Design and function of transcriptional switches in Drosophila.

Extensive genetic and biochemical analysis of Drosophila melanogaster has made this system an important model for characterization of transcriptional regulatory elements and factors. Given the striking conservation of transcriptional controls in metazoans, general principles derived from studies of Drosophila are expected to continue to illuminate transcriptional regulation in other systems, including vertebrates. With improvement in technologies for genetic manipulation of insects, research in Drosophila will also aid the design of systems for controlled expression of genes in other hosts. This review focuses on recent advances from Drosophila in analysis of the functional components of transcriptional switches, including basal promoters, enhancers, boundary elements, and maintenance elements.

Cell-type specificity of short-range transcriptional repressors.

Transcriptional repressors can be classified as short- or long-range, according to their range of activity. Functional analysis of identified short-range repressors has been carried out largely in transgenic Drosophila, but it is not known whether general properties of short-range repressors are evident in other types of assays. To study short-range transcriptional repressors in cultured cells, we created chimeric tetracycline repressors based on Drosophila transcriptional repressors Giant, Drosophila C-terminal-binding protein (dCtBP), and Knirps. We find that Giant and dCtBP are efficient repressors in Drosophila and mammalian cells, whereas Knirps is active only in insect cells. The restricted activity of Knirps, in contrast to that of Giant, suggests that not all short-range repressors possess identical activities, consistent with recent findings showing that short-range repressors act through multiple pathways. The mammalian repressor Kid is more effective than either Giant or dCtBP in mammalian cells but is inactive in Drosophila cells. These results indicate that species-specific factors are important for the function of the Knirps and Kid repressors. Giant and dCtBP repress reporter genes in a variety of contexts, including genes that were introduced by transient transfection, carried on episomal elements, or stably integrated. This broad activity indicates that the context of the target gene is not critical for the ability of short-range repressors to block transcription, in contrast to other repressors that act only on stably integrated genes.

Role of CtBP in transcriptional repression by the Drosophila giant protein.

The giant protein is a short-range transcriptional repressor that refines the expression pattern of gap and pair-rule genes in the Drosophila blastoderm embryo. Short-range repressors including knirps, Krüppel, and snail utilize the CtBP cofactor for repression, but it is not known whether a functional interaction with CtBP is a general property of all short-range repressors. We studied giant repression activity in a CtBP mutant and find that this cofactor is required for giant repression of some, but not all, genes. While targets of giant such as the even-skipped stripe 2 enhancer and a synthetic lacZ reporter show clear derepression in the CtBP mutant, another giant target, the hunchback gene, is expressed normally. A more complex situation is seen with regulation of the Krüppel gene, in which one enhancer is repressed by giant in a CtBP-dependent manner, while another is repressed in a CtBP-independent manner. These results demonstrate that giant can repress both via CtBP-dependent and CtBP-independent pathways, and that promoter context is critical for determining giant-CtBP functional interaction. To initiate mechanistic studies of the giant repression activity, we have identified a minimal repression domain within giant that encompasses residues 89-205, including an evolutionarily conserved region bearing a putative CtBP binding motif.

dCtBP-dependent and -independent repression activities of the Drosophila Knirps protein.

Transcriptional repressor proteins play essential roles in controlling the correct temporal and spatial patterns of gene expression in Drosophila melanogaster embryogenesis. Repressors such as Knirps, Krüppel, and Snail mediate short-range repression and interact with the dCtBP corepressor. The mechanism by which short-range repressors block transcription is not well understood; therefore, we have undertaken a detailed structure-function analysis of the Knirps protein. To provide a physiological setting for measurement of repression, the activities of endogenous or chimeric Knirps repressor proteins were assayed on integrated reporter genes in transgenic embryos. Two distinct repression functions were identified in Knirps. One repression activity depends on dCtBP binding, and this function maps to a C-terminal region of Knirps that contains a dCtBP binding motif. In addition, an N-terminal region was identified that represses in a CtBP mutant background and does not bind to the dCtBP protein in vitro. Although the dCtBP protein is important for Knirps activity on some genes, one endogenous target of the Knirps protein, the even-skipped stripe 3 enhancer, is not derepressed in a CtBP mutant. These results indicate that Knirps can utilize two different pathways to mediate transcriptional repression and suggest that the phenomenon of short-range repression may be a combination of independent activities.

