Chromatosome positioning on assembled long chromatin. Linker histones affect nucleosome placement on 5 S rDNA

Long chromatin containing linker histones H1 or H5 was assembled on tandemly repeated 172 or 207 base-pair nucleosome positioning sequences from a sea urchin 5 S RNA gene. The effects of H1 and H5 on spacing and positioning of nucleosomes were assessed. In the absence of linker histones, precise determinations of core particle boundaries showed that, although a large proportion of the histone octamers occupy a unique position, there is a small group of other, less populated sites located around this major site. The dominant position was found 10 to 15 base-pairs upstream from the unique position previously reported for the histone octamer on the monomer 260 base-pair sequence. Linker histones do not override the underlying DNA signals that induce the very regular spacing of nucleosomes in chromatins assembled on these strongly positioning multimer DNA sequences. They were nevertheless found to be decisive in determining the chromatosome positions and their distributions, and as such define the chromatosome as a positioning entity.     

Reconstitution experiments show that sequence-specific histone-DNA interactions are the basis for nucleosome phasing on mouse satellite DNA

The molecular basis underlying the sequence-specific positioning of nucleosomes on DNA was investigated. We previously showed that histone octamers occupy multiple specific positions on mouse satellite DNA in vivo and have now reconstituted the 234 bp mouse satellite repeat unit with pure core histones into mononucleosomes. Histones from mouse liver or chicken erythrocytes bind to the DNA in multiple precisely defined frames in perfect phase with a diverged 9 bp subrepeat of the satellite DNA. This is the first time that nucleosome positions on a DNA in vivo have been compared to those found on the same DNA by in vitro reconstitution. Most of the nucleosomes occupy identical positions in vivo and in vitro. There are, however, some characteristic differences. We conclude that sequence-dependent histone-DNA interactions play a decisive role in the positioning of nucleosomes in vivo, but that the nucleosome locations in native chromatin are subject to additional constraints.     

Nucleosome core particles of calf thymus, Tetrahymena, and the reconstituted hybrid. Their structure reflects the nature of the histone octamer

The circular dichroism spectra and the thermal denaturation profiles of the nucleosome core particles isolated by micrococcal nuclease digestion from nuclei of calf thymus and the protozoan Tetrahymena pyriformis were compared with those of the homogeneous and hybrid core particles reconstituted from calf core DNA and either calf or Tetrahymena histone octamer. The core DNA was obtained from the calf core particle, and both the histone octamers were reconstituted from the acid-extracted four core histones of calf thymus or Tetrahymena, whose amino acid sequences show the largest differences hitherto known. The reconstituted homogeneous core particle was identical in both the physical properties with the isolated calf core particle, showing that the correct reconstitution was achieved. The circular dichroism spectra of the calf and Tetrahymena core particles and the hybrid core particle showed no essential differences, indicating that the three core particles have the same overall structure. The derivative thermal-denaturation profiles, however, clearly differed; the calf core particle showed two melting transitions at 60 degrees C and 72 degrees C, while the Tetrahymena and hybrid core particles showed the same three transitions at 48-50 degrees C, 60-61 degrees C, and 72 degrees C. Thus, the thermal denaturation properties of nucleosome core particles do not reflect the nature of DNA, but rather that of the histone octamer bound to the DNA. We conclude that the Tetrahymena histones are more weakly bound to the DNA than the calf thymus histones in the same overall structure of nucleosomes.     

Effect of Z-DNA on nucleosome placement

Histone octamers were reconstituted on plasmids carrying the alternating nucleotide sequence (G-C)15. The plasmids, radioactively labeled at one of two neighboring sites near the (G-C) insert, were digested with micrococcal nuclease. Nucleosome core particles were isolated and the monomer DNA subjected to restriction analysis. Quite different results are obtained if the reconstitution is carried out with relaxed plasmids, in which the (G-C) insert is in the B form, or with supercoiled plasmids, where it is in the Z form. With supercoiled plasmids, there is a marked reduction (compared with relaxed plasmids) in the abundance of labeled monomers, the result of a large decrease in core particles carrying any (G-C) sequence. Some core particles formed on supercoiled (Z) plasmids are positioned either just outside the (G-C) sequence, or with the sequence occupying the terminal position within the core particle. In contrast, monomers obtained from relaxed plasmids incorporate the (G-C) sequence in the B form more or less randomly in the interior of the core particle; species showing discrete positioning make only a minor contribution. We conclude that DNA in the Z form cannot be incorporated within core particles, except at their termini, and that a transition from the B to the Z form in vivo might result in a significantly altered local placement of nucleosomes.     

