Perturbation of nucleosome core structure by the SWI/SNF complex persists after its detachment, enhancing subsequent transcription factor binding.

To investigate the mechanism of SWI/SNF action, we have analyzed the pathway by which SWI/SNF stimulates formation of transcription factor-bound nucleosome core complexes. We report here that the SWI/SNF complex binds directly to nucleosome cores and uses the energy of ATP hydrolysis to disrupt histone/DNA interactions, altering the preferred path of DNA bending around the histone octamer. This disruption occurs without dissociating the DNA from the surface of the histone octamer. ATP-dependent disruption of nucleosomal DNA by SWI/SNF generates an altered nucleosome core conformation that can persist for an extended period after detachment of the SWI/SNF complex. This disrupted conformation retains an enhanced affinity for the transcription factor GAL4-AH. Thus, ATP-dependent nucleosome core disruption and enhanced binding of the transcription factor can be temporally separated. These results indicate that SWI/SNF can act transiently in the remodeling of chromatin structure, even before interactions of transcription factors.

Long-range effects in a supercoiled DNA domain generated by transcription in vitro.

The translocation of a transcription complex can transiently introduce positive and negative superhelical windings into the template DNA. To gain further insight into this dynamic DNA supercoiling mechanism and its possible involvement in biological processes, we employed an in vitro system in which site-specific recombination by gammadelta resolvase is topologically coupled to transcription-induced negative supercoiling. Our kinetic experiments suggest that recombination is closely linked to the process of supercoiling by transcription. We utilized the known high speed at which two resolvase-bound recombination sites can pair to form a synaptic complex in kinetic experiments with DNA substrates containing three recombination sites. Our data provide evidence for the existence of a transient gradient of negative supercoiling. Such a gradient seems to be predominantly a consequence of DNA double helix rotation behind a translocating RNA polymerase and originates within a broad region up to two kilobase-pairs upstream of the transcriptional start site. We further demonstrate that the topological coupling between transcription and recombination is not affected when the DNA-bending protein integration host factor from E. coli is bound to multiple sites within the phage lambda attachment region. We discuss implications of our in vitro findings with respect to possible in vivo functions of the dynamic nature of transcription-induced supercoiling.

Engineering of diffraction-quality crystals of the NF-kappaB P52 homodimer:DNA complex.

The eukaryotic transcription factors NF-kappaB P50 and NF-kappaB P52 are closely related members of the Rel family. Growth of diffraction-quality NF-kappaB P52:DNA co-crystals crucially depended on (a) extensive screens for the DNA fragment of optimal length and (b) engineering of the protein based on the two known NF-kappaB P50:DNA co-crystal structures [Müller et al. (1995) Nature 373, 311-317; Ghosh et al. (1995) Nature 373, 303-310]; namely, deletion of 12 C-terminal amino acid residues. These residues are part of the Rel homology region and comprise the nuclear localization signal. The approach might be of general use for the crystallization of other Rel protein: DNA complexes and in our case yielded co-crystals which diffract beyond 2.0 A resolution.

Bacillus subtilis 168 RecR protein-DNA complexes visualized as looped structures.

The Bacillus subtilis 168 RecR protein bound to duplex DNA in the presence of ATP and divalent cations (Mg2+ and Zn2+) was visualized by electron microscopy as a nearly spherical particle. A RecR homomultimer is frequently located at the intersection of two duplex DNA strands in an interwound DNA molecule, generating DNA loops of variable length. Two individual DNA molecules bound to the same protein are seen at a very low frequency, if at all. The association of RecR with the intersection of two duplex DNA strands is more often seen in supercoiled than with relaxed or linear DNA. The RecR protein displays a slight but significant preference for negatively supercoiled over linear DNA. The minimum substrate size for RecR protein is about 150 bp in length. A possible mechanism for RecR function in DNA repair is discussed.

Maize chromosomal HMGc. Two closely related structure-specific DNA-binding proteins specify a second type of plant high mobility group box protein.

