Fig 2: Knockdown of ZNF507 leads to altered tumor growth in vivo and diminish NEPC-like phenotypes by TGF-β signaling inhibition. A Growth curve of the tumor formation after the scramble, shZNF507, Mock, or ZNF507OE DU145 cells were xenografted into Balb/C nude mice (n = 6 per each set group). For each group, three tumor samples were used for the Western blot analysis, while other three samples were used for the IF. B Tumors from scramble and shZNF507 group were extracted on the last day (day 33) and Mock and ZNF507 OE group were extracted on day 18 of the xenograft experiments conducted through injection. C Graph showing the average weight of the scramble and shZNF507 xenografted tumors (n = 6 per each group). D Histological analysis of tumors from the scramble and shZNF507 xenograft. Representative data of H&E staining and immune staining against ZNF507 (with PI), TGFBR1, MAP3K8, and FURIN in tumors, performed by immunofluorescence (Scale bar = 100 μm). E Representative protein expression data of ZNF507, TGFBR1, MAP3K8, and FURIN in the xenografted tissues as determined by western blot analysis. β-actin served as a loading control. Right panel shows the average of the relative protein levels normalized to the β-actin. At least three samples per group were used for each experiment and the three independent experiments were performed and analyzed. F Representative protein expression data of canonical TGF-β signal proteins, Smad2, Smad3, and their phosphorylated forms, assessed by western blot. β-actin was used as a normalization control. At least three samples per group were used for each experiment and the three independent experiments were performed and analyzed. G Representative protein expression levels of non-canonical TGF-β signal proteins, RAS, MEK, and p-MEK, ERK, and p-ERK, examined by western blot. β-actin was used as a normalization control. At least three samples per group were used for each experiment and the three independent experiments were performed and analyzed. H Representative data of H&E staining of the liver and lung from in vivo metastasis assay. I Representative images of the immune staining against ZNF507, TGFBR1, MAP3K8, and FURIN in liver and lung from in vivo metastasis assay, performed by immunofluorescence. J Schematics presenting the biological role of ZNF507 in TGF-β signaling, as a transcriptional activator, and related pathway in the progression of PC to its aggressive mCRPC and NEPC state. Data are presented as the Means ± SD from three independent experiments. #p < 0.05, ##p < 0.01, Mock versus the ZNF507 OE xenograft tissue and *p < 0.05, **p < 0.01, ***p < 0.001 shZNF507 versus the scramble xenograft tissue

Fig 3: Decreased cancer properties, including cell proliferation, colony formation, and metastasis from ZNF507 knockdown in PC cells. A, B Relative ZNF507 mRNA levels from shRNA-mediated knockdown compared with the scramble control in DU145 and 22Rv1 cells. Each cell lines were prepared for three independent experiments with at least three samples per group. C Representative protein expression data assessed by western blot for ZNF507 in scramble or shZNF507 DU145 or 22Rv1 cells; shZNF507 #5 treated DU145 and 22Rv1 cells. D, E CCK-8 proliferation assay of the scramble or shZNF507 DU145 or 22Rv1 cells. The optical density (OD) value at 450 nm was determined. Each experiment consists of six samples from each group of cells were performed. Three independent experiments were performed. F Colony formation of the shZNF507 treated DU145 or 22Rv1 cells performed by soft agar assay. G Statistical analysis of colony number and relative diameter analyzed from the colony formation assay data. Each experiment consists of four samples from the same group of cells were measured and the three independent experiments were conducted. H PCNA staining of the scramble or shZNF507 DU145 or 22Rv1 cells performed by immunocytochemistry (Scale bar = 50 μm). The graph in the right panel presents relative corrected total cell fluorescence (CTCF). At least 4 pictures from each sample were taken and the CTCF were calculated. I Representative images of the invasion and migration assay performed with the scramble or shZNF507 DU145 or 22Rv1 cells using trans-well plate (Scale bar = 50 μm). J Relative ratio of migration and invasion of the data from the transwell migration and invasion assay. At least four pictures from each sample were taken and the three independent experiments were performed. K, L Relative mRNA expression of Slug, Twist1, Snail, KLF8, ZEB1, and ZEB2 in the scramble or shZNF507 DU145 or 22Rv1 cells analyzed by qRT-PCR. GAPDH was used as a normalization control. The three independent experiments consist of at least four samples per group were performed. M Representative images of protein expression data of E-cadherin, N-cadherin, Vimentin, Snail, and Slug in the scramble or shZNF507 DU145 or 22Rv1 cells assessed by western blot. β-actin was used as an endogenous control. The data are presented as the Means ± SD from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001 versus scramble control

