Specific mutations in the ligand binding domain selectively abolish the silencing function of human thyroid hormone receptor beta.

Although most nuclear hormone receptors are ligand-dependent transcriptional activators, certain members of this superfamily, such as thyroid hormone receptor (TR) and retinoic acid receptor (RAR), are involved in transcriptional repression. The silencing function of these receptors has been localized to the ligand binding domain (LBD). Previously, we demonstrated that overexpression of either the entire LBD or only the N-terminal region of the LBD (amino acids 168-259) is able to inhibit the silencing activity of TR. From this result we postulated the existence of a limiting factor (corepressor) that is necessary for TR silencing activity. To support this hypothesis, we identified amino acids in the N-terminal region of the LBD of TR that are important for the corepressor interaction and for the silencing function of TR. The silencing activity of TR was unaffected by overexpression of the LBD of mutant TR (V174A/D177A), suggesting that valine at position 174 and/or aspartic acid at position 177 are important for corepressor interaction. This mutant receptor protein, V174/D177, also lost the ability to silence target genes, suggesting that these amino acids are important for silencing function. Control experiments indicate that this mutant TR maintains its wild-type hormone binding and transactivation functions. These findings further strengthen the idea that the N-terminal region of the LBD of TR interacts with a putative corepressor protein(s) to achieve silencing of basal gene transcription.

Yeast global transcriptional regulators Sin4 and Rgr1 are components of mediator complex/RNA polymerase II holoenzyme.

Sin4 and Rgr1 proteins, previously shown by genetic studies to play both positive and negative roles in the transcriptional regulation of many genes, are identified here as components of mediator and RNA polymerase II holoenzyme complexes. Results with Sin4 deletion and Rgr1 truncation strains indicate the association of these proteins in a subcomplex comprising Sin4, Rgr1, Gal11, and a 50-kDa polypeptide. Taken together with the previous genetic evidence, our findings point to a role of the mediator in repression as well as in transcriptional activation.

Antagonism of WT1 activity by protein self-association.

Germline loss-of-function mutations at the Wilms tumor (WT) suppressor locus WT1 are associated with a predisposition to WTs and mild genital system anomalies. In contrast, germ-line missense mutations within the WT1 gene encoding the DNA-binding domain often yield a more severe phenotype consisting of WT, sexual ambiguity, and renal nephropathy. In this report, we demonstrate that the products of mutant alleles that impair DNA recognition can antagonize WT1-mediated transcriptional repression. We demonstrate that WT1 can self-associate in vitro and in vivo and that the responsible domain maps to the amino-terminal region of the protein. Oligomers of full-length protein form less efficiently or produce less stable complexes than oligomers between truncated polypeptides and full-length protein. Our data suggest a molecular mechanism to explain how WT1 mutations may act in deregulating cellular proliferation and differentiation.

Cell cycle regulation of the cyclin A gene promoter is mediated by a variant E2F site.

Cyclin A is involved in the control of S phase and mitosis in mammalian cells. Expression of the cyclin A gene in nontransformed cells is characterized by repression of its promoter during the G1 phase of the cell cycle and its induction at S-phase entry. We show that this mode of regulation is mediated by the transcription factor E2F, which binds to a specific site in the cyclin A promoter. It differs from the prototype E2F site in nucleotide sequence and protein binding; it is bound by E2F complexes containing cyclin E and p107 but not pRB. Ectopic expression of cyclin D1 triggers premature activation of the cyclin A promoter by E2F, and this effect is blocked by the tumor suppressor protein p16INK4.

Regulation of yeast phospholipid biosynthetic gene expression in response to inositol involves two superimposed mechanisms.

Transcription of phospholipid biosynthetic genes in the yeast Saccharomyces cerevisiae is maximally derepressed when cells are grown in the absence of inositol and repressed when the cells are grown in its presence. We have previously suggested that this response to inositol may be dictated by regulating transcription of the cognate activator gene, INO2. However, it was also known that cells which harbor a mutant opi1 allele express constitutively derepressed levels of target genes (INO1 and CHO1), implicating the OPI1 negative regulatory gene in the response to inositol. These observations suggested that the response to inositol may involve both regulation of INO2 transcription as well as OPI1-mediated repression. We investigated these possibilities by examining the effect of inositol on target gene expression in a strain containing the INO2 gene under control of the GAL1 promoter. In this strain, transcription of the INO2 gene was regulated in response to galactose but was insensitive to inositol. The expression of the INO1 and CHO1 target genes was still responsive to inositol even though expression of the INO2 gene was unresponsive. However, the level of expression of the INO1 and CHO1 target genes correlated with the level of INO2 transcription. Furthermore, the effect of inositol on target gene expression was eliminated by deleting the OPI1 gene in the GAL1-INO2-containing strain. These data suggest that the OPI1 gene product is the primary target (sensor) of the inositol response and that derepression of INO2 transcription determines the degree of expression of the target genes.

