Alteration of CTCF associated chromatin neighborhood inhibits TAL1-driven oncogenic transcription program and leukemogenesis

Aberrant activation of the TAL1 oncogene is associated with up to 60% of T-ALL patients and is involved in CTCF mediated genome organization within the TAL1 locus, suggesting the importance of the CTCF boundary in the molecular pathogenesis of T-ALL. Here, we show that deletion of CTCF binding site (CBS) or alternation of CTCF boundary orientation alters expression of the TAL1 oncogene in a cell context dependent manner. Deletion of the CTCF binding site located at -31 Kb upstream of TAL1 (-31CBS) reduces chromatin accessibility in the +51 enhancer and the TAL1 promoter I, and blocks long-range interaction between the +51 erythroid enhancer and TAL1 promoter 1b that inhibits expression of TAL1 in erythroid cells, but not in T-ALL cells. However, in the TAL1 expressed T-ALL primary patient samples or cell line, the T-ALL prone TAL1 promoter IV specifically interacts with the +19 stem cell enhancer that is located 19 kb downstream of the TAL1 promoter and required for TAL1 transcription in the hematopoietic stem cell (HSC) stage. Inversion of -31CBS orientation, but not deletion of -31CBS, alters chromatin accessibility, enhancer/promoter histone modifications, CTCF-mediated topological associated domain (TAD), and enhancer/promoter interaction in the TAL1 locus leading to inhibition of TAL1 oncogene expression and TAL1-driven T cell leukemogenesis. Thus, our data reveal that the TAL1 +19 stem cell enhancer acts not only as stem cell enhancer, but also as a leukemia specific enhancer to activate the TAL1 oncogene in T-ALL. Manipulation of CTCF defined chromatin boundary can alter TAL1 TAD and oncogenic transcription networks in leukemogenesis. Overall design: We have finished the RNA-SEQ, ATAC-seq, ChIP-seq and HiC-seq to investigate the role of TAL1 transcription and CTCF boundaries in leukemogenesis. Jurkat leukemia cells and K562 cells were used to perform the RNA-seq, ATAC-seq, 4Cseq, CHIP-seq and HiC-seq analysis.