Natural abundance carbon-13 nuclear magnetic resonance studies of histone and DNA dynamics in nucleosome cores

Natural abundance carbon-13 nuclear magnetic resonance spectra (67.9 MHz) were obtained for native nucleosome cores: cores dissociated in 2 M NaCl and 2 M NaCl, 6 M urea; and cores degraded with DNase I plus proteinase K. Phosphorus-31 NMR spectra of native and dissociated cores and core length DNA were also obtained at 60.7 MHz. The 31P resonance and spin-lattice relaxation time (T1) of DNA were only slightly affected by packaging in nucleosome cores, in agreement with other reports, but 13C resonances of DNA were essentially unobservable. The loss of DNA spectral intensity suggests that rapid internal motions of DNA sugar carbons in protein-free DNA previously demonstrated by 13C NMR methods are partly restricted in nucleosomes. The 13C spectrum of native cores contains many narrow intense resonances assigned to lysine side chain and alpha-carbons, glycine alpha-carbons, alanine alpha- and beta- carbons, and arginine side chain carbons. Several weaker resonances were also assigned. The narrow line widths, short T1 values, and non-minimal nuclear Overhauser enhancements of these resonances, including alpha- and beta-carbons, show that some terminal chain segments of histones in nucleosomes are as mobile as small random coil polypeptides. The mobile segments include about 9% of all histone residues and 25% of all lysines, but only 10% of all arginines. The compositions of these segments indicate that mobile regions are located in amino- or carboxyl-terminal sequences of two or more histones. In addition, high mobility was observed for side chain carbons of 45-50% of all lysines (delta and epsilon carbons) and about 25% of all arginines (zeta carbon) in histones (including those in mobile segments), suggesting that basic residues in terminal histone sequences are not strongly involved in nucleosome structure and may instead help stabilize higher order chromatin structure.     

The accessibilities of histones in nucleosome cores to an arginine-specific protease

Submaxillaris protease, reportedly specific for arginine residues, was used to probe accessible arginines of chicken erythrocyte nucleosome cores. The relative rates of digestion of histones in nucleosome cores by this protease were H3 greater than H2b greater than H4 much greater than H2a. At most, 8 of 52 total arginines among the four core histones were reasonably accessible to attack. Sites most rapidly cleaved were Arg-26 in H3, a site approximately 13 residues from the NH2 terminus of H2b, and Lys-12 or Arg-17 in H4. Five sites attacked more slowly were Arg-8, -128, or -129 and Arg-49, -52, or -53 in H3; Arg-3 and Arg-17 or -19 in H4; and a site near one terminus of H2b or H3. H2a and fragments resulting from the above cleavages were highly resistant to attack, even after prolonged incubation. Similar results were obtained upon digestion of histones in intact chromatin. H1 and H5 in whole chromatin were attacked at rates comparable to H3. Seven of the eight accessible sites lie outside of structure-forming histone sequences, i.e. sequences that are immobilized in histone complexes, indicating that these sequences are inaccessible in nucleosome cores. The single exceptional site noted, approximately 50 residues from the NH2 terminus of H3, is consistent with previous observations that Glu-51 and Glu-60 of H3 in nucleosomes are accessible to attack by S. aureus protease (Rill, R. L., and Oosterhof, D. K. (1981). J. Biol. Chem. 256, 12687-12691). The relationships of these protease accessibilities to NMR assignments of mobile histone tails in nucleosome cores are discussed.     