Unique positioning of reconstituted nucleosomes occurs in one region of simian virus 40 DNA

A direct end label method was used to study the positioning of nucleosome arrays on several long (greater than 2200 base pairs) SV40 DNA fragments reconstituted in vitro with core histones. Comparison of micrococcal nuclease cutting sites in reconstituted and naked DNA fragments revealed substantial differences in one DNA region. When sufficient core histones were annealed with the DNA to form closely spaced nucleosomes over most of the molecule, a uniquely positioned array of four nucleosomes could be assigned, by strict criteria, to a 610-base pair portion of the SV40 "late region," with a precision of about +/- 20 base pairs. In some other DNA regions, a number of alternative nucleosome positions were indicated. The uniquely positioned four-nucleosome array spanned the same 610 nucleotides on two different DNA fragments that possessed different ends. Removal of a DNA region that had contained a terminal nucleosome of the array, by truncation of the fragment before reconstitution, did not affect the positioning of the other three nucleosomes. As the core histone to DNA ratio was lowered, evidence for specific positioning of nucleosomes diminished, except within the region where the four uniquely positioned nucleosomes formed. This region, however, does not appear to have a higher affinity for core histones than other regions of the DNA.     

Structural features of a regulatory nucleosome

DNA sequences from the long terminal repeat of the mouse mammary tumor virus (MMTV-LTR) position nucleosomes both in vivo and in vitro. Here, were present chromatin reconstitution experiments showing that MMTV-LTR sequences from -236 to +204 accommodate two histone octamers in positions compatible with the in vivo data. This positioning is not influenced by the length of the DNA fragment and occurs in linear as well as in closed circular DNA molecules. MMTV-LTR DNA sequences show an intrinsic bendability that closely resembles its wrapping around the histone octamer. We propose that bendability is responsible for the observed rotational nucleosome positioning. Translational nucleosome positioning seems also to be determined by the DNA sequence. These data, along with the results from reconstitution experiments with insertion mutants, support a modular model of nucleosome phasing on MMTV-LTR, where the actual positioning of the histone octamer results from the additive effect of multiple features of the DNA sequence.     

DNA distortion as a factor in nucleosome positioning.

In a previous report we constructed a synthetic DNA sequence that directed the deposition of histone octamers to a single site, and it was proposed that DNA distortion was involved in the positioning effect. In the present study we utilized the chemical probe potassium permanganate to identify sites of DNA distortion in the synthetic positioning sequence. A permanganate hypersite was identified 15 bp from the nucleosome pseudo-dyad at a site known to display DNA distortion in the mature nucleosome. The sequence of the site contained a TA step flanked by an oligo-pyrimidine tract. A series of substitutions were made in the region of the permanganate hypersite and the resulting constructs tested for affinity for histone octamers and translational positioning in in vitro studies. The results revealed that either a single base substitution at the TA step or in the adjacent homopolymeric tract dramatically affected affinity and positioning activity. The rotational orientation of the permanganate-sensitive sequence was shown to be important for functions, since altering the orientation of the site in a positioning fragment reduced positioning activity and octamer affinity, while altering the rotational orientation of the sequence in a non-positioning fragment had the opposite effects. A reconstituted 5 S rDNA positioning sequence from Lytechinus variegatus was also shown to display a permanganate hypersite 16 bp from its pseudo-dyad.     

DNA folding by histones: the kinetics of chromatin core particle reassembly and the interaction of nucleosomes with histones.

