PGC1 alpha coactivates ERG fusion to drive antioxidant target genes under metabolic stress

Abstract

The presence of ERG gene fusion; from developing prostatic intraepithelial neoplasia (PIN) lesions to hormone resistant high grade prostate cancer (PCa) dictates disease progression, altered androgen metabolism, proliferation and metastasis^1-3. ERG driven transcriptional landscape may provide pro-tumorigenic cues in overcoming various strains like hypoxia, nutrient deprivation, inflammation and oxidative stress. However, insights on the androgen independent regulation and function of ERG during stress are limited. Here, we identify PGC1α as a coactivator of ERG fusion under various metabolic stress. Deacetylase SIRT1 is necessary for PGC1α-ERG interaction and function. We reveal that ERG drives the expression of antioxidant genes; SOD1 and TXN, benefitting PCa growth. We observe increased expression of these antioxidant genes in patients with high ERG expression correlates with poor survival. Inhibition of PGC1α-ERG axis driven transcriptional program results in apoptosis and reduction in PCa xenografts. Here we report a function of ERG under metabolic stress which warrants further studies as a therapeutic target for ERG fusion positive PCa.

Fig. 1. PGC1α acts as a coactivator for ERG fusion under metabolic stress.

a VCap Cells were treated for 24 h. Relative transcript levels of genes were analyzed using qRT-PCR. b VCaP cells were treated for 24 h and the whole-cell extract was immunoblotted using indicated antibodies. c PC3 cells were cotransfected as indicated. Luciferase assay was performed post 24 h of transfection and relative firefly luciferase activity was plotted. d VCaP control (Scrambled shRNA) or VCAP PGC1α KD (PGC1α shRNA#1) stable cells transfected with SPP1 luciferase were glucose deprived(5 mM) or cultured under basal condition(25 mM) post-transfection as indicated. Luciferase assay was performed and relative firefly luciferase activity was plotted. e VCaP control or VCAP PGC1α KD stable cells transfected with SPP1 luciferase were serum-deprived or cultured under basal condition post-transfection as indicated. Luciferase assay was performed and relative firefly luciferase activity was plotted. f VCaP control or VCAP PGC1α KD stable cells were serum-starved (Left panel) or received cold stress (4 °C, Right panel) as indicated for 24 h and the whole-cell extract was immunoblotted using indicated antibodies. g VCaP control or VCAP PGC1α KD stable cells were serum-starved (Left panel) or received cold stress (4 °C, Right panel) as indicated for 24 h. Relative transcript levels of genes were analyzed using qRT-PCR. Experiments were performed as biological triplicates (mean ± SD). Two-way ANOVA with multiple comparison test used for statistical significance of 1a, 1g. For 1c, d, e; one-way ANOVA with Dunnett’s multiple comparison test used for statistical significance where ****p < .0001, ***p = 0.0001, **p = .001, *p = .01, n.s not significant.

Fig. 2. PGC1α interacts and transactivates ERG fusion.

a VCaP cells were treated as indicated. Cells were harvested after 24 h and subjected to immunoprecipitations using anti-PGC1α antibody or anti-IgG antibody, and western blots were performed for the indicated proteins. b VCaP cells were treated as indicated. Cells were harvested after 24 h and subjected to immunoprecipitations using anti-ERG antibody or anti-IgG antibody, and western blots were performed for the indicated proteins. c VCaP cells were transfected with either vector control(EV), PGC1α Flag, or ΔN-PGC1α Flag as indicated. Cells were harvested 24 h post-transfection and subjected to immunoprecipitations using anti-FLAG antibody and western blots were performed for the indicated proteins. d PC3 cells were cotransfected as indicated. Luciferase assay was performed 24 h post-transfection and relative firefly luciferase activity was plotted. Experiments were performed as biological triplicates (mean ± SD). One-way ANOVA with Dunnett’s multiple comparison test was used for statistical significance of 2d where **p = 0.001.

