalization was to β-actin and vector designated as 1, n = 3.

Figure 2

Overexpression of miR-155 selectively alters estrogen stimulated gene expression of ER-regulated genes in vitro . (A-B) Gene panel array for ER regulated and cancer associated genes, n = 2. (A) Down regulated genes and (B) up regulated genes (logarithmic scale) in MCF-7-miR-155 cells compared to vector control. Results represent average fold change ± SEM. (C-F) QPCR was performed for genes, n = 5 (C) PgR, (D) SDF-1, (E) BCL2, and (F) SERPINA3. Cells were grown in 5% CS phenol free DMEM for 48 hours before treatment with DMSO or E2 (1nM) for 24 hours. Cycle number was normalized to beta actin and MCF-7-vector control cells treated with DMSO scaled to 1. Bars represent average fold change ± SEM. * p < 0.05; ** p < 0.01; *** p < 0.001.

Figure 3

miR-155 enhanced E2 stimulated proliferation is mediated through altered mTOR signaling in vivo and in vitro . (A) Crystal violet assay for proliferation of MCF-7-vector and MCF-7-miR-155 cells treated with E2. Cells were grown in 5% CS phenol free DMEM for 48 hours prior to treatment with E2 (1nM) or vehicle control (DMSO). Each cell line was normalized to the respective vehicle control, n = 4. (B) Tumor volume for ovariectomized CB-17 SCID female mice injected bilaterally with 5X10^{^6} MCF-7-vector cells or MCF-7-miR-155 cells, n = 5 animals/group. All animals were implanted with an E2 pellet (0.72 mg) at day of cell injection. Points represent average tumor volume ± SEM. (C) MCF-7-vector and –miR-155 cells were harvested for total RNA extraction and PCR was performed for PgR following treatment with E2 (1nM), RAD001 (20nM) + E2, or vehicle control (DMSO), normalized to MCF-7-vector treated with E2 (1nM). Bars represent fold change ± SEM, n ≥ 3. * Significantly different from MCF-7-vector + E2 p < 0.05. (D) Tumor volume for ovariectomized CB-17 SCID female mice injected bilaterally with 5X10^{^6} MCF-7-miR-155 cells + E2 + RAD001 vs. MCF-7-miR-155 + E2 given placebo, n = 7. Points represent normalized tumor volume ± SEM. * p < 0.05, ** p < 0.01.

Figure 4

Rictor expression positively correlates with PgR expression in Luminal B breast cancer molecular subtype. (A-C) Targeted gene expression correlation analysis for ER, PgR, Rictor, Rheb, and Raptor derived from pooled breast cancer samples obtained from BC-GenExMiner-v3.0. Correlation maps generated based on breast cancer molecular subtype. (A) luminal A, (B) Basal-like, (C) luminal B.

Figure 5

mTORC1 induced repression of PgR is regulated by miR-155 independently of growth factor stimulation. Schematic for miR-155 induced regulation of mTOR/ER crosstalk.