Many transcription factors interact synergistically with steroid receptors.

Progesterone (PRE) or glucocorticoid receptor (GRE) DNA binding sites are often found clustered with binding sites for other transcription factors. Individual protein binding sites were tested without the influence of adjacent factors by analyzing isolated combinations of several transcription factor binding sites with PREs or GREs. All show strong synergistic effects on steroid induction. The degree of synergism is inversely related to the strength of the GRE. Thus, a steroid responsive unit can be composed of several modules that, if positioned correctly, act synergistically.

Transcriptional repression in eukaryotes.

Gene transcription can be controlled by positive or negative regulatory mechanisms; a combination of both is frequently responsible for the observed expression patterns. Analysis of a number of eukaryotic systems suggests that repressors can interfere with many and perhaps all the steps required for transcriptional activation. Transcriptional repression mechanisms can usefully be divided into three major classes: inhibition of DNA binding, blocking of activation and silencing.

CTCF is a uniquely versatile transcription regulator linked to epigenetics and disease.

CTCF is an evolutionarily conserved zinc finger (ZF) phosphoprotein that binds through combinatorial use of its 11 ZFs to approximately 50 bp target sites that have remarkable sequence variation. Formation of different CTCF-DNA complexes, some of which are methylation-sensitive, results in distinct functions, including gene activation, repression, silencing and chromatin insulation. Disrupting the spectrum of target specificities by ZF mutations or by abnormal selective methylation of targets is associated with cancer. CTCF emerges, therefore, as a central player in networks linking expression domains with epigenetics and cell growth regulation.

tau4/tau c/AF-2 of the thyroid hormone receptor relieves silencing of the retinoic acid receptor silencer core independent of both tau4 activation function and full dissociation of corepressors.

Members of the thyroid hormone (TR)-retinoic acid receptor (RAR) subfamily of nuclear hormone receptors silence gene expression in the absence of hormone. Addition of cognate ligands leads to dissociation of corepressors, association of coactivators, and transcriptional activation. Here, we used the hRAR alpha silencer core, which encompasses the ligand binding domain, including receptor regions D and E of RAR alpha without the activation function called tau4/tau c/AF-2 and without the F region, to analyze the mechanisms by which transcriptional silencing is relieved. Although the RAR silencer core is able to bind ligand, it acts as a constitutive transcriptional silencer. We have fused various small activation domains to the C terminus of the silencer core and analyzed hormone-dependent changes in receptor function. We show that nine amino acids derived from the hTRbeta are sufficient to transform the RAR silencer core into a hormone-dependent activator. Lengthening the linker between the silencer core and these nine amino acids is not critical for mediating ligand-induced relief of silencing and activation. In addition, we show that a transactivation function at the C terminus is not required for relief of silencing by the hormone, but it is required for transcriptional activation. Furthermore, we created functional silencer fusions which lose their repressive function upon addition of hormone, although the corepressors SMRT and N-CoR remain attached to the receptor.

Negative protein 1, which is required for function of the chicken lysozyme gene silencer in conjunction with hormone receptors, is identical to the multivalent zinc finger repressor CTCF.

The transcriptional repressor negative protein 1 (NeP1) binds specifically to the F1 element of the chicken lysozyme gene silencer and mediates synergistic repression by v-ERBA, thyroid hormone receptor, or retinoic acid receptor. Another protein, CCCTC-binding factor (CTCF), specifically binds to 50-bp-long sequences that contain repetitive CCCTC elements in the vicinity of vertebrate c-myc genes. Previously cloned chicken, mouse, and human CTCF cDNAs encode a highly conserved 11-Zn-finger protein. Here, NeP1 was purified and DNA bases critical for NeP1-F1 interaction were determined. NeP1 is found to bind a 50-bp stretch of nucleotides without any obvious sequence similarity to known CTCF binding sequences. Despite this remarkable difference, these two proteins are identical. They have the same molecular weight, and NeP1 contains peptide sequences which are identical to sequences in CTCF. Moreover, NeP1 and CTCF specifically recognize each other's binding DNA sequence and induce identical conformational alterations in the F1 DNA. Therefore, we propose to replace the name NeP1 with CTCF. To analyze the puzzling sequence divergence in CTCF binding sites, we studied the DNA binding of 12 CTCF deletions with serially truncated Zn fingers. While fingers 4 to 11 are indispensable for CTCF binding to the human c-myc P2 promoter site A, a completely different combination of fingers, namely, 1 to 8 or 5 to 11, was sufficient to bind the lysozyme silencer site F1. Thus, CTCF is a true multivalent factor with multiple repressive functions and multiple sequence specificities.