The chromosomal high mobility group (HMG) proteins are small and abundant non-histone proteins common to eukaryotes. We have purified the maize HMGc protein from immature kernels and characterized it by mass spectrometry and amino acid sequence analysis. HMGc could be resolved into two similar proteins by reversed phase chromatography. Cloning and characterization of the corresponding cDNAs revealed that they encode two closely related maize HMGc proteins, now termed HMGc1 and HMGc2. Their theoretical masses of 15,316 and 15,007 Da are >300 Da lower than the masses determined for the proteins purified from maize, indicating post-translational modifications of the proteins. Despite sequence similarity to maize HMGa (and previously described homologous proteins of other species) amino acid sequence alignments reveal that HMGc is in several conserved regions distinct from these proteins. Consequently, we have identified a novel type of plant protein containing an HMG box DNA binding domain and belonging to the HMG1 protein family. HMGc1 and HMGc2 were expressed in Escherichia coli, purified to homogeneity, and analyzed for their DNA binding properties. They proved to bind to DNA structure-specifically since they formed complexes with DNA minicircles at concentrations approximately 100-fold lower than the concentrations required to form complexes with linear fragments of identical sequence. Furthermore, HMGc1 and HMGc2 can constrain negative superhelical turns in plasmid DNA.

Mapping the contacts of yeast TFIIIB and RNA polymerase III at various distances from the major groove of DNA by DNA photoaffinity labeling.

The structure of the Saccharomyces cerevisiae RNA polymerase III transcription complex on the SUP4 tRNATyr gene was probed at distances of approximately 10 to approximately 23 A from the C-5 methyl of thymidine in the major groove of DNA using photoreactive aryl azides attached to deoxyuridine by variable chain lengths. The nucleotide analogs contained an azidobenzoyl group attached with chain lengths that were incrementally increased by approximately 4. 3 A by inserting 1-3 glycine residues into the chain. Another photoreactive deoxyuridine analog was made that contained a butyl chain (ABU-dUMP) to assess the effect of the chain's hydrophobicity on its ability to photoaffinity label the transcription complex. These nucleotide analogs were incorporated at base pairs (bp) -26/-21, -17, or -3/-2 on the nontranscribed strand of the SUP4 tRNATyr gene along with an [alpha-32P]dNMP by primer extension using an immobilized single-stranded DNA template annealed to specific oligonucleotides. The 27-kDa subunit of TFIIIB or the TATA box binding protein was photoaffinity labeled at bp -26/-21 with nucleotide analogs containing a approximately 19- or approximately 23-A chain and not with shorter chains of approximately 10 to approximately 15 A in length. The B" subunit of TFIIIB (Mr = 90 kDa) was photoaffinity labeled at bps -26/-21 with DNA containing a approximately 14-A chain and not with shorter or longer chains. Cross-linking of the B" subunit was inhibited by binding of RNA polymerase III (Pol III) to the TFIIIB-DNA complex and suggested that Pol III binding causes a conformational change in the TFIIIB-DNA complex resulting in the displacement of the 90-kDa subunit at bps -26/-21. Next, the chain length dependence of photoaffinity labeling the 34-kDa subunit of Pol III at bps -17 and -3/-2 indicated that the 34-kDa subunit of Pol III is slightly removed from the major groove at bp -17 in the initiation complex and makes closer contact at bps -3/-2 in a stalled elongation complex.

Crystal structure of an IHF-DNA complex: a protein-induced DNA U-turn.

Integration host factor (IHF) is a small heterodimeric protein that specifically binds to DNA and functions as an architectural factor in many cellular processes in prokaryotes. Here, we report the crystal structure of IHF complexed with 35 bp of DNA. The DNA is wrapped around the protein and bent by >160 degrees, thus reversing the direction of the helix axis within a very short distance. Much of the bending occurs at two large kinks where the base stacking is interrupted by intercalation of a proline residue. IHF contacts the DNA exclusively via the phosphodiester backbone and the minor groove and relies heavily on indirect readout to recognize its binding sequence. One such readout involves a six-base A tract, providing evidence for the importance of a narrow minor groove.

High-resolution mapping of nucleoprotein complexes by site-specific protein-DNA photocrosslinking: organization of the human TBP-TFIIA-TFIIB-DNA quaternary complex.

