Glucocorticoid receptor signalling activates YAP in breast cancer

Abstract

The Hippo pathway is an oncosuppressor signalling cascade that plays a major role in the control of cell growth, tissue homoeostasis and organ size. Dysregulation of the Hippo pathway leads to aberrant activation of the transcription co-activator YAP (Yes-associated protein) that contributes to tumorigenesis in several tissues. Here we identify glucocorticoids (GCs) as hormonal activators of YAP. Stimulation of glucocorticoid receptor (GR) leads to increase of YAP protein levels, nuclear accumulation and transcriptional activity in vitro and in vivo. Mechanistically, we find that GCs increase expression and deposition of fibronectin leading to the focal adhesion-Src pathway stimulation, cytoskeleton-dependent YAP activation and expansion of chemoresistant cancer stem cells. GR activation correlates with YAP activity in human breast cancer and predicts bad prognosis in the basal-like subtype. Our results unveil a novel mechanism of YAP activation in cancer and open the possibility to target GR to prevent cancer stem cells self-renewal and chemoresistance.

Figure 1. Glucocorticoids induce activation of the Hippo transducer YAP in vitro and in vivo.

(a) Results of the high-content screening. YAP fluorescence intensity is relative to DMSO-treated samples. (b) Representative images from the screening. MDA-MB-231 stained for Hoechst (blue) and YAP (red) after treatment with DMSO or the indicated glucocorticoids are shown. Experiment repeated two times. Scale bars, 100 μm. (c) MDA-MB-231, MCF10A and MII cells were treated with BM 1 μM alone or in combination with RU486 1 μM for 24 h. Representative blots are shown. Experiment repeated three times. (d) Quantitative PCR with reverse transcription (qRT–PCR) analysis of MDA-MB-231 transfected with indicated siRNA for 48 h and treated with 1 μM BM alone or in combination with RU486 1 μM for 24 h. siCTL is control siRNA. Error bars represent mean±s.d., from n=3 biological replicates. siYAP sequence is siYAP#1. (e) MDA-MB-231 cells were treated as in d, representative blots are shown. Experiment repeated three times. siYAP sequence is siYAP#1. (f) qRT–PCR analysis of breast epithelial tissue from control (NT) or dexamethasone (DM)-treated mice. Error bars represent mean±s.d., n=4 mice per group. *P<0.05, **P<0.01; two-tailed Student's t-test is used throughout.

Figure 2. Glucocorticoid receptor controls YAP activity in breast cancer cells.

(a) Representative images of immunofluorescence in MDA-MB-231 cells. Cells were serum starved or grown in 10% FBS for 24 h. Experiment repeated three times. Scale bars, 15 μm. (b) MDA-MB-231, BT-549 and MII cells were transfected with control (siCTL) or glucocorticoid receptor (siGR) siRNA for 48 h. Representative blots are shown. Experiment repeated three times. (c) MDA-MB-231 and BT-549 cells were transfected with control (siCTL) or glucocorticoid receptor (siGR) siRNA. The day after, cells were transfected with 8XGTII-luc reporter. After 24 h cells were collected. Data are normalized to siCTL. Error bars represent mean±s.d., from n=3 biological replicates. (d) Quantitative PCR with reverse transcription analysis of MDA-MB-231 transfected with control (siCTL) or glucocorticoid receptor (siGR) siRNA for 48 h. Error bars represent mean±s.d., from n=3 biological replicates. *P<0.05, **P<0.01; two-tailed Student's t-test is used throughout.

Figure 3. Glucocorticoids induce YAP nuclear localization.

(a) Results of the high-content screening. (b) Quantification of MDA-MB-231 cells with nuclear YAP by immunofluorescence. Cells were grown in serum-free medium (SFM) and treated with betamethasone (BM) 0.1 μM for 24 h. Error bars represent mean±s.d., from n=3 biological replicates. (c) Cells were treated as in b. Representative images are shown. Scale bars, 15 μm. (d) MDA-MB-231 and MII cells were grown in SFM and treated with BM 1 μM alone or in combination with RU486 1 μM for 24 h. Representative blots are shown. pYAP is phospho-Ser127. Experiment repeated three times. (e) MDA-MB-231 cells were grown at high confluence and treated with BM 1 μM for 24 h. Representative images are shown. Experiment repeated three times. Scale bars, 15 μm. (f) Quantification of cells with nuclear YAP by immunofluorescence. MDA-MB-231 cells were transfected with indicated siRNAs for 24 h and treated with RU486 1 μM for additional 24 h in SFM containing BM 1 μM. Error bars represent mean±s.d., from n=3 biological replicates. (g) MDA-MB-231 cells were serum starved and treated with BM 1 μM for indicated times. Representative blots are shown. *P<0.05, **P<0.01; two-tailed Student's t-test is used throughout.

