Fig. 3: ERG is a responsible transcriptional regulator of the 17q12 locus and regulates HNF1B in PCa cells and clinical tissues in a manner dependent on TMPRSS2-ERG fusion status. | Nature Communications

Fig. 3: ERG is a responsible transcriptional regulator of the 17q12 locus and regulates HNF1B in PCa cells and clinical tissues in a manner dependent on TMPRSS2-ERG fusion status.

From: Extensive germline-somatic interplay contributes to prostate cancer progression through HNF1B co-option of TMPRSS2-ERG

Fig. 3

a, b Unbiased genome-wide analysis showing ERG as the most co-expressed gene positively correlated with HNF1B in PCa specimens (n = 497). P values were assessed by the two-sided Pearson’s product-moment correlation and Spearman’s rank correlation rho tests. c Depletion of ERG results in reduced mRNA levels of HNF1B in TMPRSS2-ERG fusion positive VCaP cell. n = 3 samples; P values based on the order of appearance: 9E–06, 2E–04, 2E–06, 1E–05. d Genome browser representation of ChIP-seq signals of active enhancer marks and transcription factor ERG at HNF1B regions harboring several potential causal SNPs as indicated. e–i ChIP-qPCR validation for chromatin enrichment of ERG, HNF1B, AR, H3K4me1, H3K4me2, H3K4me3, and H3K27ac at given SNP regions in VCaP cells treated with 100 nM DHT and without (ETH-treated). (e, g) n = 16 samples, (f, h, i) n = 14 samples; (e) P values based on the order of appearance: 1,2E–05, 2,2E–05, 1,6E–03, 1,6E–03, 0,053, 2,8E–03, 1E–04, 2,4E–05, 1,2E–07, 9,7E–07, 2E–05, 9,7E–07; (f) P values based on the order of appearance: 4,5E–04, 4,6E–05, 0,13, 4,1E–03, 2,9E–05, 1,4E–06, 1,6E–04, 3,2E–07, 3,1E–04, 3,2E–07; (g) P values based on the order of appearance: 9E–04, 5,1E–05, 3,9E–03, 0,012, 0,63, 0,0204, 4,2E–04, 6,9E–06, 5,9E–06, 1,4E–06, 2,2E–05, 8,6E–06; (h) P values based on the order of appearance: 4,1E–03, 7,1E–05, 0,73, 0,021, 5,8E–05, 6,3E–07, 3,2E–05, 9,3E–06, 1,7E–04, 6,1E–05; (i) P values based on the order of appearance: 9E–03, 3,6E–05, 0,54, 4,7E–03, 0,131, 1,4E–03, 8,9E–03, 2E–04, 2,2E–04, 1E–04, 3,1E–05, 1,E–05 j, k Scatter plots showing positive correlations between ERG and HNF1B expression in PCa specimens with TMPRSS2-ERG fusion in the TCGA (n = 203) and MSKCC (n = 74) cohorts. P values were assessed by the two-sided Pearson’s product-moment correlation and Spearman’s rank correlation rho tests. Seven SNPs individually had a significant eQTL association with HNF1B expression in TMPRSS2-ERG fusion positive PCa group (n, n = 160) but not in the whole cohort (l, n = 389) or in ERG fusion negative group (m, n = 228). P values were assessed by the linear regression. In (c) and (e–i), n = 3 technical replicates, error bars, mean ± SD, * P < 0.05, ** P < 0.01, *** P < 0.001, ns: non-significant, P values were evaluated using two-tailed Student’s t tests. Source data are provided in Source Data file.

Back to article page