We have used a novel site-specific protein-DNA photocrosslinking procedure to define the positions of polypeptide chains relative to promoter DNA in binary, ternary, and quaternary complexes containing human TATA-binding protein, human or yeast transcription factor IIA (TFIIA), human transcription factor IIB (TFIIB), and promoter DNA. The results indicate that TFIIA and TFIIB make more extensive interactions with promoter DNA than previously anticipated. TATA-binding protein, TFIIA, and TFIIB surround promoter DNA for two turns of DNA helix and thus may form a "cylindrical clamp" effectively topologically linked to promoter DNA. Our results have implications for the energetics, DNA-sequence-specificity, and pathway of assembly of eukaryotic transcription complexes.

Structure of the replication terminus-terminator protein complex as probed by affinity cleavage.

The replication terminator protein (RTP) of Bacillus subtilis is a homodimer that binds to each replication terminus and impedes replication fork movement in only one orientation with respect to the replication origin. The three-dimensional structure of the RTP-DNA complex needs to be determined to understand how structurally symmetrical dimers of RTP generate functional asymmetry. The functional unit of each replication terminus of Bacillus subtilis consists of four turns of DNA complexed with two interacting dimers of RTP. Although the crystal structure of the RTP apoprotein dimer has been determined at 2.6-A resolution, the functional unit of the terminus is probably too large and too flexible to lend itself to cocrystallization. We have therefore used an alternative strategy to delineate the three dimensional structure of the RTP-DNA complex by converting the protein into a site-directed chemical nuclease. From the pattern of base-specific cleavage of the terminus DNA by the chemical nuclease, we have mapped the amino acid to base contacts. Using these contacts as distance constraints, with the crystal structure of RTP, we have constructed a model of the DNA-protein complex. The biological implications of the model have been discussed.

Estrogen receptor-induced DNA bending: orientation of the bend and replacement of an estrogen response element with an intrinsic DNA bending sequence.

The estrogen receptor (ER) binds to DNA fragments containing estrogen response elements (EREs) and causes them to bend. To characterize this ER-induced DNA bend and determine if it is involved in transcription activation, three different lines of investigation were used. Using MCF-7 human breast cancer cell extracts and circular permutation analysis, it was determined that molybdate-stabilized, unoccupied cytosolic ER was unable to bind to ERE-containing DNA fragments when maintained at 4 C, but that thermal activation enabled the cytosolic receptor to bind and bend ERE-containing DNA fragments to the same extent as ER present in whole cell extracts. DNA phasing analysis was utilized to determine that ER binding induced DNA fragments containing EREs to bend toward the major groove of the DNA helix. The orientation of this bend was the same with thermally activated, unoccupied cytosolic ER and with unoccupied ER, 17 beta-estradiol-occupied ER, and 4-hydroxytamoxifen-occupied ER present in whole cell extracts. Using transient cotransfection assays, the ability of an intrinsically bent DNA sequence to replace an ERE was tested. When a single consensus ERE, which is induced to bend 56 degrees on ER binding, was replaced with a 54 degrees intrinsic DNA bending sequence, transcription was effectively activated. Similar levels of transcription were also observed when promoters contained either a 108 degrees intrinsic DNA bending sequence or two consensus EREs. However, the 54 degrees DNA bending sequence and a single ERE were unable to cooperatively activate transcription. Because the magnitude and orientation of ER-induced DNA bends are the same with the unoccupied and occupied receptor, DNA bending alone probably does not function as a transcriptional switch to turn on gene transcription. However, DNA bending may be required to provide the architecture needed for modulation of target genes.

The use of monoclonal antibodies for studying intermediates in DNA repair by the Escherichia coli Uvr(A)BC endonuclease.

