Competitive nucleosome reconstitution of polydeoxynucleotides containing oligoguanosine tracts

Modulation of cyclobutane thymine photodimer formation in T11-tracts in rotationally phased nucleosome core particles and DNA minicircles

Cyclobutane pyrimidine dimers (CPDs) are DNA photoproducts linked to skin cancer, whose mutagenicity depends in part on their frequency of formation and deamination. Nucleosomes modulate CPD formation, favoring outside facing sites and disfavoring inward facing sites. A similar pattern of CPD formation in protein-free DNA loops suggests that DNA bending causes the modulation in nucleosomes. To systematically study the cause and effect of nucleosome structure on CPD formation and deamination, we have developed a circular permutation synthesis strategy for positioning a target sequence at different superhelix locations (SHLs) across a nucleosome in which the DNA has been rotationally phased with respect to the histone octamer by TG motifs. We have used this system to show that the nucleosome dramatically modulates CPD formation in a T₁₁-tract that covers one full turn of the nucleosome helix at seven different SHLs, and that the position of maximum CPD formation at all locations is shifted to the 5΄-side of that found in mixed-sequence nucleosomes. We also show that an 80-mer minicircle DNA using the same TG-motifs faithfully reproduces the CPD pattern in the nucleosome, indicating that it is a good model for protein-free rotationally phased bent DNA of the same curvature as in a nucleosome, and that bending is modulating CPD formation.

Homopolymer tract length dependent enrichments in functional regions of 27 eukaryotes and their novel dependence on the organism DNA (G+C)% composition

Background

DNA homopolymer tracts, poly(dA).poly(dT) and poly(dG).poly(dC), are the simplest of simple sequence repeats. Homopolymer tracts have been systematically examined in the coding, intron and flanking regions of a limited number of eukaryotes. As the number of DNA sequences publicly available increases, the representation (over and under) of homopolymer tracts of different lengths in these regions of different genomes can be compared.

Results

We carried out a survey of the extent of homopolymer tract over-representation (enrichment) and over-proportional length distribution (above expected length) primarily in the single gene documents, but including some whole chromosomes of 27 eukaryotics across the (G+C)% composition range from 20 – 60%. A total of 5.2 × 10⁷ bases from 15,560 cleaned (redundancy removed) sequence documents were analyzed. Calculated frequencies of non-overlapping long homopolymer tracts were found over-represented in non-coding sequences of eukaryotes. Long poly(dA).poly(dT) tracts demonstrated an exponential increase with tract length compared to predicted frequencies. A novel negative slope was observed for all eukaryotes between their (G+C)% composition and the threshold length N where poly(dA).poly(dT) tracts exhibited over-representation and a corresponding positive slope was observed for poly(dG).poly(dC) tracts. Tract size thresholds where over-representation of tracts in different eukaryotes began to occur was between 4 – 11 bp depending upon the organism (G+C)% composition. The higher the GC%, the lower the threshold N value was for poly(dA).poly(dT) tracts, meaning that the over-representation happens at relatively lower tract length in more GC-rich surrounding sequence. We also observed a novel relationship between the highest over-representations, as well as lengths of homopolymer tracts in excess of their random occurrence expected maximum lengths.

Conclusions

We discuss how our novel tract over-representation observations can be accounted for by a few models. A likely model for poly(dA).poly(dT) tract over-representation involves the known insertion into genomes of DNA synthesized from retroviral mRNAs containing 3' polyA tails. A proposed model that can account for a number of our observed results, concerns the origin of the isochore nature of eukaryotic genomes via a non-equilibrium GC% dependent mutation rate mechanism. Our data also suggest that tract lengthening via slip strand replication is not governed by a simple thermodynamic loop energy model.

