Sequence-dependent DNA helical rise and nucleosome stability

Background  

Nucleosomes are the basic structural units of eukaryotic chromatin and play a key role in regulation of gene expression. After resolution of the nucleosome structure, the bipartite nature of this particle has revealed itself and has disclosed the presence, on the histone surface, of a symmetric distribution of positive charges, able to interact with their negative DNA phosphate counterpart.     

The helical model of the nucleosome core.

A model of the nucleosome core is proposed based on a topologically linear array of histones attached sequentially to DNA. The linear complex folds helically forming a spring-like particle. Different variants of the particle are discussed (cylindrical springs with and without histone-histone contacts between turns of the helix, solenoidal spring). The model is consistent with known data about the nucleosome structure. Histones H3 and H4 have a special role in the model which is related also to the superstructure of chromatin.     

Preferential nucleosome occupancy at high values of DNA helical rise

Nucleosomes are the basic structural units of eukaryotic chromatin and play a key role in the regulation of gene expression. Nucleosome formation depends on several factors, including properties of the sequence itself, but also physical constraints and epigenetic factors such as chromatin-remodelling enzymes. In this view, a sequence-dependent approach is able to capture a general tendency of a region to bind a histone octamer. A reference data set of positioned nucleosomes of Saccharomyces cerevisiae was used to study the role of DNA helical rise in histone-DNA interaction. Genomic sequences were transformed into arrays of helical rise values by a tetranucleotide code and then turned into profiles of mean helical rise values. These profiles resemble maps of nucleosome occupancy, suggesting that intrinsic histone-DNA interactions are linked to helical rise. The obtained results show that preferential nucleosome occupancy occurs where the mean helical rise reaches its largest values. Mean helical rise profiles obtained by using maps of positioned nucleosomes of the Drosophila melanogaster and Plasmodium falciparum genomes, as well as Homo sapiens chromosome 20 confirm that nucleosomes are mainly located where the mean helical rise reaches its largest values.     

Two new photoaffinity polyamines appear to alter the helical twist of DNA in nucleosome core particles

Two new photoaffinity derivatives of polyamines have been synthesized by the reaction of spermine or spermidine with methyl 4-azidobenzimidate. The new compounds were purified chromatographically and characterized by several methods including proton magnetic resonance spectroscopy. The spermine derivative is N1-ABA-spermine [(azidobenzamidino)spermine], and the spermidine derivative is a mixture of N1- and N8-ABA-spermidine. ABA-spermine stabilizes nucleosome core particles in thermal denaturation experiments, with similar but not identical effects when compared with the parent polyamine, spermine. In circular dichroism experiments, ABA-spermine was capable of producing a B----Z transition in poly(dG-m5dC) at a concentration of 30 microM, compared with 5 microM required to produce the same effect with spermine. On the other hand, ANB-spermine [(azidonitrobenzoyl)spermine; Morgan, J. E., Calkins, C. C., & Matthews, H. R. (1989) Biochemistry 28, 5095-5106] stabilized the B form of poly(dG-br5dC). ABA-spermine is a potent inhibitor of ornithine decarboxylase from Escherichia coli, giving 50% inhibition at 0.12 mM, while ANB-spermine is a modest inhibitor, comparable to spermine or spermidine. Under conditions of nitrogen-limited growth, yeast take up ABA-spermine and ABA-spermidine at approximately one-third to half the rate of spermidine or spermine. In contrast, ANB-spermine was not significantly taken up. The photoaffinity polyamines were used to photoaffinity label the DNA in nucleosome core particles, and the sites of labeling were determined by exonuclease protection. All photoaffinity reagents showed both nonspecific labeling and specific sites of higher occupancy. However, the positions of the sites varied: the ANB-spermine sites confirmed those previously reported (Morgan et al., 1989); the ABA-spermine and ABA-spermidine sites were spaced at 9.8 bp intervals from the 3' end of each DNA strand. This observation, together with the effect of spermine on the circular dichroism of DNA in nucleosome core particles, implies that polyamines alter the helical twist of DNA in nucleosome core particles. The ABA-polyamines are offered as general-purpose photoaffinity polyamine reagents.     

The mechanics behind DNA sequence-dependent properties of the nucleosome

Chromatin organization and composition impart sophisticated regulatory features critical to eukaryotic genomic function. Although DNA sequence-dependent histone octamer binding is important for nucleosome activity, many aspects of this phenomenon have remained elusive. We studied nucleosome structure and stability with diverse DNA sequences, including Widom 601 derivatives with the highest known octamer affinities, to establish a simple model behind the mechanics of sequence dependency. This uncovers the unique but unexpected role of TA dinucleotides and a propensity for G|C-rich sequence elements to conform energetically favourably at most locations around the histone octamer, which rationalizes G|C% as the most predictive factor for nucleosome occupancy in vivo. In addition, our findings reveal dominant constraints on double helix conformation by H3-H4 relative to H2A-H2B binding and DNA sequence context-dependency underlying nucleosome structure, positioning and stability. This provides a basis for improved prediction of nucleosomal properties and the design of tailored DNA constructs for chromatin investigations.     

