Search

In contrast to its behavior as naked DNA, the MMTV promoter assembled in minichromosomes can be activated synergistically by the progesterone receptor and NF1 in a process involving ATP-dependent chromatin remodeling. The DNA-binding domain of NF1 is required and sufficient for stable occupancy of all receptor-binding sites and for functional synergism. Activation of purified minichromosomes is observed in the absence of SWI/SNF and can be enhanced by recombinant ISWI. Receptor binding to minichromosomes recruits ISWI and NURF38, but not brahma. We propose a two-step synergism in which the receptor triggers a chromatin remodeling event that facilitates access of NF1, which in turn stabilizes an open nucleosomal conformation required for efficient binding of further receptor molecules and full transactivation.

The chromatin accessibility complex (CHRAC) belongs to the class of nucleosome remodeling factors that increase the accessibility of nucleosomal DNA in an ATP-dependent manner. We found that CHRAC induces movements of intact histone octamers to neighboring DNA segments without facilitating their displacement to competing DNA or histone chaperones in trans. CHRAC-induced energy-dependent nucleosome sliding may, in principle, explain nucleosome remodeling, nucleosome positioning, and nucleosome spacing reactions known to be catalyzed by CHRAC. The catalytic core of CHRAC, the ATPase ISWI, also mobilized nucleosomes at the expense of energy. However, the directionality of the CHRAC- and ISWI-induced nucleosome movements differed drastically, indicating that the geometry of the native complex modulates the activity of its catalytic core.

The ATPase ISWI is a subunit of several distinct nucleosome remodeling complexes that increase the accessibility of DNA in chromatin. We found that the isolated ISWI protein itself was able to carry out nucleosome remodeling, nucleosome rearrangement, and chromatin assembly reactions. The ATPase activity of ISWI was stimulated by nucleosomes but not by free DNA or free histones, indicating that ISWI recognizes a specific structural feature of nucleosomes. Nucleosome remodeling, therefore, does not require a functional interaction between ISWI and the other subunits of ISWI complexes. The role of proteins associated with ISWI may be to regulate the activity of the remodeling engine or to define the physiological context within which a nucleosome remodeling reaction occurs.

The Drosophila GAGA factor binds specifically to the sequence GAGAG, and synergises with nucleosome remodelling factor to remodel chromatin in vitro. It consists of an N-terminal domain (POZ/BTB) which mediates protein-protein interactions, a central region which contains the DNA-binding domain, and a C-terminal glutamine-rich region. It is shown that the glutamine-rich region is responsible for the formation of fibres in vitro which, on the basis of their tinctorial properties and CD spectra, may be classified as amyloid fibres. A large structural change, probably resulting in beta-sheet structure, is observed upon fibre formation. Mutants containing the central region, either alone or together with the glutamine-rich region, are largely lacking in secondary structure but they bind specifically to the cognate DNA and are able to remodel chromatin in vitro. Consequently, neither the N-terminal domain nor the C-terminal glutamine-rich regions of the GAGA factor are necessary for chromatin remodelling in vitro., (Copyright 1999 Academic Press.)

To generate long arrays of nucleosomes within a topologically defined chromatin domain we have assembled minichromosomes on negatively supercoiled plasmid DNA with extracts from Drosophila preblastoderm embryos. These minichromosomes are dynamic substrates for energy-dependent nucleosome remodeling machines that facilitate the binding of various transcription factors but do not exhibit nucleosome positioning. In contrast, if such minichromosomes include the mouse mammary tumour virus (MMTV) promoter we find it wrapped around a nucleosome with similar translational and rotational position as in vivo . This structure precluded binding of NF1 to its cognate site at -75/-65 at salt concentrations between 60 and 120 mM, even in the presence of ATP, which rendered the NF1 site accessible to the restriction enzyme Hin fI. However, insertion of 30 bp just upstream of the NF1 site, which moves the site to the linker DNA, allowed ATP-dependent binding of NF1 to a fraction of the minichromosomes, even in the presence ofstoichiometric amounts of histone H1. The minichromosomes assembled in the Drosophila embryo extract reproduce important features of the native MMTV promoter chromatin and reveal differences in the ability of transcription factors and restriction enzymes to access their binding sites in positioned nucleosomes.