DNA sequence-dependent contributions of core histone tails to nucleosome stability: differential effects of acetylation and proteolytic tail removal

Modulation of nucleosome stability in chromatin plays an important role in eukaryotic gene expression. The core histone N-terminal tail domains are believed to modulate the stability of wrapping nucleosomal DNA and the stability of the chromatin filament. We analyzed the contribution of the tail domains to the stability of nucleosomes containing selected DNA sequences that are intrinsically straight, curved, flexible, or inflexible. We find that the presence of the histone tail domains stabilizes nucleosomes containing DNA sequences that are intrinsically straight or curved. However, the tails do not significantly contribute to the free energy of nucleosome formation with flexible DNA. Interestingly, hyperacetylation of the core histone tail domains does not recapitulate the effect of tail removal by limited proteolysis with regard to nucleosome stability. We find that acetylation of the tails has the same minor effect on nucleosome stability for all the selected DNA sequences. A comparison of histone partitioning between long donor chromatin, acceptor DNA, and free histones in solution shows that the core histone tails mediate internucleosomal interactions within an H1-depleted chromatin fiber amounting to an average free energy of about 1 kcal/mol. Thus, such interactions would be significant with regard to the free energies of sequence-dependent nucleosome positioning. Last, we analyzed the contribution of the H2A/H2B dimers to nucleosome stability. We find that the intact nucleosome is stabilized by 900 cal/mol by the presence of the dimers regardless of sequence. The biological implications of these observations are discussed.

Embryonic inheritance of the chromatin organisation of the imprinted H19 domain in mouse spermatozoa

Insulin-like growth factor 2 (Igf 2) and H19 genes are oppositely imprinted and as such have been most extensively studied imprinted genes both genetically and at the molecular level. Imprints of the H19 gene, being established during spermatogenesis, are epigenetically transmitted to the somatic cells of the embryo. Current hypotheses attempting to explain the allele-specific silence of the H19 gene include DNA methylation and chromatin condensation. In order to understand the molecular basis of H19 epigenesis, it is crucial to identify the markings in the chromatin organising the imprinted domain in spermatozoa. Using Micrococcal nuclease (MNase), DNase I and Methidiumpropyl-EDTA. iron II (MPE.Fe(II)) as chromatin probes, we demonstrate that in mouse epididymal spermatozoa, at least 4kb DNA upstream of the H19 'cap' site, containing the imprinted and differentially methylated domain (DMD), is heterochromatic. The cleavage sites in this domain (-2 to -4kb) exhibit approximately 425bp periodicity. This structure is maintained in the paternal allele of normal embryos and is disrupted at -2.2, -2.65 and at -3.5kb in embryos maternally disomic for the distal end of chromosome 7 (MatDp 7). The hypersensitive sites in chromatin precisely register the MPE.Fe(II) cleavage sites in chromosomal DNA. Therefore, the DNA sequences in the imprinted domain constrain the chromatin structure in a way similar to that of 1.688g/cm(3) Drosophila satellite chromatin. In addition, we find that condensation of the paternal allele correlates with methylation-dependent alteration in the structure of DNA sequences in DMD. These results suggest that CpG-methylation induces localised changes in DNA conformation and these facilitate consequent remodelling of chromatin thereby allowing the paternal and maternal H19 alleles to be distinguished.

Control of meiotic recombination and gene expression in yeast by a simple repetitive DNA sequence that excludes nucleosomes

Tandem repeats of the pentanucleotide 5'-CCGNN (where N indicates any base) were previously shown to exclude nucleosomes in vitro (Y. -H. Wang and J. D. Griffith, Proc. Natl. Acad. Sci. USA 93:8863-8867, 1996). To determine the in vivo effects of these sequences, we replaced the upstream regulatory sequences of the HIS4 gene of Saccharomyces cerevisiae with either 12 or 48 tandem copies of CCGNN. Both tracts activated HIS4 transcription. We found that (CCGNN)(12) tracts elevated meiotic recombination (hot spot activity), whereas the (CCGNN)(48) tract repressed recombination (cold spot activity). In addition, a "pure" tract of (CCGAT)(12) activated both transcription and meiotic recombination. We suggest that the cold spot activity of the (CCGNN)(48) tract is related to the phenomenon of the suppressive interactions of adjacent hot spots previously described in yeast (Q.-Q. Fan, F. Xu, and T. D. Petes, Mol. Cell. Biol. 15:1679-1688, 1995; Q.-Q. Fan, F. Xu, M. A. White, and T. D. Petes, Genetics 145:661-670, 1997; T.-C. Wu and M. Lichten, Genetics 140:55-66, 1995; L. Xu and N. Kleckner, EMBO J. 16:5115-5128, 1995).