Alien, a highly conserved protein with characteristics of a corepressor for members of the nuclear hormone receptor superfamily.

Some members of nuclear hormone receptors, such as the thyroid hormone receptor (TR), silence gene expression in the absence of the hormone. Corepressors, which bind to the receptor's silencing domain, are involved in this repression. Hormone binding leads to dissociation of corepressors and binding of coactivators, which in turn mediate gene activation. Here, we describe the characteristics of Alien, a novel corepressor. Alien interacts with TR only in the absence of hormone. Addition of thyroid hormone leads to dissociation of Alien from the receptor, as shown by the yeast two-hybrid system, glutathione S-transferase pull-down, and coimmunoprecipitation experiments. Reporter assays indicate that Alien increases receptor-mediated silencing and that it harbors an autonomous silencing function. Immune staining shows that Alien is localized in the cell nucleus. Alien is a highly conserved protein showing 90% identity between human and Drosophila. Drosophila Alien shows similar activities in that it interacts in a hormone-sensitive manner with TR and harbors an autonomous silencing function. Specific interaction of Alien is seen with Drosophila nuclear hormone receptors, such as the ecdysone receptor and Seven-up, the Drosophila homologue of COUP-TF1, but not with retinoic acid receptor, RXR/USP, DHR 3, DHR 38, DHR 78, or DHR 96. These properties, taken together, show that Alien has the characteristics of a corepressor. Thus, Alien represents a member of a novel class of corepressors specific for selected members of the nuclear hormone receptor superfamily.

Transcriptional repression by the insulator protein CTCF involves histone deacetylases.

The highly conserved zinc-finger protein, CTCF, is a candidate tumor suppressor protein that binds to highly divergent DNA sequences. CTCF has been connected to multiple functions in chromatin organization and gene regulation including chromatin insulator activity and transcriptional enhancement and silencing. Here we show that CTCF harbors several autonomous repression domains. One of these domains, the zinc-finger cluster, silences transcription in all cell types tested and binds directly to the co-repressor SIN3A. Two distinct regions of SIN3A, the PAH3 domain and the extreme C-terminal region, bind independently to this zinc-finger cluster. Analysis of nuclear extract from HeLa cells revealed that CTCF is also capable of retaining functional histone deacetylase activity. Furthermore, the ability of regions of CTCF to retain deacetylase activity correlates with the ability to bind to SIN3A and to repress gene activity. We suggest that CTCF driven repression is mediated in part by the recruitment of histone deacetylase activity by SIN3A.

NeP1. A ubiquitous transcription factor synergizes with v-ERBA in transcriptional silencing.

One of the chicken lysozyme gene silencers binds two transcription factors, v-ERBA or the thyroid hormone receptor and NeP1 (negative protein 1), a new silencer binding protein. NeP1 is neutral on a monomeric binding site, but mediates weak repression on a multimerized site and strong synergistic repression in conjunction with v-ERBA on the wild-type silencer. Depending on the presence or absence of ligand, synergistic induction or repression is seen with the thyroid hormone receptor. This synergism is not based on cooperative DNA-binding as measured in vitro. The NeP1 DNA-binding activity is dependent on zinc ions, the binding site is characterized by a footprint of approximately 50 bp. NeP1 has a molecular weight of 140 to 160 kDa and has been enriched by affinity columns.

Co-operative binding of the glucocorticoid receptor DNA binding domain is one of at least two mechanisms for synergism.

Steroid induction of responsive genes functions through the synergistic activity of steroid receptor binding sequences with adjacent binding sites either for other transcription factors or for further steroid receptors. Analysis of the human glucocorticoid receptor revealed that the DNA-binding domain of the receptor is sufficient to mediate co-operative binding to adjacent receptor binding sites. This is a novel feature of the domain in addition to its DNA-binding, trans-activating and trans-repressing properties. Chimaeric proteins containing the N- or C-terminal receptor halves fused to the GAL4 DNA-binding domain do not co-operate in DNA-binding, however they do functionally synergize. Thus, at least two mechanisms contribute to the synergism of the human glucocorticoid receptor bound to two adjacent receptor binding sites.

