Abstract
Chromatin condensation culminating in enucleation is a hallmark of erythropoiesis, however the mechanisms driving this process are incompletely understood. Setd8 is the sole enzyme that can mono-methylate histone H4, lysine 20 (H4K20me1) and is an important regulator of cell cycle progression, higher order chromatin structure, and genome stability. (Reviewed in Beck, Genes and Development, 2010) Setd8 and H4K20me1 are unique among epigenetic regulators in that their expression is dynamically regulated during the cell cycle. Setd8 expression peaks during G2/M, where it promotes mitotic chromatin condensation, and becomes undetectable during S-phase due to ubiquitin dependent destruction. (Oda, Mol Cell, 2010) The presence of H4K20me1 mirrors that of Setd8, peaking in G2/M, and reaching a nadir during S-phase due to removal by the histone demethylase PHF8. (Liu, Nature, 2010) Interestingly, Setd8 is expressed at levels 8- to 10- fold higher in CD71+ erythroblasts than in any other cell type, (Wu Genome Biology,2009) suggesting that it has an erythroid-specific function. We hypothesize that Setd8 drives chromatin condensation in maturing erythroblasts.
In cell lines, forced accumulation of H4K20me1 during S-phase due to perturbation of either Setd8 or PHF8 results in pre-mitotic chromatin condensation (Centore Mol Cell 2010; Liu Nature 2010). We demonstrate that primary erythroblasts express Setd8 and accumulate H4K20me1 throughout the cell cycle, suggesting that Setd8 and H4K20me1 in promote chromatin condensation during terminal maturation. We further demonstrate that Setd8 is essential for erythropoiesis, with erythroid-specific Setd8 deletion resulting in profound anemia that is lethal by E12.5. The early onset of anemia indicates a defect in the primitive erythroid lineage, which emerges from the yolk sac at E8.5, and proliferates, matures, and enucleates in the circulation as a semi-synchronous cohort. (Kingsley, Blood, 2004) Detailed analyses of Setd8-null erythroblasts revealed severe defects in cell cycle progression, increased DNA content suggesting loss of genomic integrity, accumulation of DNA damage, and a modest increase in the rate of apoptosis. Global transcriptome analyses demonstrated that Setd8-null erythroblasts had activation of checkpoint genes such as CDKN1a and Gene Set Enrichment Analyses identified significant enrichment of cell cycle and p53 signaling pathways. Despite evidence of p53 activation, concomitant p53 deletion was not able to rescue the Set8-null phenotype, indicating that Setd8 has an essential role in promoting erythroid proliferation and survival that is independent of the p53 pathway.
Consistent with our hypothesis that Setd8 drives chromatin condensation in maturing erythroblasts, the nuclear area of Setd8-null cells was nearly twice that of controls at E11.5 (119 and 69 um2, respectively p<0.0003). Transmission electron microscopy confirmed a profound defect in global chromatin condensation in the Setd8-null cells. Unexpectedly, heterochromatin was nearly absent from the Setd8-null cells, with the Setd8-null cells containing only a small amount of heterochromatin localized to the nuclear periphery. To determine the impact of Setd8 deletion on local chromatin structure, we performed ATAC-seq (Assay for Transposase Accessible Chromatin) on sorted populations of Setd8-null and control erythroblasts. Preliminary analyses of ATAC-seq data identified 364 ATAC peaks present in the Setd8-null cells but not in controls (p<0.001). Intriguingly, Gene Ontogeny analyses of the genes nearest to those regions was significant for multiple terms associated with higher order chromatin structure including "regulation of chromatin organization" and "positive regulation of histone deacetylation." Taken together, our results indicate that erythroblasts have adapted an essential cell cycle regulator to drive chromatin condensation during terminal maturation.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.