Histone H1 and the conformation of transcriptionally active chromatin

Structure and function of active chromatin and DNase I hypersensitive sites

Chromatin is by its very nature a repressive environment which restricts the recruitment of transcription factors and acts as a barrier to polymerases. Therefore the complex process of gene activation must operate at two levels. In the first instance, localized chromatin decondensation and nucleosome displacement is required to make DNA accessible. Second, sequence-specific transcription factors need to recruit chromatin modifiers and remodellers to create a chromatin environment that permits the passage of polymerases. In this review I will discuss the chromatin structural changes that occur at active gene loci and at regulatory elements that exist as DNase I hypersensitive sites.

Transcriptional frequency and cell determination

The relative base composition of DNA regulatory sequences of certain genes of undetermined multipotent progenitor cells may account for the frequency of transcription of these genes in cell determination. The sequences of these regulatory regions of cell determination genes that are more AT-rich would create the potential for transcription at a higher frequency due to their lower melting temperature, as well as propensity to bend. An increase of one or more of the high mobility group (HMG) chromatin proteins would preferentially bind the more AT-rich regulatory sequences, thereby increasing the rate of transcription. The amount of unphosphorylated H1 histone reacting with these same regulatory sites may decrease transcription frequency. The level of cell growth, i.e. total protein synthesis of a cell, is correlated positively with the synthesis of HMG proteins. H1 histone synthesis is linked to DNA replication. Unbalanced growth would alter the amounts of HMG proteins and H1 histone, thus changing transcriptional frequency. The greater the enrichment of AT sequences in the regulatory regions of the cell determination genes, the greater may be the extent of evolutionary conservation. Higher frequency of transcription of the cell determination genes with the more AT-rich regulatory sequences could account for the earlier expression of the more conserved cell determination genes during embryonic development. Preferential binding of H1 histone to the more AT-rich regulatory sequences would subsequently restrict their transcription before that of less conserved cell determination genes.

The C-terminal domain is the primary determinant of histone H1 binding to chromatin in vivo

We have used a combination of kinetic measurements and targeted mutations to show that the C-terminal domain is required for high-affinity binding of histone H1 to chromatin, and phosphorylations can disrupt binding by affecting the secondary structure of the C terminus. By measuring the fluorescence recovery after photo-bleaching profiles of green fluorescent protein-histone H1 proteins in living cells, we find that the deletion of the N terminus only modestly reduces binding affinity. Deletion of the C terminus, however, almost completely eliminates histone H1.1 binding. Specific mutations of the C-terminal domain identified Thr-152 and Ser-183 as novel regulatory switches that control the binding of histone H1.1 in vivo. It is remarkable that the single amino acid substitution of Thr-152 with glutamic acid was almost as effective as the truncation of the C terminus to amino acid 151 in destabilizing histone H1.1 binding in vivo. We found that modifications to the C terminus can affect histone H1 binding dramatically but have little or no influence on the charge distribution or the overall net charge of this domain. A comparison of individual point mutations and deletion mutants, when reviewed collectively, cannot be reconciled with simple charge-dependent mechanisms of C-terminal domain function of linker histones.

Histone H1 and chromatin interactions in human fibroblast nuclei after H1 depletion and reconstitution with H1 subfractions

BACKGROUND

Linker histones constitute a family of lysine-rich proteins associated with nucleosome core particles and linker DNA in eukaryotic chromatin. In permeabilized cells, they can be extracted from nuclei by using salt concentration in the range of 0.3 to 0.7 M. Although other nuclear proteins are also extracted at 0.7 M salt, the remaining nucleus represents a template that is relatively intact.

METHODS

A cytochemical method was used to study the affinity of reconstituted linker histones for chromatin in situ in cultured human fibroblasts. We also investigated their ability to condense chromatin by using DNA-specific osmium ammine staining for electron microscopy.

RESULTS

Permeabilized and H1-depleted fibroblast nuclei were suitable for the study of linker histone-chromatin interactions after reconstitution with purified linker histone subfractions. Our results showed that exogenous linker histones bind to chromatin with lower affinity than the native ones. We detected no significant differences between the main H1 and H1 degrees histone fractions with respect to their affinity for chromatin or in their ability to condense chromatin.

CONCLUSIONS

Linker histone interactions with chromatin are controlled also by mechanisms independent of linker histone subtype composition.

