No abstract

We recently reported the hemochromatosis-like phenotype observed in our Usf2 knockout mice. In these mice, as in murine models of hemochromatosis and patients with hereditary hemochromatosis, iron accumulates in parenchymal cells (in particular, liver and pancreas), whereas the reticuloendothelial system is spared from this iron loading. We suggested that this phenotypic trait could be attributed to the absence, in the Usf2 knockout mice, of a secreted liver-specific peptide, hepcidin. We conjectured that the reverse situation, namely overexpression of hepcidin, might result in phenotypic traits of iron deficiency. This question was addressed by generating transgenic mice expressing hepcidin under the control of the liver-specific transthyretin promoter. We found that the majority of the transgenic mice were born with a pale skin and died within a few hours after birth. These transgenic animals had decreased body iron levels and presented severe microcytic hypochromic anemia. So far, three mosaic transgenic animals have survived. They were unequivocally identified by physical features, including reduced body size, pallor, hairless and crumpled skin. These pleiotropic effects were found to be associated with erythrocyte abnormalities, with marked anisocytosis, poikylocytosis and hypochromia, which are features characteristic of iron-deficiency anemia. These results strongly support the proposed role of hepcidin as a putative iron-regulatory hormone. The animal models devoid of hepcidin (the Usf2 knockout mice) or overexpressing the peptide (the transgenic mice presented in this paper) represent valuable tools for investigating iron homeostasis in vivo and for deciphering the molecular mechanisms of hepcidin action.

We isolated full-length cDNAs encoding the 43-kD form of human upstream stimulatory factor (USF), a cellular factor required for efficient transcription of the adenovirus major late (AdML) promoter in vitro. Sequence analysis showed USF to be a member of the c-myc-related family of DNA-binding proteins. Using proteins translated in vitro, we identified a DNA-binding domain near the carboxyl terminus, which includes both a helix-loop-helix motif and a leucine repeat. We show that USF interacts with its target DNA as a dimer. The leucine repeat is required for efficient DNA binding of the intact protein and for interactions between fulllength and truncated USF proteins. Interestingly, it is not required for DNA binding of the isolated helix-loophelix domain. The structure of different cDNA clones indicates that USF RNA is differentially spliced, and alternative exon usage may regulate the levels of functional USF protein.

USF is a helix-loop-helix transcription factor that, like Myc, recognizes the DNA binding motif CACGTG. Two different forms of USF, characterized by apparent molecular weights of 43,000 and 44,000, were originally identified in HeLa cells by biochemical analysis. Clones for the 43-kDa USF were first characterized, but only partial clones for the human 44-kDa USF (USF2, or FIP) have been reported. Here we describe a complete cDNA for the 44-kDa USF from murine cells. Analysis of this clone has revealed that the various USF family members are quite divergent in their N-terminal amino acid sequences, while a high degree of conservation characterizes their dimerization and DNA-binding domains. Interestingly, the 3' noncoding region of the 44-kDa USF cDNAs displayed an unusual degree of conservation between human and mouse. In vitro transcription/translation experiments indicated a possible role for this region in translation regulation. Alternative splicing forms of the 44-kDa USF messages exist in both mouse and human. Examination of the tissue and cell-type distribution of USF by Northern blot and gel retardation assays revealed that while expression of both the 43- and 44-kDa USF species is ubiquitous, different ratios of USF homo- and heterodimers are found in different cells.

We describe a new assay system that allows a rapid, direct, and quantitative detection of promoter-dependent in vitro transcription by RNA polymerase II. The template used is a hybrid plasmid containing the adenovirus major late promoter linked to a synthetic 400-base-pair DNA fragment that lacks cytidine residues on the transcribed strand-i.e., generates a transcript with no guanosine residues. In vitro transcriptions are carried out in the absence of GTP or, if the reactions contain GTP, in the presence of RNase Ti and the chain terminator 3'-O-methyl-GTP. Under these conditions the only RNAs that can accumulate, whether from a circular or linearized DNA template, are the 400-nucleotide RNase T1-resistant transcripts resulting from accurate initiation at the major late promoter. Thus, specific transcription can be directly monitored by conventional RNA quantitation methods. Using this fast assay, we show that three basic transcription factors, TFIIB, TFiID, and TFHE, are absolutely required, in addition to the RNA polymerase II, for specific transcription initiation from the adenovirus major late promoter. Units of activity can be defied for each of these individual components. The applicability of this kind of assay to other systems is discussed.A major breakthrough for investigations of eukaryotic transcriptional control has been the development of soluble cell-free systems that mediate accurate transcription initiation on purified genes by RNA polymerase II in vitro (1-3). Initial chromatographic fractionations revealed a requirement for several protein factors in addition to the RNA polymerase (4-7). While some of these transcription factors are likely to be common to all class II promoters, the expression of some genes requires different and/or additional factors (8-10). Previous studies performed with either crude extracts or semipurified factors (11-15) have identified several steps in the initiation process. However, a better understanding of the mechanisms involved awaits the total purification of the different transcription factors.One of the major difficulties in purifying these proteins has been the lack of a fast, quantitative assay. In vitro transcription products have to be analyzed by gel electrophoresis in order to distinguish those transcripts resulting from accurate initiation at specific promoters from all of the other RNA species (resulting, for example, from random initiation, initiation at the ends of the DNA molecules, or initiation at "TATA" box-like sequences in the vector DNA). Such a time-consuming assay also necessitates freezing and thawing at successive steps during the purification procedure, and this is often deleterious to enzymatic activities. In addition, the assay is not easily amenable to rigorous quantitation. To alleviate these problems, we recently constructed a hybrid plasmid in which the internal sequences of the adenovirus major late (ML) transcription unit were replaced by a DNA fragment lacking cytidine residues on the transcribed strand. With this template...