In this review we describe how the extract-mediated chromatin assembly system derived from preblastoderm Drosophila embryos can be modified to assemble chromatin from defined histones. This approach combines the advantages of assembling (i) chromatin templates from homogeneous histones with (ii) an assembly system that generates chromatin with physiological nucleosome spacing and density and that contains the biological complexity of in vivo chromatin. We have used this technique to assemble nonacetylated and hyperacetylated histones into chromatin (K. P. Nightingale, R. Wellinger, J. Sogo, and P. B. Becker, 1998, EMBO J. 17, 2865-2876; W. A. Krajewski and P. B. Becker, 1998, Proc. Natl. Acad. Sci. USA 95, 1540-1545), and use this as an example to detail the structural and transcriptional assays used to compare and characterize these chromatin templates. The application of this procedure to assemble chromatin from recombinant histones should facilitate a wide variety of studies on the role(s) of histone mutants and variants.

Prothymosin alpha (ProTalpha) is an abundant acidic nuclear protein that may be involved in cell proliferation. In our search for its cellular partners, we have recently found that ProTalpha binds to linker histone H1. We now provide further evidence for the physiological relevance of this interaction by immunoisolation of a histone H1-ProTalpha complex from NIH 3T3 cell extracts. A detailed analysis of the interaction between the two proteins suggests contacts between the acidic region of ProTalpha and histone H1. In the context of a physiological chromatin reconstitution reaction, the presence of ProTalpha does not affect incorporation of an amount of histone H1 sufficient to increase the nucleosome repeat length by 20 bp, but prevents association of all further H1. Consistent with this finding, a fraction of histone H1 is released when H1-containing chromatin is challenged with ProTalpha. These results imply at least two different interaction modes of H1 with chromatin, which can be distinguished by their sensitivity to ProTalpha. The properties of ProTalpha suggest a role in fine tuning the stoichiometry and/or mode of interaction of H1 with chromatin.

DNA replication is initiated by binding of initiation factors to the origin of replication. Nucleosomes are known to inhibit the access of the replication machinery to origin sequences. Recently, nucleosome remodelling factors have been identified that increase the accessibility of nucleosomal DNA to transcription regulators. To test whether the initiation of DNA replication from an origin covered by nucleosomes would also benefit from the action of nucleosome remodelling factors, we reconstituted SV40 DNA into chromatin in Drosophila embryo extracts. In the presence of T-antigen and ATP, a chromatin-associated cofactor allowed efficient replication from a nucleosomal origin in vitro. In search of the energy-dependent cofactor responsible we found that purified 'chromatin accessibility complex' (CHRAC) was able to alter the nucleosomal structure at the origin allowing the binding of T-antigen and efficient initiation of replication. These experiments provide evidence for the involvement of a nucleosome remodelling machine in structural changes at the SV40 origin of DNA replication in vitro.

Chromatin has shifted into the focus of attention as a key to understanding the regulation of nuclear processes such as transcription. Protein machines have been described that use the energy of ATP to render chromatin dynamic and hence active, but which may also be involved in chromatin assembly. The discovery of three different Drosophila nucleosome remodeling complexes that contain imitation switch (ISWI), an ATPase with a high degree of sequence conservation from yeast to human, points to a central function of this ATPase in chromatin dynamics.

Transcription of ribosomal genes assembled into chromatin requires binding of the transcription termination factor TTF-I to the promoter-proximal terminator T0. To analyze the mechanism of TTF-I-mediated transcriptional activation, we have used mutant templates with altered sequence, polarity and distance of T0 with respect to the transcription start site. Transcription activation by TTF-I is chromatin specific and requires the precise positioning of the terminator relative to the promoter. Whereas termination by TTF-I depends on the correct orientation of a terminator, TTF-I-mediated transcriptional activation is orientation independent. TTF-I can bind to nucleosomal DNA in the absence of enzymatic activities that destabilize nucleosome structure. Chromatin-bound TTF-I synergizes with ATP-dependent cofactors present in extracts of Drosophila embryos and mouse cells to position a nucleosome over the rDNA promoter and the transcription start site. Nucleosome positioning correlates tightly with the activation of rDNA transcription. We suggest that transcriptional activation by TTF-I is a stepwise process involving the creation of a defined promoter architecture and that the positioning of a nucleosome is compatible with, if not a prerequisite for, transcription initiation from rDNA chromatin.

