ASH2L Regulates Genes Involved in the Survival of Leukemic T Cells through a Loci-Specific H3K4me3 Landscape and Leukemogenesis Via Direct Gene Regulation

Abstract

Acute lymphoblastic leukemia (ALL) is a deadly blood cancer and the most common hematological malignancy in pediatric patients. While treatment response has significantly improved complete remission rates in most patients, the prognosis for children with refractory and relapsed leukemia is bleak, particularly for the subtype T-cell ALL (T-ALL) that exhibits a higher risk for treatment failures and relapses. A greater understanding of T-ALL pathobiology, especially the molecular mechanisms supporting the survival of leukemia-initiating cells (LIC), a leukemic cell subset responsible for treatment failures, is a foundational step toward developing new therapeutic approaches to improve treatment response in patients with high-risk leukemia. While tri-methylation of histone 3 at lysine 4 (H3K4me3) has been shown to support a transcriptional program to preserve leukemia stem cell fate in MLL-associated leukemia, the significance and impact of the H3K4me3 epigenetic landscape in pediatric T-ALL have yet to be established. Because absent, small, or homeotic 2-like protein (ASH2L) is a core component of the mixed-lineage leukemia (MLL) family of histone methyltransferase complex involved in H3K4 methylation, we hypothesized that loss of ASH2L should disrupt the leukemic H3K4me3 landscape impairing the development and maintenance of T-ALL. Here, we show a gradual loss of leukemic T cells in the blood and overall improved survival (p-value < 0.005) in leukemic mice generated by retroviral NOTCH1L1601P-∆P transduction of hematopoietic stem/progenitor cells from mice with conditional Ash2l gene deletion (cKO) in T cells (Ash2Lfl/fl CD4-Cre+). Interestingly, haploinsufficiency and inducible gene deletion in established leukemia using the ROSA-ERCre system also showed improved survival. We found that ASH2L regulates gene expression through H3K4me3 and direct binding to DNA through a combination of RNA-seq, H3K4me3 ChIP-seq, ATAC, and ASH2L ChIP-seq of control (fl/fl) and cKO leukemic T cells (GFP+). Surprisingly, loss of ASH2L caused a gene-specific loss of the H3K4me3 landscape instead of global alterations, suggesting ASH2L regulates the expression of a subset of genes involved in T-ALL maintenance by directing the MLL complex through binding to the nucleosomal DNA. ASH2L regulated genes were further classified based upon alterations in H3K4me3 and ASH2L genome binding as either MLL-dependent or independent. Ash2l cKO NOTCH1-induced T-ALL cells displayed downregulation of gene expression of NOTCH1 and downstream targets (e.g., HES1, DTX1) associated with loss of ASH2L binding to the NOTCH1 gene. Thus, ASH2L regulates the NOTCH1 signaling independently of H3K4me3, which was validated by immunoblots. On the other hand, significant downregulation of the SLIT-ROBO Rho GTPase Activating Protein 3 (SRGAP3) in cKO T-ALL cells was associated with loss of H3K4me3 and ASH2L binding. A comparison of the gene list from mouse T-ALL with patient data sets showed that NOTCH1 and SRGAP3 are downregulated in ASH2L knockout T-ALL cells and healthy individuals compared to T-ALL patients. We are investigating how SRGAP3 regulates the proliferation and survival of leukemic T cells through cell-intrinsic and extrinsic mechanisms. Altogether, ASH2L controls the expression of genes via H3K4me3 (e.g., SRGAP3) and direct gene regulation independently of MLL activity (e.g., NOTCH1). The discovery of ASH2L contribution to T-ALL leukemogenesis will unravel potential targets for treatment.