A general strategy to enhance the potency of chimeric transcriptional activators

Efforts to increase the potency of transcriptional activators are generally unsuccessful because poor expression of activators in mammalian cells limits their delivery to target promoters. Here we report that the effectiveness of chimeric activators can be dramatically improved by expressing them as noncovalent tetrameric bundles. Bundled activation domains are much more effective at activating a reporter gene than simple monomeric activators, presumably because, at similar expression levels, up to 4 times as many the activation domains are delivered to the target promoter. These bundled activation domains are also more effective than proteins in which activation domains are tandemly reiterated in the same polypeptide chain, because such proteins are very poorly expressed and therefore not delivered effectively. These observations suggest that there is a threshold number of activation domains that must be bound to a promoter for activation, above which promoter activity is simply a function of the number of activators bound. We show that bundling can be exploited practically to enhance the sensitivity of mammalian two-hybrid assays, enabling detection of weak interactions or those between poorly expressed proteins. Bundling also dramatically improves the performance of a small-molecule-regulated gene expression system when the expression level of regulatory protein is limiting, a situation that may be encountered in gene therapy applications.

Localization of a putative transcriptional regulator (ATRX) at pericentromeric heterochromatin and the short arms of acrocentric chromosomes

ATRX is a member of the SNF2 family of helicase/ATPases that is thought to regulate gene expression via an effect on chromatin structure and/or function. Mutations in the hATRX gene cause severe syndromal mental retardation associated with α-thalassemia. Using indirect immunofluorescence and confocal microscopy we have shown that ATRX protein is associated with pericentromeric heterochromatin during interphase and mitosis. By coimmunofluorescence, ATRX localizes with a mouse homologue of the Drosophila heterochromatic protein HP1 in vivo, consistent with a previous two-hybrid screen identifying this interaction. From the analysis of a trap assay for nuclear proteins, we have shown that the localization of ATRX to heterochromatin is encoded by its N-terminal region, which contains a conserved plant homeodomain-like finger and a coiled-coil domain. In addition to its association with heterochromatin, at metaphase ATRX clearly binds to the short arms of human acrocentric chromosomes, where the arrays of ribosomal DNA are located. The unexpected association of a putative transcriptional regulator with highly repetitive DNA provides a potential explanation for the variability in phenotype of patients with identical mutations in the ATRX gene.

Analysis of a peptide hormone–receptor interaction in the yeast two-hybrid system

Interaction between a peptide hormone and extracellular domains of its receptor is a crucial step for initiation of hormone action. We have developed a modification of the yeast two-hybrid system to study this interaction and have used it to characterize the interaction of insulin-like growth factor 1 (IGF-1) with its receptor by using GAL4 transcriptional regulation with a β-galactosidase assay as readout. In this system, IGF-1 and proIGF-1 bound to the cysteine-rich domain, extracellular domain, or entire IGF-1 proreceptor. This interaction was specific. Thus, proinsulin showed no significant interaction with the IGF-1 receptor, while a chimeric proinsulin containing the C-peptide of IGF-1 had an intermediate interaction, consistent with its affinity for the IGF-1 receptor. Over 2000 IGF-1 mutants were generated by PCR and screened for interaction with the color assay. About 40% showed a strong interaction, 20% showed an intermediate interaction, and 40% give little or no signal. Of 50 mutants that were sequenced, several (Leu-5 → His, Glu-9 → Val, Arg-37 → Gly, and Met-59 → Leu) appeared to enhance receptor association, others resulted in weaker receptor interaction (Tyr-31 → Phe and Ile-43 → Phe), and two gave no detectable signal (Leu-14 → Arg and Glu-46 → Ala). Using PCR-based mutagenesis with proinsulin, we also identified a gain of function mutant (proinsulin Leu-17 → Pro) that allowed for a strong IGF-1–receptor interaction. These data demonstrate that the specificity of the interaction between a hormone and its receptor can be characterized with high efficiency in the two-hybrid system and that novel hormone analogues may be found by this method.

