JOURNAL OF CELLULAR PHYSIOLOGY 205:1–9 (2005) REVIEW ARTICLES Developmental Control via GATA Factor Interplay at Chromatin Domains EMERY H. BRESNICK,* MELISSA L. MARTOWICZ, SAUMEN PAL, AND KIRBY D. JOHNSON Department of Pharmacology, University of Wisconsin Medical School, Molecular and Cellular Pharmacology Program, Madison, Wisconsin Despite the extraordinary task of packaging mammalian DNA within the constraints of a cell nucleus, individual genes assemble into cell type-specific chromatin structures with high fidelity. This chromatin architecture is a crucial determinant of gene expression signatures that distinguish specific cell types. Whereas extensive progress has been made on defining biochemical and molecular mechanisms of chromatin modification and remodeling, many questions remain unanswered about how cell type- specific chromatin domains assemble and are regulated. This mini-review will discuss emerging studies on how interplay among members of the GATA family of transcription factors establishes and regulates chromatin domains. Dissecting mechanisms underlying the function of hematopoietic GATA factors has revealed fundamental insights into the control of blood cell development from hematopoietic stem cells and the etiology of pathological states in which hematopoiesis is perturbed. J. Cell. Physiol. 205: 1 – 9, 2005. ß 2005 Wiley-Liss, Inc. GATA FACTOR STRUCTURE/FUNCTION The identification of the first mammalian GATA transcription factor, GATA-1, as a regulator of b-globin transcription (Evans et al., 1988; Evans and Felsenfeld, 1989; Tsai et al., 1989) set the stage for the discovery of five homologous factors (GATA-2–6) (Ho et al., 1991; Joulin et al., 1991; Lee et al., 1991; Zon et al., 1991b; Dorfman et al., 1992; Arceci et al., 1993; Kelley et al., 1993; Molkentin, 2000), which constitute an important transcription factor family. Besides regulating tran- scription of b-globin and other erythroid cell-specific genes, GATA-1 is required for differentiation of ery- throid cells (Pevny et al., 1991; Simon et al., 1992; Weiss et al., 1994; Takahashi et al., 1997; Weiss et al., 1997). The dual role of regulating differentiation as well as transcription in differentiated cells is a reoccurring theme with GATA factors. GATA factors elicit biological activities through both the activation and repression of target genes. Essential structural features of GATA factors required for transcriptional regulation include two highly conserved Cys4 zinc fingers (Fig. 1A). The C-terminal finger of GATA-1 mediates sequence-specific DNA binding to A/ TGATAA/G (WGATAR) motifs (Ko and Engel, 1993; Merika and Orkin, 1993). By contrast, the N-terminal finger interacts with Friend of GATA (FOG) coregula- tors (Tsang et al., 1997; Svensson et al., 1999; Tevosian et al., 1999), stabilizes GATA factor binding on certain DNA motifs (Trainor et al., 1996), and binds GATC motifs (Pedone et al., 1997; Newton et al., 2001). The N- and C-terminal regions of GATA factors are far less conserved than the DNA binding domains (Fig. 1B). The N-terminus of GATA-1 has been implicated in mediating transactivation in transfection assays with reporter genes (Visvader et al., 1995). Furthermore, initiating mutations in human megakaryoblastic leukemia in Down syndrome patients result in expression of N- terminally truncated GATA-1 (Wechsler et al., 2002; Mundschau et al., 2003). These findings suggest an important functional role for the GATA-1 N-terminus, but further mechanistic analysis is required to test this possibility. To limit the scope of this minireview, we shall focus on the hematopoietic GATA factors (GATA-1 – 3) (Weiss and Orkin, 1995). Transcriptional regulation (both acti- vation and repression) by these factors often requires FOG-1, but GATA-1 and GATA-2 can also activate transcription in a FOG-1-independent manner (Cris- pino et al., 1999). Individual amino acids within the N- terminal finger of GATA-1 (for example, V205) are important for conferring high-affinity FOG-1 binding (Crispino et al., 1999; Nichols et al., 2000; Liew et al., 2005). FOG-1 sequences mediating GATA-1 binding appear to be more complex, since FOG-1 has nine zinc fingers, and mutational disruption of four fingers is required to abolish GATA-1 binding (Fox et al., 1999; Cantor et al., 2002). Despite the multiple zinc fingers, DNA binding activity of FOG-1 has not been demon- strated. Thus, it is unclear whether the zinc fingers are exclusively protein–protein interaction modules or if they impart additional structural and/or functional properties. Studies of GATA factor and FOG-1 interactions with chromatin have revealed that FOG-1 colocalizes with GATA-2 at chromatin sites (Pal et al., 2004a,b). Intriguingly, FOG-1 facilitates chromatin occupancy by GATA-1 at certain chromosomal sites (Letting et al., 2004; Pal et al., 2004a) and is necessary for GATA switches in which GATA-1 displaces GATA-2 from such sites (Pal et al., 2004a). This novel coregulator activity to facilitate GATA switches has been deemed ‘‘chromatin occupancy facilitator’’ (COF) activity (Pal et al., 2004a). Besides mediating GATA switches, FOG-1 has a con- served N-terminal region consisting of twelve amino acids that confer transcriptional repression in transfec- tion assays (Svensson et al., 1999; Lin et al., 2004). In ß 2005 WILEY-LISS, INC. Contract grant sponsor: NIH (to E.H.B.); Contract grant numbers: DK55700, DK50107; Contract grant sponsor: NIH (to K.D.J.); Contract grant number: NRSA T32 NL07936; Contract grant sponsor: American Heart Association. *Correspondence to: Emery H. Bresnick, University of Wisconsin Medical School, Department of Pharmacology, Molecular and Cellular Pharmacology Program, 383 Medical Sciences Center, 1300 University Avenue, Madison, WI 53706. E-mail: ehbresni@wisc.edu Received 16 January 2005; Accepted 21 January 2005 DOI: 10.1002/jcp.20393