Sequence directed flexibility of DNA and the role of cross-strand hydrogen bonds

Persistence length and torsional rigidity for different B-DNA sequences have been calculated by analysing crystal structure database. The values of these parameters for mixed sequence DNA are in good agreement with those estimated by others. Persistence lengths for the homopolymeric sequences, namely poly(dA).poly(dT) and poly(dG).poly(dC), are significantly large compared to those of others as expected from the inability of these sequences to form nucleosome under normal conditions. The heteropolymeric sequences poly(dA-dC).poly(dG-dT) and poly(dG-dC).poly(dG-dC), on the other hand, have smaller persistence lengths. This implies larger flexibility of the d(AC).d(GT), d(CA).d(TG), d(GC).d(GC) and d(CG).d(CG) doublets, some of which constitute the genetic disease forming triplet repeats d(CTG).d(CAG) and d(CGG).d(CCG). Thus it is expected that these triplet repeat sequences are also flexible and wrap around the histone octamer efficiently. Persistence length calculations also indicate larger flexibility for these triplet repeat sequences. Furthermore, our computations reveal that the rigidity of a given DNA sequence is controlled by its ability to form cross-strand bifurcated hydrogen bonds between the successive base pairs. Molecular orbital calculations suggest that these hydrogen bonds are generally extended with bond lengths around 3A.

Architectural DNA-binding properties of the spermatidal transition proteins 1 and 2

Mammalian spermiogenesis is characterized by replacement of somatic histones by a set of basic nuclear transition proteins thought to be actively involved in the chromatin remodeling process. The two major transition proteins of the elongating spermatids, namely TP1 and TP2, were expressed and purified using a bacterial expression system. Both topoisomerase and ligase-mediated supercoiling assays demonstrated that TP1, as well as TP2, did not produce detectable changes in the twist and/or writhe of DNA molecules upon binding. Ligase-mediated circularization assay further demonstrated that neither of the transition proteins under study produced bends in linear DNA but that they both have the capacity to stimulate oligomerization of linear DNA fragments. We further established that the transition proteins are in vitro substrates for the Ca+2-phospholipid-dependent protein kinase (PKC) as well as the cAMP-dependent protein kinase (PKA). PKC phosphorylation was found to strongly weaken the DNA-condensing ability of TP2. These results suggest that the major transition proteins represent architectural factors able to stabilize DNA in a nonsupercoiled state, thereby promoting DNA condensation.

Nucleosome assembly on telomeric sequences

The organization of telomeric chromatin differs from that of bulk chromatin in some peculiar features, such as the unusually short nucleosomal spacing found in vertebrates. Telomeric DNAs are straight, since they consist mostly of 6-8-bp repeated sequences, therefore out of phase with the B DNA period. This feature should be of relevance in nucleosome formation, suggesting the usefulness of studying simple model systems of nucleosome assembly. We reconstituted nucleosomes in vitro, by using purified histone octamers and/or by octamer transfer from chicken erythrocyte nucleosomes, onto telomeric sequences from human, Arabidopsis thaliana, and Saccharomyces cerevisiae. All of these telomeres contain GGG and GGT triplets but are characterized by different repeat lengths (6, 7, and 8 bp). The free energies involved in the association process are the highest among the biological sequences so far assayed, suggesting a main role of DNA flexibility in the assembly of telomeric chromatin. Digestion studies with DNase I, hydroxyl radicals, exonuclease III, and lambda exonuclease indicate that telomeric nucleosomes are characterized by multiple translational positioning without rotational phasing, whereas the telomeric DNA folding around the histone octamer shows the canonical periodicity of about 10.2 bp. The experimental results and a theoretical simulation of DNase I digestion indicate a multiple nucleosome positioning with the periodicity of telomeric DNA. This suggests a main role of local chemical recognition between telomeric sequences and the histone octamer in nucleosome assembly.

