Histone acetylation and heterochromatin content of cultured Peromyscus cells

In order to explore the relationship between unacetylated arginine-rich histones and condensed chromatin structure, the extent of histone acetylation was examined in cultured cell lines derived from three species of deer mice. These species differ considerably in their genomic content of heterochromatin but contain essentially the same euchromatin content. Cells of Peromyscus eremicus, containing 34–36% more constitutive heterochromatin than Peromyscus boylii or Peromyscus crinitus cells were found to contain 28–35% more unacetylated histone H4, 22–29% more unacetylated histone H3, and 18–22% more unacetylated histone H2B. This relationship between unacetylated histones and heterochromatin content was further explored by inducing hyperacetylation of P. eremicus and P. boylii histones through treatment of cells with 15 mM sodium butyrate for 24 h. It was found that the percentages of unacetylated histones H3 and H4 remaining after butyrate treatment were proportional to the amount of constitutive heterochromatin in the genome. These data support the concept that a small core of histones in constitutive heterochromatin is inaccessible to acetylation. It was also found that the acetylated state of isolated histones was sensitive to the method of histone extraction. Thus concern must be given to preparative procedures when studying histone acetylation in order to minimize these acetate losses.     

Heterochromatin and histone phosphorylation

Histone phosphorylation and nuclear structure have been compared in cultured cell lines of two related species of deer mice, Peromyscus crinitus and Peromyscus eremicus, which differ greatly in their heterochromatin contents but which contain essentially the same euchromatin content. Flow microfluorometry measurements indicated that P. eremicus contained 36% more DNA than did P. crinitus, and C-band chromosome staining indicated that the extra DNA of P. eremicus existed as constitutive heterochromatin. Two striking differences in interphase nuclear structure were observed by electron microscopy. Peromyscus crinitus nuclei contained small clumps of heterochromatin and a loose, amorphous nucleolus, while P. eremicus nuclei contained large, dense clumps of heterochromatin and a densely structured, well defined, nucleolonema form of nucleolus. Incorporation of ³²PO₄ into histones indicated that the steady-state phosphorylation of H1 was identical in P. crinitus and P. eremicus cells. In contrast, the phosphorylation rate of H2a was 58% greater in the highly heterochromatic chromatin of P. eremicus cells than in the lesser heterochromatic chromatin of P. crinitus cells, suggesting an involvement of H2a phosphorylation in heterochromatin structure. It is suggested that the three histone phosphorylations related to cell growth (H1, H2a, and H3) may be associated with different levels of chromatin organization: H1 interphase phosphorylation with some submicroscopic (molecular) level of organization, H2a phosphorylation with a higher level of chromatin organization found in heterochromatin, and H3 and H1 superphosphorylation with the highest level of chromatin organization observed in condensed chromosomes.     

Histone variants and histone modifications in chromatin fractions from heterochromatin-rich Peromyscus cells

In order to investigate the relationship between condensed heterochromatin and histone modification by acetylation, phosphorylation and amino acid variation, chromatin from cultured Peromyscus eremicus cells, containing 35% constitutive heterochromatin, was fractionated into heterochromatin-enriched and heterochromatin-depleted fractions. The constitutive heterochromatin content of these fractions was determined from satellite DNA content. The distribution of phosphorylated and acetylated histones and amino acid variants of histone H2A in these chromatin fractions was examined by gel electrophoresis. Fractionation of histones demonstrated that endogenous histone phosphatase activity was high in chromatin fractions and could not be inhibited sufficiently to allow accurate histone phosphorylation measurements. However, sodium butyrate did inhibit deacetylation activity in the fractions, allowing histone acetylation measurements to be made. It was found that the constitutive heterochromatin content of these fractions was proportional to both their unacetylated H4 content and their more-hydrophobic H2A content. These observations support, by direct measurement, earlier experiments (Exp cell res 111 (1978) 373; 125 (1980) 377; 132 (1981) 201) suggesting that constitutive heterochromatin is enriched in unacetylated arginine-rich histones, and in the more hydrophobic variant of histone H2A.     

