Scl/Tal1 Directly Activates Hematopoiesis and Represses Cardiogenesis During Mesodermal Diversification

Abstract 3446

Understanding the mechanisms of mesoderm specification into the different lineages during embryogenesis holds a great potential to advance the development of cell-based regenerative therapies for cardiovascular and blood disorders. The divergence of the developmental fates is dictated by transcription factors that induce lineage-specific gene expression programs. The basic helix-loop-helix transcription factor Scl is known as the master regulator for the specification of the hematopoietic fate. We recently discovered that, in addition to positive effects of Scl in promoting the establishment of hemogenic endothelium and hematopoietic stem/progenitor cells development, it is also required to repress cardiogenesis in hematopoietic tissues during developmentally defined window (Van Handel, Montel-Hagen, et al, Cell, 2012). However, how Scl regulates hematopoiesis and cardiogenesis remains unknown.

To identify Scl's direct target genes during mesoderm diversification, we determined the genome-wide Scl binding sites in Flk+ mesoderm from embryoid bodies using ChIP-sequencing. This analysis identified ∼4600 Scl binding sites throughout the genome, with predominance in inter-genic regions. Comparison with previously published Scl ChIP-seq datasets during later stages of development (HPC7 hematopoietic progenitor cell-line, Wilson et al. 2010, and red blood cells from fetal liver, Kassouf et al. 2010) revealed that the majority of the binding sites are developmental stage specific. Using nearest gene approach to intersect ChIP-seq data with gene expression data showed that the regulating regions of about 35% of Scl activated and 20% of repressed genes in Flk+ mesoderm were bound by Scl. Similar to later stages of hematopoietic development, robust binding of Scl to key hematopoietic transcription factors downstream of Scl, such as Runx1, Gata1, Gata2, Lyl1, Eto2, Erg, Fli1, Hhex, Gfi1, Gfi1b and Myb was observed during mesoderm specification. Interestingly, genomic regions enrichment analysis of Scl binding sites unique to Flk+ mesoderm showed enrichment for genes implicated in mesoderm formation and heart development, such as Gata4, Gata6, Msx1, Myocd, Nkx2–5 and Tbx5 indicating Scl functions as a direct repressor for cardiogenic transcriptional program.

We then went on to investigate the mechanism of how Scl distinguishes between cardiac and hematopoietic genes to repress or activate them. Previous studies have shown that Scl forms complex with Gata1/2 transcription factors to activate red cell transcriptional program in erythroid cells. To clarify whether Gata1/2 are required in Scl binding and functional distinction between activation and repression during mesoderm specification, we performed Scl ChIP sequencing on Flk+ mesoderm from embryoid bodies induced from Gata1/2 double KO ES cells. Unexpectedly, Scl still bound to most hematopoietic as well as cardiac sites. However, some specific binding sites around key hematopoietic genes were completely lost or significantly reduced, such as ∼+300kb Runx1, ∼+30kb Myb and ∼TSS of Pu.1, and the expression of these genes was also down regulated in agreement with the loss of CD41+ hematopoietic progenitors in Day4.5 embryonic bodies. This suggests that Gata1/2 are required for Scl binding to the regulatory regions of a subset of crucial hematopoietic genes.

These studies show that Scl has a direct critical function both as an activator of hematopoietic fate and a repressor of cardiac fate during mesoderm diversification, and only a fraction of Scl binding sites are Gata1 and/or Gata2 dependent.