SRA

Human silencers have been shown to exist and regulate developmental gene expression. However, the functional importance of human silencers need to be elucidated such as the working mode and whether they can form “super-silencers”. Here, through interrogating two putative silencer components of FGF18 gene, we found that two silencers can cooperate via compensated chromatin interactions to form the “super-silencer”. Furthermore, double knock-out of two silencers exhibited synergistic upregulation of FGF18 expression and changes of cell identity. To disturb the “super-silencers”, we applied the combinational treatment of GSK343 and X5050 (“GR”). We found that GR led to severe loss of TADs and loops, while any single treatment only had mild changes. Such changes of TADs and loops may due to the decreased CTCF proteins upon the GR treatment. Moreover, GSK343 and X5050 can work synergistically to upregulated the super-silencer controlled apoptotic genes, thus gave rise to antitumor effects including apoptosis, cell cycle arrest and tumor growth inhibition. Overall, our data demonstrated the first example of “super-silencer” and showed that combinational usage of GSK343 and X5050 maybe the potential drug to curing cancers through disruption of “super-silencers”. Overall design: Silencer knockout (KO) Hi-C data has two biological replicates for each condition including EV, S1KO, S2KO, and DKO. Drug treatment Hi-C data has two biological replicates for each drug-treated condition, including DMSO, GSK343, X5050, GR 8h, GR 24h, and GR 72h. Drug treatment Hi-C data has two biological replicates for each DMSO- and etoposide-treated K562 cells (72 hr). siRNA Hi-C data has two biological replicates for each siScramble- and siCTCF-treated K562 cells (48 h). Silencer KO 4C has two biological replicates for each condition (EV, S1KO, and DKO). Silencer KO RNA-seq has two biological replicates for different KO cells (EV, S1KO S2KO, and DKO). H3K27me3 and H3K27ac ChIP-seq for silencer KO cells (EV, S1KO, and DKO) have two biological replicates. H3K27me3 HiChIP in wild-type K562 cells has one biological replicate. Cut and Run CTCF has two biological replicates for CTCF antibodies in EV, S1KO, S2KO, and DKO cells. Cut and Run H3K27me3 has two biological replicates for H3K27me3 antibodies (24 h) in THP1 and K562 cells. ATAC-seq has two biological replicates for each EV, S1KO, S2KO, and DKO cells. ATAC-seq has two biological replicates for each DMSO- and GR-treated K562 cells (72 h). Drug-treated K562 cells RNA-seq has two biological replicates including DMSO, GSK343, X5050, GR 8 h, GR 24 h, and GR 72 h.