Studies in heterochromatin DNA: accessibility of late replicating heterochromatin DNA in chromatin to micrococcal nuclease digestion

Heterochromatin DNA in cactus mouse (Peromyscus eremicus) replicates in the late S phase of cell cycle. A method of obtaining cells which contain DNA preferentially labeled at heterochromatic areas by a pulse-labeling of late replicating DNA is described. When the nuclei of P. eremicus cells containing radioactively labeled DNA in heterochromatin were digested with micrococcal nuclease and the resultant nucleosomal DNA was separated by gel electrophoresis, it was found that the repeat length of nucleosomal DNA in the heterochromatin DNA is not different from that of the bulk of the genomic DNA. Furthermore, there was no significant difference in the accessibility to digestion by micrococcal nuclease between the late replicating heterochromatin DNA and the total DNA under our digestion conditions. Two dimensional gel electrophoresis patterns of nucleosomal DNAs isolated from micrococcal nuclease digested nuclei from P. eremicus, P. collatus, and P. crinitus cells in culture were very similar. Cytogenetic data showed that these three species are different in heterochromatin but similar in euchromatin.     

Content of histone H1 and histone phosphorylation in relation to the higher order structures of chromatin in Drosophila

The amount of histone H1 relative to core histones has been determined in three Drosophila species (D. melanogaster, D. texana and D. virilis) in chromatin from several tissues differing in chromatin structure and genetic activity. Low levels of H1 were found in relatively undifferentiated, early embryos as well as in a line of cultured cells. In late embryos the content of H1 was highest in D. virilis which possesses larger amounts of and a partially more compacted constitutive heterochromatin than the two other species. Polytene chromatin from larval salivary glands showed increased levels of H1 compared with diploid chromatin and the degree of phosphorylation of this histone was relatively low. The degree of phosphorylation of H2A was found to be drastically reduced in polytene as compared with diploid embryonic chromatin, which parallels the extensive underreplication of constitutive heterochromatin. Also, in diploid chromatin a qualitative correlation was observed between the relative amounts of heterochromatin and the levels of H2A phosphorylation. These findings suggest a connection between H2A phosphorylation and heavy compaction of interphase chromatin.     

Histone H1 heterogeneity in the midge, Chironomus thummi. Structural comparison of the H1 variants in an organism where their intrachromosomal localization is possible

1. Seven subfractions of histone H1 have been isolated and purified from larvae of Chironomus thummi (Diptera). They have been denominated I-1, II-1, II-2, II-3, III-1, III-2, and III-3, according to the order of migration in two steps of preparative electrophoresis. 2. The amino acid compositions are similar to those of other H1 histones. Subfractions I-1 and II-1 were found to contain one methionine and two tyrosine residues, II-2 contained two methionine and three tyrosine residues, and III-1 one methionine and three tyrosine residues. The other subfractions contained one or two methionine and two or three tyrosine residues. For subfractions I-1 and II-1 a chain length of about 252 amino acids was estimated. 3. Peptide pattern analyses after chemical cleavage at the methionine and tyrosine residues, and enzymatic cleavage with thrombin and chymotrypsin, respectively, showed that all subfractions have different individual primary structures. A comparison of peptide sizes and of the positions in the peptide patterns of epitopes recognized by monoclonal antibodies was made to check whether some of the subfractions could arise by proteolytic degradation of others. This possibility can be excluded for five of the subfractions and is very improbable for the two others. Treatment of C. thummi H1 with alkaline phosphatase did not change the pattern of subfractions, while the phosphorylated subfraction of histone H2A disappeared after this treatment. Most and very probably all subfractions are thus H1 sequence variants. 4. Inbred strains and individual larvae of C. thummi were found to comprise all seven variants. The H1 heterogeneity can therefore not be due to allelic polymorphism. Salivary gland nuclei were found to contain variant I-1 and at least some of the other variants. 5. H1 from Drosophila melanogaster and from calf thymus were used as reference molecules in all cleavage experiments and yielded the peptide patterns expected from the sequence. The comparison discriminates the group of C. thummi H1 histones clearly from Drosophila and calf thymus H1. Limited trypsin digestion yielded a protected peptide of uniform size in six of the seven variants which was considerably smaller than the protected central domain of calf thymus H1. 6. Two other species of Chironomidae, C. pallidivittatus and Glyptotendipes barbipes were found to contain five and three H1 subfractions, respectively.(ABSTRACT TRUNCATED AT 400 WORDS)     