The orientation of the AP-1 heterodimer on DNA strongly affects transcriptional potency

Activation of gene transcription in eukaryotes requires the cooperative assembly of an initiation complex containing many protein subunits. The necessity that these components contact each other and the promoter/enhancer in defined ways suggests that their spatial arrangement might influence the activation response. Indeed, growing evidence indicates that DNA architecture can profoundly affect transcriptional potency. Much less is known about the influence of protein architecture on transcriptional activation. Here, we examine the architectural dependence of activator function through the analysis of matched pairs of AP-1•DNA complexes differing only in their orientation. Mutation of a critical Arg residue in the basic-leucine zipper domain of either Fos or Jun yielded single point-mutant heterodimers that bind DNA in a single defined orientation, as determined directly by native chemical ligation/affinity cleavage; by contrast, the corresponding wild-type protein binds DNA as a roughly equal mixture of two isomeric orientations, which are related by subunit interchange. The stereochemistry of the point-mutant heterodimers could be switched by inversion of a C•G base pair in the center of the AP-1 site, thus providing access to both fixed orientational isomers. Yeast reporter gene assays consistently revealed that one orientational isomer activates transcription at least 10-fold more strongly than the other. These results suggest that protein architecture, especially the spatial relationship of the activation domain to the promoter, can exert a powerful influence on activator potency.

Transcriptional coupling between the divergent promoters of a prototypic LysR-type regulatory system, the ilvYC operon of Escherichia coli

The twin-domain model [Liu, L. F. & Wang, J. C. (1987) Proc. Natl. Acad. Sci. USA 84, 7024–7027] suggests that closely spaced, divergent, superhelically sensitive promoters can affect the transcriptional activity of one another by transcriptionally induced negative DNA supercoiling generated in the divergent promoter region. This gene arrangement is observed for many LysR-type-regulated operons in bacteria. We have examined the effects of divergent transcription in the prototypic LysR-type system, the ilvYC operon of Escherichia coli. Double-reporter constructs with the lacZ gene under transcriptional control of the ilvC promoter and the galK gene under control of the divergent ilvY promoter were used to demonstrate that a down-promoter mutation in the ilvY promoter severely decreases in vivo transcription from the ilvC promoter. However, a down-promoter mutation in the ilvC promoter only slightly affects transcription from the ilvY promoter. In vitro transcription assays with DNA topoisomers showed that transcription from the ilvC promoter increases over the entire range of physiological superhelical densities, whereas transcription initiation from the ilvY promoter exhibits a broad optimum at a midphysiological superhelical density. Evidence that this promoter coupling is DNA supercoiling-dependent is provided by the observation that a novobiocin-induced decrease in global negative superhelicity results in an increase in ilvY promoter activity and a decrease in ilvC promoter activity predicted by the in vitro data. We suggest that this transcriptional coupling is important for coordinating basal level expression of the ilvYC operon with the nutritional and environmental conditions of cell growth.

SMAD3/4-dependent transcriptional activation of the human type VII collagen gene (COL7A1) promoter by transforming growth factor β

The human type VII collagen gene (COL7A1) recently has been identified as an immediate-early response gene for transforming growth factor β (TGF-β)/SMAD signaling pathway. In this study, by using MDA-MB-468 SMAD4−/− breast carcinoma cells, we demonstrate that expression of SMAD4 is an absolute requirement for SMAD-mediated promoter activity. We also demonstrate that the SMAD binding sequence (SBS) representing the TGF-β response element in the region −496/−444 of the COL7A1 promoter functions as an enhancer in the context of a heterologous promoter. Electrophoretic mobility-shift assays with nuclear extracts from COS-1 cells transfected with expression vectors for SMADs 1–5 indicate that SMAD3 forms a complex with a migration similar to that of the endogenous TGF-β-specific complex observed in fibroblast extracts. Electrophoretic mobility-shift assays using recombinant glutathione S-transferase-SMAD fusion proteins indicate that both SMAD4 and C-terminally truncated SMAD3, but not SMAD2, can bind the COL7A1 SBS. Coexpression of SMAD3 and SMAD4 in COS-1 cells leads to the formation of two complexes: a DNA/protein complex containing SMAD3 alone and another slower-migrating complex containing both SMAD3 and SMAD4, the latter complex not being detected in fibroblasts. Maximal transactivation of COL7A1 SBS-driven promoters in either MDA-MB-468 carcinoma cells or fibroblasts requires concomitant overexpression of SMAD3 and SMAD4. These data may represent the first identification of a functional homomeric SMAD3 complex regulating a human gene.