Replication timing and cell differentiation

Cell differentiation may depend in part upon a type of unbalanced growth in which several cell cycles occur with a reduced level of total protein synthesis. During this period the synthesis of the chromatin protein HMG-I/Y is reduced since its synthesis is correlated with that of total protein. The synthesis of histone H1 shows less reduction since its synthesis is entrained with that of DNA. This greater reduction of HMG-I/Y than of histone H1 is thought to delay or prevent replicon initiations within AT-enriched isochores. This shifts their time of replication from early to late S phase. This may restrict certain pathways of cell differentiation in multipotent progenitor cells and allow one particular type of differentiation.

The type of DNA attachment sites recovered from nuclear matrix depends on isolation procedure used

A large variety of DNA sequences have been described in nuclear matrix attachment regions. It could be most likely a result of the different methods used for their isolation. The idea about how different types of known DNA sequences (strongly attached to the nuclear matrix, weakly attached, or not attached) directly participate in anchoring DNA loops to the nuclear matrices isolated by different experimental procedures was tested in this study. Matrix-attached (M) and matrix-independent or loop (L) fractions as well as nuclear matrices were isolated using extractions of nuclei with 25 mM lithium 3,5-diiodosalicylate (LIS), 2 M NaCl, 0.65 M ammonium sulphate containing buffers followed by DNase I/RNase A digestion, or according to so designated conventional method. Using PCR-based and in vitro binding assays it was established that LIS and ammonium sulphate extractions gave similar results for the type of attachment of sequences investigated. The harsh extraction with 2 M NaCl or the conventional procedure led to some rearrangements in the attachment of DNA loops. As a result a big part of matrix attached sequences were found detached in the loop fractions. However, the in vitro binding abilities of the MARs to the nuclear matrices isolated by different methods did not change.

Core histones of the amitochondriate protist, Giardia lamblia

Genes coding for the core histones H2a, H2b, H3, and H4 of Giardia lamblia were sequenced. A conserved organism- and gene-specific element, GRGCGCAGATTTVGG, was found upstream of the coding region in all core histone genes. The derived amino acid sequences of all four histones were similar to their homologs in other eukaryotes, although they were among the most divergent members of this protein family. Comparative protein structure modeling combined with energy evaluation of the resulting models indicated that the G. lamblia core histones individually and together can assume the same three-dimensional structures that were established by X-ray crystallography for Xenopus laevis histones and the nucleosome core particle. Since G. lamblia represents one of the earliest-diverging eukaryotes in many different molecular trees, the structure of its histones is potentially of relevance to understanding histone evolution. The G. lamblia proteins do not represent an intermediate stage between archaeal and eukaryotic histones.

An upstream Oct-1- and Oct-2-binding silencer governs B29 (Ig beta) gene expression

The B cell-specific B29 (Igbeta) gene is activated in the earliest B cell precursors and is expressed throughout B cell development. Tissue-specific expression of the murine B29 gene is controlled by a B cell-specific promoter whose activity is governed by a cassette of upstream transcriptional silencers. This study describes a potent new silencer that is located 5' of the previously identified B29 silencer elements, FROG and TOAD. Like these known elements, the new B29 silencer is not restricted to the B29 promoter. Nuclear proteins from all cell lines tested interacted with this A+T-rich sequence, which closely resembled a noncanonical octamer binding motif and also conformed to the consensus sequence for nuclear matrix attachment regions. Interaction of Oct-1 and Oct-2 with the B29 A+T-rich sequence was confirmed using octamer-specific Abs. Oct-1/Oct-2 binding was required for the inhibitory activity of this sequence because mutations that blocked Oct-1/Oct-2 binding also eliminated inhibition of the B29 promoter. This B29 A+T-rich sequence specifically interacted with isolated nuclear matrix proteins in vitro, suggesting that it may also function as a matrix attachment region element. Maintenance of the level of B29 gene expression through the interaction of the minimal promoter and the upstream silencer elements FROG, TOAD, and the A+T-rich Oct-1/Oct-2 binding motif may be essential for normal B cell development and/or function.