DNase I hypersensitive sites far upstream of the rat tryptophan oxygenase gene direct developmentally regulated transcription in livers of transgenic mice.

Expression of the gene coding for tryptophan oxygenase (TO) is switched on in rat liver about two weeks after birth. We identified two clusters of DNaseI hypersensitive (HS) sites in the TO gene upstream region; one near the promoter, the other at a distant upstream location (-8.5 kb). Hypersensitivity of upstream sites was present in adult and in 7 day old rat liver, but absent in kidney. To investigate their role in transcriptional regulation, a reporter gene controlled by both HS site regions was used to generate transgenic mice. In these animals the transgene followed the cell specific and developmental regulation of the endogenous gene: inactive after birth and active in adult liver. Transgenes containing only the promoter proximal HS site were non-functional.

Cell-specific inhibition of retinoic acid receptor-alpha silencing by the AF2/tau c activation domain can be overcome by the corepressor SMRT, but not by N-CoR.

The human retinoic acid receptor alpha (hRAR alpha) exhibits cell-specific transcriptional activity. Previously, it was shown that in the absence of hormone the wild-type receptor is a transcriptional silencer in L cells, whereas it lacks silencing function and is a weak activator in CV1 cells. Addition of hormone leads to a further increase in transactivation in CV1 cells. Thus, the retinoic acid response mediated by RAR alpha is weak in these cells. It was shown that the CV1-specific effect is due to the receptor C terminus. We show, that the failure of silencing by RAR is not due to a general lack of corepressors in CV1 cells, since the silencing domain of RAR is functionally active and exhibits active repression in these cells. Furthermore, we show that the conserved AF2/tau c activation function of RAR is responsible for the cell-specific inhibition of silencing. Thereby, the CV1 cell specificity was abolished by replacing AF2/tau c of RAR with the corresponding sequence of the thyroid hormone receptor. Thus, we find a new role of the C-terminal conserved activation function AF2/tau c in that, specifically, the RAR AF2/tau c-sequence is able to prevent silencing of RAR in a cell-specific manner. In addition, we show that the inhibitory effect of AF2/tau c in CV1 cells can be overcome by expression of the corepressor SMRT (silencing mediator of retinoic acid and thyroid hormone receptor), but not by that of N-CoR (nuclear receptor corepressor). The expression of these two corepressors, however, had no measurable effect on RAR-mediated silencing in L cells. Thus, the expression of a corepressor can lead to a dramatic increase of hormonal response in a cell-specific manner.

A thyroid hormone receptor-dependent glucocorticoid induction.

Glucocorticoid and thyroid hormones exert their effects in many body tissues by binding to their respective receptors. The search for possible cross-talking mechanisms in overlapping target cells led to the discovery of synergism between a thyroid hormone receptor-binding site and a cryptic glucocorticoid-responsive element. Glucocorticoid responsiveness could only be detected in the presence of thyroid hormone and its receptor. This synergism requires the glucocorticoid receptor (GR) DNA-binding domain and is mediated by the transactivation domains. We found that synergism also occurs when the thyroid hormone receptor is replaced by the retinoic acid receptor or the GR is replaced by the progesterone receptor. Synergism is qualitatively independent of the type of thyroid hormone receptor-binding site and promoter. In several combinations of promoter and response elements, including a retinoic acid response element, T3 induction was only seen in the presence of the cryptic glucocorticoid-responsive element, GR, and glucocorticoids.

Multiple domains of the glucocorticoid receptor involved in synergism with the CACCC box factor(s).

Steroid induction of responsive genes functions through the synergistic activity of steroid receptor-binding sequences with adjacent transcription factor-binding sites. To analyze the mechanism of synergy we tested different human glucocorticoid receptor mutants for synergistic function with another transcription factor in comparison with intrinsic trans-activation obtained with a single receptor binding site (glucocorticoid response element). Multiple domains were found to be involved in synergistic activity of the glucocorticoid receptor with the CACCC box factor. Deletions within the N-terminal receptor half affected simultaneously intrinsic trans-activation and synergism. However, deletion of the hormone-binding domain mainly impaired synergism rather than intrinsic trans-activation, clearly showing that this domain synergizes by a mechanism independent of intrinsic activation. A chimeric protein where the DNA-binding domain of the glucocorticoid receptor was replaced by that of the yeast GAL4 protein also showed functional synergism. These data suggest that some of the receptor domains outside the DNA-binding domain synergize by their intrinsic trans-activating property, but the hormone-binding domain contributes to synergism by a different mechanism.