The amino-terminal domain of the prokaryotic enhancer-binding protein XylR is a specific intramolecular repressor.

The mechanism under which the signal-reception amino-terminal portion (A domain) of the prokaryotic enhancer-binding protein XylR controls the activity of the regulator has been investigated through complementation tests in vivo, in which the various protein segments were produced as independent polypeptides. Separate expression of the A domain repressed the otherwise constitutive activity of a truncated derivative of XylR deleted of its A domain (XylR delta A). Such inhibition was not released by m-xylene, the natural inducer of the system. Repression caused by the A domain was specific for XylR because it did not affect activation of the sigma 54 promoter PnifH by a derivative of its cognate regulator, NifA, deleted of its own A domain. The A domain was also unable to repress the activity of a NifA-XylR hybrid protein resulting from fusing two-thirds of the central domain of NifA to the carboxyl-terminal third of XylR, which includes its DNA-binding domain. The inhibitory effect caused by the A domain of XylR on XylR delta A seems, therefore, to result from specific interactions in trans between the two truncated proteins and not from mere hindering of an activating surface.

Simian virus 40 late gene expression is regulated by members of the steroid/thyroid hormone receptor superfamily.

Transcription of the late genes of simian virus 40 (SV40) is repressed during the early phase of the lytic cycle of infection of binding of cellular factors, called IBP-s, to the SV40 late promoter; repression is relieved after the onset of viral DNA replication by titration of these repressors. Preliminary data indicated that one of the major components of IBP-s was human estrogen-related receptor 1 (hERR1). We show here that several members of the steroid/thyroid hormone receptor superfamily, including testis receptor 2, thyroid receptor alpha 1 in combination with retinoid X receptor alpha, chicken ovalbumin upstream promoter transcription factors 1 and 2 (COUP-TF1 and COUP-TF2), as well as hERR1, possess the properties of IBP-s. These receptors bind specifically to hormone receptor binding sites present in the SV40 major late promoter. Recombinant COUP-TF1 specifically represses transcription from the SV40 major late promoter in a cell-free transcription system. Expression of COUP-TF1, COUP-TF2, or hERR1 in monkey cells results in repression of the SV40 late promoter, but not the early promoter, in the absence of the virally encoded large tumor antigen. Overexpression of COUP-TF1 leads to a delay in the early-to-late switch in SV40 gene expression during the lytic cycle of infection. Thus, members of this superfamily can play major direct roles in regulating expression of SV40. Possibly, natural or synthetic ligands to these receptors can serve as antiviral drugs. Our findings also provide the basis for the development of assays to screen for the ligands to testis receptor 2 and hERR1.

Contrasting roles for Myc and Mad proteins in cellular growth and differentiation.

The positive effects of Myc on cellular growth and gene expression are antagonized by activities of another member of the Myc superfamily, Mad. Characterization of the mouse homolog of human mad on the structural level revealed that domains shown previously to be required in the human protein for anti-Myc repression, sequence-specific DNA-binding activity, and dimerization with its partner Max are highly conserved. Conservation is also evident on the biological level in that both human and mouse mad can antagonize the ability of c-myc to cooperate with ras in the malignant transformation of cultured cells. An analysis of c-myc and mad gene expression in the developing mouse showed contrasting patterns with respect to tissue distribution and developmental stage. Regional differences in expression were more striking on the cellular level, particularly in the mouse and human gastrointestinal system, wherein c-Myc protein was readily detected in immature proliferating cells at the base of the colonic crypts, while Mad protein distribution was restricted to the postmitotic differentiated cells in the apex of the crypts. An increasing gradient of Mad was also evident in the more differentiated subcorneal layers of the stratified squamous epithelium of the skin. Together, these observations support the view that both downregulation of Myc and accumulation of Mad may be necessary for progression of precursor cells to a growth-arrested, terminally differentiated state.

