A Potential Epigenetic Bookmarking Function for the Hematopoietic Transcription Factor GATA-1.

Abstract 2601

Hematopoietic lineage choice decisions are stably maintained through many cell divisions. For example, erythroid precursor cells undergo several rounds of cell division during their maturation. During each mitosis, most transcription factors separate from chromatin causing transcription to cease globally. Mitosis therefore poses a challenge for transcription factors to re-associate with the appropriate target sites in chromatin of newborn cells. The epigenetic mechanisms that cement lineage stability and resist cell reprogramming during mitosis are poorly understood, although recent evidence supports the idea that “bookmarking” factors that remain associated with mitotic chromatin may play a role in this process. We therefore investigated whether the hematopoietic transcription factor GATA-1 might be retained at specific sites during mitosis. GATA-1 controls the expression of essentially all erythroid-specific genes and might therefore play a role in maintaining erythroid gene expression programs throughout the cell cycle. Surprisingly, we found that while a substantial fraction of GATA-1 dissociates from chromatin in mitosis, foci of high GATA-1 density are present within mitotic chromatin. To determine the exact locations of GATA-1 binding during mitosis, we developed a method to highly purify mitotic erythroid cells in sufficient quantities for chromatin immunoprecipitation followed by high-throughput sequencing (ChIP-Seq). These experiments revealed that a subset of sites bound by GATA-1 during interphase is occupied continuously throughout mitosis. Importantly, continuously GATA-1-occupied sites are enriched at promoters and cis-regulatory elements of genes coding for key developmental regulators of hematopoiesis (e.g., Fog1/Zfpm1, Gata2, Lyl1) but are notably absent at erythroid physiological and structural genes (e.g., Hba, Hbb, Epb4.9). To examine the importance of mitotic chromatin binding by GATA-1, we engineered a version of GATA-1 bearing a mitosis-specific degron that targets GATA-1 for degradation during mitosis but not interphase. Preliminary results show that mitotically degraded GATA-1 fails to induce differentiation when expressed in GATA-1-null erythroblasts. This suggests an important mitotic function for GATA-1. Current work focuses on delineating the mechanism by which continuous chromatin occupancy of GATA-1 throughout mitosis ensures proper erythroid differentiation. The results will be presented and discussed at the meeting.