Oligoadenosine tracts favor nucleosome formation

We have measured the ability of oligoadenosine tracts 25 base pairs in length to influence nucleosome formation. Such tracts can cause DNA to bind in nucleosomes at higher temperatures with a free energy up to 1 kcal/mol more favorable than heterogenous-sequence DNA. Furthermore, the position of the oligoadenosine tract affects the free energy of binding, with the most favorable position occurring at the dyad axis.

DNA damage can alter the stability of nucleosomes: effects are dependent on damage type

We have investigated the effects of DNA damage by (+/-)-anti-benzo[a]pyrene diol epoxide (BPDE) and UV light on the formation of a positioned nucleosome in the Xenopus borealis 5S rRNA gene. Gel-shift analysis of the reconstituted products indicates that BPDE damage facilitates the formation of a nucleosome onto this sequence. Competitive reconstitution experiments show that average levels of 0.5, 0.9, and 2.1 BPDE adducts/146 bp of 5S DNA (i.e., the size of DNA associated with a nucleosome core particle) yield changes of -220, -290, and -540 cal/mol, respectively, in the free energy (delta G) of nucleosome formation. These values yield increases of core histone binding to 5S DNA (K(a)) of 1.4-, 1.6-, and 2.5-fold, compared with undamaged DNA. Conversely, irradiation with UV light decreases nucleosome formation. Irradiation at either 500 or 2500 J/m2 of UV light [0.6 and 0.8 cyclobutane pyrimidine dimer/146 bp (on average), respectively] results in respective changes of +130 and +250 cal/mol. This translates to decreases in core histone binding to irradiated 5S DNA (K(a)) of 1.2- and 1.5-fold compared with undamaged DNA. These results indicate that nucleosome stability can be markedly affected by the formation of certain DNA lesions. Such changes could have major effects on the kinetics of DNA processing events.

Chemical probes of DNA structure in chromatin

Understanding the way genes work requires detailed knowledge of the organization of DNA in the chromatin complex. The difficulties associated with the study of this large macromolecular assembly present an interesting challenge to both biologists and chemists.

An overabundance of long oligopurine tracts occurs in the genome of simple and complex eukaryotes

A search of sequence information in the GenBank files shows that tracts of 15-30 contiguous purines are greatly overrepresented in all eukaryotic species examined, ranging from yeast to human. Such an overabundance does not occur in prokaryotic sequences. The large increase in the number of oligopurine tracts cannot be explained as a simple consequence of base composition, nearest-neighbor frequencies, or the occurrence of an overabundance of oligoadenosine tracts. Oligopurine sequences have previously been shown to be versatile structural elements in DNA, capable of occuring in several alternate conformations. Thus the bias toward long oligopurine tracts in eukaryotic DNA may reflect the usefulness of these structurally versatile sequences in cell function.

