Combinatorial regulation of tissue specification by GATA and FOG factors

Abstract

The development of complex organisms requires the formation of diverse cell types from common stem and progenitor cells. GATA family transcriptional regulators and their dedicated co-factors, termed Friend of GATA (FOG) proteins, control cell fate and differentiation in multiple tissue types from Drosophila to man. FOGs can both facilitate and antagonize GATA factor transcriptional regulation depending on the factor, cell, and even the specific gene target. In this review, we highlight recent studies that have elucidated mechanisms by which FOGs regulate GATA factor function and discuss how these factors use these diverse modes of gene regulation to control cell lineage specification throughout metazoans.

Fig. 1.

Conservation of FOG proteins throughout evolution. (A) The six mammalian GATA factors, which are well conserved throughout vertebrates. Each GATA factor contains two highly conserved zinc finger domains (N and C, red), a conserved nuclear localization signal (NLS) that lies within a basic region C-terminal to the C-finger (black), and a poorly conserved N-terminal activation domain (AD). The shading of the N-terminal activation domains indicates the divergence in sequence of this domain. (B) FOG family members in mouse, frog, zebrafish and fruit fly. Zinc finger domains are represented by colored vertical bars: red, C2HC; blue, C2H2. The black bars indicate the presence of a conserved co-repressor interaction motif (NuRD or CtBP). The sizes and distributions of the zinc fingers are approximations. The question mark in zebrafish Fog1 indicates the presence of a predicted zinc finger domain that is not conserved in mammalian FOG proteins.

Fig. 2.

Control of mammalian hematopoiesis by GATA and FOG factors. A classical representation of mammalian hematopoiesis is depicted, and the roles of GATA and FOG proteins at each step are indicated. HSC, hematopoietic stem cell; CMP, common myeloid progenitor; MEP, megakaryocyte-erythrocyte progenitor; GMP, granulocyte/macrophage progenitor; CLP, common lymphoid progenitor.

Fig. 3.

FOG1 controls GATA1 transcriptional activity through diverse mechanisms. (A) FOG1 is required for activation of a subset of GATA1 target genes, including (i) β-globin and (ii) Band3. At the β-globin gene, GATA1 and FOG1 occupy the locus control region (LCR), recruit the SCL and NuRD complexes, and participate in a chromatin loop that brings the LCR into close proximity with the β-globin promoter. At the Band3 gene, FOG1 facilitates GATA1 chromatin occupancy to induce gene activation. (B) FOG1 is required for repression of a subset of GATA1 target genes, including (i) Gata2 and (ii) Kit. At Gata2, GATA1 and FOG1 eject GATA2 from cis-regulatory elements (GATA switch), and FOG1 recruits the NuRD repressor to induce locus-wide histone deacetylation. At Kit, FOG1 is required for GATA1 to alter chromatin loops involving distal enhancer elements. (C) A subset of GATA1-activated genes, including Fog1 itself, do not require FOG1. (D) A subset of GATA1-repressed genes, including Lyl1, do not require FOG1. (E) FOG1 inhibits GATA1 activation of some target genes. In some cases, FOG1 accomplishes this regulation by prohibiting GATA1 chromatin occupancy. Several mast cell-specific genes, including Fcer1b (Ms4a2), are regulated by this mechanism.