Transcriptional repression by the Drosophila giant protein: cis element positioning provides an alternative means of interpreting an effector gradient.

Early developmental patterning of the Drosophila embryo is driven by the activities of a diverse set of maternally and zygotically derived transcription factors, including repressors encoded by gap genes such as Krüppel, knirps, giant and the mesoderm-specific snail. The mechanism of repression by gap transcription factors is not well understood at a molecular level. Initial characterization of these transcription factors suggests that they act as short-range repressors, interfering with the activity of enhancer or promoter elements 50 to 100 bp away. To better understand the molecular mechanism of short-range repression, we have investigated the properties of the Giant gap protein. We tested the ability of endogenous Giant to repress when bound close to the transcriptional initiation site and found that Giant effectively represses a heterologous promoter when binding sites are located at -55 bp with respect to the start of transcription. Consistent with its role as a short-range repressor, as the binding sites are moved to more distal locations, repression is diminished. Rather than exhibiting a sharp 'step-function' drop-off in activity, however, repression is progressively restricted to areas of highest Giant concentration. Less than a two-fold difference in Giant protein concentration is sufficient to determine a change in transcriptional status of a target gene. This effect demonstrates that Giant protein gradients can be differentially interpreted by target promoters, depending on the exact location of the Giant binding sites within the gene. Thus, in addition to binding site affinity and number, cis element positioning within a promoter can affect the response of a gene to a repressor gradient. We also demonstrate that a chimeric Gal4-Giant protein lacking the basic/zipper domain can specifically repress reporter genes, suggesting that the Giant effector domain is an autonomous repression domain.

Long-range repression in the Drosophila embryo.

Transcriptional repressors can be characterized by their range of action on promoters and enhancers. Short-range repressors interact over distances of 50-150 bp to inhibit, or quench, either upstream activators or the basal transcription complex. In contrast, long-range repressors act over several kilobases to silence basal promoters. We describe recent progress in characterizing the functional properties of one such long-range element in the Drosophila embryo and discuss the contrasting types of gene regulation that are made possible by short- and long-range repressors.

The gap protein knirps mediates both quenching and direct repression in the Drosophila embryo.

Transcriptional repression is essential for establishing localized patterns of gene expression during Drosophila embryogenesis. Several mechanisms of repression have been proposed, including competition, quenching and direct repression of the transcription complex. Previous studies suggest that the knirps orphan receptor (kni) may repress transcription via competition, and exclude the binding of the bicoid (bcd) activator to an overlapping site in a target promoter. Here we present evidence that kni can quench, or locally inhibit, upstream activators within a heterologous enhancer in transgenic embryos. The range of kni repression is approximately 50-100 bp, so that neighboring enhancers in a modular promoter are free to interact with the transcription complex (enhancer autonomy). However, kni can also repress the transcription complex when bound in promoter-proximal regions. In this position, kni functions as a dominant repressor and blocks multiple enhancers in a modular promoter. Our studies suggest that short-range repression represents a flexible form of gene regulation, exhibiting enhancer- or promoter-specific effects depending on the location of repressor binding sites.

The eve stripe 2 enhancer employs multiple modes of transcriptional synergy.

Previous studies have provided a detailed model for the regulation of even-skipped (eve) stripe 2 expression in the Drosophila embryo. The bicoid (bcd) regulatory gradient triggers the expression of hunchback (hb); these work synergistically to activate the stripe in the anterior half of the embryo, bcd also coordinates the expression of two repressors, giant (gt) and Kruppel (Kr), which define the anterior and posterior borders of the stripe, respectively. Here, we report the findings of extensive cis- and trans- complementation analyses using a series of defective stripe 2 enhancers in transgenic embryos. This study reaches two primary conclusions. First, the strip 2 enhancer is inherently 'sensitized' for repression by gt. We propose that gt specifies the sharp anterior stripe border by blocking two tiers of transcriptional synergy, cooperative binding to DNA and cooperative contact of bound activators with the transcription complex. Second, we find that the synergistic activity of hb and bcd is 'promiscuous'. For example, a maternally expressed Gal4-Sp1 fusion protein can functionally replace hb in the stripe 2 enhancer. This finding challenges previous proposals for dedicated hb and bcd interactions in the segmentation process.