A number of activators are known to increase transcription by RNA polymerase (pol) II through protein acetylation. While the physiological substrates for those acetylases are poorly defined, possible targets include general transcription factors, activator proteins and histones. Using a cell-free system to reconstitute chromatin with increased histone acetylation levels, we directly tested for a causal role of histone acetylation in transcription by RNA pol II. Chromatin, containing either control or acetylated histones, was reconstituted to comparable nucleosome densities and characterized by electron microscopy after psoralen cross-linking as well as by in vitro transcription. While H1-containing control chromatin severely repressed transcription of our model hsp26 gene, highly acetylated chromatin was significantly less repressive. Acetylation of histones, and particularly of histone H4, affected transcription at the level of initiation. Monitoring the ability of the transcription machinery to associate with the promoter in chromatin, we found that heat shock factor, a crucial regulator of heat shock gene transcription, profited most from histone acetylation. These experiments demonstrate that histone acetylation can modulate activator access to their target sites in chromatin, and provide a causal link between histone acetylation and enhanced transcription initiation of RNA pol II in chromatin.

Increased acetylation at specific N-terminal lysines of core histones is a hallmark of active chromatin in vivo, yet the structural consequences of acetylation leading to increased gene activity are only poorly defined. We employed a new approach to characterize the effects of histone acetylation: A Drosophila embryo-derived cell-free system for chromatin reconstitution under physiological conditions was programmed with exogenous histones to assemble hyperacetylated or matching control chromatin of high complexity. Hyperacetylated chromatin resembled unmodified chromatin at similar nucleosome density with respect to its sensitivity toward microccal nuclease, its nucleosomal repeat length, and the incorporation of the linker histone H1. In contrast, DNA in acetylated chromatin showed an increased sensitivity toward DNase I and a surprisingly high degree of conformational flexibility upon temperature shift pointing to profound alterations of DNA/histone interactions. This successful reconstitution of accessible and flexible chromatin outside of a nucleus paves the way for a thorough analysis of the causal relationship between histone acetylation and gene function.

Transcription by RNA polymerase II is highly regulated at the level of initiation and elongation. Well-documented transcription activation mechanisms, such as the recruitment of TFIID and TFIIB, control the early phases of preinitiation complex formation. The heat shock genes provide an example for transcriptional regulation at a later step: in nuclei TFIID can be detected at the TATA box prior to heat induction. Using cell-free systems for chromatin reconstitution and transcription, we have analyzed the mechanisms by which heat shock factor (HSF) increases transcription of heat shock genes in chromatin. HSF affected transcription of naked DNA templates in multiple ways: (i) by speeding up the rate of preinitiation complex formation, (ii) by increasing the number of productive templates, and (iii) by increasing the reinitiation rate. Under the more physiological conditions of potentiated chromatin templates, HSF affected only the reinitiation rate. Activator-dependent reinitiation of transcription, obviating the slow assembly of the TFIID-TFIIA complex on a promoter, may be especially crucial for genes requiring a fast response to inducers.

To restore full genomic integrity in a eukaryotic cell, DNA repair processes have to be coordinated with the resetting of nucleosomal organization. We have established a cell-free system using Drosophila embryo extracts to investigate the mechanism linking de novo nucleosome formation to nucleotide excision repair (NER). Closed-circular DNA containing a uniquely placed cisplatin-DNA adduct was used to follow chromatin assembly specifically from a site of NER. Nucleosome formation was initiated from a target site for NER. The assembly of nucleosomes propagated bidirectionally, creating a regular nucleosomal array extending beyond the initiation site. Furthermore, this chromatin assembly was still effective when the repair synthesis step in the NER process was inhibited.