Identification of the Bacillus subtilis pur operon repressor.

Transcription of the Bacillus subtilis pur operon is repressed in response to a signal of excess adenine. We have purified the repressor protein and have identified, cloned, and overexpressed the purR regulatory gene that controls transcription initiation of the operon. B. subtilis purR encodes a 62-kDa homodimer that binds to the pur operon control region. The PurR binding site which overlaps the promoter encompasses approximately 110 bp. The protein-DNA interaction is inhibited by 5-phosphoribosyl 1-pyrophosphate. A mutation that deletes the repressor binding site or one that disrupts purR abolishes binding activity in vitro and repression of transcription in vivo in response to the excess adenine signal. These results lead to a model in which an excess-adenine signal is transmitted to PurR via the 5-phosphoribosyl 1-pyrophosphate pool. In addition, purR is autoregulated. There is no structural or mechanistic similarity between the B. subtilis and Escherichia coli purine repressors.

Enhancers increase the probability but not the level of gene expression.

We have studied enhancer function in transient and stable expression assays in mammalian cells by using systems that distinguish expressing from nonexpressing cells. When expression is studied in this way, enhancers are found to increase the probability of a construct being active but not the level of expression per template. In stably integrated constructs, large differences in expression level are observed but these are not related to the presence of an enhancer. Together with earlier studies, these results suggest that enhancers act to affect a binary (on/off) switch in transcriptional activity. Although this idea challenges the widely accepted model of enhancer activity, it is consistent with much, if not all, experimental evidence on this subject. We hypothesize that enhancers act to increase the probability of forming a stably active template. When randomly integrated into the genome, enhancers may affect a metastable state of repression/activity, permitting expression in regions that would not permit activity of an isolated promoter.

The GA5 locus of Arabidopsis thaliana encodes a multifunctional gibberellin 20-oxidase: molecular cloning and functional expression.

The biosynthesis of gibberellins (GAs) after GA12-aldehyde involves a series of oxidative steps that lead to the formation of bioactive GAs. Previously, a cDNA clone encoding a GA 20-oxidase [gibberellin, 2-oxoglutarate:oxygen oxidoreductase (20-hydroxylating, oxidizing), EC 1.14.11.-] was isolated by immunoscreening a cDNA library from liquid endosperm of pumpkin (Cucurbita maxima L.) with antibodies against partially purified GA 20-oxidase. Here, we report isolation of a genomic clone for GA 20-oxidase from a genomic library of the long-day species Arabidopsis thaliana Heynh., strain Columbia, by using the pumpkin cDNA clone as a heterologous probe. This genomic clone contains a GA 20-oxidase gene that consists of three exons and two introns. The three exons are 1131-bp long and encode 377 amino acid residues. A cDNA clone corresponding to the putative GA 20-oxidase genomic sequence was constructed with the reverse transcription-PCR method, and the identity of the cDNA clone was confirmed by analyzing the capability of the fusion protein expressed in Escherichia coli to convert GA53 to GA44 and GA19 to GA20. The Arabidopsis GA 20-oxidase shares 55% identity and > 80% similarity with the pumpkin GA 20-oxidase at the derived amino acid level. Both GA 20-oxidases share high homology with other 2-oxoglutarate-dependent dioxygenases (2-ODDs), but the highest homology was found between the two GA 20-oxidases. Mapping results indicated tight linkage between the cloned GA 20-oxidase and the GA5 locus of Arabidopsis. The ga5 semidwarf mutant contains a G-->A point mutation that inserts a translational stop codon in the protein-coding sequence, thus confirming that the GA5 locus encodes GA 20-oxidase. Expression of the GA5 gene in Ara-bidopsis leaves was enhanced after plants were transferred from short to long days; it was reduced by GA4 treatment, suggesting end-product repression in the GA biosynthetic pathway.

Binding of purified multiple antibiotic-resistance repressor protein (MarR) to mar operator sequences.

