Genome-Wide Analysis Reveals Conserved and Divergent Features of Notch1/RBPJ Binding in Human and Murine T Lymphoblastic Leukemia Cells

Abstract 5236

Activated Notch1 regulates gene expression by associating with the DNA-binding factor RBPJ and is an important oncoprotein in murine and human T cell acute lymphoblastic leukemia/lymphoma (T-ALL), yet the interplay between Notch1 and other factors that regulate the transcriptional output of T-ALL cells is poorly understood. Using ChIP-Seq and starting with Notch1-dependent human and murine T-ALL cell lines, we find that Notch1 binds preferentially to promoters, to RBPJ binding sites, and near sites for ZNF143, as well as Ets and Runx factors. By ChIP-Seq, ZNF143 binds to ∼40% of Notch1 sites, whereas Ets1 binding is observed within 100 basepairs of ∼70% of genomic Notch1 binding sites. Notch1/ZNF143 “co-sites” have high Notch1 and ZNF143 signals, frequent co-binding of RBPJ to sites embedded within ZNF143 motifs, strong promoter bias, and low mean levels of “activated” chromatin marks. RBPJ and ZNF143 binding to DNA is mutually exclusive in vitro, suggesting RBPJ/Notch1 and ZNF143 complexes exchange on these sites in T-ALL cell lines. In contrast, Ets1 binding sites flank RBPJ/Notch1 binding sites and are associated with high levels of activated chromatin marks, whereas Runx sites are predominantly intergenic.

Although Notch1 predominantly binds promoters, ∼75% of direct Notch1 target genes lack promoter binding and appear to be regulated by enhancers, which were identified near MYC, DTX1, IGF1R, IL7R and the GIMAP gene cluster. Both Ets1 and Notch1 binding to an intronic enhancer located in DTX1 were required for expression of this well characterized Notch1 target gene, suggesting that these two factors coordinately regulate DTX1 expression. Although the association of Notch1 binding with ZNF143, Ets, and Runx sites was highly conserved, binding near certain important genes showed substantial divergence. For example, in human T-ALL lines Notch1/RBPJ bind a 3' enhancer near the IL7R gene, whereas in murine T-ALL lines no binding was observed near Il7r. Similarly, in human T-ALL lines Notch1/RBPJ bound an enhancer located ∼565 kb 5' of MYC, whereas in murine T-ALL cells Notch1/RBPJ bound an enhancer located ∼1 Mb 3' of Myc. Human and murine T-ALL genomes also have many sites that bind only RBPJ. Murine RBPJ “only” sites are highly enriched for imputed sites for the corepressor REST, whereas human RPBJ “only” sites lack REST motifs and are more highly enriched for imputed CREB binding sites. Thus, there is a conserved network of cis-regulatory factors that interacts with Notch1 to regulate gene expression in T-ALL cell lines, as well as novel classes of divergent RBPJ “only” sites that also likely regulate transcription.

To extend these findings to normal and pathophysiologic tissues, ChIP-Seq was used to identify RBPJ/Notch1 binding sites in primary murine thymocytes and primary murine T-ALL associated with Notch1 gain-of-function mutations. Early findings appear to indicate that primary T-ALLs closely resemble normal DN3a thymocytes in terms of the distribution of Notch1 binding sites and associated chromatin marks. These data suggest that Notch1-driven T-ALLs epigenetically resemble the DN3a stage of T cell development, during which Notch1 signaling is high and cells are rapidly proliferating.