Specific transcriptional activation in vitro by the herpes simplex virus protein VP16.

The herpes simplex virus protein VP16 interacts with cellular factors, including the protein Oct-1, to activate viral immediate early (IE) gene transcription. We have reproduced this effect by addition of purified, full-length VP16 and the DNA-binding 'POU' domain of Oct-1 (Oct-1/POU) to a HeLa cell in vitro transcription system. Stimulation of transcription was dependent on the IE-specific element, TAATGARAT. In agreement with earlier observations from electrophoretic mobility shift assays, activation was not observed when Oct-2/POU, the DNA-binding domain from the Oct-2 protein, was substituted for Oct-1/POU. Single round transcription assays revealed that, together, VP16 and Oct-1/POU facilitate the assembly of pre-initiation complexes at target gene promoters.

Spacing ensures autonomous expression of different stripe enhancers in the even-skipped promoter.

The even-skipped (eve) promoter contains a series of enhancers that control the expression of different segmentation stripes in the Drosophila embryo. The stripe 3 enhancer is located 1.7 kb upstream of the stripe 2 enhancer. Here we demonstrate that these enhancers must be physically separated by a minimum distance for proper stripe expression. When they are directly coupled in either orientation, the enhancers generate abnormal patterns of expression in the early embryo. For example, the levels of stripe 2 expression are augmented and there is a posterior expansion of the pattern when the stripe 3 enhancer is positioned immediately upstream of the stripe 2 enhancer. Despite this spacing requirement, the order of the enhancers within the eve promoter can be reversed without affecting the normal expression pattern. These results suggest that spacing maintains the autonomous activities of the stripe enhancers and that interactions between enhancers can generate novel patterns of gene expression.

Conserved cysteine residues of Oct-2 POU domain confer sensitivity to oxidation but are dispensable for sequence-specific DNA binding.

The POU family of proteins, including the Oct-2 transcription factor, is characterized by a highly conserved bipartite DNA binding domain containing a 'POU homeodomain', distantly related to homeodomains of other DNA binding proteins, and a 'POU specific' domain unique to this class of factors. Prompted by the finding that in vitro DNA binding by Oct-2 is reversibly inhibited by oxidation of the protein, we investigated the role of the cysteine residues in the POU domain. All POU homeodomains identified contain a cysteine in the helix 3 region presumed to contact DNA directly; many (including Oct-2) also contain a less-well conserved cysteine residue(s) in the POU specific domain. Replacement of these cysteines with serine residues rendered the DNA binding domain resistant to oxidation but did not appreciably change the binding to a canonical octamer sequence, suggesting that the conserved cysteine residues are not required for sequence-specific DNA contacts, but may be important for another function.

Oct-2 facilitates functional preinitiation complex assembly and is continuously required at the promoter for multiple rounds of transcription.

Octamer factor 2 (Oct-2, OTF-2, NF-A2) is an 'upstream' promoter factor that binds to the octamer motif (ATGCAAAT) implicated in control of immunoglobulin gene transcription in B-lymphocytes. We have studied the role of Oct-2 in the process of transcription initiation in vitro using both nuclear extracts and purified basal transcription factors. Oct-2 specifically stimulates transcription from octamer-containing promoters in both systems. Thus, Oct-2 is a 'true activator', rather than merely an 'anti-repressor' counteracting the effect of histones. In order-of-addition experiments, Oct-2 is required early, together with TFIID, to allow formation of a preinitiation complex. Oct-2 cannot functionally interact with cloned TATA binding protein (TBP) but rather requires 'coactivators' found in the TFIID fraction. In single-round transcription experiments, early competition for Oct-2 by an octamer oligonucleotide is deleterious, but no effect is seen after assembly of a complete preinitiation complex. However, for multiple rounds of transcription, Oct-2 is continuously required at the promoter; this result argues against a 'hit-and-run' mechanism whereby the activator becomes dispensible after organizing a TFIID-promoter complex. In agreement with our previous studies in vivo, the N-terminal glutamine-rich activation domain of Oct-2 is required for full activity in vitro, indicating that this domain directly interacts with basal transcription factors.