Macromolecular and ultrastructural organization of the mitotic chromosome scaffold

Using electron microscopy (EM), we have examined three structural domains of the mitotic chromosome scaffold of mouse erythroleukemia (MEL) Friend cells with different morphologic organization: centromeric, intermediate, and telomeric. The intermediate, most extensive, domain exhibited a specific fibrogranular structure representing tightly packed granular bodies with diameters between 20 and 60 nm. The chromosome scaffold contained three main components: proteins (81%), RNA (12%), and DNA (7%). The residual DNA extracted from the scaffold represented short fragments, 300 bp on average, belonging to the class of tandemly arranged repetitive DNA. In situ hybridization experiments confirmed its typical centromeric location. Scaffold RNA represented three fractions: a major RNA fraction with an electrophoretic mobility corresponding to that of 5S RNA and two minor fractions with electrophoretic mobilities somewhat lower than that of 18S RNA. Scaffold RNA was localized mainly in the centromeric region. We show that the newly synthesized protein component of the chromosome scaffolds migrates slowly to the chromosomes, reaching a maximum specific radioactivity 12 h from the onset of the chase period.

Folding of chromatin in the presence of heterogeneous histone H1 binding to nucleosomes

We have reconstituted oligonucleosome complexes containing histone H1 starting from a synthetic DNA template, consisting of 12 tandemly arranged 208-base pair fragments of the 5 S rRNA gene, purified HeLa histone octamers, and histone H1. A ratio of histone H1 per histone octamer used in the reconstitution (0.8-0.9 mol of histone H1/mol of histone octamer) similar to that observed in vivo was used. The reconstituted chromatin complexes exhibit a salt-dependent folding, which is almost indistinguishable from that exhibited by chromatin fragments obtained from nuclease digestion of native chromatin. The folding of this reconstituted chromatin complex seems to be rather independent of the symmetrical or asymmetrical position occupied by H1 in the individual nucleosomes. Binding of histone H1 to the oligonucleosome complexes, under the stoichiometric binding conditions used, had no inhibitory effect on the transcriptional potential of these complexes.

Matrix attachment regions and transcription units in a polygenic mammalian locus overlapping two isochores

Eukaryotic chromosomes are ponctuated by specialized DNA sequences (MARs) characterized by their ability to bind the network of nonhistone proteins that form the nuclear matrix or scaffold. We previously described an amplifiable cluster of genes with different tissue-specific expression patterns, located on Chinese hamster chromosome 1q. This model is especially appropriate to study the relationships between MARs and transcription units. We show here that four attachment regions, with sequences exhibiting motifs specific to MARs, are present within the 100 kb of screened DNA. Three of them are relatively short sequences localized in intergenic regions. The last one extends over one of the transcription units and contains a region previously identified as a recombination hot spot. Moreover, the analysis of a DNA sequence extending over some 50 Kb of this region and spanning at least four genes, disclosed a strikingly sharp change in G + C content. This strongly suggests that the studied region contains the boundary of two isochores. We propose that the frequency and the size of MARs are correlated to their localization in G + C rich or poor domains.

Alleviation of histone H1-mediated transcriptional repression and chromatin compaction by the acidic activation region in chromosomal protein HMG-14

Histone H1 promotes the generation of a condensed, transcriptionally inactive, higher-order chromatin structure. Consequently, histone H1 activity must be antagonized in order to convert chromatin to a transcriptionally competent, more extended structure. Using simian virus 40 minichromosomes as a model system, we now demonstrate that the nonhistone chromosomal protein HMG-14, which is known to preferentially associate with active chromatin, completely alleviates histone H1-mediated inhibition of transcription by RNA polymerase II. HMG-14 also partially disrupts histone H1-dependent compaction of chromatin. Both the transcriptional enhancement and chromatin-unfolding activities of HMG-14 are mediated through its acidic, C-terminal region. Strikingly, transcriptional and structural activities of HMG-14 are maintained upon replacement of the C-terminal fragment by acidic regions from either GAL4 or HMG-2. These data support the model that the acidic C terminus of HMG-14 is involved in unfolding higher-order chromatin structure to facilitate transcriptional activation of mammalian genes.

