Megakaryocytes (MK) generate platelets. Recently, we and others, have reported MK also regulate hematopoietic stem cells (HSC). Here we show high ploidy large cytoplasmic megakaryocytes (LCM) are critical negative regulators of HSC and critical for platelet formation. Using a mouse knockout model (Pf4-Srsf3Δ/Δ) with normal MK numbers, but essentially devoid of LCM, we demonstrate a pronounced increase in BM HSC concurrent with endogenous mobilization and extramedullary hematopoiesis. Severe thrombocytopenia is observed in animals with diminished LCM, although there is no change in MK ploidy distribution, uncoupling endoreduplication and platelet production. When HSC isolated from a microenvironment essentially devoid of LCM reconstitute hematopoiesis in lethally irradiated mice, the absence of LCM increases HSC in BM, blood and spleen, and the recapitulation of thrombocytopenia. In contrast, following a competitive transplant using minimal numbers of WT HSC together with HSC from a microenvironment with diminished LCM, sufficient WT HSC-generated LCM regulates a normal HSC pool and prevents thrombocytopenia. Importantly, LCM are conserved in humans.

The ZEB2 transcription factor has been demonstrated to play important roles in hematopoiesis and leukemic transformation. ZEB1 is a close family member of ZEB2 but has remained more enigmatic concerning its roles in hematopoiesis. Here, we show using conditional loss-of-function approaches and bone marrow (BM) reconstitution experiments that ZEB1 plays a cell-autonomous role in hematopoietic lineage differentiation, particularly as a positive regulator of monocyte development in addition to its previously reported important role in T-cell differentiation. Analysis of existing single-cell (sc) RNA sequencing (RNA-seq) data of early hematopoiesis has revealed distinctive expression differences between Zeb1 and Zeb2 in hematopoietic stem and progenitor cell (HSPC) differentiation, with Zeb2 being more highly and broadly expressed than Zeb1 except at a key transition point (short-term HSC [ST-HSC]➔MPP1), whereby Zeb1 appears to be the dominantly expressed family member. Inducible genetic inactivation of both Zeb1 and Zeb2 using a tamoxifen-inducible Cre-mediated approach leads to acute BM failure at this transition point with increased long-term and short-term hematopoietic stem cell numbers and an accompanying decrease in all hematopoietic lineage differentiation. Bioinformatics analysis of RNA-seq data has revealed that ZEB2 acts predominantly as a transcriptional repressor involved in restraining mature hematopoietic lineage gene expression programs from being expressed too early in HSPCs. ZEB1 appears to fine-tune this repressive role during hematopoiesis to ensure hematopoietic lineage fidelity. Analysis of Rosa26 locus-based transgenic models has revealed that Zeb1 as well as Zeb2 cDNA-based overexpression within the hematopoietic system can drive extramedullary hematopoiesis/splenomegaly and enhance monocyte development. Finally, inactivation of Zeb2 alone or Zeb1/2 together was found to enhance survival in secondary MLL-AF9 acute myeloid leukemia (AML) models attesting to the oncogenic role of ZEB1/2 in AML.