Histone content in relation to amount of heterochromatin and developmental stage in three species of Drosophila

Relative amounts of various histone fractions in Drosophila chromatin were estimated densitometrically on electrophoretic gel separations. Several consistent and highly significant differences were obtained between larval and adult chromatin. The arginine-rich histones showed the most conspicuous changes: higher amounts of H4 in larvae, higher H3 in adults. The level of modification of these histones was clearly higher in larval than in adult chromatin. The modification of the two slower subfractions of H4 involved, in all probability, phosphorylation as well as acetylation. In all types of Drosophila chromatin studied 50% or more of the H2a molecules were phosphorylated—a remarkably high proportion. The species differences observed in relative amounts of histone were consistent in both stages of development. D. melanogaster differed from D. hydei and D. virilis in all histones except H2b, while the latter two species were generally similar. The interspecific variation in histone pattern was generally not correlated to differences in content of heterochromatin. The level of modification of H3 was, however, presumably an exception, as it was significantly lower for both larvae and adults in D. virilis than in the other two species. These differ from D. virilis in containing appreciably lower proportions of heterochromatic chromosome segments.     

Two new chromodomain-containing proteins that associate with heterochromatin in <Emphasis Type="Italic">Sciara coprophila</Emphasis> chromosomes

We report here the molecular and cytological characterization of two proteins, ScoHET1 and ScoHET2 (for Sciara coprophila heterochromatin), which associate to constitutive heterochromatin in the dipteran S. coprophila. Both proteins, ScoHET1 of 37 kDa and ScoHET2 of 44 kDa, display two chromodomain motifs that contain the conserved residues essential for the recognition of methylated histone H3 at lysine 9. We raised antibodies to analyze the chromosomal location of ScoHET1 and ScoHET2 in somatic and germline cells. In S. coprophila polytene chromosomes, both proteins associate to the pericentromeric regions and to the heterochromatic subterminal bands of the chromosomes. In germinal nuclei, ScoHET1 and ScoHET2 proteins distribute to the heterochromatic regions of the regular chromosome complement and are abundantly present along the heterochromatic germline-limited “L” chromosomes. We investigated histone methylation modifications and found that all heterochromatic regions enriched in ScoHET1/ScoHET2 proteins exhibit high levels of di- and tri-methylated histone H3 at lysine 9. Taken together, our results support that the association of ScoHET1/ScoHET2 to heterochromatin is mediated by histone H3K9 methylation. Using 5-methylcytosine antibodies, we proved the cytological detection of DNA methylation in S. coprophila. From our observations in L germline chromosomes, heterochromatin in S. coprophila is highly enriched in DNA 5-methylcytosine residues. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Susceptibility of heterochromatin to aphidicolin-induced chromosomal breakage

The distribution of aphidicolin-induced chromosomal lesions was analyzed to determine the relative breakage susceptibility of euchromatin and heterochromatin in the cactus mouse, Peromyscus eremicus. The observed breakage was tested against expected distributions corresponding to the karyotypic proportions of autosomal euchromatin, autosomal heterochromatin, X-chromosomal euchromatin, and X-chromosomal heterochromatin. The distribution of induced breakage was independent of sex but dependent on the individual. In all individuals, there was a highly significant (P0.0001) deficiency in the number of breaks observed as compared to expected in autosomal heterochromatin. Sparse observations in the X chromosome and the absence of breaks in the Y chromosome precluded valid statistical tests of the sex-chromosomal distribution of induced breakage. These data indicate that the autosomal heterochromatin of Peromyscus is resistant to aphidicolin-induced chromosomal breakage and argue against a simple relationship between late replication and a general mechanism for chromosomal fragility.     

