Ftz-F1 is a cofactor in Ftz activation of the Drosophila engrailed gene

Transcription factor CP2 is essential for lens-specific expression of the chicken alphaA-crystallin gene

BACKGROUND

Lens-specific transcriptional activation of the chicken alphaA-crystallin gene is controlled by the distal and proximal enhancers, alphaCE1 and alphaCE2, respectively. Analysis using specific monoclonal antibodies against purified alphaCE1-binding factor alphaCEF1 revealed that alphaCEF1 is composed of two distinct subunits.

RESULTS

We have demonstrated that one of the subunits of alphaCEF1 is encoded by chicken ubiquitous transcription factor CP2 (cCP2), which is homologous to mouse CP2, and human CP2/LBP-1/LSF-1. Electrophoretic mobility shift assays and cross-linking experiments showed that alphaCEF1 and bacterially expressed cCP2 form a tetramer. Overexpression of cCP2 activates transcription through alphaCE1, but a mutant cCP2 lacking the DNA-binding domain reduced the transcription to basal levels. Although cCP2 binds to the CP2 template from the mouse alpha-globin promoter, it fails to promote transcription through this template. Element substitution experiments between alphaCE1 and the CP2 template revealed that the lens-specific enhancer activity of alphaCE1 is due to the 6 bp sequence (-139/-134; lens-specific element (LSE)) adjacent to the 3' of the cCP2 binding site within alphaCE1.

CONCLUSION

We have shown that the tetrameric transcription factor cCP2 is essential for lens-specific transcription of the chicken alphaA-crystallin gene, although it is ubiquitously expressed. We propose a model where cCP2 cooperates with a putative lens-specific factor which binds to LSE.

The cardiac tissue-restricted homeobox protein Csx/Nkx2.5 physically associates with the zinc finger protein GATA4 and cooperatively activates atrial natriuretic factor gene expression

Specification and differentiation of the cardiac muscle lineage appear to require a combinatorial network of many factors. The cardiac muscle-restricted homeobox protein Csx/Nkx2.5 (Csx) is expressed in the precardiac mesoderm as well as the embryonic and adult heart. Targeted disruption of Csx causes embryonic lethality due to abnormal heart morphogenesis. The zinc finger transcription factor GATA4 is also expressed in the heart and has been shown to be essential for heart tube formation. GATA4 is known to activate many cardiac tissue-restricted genes. In this study, we tested whether Csx and GATA4 physically associate and cooperatively activate transcription of a target gene. Coimmunoprecipitation experiments demonstrate that Csx and GATA4 associate intracellularly. Interestingly, in vitro protein-protein interaction studies indicate that helix III of the homeodomain of Csx is required to interact with GATA4 and that the carboxy-terminal zinc finger of GATA4 is necessary to associate with Csx. Both regions are known to directly contact the cognate DNA sequences. The promoter-enhancer region of the atrial natriuretic factor (ANF) contains several putative Csx binding sites and consensus GATA4 binding sites. Transient-transfection assays indicate that Csx can activate ANF reporter gene expression to the same extent that GATA4 does in a DNA binding site-dependent manner. Coexpression of Csx and GATA4 synergistically activates ANF reporter gene expression. Mutational analyses suggest that this synergy requires both factors to fully retain their transcriptional activities, including the cofactor binding activity. These results demonstrate the first example of homeoprotein and zinc finger protein interaction in vertebrates to cooperatively regulate target gene expression. Such synergistic interaction among tissue-restricted transcription factors may be an important mechanism to reinforce tissue-specific developmental pathways.

