Under normal homeostatic conditions, adult hematopoietic stem cells (HSCs) are usually quiescent. Hematopoietic stress, such as blood loss, infection, inflammation, or chemotherapy, can drive HSCs into cycle. When adult HSCs divide multiple times, they lose self-renewal capacity. Inflammatory cytokines, such as interleukin-1 (IL-1), can accelerate the loss of HSC self-renewal capacity by activating PU.1 and promoting myeloid commitment. This raises the question of whether intrinsic tumor suppressor genes can modulate sensitivity to inflammatory cytokines, and whether loss of these tumor suppressors can allow HSCs to evade commitment programs that would otherwise limit self-renewal capacity.
The KMT2C tumor suppressor is located on chromosome 7q within a region that is frequently deleted in myelodsplastic syndrome (MDS) and therapy-related acute myeloid leukemia (AML). It encodes MLL3, a histone methyltransferase that activates enhancer elements and promotes transcription. Haploid KMT2C deletion has previously been shown to activate self-renewal programs and accelerate AML formation. This raised the question of whether KMT2C/MLL3 regulates normal HSC self-renewal and whether KMT2C deletion conveys a selective advantage to HSCs in contexts that would otherwise deplete the HSC pool. By protecting HSCs from exhaustion, KMT2C deletions may indirectly facilitate 7q-deficient MDS/AML.
To understand whether and how Kmt2c regulates HSC self-renewal, we developed novel germline and conditional Kmt2c knockout mouse alleles. Mono- and bi-allelic Kmt2c deletions led to a modest increase in adult HSC numbers and a significant reduction in committed hematopoietic progenitors (HPCs). Kmt2c deletions markedly enhanced HSC self-renewal capacity, but HSC proliferation rates were not altered. To mimic conditions that lead to therapy-related AML, we deleted a single Kmt2c allele in a minority of HSCs. We then tested whether the Kmt2c-deleted HSC population expanded, relative to wild type HSCs, under native and stressed conditions. Under native conditions, the percentage of Kmt2c-heterozygous HSCs remained stable. However, after several cycles of chemotherapy, the Kmt2c mutant HSCs predominated within the marrow. In mechanistic studies, RNA-sequencing showed that Kmt2c-deficient HSCs expressed genes associated with innate immune signaling, including the receptor for interleukin-1 (IL1R), at lower levels than wild type HSCs. This suggested that Kmt2c mutations might sustain self-renewal capacity in multiply divided HSCs by dampening IL-1 driven myeloid commitment. In support of this hypothesis, Kmt2c-deficient HSCs retained multilineage potential when they were cultured with IL-1, and they failed to activate JNK and p38 upon exposure to IL-1. Altogether, our data suggest a mechanism to explain how KMT2C deletions, in the context of larger 7q deletions, may promote therapy related MDS/AML. When HSCs acquire a KMT2C deletion, they can then escape IL-1-mediated exhaustion when they are driven into cycle by chemotherapy or other stressors. In lieu of chemotherapy-induced stress, the same clones may remain relatively indolent.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.