At least three subdomains of v-erbA are involved in its silencing function.

Several members of the thyroid hormone receptor (TR) family are able to switch from a transcriptional repressor to a transcriptional activator upon binding of their ligand. The oncogene v-erbA is a variant form of the TR unable to bind hormone and thus acts as a constitutive repressor. We demonstrate, using fusion proteins between the DNA-binding domain of the yeast factor GAL4 and the silencing domains of v-erbA and TR beta, that point mutations in three different regions severely affect their repression function. Furthermore, the three regions, each as an inactive fusion protein with the GAL4 DNA-binding domain, restore silencing activity when assembled on the same promoter. These observations define at least three silencing subdomains, SSD1-SSD3, which are involved in the silencing function of v-erbA. We propose a model in which full silencing activity is brought about by the combined interaction of each silencing subdomain with corepressors and/or basal transcription factors.

Silencing subdomains of v-ErbA interact cooperatively with corepressors: involvement of helices 5/6.

Members of the thyroid hormone receptor (TR) family act on vertebrate development and homeostasis by activating or repressing transcription of specific target genes in a ligand-dependent way. Repression by TR in the absence of ligand is mediated by an active silencing mechanism. The oncogene v-ErbA is a variant form of TR unable to bind hormone and thus acts as a constitutive repressor. Functional studies and mutation analysis revealed that the TR/v-ErbA silencing domain is composed of three silencing subdomains (SSD1-3) which, although nonfunctional individually, synergize such that silencing activity is restored when they are combined in a heteromeric complex. Here we demonstrate, using protein interaction assays in vitro and in vivo, that the inactive v-ErbA point mutant L489R within helix 5/6 in SSD2 fails to interact with the two corepressors N-CoR (nuclear receptor corepressor) or SMRT (silencing mediator of retinoic acid and thyroid hormone receptor). Furthermore, mutants in SSD1 and SSD3 exhibit a reduced corepressor recruitment corresponding to their weak residual silencing activity. In mammalian two-hybrid assays, only the combination of all three silencing subdomains, SSD1-3, leads to a cooperative binding to the corepressors N-CoR or SMRT comparable to that of the full-length v-ErbA repression domain. In conclusion, full silencing activity requires corepressor interaction with all three silencing subdomains, SSD1-3. Among these, SSD2 is a new target for N-CoR and SMRT and is essential for corepressor binding and function.

The DNA-binding and tau2 transactivation domains of the rat glucocorticoid receptor constitute a nuclear matrix-targeting signal.

Using an ATP-depletion paradigm to augment glucocorticoid receptor (GR) binding to the nuclear matrix, we have identified a minimal segment of the receptor that constitutes a nuclear matrix targeting signal (NMTS). While previous studies implicated a role for the receptor's DNA-binding domain in nuclear matrix targeting, we show here that this domain of rat GR is necessary, but not sufficient, for matrix targeting. A minimal NMTS can be generated by linking the rat GR DNA-binding domain to either its tau2 transactivation domain in its natural context, or a heterologous transactivation domain derived from the Herpes simplex virus VP16 protein. The transactivation and nuclear matrix-targeting activities of tau2 are separable, as transactivation mutants were identified that either inhibited or had no apparent effect on matrix targeting of tau2. A functional interaction between the NMTS of rat GR and the RNA-binding nuclear matrix protein hnRNP U was revealed in cotransfection experiments in which hnRNP U overexpression was found to interfere with the transactivation activity of GR derivatives that possess nuclear matrix-binding capacity. We have therefore ascribed a novel function to a steroid hormone transactivation domain that could be an important component of the mechanism used by steroid hormone receptors to regulate genes in their native configuration within the nucleus.

The (gt)n(ga)m containing intron 2 of HLA-DRB alleles binds a zinc-dependent protein and forms non B-DNA structures.

