Fig 2: miR‐144‐3p targeted the ONECUT2 gene. (A) Venn diagram demonstrated ONECUT2‐targeted miRNAs. (B) Bioinformatics and dual luciferase reporter gene assay analyzed the binding sites of ONECUT2 to miR‐144‐3p. (C) RIP was used to assess the effects of miR‐144‐3p mimic on ONECUT2 enrichment in Anti‐IgG and Anti‐Ago2. (D) The expression of miR‐144‐3p was calculated by qRT‐PCR. *P < 0.05 vs Control group.

Fig 3: NEPC is highly hypoxic and sensitive to hypoxia-directed treatment. a, b Hypoxia marker gene CA9 IHC staining in two independent tissue microarray sets. The combined H-score, taking into account of both staining intensity and percentage, was used to quantify CA9 protein levels. (c) Representative images of tissue microarray analysis of AR, SYP, and CA9 IHC staining in benign prostate, primary PCa, adeno-CRPC, and NEPC tissues. Scale bar = 50 µm. d Correlation between ONECUT2 upregulated genes (summarized in z-scores) and hypoxia scores in Beltran adeno-CRPC/NEPC dataset. e PC3 xenograft tumor growth in response to hypoxia-activated prodrug TH-302 treatment, with and without silencing of ONECUT2. n = 7 for shCtrl_Vehicle; n = 6 for shCtrl_TH302; n = 8 for shOC2_Vehicle; n = 7 for shOC2_TH302. f Representative images of PIMO IHC staining of V16A xenograft tumors with and without overexpression of ONECUT2. Scale bar = 200 µm. g V16A xenograft tumor growth in response to TH-302 with and without overexpression of ONECUT2. n = 7 for EV_Vehicle; n = 8 for EV_TH302; n = 6 for OC2_Vehicle; n = 6 for OC2_TH302. h Relative expression levels of related genes in the adeno-CRPC (LTL484) and NEPC (LTL545) PDX models determined by microarray analysis. i TH-302 treatment suppresses tumor growth in NEPC patient-derived xenografts. n = 4 for vehicle group and n = 4 for TH-302 group in NEPC PDXs; n = 5 for vehicle group and n = 5 for TH-302 group in adeno-CRPC PDXs. In e, g, and i, inhibition rate (IR) of tumor growth was calculated as (Tumor volume Vehicle-Tumor volume TH-302)/Tumor volume Vehicle. P values were determined by mixed-effects models of repeated-measures ANOVA for e, g, and i. Source data are provided as a Source Data file

Fig 4: TGF-β–dependent activation of the miR-218-5p ceRNA network. (A) qPCR analysis of TGF-β family members and TGF-β receptors. Values were normalized to H2122-P. (B) TGF-β1 ELISA assay on CM samples. (C) WB analysis of p-Smad3 (Ser423/425) and total Smad2/3 in cell lysates. α-Tubulin served as loading control. (D) qPCR analysis of TGF-β downstream targets (CTGF, PAI-1, and LEF1) and predicted miR-218-5p targets (LEF1, OC2, N-cad, and PI4KIIα) in H2122-P cells that had been treated for 5 d with DMSO or 10 ng/mL TGF-β1. Values were normalized to DMSO control. (E) WB analysis of cells in D. β-Actin served as loading control. (F) WB analysis of H2122-BT cells treated for 72 h with DMSO or 5 µM ALK5i. (G and H) WB analysis of H2122-P cells that were treated for 72 h with 10 ng/mL TGF-β1 and then transfected with miR-218-5p mimics (G) or siRNAs against OC2 or N-cad (H). Values are shown as mean ± SEM of triplicate biological replicates. P values: one-way ANOVA in A and B and two-sided t test in D. *P < 0.05, **P < 0.01, and ***P < 0.001.

Fig 5: Pan-cancer analysis identifies ONECUT2 as a potential master transcriptional regulator of neuroendocrine tumors. a Schematic illustration of pan-NET analysis (see Methods section and Supplementary Figure 1a for details). Ninety-three genes, including nine transcription factors, were commonly upregulated in NETs compared to the non-NET counterparts within the same cancer types. NET neuroendocrine tumor, NEPC neuroendocrine prostate cancer, Adeno-CRPC castration-resistant prostatic adenocarcinoma, SCLC small-cell lung cancer, NSCLC non-small-cell lung cancer (exclusive of large cell lung cancer), NB neuroblastoma, CCLE Cancer Cell Line Encyclopedia. b Network analysis of nine TFs and 79 non-TF genes commonly upregulated in poorly differentiated NETs. These 88 genes were classified into two communities using GEPHI, as labeled by color (orange and cyan). c Dysregulation of the five TFs from the major community from panel b in 20 cancer types from TCGA. log2 transformed fold changes represent difference in expression between tumor and normal tissue. Colored cells indicate significant expression changes (absolute-value of the log2(fold change) > 1 and P < 0.05). The Wilcoxon rank sum test was used to calculate P-values. d ONECUT2 expression in benign prostate tissues and prostate tumors. RNA-Seq data of 52 benign prostate tissues and 333 primary PCa were retrieved from TCGA. RNA-Seq data of 34 adeno-CRPC and 15 NEPC were retrieved from the Beltran dataset. Expression levels as measured by RPKM were normalized by housekeeping gene ACTB