Different mechanism for in vitro formation of nucleosome core particles

The interaction of different histone oligomers with nucleosomes has been investigated by using nondenaturing gel electrophoresis. In the presence of 0.2 M NaCl, the addition of the pairs H2A,H2B or H3,H4 or the four core histones to nucleosome core particles produces a decrease in the intensity of the core particle band and the appearance of aggregated material at the top of the gel, indicating that all these histone oligomers are able to associate with nucleosomes. Equivalent results were obtained by using oligonucleosome core particles. Additional electrophoretic results, together with second-dimension analysis of histone composition and fluorescence and solubility studies, indicate that H2A,H2B, H3,H4, and the four core histones can migrate spontaneously from the aggregated nucleosomes containing excess histones to free core DNA. In all cases the estimated yield of histone transfer is very high. Furthermore, the results obtained from electron microscopy, solubility, and supercoiling assays demonstrate the transfer of excess histones from oligonucleosomes to free circular DNA. However, the extent of solubilization obtained in this case is lower than that observed with core DNA as histone acceptor. Our results demonstrate that nucleosome core particles can be formed in 0.2 M NaCl by the following mechanisms: (1) transfer of excess core histones from oligonucleosomes of free DNA, (2) transfer to excess H2A,H2B and H3,H4 associated separately with oligonucleosomes to free DNA, (3) transfer to excess H2A,H2B initially associated with oligonucleosomes to DNA, followed by the reaction of the resulting DNA-(H2A,H2B) complex with oligonucleosomes containing excess H3,H4, and (4) a two-step transfer reaction similar to that indicated in (3), in which excess histones H3,H4 are transferred to DNA before the reaction with oligonucleosomes containing excess H2A,H2B. The possible biological implications of these spontaneous reactions are discussed in the context of the present knowledge of the nucleosome function.     

The action of chymotrypsin on nucleosome cores. Histone products and conformational effects of limited digestion

alpha-Chymotrypsin was used to probe accessible hydrophobic amino acid residues in nucleosome cores. Small amounts of chymotrypsin rapidly and selectively cleaved at leucine 20 of histone H3. Cleavage at this site caused partial unfolding of the nucleosome core at low ionic strengths indicated by a small decrease in sedimentation coefficient and increase in circular dichroism in the 265-285-nm range. Unfolding did not occur at moderate ionic strengths, probably because of more effective electrolyte screening of residual negative charge on the nucleosome core. More extensive treatment with chymotrypsin partially degraded other core histones in nucleosome cores at similar rates. The primary sites of cleavage were assigned to Leu115 of H2a, Val18 or Gln22 of H2b, and Leu10 plus Leu22 of H4. We conclude that these primary sites of chymotrypsin cleavage of the four core histones lie on or near the nucleosome core surface, while the large number of other hydrophobic histone residues located in more central sequences must be inaccessible. Extensive chymotrypsin treatment yielded a set of "limit" products approximately 80-100 residues long that were similar to the limit products of trypsin digestion. Sedimentation coefficients and circular dichroism spectra of nucleosome cores treated to near limits with chymotrypsin or chymotrypsin followed by trypsin were not consistent with significant unfolding of the proteolyzed cores at moderate ionic strength. These results indicate that the amino-terminal 20-30 residues of H2b, H3, and H4 and the amino- and carboxyl-terminal approximately 12 residues of H2a, in toto, interact weakly if at all with DNA of isolated nucleosome cores. These histone termini stabilize less than two turns and perhaps only one turn on each DNA terminus.     

Localization of testis-variant histones in rat testis chromatin.

Nucleosome core particles and oligonucleosomes were isolated by digesting rat testis nuclei with micrococcal nuclease to 20% acid-solubility, followed by fractionation of the digest on a Bio-Gel A-5m column. The core particles thus isolated were characterized on the basis of their DNA length of 151 +/- 5 base-pairs and sedimentation coefficient of 11.4S. Analysis of the acid-soluble proteins of the core particles indicated that histones TH2B and X2 are constituents of the core particles, in addition to the somatic histones H2A, H2B, H3 and H4. The acid-soluble proteins of the oligonucleosomes comprised all the histones, including both the somatic (H1, H2A, H2B, H3, H4 and X2) and the testis-specific ones (TH1 and TH2B). It was also observed that histones TH1 and H1 are absent from the core particles and were readily extracted from the chromatin by 0.6 M-NaCl, which indicated that both of them are bound to the linker DNA.     

