Deciphering the Transcriptional Networks Regulated by Two Hematopoietic Master Regulatory Transcription Factors, PU.1 and GATA-1

Abstract

Lineage-specification involves the activation of lineage-specific genes and the inhibition of alternative lineages by master regulatory transcription factors (MRTFs). Previous work from our lab showed PU.1, a MRTF of myelopoiesis, directly interacts with GATA-1, the master regulator of erythropoiesis, on GATA-1 target genes and recruits a repressive complex that silences these genes, thereby inhibiting the erythroid program (Rekhtman et al., 2003; Rekhtman et al., 1999; Stopka et al., 2005). Building on this initial observation, this thesis presents genome-wide occupancy studies of PU.1 and GATA-1 in hematopoietic cells coupled with gene expression analysis that yield novel insights into the transcriptional networks regulated by these MRTFs.;By comparing our genome-wide occupancy data of GATA-1 with published ChIP-Seq and gene expression analysis for SCL and Klf1, two other essential erythroid promoting transcription factors (EryTFs), we identify a "core erythroid network". The three EryTFs co-occupy many erythroid-specific genes and positively regulate the expression of these genes. Furthermore, using our newly acquired PU.1 ChIP-Seq and PU.1-dependent gene expression datasets, we also incorporate a negative regulator of erythroid development into this core erythroid network. We find that PU.1 is a highly integrated component of this core network, which antagonizes the actions of the three EryTFs. Moreover, PU.1 represses the expression of critical GATA-1 associated co-factors. These studies reveal that PU.1 not only inhibits the core erythroid transcriptional network but it also, at least partially, antagonizes an erythroid proteomic network.;These studies also reveal that PU.1 regulates a large set of genes independently of GATA-1 in erythroid cells. Analysis of the PU.1 ChIP-Seq and gene expression data in erythroid cells showed that there is enrichment for pathways that control cell proliferation and survival. Consistent with these findings, we observe a depletion of an early erythroid progenitor population in a hypomorphic mouse mutant that reduces PU.1 levels. Furthermore, many of these functions are carried out by PU.1 in other hematopoietic cells. These studies highlight the role of MRTFs in not only promoting lineage-specific gene expression, but also their role in maintaining normal homeostasis of cells.;Interestingly, PU.1 binds to the promoters of many lymphoid/myeloid-specific genes in erythroid cells, though these genes are not highly expressed in these cells. However, by overexpressing PU.1 in erythroid progenitor cells, many lymphoid and myeloid-specific genes are robustly upregulated. To understand why higher levels of PU.1 are required to upregulate these genes, we analyzed the genome-wide binding patterns of PU.1 in primary macrophages and B-cells, which normally contain higher levels of PU.1 than erythroid cells. Strikingly, a large fraction of PU.1 bound sites in macrophages and B-cells are located distal to the proximal promoter regions of genes, whereas this is not the case in erythroid cells. By generating a genome-wide occupancy map of overexpressed PU.1 in erythroid cells, we observe that these differences are primarily due to the level of PU.1. We propose a molecular mechanism that could account for why high levels of PU.1 are required to bind to these distal sites.;Through deciphering the transcriptional networks regulated by PU.1 and GATA-1 in hematopoietic cells, the work presented in this thesis validates, extends, and provides novel mechanisms that underlie lineage-specification.