Fig. 3. Physiological role of ERG-PGC1α axis during metabolic stress.

a VCaP cells were transfected with either SiRNA Negative control (SiNeg) or SiRNA Sirt1. Post 24 h of transfection, cells were glucose deprived as indicated, harvested, and subjected to immunoprecipitations using anti- PGC1α antibody. Western blots were performed for the indicated proteins. b VCaP cells were glucose deprived and subjected to either 0.1% DMSO or 10 μM of Sirt1 inhibitor (EX-527) treatment for 24 h. Cells were harvested and subjected to immunoprecipitations using an anti-ERG antibody, and western blots were performed for the indicated proteins. c SPP1 luciferase transfected VCaP cells were glucose deprived and subjected to either 0.1% DMSO or 10 μM of SIRT1 inhibitor as indicated. Luciferase assay was performed and relative firefly luciferase activity was plotted. d SPP1 luciferase transfected VCaP cells were glucose deprived post-ERG transfection and subjected to either 0.1%DMSO or 10 μM of SIRT1 inhibitor as indicated. Luciferase assay was performed and relative firefly luciferase activity was plotted. e ROS levels in treated cells were measured by FACS analysis after staining with fluorescent dye DCFDA. f ROS levels of treated cells were measured by flow cytometry after staining with the fluorescent dye DCFDA. g ROS levels in glucose deprived VCaP cells treated with either 0.1% DMSO or 10 μM of Sirt1 inhibitor were measured by flow cytometry after staining with the fluorescent dye DCFDA. h ROS levels in treated cells post-transfection with indicated constructs were measured by FACS analysis after staining with fluorescent dye DCFDA. i Relative caspase3/7 activity of glucose-deprived VCaP control or VCAP PGC1α KD stable cells were measured at the indicated time. ROS scavenger; Tiron (5 mM) was added for the last 12 h as indicated. j Relative caspase3/7 activity of glucose-deprived VCaP cells treated with either DMSO or 10 μM SIRT1 inhibitor for an indicated time were measured. ROS scavenger; Tiron (5 mM) was added for the last 12 h as indicated. Experiments were performed as biological triplicates (mean ± SD). For 3c–j one-way ANOVA with Dunnett’s multiple comparison test used for statistical significance were performed; where ****p < 0.0001, ***p = 0.0001, **p = 0.001, *p = 0.01, n.s not significant.

Fig. 4. PGC1α mediated transactivation of ERG fusion revealed novel antioxidant target genes.

a Enrichment plot from GSEA was conducted with datasets of ERG Knockdown (top panel) and ERG overexpression (bottom panel). Plots represent statistically significant enrichment (FDR <0.25) of “reactive oxygen species pathway” genes. b Heatmap of core enriched genes in “reactive oxygen species pathway” from GSEA analysis as indicated. c Aligned chromosomal peak regions and enriched ChIP signals (ERG ChIP left and AR ChIP right) in bigwig format were visualized using the UCSC browser for indicated samples. d ChIP assay performed with either ERG or control IgG antibody on glucose-deprived cells. The fold enrichment of co-precipitating DNA was determined by qPCR for indicated promoters. e VCaP control or VCAP PGC1α KD cells treated for 24 h. Relative transcript levels of indicated genes were analyzed using qRT-PCR. f ROS levels of treated VCaP cells were measured by flow cytometry after staining with the fluorescent dye DCFDA. g Relative caspase3/7 activity of treated VCaP cells were measured. h Representative cleaved-caspase3 staining (20X top, 40X bottom) in VCaP control and VCaP PGC1α KD xenograft following STF on day 21. i VCaP (control) and VCaP(PGC1α KD) xenograft tumor volume (mm³) following STF on days 4, 21, and 42. Lines indicate the mean ± SD (n = 6 per group). j Violin plots for transcript level in PRAD patient data set (TCGA targeted GTEx) were accessed using the Xena browser. Data were Z-score normalized and p value calculated using Welch’s t-test. k Kaplan–Meier curves for survival probability of TCGA PRAD data separated into high- and low-risk groups. Survival analysis was performed using Surv Express software. Experiments were performed as biological triplicates (mean ± SD). Two-way ANOVA with multiple comparison tests used for statistical significance of 4e. For 4 d, f, g, i one-way ANOVA with Dunnett’s multiple comparison test used for statistical significance were performed; where ****p < 0.0001, ***p = 0.0001, **p = 0.001, *p = 0.01, n.s not significant.