The mechanism of nucleosome assembly onto oligomers of the sea urchin 5 S DNA positioning sequence

We have used a model system composed of tandem repeats of Lytechinus variegatus 5 S rDNA (Simpson, R. T., Thoma, F., and Brubaker, J. M. (1985) Cell 42, 799-808) reconstituted into chromatin with chicken erythrocyte core histones to investigate the mechanism of chromatin assembly. Nucleosomes are assembled onto the DNA template by mixing histone octamers and DNA in 2 M NaCl followed by stepwise dialysis into very low ionic strength buffer over a 24-h period. By 1.0 M NaCl, a defined intermediate composed of arrays of H3.H4 tetramers has formed, as shown by analytical and preparative ultracentrifugation. Digestion with methidium propyl EDTA.Fe(II) indicates that these tetramers are spaced at 207 base pair intervals, i.e. one/repeat length of the DNA positioning sequence. In 0.8 M NaCl, some H2A.H2B has become associated with the H3.H4 tetramers and DNA. Surprisingly, under these conditions DNA is protected from methidium propyl EDTA.Fe(II) digestion almost as well as in the complete nucleosome, even though these structures are quite deficient in H2A.H2B. By 0.6 M NaCl, nucleosome assembly is complete, and the MPE digestion pattern is indistinguishable from that observed for oligonucleosomes at very low ionic strength. Below 0.6 M NaCl, the oligonucleosomes are involved in various salt-dependent conformational equilibria: at approximately 0.6 M, a 15% reduction in S20,w that mimics a conformational change observed previously with nucleosome core particles; at and above 0.1 M, folding into a more compact structure(s); at and above 0.1 M NaCl, a reaction involving varying amounts of dissociation of histone octamers from a small fraction of the DNA templates. In low ionic strength buffer (less than 1 mM NaCl), oligonucleosomes are present as fully loaded templates in the extended beads-on-a-string structure.     

Histone H4 K16Q mutation, an acetylation mimic, causes structural disorder of its N-terminal basic patch in the nucleosome

Histone tails and their posttranslational modifications play important roles in regulating the structure and dynamics of chromatin. For histone H4, the basic patch K(16)R(17)H(18)R(19) in the N-terminal tail modulates chromatin compaction and nucleosome sliding catalyzed by ATP-dependent ISWI chromatin remodeling enzymes while acetylation of H4 K16 affects both functions. The structural basis for the effects of this acetylation is unknown. Here, we investigated the conformation of histone tails in the nucleosome by solution NMR. We found that backbone amides of the N-terminal tails of histones H2A, H2B, and H3 are largely observable due to their conformational disorder. However, only residues 1-15 in H4 can be detected, indicating that residues 16-22 in the tails of both H4 histones fold onto the nucleosome core. Surprisingly, we found that K16Q mutation in H4, a mimic of K16 acetylation, leads to a structural disorder of the basic patch. Thus, our study suggests that the folded structure of the H4 basic patch in the nucleosome is important for chromatin compaction and nucleosome remodeling by ISWI enzymes while K16 acetylation affects both functions by causing structural disorder of the basic patch K(16)R(17)H(18)R(19).     

Role of the histone "tails" in the folding of oligonucleosomes depleted of histone H1.

An oligonucleosome 12-mer was reconstituted in the absence of linker histones, onto a DNA template consisting of 12 tandemly arranged 208-base pair fragments of the 5 S rRNA gene from the sea urchin Ly-techinus variegatus (Simpson, R. T., Thoma, F. S., and Burbaker, J. M. (1985) Cell 42, 799-808). The ionic strength-dependent folding of this nucleohistone complex was compared with that of a native oligonucleosome fraction obtained from chicken erythrocyte chromatin, which had been carefully stripped of linker histones and fractionated in sucrose gradients. The DNA of this native fraction exhibited a narrow size distribution centered around the length of the 208-12 DNA template used in the reconstituted complex. These two complexes displayed very similar hydrodynamic behavior as judged by sedimentation velocity analysis. By combining these data with electron microscopy analysis, it was shown that the salt-dependent folding of oligonucleosomes in the absence of linker histones involves the bending of the linker DNA region connecting adjacent nucleosomes. It was also found that selective removal by trypsin of the N-terminal regions ("tails") of the core histones prevents the oligonucleosome chains from folding. Thus, in the absence of these histone domains, the bending of the linker DNA necessary to bring the nucleosomes in contact is completely abolished. In addition to the complete lack of folding, removal of the histone tails results in an unwinding at low salt of a 20-base pair region at each flanking side of the nucleosome core particle. The possible functional relevance of these results is discussed.     

