The maturation of a committed erythroid progenitor to a functional erythrocyte is characterized by a global decline in transcription and progressive nuclear condensation that ultimately culminates in enucleation. At the molecular level, erythroid maturation is driven by erythroid transcription factors and chromatin modifying enzymes that work in a coordinated manner to drive the expression of erythroid-specific genes, while silencing most other genes during the process of chromatin condensation and enucleation. Setd8 is the sole histone methyltransferase in mammals capable of mono-methylating histone H4 Lysine 20 (H4K20me1). In vitro studies suggest that Setd8 and H4K20me1 play critical roles in cell cycle progression, nuclear condensation, and DNA damage response (Reviewed in Beck et al, Genes and Development, 2012). In vivo studies on the function of Setd8 and H4K20me1 have been limited by the early embryonic lethality of constitutive Setd8 deletion (Oda et al, MCB, 2009). Although Setd8 is broadly expressed in tissues, there is a striking increase in Setd8 expression in CD71+ erythroid cells (Wu et al, Genome Biology, 2009), suggesting that Setd8 may have erythroid specific functions. Initial studies from our lab and others suggest that Setd8 regulates erythroid maturation and represses Gata2 expression (Malik et al MCB 2015; DeVilbiss et al MCB 2015).

To delineate the function of Setd8 in vivo, we generated an erythroid-specific Setd8 deletion by crossing mice with flox sites flanking exon 7 of Setd8 (Setd8 fl/fl; Oda et al, MCB, 2009) with mice expressing a Cre-Recombinase GFP fusion protein under the control of the endogenous erythropoietin receptor promoter (ErGFPCre; Heinrich et al, Blood, 2004). Setd8Δ/Δ;ErGFPCre embryos demonstrated visible anemia starting at E9.5, with death occurring at E12.5 due to severe anemia. The early onset of anemia is consistent with a defect in primitive erythropoiesis. Cytospins and imaging flow cytometric analyses demonstrated a block in primitive erythroblast maturation and a profound impairment in nuclear condensation, with a nuclear area of 96 um2 in the Setd8 null erythroblasts and 56um2 in the littermate control erythroblasts at E10.5. Setd8 null erythroblasts also had abnormalities in cell cycle progression as well as increased staining for γH2AX, suggesting accumulation of DNA damage. There were similar numbers of Erythro-myeloid progenitors, (EMP; defined as defined as Ter119-negative, kit-hi,CD41-hi cells) in both the Setd8Δ/Δ;ErGFPCre and littermate control embryos, however the Setd8 null progenitors failed to differentiate into definitive erythroblasts.

The role of Setd8 in transcriptional regulation is controversial, with some studies suggesting it is an activator and others suggesting it is a repressor. Global transcriptome analyses on sorted populations of Setd8 null and littermate control erythroblasts suggest that Setd8 acts primarily as a repressor in erythroid cells, with the majority of transcripts upregulated following Setd8 deletion. RNA-seq analyses further demonstrated a profound increase in the expression of cell cycle checkpoint enzymes such as P21 (fold change 73, p< e-53) and a significant increase in Gata2 expression (fold change 16, P<10-15) in the Setd8 null erythroblasts. The top networks identified by Ingenuity Pathway Analyses were Cell Death and Survival (Score 41, Focus Molecules 31) and Cell Cycle and Embryonic Development (Score 26; Focus Molecules 21). ChIP-seq for Setd8 in primary human erythroblasts demonstrated that sites of Setd8 occupancy were distributed throughout the erythroid genome, including over key regulatory regions of the Gata2 locus. Further supporting a role for Setd8 in transcriptional regulation, sites of Setd8 occupancy frequently co-localized with sites of Gata1, Klf1, P300, and H3K4me2 occupancy. Taken together, our results demonstrate that Setd8 is essential for mammalian erythropoiesis and suggest that Setd8 deletion results in defective proliferation and maturation of erythroid precursors that culminates in cell cycle arrest and apoptosis.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution