Thermal denaturation of nucleosomal core particles

Heats of thermal denaturation of chromatin core particles and core particles with covalently crosslinked histones were measured by differential scanning calorimetry. The additional stabilization of the nucleoprotein complex by crosslinking is not reflected in the transition enthalpy. The contribution of protein denaturation to the total heat was estimated by comparison of core particles with core particle DNA in a hifh-alt solution. By taking into account the temperature dependence of the transition entialpy of DNA, we conclude that the enthalpy change for denaturation of DNA in core particles is nearly the same as that for naked DNA in solution. The nature of the forces responsible for maintaining DNA in a condensed state in the cell nucleus is not yet understood. It is established that the first order of DNA compaction is achieved by the wrapping of from 145 to about 200 base pair length of DNA around histone cores containing two molecules each of the four smaller histones (1-3). These structures, called nucleosomes, are the fundamental repeating units of chromatin (1-3). As a first step toward understanding the DNA compaction process, it is of interest to know the stability of nucleosomal DNA relative to that of naked DNA, as well as the enthalpic and entropic contributions to the free energy of stabilization. In this paper we report the results of a thermodynamic study of the stability of nucleosome core particles, employing differential scanning calorimetry. MATERIALS AND METHODSPreparation of Core Particles. Chicken erythrocyte nuclei were prepared as described by Hymer and Kuff (4). The nuclei were swollen in 0.25 mM EDTA and the very lysine rich histones (Hi, H5) were removed by repeated suspension of the chromatin in 0.7 M NaCl/0.01 M Tris-HCI, pH 8.0/1 mM EDTA (5). The chromatin was subsequently digested with micrococcal nuclease (Worthington) and the digest was fractionated by sucrose gradient centrifugation as described (5). Core particles thus purified contain the four smaller histones in equal amounts, approximately 145 base pairs of DNA, and no HI or H5 histones.

Section: Methodssupporting

confidence: 92%

Section: Methodssupporting

confidence: 84%

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Stability of DNA in nucleosomes.

Bina¹,

Sturtevant²,

Stein³

1980

“…Perhaps the entire complement of DNA within the particle behaves in this way, but it is also possible that only a certain fraction, near the ends, can break loose from the surface of the core particle so that it effectively becomes part of the linker DNA (see for example refs. [13][14][15]. Our results would then suggest that the portion of DNA remaining bound to the core must have an average superhelix density close to -0. superhelix density averaged over 230 bp is about -0.046.…”

Section: Resultsmentioning

confidence: 95%

Effects of DNA supercoiling on the topological properties of nucleosomes.

Garner¹,

Felsenfeld²,

O’Dea³

et al. 1987

In the nucleosome core particle, at least 145 base pairs of DNA are bound to the histone octamer in a superhelical conformation. We have asked what effect the presence of these particles has on the ability of DNA gyrase to supercoil DNA. Synthetic minichromosomes, constructed by reconstituting complexes of core histones with the closed circular plasmid pBR322, were treated with various amounts of DNA gyrase. We have found that the maximum level of supercoiling that is attainable is nearly identical for proteinfree plasmids and for plasmids half-saturated with core histones, even though supercoiling does not result in a loss of histones from the complex. It appears that, at sufficiently high levels of supercoiling, the core particle is disrupted in such a way that the DNA bound to histones is no longer constrained.Although there is considerable evidence linking the biological function of prokaryotic DNA to its topological state, the role of DNA supercoiling in eukaryotes is more obscure. Data from transfection experiments in eukaryotic cells suggest that expression ofgenes introduced on plasmids depends upon the presence of intact closed circular DNA (1). Other evidence suggests that chromatin domains, either genomic (2) or extrachromosomal (3), may exist in some circumstances in torsionally stressed states; it has been argued that such states are associated with expression ofgenes within the domain. In every case, however, the assessment of the role of supercoiling is complicated because of the presence of histones, which alter torsional stress when they bind to topologically isolated DNA domains.In the earliest experiments designed to measure the topological effects of histone binding, complexes of core histone octamers and relaxed closed circular DNA were incubated with topoisomerase I, and the resulting linking number change (AL) was measured after the DNA had been freed of protein (4). A recent determination by this method gives a value of AL = -1.0/core particle (5). The binding of a histone octamer thus has a topological effect equivalent to the unwinding of one turn of DNA duplex. Since the final linking number measurements were made after treatment with topoisomerase I, the observed values refer to a DNA-protein complex that is under no torsional stress.DNA within the nucleus is not necessarily in such a relaxed state, however. As noted above, there is some evidence that certain regions of the genome are under stress. This could in principle arise either through the removal ofhistone octamers from a topologically isolated domain or by the action of a putative DNA gyrase. In either case, the way in which the stress is distributed through the domain will be determined by the properties of the DNA in the remaining nucleosomes of the domain. Something is known of the effect of bound histone octamers on the torsional properties of the surrounding DNA when the DNA is relaxed (6), but there have been no measurements of the properties of DNA within nucleosomes associated with plasmids that are superhelic...