Steroidogenic factor-1 contains a carboxy-terminal transcriptional activation domain that interacts with steroid receptor coactivator-1

The orphan nuclear receptor, steroidogenic factor-1 (SF-1), plays an important role in the development of the adrenal gland and in sexual differentiation. SF-1 regulates the transcription of variety of genes, including several steroidogenic enzymes, Müllerian inhibiting substance, and gonadotropin genes. In this report, we sought to identify domains in SF-1 that are required for transactivation and to determine whether SF-1 interacts with a subset of known coactivators. Natural variants of the FTZ-F1 locus include embryonal long terminal repeat-binding protein (ELP)-1, ELP-2, and SF-1, which share the DNA-binding domain. Analyses of the transcriptional activity of these variants revealed that the activity of ELP-2 and SF-1 was much greater than ELP-1, which contains a distinct carboxy terminus. Further studies were performed using GAL4-SF-1 fusion proteins that were constructed by replacement of the zinc finger region and FTZ-F1 box of SF-1 with the DNA-binding domain of GAL4. Elimination of the putative AF-2 domain at the carboxy terminus of GAL4-SF-1 proteins resulted in a complete loss of transactivation. Several lines of evidence demonstrated that SF-1 interacts with steroid receptor coactivator-1 (SRC-1). Full-length SRC-1 enhanced GAL4-SF-1-mediated transactivation, whereas a dominant negative form of SRC-1, consisting of its interaction domain alone, inhibited the activity of GAL4-SF-1. In mammalian two-hybrid assays, fusion of the VP16 activation domain to the interaction domain of SRC-1 confirmed the interaction between SRC-1 and GAL4-SF-1 and demonstrated that the AF-2 domain is required for interaction with SRC-1. Furthermore, SRC-1, together with the cAMP responsive element binding protein (CBP) or a closely related factor, p300, synergistically enhanced transcriptional activity of GAL4-SF-1. We conclude that the carboxy-terminal AF-2 region of SF-1 functions as an activation domain and that SRC-1 and CBP/p300 are components of the coactivator complex with SF-1.

Armadillo and dTCF: a marriage made in the nucleus

Within the past year, Armadillo and beta-catenin's role in transducing the Wingless/Wnt signal has been substantially clarified. It is now clear that Armadillo and beta-catenin bind directly to members of the T-cell factor/lymphoid enhancer factor subfamily of HMG box DNA-binding proteins, forming bipartite transcription factors that regulate Wingless/Wnt responsive genes in both Drosophila and vertebrates. These partners not only play key roles in a variety of cell fate decisions during normal development but, when inappropriately activated, contribute to both colon cancer and melanoma.

Armadillo coactivates transcription driven by the product of the Drosophila segment polarity gene dTCF

The vertebrate transcription factors TCF (T cell factor) and LEF (lymphocyte enhancer binding factor) interact with beta-catenin and are hypothesized to mediate Wingless/Wnt signaling. We have cloned a maternally expressed Drosophila TCF family member, dTCF. dTCF binds a canonical TCF DNA motif and interacts with the beta-catenin homolog Armadillo. Previous studies have identified two regions in Armadillo required for Wingless signaling. One of these interacts with dTCF, while the other constitutes a transactivation domain. Mutations in dTCF and expression of a dominant-negative dTCF transgene cause a segment polarity phenotype and affect expression of the Wingless target genes engrailed and Ultrabithorax. Epistasis analysis positions dTCF downstream of armadillo. The Armadillo-dTCF complex mediates Wingless signaling as a bipartite transcription factor.

The nuclear receptor homologue Ftz-F1 and the homeodomain protein Ftz are mutually dependent cofactors

Nuclear hormone receptors and homeodomain proteins are two classes of transcription factor that regulate major developmental processes. Both depend on interactions with other proteins for specificity and activity. The Drosophila gene fushi tarazu (ftz), which encodes a homeodomain protein (Ftz), is required zygotically for the formation of alternate segments in the developing embryo. Here we show that the orphan nuclear receptor alphaFtz-F1 (ref. 3), which is deposited in the egg during oogenesis, is an obligatory cofactor for Ftz. The two proteins interact specifically and directly, both in vitro and in vivo, through a conserved domain in the Ftz polypeptide. This interaction suggests that other nuclear receptor/homeodomain protein interactions maybe important and common in developing organisms.