Histone-histone associations within chromatin. Cross-linking studies using tetranitromethane.

Treatment of chromatin with the protein cross-linker tetranitromethane (TNM) results in a product identified as an F2a1-F2b dimer. The same product appears after treatment with TNM of HeLa cells growing in culture. Furthermore acid-extracted histones which have been fractionated into the five separate species can be recombined and mixed with DNA to produce a nucleohistone preparation which is also cross-linked by TNM to give the F2a1-F2b dimer. F1 and F3 can be excluded from the reconstitution mixture without effect on the dimer production. In contrast, the presence of F2a2 is essential to the proper reconstitution of F2a1 and F2b with DNA. The specificity of TNM and the characteristics of the reaction suggest that F2a1 and F2b are cross-linked at their specific binding sites. These results provide evidence that F2a1, F2a2, and F2b interact specifically in chromatin.     

Histone-histone interactions as revealed by formaldehyde treatment of chromatin

Histone oligomers produced by formaldehyde treatment of chromatin were studied. It was shown that these histone oligomers could be converted into monomers by boiling in 0.1 NH₂SO₄. Dimers H2B-H4 and H2B-H2A were identified by two-dimensional polyacrylamide gel electrophoresis.Abbreviations. Histones Nomenclature H1 formerly histone F1 - H2B formerly histone F2b - H2A formerly histone F2a2 - H3 formerly histone F3 - H4 formerly histone F2a1 This nomenclature has been proposed at the recent Symposium on the Structure and Function of Chromatin. Ciba Foundation. London. April 1974.     

Histone-DNA interactions within chromatin. Isolation of histones from DNA-histone adducts induced in nuclei by UV light.

We have developed a method by which to isolate histones that have been crosslinked to DNA following irradiation of calf thymus nuclei by UV light. The procedure involves separation of protein-DNA adducts from uncrosslinked protein by Sepharose 4B chromatography under dissociating conditions. Histones which are crosslinked to DNA are released by chemical hydrolysis of the DNA and identified by SDS gel electrophoresis. The results indicate that, of the histones, H1 and H3 become crosslinked to the DNA most readily under our irradiation conditions.     

Mapping physical interactions within chromatin by proteomic approaches

Our ability to study protein-protein interactions has grown by leaps and bounds in recent years, enabling numerous large-scale studies to be performed in a variety of organisms. Despite this success, some classes of proteins, including those bound to chromatin, remain difficult to characterize through proteomic approaches. Some of the problems faced by researchers studying chromatin-bound proteins include low complex solubility, heterogeneous sample composition, and numerous transient interactions, which can be further complicated by the presence of DNA itself. To tackle these issues, a number of innovative protocols have been developed to better study the various facets of chromatin biology. In this review, we will discuss novel approaches to study protein-DNA interactions as well as protein complexes affecting chromatin.     

DNA-protein interactions in nucleosomes and in chromatin. Structural studies of chromatin stabilized by ultraviolet-light induced crosslinking.

Crosslinking induced by ultraviolet light irradiation at 254 nm has been utilized to investigate the structure of chromatin and isolated nucleosomes. The results presented here imply that the four core histones, as well as histone H1, have reactive groups within a bond length of the DNA bases. In nucleosomes depleted of H1, all of the core histones react similarly with the DNA and form crosslinks. In chromatin, the rate of crosslinking of all histones to DNA is essentially similar. Comparison of mononucleosomes, dinucleosomes and whole chromatin shows that the rate of crosslinking increases significantly with increasing number of connected nucleosomes. These differences in the rate of crosslinking are interpreted in terms of interactions between neighbouring nucleosomes on the chromatin fiber, which are absent in an isolated mononucleosome.     

Induction of stable protein-deoxyribonucleic acid adducts in Chinese hamster cell chromatin by ultraviolet light.

Ultraviolet (uv)-light-mediated formation of protein-DNA adducts in Chinese hamster cell chromatin was investigated in an attempt to compare chromatin alterations induced in vitro with those observed in vivo. Three independent methods of analysis indicated stable protein-DNA associations: (1) a membrane filter assay which retained DNA on the filter in the presence of high salt-detergent; (2) a Sepharose 4B column assay in which protein eluted cincident with DNA; and (3) a CsCl density gradient equilibrium assay which showed both protein and DNA banding at densities other than their respective native densities. Treatment of the irradiated chromatin with DNase provided further evidence that protein-DNA and not protein-protein adducts were being observed in the column assay. There is a fluence-dependent response of protein-DNA adduct formation when the chromatin is irradiated at low ionic strength and is linear for protein over the range studied. When the chromatin is exposed to differing conditions of pH, ionic strength, or divalent metal ion concentration, the quantity of adduct formed upon uv irradiation varies. Susceptibility to adduct formation can be partially explained in terms of the condensation state of the chromatin and other factors such as rearrangement, denaturation, and dissociation of the chromatin components. Besides providing information on the biological significance of these types of uv-induced lesions, this technique may be useful as a probe of chromatin structure.     

