Mixed lineage leukemia (MLL) is a very aggressive therapy resistant leukemia. MLL-AF9 is one of the most common MLL-fusion proteins that recruits disruptor of telomeric silencing 1-like (DOT1L), to MLL-target genes HOXA9/MEIS1. DOT1L is a histone methyltransferase that catalyzes methylation of H3K79, a gene activating mark, that is essential for leukemogenesis and normal hematopoiesis. Our lab and others mapped the 10 amino acid (865 - 874) binding site on DOT1L, which binds to the ANC1 homology domain (AHD) domain of AF9. We also identified a point mutation, I867A, sufficient to disrupt the AF9-DOT1L interaction in vitro. We demonstrated that DOT1L lacking the 10 amino acids (Δ10) were unable to support transformation by ML-AF9. We hypothesize that by disrupting the AF9-DOT1L protein-protein interaction (PPI), we will be able to inhibit leukemogenesis and potentially circumvent the negative consequences of DOT1L loss in normal bone marrow. We used a genetic approach to explore the role of DOT1L recruitment in leukemogenesis and normal hematopoiesis.

We generated cell lines derived from MLL-AF9-driven DOT1Lf/f CreER(T2)+ bone marrow by transducing DOT1L constructs, including wild type (WT-DOT1L), deletion of 10 amino acids (Δ10-DOT1L), point mutant (I867A-DOT1L), and an enzymatically inactive mutant (RCR-DOT1L). Both Δ10 and I867A mutants inhibited leukemogenesis similarly to the RCR mutant. This inhibited proliferation was coupled with a decrease in H3K79me2, a decrease in HOXA9/Meis1 gene expression, cellular differentiation, and induction of apoptosis demonstrating that disruption of the AF9-DOT1L PPI mimics the phenotype of as the enzymatic inhibition, deeming it as a potential therapeutic target for this cancer.

To explore the impact of disrupting the AF9-DOT1L interaction in normal bone marrow, we transduced Mx-Cre+ DOT1Lf/f bone marrow with either WT-DOT1L, Δ10-DOT1L, or RCR-DOT1L constructs and injected these cells into lethally irradiated B6-CD45.1 recipients. After hematopoietic reconstitution, recipients were treated with poly(I:C) to inactivate endogenous DOT1L. In the absence of DOT1L's methyltransferase activity (RCR-DOT1L), transduced hematopoietic progenitors underwent rapid failure. Interestingly, expression of the Δ10-DOT1L mutant was sufficient to support short-term hematopoiesis but insufficient to sustain long-term hematopoietic stem cells (LT-HSCs). Together these results show that blocking DOT1L's recruitment eliminates the transformative potential of the MLL-AF9 oncogene and that while DOT1L's function is critical for all hematopoietic progenitor cell survival, removing the DOT1L AF9 binding domain allows for short term hematopoiesis.

To further explore the role of AF9 role in recruiting DOT1L in MLL-AF9 leukemia and normal hematopoiesis, using high throughput screening approach, we identified small molecules that bind to the AF9 AHD domain. These compounds inhibit proliferation of cells harboring the MLL-AF9 fusion, decrease H3K79me2, and induce apoptosis similar to Δ10-DOT1L and I867A-DOT1L mutants. We are currently working to further characterize their on-target activity in cells, and test their effects on normal bone marrow, with the ultimate goal to validate additional avenues for therapeutic intervention for MLL-rearranged leukemia.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.