High expression of NR3C1 and ESRRG are associated with better patient outcomes. (A) Relative cell proliferation from EtOH-treated (gray line), Dex-treated (blue line), DHT-treated (red line), and DexDHT-treated (black line) VCaP cells. Data are normalized to day 0 and represent the mean ± SD of eight to 10 images per day. Statistical significance calculated with two-way ANOVA with a Bonferroni post hoc test, and comparisons between DHT with other treatment are shown denoted with colored stars. (B) Genome browser track depicts chromatin binding of AR and GR at the ESRRG locus. (C) Violin plot represents the single-cell expression level of ESRRG in EtOH-, Dex-, DHT-, and DexDHT-treated VCaP cells. (D) Percentage of ESRRG-expressing single cells in EtOH-, Dex-, DHT-, and DexDHT-treated VCaP cells. Statistical significance calculated with a chi-squared test. (E) Disease-free survival of TCGA PCa patients with combination of high and/or low expression levels of NR3C1 and ESRRG. Statistical significance calculated with a log-rank test. (F) Percentage of TCGA PCa patients with or without biochemical recurrence event divided based on high NR3C1 and high ESRRG (n = 124), high NR3C1 and low ESRRG (n = 123), low NR3C1 and high ESRRG (n = 123), or low NR3C1 and low ESRRG (n = 123) expression. White bars represent no biochemical recurrence event (w/o), and black bars represent biochemical recurrence event (w/). Statistical significance calculated with a chi-squared test. (G) Box plots represent normalized proliferation marker MKI67 expression of TCGA PCa patients with high NR3C1 and high ESRRG (n = 124), high NR3C1 and low ESRRG (n = 123), low NR3C1 and high ESRRG (n = 123), or low NR3C1 and low ESRRG (n = 123) expression. (H) Model of AR assisting the loading of GR to chromatin through modulation of chromatin accessibility and enhancer activation (top), and expansion of AR-regulated transcriptome by GR that leads to restriction of cell proliferation and PCa progression (bottom). Statistical significance calculated using one-way ANOVA with a Bonferroni post hoc test. All genome browser tracks are normalized to a total of 10 million reads. (*) P < 0.05, (**) P < 0.01, (***) P < 0.001.