Alignment of (dA).(dT) homopolymer tracts in gene flanking sequences suggests nucleosomal periodicity in D. discoideum DNA

It has been shown that the frequency versus size distribution of A and T overlapping and non-overlapping homopolymer tracts of N > 5 in D. discoideum gene flanking and intron regions are significantly greater than in coding regions(1). In the present report, we demonstrate, that a spatial periodicity exists in long A and T tracts (N > 10) in long flanking sequences by scored alignments of those tracts (N > 10) with the nucleosomal repeat. A tract spacing was found at 185-190 bp that corresponds to a maximum alignment score. This is exactly the average spacing of D. discoideum nucleosomes determined experimentally. A majority of A and T tracts in flanking sequences are often spaced by short DNA stretches and the total length of adjacent A and T tracts plus the interrupting short DNA stretch corresponds closely to the average experimentally measured nucleosomal linker DNA size in D. discoideum-42 bp. These data suggest a model which has A and T runs of N > 10 bp in flanking DNA of D. discoideum organized in a regular phase with nonhomopolymer sequences along the DNA. This model has functional implications for A and T tracts, suggesting that they are found in nucleosomal linker DNA regions of chromatin during some necessary portion(s) of the life of the cell.

Characteristics of the large (dA).(dT) homopolymer tracts in D. discoideum gene flanking and intron sequences

D. discoideum, the slime mold, is one of the most AT rich eukaryotic genomes known. In this paper we examine this organism's database for overlapping N-tuples of high frequency and find A and T tracts possess among the highest frequencies in flanking sequences but not in coding sequences. We examined both overlapping and non-overlapping frequencies of the A, T, G and C homopolymer tracts of 2 < N < 6. Overlapping (dG).(dC) and (dA).(dT) tracts occur at greater frequencies than expected, based on random occurrence. Long (dA).(dT) tracts of N > 10 occur at well above expected frequencies in flanking and intron regions, while (dG).(dC) tracts above N = 5 are rarely found. Some of the implications of these findings for tract origins in slip-strand replication and for chromatin structure are discussed.

Structure of DNA in a nucleosome core at high salt concentration and at high temperature

We have used hydroxyl radical cleavage of DNA to probe the organization of the nucleosome core at high salt concentration and high temperature. The rotational and translational positioning of a DNA fragment, containing part of the Xenopus borealis 5S RNA gene, on the histone octamer is maintained between salt concentrations of 0.0 and 0.8 M NaCl and between temperatures of 0 and 75 degrees C. These results provide evidence that the energy of bending DNA around the nucleosome is independent of salt concentration and temperature in this range. They illustrate the severe energetic requirements necessary to displace DNA from previously organized contacts with histones in the nucleosome core.

Isolated oligopurine tracts do not significantly affect the binding of DNA to nucleosomes

Nucleosomal-length DNA was constructed to contain one of two 10 bp oligopurine-oligopyrimidine sequences, either d(A10.T10) or d(G10.C10). The 146 base pair (bp) sequences were then each tandemly cloned. This allowed for the production of circularly-permuted sequence variants in which the oligopurine tract was located at eight different positions. The permuted sequences were then assayed for their ability to reconstitute into nucleosomes by competitive reconstitution. The results of the assay indicate that the free energy of nucleosome formation differs only by several tenths of a kilocalorie per mole for an oligopurine tract at any position along the DNA, including the central dyad region.

Poly[d(A.T)] and other synthetic polydeoxynucleotides containing oligoadenosine tracts form nucleosomes easily

Synthetic double-stranded polydeoxynucleotides of the general form poly[d(AnT).d(ATn)], with n ranging from 3 to 11, have been synthesized. The conformation of the polymers was investigated by circular dichroism spectroscopy and the polymers were examined for their ability to form nucleosomes. Although spectra show that a circular dichroism band characteristic of poly[d(A.T)] appears in the polymer family for n greater than 7, we demonstrate that even polynucleotides with the longest tracts of contiguous adenosine bases (n = 11) are able to form nucleosomes when reconstituted using a histone exchange procedure. Thus resistance to nucleosome formation does not coincide with the appearance of features similar to that of poly[d(A.T)] over the bulk of the nucleosomal DNA. Furthermore, we show that an approximately 150 base-pair poly[d(A.T)] itself, long thought to be refractory to nucleosome formation, can assemble into such a protein-DNA complex when reconstituted by a low-salt exchange procedure. Competitive assays show that the homopolymer reconstitutes about as well as heterogeneous sequences DNA. Our work, therefore, suggests that highly adenosine-rich sequences in vivo apparently have a function that operates at a level other than that of nucleosome structure.