The Escherichia coli Uvr(A)BC endonuclease acts in a progression of several distinct steps accompanied by changes in the conformation of macromolecular constituents, the overall architecture of the complex, and its stoichiometry. In order to probe these structural changes, we generated monoclonal antibodies (mAbs) to Uvr proteins. The anti-UvrA mAb, A2D1, recognizing the N-terminal zinc-finger region of UvrA, and the anti-UvrB mAb, B2E2, having an epitope within the 43 C-terminal amino acids of UvrB, were purified and further characterized. It was found that A2D1 mAb interacts in solution both with UvrA-UvrB and UvrA-DNA complexes in the presence of the requisite ATP. This implies that the N-terminal zinc-finger of UvrA doesn't play a direct role in its interactions with UvrB and DNA. On the other hand, A2D1 does inhibit formation of the UvrB-damaged DNA preincision complex, apparently by preventing UvrB delivery by UvrA. The interaction of B2E2 with UvrA-UvrB and nucleoprotein complexes, including UvrB, suggests that the highly hydrophobic C-terminal domain of UvrB (i) doesn't participate in its interaction with UvrA, (ii) is accessible to this mAb in an intermediate UvrA-UvrB DNA helix-tracking complex, and (iii) seems to be directly involved in the formation of the preincision complex. These conclusions are supported by the finding that the neutralizing effect of A2D1 and B2E2 on the Uvr(A)BC endonuclease is significantly decreased if the preincision complex is preformed prior to mAbs addition.

Crystal structure of the Oct-1 POU domain bound to an octamer site: DNA recognition with tethered DNA-binding modules.

The structure of an Oct-1 POU domain-octamer DNA complex has been solved at 3.0 A resolution. The POU-specific domain contacts the 5' half of this site (ATGCAAAT), and as predicted from nuclear magnetic resonance studies, the structure, docking, and contacts are remarkably similar to those of the lambda and 434 repressors. The POU homeodomain contacts the 3' half of this site (ATGCAAAT), and the docking is similar to that of the engrailed, MAT alpha 2, and Antennapedia homeodomains. The linker region is not visible and there are no protein-protein contacts between the domains, but overlapping phosphate contacts near the center of the octamer site may favor cooperative binding. This novel arrangement raises important questions about cooperativity in protein-DNA recognition.

Structure and function of RecA-DNA complexes.

While the E. coli RecA protein has been the most intensively studied enzyme of homologous recombination, the unusual RecA-DNA filament has stood alone until very recently. It now appears that this protein is part of a universal family that spans all of biology, and the filament that is formed by the protein on DNA is a universal structure. With RecA's role in recombination given new and greatly increased significance, we focus in this review on the energetics of the RecA-mediated strand exchange and the relation between the energetics and recombination spanning heterologous inserts.

High-mobility-group 1 protein mediates DNA bending as determined by ring closures.

High-mobility-group 1 protein (HMG1) is an abundant eukaryotic DNA-binding protein, the cellular role of which remains ill-defined. To test the ability of HMG1 itself to mediate curvature in double-stranded DNA, we examined its effect on the phage T4 DNA ligase-dependent cyclization of short DNA fragments. HMG1 caused circle formation for fragments > or = 87 bp. Fragments of 123, 100, 92, and 87 bp did not cyclize in the absence of protein but formed covalently closed circular monomers efficiently in the presence of HMG1, indicating that the protein is capable of introducing bends into the duplex. The bending activity was maintained by a 79-amino acid polypeptide corresponding to a single HMG-box domain of HMG1. The binding affinity for the DNA minicircle was greater than for the corresponding linear fragment. These findings indicate that the role of HMG1 could involve both structure-specific recognition of prebent DNA and distortion of the DNA helix by bending and that the HMG-box domain may actually be responsible for this activity.

[The ultrastructural characteristics of the metaphase chromosomal regions (telomere, centromere, nucleolar organizer) in the pig]. / Osobennosti ul'trastruktury nekotorykh raionov metafaznykh khromosom svin'i (telomera, tsentromera, iadryshkovyi organizator).

The ultrastructural organization of some chromosome regions (telomere, centromere, nucleolus organizer region), which can be identified undoubtedly under electron microscope after in situ fixation, was studied using serial ultrathin sections of pig embryo kidney cells. The ultrastructure of these regions was compared with patterns of their differential staining revealed by C-method and by fluorochromes, chromomycin A3 and DAPI, specific for GC- and AT-rich base pair regions. It is shown that telomere regions of chromosome are organized primarily by 20 nm fibrils of DNP, which form a clear chromonema--a thread of chromatin with thickness nearly 100 nm. In the centromere regions of metacentric chromosomes, fibrils 20 nm in diameter predominate, while in several acrocentric chromosomes--fibrils 10 nm thick are prevailing. It was suggested that the observed differences may depend on the DNA composition in these regions. The nucleolus organizer regions contain the fibrillar material 5-11 nm in diameter.