Repressive chromatin structures need to be unravelled to allow DNA-binding proteins access to their target sequences. This de-repression constitutes an important point at which transcription and presumably other nuclear processes can be regulated. Energy-consuming enzyme complexes that facilitate the interaction of transcription factors with chromatin by modifying nucleosome structure are involved in this regulation. One such factor, nucleosome-remodelling factor (NURF), has been isolated from Drosophila embryo extracts. We have now identified a chromatin-accessibility complex (CHRAC) which uses energy to increase the general accessibility of DNA in chromatin. However, unlike other known chromatin remodelling factors, CHRAC can also function during chromatin assembly: it uses ATP to convert irregular chromatin into a regular array of nucleosomes with even spacing. CHRAC combines enzymes that modulate nucleosome structure and DNA topology. Using mass spectrometry, we identified two of the five CHRAC subunits as the ATPase ISWI, which is also part of NURF, and topoisomerase II. The presence of ISWI in different contexts suggests that chromatin remodelling machines have a modular nature and that ISWI has a central role in different chromatin remodelling reactions.

The biochemical analysis of chromatin structure and function is greatly facilitated by the availability of cell-free systems that assemble chromatin under physiological conditions. One such system that has shown great potential is derived from extracts of early Drosophila embryos. These embryos contain large maternal stocks of chromatin constituents, such as histones and assembly factors. Chromatin assembled in these extracts resembles native chromatin in many respects: it displays physiological nucleosome repeat lengths, it is complex, containing a wealth of nonhistone proteins as well as enzymatic activities, and it has dynamic properties that allow the interaction of DNA-binding proteins that regulate important cellular processes. Most importantly, chromatin with variant properties, e.g., with respect to the basic geometry of the nucleosomal array, histone modifications, and its content of linker histones or nonhistone proteins, can be obtained by manipulating the reconstitution conditions. The synthesis of uniform chromatin with specific characteristics should allow the analysis of the functional significance of the structural and biochemical heterogeneity observed in vivo.

Eukaryotic ribosomal gene promoters are preceded by a terminator element which is recognized by the transcription termination factor TTF-I. We have studied the function of this promoter-proximal terminator and show that binding of TTF-I is the key event which leads to ATP-dependent nucleosome remodeling and transcriptional activation of mouse rDNA pre-assembled into chromatin. We have analyzed TTF-I mutants for their ability to bind to free or nucleosomal DNA, and show that the DNA binding domain of TTF-I on its own is not sufficient for interaction with chromatin, indicating that specific protein features exist that endow a transcription factor with chromatin binding and remodeling properties. This first analysis of RNA polymerase I transcription in chromatin provides a clue for the function of the upstream terminator and establishes a dual role for TTF-I both as a termination factor and a chromatin-specific transcription activator.

The unmatched power of Drosophila genetics revealed the complex regulatory network of gene activities that governs the development of higher eukaryotes. An understanding of gene control at the level of transcription requires insight into the protein/DNA interactions that regulate transcription in the developing embryo. Genomic footprinting allows the direct visualization of these protein/DNA interactions within intact nuclei or cells. In combination with other in vivo assays such as protein/DNA crosslinking and classical biochemistry, genomic footprinting can give valuable insight into the architecture of promoters in various states of activity. In this article we summarize our experience in analyzing Drosophila embryos by genomic footprinting and describe modifications of the ligation-mediated PCR procedure that have improved this analysis. Applications of genomic footprinting to embryos are currently limited by the fact that all target nuclei must be uniform with respect to the protein/DNA interactions at the chosen site. We discuss strategies that should allow the analysis of small numbers of cells derived from heterogeneous populations and tissues.

For the analysis of protein-DNA interactions by coupled gel-shift/footprinting, DNA fragments need to be extracted from polyacrylamide gels and subsequently separated on high resolution gels. Due to impurities in the extracted DNA, single nucleotide resolution is frequently not achieved. We now describe an improved experimental strategy that employs transient coupling of DNA fragments to a solid support in order to extract DNA of high purity quantitatively, rapidly and reliably. As an example, we describe the application of our protocol to the 'in-gel footprinting' by copper phenanthroline. The method should also find application to the chemical interference assays.

