Fig. 10

CREB repression inhibits growth of NEPC xenografts, and blocks castration-activated EZH2 axis and angiogenesis in vivo. a NOD/SCID mice were injected with luciferase-labelled LNCaP or NE1.3 cells, which were then treated daily with saline or 2 mg kg⁻¹ propranolol for 25 days, as indicated (6 mice in each group, 2 tumors in each mouse, n = 12). The tumor growth was monitored by bioluminescent imaging (BLI) with IVIS instrument. Presented on Y-axis are average fold changes of BLI signals for each tumor at day 25, relative to BLI signals at day 1 of treatments. b Left 4 panels: representative images of IHC staining for pCREB-S133 and H3K27me3 in LNCaP-derived xenografts treated daily for 25 days with saline or 10 mg kg⁻¹ of ISO (tumors from our previous study³⁶). Right 4 panels: representative images of IHC staining for pCREB-S133 and H3K27me3 in NE1.3-derived xenografts treated daily with saline or 2 mg kg⁻¹ of PRO. c Western blots of xenograft tumors from LNCaP with or without ISO treatments (left) or from NE1.3 with or without PRO treatment (right) for the CREB/EZH2/TSP1 pathway proteins as indicated. d Top two panels: representative IHC staining images of angiogenesis marker CD31 in LNCaP CDX treated with saline or ISO. Bottom 2 panels: representative IHC staining images of CD31 in NE1.3 CDX treated with saline or PRO. The arrows indicated typical CD31 + microvessels. e Left panel: dot plots of tumor weights in the three indicated experimental groups that are collected at the end of the experiment. Right panel: photographic picture of LNCaP xenograft tumors in the three groups at sacrifice. f Western blotting of the proteins on the CREB/EZH2/TSP1/NE pathway in the indicated three groups of LNCaP xenograft tumors. g A summary model of the key findings in this study. ADT activates CREB, in part through PKA, which in turn enhances the PRC2 activity of EZH2. EZH2 is critical for ADT/CREB-induced neuroendocrine differentiation and angiogenesis of prostate cancer cells. EZH2 enhances angiogenesis through epigenetically repressing anti-angiogenic factor TSP1. Additional data related to this Figure are in Supplementary Fig. 5. Statistical significance was determined by using unpaired two-sided Student’s t test and shown as mean with s.d. *P < 0.05. Scale bar = 100 µm