Primitive erythroblasts (EryP) are the first cell type specified from mesodermal progenitors in the mammalian embryo. They are found in the mouse yolk sac from embryonic day (E) ∼E7.5–8.5 and, as circulation initiates, they begin to differentiate to erythroblasts that enter the bloodstream and continue to mature in a stepwise, synchronous fashion until their enucleation several days later. We have purified these first hematopoietic-committed progenitors from staged embryos based on the expression of a nuclear GFP transgene that is expressed specifically within the EryP lineage as early as E7.5. Genome-wide expression profiling allowed us to define the transcriptome from each stage of development and revealed highly dynamic changes during the progression from progenitor to maturing erythroblast. We focused on the emergence of EryP progenitors in the yolk sac and on the transition to circulation stage, when progenitor activity is lost and a peak is observed in the number of genes whose expression changes. TRANSFAC analysis of promoters of differentially expressed genes allowed us to identify candidate transcriptional regulators, some of which have not previously been implicated in erythroid development (e.g. Nkx3.1, known previously as a regulator of prostate stem cells). We designed experiments to test predictions from our microarray analysis and found that EryP progenitor numbers are regulated by TGF-beta1 and hypoxia. In most mammalian cells, the response to hypoxia is mediated by the transcription factor HIF-1. Hif-1 is apparently not expressed in EryP. Howver, Hif3a/Ipas, a Hif-1 target gene that encodes a dominant negative regulator of HIFs and that is thought to function as a feedback regulator in response to hypoxia, is expressed in EryP as early as E7.5 and is upregulated as the cells mature. These findings suggest that the response to hypoxia by EryP may involve a pathway that is distinct from that of most other cells. EryP progenitors express genes associated with aerobic glucose metabolism (the Warburg effect), a phenotype characteristic of cancer and other rapidly proliferating cells. Whether this glycolytic profile reflects the energy needs of these cells or a more unique feature of primitive erythropoiesis is under investigation. Currently we are using computational methods to identify transcription factor (ChEA, ChIP Enrichment Analysis) and kinase (KEA, Kinase Enrichment Analysis) networks that may play a role in the regulation of primitive erythroid development. This study is the first lineage specific transcription profiling of a differentiating cell type in the early mouse embryo and will provide a strong basis for future work on normal erythropoiesis throughout ontogeny. It may also help guide efforts to direct the differentiation of stem/progenitor and cells of other lineages to an erythroid cell fate.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.