The hepatocyte nuclear factor-1 (HNF1) is a homeodomain transcription factor that binds DNA as a dimer. HNF1 dimers associate with two molecules of DCoH, a bifunctional protein that also has an enzymatic function in the tetrahydrobiopterin regeneration, to form stable heterotetramers also capable of DNA binding. Employing purified, recombinant HNF1, HNF1/DCoH heterotetramers and DCoH homotetramers we investigated whether DCoH affects interactions of HNF1 with nucleic acids. Although we detected no direct binding of DCoH to DNA or RNA, DCoH stabilized HNF1/DNA complexes and promoted interactions with sub-optimal DNA target sequences such as the human alpha1-antitrypsin TATA box region. Importantly, we also observed interactions of HNF1 with RNA, but these interactions were completely abolished when HNF1 was complexed with DCoH. Interestingly, DCoH retains its enzymatic activity while complexed with HNF1. Our results document intermolecular regulation of HNF1 binding to nucleic acids by DCoH.

The distances between the nucleosomes in eukaryotic chromatin that define the nucleosome repeat length are not universally constant, but vary between different cell types and activity states. We have previously established in a cell-free system that nucleosome spacing is essentially governed by electrostatic principles, most likely through charge neutralisation of linker DNA by cations either free in solution or on flexible histone domains. On the basis of the tight correlation between the parameters that affect nucleosome spacing and those that influence the folding of the nucleosomal fiber into higher order structures, we suggested that there is an intimate relationship between nucleosome spacing and chromatin folding. Here we describe DNA topology as a new parameter that influences nucleosome spacing in a predictable way. The effects of topology and cation concentrations integrate to define the final repeat length. The phenomenon of "nucleosome phasing" describes nucleosomal arrays that are generated through positioning of nucleosomes by the underlying DNA sequence. To determine the relative contribution of DNA sequence and the parameters intrinsic to physiological chromatin for nucleosomal positions, we created situations where these two principles were in conflict. We found that nucleosome repeats directed by a strong positioning sequence are dominated by the cation-induced spacing as well as by the effects of topology. We conclude that the DNA sequence effects nucleosome spacing only by "fine tuning" of nucleosome positions within the framework of a repeat pattern that is established by other principles.

Transcriptional activation of the Drosophila hsp27 gene in response to heat shock critically relies on binding sites for heat shock factor (HSF) about 300 bp upstream of the transcription start site. In contrast to the well-characterised heat-inducible hsp70 and hsp26 promoters, no other transcription factor binding sites have been identified closer to the TATA box. In order to understand the structural requirements for activation from a distance we studied the protein-DNA interactions at the hsp27 promoter in Drosophila embryos and tissue culture cells before and after heat induction. Genomic footprinting with nucleases and a chemical probe, the 1,10-phenanthroline cuprous complex (OP-Cu), suggests that the DNA between the TATA box and the heat shock elements (HSEs) is constitutively organised by a positioned nucleosome, effectively shortening the distance between the distal HSEs and the TATA box. Protection of the TATA element from nuclease attack and the OP-Cu reactivity pattern around the start site of transcription is consistent with the constitutive presence of TFIID and the "poised polymerase", a transcription machinery blocked in an early phase of elongation. The general transcription factors at the TATA box and the positioned nucleosome are separated by a stable structure, presumably a protein bound to a palindromic sequence. These constitutive features define the "preset" architecture of the promoter within which the induced binding of HSF in vivo is observed. Our study highlights the importance of positioned nucleosomes as architectural elements within promoters and identifies a new regulatory sequence that may function either to direct a nucleosome boundary or to mediate signals of distant activator proteins.

