Fig 2: GREB1 amplifies AR transcriptional activity by enhancing AR binding to chromatin.(A) GREB1 overexpression in LNCaP ARsig-lo cells with stable integration of a GREB1 lentiviral vector containing an amino-terminal HA-tag. (B) LNCaP ARsig-lo cells with GREB1 overexpression show higher induction of AR target genes in response to DHT treatment. The q-PCR data is presented as mean fold change ±SD relative to the DMSO control. NS = not significant, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001, unpaired t-test compared to the control cells. (C) GREB1 overexpression in LNCaP ARsig-lo cells increases protein levels of AR target genes without affecting AR level. (D) Co-immunoprecipitation using nuclear extracts shows an interaction between AR and GREB1 (HA) in LNCaP ARsig-lo cells. (E) ChIP against the HA-tag shows GREB1 binding on the KLK3 and FKBP5 enhancer regions in LNCaP ARsig-lo cells. *p<0.05, unpaired t-test. (F–G) LNCaP ARsig-hi cells have increased EP300 binding on the KLK3 and FKBP5 enhancer regions in a GREB1 dependent manner. *p<0.05, **p<0.01, unpaired t-test. (H) GREB1 knockdown or CRISPR decreases AR binding to KLK3 enhancer in LNCaP ARsig-hi cells. *p<0.05, **p<0.01, ***p<0.001, unpaired t-test (shRenilla vs. shGREB1), one-way ANOVA (SgNT vs. SgGREB1). The ChIP q-PCR data (E–H) is presented as mean percentage input ±SD. (I) Overlap of AR ChIP-sequencing peaks shows that AR peaks are disrupted by GREB1 knockdown in LNCaP ARsig-hi cells. (J) ChIP-sequencing summary plot shows that AR enrichment across the AR-binding sites is reduced by GREB1 knockdown. (K) Example of AR genomic peaks at NKX3-1.

Fig 3: AR is regulated by Akt1 in the castration-resistant prostate cancer cell line. (A) Western blot analysis of p-Akt1, total Akt1, AR, PSA and FKBP5 expression in C4-2 cells transfected with miR-149 mimic and miR-149 inhibitor. (B) Semi-quantification of western blotting results demonstrated the expression of p-Akt1/total Akt1, AR/β-actin, PSA/β-actin and FKBP5/β-actin. Data are expressed as the mean ± standard deviation (n=3). **P<0.01 vs. Akt1 NC + miR-149 mimic NC group. AR, androgen receptor; miR, microRNA; p-, phosphorylated; NC, negative control; FKBP5, peptidyl-prolyl cis-trans isomerase FKBP5; KD, knockdown.

Fig 4: Effects of docetaxel and YK-4-279 alone or in combination on the levels of ETV1, AR, PSA, p-STAT3, survivin, Bcl-2 and p-Akt in LNCaP and PC-3 cells. LNCaP and PC-3 cells were cultured at a density of 1 × 105 cells/ml in 100 mm tissue culture dishes (10ml/dish) for 24 h. The cells were then treated with docetaxel (0.5nM) and YK-4-279(0.5µM) alone or in combination for 24h (for analysis of p-Akt, survivin, PSA, p-STAT3 and AR) and 48 h (for analysis of Bcl-2). The levels of AR,Bcl-2, p-STAT3, PSA,p-Akt and survivin were determined by the Western blot analysis. The band density was measured and normalized for actin.

Fig 5: Isolation of CTC clusters from ovarian and prostate cancer patients.a Fluorescence microscope images of the isolated CTC clusters from ovarian cancer patients. Captured cells were stained with Cytokeratin, Vimentin (green), CD45 (red), and DAPI (nuclei, blue). Scale bars, 20 µm. b Images of the isolated CTC clusters from metastatic prostate cancer patients. Captured cells were stained with Cytokeratin, Vimentin, PSA/KLK3, EpCAM (green), CD45(red), and DAPI (nuclei, blue). Scale bars, 20 µm. c Clusters isolated from the peripheral blood and ascites sample drawn from the peritoneal cavity of an ovarian cancer patient at the same timepoint. The isolated CTC cluster was morphologically different compared to the tumor spheroids isolated from ascites sample of the same patient. The spheroids from ascites sample were fixed and stained with Cytokeratin, EpCAM (red), CD45 (green) and DAPI (nuclei, blue), and CTC clusters isolated from blood sample were fixed and stained with Cytokeratin, Vimentin (green), CD45 (red) and DAPI (nuclei, blue). Scale bars, 20 µm. d Percentage of prostate (n = 8) and ovarian (n = 9) cancer patients that are observed to have CTC cluster(s) in their blood samples. e Number of CTC clusters observed per milliliter of blood samples from prostate and ovarian cancer patients. f Distribution of the number of cells observed in the prostate and ovarian CTC clusters. The box plots show the 25th and 75th percentiles, line shows median, and whiskers show maxima and minima points. g Percentage of pure and white blood cell associated CTC clusters for each disease type. h Images of the CTC clusters isolated from an ovarian cancer patient and subjected to a fluorescence-based viability assay. The images show CTC clusters that were composed of (i) both live and dead (yellow) cells, and (ii) live cells only. Scale bars, 20 µm. Source data are provided as a Source Data file.