Mapping the interaction surface of linker histone H1(0) with the nucleosome of native chromatin in vivo

H1 linker histones stabilize the nucleosome, limit nucleosome mobility and facilitate the condensation of metazoan chromatin. Here, we have combined systematic mutagenesis, measurement of in vivo binding by photobleaching microscopy, and structural modeling to determine the binding geometry of the globular domain of the H1(0) linker histone variant within the nucleosome in unperturbed, native chromatin in vivo. We demonstrate the existence of two distinct DNA-binding sites within the globular domain that are formed by spatial clustering of multiple residues. The globular domain is positioned via interaction of one binding site with the major groove near the nucleosome dyad. The second site interacts with linker DNA adjacent to the nucleosome core. Multiple residues bind cooperatively to form a highly specific chromatosome structure that provides a mechanism by which individual domains of linker histones interact to facilitate chromatin condensation.     

Counting nucleosomes in living cells with a combination of fluorescence correlation spectroscopy and confocal imaging

Although methods for light microscopy of chromatin are well established, there are no quantitative data for nucleosome concentrations in vivo. To establish such a method we used a HeLa clone expressing the core histone H2B fused to the enhanced yellow fluorescent protein (H2B-EYFP). Quantitative gel electrophoresis and fluorescence correlation spectroscopy (FCS) of isolated oligonucleosomes show that 5% of the total H2Bs carry the fluorescent tag and an increased nucleosome repeat length of 204 bp for the fluorescent cells. In vivo, the mobility and distribution of H2B-EYFP were studied with a combination of FCS and confocal imaging. With FCS, concentration and brightness of nascent molecules were measured in the cytoplasm, while in the nucleoplasm a background of mobile fluorescent histones was determined by continuous photobleaching. Combining these results allows converting confocal fluorescence images of nuclei into calibrated nucleosome density maps. Absolute nucleosome concentrations in interphase amount up to 250 microM locally, with mean values of 140(+/-28)microM, suggesting that a condensation-controlled regulation of site accessibility takes place at length scales well below 200 nm.     

Secondary structure of acyl carrier protein as derived from two-dimensional 1H NMR spectroscopy

Sequence-specific assignments of 1H NMR resonances were obtained for the backbone protons in acyl carrier protein (ACP) from Escherichia coli, a protein of 77 residues. The observations, in the NOESY spectra, of 1H-1H sequential and medium-range connectivities indicate the presence of three or four alpha-helical segments joined by short sequences of mixed conformations. The observations are used to refine a secondary structure model previously proposed on the basis of a Chou-Fasman algorithm [Rock, C. O., & Cronan, J. E., Jr. (1979) J. Biol. Chem. 254, 9778-9785].     

Nucleosome linking number change controlled by acetylation of histones H3 and H4

High levels of acetylation of lysines in the amino-terminal domains of all four core histones, H2A, H2B, H3, and H4, have been shown to reduce the linking number change per nucleosome core particle in reconstituted minichromosomes (Norton, V. G., Imai, B. S., Yau, P., and Bradbury, E. M. (1989) Cell 57, 449-457). Because there is evidence to suggest that the acetylations of H3 and H4 have functions that are distinct from those of H2A and H2B, we have determined the nucleosome core particle linking number change in minichromosomes containing fully acetylated H3 and H4 and very low levels of acetylation in H2A and H2B. This linking number change was -0.81 +/- 0.05, in close agreement with the linking number change for hyperacetylated nucleosome core particles which contain high levels of acetylation in all four core histones (approximately 70% of full acetylation in H3 and H4). Therefore, high levels of acetylation of H3 and H4 alone are responsible for the reduction in the linking number change per nucleosome core particle.     