A qualitative and quantitative study of subfractions of the histone H10 in various mammalian tissues.

The histone H10 was examined from seven mammalian species. All tissues were shown to contain two subfractions of H10, except for those of rabbit, in which little or no H10 was found. The subfraction composition was compared quantitatively in different mouse and hamster tissues, with the conclusion that this composition is tissue-specific. It is proposed that the wide occurrence of H10, together with the evidence of no more or less than two subfractions wherever it occurs, and the tissue-specific nature of the ratio of subfractions, signify that these two subfractions have specific individual functions.     

Interplay between histone deacetylase SIR-2, linker histone H1 and histone methyltransferases in heterochromatin formation

Maintenance of intact heterochromatin structure through epigenetic mechanisms is essential for cell survival. Defects in heterochromatin formation caused by loss of chromatin-modifying enzymes lead to genomic instability and cellular senescence. The NAD+-dependent histone deacetylase SIR-2 and the H1 linker histone are intriguing chromatin elements that are connected to chromatin regulation and cell viability in the single cellular eukaryotic organism yeast. However, it remains an open question how SIR-2 and H1 mediate heterochromatin formation in simple multi-cellular organisms such as C. elegans and in even more complex organisms such as mammals. Recently we have identified SIR-2.1 and the H1 histone subtype, HIS-24 as factors involved in heterochromatin regulation at subtelomeric regions in C. elegans. In addition we show that SIR-2.1, HIS-24, and MES-2, a ortholog to Enhancer of zeste E(Z) are functionally related in heterochromatin formation contributing to fertility and embryogenesis. Here we discuss the interplay between SIR-2, H1 histone and histone methyltransferases in modulation of chromatin structure in further detail.     

Cathepsin L stabilizes the histone modification landscape on the Y chromosome and pericentromeric heterochromatin

Posttranslational histone modifications and histone variants form a unique epigenetic landscape on mammalian chromosomes where the principal epigenetic heterochromatin markers, trimethylated histone H3(K9) and the histone H2A.Z, are inversely localized in relation to each other. Trimethylated H3(K9) marks pericentromeric constitutive heterochromatin and the male Y chromosome, while H2A.Z is dramatically reduced at these chromosomal locations. Inactivation of a lysosomal and nuclear protease, cathepsin L, causes a global redistribution of epigenetic markers. In cathepsin L knockout cells, the levels of trimethylated H3(K9) decrease dramatically, concomitant with its relocation away from heterochromatin, and H2A.Z becomes enriched at pericentromeric heterochromatin and the Y chromosome. This change is also associated with global relocation of heterochromatin protein HP1 and histone H3 methyltransferase Suv39h1 away from constitutive heterochromatin; however, it does not affect DNA methylation or chromosome segregation, phenotypes commonly associated with impaired histone H3(K9) methylation. Therefore, the key constitutive heterochromatin determinants can dynamically redistribute depending on physiological context but still maintain the essential function(s) of chromosomes. Thus, our data show that cathepsin L stabilizes epigenetic heterochromatin markers on pericentromeric heterochromatin and the Y chromosome through a novel mechanism that does not involve DNA methylation or affect heterochromatin structure and operates on both somatic and sex chromosomes.     