The kinetics of the chromatin core particle reassembly reaction in solution were quantitatively studied under conditions such that nucleohistone aggregation did not occur. Core particles, salt-jumped rapidly by dilution from 2.5 m-NaCl (in which DNA and histones do not interact) to 0.6 m-NaCl (in which core particles are nearly intact), reassemble in two distinct time ranges. Approximately 75% of the DNA refolds into core particle-like structures “instantaneously” as measured by several physical and chemical techniques with dead times in the seconds to minutes time range. The remaining DNA refolds with relaxation times ranging from 250 minutes at 0 °C to 80 minutes at 37 °C; this slow effect cannot be attributed to sample heterogeneity. The fraction of slowly refolding DNA and the slow relaxation time are independent of the core particle concentration. Transient intermediates present during the slow phase of refolding were identified as free DNA and core particle-like structures containing excess histone. Mixing experiments with DNA, histones, and core particles showed that core particle-histone interactions are responsible for the slow kinetics of DNA refolding. Upon treatment of reassembling core particles with the protein crosslinking reagent, dimethylsuberimidate, the slow phase of the reassembly reaction was arrested and a 13 S particle containing DNA and two octamers of histone was isolated. Consistent with the nature of this kinetic intermediate, it is shown that in 0.6 m-NaCl, core particles co-operatively bind at least one additional equivalent of histones with high affinity in the form of excess octamers. Also, core particles continue to adsorb considerably more histones with a weaker association constant of the order 10⁵m^?1 (in units of octamers) to a maximum value of 12 ± 2 equivalents (octamers) per core particle. The sedimentation coefficient increases with the two-thirds power of the molecular weight of the complex, as it would in the case of clustered spheres.A reassembly mechanism consistent with the data is presented, and other simple mechanisms are excluded. In the proposed mechanism, core particles reassemble very rapidly and compete effectively with DNA for histones such that approximately one-third of the particles initially formed are complexed with an excess octamer of histones, and 25% of the total DNA remains uncomplexed. The amount of this unusual reaction intermediate decays slowly to an equilibrium value of about 10%, thereby leaving 9% of the total DNA uncomplexed. Approximate values are calculated for the free energies, rate constants, and two of the activation energies which characterize this migrating octamer mechanism. This mechanism provides a means whereby histone octamers can be temporarily stripped off DNA at a modest free energy cost, approximately 2.6 kcal per nucleosome. Also, the properties of excess histone adsorption by chromatin and octamer migration suggest an efficient mechanism, consistent with observations by others, for nucleosome assembly in vivo during replication.     

S(T)PXX motifs promote the interaction between the extended N-terminal tails of histone H2B with "linker" DNA.

The assembly of hybrid core particles onto long chicken DNA with histone H2B in the chicken histone octamer replaced with either wheat histone H2B(2) or sea urchin sperm histone H2B(1) or H2B(2) is described. All these histone H2B variants have N-terminal extensions of between 18 and 20 amino acids, although only those from sea urchin sperm have S(T)PXX motifs present. Whereas chicken histone octamers protected 167 base pairs (bp) (representing two full turns) of DNA against micrococcal nuclease digestion (Lindsey, G. G., Orgeig, S., Thompson, P., Davies, N., and Maeder, D. L. (1991) J. Mol. Biol. 218, 805-813), all the hybrid histone octamers protected an additional 17-bp DNA against nuclease digestion. This protection was more marked in the case of hybrid octamers containing sea urchin sperm histone H2B variants and similar to that described previously (Lindsey, G. G., Orgeig, S., Thompson, P., Davies, N., and Maeder, D. L. (1991) J. Mol. Biol. 218, 805-813) for hybrid histone octamers containing wheat histone H2A variants all of which also have S(T)PXX motifs present. Continued micrococcal nuclease digestion reduced the length of DNA associated with the core particle via 172-, 162-, and 152-bp intermediates until the 146-bp core particle was obtained. These DNA lengths were approximately 5 bp or half a helical turn longer than those reported previously for stripped chicken chromatin and for core particles containing histone octamers reconstituted using "normal" length histone H2B variants. This protection pattern was also found in stripped sea urchin sperm chromatin, demonstrating that the assembly/digestion methodology reflects the in vivo situation. The interaction between the N-terminal histone H2B extension and DNA of the "linker" region was confirmed by demonstrating that stripped sea urchin sperm chromatin precipitated between 120 and 500 mM NaCl in a manner analogous to unstripped chromatin whereas stripped chicken chromatin did not. Tryptic digestion to remove all the histone tails abolished this precipitation as well as the protection of DNA outside of the 167-bp core particle against nuclease digestion.     

The in vitro reconstitution of nucleosome and its binding patterns with HMG1／2 and HMG14／17 proteins

Using atomic force microscopy (AFM), the dynamic process of the in vitro nucleosome reconstitution followed by slow dilution from high salt to low salt was visualized. Data showed that the histone octamers were dissociatedfrom DNA at 1M NaC1. When the salt concentration was slowly reduced to 650 mM and 300 mM, the core histones bound to the naked DNA gradually. Once the salt concentration was reduced to 50 mM the classic “beads-on-a-string“ structure was clearly visualized. Furthermore, using the technique of the in vitro reconstitution of nucleosome,the mono- and di- nucleosomes were assembled in vitro with both HS2core (-10681 to -10970 bp) and NCR2 (-372to -194 bp) DNA sequences in the 5‘flanking sequence of human b-globin gene. Data revealed that HMG 1/2 and HMG 14/17 proteins binding to both DNA sequences are changeable following the assembly and disassembly of nucleosomes. We suggest that the changeable binding patterns of HMG 14/17 and HMG1/2 proteins with these regulatory elements may be critical in the process of nucleosome assembly, recruitment of chromatin-modifying activities, and the regulation of human b-globin gene expression.     