DNase I site mapping and micrococcal nuclease digestion of pachytene chromatin reveal novel structural features

A comparison of the DNase I digestion products of the 32P-5'-end-labeled pachytene nucleosome core particles (containing histones H2A, TH2A, X2, H2B, TH2B, H3, and H4) and liver nucleosome core particles (containing somatic histones H2A, H2B, H3, and H4) revealed that the cleavage sites that are 30, 40, and 110 nucleotides away from the 5'-end are significantly more accessible in the pachytene core particles than in the liver core particles. These cleavage sites correspond to the region wherein H2B interacts with the nucleosome core DNA. These results, therefore, suggest that the histone-DNA interaction at these sites in the pachytene core particles is weaker, possibly because of the presence of the histone variant TH2B interacting at similar topological positions in the nucleosome core as that of its somatic counterpart H2B. Such a loosened structure may also be maintained even in the native pachytene chromatin since micrococcal nuclease digestion of pachytene nuclei resulted in a higher ratio of subnucleosomes (SN4 + SN7) to mononucleosomes than that observed in liver chromatin.     

Binding of linker histones to the core nucleosome

Binding of chicken erythrocyte linker histones H1/H5 to the core nucleosome has been studied. Histones H1/H5 bind very efficiently to the isolated core nucleosome in vitro. The binding of linker histones to the core nucleosome is associated with aggregation of the particles. Approximately one molecule of linker histone binds per core nucleosome in the aggregates, irrespective of the concentration of the linker histones and the salt used. Histone H5 shows greater binding affinity to the core nucleosome as compared to H1. The carboxyl-terminal fragment of the linker histones binds strongly to the core nucleosome while the binding of the central globular domain is weak. Each core nucleosome is capable of binding two molecules of carboxyl-terminal fragment of linker histone. The core nucleosome containing one molecule of carboxyl-terminal fragment of linker histone requires higher salt concentration for aggregation while the core nucleosome containing two molecules of carboxyl-terminal fragment of linker histone can self-associate even at lower salt concentrations. On the basis of these results we are proposing a novel mechanism for the condensation of chromatin by linker histones and other related phenomena.     

Conformational change of adrenodoxin induced by reduction of iron-sulfur cluster. Proton nuclear magnetic resonance study.

Bovine adrenodoxin in the reduced form has been measured by one- and two-dimensional 1H NMR spectroscopy. By comparing the spectrum of reduced adrenodoxin with that of the oxidized protein, resonances have been assigned for the aromatic residues. The spin-lattice relaxation time for the resonances due to histidine residues was found to depend on the reduction state of adrenodoxin. The distance from the paramagnetic center is calculated by using the Solomone-Bloembergen equation. The resonances from Tyr-82 and Ala-81 show large chemical shift changes upon reduction of adrenodoxin. The conformational change of adrenodoxin manifested by chemical shift difference between reduced and oxidized forms is found in the sequence around Tyr-82 and Ala-81. Modification with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide at Glu-74, Asp-79, and Asp-86 inhibited the interaction with both adrenodoxin reductase and cytochrome P-450scc (Lambeth, D. J., Geren, L. M., and Millett, F. (1984) J. Biol. Chem. 259, 10025-10029; Geren, L. M., O'Brien, P., Stonehuerner, J., and Millett, F. (1984) J. Biol. Chem. 259, 2155-2160). Thus, the sequence of these amino acids was assigned to the interaction site with the redox partners. The present 1H NMR investigation of adrenodoxin demonstrates that a conformational change upon reduction of the iron-sulfur cluster occurs in the sequence of negatively charged amino acids that is a putative site for interaction with redox partners. This could offer the structural basis of the electron transfer mechanism in which adrenodoxin functions as a mobile electron carrier.     

Characterization of nucleosomes consisting of the human testis/sperm-specific histone H2B variant (hTSH2B)

We have reported earlier the occurrence of a specific histone H2B variant in human testis and sperm. Here we have structurally characterized this protein, its association with the rest of the histone octamer, and its effects on the nucleosome structure. We show that a reconstituted octamer consisting of hTSH2B and a stoichiometric complement of histones H2A, H3, and H4 exhibits a lower stability compared to the reconstituted native counterpart consisting of H2B. In contrast, the hTSH2B containing octamers are able to form nucleosome core particles which are structurally and dynamically indistinguishable from those reconstituted with octamers consisting of only native histones. Furthermore, the presence of hTSH2B in the nucleosome does not affect its ability to bind to linker histones.