Correlation between histone phosphorylation and tumor ageing in Ehrlich ascites tumor cells

The histone fraction F1 can be divided into subfractions by gel electrophoresis. The microheterogeneity of F1 histone has been investigated in EAT cells in mice between 3 and 16 days after inoculation. The cell number per mouse increases during the first 8 days (proliferation phase); thereafter it remains constant (non-proliferating phase). We could demonstrate that the number of F1 subfractions is reduced from 5 in proliferating cells to 3 in non-proliferating ones. In short term experiments using [³²P]phosphate the label was only found in F1 histone from proliferating cells but not in that from resting cells. However, F2a1 histone, which is the other phosphorylated histone in interphase cells, was labelled in young and old cell populations. When ³²P-labelled F1 histone was treated with alkaline phosphatase not only was the label split off but also the number of subfractions was reduced from 5 to 3. These results lend additional evidence to the hypothesis that at least some of the F1 subfractions are derived from the same protein by phosphorylation.     

Study of the relationship between the subfraction composition of histone F1 and the type of tissue in birds.

The subfraction composition of lysine-rich histone has been studied with the aid of polyacrylamide gel electrophoresis. The subfraction compositions of the histone F1 of several tissues from the chicken, pigeon, and titmouse have been compared. The histone F1 from the tissues investigated consists of four or five subfractions of similar number and electrophoretic mobility (1, 1a, 2, 3, and 4). In the different avian species each subfraction varied its mobility independently of the others. The chicken tissues investigated can be divided into two classes, depending on the relative concentration of subfractions 2 and 3 (A and B): Class A (subfraction 2 is smaller than 3) includes the brain, liver, skeletal muscle, heart, muscular layer of the stomach, and pancreas, and class B (subfraction 2 is larger than 3) includes the intestinal mucosa, thymus, and testes, as well as the liver, heart, and pancreas from a 21-day embryo. Such a division of the tissues corresponds to the varying rate of their cellular renewal. In a parallel examination of the relative concentrations of the individual subfractions in the same tissues from the three avian species it has been found that the relative concentration of subfractions 3 and 2 is increased in the skeletal muscles, heart, brain, and liver, that subfraction 2 is increased in the intestinal mucosa, that subfractions 4 and 3 are increased in the pancreas, and that subfractions 1, 1a, and 4 are increased in the erythrocytes. The results obtained may be interpreted as a consequence of some relationship between the subfraction composition of histone F1 and the type of tissue of the source.     

Does high-mobility-group non-histone protein HMG 1 interact specifically with histone H1 subfractions?

The interaction of the non-histone chromosomal protein HMG (high-mobility group) 1 with histone H1 subfractions was investigated by equilibrium sedimentation and n.m.r. sectroscopy. In contrast with a previous report [Smerdon & Isenberg (1976) Biochemistry 15, 4242--4247], it was found, by using equilibrium-sedimentation analysis, that protein HMG 1 binds to all three histone H1 subfractions CTL1, CTL2, and CTL3, arguing against there being a specific interaction between protein HMG 1 and only two of the subfractions, CTL1 and CTL2. Raising the ionic strength of the solutions prevents binding of protein HMG 1 to total histone H1 and the three subfractions, suggesting that the binding in vitro is simply a non-specific ionic interaction between acidic regions of the non-histone protein and the basic regions of the histone. Protein HMG 1 binds to histone H5 also, supporting this view. The above conclusions are supported by n.m.r. studies of protein HMG 1/histone H1 subfraction mixtures. When the two proteins were mixed, there was little perturbation of the n.m.r. spectra and there was no evidence for specific interaction of protein HMG 1 with any of the subfractions. It therefore remains an open question as to whether protein HMG 1 and histone H1 are complexed together in chromatin.     

Intranuclear distribution of mouse liver nuclear DNA polymerase

Adult mouse liver nuclei and their subfractions corresponding to heterochromatin, nucleoli, membranes, and euchromatin were studied for DNA-polymerase activity. The intact nuclei and the two heavy nuclear fractions contained rather low activity while the two light fractions (membranes and euchromatin) had no activity at all. In the two heavy fractions, the activity was stimulated by β-mercaptoethanol and depressed by p-hydroxymercuribenzoate and by omission of one or more nucleotides. A nuclease activity, detected in the intact nuclei, may also be present in the nuclear subfractions. DNA-polymerase activity in the heavy fractions of mouse liver nuclei is discussed in relation to other published results.