USF1 and USF2 are basic helix-loop-helix transcription factors implicated in the control of cellular proliferation. In HeLa cells, the USF proteins are transcriptionally active and their overexpression causes marked growth inhibition. In contrast, USF overexpression had essentially no effect on the proliferation of the Saos-2 osteosarcoma cell line. USF1 and USF2 also lacked transcriptional activity in Saos-2 cells when assayed by transient cotransfection with USF-dependent reporter genes. Yet, there was no difference in the expression, subcellular localization, or DNA-binding activity of the USF proteins in HeLa and Saos-2 cells. Furthermore, Gal4-USF1 and Gal4-USF2 fusion proteins activated transcription similarly in both cell lines. Mutational analysis and domain swapping experiments revealed that the small, highly conserved USF-specific region (USR) was responsible for the inactivity of USF in Saos-2 cells. In HeLa, the USR serves a dual function. It acts as an autonomous transcriptional activation domain at promoters containing an initiator element and also induces a conformational change that is required for USF activity at promoters lacking an initiator. Taken together, these results suggest a model in which the transcriptional activity of the USF proteins, and consequently their antiproliferative activity, is tightly controlled by interaction with a specialized coactivator that recognizes the conserved USR domain and, in contrast to USF, is not ubiquitous. The activity of USF is therefore context dependent, and evidence for USF DNA-binding activity in particular cells is insufficient to indicate USF function in transcriptional activation and growth control.

Transcription of protein-encoding genes by human RNA polymerase II requires multiple ancillary proteins (transcription factors). Interactions between these proteins and the promoter DNA of a viral class II gene (the major late transcription unit of adenovirus) were investigated by enzymatic and chemical footprinting. The experiments indicated that the assembly of functionally active RNA polymerase II-containing transcription preinitiation complexes requires a complete set of transcription factors, and that both specific protein-DNA and protein-protein interactions are involved. This allows individual steps along the transcription reaction pathway to be tested directly, thus providing a basis for understanding basic transcription initiation mechanisms as well as the regulatory processes that act on them.

USF is a family of basic helix-loop-helix transcription factors that recognizes DNA-binding sites similar to those of the Myc oncoproteins. Here, various functional domains in the mouse USF2 protein were identified and characterized. Indirect immunofluorescence studies with transiently transfected cells revealed that both the basic region and the highly conserved USF-specific region (USR) are involved in the nuclear localization of USF2. Cotransfection assays with deletion mutants containing the DNA-binding domain of either USF2 or GAL4 identified two distinct transcriptional activation domains in USF2, the USR and the exon 5-encoded region. Activity of the exon 5 activation domain was detectable in both assay systems. Within USF2, however, its potency varied with the conformation induced by the surrounding regions, especially that encoded by alternatively spliced exon 4. In contrast, the USR activated transcription only in its natural context upstream of the USF2 basic region and only with reporter constructs containing the adenovirus major late minimal promoter but not the E1b minimal promoter. However, insertion of an initiator element downstream of the TATA box rescued the activity of the USR on E1b-driven reporters. The USR therefore represents a new type of activation domain whose function depends very strongly on the core promoter context.Transcription factor USF was originally identified by its ability to bind to the adenovirus major late (ML) promoter and stimulate transcription in vitro (3,31,42). In HeLa cells, USF was shown to consist of two polypeptides with apparent molecular masses of 43 and 44 kDa (41, 43). cDNA clones encoding these two proteins, termed, respectively, USF1 and USF2, were isolated from both humans and mice (14,45,46), and other family members were subsequently cloned from sea urchins and Xenopus laevis (18,22). Analysis of these clones demonstrated that USF belongs to the Myc family of regulatory proteins characterized by a C-terminal basic-region (BR)-helix-loop-helix (HLH)-leucine zipper (zip) structure responsible for dimerization and DNA binding (16,32). The different USF family members are all extremely similar in the BR-HLHzip domain, while the N-terminal regions, which are possibly involved in transcriptional activation, are highly divergent (45,46). USF recognizes sites on the DNA that contain a CACGTG core sequence. Cocrystallization of the C-terminal DNA-binding domain of human USF1 with its specific DNA-binding site revealed that USF dimers bind DNA as a four-helix bundle, with the basic domain from each monomer contacting half of the DNA-binding site (10). Transcriptional activation by USF can be demonstrated both in vitro and in vivo (9,21,25,35,38,42), and the possible involvement of USF in the transcriptional regulation of many different genes has been suggested. Unfortunately, since the putative USF target sequences may also be recognized by several other BR-HLH-zip proteins in vivo, it has been difficult to assess the direct involvement of USF in the regulation o...