Elevated expression of the marORAB multiple antibiotic-resistance operon enhances the resistance of Escherichia coli to various medically significant antibiotics. Transcription of the operon is repressed in vivo by the marR-encoded protein, MarR, and derepressed by salicylate and certain antibiotics. The possibility that repression results from MarR interacting with the marO operator-promoter region was studied in vitro using purified MarR and a DNA fragment containing marO. MarR formed at least two complexes with marO DNA, bound > 30-fold more tightly to it than to salmon sperm DNA, and protected two separate 21-bp sites within marO from digestion by DNase I. Site I abuts the downstream side of the putative -35 transcription-start signal and includes 4 bp of the -10 signal. Site II begins 13 bp downstream of site I, ending immediately before the first base pair of marR. Site II, approximately 80% homologous to site I, is not required for repression since a site II-deleted mutant (marO133) was repressed in trans by wild-type MarR. The absence of site II did not prevent MarR from complexing with the site I of marO133. Salicylate bound to MarR (Kd approximately 0.5 mM) and weakened the interaction of MarR with sites I and II. Thus, repression of the mar operon, which curbs the antibiotic resistance of E. coli, correlates with the formation of MarR-site I complexes. Salicylate appears to induce the mar operon by binding to MarR and inhibiting complex formation, whereas tetracycline and chloramphenicol, which neither bind MarR nor inhibit complex formation, must induce by an indirect mechanism.

Yeast histone H3 and H4 N termini function through different GAL1 regulatory elements to repress and activate transcription.

Previous work has shown that N-terminal deletions of yeast histone H3 cause a 2- to 4-fold increase in the induction of GAL1 and a number of other genes involved in galactose metabolism. In contrast, deletions at the H4 N terminus cause a 10- to 20-fold decrease in the induction of these same GAL genes. However, H3 and H4 N-terminal deletions each decrease PHO5 induction only 2- to 4-fold. To define the GAL1 gene regulatory elements through which the histone N termini activate or repress transcription, fusions were made between GAL1 and PHO5 promoter elements attached to a beta-galactosidase reporter gene. We show here that GAL1 hyperactivation caused by the H3 N-terminal deletion delta 4-15 is linked to the upstream activation sequence. Conversely, the relative decrease in GAL1 induction caused by the H4N-terminal deletion delta 4-28 is linked to the downstream promoter which contains the TATA element. These data indicate that the H3 N terminus is required for the repression of the GAL1 upstream element, whereas the H4N terminus is required for the activation of the GAL1 downstream promoter element.

The su(Hw) protein bound to gypsy sequences in one chromosome can repress enhancer-promoter interactions in the paired gene located in the other homolog.

The suppressor of Hairy-wing [su(Hw)] protein exerts a polar effect on gene expression by repressing the function of transcriptional enhancers located distally from the promoter with respect to the location of su(Hw) binding sequences. The directionality of this effect suggests that the su(Hw) protein specifically interferes with the basic mechanism of enhancer action. Moreover, mutations in modifier of mdg4 [mod(mdg4)] result in the repression of expression of a gene when the su(Hw) protein is bound to sequences in the copy of this gene located in the homologous chromosome. This effect is dependent on the presence of the su(Hw) binding region from the gypsy retrotransposon in at least one of the chromosomes and is enhanced by the presence of additional gypsy sequences in the other homology. This phenomenon is inhibited by chromosomal rearrangements that disrupt pairing, suggesting that close apposition between the two copies of the affected gene is important for trans repression of transcription. These results indicate that, in the absence of mod-(mdg4) product, the su(Hw) protein present in one chromosome can act in trans and inactivate enhancers located in the other homolog.

Cyclin-dependent protein kinase and cyclin homologs SSN3 and SSN8 contribute to transcriptional control in yeast.

The SSN3 and SSN8 genes of Saccharomyces cerevisiae were identified by mutations that suppress a defect in SNF1, a protein kinase required for release from glucose repression. Mutations in SSN3 and SSN8 also act synergistically with a mutation of the MIG1 repressor protein to relieve glucose repression. We have cloned the SSN3 and SSN8 genes. SSN3 encodes a cyclin-dependent protein kinase (cdk) homolog and is identical to UME5. SSN8 encodes a cyclin homolog 35% identical to human cyclin C. SSN3 and SSN8 fusion proteins interact in the two-hybrid system and coimmunoprecipitate from yeast cell extracts. Using an immune complex assay, we detected protein kinase activity that depends on both SSN3 and SSN8. Thus, the two SSN proteins are likely to function as a cdk-cyclin pair. Genetic analysis indicates that the SSN3-SSN8 complex contributes to transcriptional repression of diversely regulated genes and also affects induction of the GAL1 promoter.