The DNA architectural protein HMGB1 displays two distinct modes of action that promote enhanceosome assembly

HMGB1 (also called HMG-1) is a DNA-bending protein that augments the affinity of diverse regulatory proteins for their DNA sites. Previous studies have argued for a specific interaction between HMGB1 and target proteins, which leads to cooperative binding of the complex to DNA. Here we propose a different model that emerged from studying how HMGB1 stimulates enhanceosome formation by the Epstein-Barr viral activator Rta on a target gene, BHLF-1. HMGB1 stimulates binding of individual Rta dimers to multiple sites in the enhancer. DNase I and hydroxyl radical footprinting, electrophoretic mobility shift assays, and immobilized template assays failed to reveal stable binding of HMGB1 within the complex. Furthermore, mutational analysis failed to identify a specific HMGB1 target sequence. The effect of HMGB1 on Rta could be reproduced by individual HMG domains, yeast HMO1, or bacterial HU. These results, combined with the effects of single-amino-acid substitutions within the DNA-binding surface of HMGB1 domain A, argue for a mechanism whereby DNA-binding and bending by HMGB1 stimulate Rta-DNA complex formation in the absence of direct interaction with Rta or a specific HMGB1 target sequence. The data contrast with our analysis of HMGB1 action on another BHLF-1 regulatory protein called ZEBRA. We discuss the two distinct modes of HMGB1 action on a single regulatory region and propose how HMGB1 can function in diverse contexts.     

EcoRV restriction endonuclease binds all DNA sequences with equal affinity

In the presence of MgCl2, the EcoRV restriction endonuclease cleaves its recognition sequence on DNA at least a million times more readily than any other sequence. In this study, the binding of the EcoRV restriction enzyme to DNA was examined in the absence of Mg2+. With each DNA fragment tested, several DNA-protein complexes were detected by electrophoresis through polyacrylamide. No differences were observed between isogenic DNA molecules that either contained or lacked the EcoRV recognition site. The number of complexes with each fragment varied with the length of the DNA. Three complexes were formed with a DNA molecule of 55 base pairs, corresponding to the DNA bound to 1, 2, or 3 molecules of the protein, while greater than 15 complexes were formed with a DNA of 381 base pairs. A new method was developed to analyze the binding of a protein to multiple sites on DNA. The method showed that the EcoRV enzyme binds to all DNA sequences, including the EcoRV recognition site, with the same equilibrium constant, though two molecules of the protein bind preferentially to adjacent sites on the DNA in a cooperative fashion. All of the complexes with a substrate that contained the EcoRV site dissociated upon addition of competitor DNA, but when the competitor was mixed with MgCl2, a fraction of the substrate was cleaved at the EcoRV site. The fraction cleaved was due mainly to the translocation of the enzyme from nonspecific sites on the DNA to the specific site.     

From Nonspecific DNA–Protein Encounter Complexes to the Prediction of DNA–Protein Interactions

DNA–protein interactions are involved in many essential biological activities. Because there is no simple mapping code between DNA base pairs and protein amino acids, the prediction of DNA–protein interactions is a challenging problem. Here, we present a novel computational approach for predicting DNA-binding protein residues and DNA–protein interaction modes without knowing its specific DNA target sequence. Given the structure of a DNA-binding protein, the method first generates an ensemble of complex structures obtained by rigid-body docking with a nonspecific canonical B-DNA. Representative models are subsequently selected through clustering and ranking by their DNA–protein interfacial energy. Analysis of these encounter complex models suggests that the recognition sites for specific DNA binding are usually favorable interaction sites for the nonspecific DNA probe and that nonspecific DNA–protein interaction modes exhibit some similarity to specific DNA–protein binding modes. Although the method requires as input the knowledge that the protein binds DNA, in benchmark tests, it achieves better performance in identifying DNA-binding sites than three previously established methods, which are based on sophisticated machine-learning techniques. We further apply our method to protein structures predicted through modeling and demonstrate that our method performs satisfactorily on protein models whose root-mean-square Cα deviation from native is up to 5 Å from their native structures. This study provides valuable structural insights into how a specific DNA-binding protein interacts with a nonspecific DNA sequence. The similarity between the specific DNA–protein interaction mode and nonspecific interaction modes may reflect an important sampling step in search of its specific DNA targets by a DNA-binding protein.     

Bacteriophage lambda gpNu1 and Escherichia coli IHF proteins cooperatively bind and bend viral DNA: implications for the assembly of a genome-packaging motor

Terminase enzymes are common to both prokaryotic and eukaryotic double-stranded DNA viruses and are responsible for packaging viral DNA into the confines of an empty procapsid shell. In all known cases, the holoenzymes are heteroligomers composed of a large subunit that possesses the catalytic activities required for genome packaging and a small subunit that is responsible for specific recognition of viral DNA. In bacteriophage lambda, the DNA recognition protein is gpNu1. The gpNu1 subunit interacts with multiple recognition elements within cos, the packaging initiation site in viral DNA, to site-specifically assemble the packaging machinery. Motor assembly is modulated by the Escherichia coli integration host factor protein (IHF), which binds to a consensus sequence also located within cos. On the basis of a variety of biochemical data and the recently solved NMR structure of the DNA binding domain of gpNu1, we proposed a novel DNA binding mode that predicts significant bending of duplex DNA by gpNu1 (de Beer et al. (2002) Mol. Cell 9, 981-991). We further proposed that gpNu1 and IHF cooperatively bind and bend viral DNA to regulate the assembly of the packaging motor. Here, we characterize cooperative gpNu1 and IHF binding to the cos site in lambda DNA using a quantitative electrophoretic mobility shift (EMS) assay. These studies provide direct experimental support for the long presumed cooperative assembly of gpNu1 and IHF at the cos sequence of lambda DNA. Further, circular permutation experiments demonstrate that the viral and host proteins each introduce a strong bend in cos-containing DNA, but not nonspecific DNA substrates. Thus, specific recognition of viral DNA by the packaging apparatus is mediated by both DNA sequence information and by structural alteration of the duplex. The relevance of these results with respect to the assembly of a viral DNA-packaging motor is discussed.     

