TMPRSS2-ERG activates NO-cGMP signaling in prostate cancer cells

Figures

• 

  Fig. 1. Identification of GUCY1A1 and GUCY1B1 as novel ERG-regulated genes that are clinically associated with TMPRSS2-ERG fusion in PCa. a ERG-upregulated genes were identified using Affymetrix microarray analyzes (Human Genome U133 Plus 2.0 Array) on VCaP cells stably infected with lentiviral shRNA against non-target-control or ERG and subsequently treated with ethanol or 10 nM DHT for 24 h. These genes were further examined for clinical correlation with ERG expression using TCGA PCa dataset. Note: The stable cells were hormone-starved for 3 days (d) before the treatment. b The mRNA expression correlation analyzes on GUCY1B1 vs. ERG or GUCY1A1 vs. ERG using TCGA PCa dataset (N = 333) from cBioPortal. The Pearson and Spearman rank correlation coefficient were shown and (*) indicates the p-value < 0.05. c, d Co-expressed genes associated with (c) GUCY1B1 or d GUCY1A1. e Co-exist mutations associated with the expressions of both GUCY1B1/A1. f Box plots for GUCY1B1 or A1 expression in ERG fusion-positive PCa vs. ERG fusion-negative PCa samples

• 

  Fig. 2. ERG directly regulates the expression of sGC in PCa cells. a, b VCaP cells transfected with siRNA against non-target-control (siNTC) or ERG (siERG1 or 2 targeting different regions of ERG) were subjected to a real-time RT-PCR or b immunoblotting. c Four PCa cell lines were hormone-starved for 3d and stimulated with DHT (10 nM) for 1d, and then subjected to immunoblotting. d ERG ChIP-seq binding peaks on gene loci of GUCY1B1/A1 in VCaP cells. e ChIP-qPCR for ERG binding at indicated gene loci. f, g LNCaP stable cells expressing doxycycline (dox)-inducible ERG (LNCaP-tetERG) were treated with dox for 2d, and then subjected to f immunoblotting or g real-time RT-PCR. h ChIP-qPCR for ERG binding (against HA tag) in LNCaP-tetERG cells after dox treatment for 2d

• 

  Fig. 3. ERG activates cGMP synthesis in PCa cells in vitro and in vivo. a VCaP and LNCaP cells treated with 0, 10 or 50 μM riociguat for 24 h were subjected to the measurement of cellular cGMP level using ELISA kit. b, c VCaP cells stably infected with lentiviral shNTC or two independent shRNA against ERG (shERG-1,2) were b subjected to immunoblotting or c treated with or without riociguat (50 μM for 24 h), followed with cGMP ELISA assay. d LNCaP-tetERG cells pretreated with doxycycline for 3d and then treated with 50 μM riociguat for 24 h were subjected to cGMP ELISA assay. e Subcutaneous xenograft tumors derived from LNCaP (N = 4), CWR22-RV1 (RV1) (N = 4), or VCaP (N = 7) were established and then tumor biopsies were taken to examine the intratumoral cGMP level. f–h Equal amount of VCaP-shNTC or VCaP-shERG stable cells were subcutaneously injected into the left or right flank of the same mouse (N = 4) to allow xenograft tumor establishment. When the first xenograft reached ~1 cm in diameter, the group of mice were sacrificed and f tumor volumes were measured. Tumor biopsies from two mice were then taken for subsequent analyzes with g immunoblotting and h cGMP ELISA asay

• 

  Fig. 4. Global identification of sGC-regulated genes in TMPRSS2-ERG positive PCa cells. a–c VCaP cells transfected with siNTC or siRNA against GUCY1B1/A1 (siB1/siA1) individually or in combination were subjected to a immunoblotting, b real-time RT-PCR, or c cGMP ELISA assay with/out riociguat stimulation (12 h). d, e VCaP cells transfected with siNTC, siA1, or siB1 were subjected to RNA-seq analysis. d Venn diagram for overlapping or differentially regulated genes by GUCY1B1/A1 (cutoff 1.5-fold) and e KEGG pathway analyzes on GUCY1B1 or A1 regulated genes were presented. f A panel of GUCY1B1 upregulated genes were selected for subsequent validation. VCaP cells transected with siNTC, siA1 or siB1 (upper panel) or VCaP cells treated with NS2028 or KT5823 (1d) (lower panel) were subjected for real-time RT-PCR analyzes

• 

  Fig. 5. sGC-mediated cGMP signaling promotes cell proliferation in TMPRSS2-ERG-positive PCa cells. a VCaP and LNCaP cells treated with riociguat (0–10 μM) for 5d, and cell growth was measured by flow cytometry-based MUSE proliferation assay. b VCaP, CWR22-RV1, or LNCaP cells treated with NS2028 (0–10 μM) for 5d were examined for cell proliferation. c VCaP cells treated with NS2028 (0–10 μM) for 2d were subjected to cGMP ELISA assay. d, e VCaP cells pretreated with d riociguat (0–50 μM) or e 8-Br-cGMP (0~ 50 μM) were treated with NS2028 (0~ 10 μM) for 5d, followed by cell proliferation assay. f VCaP or CWR22-RV1 cells treated with KT5823 (0~ 50 μM) for 5d were examined for cell proliferation. g VCaP cells treated with KT5823 (0~ 50 μM) with/out 8-Br-cGMP (50 μM) for 5d were examined for cell proliferation

• 

  Fig. 6. sGC inhibitor treatment decreases PCa tumor growth in vivo. a, b Male mice bearing VCaP xenograft tumors (N = 8) were treated with vehicle or NS2028 (25 mg/kg) through daily intraperitoneal injection. a Tumor growth was measured at indicated time points and b Intratumoral cGMP levels were measured in tumor biopsies at 19d. c–e Male mice bearing VCaP xenograft tumors (N = 6) were treated daily with vehicle, NS2028 alone (50 mg/kg), enzalutamide alone (10 mg/kg), or in combination. c Tumor growth was measured at indicated time points, d the expression of ERG, GUCY1B1, and GUCY1A1 was measured in tumor biopsies at 13d using qRT-PCR, and e Intratumoral cGMP levels were measured using the same group of biopsies. f, g Castrated male mice bearing LuCaP35CR xenograft tumors (N = 6) were treated daily with vehicle or NS2028 (50 mg/kg). f Tumor growth was measured at indicated time points and g intratumoral cGMP levels were measured in tumor biopsies at 19d