Fig 4: Increased ZNF507 expression in various PC databases and PC tissues. A ZNF507 expression data in normal (n = 42) and PC tissues (n = 277) as retrieved from the GENT2 database. B ZNF507 expression data in normal (n = 13) and PC tissues (n = 44) as retrieved from the EMBL-EBI database (Liu et al.). C The survival rate of PC patients compared by low-ZNF507 (n = 246) expressed and high-ZNF507 (n = 246) expressed groups from the GEPIA database. TPM; Transcripts per million, HR; hazard ratio. The median value was used for the group classification. p = 0.82, HR = 0.86. D Relative ZNF507 expression in non-recurrent (primary) PC (n = 37) and recurrent PC tissues (n = 45) as retrieved from the EMBL-EBI database (Sun et al.). E ZNF507 expression in the normal and Gleason graded PC tissues as retrieved from the EMBL-EBI database (Liu et al.) (NA, n = 13; grade 6, n = 13; grade 7, n = 16; grade 8, n = 10; grade 8+, n = 5; *, compared to NA group; #, compared to grade 6 group; §, compared to grade 7 group). F ZNF507 expression in normal (n = 28), primary PC (n = 59), and mCRPC (n = 34) as retrieved from the GEO database (*, compared to normal group; #, compared to primary PC group). G Genomic alteration frequency of ZNF507 in NEPC, mCRPC, and PA as retrieved from the cBioPortal database. H Average of the relative ZNF507 expression from benign hyperplasia (n = 6) and Gleason score 8 patient tissues (n = 15) specimens compared to its related β-actin intensity. I Representative western blot data showing ZNF507 expression from benign hyperplasia and Gleason score over 8 PC patient tissues. J Relative mRNA expression of ZNF507 from benign hyperplasia (n = 16) and Gleason score 8 PC tissues from patients (n = 14). K Representative IHC staining of ZNF507 in benign hyperplasia and Gleason grade 4 and 5 PC tumor tissues (Scale bar = 100 μm). The data are presented as the Means ± SD from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001 versus normal control

Fig 5: ZNF507 as an activator of TGFBR1, MAP3K8, and FURIN transcription, and their positive correlation in PC. A Spearman correlation analysis of ZNF507 and TGFBR1 levels compared to the Gapdh levels in PC samples based on the GEPIA database (TPM, Transcripts per Kilobase Million, n = 472). B Spearman correlation analysis of ZNF507 and MAP3K8 expression normalized to the Gapdh levels in PC based on the GEPIA database (n = 472). C Spearman correlation analysis of ZNF507 and FURIN normalized to the Gapdh levels in PC from the GEPIA database (n = 472). D Representative protein expression data of ZNF507, TGFBR1, MAP3K8, and FURIN in the benign hyperplasia and PC tissue (Gleason score over 8) specimens assessed by western blot. β-actin was used as a normalization control. E Average of relative protein expression of ZNF507, TGFBR1, MAP3K8, and FURIN compared with the intensity of β-actin in the benign hyperplasia (n = 6) and PC tissue (Gleason score over 8m n = 15) specimens. Three independent experiments were performed and the average rate of each sample were analyzed. F Relative mRNA expression of ZNF507, TGFBR1, MAP3K8, and FURIN determined by qRT-PCR analysis of the ZNF507 overexpression (ZNF507 OE) DU145 cells compared to the mock control.G Representative images of the protein levels of ZNF507, TGFBR1, MAP3K8, and FURIN in the mock and ZNF507 OE DU145 cells examined by western blot. β-actin was used as a normalization control. H, I Schematics of TGFBR1 and MAP3K8 promoter regions and related four sets of ChIP qRT-PCR primers. J Dual reporter luciferase assay conducted with pGL3-Basic or pGL3 containing TGFBR1 promoter vectors and pcDNA3.1-Mock or pcDNA3.1-ZNF507 vectors. K Dual reporter luciferase assay conducted with pGL3-Basic or pGL3 containing MAP3K8 promoter vectors and pcDNA3.1-Mock or pcDNA3.1-ZNF507 vectors. L qRT-PCR analysis of TGFBR1 promoter regions conducted with 4 sets of specific primers after a ChIP assay with DU145 cells, with anti-ZNF507 antibody or the IgG isotype control. The results presented are fold enrichments relative to IgG control. M qRT-PCR analysis of MAP3K8 promoter regions conducted with 3 sets of specific primers after a ChIP assay in DU145 cells using an anti-ZNF507 antibody to query binding, with IgG used as an antibody control. The results presented are fold enrichment relative to the IgG control. Data are presented as the Means ± SD from three independent experiments including at least 3 independents samples. *p < 0.05, **p < 0.01, ***p < 0.001 versus normal tissue or Mock control