The influence of DNA stiffness upon nucleosome formation

The rotational and translational positioning of nucleosomes on DNA is dependent to a significant extent on the physicochemical properties of the double helix. We have investigated the influence of the axial flexibility of the molecule on the affinity for the histone octamer by substituting selected DNA sequences with either inosine for guanosine or diaminopurine for adenine. These substitutions, respectively, remove or add a purine 2-amino group exposed in the minor groove and, respectively, decrease and increase the apparent persistence length. We observe that for all sequences tested inosine substitution, with one exception, increases the affinity for histone binding. Conversely diaminopurine substitution decreases the affinity. In the sole example where replacement of guanosine with inosine decreases the persistence length as well as the affinity for histones, the substitution concomitantly removes an intrinsic curvature of the DNA molecule. We show that, to a first approximation, the binding energy of DNA to histones at 1M NaCl is directly proportional to the persistence length. The data also indicate that a high local flexibility of DNA can favour strong rotational positioning.     

Chromatin reconstitution on small DNA rings. II. DNA supercoiling on the nucleosome

DNA supercoiling on the nucleosome was investigated by relaxing with topoisomerase I mono- and dinucleosomes reconstituted on small DNA rings. Besides 359 base-pair (bp) rings whose linking differences were integers, two additional series of rings with fractional differences, 341 and 354 bp in size, were used. Mononucleosomes reconstituted on 359 bp rings were found to relax into a single mononucleosome form. In contrast, 341 and 354 bp mononucleosomes relaxed into a mixture of two forms, corresponding to two adjacent topoisomers. The observation that the ratio between these two forms was, within each ring series, virtually independent of the initial linking number of the topoisomer used for the reconstitution suggested that each partition reflected an equilibrium. Comparison with the equilibria observed for the same rings in the absence of histones showed that the formation of a single nucleosome is associated with a linking number change of -1.1(+/-0.1) turn. Dinucleosomes, in contrast, were not relaxed to completion and do not reach equilibria. The corresponding linking number change per nucleosome was, however, estimated to be similar to the above figure, in agreement with previous data from the literature obtained with circular chromatins containing larger numbers of nucleosomes. DNA structure in mononucleosomes was subsequently investigated by means of high-resolution electron microscopy and gel electrophoresis. It was found that the above linking number reduction could be ascribed to a particle with a large open extranucleosomal DNA loop and with no more than 1.5 turns of a superhelix around the histone core. A theoretical model of a nucleosome on a small ring was constructed in which one part of the DNA was wrapped around a cylinder and the other part was free to vary both in torsion and flexion. The linking number reduction predicted was found to be most consistent with experimental data when the twist of the DNA in the superhelix was between 10.5 and 10.65 pb per turn, suggesting that wrapping on the nucleosome does not alter the twist of the DNA significantly. A lower estimate of the linking number reduction associated with a two-turn nucleosome was also derived, based on an analysis of recent data obtained upon treatment of reconstituted minichromosomes with gyrase. The value, 1.6 turns, set a lower limit of 10.44 bp per turn for the twist of nucleosomal DNA, in agreement with the above estimate.(ABSTRACT TRUNCATED AT 400 WORDS)     

The number of charge-charge interactions stabilizing the ends of nucleosome DNA.

It has been shown by others that the melting of DNA in the nucleosome core particle is biphasic (ref. 1) and that the initial denaturation phase is due to melting of the DNA termini (refs. 1 & 2). We analyze the salt dependence of the melting temperature of this first transition and estimate that only 15% of the phosphates of the DNA termini are involved in intimate charge-charge interactions with histones. (The simplest model yields approximately 9%, whereas a calculated overestimate yields approximately 21% neutralization.) This is a surprisingly small number of interactions but we suggest that it may nonetheless be representative of all the core particle DNA.     

Characteristics of nucleosome core DNA and their applications in predicting nucleosome positions

By analyzing dinucleotide position-frequency data of yeast nucleosome-bound DNA sequences, dinucleotide periodicities of core DNA sequences were investigated. Within frequency domains, weakly bound dinucleotides (AA, AT, and the combinations AA-TT-TA and AA-TT-TA-AT) present doublet peaks in a periodicity range of 10-11 bp, and strongly bound dinucleotides present a single peak. A time-frequency analysis, based on wavelet transformation, indicated that weakly bound dinucleotides of core DNA sequences were spaced smaller (∼10.3 bp) at the two ends, with larger (∼11.1 bp) spacing in the middle section. The finding was supported by DNA curvature and was prevalent in all core DNA sequences. Therefore, three approaches were developed to predict nucleosome positions. After analyzing a 2200-bp DNA sequence, results indicated that the predictions were feasible; areas near protein-DNA binding sites resulted in periodicity profiles with irregular signals. The effects of five dinucleotide patterns were evaluated, indicating that the AA-TT pattern exhibited better performance. A chromosome-scale prediction demonstrated that periodicity profiles perform better than previously described, with up to 59% accuracy. Based on predictions, nucleosome distributions near the beginning and end of open reading frames were analyzed. Results indicated that the majority of open reading frames’ start and end sites were occupied by nucleosomes.     