Homogeneous reconstituted oligonucleosomes, evidence for salt-dependent folding in the absence of histone H1

Using the method of salt dialysis, we have reconstituted histone octamers onto DNA templates consisting of 12 tandem repeats, each containing a fragment of the sea urchin 5S rRNA gene [Simpson, R.T., Thoma, F., & Brubaker, J.M. (1985) Cell 42, 799-808]. In these templates, each sea urchin repeat contains a sequence for preferred nucleosome positioning. Sedimentation velocity and sedimentation equilibrium studies in the analytical ultracentrifuge indicate that at molar histone/DNA ratios of 1.0-1.1 extremely homogeneous preparations of fully loaded oligonucleosomes (12 nucleosomes/template) can be regularly obtained. Digestion of the oligonucleosomes with micrococcal nuclease, followed by restriction mapping of purified nucleosome-bound DNA sequences, yields a complicated but consistent pattern of nucleosome positioning. Roughly 50% of the nucleosomes appear to be phased at positions 1-146 of each repeat, while the remainder of the nucleosomes occupy a number of other minor discrete positions along the template that differ by multiples of 10 bp. From sedimentation velocity studies of the oligonucleosomes in 0-0.2 M NaCl, we observe a reversible increase in mean sedimentation coefficient by almost 30%, accompanied by development of heterogeneity in sedimentation. These results, in combination with theoretical predictions, indicate that linear stretches of chromatin in the absence of lysine-rich histones exist in solution in a salt-dependent equilibrium between an extended (low salt) conformation and one or more folded (high salt) structures. In addition, by 100 mM NaCl, salt-dependent dissociation of histone octamers from these linear oligonucleosomes is observed.     

Sequence-specific interaction of histones with the simian virus 40 enhancer region in vitro

DNA fragments containing either one or both of the 72-base pair (bp) elements which constitute the SV40 enhancer and the three adjacent 21-bp repeats were associated with histone octomers from chicken erythrocytes in vitro. Both fragments formed complexes with electrophoretic mobilities of nucleosomes containing the appropriate length of DNA. Analysis of DNase I cutting of uniquely end-labeled complexes suggests that the fragment containing a single 72-bp element forms a positioned core particle. Control experiments show that positioning is not due to the 21-bp repeats or to end effects. The fragment with a tandem repeat of the 72-bp element also does not associate randomly with histones. The data are consistent with formation of a core particle on one or the other of the repeated enhancer sequences. We discuss possible functional consequences of such nucleosome positioning.     

Construction of nucleosome cores from defined DNA sequences of prokaryotic origin.

A procedure for the de novo construction of nucleosome core particles from defined DNA sequences of prokaryotic origin is described. Efficient de novo reconstitution without added carrier DNA is demonstrated. DNase I and exonuclease III analysis of a nucleosome core prepared from a 154 base pair fragment extending from base 853 to base 1006 of pBR322 indicates a non-random positioning of the histone core along the DNA. As bacteria have no histones, their DNA cannot be expected to have a histone core positioning signal encoded in it, the efficient formation of a uniquely positioned core particle is not self evident. The possibility that a phosphate end group positions DNA fragments on the histone is considered. The de novo reconstitution of carrier-less defined nucleosome core particles should facilitate the physicochemical study of nucleosomes on the fine structural level.     

The site of binding of linker histone to the nucleosome does not depend upon the amino termini of core histones.

Using nucleosomes reconstituted on a defined sequence of DNA, we have investigated the question as to whether the N-terminal tails of core histones play a role in determining the site of binding of a linker histone. Reconstitutes used histone cores of three types: intact, lacking the N-terminal H3 tails, or lacking all tails. In each case the same, single defined position for the histone core was observed, using high-resolution mapping. The affinity for binding of linker histone H1(o) was highest for the intact cores, lowest for the tailless cores. However, the location of the linker histone, as judged by micrococcal nuclease protection, was exactly the same in each case, an asymmetric site of about 17 bp to one side of the core particle DNA.     