Figure 4. Glucocorticoid receptor activates YAP by inducing FN1 expression.

(a,b) Quantification of cells with nuclear YAP (a) and representative immunofluorescence (b) of serum-starved MDA-MB-231 treated with 1 μM betamethasone (BM) for the indicated times. (c) Quantitative PCR with reverse transcription (qRT–PCR) analysis of serum-starved MDA-MB-231 treated with 1 μM BM for the indicated times. Left y axis is relative to GILZ, right y axis is relative to ANKRD1. (d) qRT–PCR analysis of serum-starved MDA-MB-231 treated with 1 μM BM alone or in combination with RU486 1 μM for 24 h. Error bars represent mean±s.d., from n=3 biological replicates. (e) MDA-MB-231 cells were treated with BM 1 μM alone or in combination with RU486 1 μM for 24 h. Representative blots are shown. Experiment repeated three times. (f) MDA-MB-231 cells were serum starved and treated with BM 1 μM for 24 h. Extracellular fibronectin was stained by immunofluorescence. Experiment repeated three times. Scale bars, 15 μm. (g) Bright-field microscopy image of MDA-MB-231 cells treated with BM 1 μM alone or in combination with RU486 1 μM for 24 h. Experiment repeated three times. Scale bars, 15 μm. (h) MDA-MB-231 were serum starved and treated with BM 1 μM alone or in combination with RU486 1 μM for 6 h. Representative blots are shown. pYAP is phospho-Ser127. Experiment repeated three times. (i,j) Quantification (i) and representative images (j) of MDA-MB-231 cells with nuclear YAP by immunofluorescence. Cells were grown in serum-free medium and treated with BM 1 μM alone or in combination with RGD 500 μg ml⁻¹ for 6 h. Scale bars, 15 μm. Error bars represent mean±s.d., from n=3 biological replicates. *P<0.05, **P<0.01; two-tailed Student's t-test is used throughout.

Figure 5. Glucocorticoids activate YAP via FAK/Src-dependent actin cytoskeleton remodelling.

(a) MDA-MB-231 cells were grown in serum-free medium and treated with betamethasone (BM) 1 μM alone or in combination with RU486 1 μM for 24 h. Representative images of immunofluorescence are shown. Experiment repeated three times. Scale bars, 15 μm. (b) MDA-MB-231 cells were treated as in a. Representative blots are shown. Experiment repeated three times. (c) Quantification of MDA-MB-231 cells with nuclear YAP by immunofluorescence. Cells were grown in serum-free medium in presence of BM 1 μM alone or in combination with dasatinib (DAS) 0.1 μM or saracatinib (SAR) 0.1 μM or PF573228 (PF) 5 μM for 24 h. Error bars represent mean±s.d., from n=3 biological replicates. (d) MDA-MB-231 cells were grown in serum-free medium in presence of BM 1 μM alone or in combination with DAS 0.1 μM or PF 5 μM for 24 h. Representative blots are shown. Experiment repeated three times. (e) MDA-MB-231 cells were trypsinized and then maintained in suspension and treated with BM 1 μM for indicated times. Representative blots are shown. (f) Quantification of MDA-MB-231 cells with nuclear YAP by immunofluorescence. Cells were transfected with control siRNA (siCTL) or a combination of siRNA targeting LATS1 and 2 (siLATS1/2). The day after cells were grown in serum-free medium in presence of BM 1 μM alone or in combination with DAS 0.1 μM for 24 h. Error bars represent mean±s.d., from n=3 biological replicates. (g) Representative images of immunofluorescence. MDA-MB-231 cells were grown in serum-free medium and treated with BM 1 μM alone or in combination with RU486 1 μM for 24 h. Experiment repeated three times. Scale bars, 15 μm. (h) Quantification of MDA-MB-231 cells with nuclear YAP by immunofluorescence. Cells were grown in serum-free medium in presence of BM 1 μM alone or in combination with latrunculin A (lat.A) 0.5 μM for 24 h. Error bars represent mean±s.d., from n=3 biological replicates. (i) Quantitative PCR with reverse transcription analysis of MDA-MB-231 cells grown in serum-free medium in presence of BM 1 μM alone or in combination with lat.A 0.5 μM for 24 h. Error bars represent mean±s.d., from n=3 biological replicates. (j) MDA-MB-231 cells were treated as in g. Representative blots are shown. Experiment repeated three times. *P<0.05, **P<0.01; two-tailed Student's t-test is used throughout.

