Collaboration Between RUNX and NOTCH Pathways in T-Cell Acute Lymphoblastic Leukemia

Abstract 1279

T-cell acute lymphoblastic leukemia (T-ALL) is an aggressive malignancy characterized by the clonal outgrowth of developmentally arrested T-lymphoid blasts. Notch signaling is activated by mutation of NOTCH1 and/or FBW7 in over half of cases, and ultimately results in increased expression of target genes via the NOTCH/CSL transcriptional complex. Enforced expression of activated NOTCH1 in mouse hematopoietic progenitors leads to the development of clonal T-cell leukemias, suggesting that collaborating mutations are required for establishment and/or propagation of malignant clones. To identify candidate collaborating loci, Beverly and Capobianco performed a retroviral insertional mutagenesis screen in mice expressing a relatively weak activated Notch1 transgene and found recurrent insertions into Ikaros (Ikzf1). These insertions resulted in expression of dominant negative isoforms of Ikaros and thus potentiated Notch signaling since Ikaros and Notch/CSL compete for occupancy at target gene regulatory elements.

In an attempt to identify collaborating mutations outside of the Notch pathway, we performed a similar screen, but employed instead a very potent activated NOTCH1 allele (ΔE) in hopes of saturating the Notch signaling pathway. We thus cloned out the insertion sites from 88 primary mouse leukemias generated by transduction of bone marrow with ΔE retrovirus. While recurrent insertions into Ikzf1 were again identified, we also observed frequent insertions into other regions including the Runx3 locus. The Runx3 integrations were tightly clustered in a region 40–60kb upstream of the transcriptional start site, suggesting the retroviral LTR might be inducing an increase in Runx3 expression. A single integration upstream of Runx1 was also identified in a region frequently mutated in similar screens. Of note, analysis of publically available gene expression profile data revealed that RUNX1 and RUNX3 are ubiquitously expressed in patient T-ALL samples.

In order to functionally characterize the roles of RUNX1 and RUNX3 in T-ALL, we utilized lentiviral shRNAs to knock down RUNX1 and/or RUNX3 across a broad panel of 26 human T-ALL cell lines. Despite recent studies suggesting RUNX1 may act as a tumor suppressor in T-ALL, we observed the overwhelming majority of cell lines to show substantial growth defects after knock-down of RUNX1/3 as measured by competitive growth assay. These results were confirmed in a subset of cell lines and also in xenograft-expanded primary T-ALL samples by BrdU incorporation/DNA content assays which showed reduced proliferation/G1 cell cycle arrest following RUNX1/3 knock-down. Conversely, overexpression of RUNX3 induced T-ALL cells to proliferate more rapidly and to resist ABT-263-induced apoptosis.

To explore potential target genes responsible for these pro-growth/survival effects, we mined available ChIP-Seq data and found NOTCH1/CSL and RUNX1 binding sites to co-localize within IGF1R and IL7R loci at intronic enhancer-like regions with associated H3K4me1>H3K4me3 marks and reduced H3K27me3 marks. Importantly, we show that NOTCH1 and RUNX factors co-regulate surface protein expression of IGF1R and IL7R in a synergistic/additive manner. As we and others have previously demonstrated important roles for both IGF1R and IL7R in T-ALL cell growth and leukemia-initiating activity, these studies reveal a novel collaborative mechanism between NOTCH1 and RUNX proteins in supporting propagation of established T-ALL disease.