Native bulk chromatin is characterized by regular arrays of nucleosomes with defined internucleosomal distances. The nucleosome repeat length is not a constant but varies between species and cell-types, during differentiation and during gene activation. Previous studies have highlighted the importance of linker histones as a major determinant of nucleosome repeat length in vivo. We used a physiological reconstitution system derived from Drosophila embryos to study nucleosome spacing. In these extracts, histone H1 incorporation increases the apparent linker length in a gradual way. Manipulation of the chromatin assembly conditions in vitro allowed us to define additional parameters that modulate nucleosomal distances, such as protein phosphorylation events and the precise ionic conditions during the reconstitution. Interestingly, moderate changes in the concentrations of mono-, di-, and multivalent cations affect the precise distances between nucleosome cores remarkably. These changes in the ionic environment are unlikely to affect the association of linker proteins but are known to influence the folding of the nucleosomal fiber by modulation of electrostatic forces. Our results suggest electrostatic interactions in chromatin units as major determinants of nucleosome spacing. Nucleosome spacing and the folding of the nucleosomal fiber can therefore be explained by common principles, most notably the neutralization of charges in linker DNA.

The association of DNA with nucleosomes in chromatin severely restricts the access of the regulatory factors that bring about transcription. In vivo active promoters are characterised by altered, almost transparent chromatin structures that allow the interaction of the transcriptional machinery. Recently, enzymatic activities have been discovered that facilitate the binding of transcription factors to chromatin by modifying nucleosomal structures in a process that requires energy. The mechanisms by which chromatin is remodelled may involve nucleosome movements, their transient unfolding, their partial or even complete disassembly. The dynamic properties of chromatin that underlie these structural changes are fundamental to the process of regulated gene expression.

In Drosophila transcription is differentially repressed by various aspects of chromatin, thought to represent distinct structural levels: the nucleosome core particle, the linker histone H1 and as yet undefined higher order structures. Heterochromatin serves as a paradigm for the latter level of organization, but maintenance of silencing at homeotic gene loci may also fall into this class. Recently GAGA transcription factor (GAF) was shown to counteract chromatin repression at all levels. The various effects of GAF could be explained by its ability to rearrange nucleosomal positions. Chromatin remodelling by GAF and other factors in vitro require activities that maintain a highly dynamic state of chromatin.

Efficient heat shock induction of Drosophila hsp26 gene transcription in vivo requires binding sites for heat shock factor (HSF) and GAGA factor (GAF) close to the TATA box (proximal elements) as well as 350 bp upstream of the start site of transcription (distal elements). We have evaluated the contribution of hsp26 promoter sequences to transcriptional activity in extracts from either heat shocked or unstressed fly embryos. Efficient transcription in either extract was governed by distinct regulatory principles. Transcription in extracts from unstressed embryos relied solely on GAGA elements which efficiently counteracted repression by abundant non-specific DNA-binding proteins. Transcription in extracts from heat shocked embryos depended only a little on GAGA elements, relying mainly on functional HSEs. Constitutively active recombinant HSF or native factor in an extract from heat shocked embryos was able to truly activate transcription essentially via proximal HSEs, but not when bound to distal sites. These two modes of regulation in vitro may correspond to the two functional states of the promoter before and after heat shock in vivo.

Chromatin structure must be flexible to allow the binding of regulatory proteins and to accommodate different levels of gene activity. Chromatin assembled in a cell-free system derived from Drosophila embryos contains an activity that hydrolyses ATP to render entire nucleosome arrays mobile. Nucleosome movements, most likely their sliding, occurred even in the presence of the linker histone H1. The dynamic state of chromatin in the presence of the activity and ATP globally increased the accessibility of nucleosomal DNA to incoming proteins. This unprecedented demonstration of energy-dependent nucleosome mobility identifies a new principle which is likely to be fundamental to the mechanism of chromatin remodelling and the binding of regulatory proteins.

The chromatin structure at the Drosophila hsp26 promoter in vivo is characterized by two DNase I-hypersensitive (DH) sites harboring regulatory elements. Proximal and distal DH sites are separated by a positioned nucleosome. To study the contribution of transcription factors to the establishment of this specific chromatin configuration we assembled nucleosomes on the hsp26 promoter using a cell-free reconstitution system derived from fly embryos. Both DH sites were readily reconstituted from extract components. They were separated by a nucleosome which was less strictly positioned than its in vivo counterpart. The interactions of GAGA factor and heat shock factor with their binding sites in chromatin occurred in two modes. Their interaction with binding sites in the nucleosome-free regions did not require ATP. In the presence of ATP both factors interacted also with nucleosomal binding sites, causing nucleosome rearrangements and a refinement of nucleosome positions. While chromatin remodeling upon transcription factor interaction has previously been interpreted to involve nucleosome disruption, the data suggest energy-dependent nucleosome sliding as main principle of chromatin reorganization.