Functional antioxidant responsive elements

Exposure of human and rodent cells to a wide variety of chemoprotective compounds confers resistance against a broad set of carcinogens. For a subset of the chemoprotective compounds, protection is generated by an increase in the abundance of protective enzymes like glutathione S-transferases (GST). Antioxidant responsive elements (AREs) mediate the transcriptional induction of a battery of genes which comprise much of this chemoprotective response system. Past studies identified a necessary ARE “core” sequence of RTGACnnnGC, but this sequence alone is insufficient to mediate induction. In this study, the additional sequences necessary to define a sufficient, functional ARE are identified through systematic mutational analysis of the murine GST Ya ARE. Introduction of the newly identified necessary nucleotides into the regions flanking a nonresponsive, ARE-like, GST-Mu promoter sequence produced an inducible element. A screen of the GenBank database with the newly identified ARE consensus identified 16 genes which contained the functional ARE consensus sequence in their promoters. Included within this group was an ARE sequence from the murine ferritin-L promoter that mediated induction when tested. In an electrophoretic mobility-shift assay, the ferritin-L ARE was bound by ARE–binding protein 1, a protein previously identified as the likely mediator of the chemoprotective response. A three-level ARE classification system is presented to account for the distinct induction strengths observed in our mutagenesis studies. A model of the ARE as a composite regulatory site, where multiple transcription factors interact, is presented to account for the complex characteristics of ARE-mediated chemoprotective gene expression.

Scanning mutagenesis of the putative transmembrane segments of Kir2.1, an inward rectifier potassium channel

Structural models of inward rectifier K+ channels incorporate four identical or homologous subunits, each of which has two hydrophobic segments (M1 and M2) which are predicted to span the membrane as α helices. Since hydrophobic interactions between proteins and membrane lipids are thought to be generally of a nonspecific nature, we attempted to identify lipid-contacting residues in Kir2.1 as those which tolerate mutation to tryptophan, which has a large hydrophobic side chain. Tolerated mutations were defined as those which produced measurable inwardly rectifying currents in Xenopus oocytes. To distinguish between water-accessible positions and positions adjacent to membrane lipids or within the protein interior we also mutated residues in M1 and M2 individually to aspartate, since an amino acid with a charged side chain should not be tolerated at lipid-facing or interior positions, due to the energy cost of burying a charge in a hydrophobic environment. Surprisingly, 17 out of 20 and 17 out of 22 non-tryptophan residues in M1 and M2, respectively, tolerated being mutated to tryptophan. Moreover, aspartate was tolerated at 15 out of 22 and 15 out of 21 non-aspartate M1 and M2 positions respectively. Periodicity in the pattern of tolerated vs. nontolerated mutations consistent with α helices or β strands did not emerge convincingly from these data. We consider the possibility that parts of M1 and M2 may be in contact with water.