POU-specific domain of Oct-2 factor confers 'octamer' motif DNA binding specificity on heterologous Antennapedia homeodomain.

The bipartite DNA binding domain of the POU family of transcription factors contains a 'POU-specific' domain unique to this class of factors and a 'POU homeodomain' homologous to other homeodomains. We compared DNA binding of the Oct-2 factor POU domain and the Antennapedia (Antp) homeodomain with a chimeric Oct-2/Antp protein in which the distantly related Antp homeodomain was substituted for the Oct-2 POU homeodomain. The Oct-2/Antp chimeric protein bound both the octamer and the Antp sites efficiently, indicating that DNA binding specificity is contributed by both components of the POU domain.

Regulation of Escherichia coli sigma F RNA polymerase by flhD and flhC flagellar regulatory genes.

The sigma F RNA polymerase has been characterized biochemically and is known to transcribe several flagellar genes in Escherichia coli. It was found that while the flagellar regulatory genes flhD and flhC are required for sigma F activity, the sizes of their corresponding gene products are inconsistent with their encoding sigma F itself.

An apaH mutation causes AppppA to accumulate and affects motility and catabolite repression in Escherichia coli.

apaH- mutants lack the hydrolase responsible for degradation of AppppN dinucleotides in Escherichia coli and show a greater than or equal to 16-fold increase in AppppA under nonstress conditions. These mutants lack detectable activity of sigma F, a factor required for transcription of motility and chemotaxis genes. Expression of the flbB/flaI operon, thought to encode sigma F, is decreased in apaH- mutants, and there appears to be a general decrease in expression of genes regulated by cAMP-binding protein and cAMP as well.

Secondary sigma factor controls transcription of flagellar and chemotaxis genes in Escherichia coli.

The genes specifying chemotaxis, motility, and flagellar function in Escherichia coli are coordinately regulated and form a large and complex regulon. Despite the importance of these genes in controlling bacterial behavior, little is known of the molecular mechanisms that regulate their expression. We have identified a minor form of E. coli RNA polymerase that specifically transcribes several E. coli chemotaxis/flagellar genes in vitro and is likely to carry out transcription of these genes in vivo. The enzyme was purified to near homogeneity based on its ability to initiate transcription of the E. coli tar chemotaxis gene at start sites that are used in vivo. Specific tar transcription activity is associated with a polypeptide of apparent 28-kDa molecular mass that remains bound to the E. coli RNA polymerase throughout purification. This peptide behaves as a secondary sigma factor--designated sigma F--because it restores specific tar transcription activity when added to core RNA polymerase. The sigma F holoenzyme also transcribes the E. coli tsr and flaAI genes in vitro as well as several Bacillus subtilis genes that are transcribed specifically by the sigma 28 form of B. subtilis RNA polymerase. The latter holoenzyme is implicated in transcription of flagellar and chemotaxis genes in B. subtilis. Hence E. coli sigma F holoenzyme appears to be analogous to the B. subtilis sigma 28 RNA polymerase, both in its promoter specificity and in the nature of the regulon it controls.

Characterization of heat shock in Bacillus subtilis.

We characterized the general properties of the heat shock response in Bacillus subtilis W168, B. subtilis JH642, and an spo0A mutant by using pulse-labeling of bacterial proteins and one-dimensional sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The transfer of cells from 37 to 50 degrees C repressed synthesis of most cellular proteins and led to the induction of at least 26 distinct heat shock proteins after about 3 min. Ethanol (4% [vol/vol]) induced a similar set of proteins, but somewhat more slowly. Synthesis of the majority of heat shock proteins at 50 degrees C returned to a steady-state level 20 to 40 min after the shock. Although no B. subtilis heat shock protein has yet been extensively characterized, three of these proteins were found to be immunologically related to the Escherichia coli heat shock proteins Dnak, Lon, and GroEL. Synthesis of both sigma 28 and sigma 43 proteins was sharply reduced during heat shock. Although a spo0A amber mutation blocks transcription from promoters used by at least two minor B. subtilis sigma factors, it did not alter the kinetics or general properties of the heat shock response.