Interaction of nuclear factor EF-1A with the polyomavirus enhancer region.

Enhancer factor 1A (EF-1A) is a mammalian nuclear protein that previously was shown to bind cooperatively to the repeated core enhancer element I sequence in the adenovirus E1A enhancer region. We now have characterized three binding sites for EF-1A in the polyomavirus A2 (Py) enhancer region. Site 1 resides in the Py A enhancer domain, and sites 2 and 3 reside in the Py B enhancer domain. EF-1A binding to Py site 1 is independent of cooperation with other EF-1A sites or the adjacent binding sites for PEA-1 and PEA-2, two murine nuclear factors that bind in the Py A enhancer domain. EF-1A binding to Py sites 2 and 3, in contrast, is cooperative, similar to the situation previously observed with binding sites in the adenovirus E1A enhancer region. In a transient replication assay, EF-1A site 1 functions synergistically with the PEA-1 and PEA-2 sites in the A enhancer domain to enhance Py replication. The functional cooperativity observed with the EF-1A, PEA-1, and PEA-2 sites in vivo does not reflect cooperative DNA binding interactions, as detected in vitro. Py EF-1A site 1 alone is capable of weakly stimulating Py replication. EF-1A site 1 overlaps with the binding sites for the murine nuclear protein PEA-3 and the ets family of oncoproteins.     

Mechanisms of specific and nonspecific binding of architectural proteins in prokaryotic gene regulation

IHF and HU are small basic proteins of eubacteria that bind as homodimers to double-stranded DNA and bend the duplex to promote architectures required for gene regulation. These architectural proteins share a common alpha/beta fold but exhibit different nucleic acid binding surfaces and distinct functional roles. With respect to DNA-binding specificity, for example, IHF is sequence specific, while HU is not. We have employed Raman difference spectroscopy and gel mobility assays to characterize the molecular mechanisms underlying such differences in DNA recognition. Parallel studies of solution complexes of IHF and HU with the same DNA nonadecamer (5' --> 3' sequence: TC TAAGTAGTTGATTCATA, where the phage lambda H1 consensus sequence of IHF is underlined) show the following. (i) The structure of the targeted DNA site is altered much more dramatically by IHF than by HU binding. (ii) In the IHF complex, the structural perturbations encompass both the sugar-phosphate backbone and the bases of the consensus sequence, whereas only the DNA backbone is altered by HU binding. (iii) In the presence of excess protein, complexes of order higher than 1 dimer per duplex are detected for HU:DNA, though not for IHF:DNA. The results differentiate structural motifs of IHF:DNA and HU:DNA solution complexes, provide Raman signatures of prokaryotic sequence-specific and nonspecific recognition, and suggest that the architectural role of HU may involve the capability to recruit additional binding partners to even relatively short DNA sequences.     

Directional binding of HMG-I(Y) on four-way junction DNA and the molecular basis for competitive binding with HMG-1 and histone H1

Histone H1, HMG-1 and HMG-I(Y) are mammalian nuclear proteins possessing distinctive DNA-binding domain structures that share the common property of preferentially binding to four-way junction (4H) DNA, an in vitro mimic of the in vivo genetic recombination intermediate known as the Holliday junction. Nevertheless, these three proteins bind to 4H DNA in vitro with very different affinities and in a mutually exclusive manner. To investigate the molecular basis for these distinctive binding characteristics, we employed base pair resolution hydroxyl radical footprinting to determine the precise sites of nucleotide interactions of both HMG-1 and histone H1 on 4H DNA and compared these contacts with those previously described for HMG-I(Y) on the same substrate. Each of these proteins had a unique binding pattern on 4H DNA and yet shared certain common nucleotide contacts on the arms of the 4H DNA molecule near the branch point. Both the HMG-I(Y) and HMG-1 proteins made specific contacts across the 4H DNA branch point, as well as interacting at discrete sites on the arms, whereas the globular domain of histone H1 bound exclusively to the arms of the 4H DNA substrate without contacting nucleotides at the crossover region. Experiments employing the chemical cleavage reagent 1, 10-orthophenanthroline copper(II) attached to the C-terminal end of a site-specifically mutagenized HMG-I(Y) protein molecule demonstrated that this protein binds to 4H DNA in a distinctly polar, direction-specific manner. Together these results provide an attractive molecular explanation for the observed mutually exclusive 4H DNA-binding characteristics of these proteins and also allow for critical assessment of proposed models for their interaction with 4H DNA substrates. The results also have important implications concerning the possible in vivo roles of HMG-I(Y), histone H1 and HMG-1 in biological processes such as genetic recombination and retroviral integration.