“…In this model there can be interaction of DNA in one core (18), that amounts up to 20 bp at each end could be so loosened.…”

Section: Discussionmentioning

confidence: 99%

Organization of spacer DNA in chromatin.

Lohr¹,

Holde²

1979

Detailed analysis of the DNA fragment patterns produced by DNase I digestion of yeast, HeLa, and chicken erythrocyte nuclei reveals surprising features of nucleosome phasing. First, the spacer regions in phased yeast chromatin must be of lengths (lOi + 5) base pairs, where m = 0, 1, 2,2.... This feature is not seen in parallel studies of chicken erythrocyte chromatin. The 5-base pair increment in the yeast spacer imposes interesting restraints on the higher order structure of yeast chromatin. Second DNase I digestion of yeast, chicken erythrocyte, and HeLa nuclei produces "extended ladders" of bands of single-strand DNA fragments extending to well beyond the size of a nucleosomal repeat (1). This shows that the interparticle DNA, so-called linker or spacer DNA, occurs in discrete sizes in at least some fraction of the chromatin of these species. We have used the term "phasing" to denote this characteristic of spacer organization, because a discrete spacer requires that adjacent core particles be in a precise location relative to one another. Phasing, as used here, is not meant to suggest there is a precise location of a nucleosome with respect to a given DNA sequence. We present in this paper further observations on nucleosome phasing that provide at least partial explanation for many of the details of DNase I cleavage patterns in yeast and other eukaryotes.EXPERIMENTAL PROCEDURES HeLa, chicken erythrocyte, and yeast nuclei (the last from growing and stationary cells) were isolated as described (2, 3). DNase I digestion and DNA extraction were performed as in ref. 2. DNA was electrophoresed on either 5.5% or 8% polyacrylamide/98% formamide gels as described (4) or 8, 12, 13, or 15.5% polyacrylamide/urea gels prepared as described (5). Gels were analyzed as in ref. The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact. the two series overlap. In fact, a plot of band number vs. gel mobility shows precisely this behavior (Fig. 2). Both series are quite regular; however, in the overlap region there is a transition from the "low" (10-z110 b) pattern to the "high" (>140 b) series.We have accurately calibrated the band sizes of the yeast and chicken erythrocyte DNA and of our PM2 Hae III restriction fragments against OX174 Hae III, simian virus 40 Hae III, and pBR322 Hae III restriction fragments, whose sizes are exactly known from DNA sequence determinations (6-8). Results are shown in Table 1. Notice that (i) in the overlap region the two patterns differ by about 4-6 b, quantitatively showing the amount of DNA by which the high and low series of bands are out of phase and (ii) clearly the interval between bands in both the low and high series slightly exceeds 10 b.A graphical representation of the size data (Fig. 3) shows a clear discontinuity in the sizes of the two series of bands obtained by DNase I digestion of yeast nuclei. The lower series can be fi...