The organization of eukaryotic genomes as chromatin provides the framework within which regulated transcription occurs in the nucleus. The association of DNA with chromatin proteins required to package the genome into the nucleus is, in general, inhibitory to transcription, and therefore provides opportunities for regulated transcriptional activation. Granting access to the cis-acting elements in DNA, a prerequisite for any further action of the trans-acting factors involved, requires the establishment of local heterogeneity of chromatin and, in some cases, extensive remodeling of nucleosomal structures. Challenging problems relate to the establishment of this heterogeneity at the level of the single nucleosome and to the mechanisms that operate when nucleosomal arrays are reorganized. Recent developments indicate that chromatin reconstitution in cell-free systems allows the biochemical analysis of the interplay between transcription factors and chromatin components that brings about regulated transcription.

Genetic control elements are usually situated in local regions of chromatin that are hypersensitive to structural probes such as DNase I. We have reconstructed the chromatin structure of the hsp70 promoter using an in vitro nucleosome assembly system. Binding of the GAGA transcription factor on existing nucleosomes leads to nucleosome disruption, DNase I hypersensitivity at the TATA box and heat-shock elements, and rearrangement of adjacent nucleosomes. ATP hydrolysis facilitates this process, suggesting that an energy-dependent pathway is involved in chromatin remodelling.

Chromatin reconstituted in an extract from preblastoderm Drosophila embryos represses transcription by RNA polymerase II. We have assembled regularly spaced nucleosomes on DNA attached to paramagnetic beads enabling the efficient purification of chromatin templates for transcription studies. We have used diagnostic salt extractions to establish that transcriptional repression of immobilized chromatin was largely due to nucleosome cores. When purified H1 was incorporated into chromatin, resulting in increased repeat lengths to 200-220 bp, the contribution of H1 to transcriptional repression was negligible. If more H1 was added no regularly spaced chromatin was obtained and only under these conditions was transcriptional inhibition by H1 apparent. We conclude that efficient repression of transcription by polymerase II in this system does not require the presence of histone H1.

We describe a cell-free system, derived from preblastoderm Drosophila embryos, for the efficient assembly of cloned DNA into chromatin. The chromatin assembly system utilizes endogenous core histones and assembly factors and yields long arrays of regularly spaced nucleosomes with a repeat length of 180 bp. The assembly system is also capable of complementary-strand DNA synthesis accompanied by rapid nucleosome formation when the starting template is single-stranded circular DNA. Chromatin assembled with the preblastoderm embryo extract is naturally deficient in histone H1, but exogenous H1 can be incorporated during nucleosome assembly in vitro. Regular spacing of nucleosomes with or without histone H1 is sufficient to maximally repress transcription from hsp70 and fushi tarazu gene promoters. The Drosophila assembly system should be particularly useful for in vitro studies of chromatin assembly during DNA synthesis and for elucidating the action of transcription factors in the context of native chromatin.

Specific DNA sequences from several DNase I hypersensitive sites located upstream of the tyrosine aminotransferase (TAT) gene are bound by ubiquitous nuclear factors in vitro. Genomic footprinting has shown, however, that proteins are excluded from their potential binding sites in cells where the gene is inactive and that the absence of in vivo footprints is correlated with CpG methylation and altered chromatin structures at these sites. In vitro, interactions of proteins with sequences of the TAT gene, including binding of the transcription factor CREB to the cAMP-responsive element (CRE), are prevented by a methylated CpG dinucleotide in the respective binding sites, suggesting that methylation of DNA might be sufficient to exclude proteins from their sites in vivo. To test directly whether the absence of in vivo footprints is the result of DNA methylation, we treated two different cell lines with 5-azacytidine to demethylate CpG dinucleotides. While genomic sequencing confirmed demethylation at two widely separated regions upstream of the TAT promoter, no footprints appeared in these cell lines, even though proteins capable of binding these sites in vitro were present in the nuclei. Thus, the simple model whereby protein exclusion in vivo is caused solely by DNA methylation is not appropriate in this case. The nucleosomal organization of the potential binding sites suggests that chromatin structure is a dominant determinant in maintaining the inactive state of these sites.