Figure 6. YAP is required for glucocorticoids-induced stem cells traits in breast cancer cells.

(a) Upper panel: schematic representation of the experiment. Cells were grown in 2-day culture and treated with indicated compounds for 5 days, then cells were cultured in mammosphere conditions. Lower panel: number of secondary mammospheres generated from the indicated breast cancer cell lines treated with vehicle (NT) or betamethasone 1 μM (BM) alone or in combination with RU486 1 μM. Error bars represent mean±s.d., from n=3 biological replicates. (b) Upper panel: number of secondary mammospheres generated from MII cells overexpressing control vector (CTL) or YAP5SA, transfected with control (siCTL) or YAP (siYAP) siRNA and treated as indicated. HC is hydrocortisone 1 μM. Lower panel: representative images of mammospheres. Error bars represent mean±s.d., from n=3 biological replicates. siYAP sequence is siYAP#1. (c) Quantitative PCR with reverse transcription analysis of MDA-MB-231 transfected with indicated siRNA for 48 h and treated with 1 μM BM alone or in combination with RU486 1 μM for 24 h. siCTL is control siRNA. Error bars represent mean±s.d., from n=3 biological replicates. siYAP sequence is siYAP#1. (d) Number of secondary mammospheres generated from MII cells overexpressing control vector (CTL) or YAP5SA, transfected with indicated siRNA and treated as in a. Error bars represent mean±s.d., from n=3 biological replicates. (e) Tumour volumes 21 days after injection of indicated number of MDA-MB-231-shCTL or MDA-MB-231-shGR cell into mammary fat pad of mice. Error bars represent mean±s.e.m. (f) Lysates of tumours from 125,000 MDA-MB-231-shCTL or MDA-MB-231-shGR cells injected in mice were immunoblotted with the indicated antibodies. The numbers are mice identificative numbers. (g) Schematic representation summarizing drugs and their targets. (h) Number of secondary mammospheres from MII cells overexpressing control vector (CTL) or YAP5SA. Cells were treated as in a with indicated compounds. Error bars represent mean±s.d., from n=3 biological replicates. *P<0.05, **P<0.01; two-tailed Student's t-test is used throughout.

Figure 7. Glucocorticoid receptor activation correlates with YAP activity in breast cancer and is involved in chemoresistance.

(a) Primary human breast cancers of the metadata set were stratified according to high or low GR activity signature (left panel; right panel25) and then, the levels of the YAP activity signature score were determined in the intrinsic molecular subtypes (PAM50). YAP activity is significantly higher in breast cancers with higher levels of the GR activity signature, as visualized by the box plot. Signature scores have been obtained, summarizing the standardized expression levels of signature genes into a combined score with zero mean. The values shown in graphs are thus adimensional. The bottom and top of the box are the first and third quartiles, and the band inside the box is the median; whiskers represent 1st and 99th percentiles; values that are lower and greater are shown as circles (****P<0.0001, two-tailed Student's t-test). (b) MDA-MB-231 cells transfected with control (siCTL) or YAP (siYAP) siRNA and treated with vehicle (NT) or betamethasone 1 μM (BM) and paclitaxel 0,1 μM (PX) alone or in combination for 48 h. Representative blots are shown. Experiment repeated three times. siYAP sequence is siYAP#1. (c) Number of secondary mammospheres generated from MII cells overexpressing control vector (CTL) or YAP5SA and treated as in a, with indicated compounds. Error bars represent mean±s.d., from n=3 biological replicates. (d) Kaplan–Meier analysis representing the probability of disease-specific survival in basal breast cancer from the metadata set stratified according to high or low GR signature score (left panel; right panel25). The log-rank test P value reflects the significance of the association between high levels of the GR signature score and shorter survival in GR signature high as compared with GR signature low patients (P<0.05). (e) Proposed working model. *P<0.05, **P<0.01; two-tailed Student's t-test is used throughout.