Crystal structure disposition of thymidylic acid tetramer in complex with ribonuclease A.

The crystal structure of ribonuclease A with bound thymidylic acid tetramer is reported at 2.5-A resolution. The diffusion of the tetramer into native orthorhombic crystals of the ribonuclease allows for the formation of a structurally stable complex where the single-stranded nucleic acid enters and leaves the enzyme's catalytic region in a persistent 5'-3' direction. The binding of the tetramer to the enzyme's surface is facilitated and mediated by electrostatic interactions between basic protein residues and nucleotide phosphates. Two pyrimidine nucleotides are bound to the enzyme's active site in a manner similar to that observed for other complexes between ribonuclease A and nucleic acid oligomers.

Bending of the Saccharomyces cerevisiae 5S rRNA gene in transcription factor complexes.

Bending of the yeast 5S rRNA gene by its transcription factors (TF) IIIA, IIIC, and IIIB has been investigated by two electrophoretic methods that exploit the anomalous mobility of bent DNA in tight gel networks. A minor bend is induced by TFIIIA, and a very strong bend, centered at its upstream DNA-binding site, is induced by TFIIIB. Despite binding to different DNA sequences on the 5S rRNA gene and the previously analyzed tRNA(Glu) gene, TFIIIB generates nearly identical bends in each site. Fully assembled transcription factor complexes bend the 5S rRNA gene in the same net direction as does TFIIIB alone.

Structural data suggest that the active and inactive forms of the RecA filament are not simply interconvertible.

We have used electron microscopy to examine the two major conformational states of the helical filament formed by the RecA protein of Escherichia coli. The compressed filament, formed in the absence of a nucleotide cofactor either as a self-polymer or on a single-stranded DNA molecule, is characterized in solution by about 6.1 subunits per turn of a 76 A pitch helix, and appears to be inactive with respect to all RecA activity. The active state of the filament, formed with ATP or an ATP analog on either a single or double-stranded DNA substrate, has about 6.2 subunits per turn of a 94 A pitch helix. Measurements of the contour length of RecA-covered single-stranded DNA circles in ice, formed in the absence of nucleotide cofactor, indicate that each RecA subunit binds five bases, in contrast to the three bases or base-pairs per subunit in the active state. The different stoichiometries of DNA binding suggests that the two polymeric forms are not interconvertible, as has been suggested on biochemical grounds. A three-dimensional reconstruction of the inactive state shows the same general features as the 83 A pitch filament present in the RecA crystal. This structural similarity and the fact that the crystal does not contain ATP or DNA suggests that the crystal structure is more similar to the compressed filament than the active, extended filament.

DNA-bridging by a palindromic alpha-helix.

The nucleosomal DNA repeat of 240 base pairs in the chromatin structure of sea urchin sperm is exceptionally long and is accompanied by the presence of a histone H1 molecule larger than is usual in most species of chromatin. I propose how these two features are correlated and how they fit into the solenoidal model for the 300-A-diameter fiber of chromatin. Comparison of the sequence of spermatogenous H1 with other H1 sequences reveals an insert of 55 amino acid residues (residues 122-176). A 37-residue sequence in the insert (residues 140-176) has a palindromic character. I propose that each half of the palindromic sequence constitutes an alpha-helical DNA-binding unit and that the continuous alpha-helix made up of the two halves, by virtue of its palindromic nature, stabilizes the formation of an extra superhelical turn by the long linker DNA between two nucleosome cores. The N-terminal-C-terminal "polarity" of each alpha-helical section of half the palindromic sequence indicates how the arginine/lysine-rich DNA-binding surface of the alpha-helical section is used. The polarity of the H1 insertion sequence supports the so-called "reverse-loop" model or a "figure-eight" model for the path of the DNA within the solenoid structure; i.e., the linker DNA forms a right-handed superhelical turn toward the center of the solenoid structure. This use of a pair of a palindromically related alpha-helical sections has a similarity with the "scissors-grip" model for the interaction of the leucine-zipper proteins with DNA.