We studied protein binding and structural features of perfect and imperfect composite (gt)n(ga)m blocks from different HLA-DRB1 alleles in their original genomic and artificial environments. The major retarded protein/DNA complex of the genomic (gt)n(ga)m fragments comprises a zinc-dependent protein present in nuclear extracts from different cell types. The protein binding is characterized by moderate affinities independent of the polymorphic form of the physiological microsatellite allele. The binding affinity depends on the 5' and 3' adjacent single copy parts. DNase I footprinting of genome-derived fragments revealed that the 5' adjacent sequence and the (gt)n repeat are preferentially protected on the (gt)n(ga)m strand. Comparing three alleles, a regular pattern of footprints was not detectable in the (gt)n part, indicating that the zinc-dependent protein recognizes structural rather than sequence-specific features in this region. Chemical probing resulted in a pattern characteristic for Z-DNA in the (gt)n tract of the fragments. However, EMSA experiments using the Z-DNA specific monoclonal antibody mABZ-22 did not prove the presence of Z-DNA. As demonstrated by chemical modifications of the different (ga)m targets, only one of three (gt)n(ga)m fragments formed intramolecular triplexes of the type H-y3 and H-y5. DNase I footprinting revealed only weak protection, if any, in the homopurine tract. Rather, the (tc)m strands are hypersensitive for DNase I. This is probably due to structural conversions into intramolecular *H-triplexes after binding of HIZP.

Complex protein binding to the mouse M-lysozyme gene downstream enhancer involves single-stranded DNA binding.

The mouse M-lysozyme downstream enhancer has been previously characterized on several levels of gene regulation. The enhancer was co-localized with a DNase I hypersensitive site in the chromatin of mature macrophages, the in vivo interaction of transcription factor GABP with the enhancer core (MLDE) demonstrated binding being restricted to mature macrophage cells, and analysis of the MLDE methylation state revealed a correlation between demethylation of CpG dinucleotides and the in vivo GABP binding. Here, we analyzed in detail the full-length enhancer in addition to the core element. We identified a total of nine binding sites for nuclear factors. Most of these factors are found ubiquitously in all cell types tested. These factors include several unknown proteins as well as the transcription factor NF-Y. In addition, three binding sites for a new single-stranded DNA binding protein were found. The presence of this factor in mature macrophages correlates with the in vivo DNA melting of one of the binding sites and with the enhancer strength.

The insulator protein CTCF represses transcription on binding to the (gt)(22)(ga)(15) microsatellite in intron 2 of the HLA-DRB1(*)0401 gene.

The insulator and transcription factor CTCF is a highly conserved 11 zinc finger protein possessing multiple specifities in DNA sequence recognition. CTCF regulates transcription of several genes, like the human oncogene c-myc or the chicken lysozyme gene by binding extremely divergent DNA sequences with different sets of its 11 zinc fingers. Recently, an insulator function was demonstrated for several CTCF binding elements. Here we show that CTCF binds to the (gt)(22)(ga)(15) microsatellite repeat A9 in intron 2 of the HLA-DRB1(*)0401 gene. Reporter gene activity is repressed by the A9 element. This repression is dependent on coexpressed CTCF and is even stronger compared with the CTCF binding site F1 of the chicken lysozyme gene, for which a silencer activity has been shown. This is the first report suggesting a function for microsatellite sequences in regulating specific gene expression.

Lysozyme gene expression and regulation.

Analysis of lysozyme gene expression in chicken and mouse identified two evolutionarily different mechanisms of lysozyme gene regulation. The lysozyme gene in chicken is expressed in the oviduct and macrophage cells with expression regulated by different, partially overlapping sets of tissue specific cis-acting elements. In contrast to chicken, the mouse genome contains two lysozyme genes generated by a gene duplication event allowing each gene to be regulated by its own regulatory region. One gene is expressed in macrophages, the other in Paneth cells of the small intestine. The macrophage-specific gene contains a single strong enhancer in the 3'-flanking sequences that interacts with ubiquitously factors. Cytosine methylation of the core enhancer sequence has been implicated in the regulation of the enhancer activity. In spite of these evolutionary regulatory differences, the chicken lysozyme transgene retains macrophage-specific expression in mice.