To investigate the mechanisms of transcriptional regulation of Drosophila heat shock genes we studied the activity of a heat shock promoter in vitro after reconstitution into chromatin. Increasing the duration of nucleosome assembly progressively inactivated a plasmid template when it was transcribed with extracts of either unshocked or heat-shocked Drosophila embryos, despite induction of the transcriptional activator heat shock factor. Addition of the general transcription factor IID (TFIID) before nucleosome assembly did not significantly relieve nucleosomal inhibition, but TFIID potentiated the promoter to be responsive to activation by heat shock factor in the heat shock transcription extract. The potentiation by TFIID could be related to the nucleosome-free, hypersensitive state of heat shock promoters previously observed in vivo before heat shock induction and may be necessitated by the need to expedite activation of heat shock genes in response to environmental stress.

We report the cloning of the transcriptional activator of heat shock genes, HSF, from Drosophila. The predicted sequence of Drosophila HSF protein is surprisingly divergent from that of yeast HSF, except in regions important for DNA binding and oligomerization. A segment of the DNA binding domain of HSF bears an intriguing similarity to the putative DNA recognition helix of bacterial sigma factors, while the oligomerization domain contains an unusual arrangement of conserved hydrophobic heptad repeats. Drosophila HSF produced in E. coli under nonshock conditions forms a hexamer that binds specifically to DNA with high affinity and activates transcription from a heat shock promoter in vitro. In contrast, when HSF is expressed in Xenopus oocytes, maximal DNA binding affinity is observed only after heat shock induction. These results suggest that Drosophila HSF has an intrinsic affinity for DNA, which is repressed under nonshock conditions in vivo.

Transcriptional activation of gene expression by glucocorticoid hormones is mediated by the interaction of hormone-receptor complexes with specific DNA sequences called glucocorticoid responsive elements (GREs) (refs 1-3, see ref. 4 for review). Deletion of this sequence abolishes glucocorticoid induction of transcription. According to a current model, activation of the cytoplasmic receptor protein by hormone binding leads to its increased affinity for and translocation to the nucleus. However, recent reports that the oestradiol and progesterone receptors are localized in the nucleus in the absence of steroid led us to examine whether the free receptor interacts in vivo with its DNA binding site in the absence of hormone binding. We used the genomic footprinting technique to show that changes in in vivo protein-DNA interactions within the GREs of the tyrosine aminotransferase gene (TAT) can be detected only after hormone treatment in hepatoma cells. Such changes are not detected in fibroblast cells, in which the TAT gene is not expressed. Many of the changes in dimethylsulphate reactivity observed in the living cell are also found in vitro using cloned DNA and a partially purified glucocorticoid receptor.

Using in vivo dimethylsulfate footprinting, we have analyzed protein-DNA interactions within two regions upstream of the tyrosine aminotransferase (TAT) gene that are characterized by an altered chromatin structure in TAT-expressing as compared to nonexpressing cells. All the identified protein contacts to DNA are found exclusively in the TAT-expressing hepatoma cells. In vitro analyses of specific DNA-binding factors in crude nuclear extracts yield DNAase I footprints that correlate well with the binding sites in vivo. Surprisingly, all DNA-binding activities are present in nuclei of TAT-expressing and nonexpressing cells, indicating that the mere presence of factors is not sufficient for their interaction with a binding site in vivo. Genomic sequencing reveals methylation of CpG dinucleotides in the regions analyzed in nonexpressing cells, whereas no methylation is found in TAT-expressing cells. In vitro methylation at a cytosine residue within a footprint region prevents the interaction of a factor with its binding site.