Switching sides: How MLL-rearranged leukemias leverage oncoproteins to evade treatment

We often think of blood as the uniform red liquid encountered after a kitchen accident or at the doctor’s office. While this is true, on a microscopic level blood is made up of highly heterogeneous populations of cells that give rise to our immune system, carry oxygen and nutrients throughout the body, and remove waste products. There are two major families of blood cells: the myeloid lineage and the lymphoid lineage. Proper blood composition and function requires that myeloid and lymphoid precursor cells differentiate into the blood cells that perform these life-sustaining tasks. Sometimes, these precursor cells acquire genetic mutations that prevent their differentiation and promote growth of the precursors. A buildup of these mutant precursors is known as leukemia.

Potent mutations that can cause both myeloid- and lymphoid-lineage leukemias occur at the Mixed Lineage Leukemia (MLL) locus. Normally, the MLL gene influences which genes are transcribed in healthy cells. When the gene is mutated, MLL fuses with other partner genes that bind to DNA incorrectly and turn on transcription of genes that drive leukemia instead of healthy genes. Despite falling under the same genetic umbrella, MLL-rearranged leukemias differ from patient to patient and have diverse responses to cancer treatments. These frequently unpredictable responses mean that the leukemias can become treatment resistant, posing a threat to long-term patient survival. “Even though these are being lumped together as the same genetic subtypes of leukemia, there’s still a lot of heterogeneity in that leukemia,” says Dr. Derek Janssens, lead author of the study. To add to the complexity, MLL-rearranged leukemias can switch from the lymphoid to myeloid lineage following treatment. This switch results in a highly aggressive form of myeloid leukemia, and the mechanism underlying the switch is still unclear.

To clarify how the switch happens and why it results in such aggressive cancer, Drs. Derek Janssens and Steven Henikoff in the Fred Hutch Basic Sciences Division set out to define where the mutant MLL-rearranged proteins bind in the genome. To do this, they used CUT&RUN to sequence the regions of DNA that are bound by healthy and mutant MLL proteins. “We wanted to develop this one size fits all approach so that we could profile the oncoprotein regardless of the fusion partner,” says Janssens, lead author of the study. They found that mutant MLL proteins bind at different sites in myeloid and lymphoid lineage leukemias, underscoring how the lineage context can influence the genetic targets of mutant MLL proteins in leukemia.

After demonstrating that the DNA binding patterns for MLL differ in lymphoid and myeloid leukemias, the group wanted to know how the abundance of MLL protein influenced target gene expression. They found that samples with less MLL protein had less binding and activation of target genes and samples with more MLL protein had more binding. This supports the idea that high levels of mutant MLL protein can activate transcription of target genes. Whether the MLL fusion partner could influence whether leukemia comes from a myeloid or lymphoid lineage was also unclear. When they grouped MLL-rearranged leukemia samples according to their fusion partners, they found that fusion partners that are more frequent in myeloid or lymphoid leukemia activate their lineage-specific genes that drive leukemia. “The model we started coming to with this is that certain mutations are not more likely to occur in one lineage or another, but the lineage provides a context where the mutations can be very potent or not,” says Janssens.

Equipped with new understanding of how mutant MLL dosage, fusion partner, and lineage context influence which pro-leukemic genes are activated, the group next asked whether the target genes change when leukemia undergoes a lineage-switching event. They identified patients with lymphoid leukemia that switched to myeloid leukemia after initial treatments. They found that there was significantly less mutant MLL protein in the myeloid leukemia patient samples. When they interrogated which genes were activated in both leukemias, they found that the MLL protein switched from activating lymphoid-specific genes to activating myeloid-specific genes. The group identified new mechanisms for a well-known leukemia mutation across two very different lineages of leukemia. This revelatory epigenetic mechanism for mutant MLL proteins could potentially explain why MLL-rearranged leukemias continue to stump researchers with their unpredictable response to therapies. In the future, Janssens hopes this work will inform better treatments to prevent lineage switching events in MLL-rearranged leukemias to improve patient outcomes.

This work was supported by funding from the Howard Hughes Medical Institute, the Hartwell Foundation, the National Institutes of Health, and the American Lebanese Syrian Associated Charities of St. Jude Children’s Research Hospital.

Fred Hutch/University of Washington/Seattle Children's Cancer Consortium Members Drs. Soheil Meshinchi, Jay Sarthy, and Steven Henikoff contributed to this work.

Janssens DH, Duran M, Otto DJ, Wu W, Xu Y, Kirkey D, Mullighan CG, Meshinchi S, Sarthy JF, Ahmad K, Henikoff S. 2024. MLL oncoprotein levels influence leukemia lineage identities. Nat Commun. (15) 9341.