Human NOTCH4 Is a Key Target of RUNX1 in Megakaryocytic Differentiation

Megakaryocytes (MK), which produce platelets, play important roles in blood coagulation and hemostasis. The master transcription factor RUNX1 regulates lineage-specific transcriptional targets and key signaling pathways, and is known to be essential for megakaryopoiesis. Mono-allelic RUNX1 mutations lead to familial platelet disorder (FPD), which is characterized by thrombocytopenia and abnormal platelet functions. A high percentage (~50%) of these FPD patients later develop myelodysplastic syndromes and acute myeloid leukemia. The exact mechanisms underlying deregulated megakaryopoiesis in FPD remain unclear, partially due to the lack of an adequate experimental model mimicking the human disease. For example, engineered laboratory mice and zebrafish with only one copy of the Runx1 gene do not develop bleeding disorders or leukemia. Using an in vitro hematopoietic differentiation system, we found that megakaryocytic differentiation from FPD-derived induced pluripotent stem cells (iPSCs) were defective (Connelly et al., 2014). Targeted correction of the mutated RUNX1 allele by genome editing restored the MK production and functions, validating the central role of RUNX1 in megakaryopoiesis (Connelly et al., 2014). In this new study, we pursued the hypothesis that direct target genes regulated by RUNX1 play important roles in human megakaryopoiesis.

We first performed RNA-Seq analysis on differentiated hematopoietic cells from FPD-iPSCs (harboring a mono-allelic RUNX1 mutation) and RUNX1-corrected isogenic iPSCs. Seventy-nine genes were expressed at a significantly higher level (p<0.01, FDR<0.05) while 93 genes were expressed at a significantly lower level (p<0.01, FDR<0.05) in the RUNX1-corrected cells as compared to the FPD-iPSCs. To determine whether these differentially expressed genes (DEGs) are the direct targets of RUNX1, we additionally performed genome-wide location analysis of RUNX1 by ChIP-Seq using the same hematopoietic cell population differentiated from the RUNX1-corrected isogenic iPSCs. We detected 5266 (FDR<0.05) binding sites in 4526 gene loci. Combined with the DEG data from RNA-Seq analyses, we further identified 37 up-regulated genes (such as ITGB3 and PF4) and 27 down-regulated genes with RUNX1 binding to the gene's proximity. Among the 64 differentially expressed genes with RUNX1 binding, Gene Ontology (GO) analysis revealed that only 13 genes including PF4 have been reported to be relevant to megakaryopoiesis. In order to verify the roles of these RUNX1 target genes in hematopoiesis and megakaryopoiesis, we carried out gene knockout (KO) experiments by CRISPR-Cas9 in normal human iPSCs followed by in vitro hematopoietic differentiation assays. We first focused on the "down-regulated" genes by RUNX1 binding, with the hypothesis that their KO may enhance hematopoiesis and/or megakaryopoiesis from normal iPSCs. One of such genes is NOTCH4, a member of NOTCH receptor family that plays important roles in development and cell fate determination. A previous study showed that NOTCH signaling specifies MK development from mouse hematopoietic progenitor cells (Mercher et al., 2008), while we have not seen publications on the NOTCH4 in human MK development.

Using the improved CRISPR technology, we successfully achieved KO of one copy of NOTCH4 in the wildtype iPSCs. We found that heterozygous KO of NOTCH4 increased MK (progenitor) production by 95% (p<0.05), while the production of CD34⁺ multipotent hematopoietic progenitor cells were not affected. To further verify its function, we inhibited NOTCH4 signaling with a gamma-secretase inhibitor. Notably, inhibition of NOTCH4 signal starting at day 2 of hematopoietic differentiation improved the efficiency of MK progenitor production by 50% (p<0.05) and more mature MK production by 70% (p<0.05). Taken together, we conclude that NOTCH4, a newly discovered RUNX1 target gene, negatively regulates megakaryopoiesis in a developmental-stage specific manner. Unlocking this inhibitory effect by small molecule inhibitors can promote MK production ex vivo. The described approach will enable us to discover additional novel genes that influence human hematopoiesis and megakaryopoiesis, which in turn will help to promote ex vivo generation of MKs from human iPSCs or postnatal hematopoietic stem/progenitor cells.