Recruitment of SMRT/N-CoR-mSin3A-HDAC-repressing complexes is not a general mechanism for BTB/POZ transcriptional repressors: The case of HIC-1 and γFBP-B

Hypermethylated in cancer (HIC-1), a new candidate tumor suppressor gene located in 17p13.3, encodes a protein with five C2H2 zinc fingers and an N-terminal broad complex, tramtrack, and bric à brac/poxviruses and zinc-finger (BTB/POZ) domain found in actin binding proteins or transcriptional regulators involved in chromatin modeling. In the human B cell lymphoma (BCL-6) and promyelocityc leukemia (PLZF) oncoproteins, this domain mediates transcriptional repression through its ability to recruit a silencing mediator of retinoid and thyroid hormone receptor (SMRT)/nuclear receptor corepressor (N-CoR)-mSin3A-histone deacetylase (HDAC) complex, a mechanism shared with numerous transcription factors. HIC-1 appears unique because it contains a 13-aa insertion acquired late in evolution, because it is not found in its avian homologue, γF1-binding protein isoform B (γFBP-B), a transcriptional repressor of the γF-crystallin gene. This insertion, located in a conserved region involved in the dimerization and scaffolding of the BTB/POZ domain, mainly affects slightly the ability of the HIC-1 and γFBP-B BTB/POZ domains to homo- and heterodimerize in vivo, as shown by mammalian two-hybrid experiments. Both the HIC-1 and γFBP-B BTB/POZ domains behave as autonomous transcriptional repression domains. However, in striking contrast with BCL-6 and PLZF, both HIC-1 and γFBP-B similarly fail to interact with members of the HDAC complexes (SMRT/N-CoR, mSin3A or HDAC-1) in vivo and in vitro. In addition, a general and specific inhibitor of HDACs, trichostatin A, did not alleviate the HIC-1- and γFBP-B-mediated transcriptional repression, as previously shown for BCL-6. Taken together, our studies show that the recruitment onto target promoters of an HDAC complex is not a general property of transcriptional repressors containing a conserved BTB/POZ domain.

Caenorhabditis elegans Mediator complexes are required for developmental-specific transcriptional activation

Mediator proteins are required for transcriptional regulation of most genes in yeast. Mammalian Mediator homologs also function as transcriptional coactivators in vitro; however, their physiological role in gene-specific transcription is not yet known. To determine the role of Mediator proteins in the development of complex organisms, we purified putative Mediator complexes from Caenorhabditis elegans and analyzed their phenotypes in vivo. C. elegans Mediator homologs were assembled into two multiprotein complexes. RNA interference assays showed that the CeMed6, CeMed7, and CeMed10/CeNut2 gene products are required for the expression of developmentally regulated genes, but are dispensable for expression of the ubiquitously expressed genes tested in this study. Therefore, the gene-specific function of Mediator as an integrator of transcriptional regulatory signals is evolutionarily conserved and is essential for C. elegans development.

Long-distance transcriptional enhancement by the histone acetyltransferase PCAF

Enhancers are defined by their ability to stimulate gene activity from remote sites and their requirement for promoter-proximal upstream activators to activate transcription. Here we demonstrate that recruitment of the p300/CBP-associated factor PCAF to a reporter gene is sufficient to stimulate promoter activity. The PCAF-mediated stimulation of transcription from either a distant or promoter-proximal position depends on the presence of an upstream activator (Sp1). These data suggest that acetyltransferase activity may be a primary component of enhancer function, and that recruitment of polymerase and enhancement of transcription are separable. Transcriptional activation by PCAF requires both its acetyltransferase activity and an additional activity within its N terminus. We also show that the simian virus 40 enhancer and PCAF itself are sufficient to counteract Mad-mediated repression. These results are compatible with recent models in which gene activity is regulated by the competition between deacetylase-mediated repression and enhancer-mediated recruitment of acetyltransferases.

A transcriptional activating region with two contrasting modes of protein interaction

A C-terminal segment of the yeast activator Gal4 manifests two functions: When tethered to DNA, it elicits gene activation, and it binds the inhibitor Gal80. Here we examine the effects on these two functions of cysteine and proline substitutions. We find that, although certain cysteine substitutions diminish interaction with Gal80, those substitutions have little effect on the activating function in vivo and interaction with TATA box-binding protein (TBP) in vitro. Proline substitutions introduced near residues critical for Gal80 binding abolish that interaction but once again have no effect on the activating function. Crosslinking experiments show that a defined position in the activating peptide is in close proximity to TBP and Gal80 in the two separate reactions and show that binding of the inhibitor blocks binding to TBP. Thus, the same stretch of amino acids are involved in two quite different protein–protein interactions: binding to Gal80, which depends on a precise sequence and the formation of a defined secondary structure, or interactions with the transcriptional machinery in vivo, which are not impaired by perturbations of either sequence or structure.