Histone-histone interactions. II. Structural stability of the histone H3-H4 complex.

The stability of the histone H3-H4 complex toward urea, changes in pH and ionic strength, and certain chemical modifications have been examined by gel electrophoresis anc circular dichronism. When uncomplexed, the two cysteine residues of histone H3 become rapidly oxidized, forming an intramolecular disulfide bridge which apparently blocks complex formation on return to complexing conditions. The complex was found to be unstable toward low values of pH and ionic strength, concentrations of urea exceeding 1 M, modifications of the cysteine residues, and fragmention in which the C terminal portions of either H3 or H4 are removed. A possible structure for this complex is proposed.     

Structural studies on Allium cepa L. chromatin: enhanced stability of internucleosome interactions in plant chromatin

The pattern of micrococcal nuclease digestion of chromatin from different organs of Allium cepa has been studied. The DNA from small oligonucleosomes appears to be highly degraded and heterogeneous. In solutions of intermediate ionic strength (0.15 M NaCl) histones H2A and H2B form dimers, however at high salt concentrations (2 M NaCl) they tend to form complexes of higher order, such as tetramers. It is proposed that a correlation exists between the ability of these proteins to form tetramers and the particular stability of internucleosome interactions.     

The use of ultraviolet light in the fractionation of chromatin containing unsubstituted and bromodeoxyuridine-substituted DNA.

Two procedures are described for the fractionation of chromatin containing unsubstituted (LL) DNA and DNA unifilarly substituted with bromodeoxyuridine (HL). The two procedures rely upon the sensitivity of bromodeoxyuridine-containing DNA to UV light to induce either strand breakage or protein crosslinking. When a mixture of LL and HL chromatin is irradiated with UV light, the HL DNA fragments into molecules of smaller molecular weight than the LL DNA and crosslinks more chromosomal protein than the LL DNA. LL and HL chromatin can be fractionated on the basis of size by centrifuging through a neutral sucrose gradient. The HL DNA-protein adducts that are generated by the UV light have a unique buoyant density and may be isolated by isopycnic centrifugation in CS2SO4. The ability to fractionate LL and HL chromatin permits certain studies on the structure of replicating chromatin.     

Crosslinking proteins to nucleic acids by ultraviolet laser irradiation.

Ultraviolet (UV) irradiation can initiate complex formation between proteins and DNA or RNA and so can be used to study such interactions. However, crosslink formation by standard UV light sources can take up to several hours. More recently, a beam of monochromatic UV light from a laser has been used to initiate crosslinking in nano- and picosecond time intervals. As noted in an earlier TIBS article 'the advantages of short pulse times and high-energy fluxes should make this a valuable technique in the future'. In this review we characterize laser-induced crosslinking and explore the applications of this method.     

Amino acid contacts between histones are the same for plants and mammals. Binding-site studies using ultraviolet light and tetranitromethane.

Leek chromatin has been cross-linked by UV light and tetranitromethane. The same major H2A--H2B and H2B--H4 cross-linked dimers are formed as in mammalian chromatin. CNBr peptide mapping shows that the cross-links occur in the same regions of the histone sequence for both plants and mammals. Interspecies complexes formed between leek and calf H2A and H2B can be cross-linked by UV light with the same specificity as intraspecies H2A--H2B complexes. We conclude that certain geometric features of histone-histone binding sites are conserved precisely during evolution despite large changes in the overall histone sequence. Moreover, our data show that identification of cross-linked amino acids using binding-site probes such as UV light and tetranitromethane can yield significant information about thermodynamically important contacts within histone-histone binding sites.     

Analysis of interactions between mutagens, II. Ethyl methanesulfonate and ultraviolet light in Saccharomyces cerevisiae

The results of this study indicate the existence of a strong interaction between ethyl methanesulfonate (EMS) and ultraviolet light (UV) for cell killing in the yeast Saccharomyces cerevisiae. Conversely, mutation and gene conversion frequencies observed for the combined treatment of EMS and UV do not deviate significantly from that expected on the basis of simple additivity. Studies involving repair-deficient mutants (rad mutants) reveal that the synergistic interaction for cell killing depends on RAD52 function (recombinational repair), but not on RAD3 function (excision repair). On the basis of this analysis, the interaction between EMS and UV in S. cerevisiae might arise from the inhibition of double-strand break repair by one, or both agents.     

Specific interactions of chromatin with the nuclear envelope: positional determination within the nucleus in Drosophila melanogaster.

