A therapeutically crucial sub-set of leukemia cells are those that self-renew (leukemia stem-cells or LSC). LSC could be progenitors that abnormally self-renew from impairment of the terminal differentiation that usually regulates progenitor proliferation. In the present study, we demonstrate a molecular pathway that explains how Runx1 disruption (a frequent event in leukemia) can produce this phenotype, and its therapeutic implications. Hematopoietic differentiation is dependent on lineage-determining factors such as Pu.1. Pu.1 can repress or activate genes, depending on its interaction partners. One key interaction partner of Pu.1 is the non-lineage specific hematopoietic transcription factor Runx1, which is a frequent target of inherited and acquired mutations in myelodysplastic syndrome and leukemia. We demonstrate that knock-down of Runx1 in a hematopoietic model of inducible PU.1 function (PUERhi – gift of Harinder Singh), preserves the ability of Pu.1 to repress transcription of genes associated with self-renewal (Hoxb4, c-kit, Bmi-1). Instead, Runx1 deficiency specifically inhibited Pu.1 mediated upregulation of genes associated with differentiation (m-csfr, gm-csfr, F4/80). In chromatin-immunoprecipitation assays (ChIP), this failure in Pu.1 mediated gene transactivation was associated with decreased histone acetylation in the m-csfr and gm-csfr promoters which contain DNA binding motifs for Pu.1 and Runx1 in proximity. This suggests that Runx1 could regulate co-repressor/coactivator recruitment to Pu.1. Therefore, in co-immunoprecipitation and confocal assays, we examined the effects of RUNX1 on Pu.1 interactions with the co-repressors SIN3A and ETO2. RUNX1 prevented co-repressor recruitment to Pu.1. However, removal of the C-terminus transcription regulating region of RUNX1, or the replacement of this domain with ETO (modeling the leukemia fusion protein RUNX1-ETO), allowed formation of a RUNX1/Pu.1/co-repressor tri-partite complex. These findings, taken together with the histone acetylation changes noted at Pu.1/Runx1 composite DNA-binding sites upon Runx1 knock-down, suggest that RUNX1, but not RUNX1 isoforms that lack the C-terminal transcription regulating domain, prevent co-repressor recruitment to a promoter specific Pu.1/RUNX1 complex. This explains retained Pu.1 mediated transrepression, but impaired Pu.1 mediated transactivation, in the context of Runx1 deficiency or leukemia associated RUNX1 abnormalities. In murine models with Runx1 disruption, one feature is an increase in myeloid progenitors but decrease in stem-cells. Our findings show that Runx1 deficiency allows some phenotype transition from stem-cell to progenitor through intact Pu.1 repression of pro-self-renewal genes, but aberrant histone deacetylation prevents differentiation gene expression and terminal differentiation. Therefore, we examined if histone deacetylase inhibitor (HDACi) treatment could renew terminal differentiation of the Runx1 deficient cells. HDACi renewed terminal differentiation of the abnormally proliferating cells. In contrast, HDACi increased self-renewal of wild-type cells by preventing Pu.1 mediated repression of pro-self-renewal genes, the initial step in Pu.1 mediated terminal differentiation. Thus, HDACi had opposite effects on abnormally self-renewing progenitors versus normal self-renewing stem-cells. These finding could explain the accumulation of progenitors, and possibly LSC, with Runx1 disruption, and the potential for LSC-specific therapy that spares normal HSC.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author