ISWI induces nucleosome sliding on nicked DNA.

The ATPase ISWI is the molecular motor of several remodeling factors that trigger nucleosome sliding in vitro. In search for the underlying mechanism, we found that unilateral binding of ISWI to a model nucleosome correlated with directional movement of the nucleosome toward the enzyme. It has been proposed that ISWI might loosen histone-DNA interactions through twisting DNA. However, nucleosome sliding assays on nicked DNA substrates suggest that propagation of altered twist is not involved. Surprisingly, nicks in the linker DNA in front of the nucleosome facilitate sliding. These data suggest that the rate of nucleosome sliding is limited by a conformational change other than twisting, such as the formation of a short loop, of DNA at the entry into the nucleosome.     

Effect of DNA length on the nucleosome low salt transition.

The effect of DNA length on the low salt unfolding transition of nucleosomes has been studied by the use of fluorescently labeled histones. Nucleosomes were formed by the reconstitution of bulk DNA fragments averaging 173 and 250 base pairs in length. These nucleosomes exhibited a conformational change in a transition centered at about 7 mM ionic strength, very different from that observed for the standard 145 bp nucleosomes (1-3mM). In addition, the conformational change of the 173 and 250 bp nucleosomes involves twice as many ions as that of the 145 bp nucleosomes.     

Sequence structure of human nucleosome DNA

Positional distributions of various dinucleotides in experimentally derived human nucleosome DNA sequences are analyzed. Nucleosome positioning in this species is found to depend largely on GG and CC dinucleotides periodically distributed along the nucleosome DNA sequence, with the period of 10.4 bases. The GG and CC dinucleotides oscillate counterphase, i.e., their respective preferred positions are shifted about a half-period from one another, as it was observed earlier for AA and TT dinucleotides. Other purine-purine and pyrimidine-pyrimidine dinucleotides (RR and YY) display the same periodical and counterphase pattern. The dominance of oscillating GG and CC dinucleotides in human nucleosomes and the contribution of AG(CT), GA(TC), and AA(TT) suggest a general nucleosome DNA sequence pattern - counterphase oscillation of RR and YY dinucleotides. AA and TT dinucleotides, commonly accepted as major players, are only weak contributors in the case of human nucleosomes.     

Enhancer elements share local homologous twist-angle variations with a helical periodicity

Several experiments have shown that some enhancers can be exchanged between different genomes. The transferred enhancers were functional (cf. Levinson, B., Khoury, G., Vande Woude, G. and Gruss, P. (1982) Nature 295, 568-572). This argues that these exchanged fragments are recognized as enhancers and possess some common characteristics which other sequences lack. Extensive comparisons of enhancers yielded only very limited nucleotide sequence homology, which appears to be insufficient for enhancer recognition. We suggest that the enhancers located and sequenced to date have recurring, periodic homologous twist-angle (tg) patterns. This helical periodicity and the symmetric nature of the repeating twist-angle features present a recurring spatial geometry. It also offers a possible explanation of the fact that inverted enhancers are still functional. Regions of large twist, roll or main-chain torsion angle delta deviations from regular B-DNA may facilitate enhancer recognition especially when distant from promoter elements. Tissue specificity may be encoded in additional sequence or structural features.     

Enhancer elements share local homologous twist-angle variations with a helical periodicity

Several experiments have shown that some enhancers can be exchanged between different genomes. The transferred enhancers were functional (cf. Levinson, B., Khoury, G., Vande Woude, G. and Gruss, P. (1982) Nature 295, 568–572). This argues that these exchanged fragments are recognized as enhancers and possess some common characteristics which other sequences lack. Extensive comparisons of enhancers yielded only very limited nucleotide sequence homology, which appears to be insufficient for enhancer recognition. We suggest that the enhancers located and sequenced to date have recurring, periodic homologous twist-angle (t_g) patterns. This helical periodicity and the symmetric nature of the repeating twist-angle features present a recurring spatial geometry. It also offers a possible explanation of the fact that inverted enhancers are still functional. Regions of large twist, roll or main-chain torsion angle δ deviations from regular B-DNA may facilitate enhancer recognition especially when distant from promoter elements. Tissue specificity may be encoded in additional sequence or structural features.