Graded transcriptional response to different concentrations of a single transactivator

Threshold mechanisms of transcriptional activation are thought to be critical for translating continuous gradients of extracellular signals into discrete all-or-none cellular responses, such as mitogenesis and differentiation. Indeed, unequivocal evidence for a graded transcriptional response in which the concentration of inducer directly correlates with the level of gene expression in individual eukaryotic cells is lacking. By using a novel binary tetracycline regulatable retroviral vector system, we observed a graded rather than a threshold mechanism of transcriptional activation in two different model systems. When polyclonal populations of cells were analyzed at the single cell level, a dose-dependent, stepwise increase in expression of the reporter gene, green fluorescent protein (GFP), was observed by fluorescence-activated cell sorting. These data provide evidence that, in addition to the generally observed all-or-none switch, the basal transcription machinery also can respond proportionally to changes in concentration of extracellular inducers and trancriptional activators.

Silencer elements controlling the B29 (Igβ) promoter are neither promoter- nor cell-type-specific

The murine B29 (Igβ) promoter is B cell specific and contains essential SP1, ETS, OCT, and Ikaros motifs. Flanking 5′ DNA sequences inhibit B29 promoter activity, suggesting this region contains silencer elements. Two adjacent 5′ DNA segments repress transcription by the murine B29 promoter in a position- and orientation-independent manner, analogous to known silencers. Both these 5′ segments also inhibit transcription by several heterologous promoters in B cells, including mb-1, c-fos, and human B29. These 5′ segments also inhibit transcription by the c-fos promoter in T cells suggesting they are not B cell-specific elements. DNase I footprint analyses show an approximately 70-bp protected region overlapping the boundary between the two negative regulatory DNA segments and corresponding to binding sites for at least two different DNA-binding proteins. Within this footprint, two unrelated 30-bp cis-acting DNA motifs (designated TOAD and FROG) function as position- and orientation-independent silencers when located directly 5′ of the murine B29 promoter. These two silencer motifs act cooperatively to restrict the transcriptional activity of the B29 promoter. Neither of these motifs resembles any known silencers. Mutagenesis of the TOAD and FROG motifs in their respective 5′ DNA segments eliminates the silencing activity of these upstream regions, indicating these two motifs as the principal B29 silencer elements within these regions.

Spatio-temporally controlled site-specific somatic mutagenesis in the mouse

The efficient introduction of somatic mutations in a given gene, at a given time, in a specific cell type will facilitate studies of gene function and the generation of animal models for human diseases. We have shown previously that conditional recombination–excision between two loxP sites can be achieved in mice by using the Cre recombinase fused to a mutated ligand binding domain of the human estrogen receptor (Cre-ERT), which binds tamoxifen but not estrogens. DNA excision was induced in a number of tissues after administration of tamoxifen to transgenic mice expressing Cre-ERT under the control of the cytomegalovirus promoter. However, the efficiency of excision varied between tissues, and the highest level (≈40%) was obtained in the skin. To determine the efficiency of excision mediated by Cre-ERT in a given cell type, we have now crossed Cre-ERT-expressing mice with reporter mice in which expression of Escherichia coli β-galactosidase can be induced through Cre-mediated recombination. The efficiency and kinetics of this recombination were analyzed at the cellular level in the epidermis of 6- to 8-week-old double transgenic mice. We show that site-specific excision occurred within a few days of tamoxifen treatment in essentially all epidermis cells expressing Cre-ERT. These results indicate that cell-specific expression of Cre-ERT in transgenic mice can be used for efficient tamoxifen-dependent, Cre-mediated recombination at loci containing loxP sites to generate site-specific somatic mutations in a spatio-temporally controlled manner.