Specific interactions of chromatin with the nuclear envelope (NE) in early embryos of Drosophila melanogaster have been mapped and analyzed. Using fluorescence in situ hybridization, the three-dimensional positions of 42 DNA probes, primarily to chromosome 2L, have been mapped in nuclei of intact Drosophila embryos, revealing five euchromatic and two heterochromatic regions associated with the NE. These results predict that there are approximately 15 NE contacts per chromosome arm, which delimit large chromatin loops of approximately 1-2 Mb. These NE association sites do not strictly correlate with scaffold-attachment regions, heterochromatin, or binding sites of known chromatin proteins. Pairs of neighboring probes surrounding one NE association site were used to delimit the NE association site more precisely, suggesting that peripheral localization of a large stretch of chromatin is likely to result from NE association at a single discrete site. These NE interactions are not established until after telophase, by which time the nuclear envelope has reassembled around the chromosomes, and they are thus unlikely to be involved in binding of NE vesicles to chromosomes following mitosis. Analysis of positions of these probes also reveals that the interphase nucleus is strongly polarized in a Rabl configuration which, together with specific targeting to the NE or to the nuclear interior, results in each locus occupying a highly determined position within the nucleus.     

Normal reconstruction of DNA supercoiling and chromatin structure in cockayne syndrome cells during repair of damage from ultraviolet light.

The chromatin of human cells undergoes structural rearrangements during excision repair of ultraviolet damage in DNA that were detected by transient relaxation of DNA supercoiling and increased staphylococcal nuclease digestibility of repaired sites. Inhibition of polymerization and/or ligation of repaired regions with inhibitors of DNA polymerase alpha (cytosine arabinoside and aphidicolin) resulted in the accumulation of single-strand breaks, delayed reconstruction of DNA supercoiling, and maintenance of the staphylococcal nuclease digestibility. These observations suggest that reconstruction of the native chromatin state requires completion of repaired regions with covalent ligation into the DNA strands. Although previous claims have been made that a late stage associated with ligation of repaired regions may be defective in cells from patients with Cockayne syndrome, complete reconstruction of the native chromatin occurred in cells from three unrelated patients after ultraviolet irradiation. No abnormality in repair was therefore detected in Cockayne syndrome cells. The hypersensitivity of cell survival and semiconservative DNA replication to damage by ultraviolet light in this human disorder must therefore be regarded as features of a primary defect in DNA metabolism unrelated to DNA repair.     

Inhibition of DNA replication by ultraviolet light.

DNA replication in ultraviolet-irradiated HeLa cells was studied by two different techniques: measurements of the kinetics of semiconservative DNA synthesis, and DNA fiber autoradiography. In examining the kinetics of semiconservative DNA synthesis, density label was used to avoid measuring the incorporation due to repair replication. The extent of inhibition varied with time. After doses of less than 10J/m2 the rate was initially depressed but later showed some recovery. After higher doses, a constant, low rate of synthesis was seen for at least the initial 6 h. An analysis of these data indicated that the inhibition of DNA synthesis could be explained by replication forks halting at pyrimidine dimers. DNA fiber autoradiography was used to further characterize replication after ultraviolet irradiation. The average length of labeled segments in irradiated cells increased in the time immediately after irradiation, and then leveled off. This is the predicted pattern if DNA synthesis in each replicon continued at its previous rate until a lesion is reached, and then halted. The frequency of lesions that block synthesis is approximately the same as the frequency of pyrimidine dimers.     

Complex interactions of caffeine and its structural analogs with ultraviolet light in cell killing

We measured the clonogenic survival response of cultured mouse $\text{10}\text{T}{}^{\mathrm{<mtext>1</mtext><mtext>2</mtext>}}$ cells exposed to UV light and caffeine post-treatment. When 0.5 and 1 mM caffeine were present for 24 h immediately following UV, the D₀ values of the biphasic survival curves suggest that one subpopulation was sensitized and one subpopulation was protected from killing by UV light. A cloned survivor from the radioprotected subpopulation responded to UV plus caffeine in identical manner as the parent cells. When the caffeine exposure was prolonged to 48 h, only the radiosensitizing effect was observed. Two demethylated analogs of caffeine were also tested. The response of $\text{10}\text{T}{}^{\mathrm{<mtext>1</mtext><mtext>2</mtext>}}$ cells to 1 mM theophylline present for 24 h after UV irradiation was approximately the same as that for the same treatment with 1mM caffeine. However, prolonging the theophylline exposure to 48 h failed to produce the same kind of potentiation of cell killing as that observed for caffeine. Xanthine by itself was as toxic to $\text{10}\text{T}{}^{\mathrm{<mtext>1</mtext><mtext>2</mtext>}}$ cells as caffeine, but had no synergistic effect as caffeine when given to UV-irradiated cells for 24 or 48 h. It is therefore unlikely that all the effects of caffeine on UV-irradiated cells are mediated by its demethylated metabolites.