Eukaryotic DNA does not form nucleosomes as readily as some prokaryotic DNA

Nucleosomal-length DNA was prepared from the genomic DNA of various prokaryotic and eukaryotic organisms by limited nuclease digestion after reconstitution with core histones. The DNAs ranged in base composition from 26.5% to 72% guanosine-plus-cytosine (%GC). The nucleosomal-length DNAs were then used in a competitive reconstitution assay in order to quantitatively determine their relative abilities to form nucleosomes. The results of the assay indicate a linear dependence of the free energy of nucleosome formation on base composition and, surprisingly, show that several prokaryotic DNAs form nucleosomes as well as or better than eukaryotic DNAs.

Can one measure the free energy of binding of the histone octamer to different DNA sequences by salt-dependent reconstitution?

I explain why many recently reported measurements for the "free energy" of positioning of the histone octamer on different DNA sequences are likely to be in error: i.e. because histone octamers do not equilibrate between different DNA molecules at low salt, but only at high salt. Thus, the reported "free energies" refer to an equilibrium at high salt, under nearly dissociating conditions between protein and DNA, and they are likely to be much too small on an absolute scale. There are many other lines of evidence to suggest that the preferences of the histone octamer for different DNA sequences are rather strong and of importance in biological systems.

Local supercoil-stabilized DNA structures

The DNA double helix exhibits local sequence-dependent polymorphism at the level of the single base pair and dinucleotide step. Curvature of the DNA molecule occurs in DNA regions with a specific type of nucleotide sequence periodicities. Negative supercoiling induces in vitro local nucleotide sequence-dependent DNA structures such as cruciforms, left-handed DNA, multistranded structures, etc. Techniques based on chemical probes have been proposed that make it possible to study DNA local structures in cells. Recent results suggest that the local DNA structures observed in vitro exist in the cell, but their occurrence and structural details are dependent on the DNA superhelical density in the cell and can be related to some cellular processes.

Effects of DNA sequence and histone-histone interactions on nucleosome placement

Using competitive reconstitution, we have refined the parameters for the binding of histone octamers to artificial nucleosome-positioning sequences of the form: (A/T3nn(G/C)3nn. We find that the optimal period between flexible segments is approximately 10.1 base-pairs, supporting the view that the DNA on the nucleosome surface is overwound. The strongest requirement for flexible DNA is near the protein dyad. However, we see no indication of changes in DNA helical repeat in this region. Using a series of repetitive sequences, we confirm that neither all A/T-rich nor all G/C-rich regions are identical in promoting nucleosome formation. Surprisingly, A/T-rich segments containing the TpA step, subject to purine-purine clash in the minor groove, favor nucleosome formation over sequences lacking this step. Short tracts of adenine residues are found to position on the histone surface like other A/T-rich regions, in the manner predicted by the direction of their sequence-directed bends as determined by electrophoretic methods. Tracts containing five adenine residues are extremely aniostropic in their flexibility and are strongly detrimental to nucleosome formation when positioned for major groove compression. Longer adenine tracts are found to position near the ends of the nucleosomal DNA. However, other positions may be occupied by an A12 tract, with only a minor penalty in the free energy of nucleosome formation. Overall, reconstituted nucleosome positions are translationally degenerate, suggesting a weak dependence on DNA flexibility for nucleosome positioning. Dinucleosomal reconstitutions on tandem dimers of the 5 S RNA gene of Lytechinus variegatus demonstrate a weak phasing dependence for the interaction between nucleosomes. This interaction is maximal for the 202 base-pair repeat and suggests a co-operative mechanism for the formation of ordered nucleosomal arrays based on a combination of DNA flexibility and nucleosome-nucleosome interactions.