DNA engineering shows that nucleosome phasing on the African green monkey alpha-satellite is the result of multiple additive histone-DNA interactions

The mechanism underlying sequence-specific positioning of nucleosomes on DNA was investigated. African green monkey alpha-satellite DNA was reconstituted in vitro with histones. Histone octamers were found to adopt one major and several minor positions on the satellite repeat unit, very similar to those positions found previously in vitro, demonstrating that sequence-specific histone-DNA interactions are responsible for nucleosome positioning on this DNA. In order to understand the nature of these interactions in more detail, we have constructed a variant satellite fragment containing an insertion of half a helical DNA turn. The parent fragment directs histones to one major and two overlapping minor positions that are all affected by the insertion. All three frames respond in a unique fashion to the additional five base-pairs. From a quantitative analysis of the nucleosome positions on the engineered fragment, consensus "phasing boxes" as the basis for nucleosome positioning can be ruled out. Instead, our results argue very strongly that nucleosome positioning is due to the independent contribution of many different DNA-histone contacts along the entire core particle, in an apparently additive fashion.     

Limitations of the poly(glutamic acid) reconstitution method in the reassembly of mono- and dinucleosomes

Reconstitution of mononucleosomes and dinucleosomes at physiological ionic strength by means of poly(glutamic acid) is not efficient at physiological histone octamer:DNA ratios, unlike that with the salt dialysis method. The shorter the DNA is, the less transfer of octamers from poly(glutamic acid) to DNA occurs. By increasing the octamer:DNA ratio it is possible to involve all the DNA in the assembly, but for DNA longer than core particle length, nucleoprotein particles containing extra histones are concomitantly generated. Except for core particle and chromatosome lengths of DNA reassembled at 0.6:1 or 1:1 octamer:DNA ratio (and thus with low yield), reconstituted nucleoprotein particles proved to be different from native nucleosomes by their insolubility upon isolation. In the aggregates, DNA ends seemed to be sufficiently loose to allow exonuclease III digestion up to a certain limit. This resulted in patterns that for some cloned DNA fragments could give the impression, without knowledge of the above, of resulting from a unique octamer position. In view of the small range of length of DNA and the low yield of faithful reconstitution, the assembly method using poly(glutamic acid) is only of limited use in mono- or dinucleosome reconstitution experiments, at least in our hands.     

Acetylated nucleosome assembly on telomeric DNAs

The role of histone N-terminal domains on the thermodynamic stability of nucleosomes assembled on several different telomeric DNAs as well as on 'average' sequence DNA and on strong nucleosome positioning sequences, has been studied by competitive reconstitution. We find that histone tails hyperacetylation favors nucleosome formation, in a similar extent for all the examined sequences. On the contrary, removal of histone terminal domains by selective trypsinization causes a decrease of nucleosome stability which is smaller for telomeres compared to the other sequences examined, suggesting that telomeric sequences have only minor interactions with histone tails. Micrococcal nuclease kinetics shows enhanced accessibility of acetylated nucleosomes formed both on telomeric and 'average' sequence DNAs. These results suggest a more complex role for histone acetylation than the decrease of electrostatic interactions between DNA and histones.     

Extended C-terminal tail of wheat histone H2A interacts with DNA of the "linker" region

The preparation of hybrid histone octamers with wheat histone H2A variants replacing chicken H2A in the chicken octamer is described. The fidelity of the reconstituted hybrid octamers was confirmed by dimethyl suberimidate cross-linking. Polyglutamic-acid-mediated assembly of these octamers on long DNA and subsequent micrococcal nuclease (MNase) digestion demonstrated that, whereas chicken octamers protected 167 base-pairs (representing 2 full turns of DNA), hybrid histone octamers containing wheat histone H2A(1) with its 19 amino acid residue C-terminal extension protected an additional 16 base pairs of DNA against nuclease digestion. The protection observed by hybrid histone octamers containing wheat histone H2A(3) with both a 15 residue N-terminal and a 19 residue C-terminal extension was identical with that observed with H2A(1)-containing hybrid histone octamers with only the 19 residue C-terminal extension. These results suggest that the role of the C-terminal extension is to bind to DNA of the "linker" region. The thermal denaturation of chicken and hybrid core particles was identical in 10 mM-Tris.HCl.20 mM-NaCl, 0.1 mM-EDTA, confirming that there was no interaction between the basic C-terminal extension and DNA of the core particle. Denaturation in EDTA, however, showed that hybrid core particles had enhanced stability, suggesting that the known conformational change of core particles at very low ionic strength allows the C-terminal extension to bind to core particle DNA under these conditions. A model accounting for the observed MNase protection is presented.