Phosphorylation of linker histone is associated with transcriptional activation in a normally silent nucleus

Previous studies have suggested that micronuclear linker histones are phosphorylated by cAMP-dependent protein kinase (PKA) in Tetrahymena (Sweet, M.T., and C.D. Allis. 1993. Chromosoma. 102: 637-647). In this study, we report that a rapid and dramatic phosphorylation of the micronuclear linker histone, delta, occurs early in the sexual pathway, conjugation. Phosphorylated isoforms of delta are detected as early as 30 min after mixing cells of different mating types; blocking pair formation abolishes this induction completely. Phosphorylation of delta is stimulated by the addition of N6-benzoyladenosine 3':5' cyclic monophosphate to starved (nonmating) cells, suggesting that a PKA/cAMP signal transduction pathway is involved. Maximal phosphorylation of delta is observed during meiotic prophase, a period when micronuclei become transcriptionally active. In situ staining, using phospho-delta-specific antibodies combined with [3H]uridine autoradiography, shows that decondensed micronuclear chromatin undergoing active transcription is enriched in phosphorylated delta isoforms. In contrast, condensed inactive micronuclear chromatin is enriched in dephosphorylated delta. These results strongly suggest that phosphorylation of linker histone plays an important and previously unsuspected role in establishing transcriptional competence in micronuclei.

Different effects of histone H1 on de novo DNA methylation in vitro depend on both the DNA base composition and the DNA methyltransferase

We have characterized the inhibition exerted by histone H1 on the activity of human placenta DNA (cytosine-5-)-methyltransferase. Our experiments demonstrate that the extent of inhibition depends on the DNA base composition, AT-rich substrates being more severely affected than GC-rich substrates and CpG-rich islands. With bacterial SssI methylase, the effect is completely reversed since its activity on AT-rich substrates undergoes a 4-5-fold stimulation upon the addition of H1. Poly(L-lysine) mimicks H1 effects, suggesting an essential role of lysine residues in both the inhibitory and stimulatory effects of H1. By comparison of the different behaviors of the two enzymes, the inhibitory effect over the eukaryotic enzyme might be accounted for by hypothesizing a competition between minor groove-binding motifs (SPKK-like) present in placenta methylase as well as in histone H1.

Either of the major H2A genes but not an evolutionarily conserved H2A.F/Z variant of Tetrahymena thermophila can function as the sole H2A gene in the yeast Saccharomyces cerevisiae

H2A.F/Z histones are conserved variants that diverged from major H2A proteins early in evolution, suggesting they perform an important function distinct from major H2A proteins. Antisera specific for hv1, the H2A.F/Z variant of the ciliated protozoan Tetrahymena thermophila, cross-react with proteins from Saccharomyces cerevisiae. However, no H2A.F/Z variant has been reported in this budding yeast species. We sought to distinguish among three explanations for these observations: (i) that S. cerevisiae has an undiscovered H2A.F/Z variant, (ii) that the major S. cerevisiae H2A proteins are functionally equivalent to H2A.F/Z variants, or (iii) that the conserved epitope is found on a non-H2A molecule. Repeated attempts to clone an S. cerevisiae hv1 homolog only resulted in the cloning of the known H2A genes yHTA1 and yHTA2. To test for functional relatedness, we attempted to rescue strains lacking the yeast H2A genes with either the Tetrahymena major H2A genes (tHTA1 or tHTA2) or the gene (tHTA3) encoding hv1. Although they differ considerably in sequence from the yeast H2A genes, the major Tetrahymena H2A genes can provide the essential functions of H2A in yeast cells, the first such case of trans-species complementation of histone function. The Tetrahymena H2A genes confer a cold-sensitive phenotype. Although expressed at high levels and transported to the nucleus, hv1 cannot replace yeast H2A proteins. Proteins from S. cerevisiae strains lacking yeast H2A genes fail to cross-react with anti-hv1 antibodies. These studies make it likely that S. cerevisiae differs from most other eukaryotes in that it does not have an H2A.F/Z homolog. A hypothesis is presented relating the absence of H2A.F/Z in S. cerevisiae to its function in other organisms.

Purification and molecular cloning of the scaffold attachment factor B (SAF-B), a novel human nuclear protein that specifically binds to S/MAR-DNA

We have purified to near homogeneity a novel nuclear protein from HeLa cells, that specifically binds to scaffold or matrix attachment region DNA elements (S/MAR DNA). The protein, designated SAF-B for scaffold attachment factor B, is an abundant component of chromatin, but not of the nuclear matrix and is expressed in all human tissues investigated. Antibodies against the purified protein were raised in rabbit and used to isolate the complete cDNA encoding SAF-B by immunoscreening. As predicted from the cDNA sequence, SAF-B contains 849 amino acids (96 696 Da), without significant homology to any known protein. SAF-B is rich in charged residues, leading to an aberrant migration on SDS gels, and has two putative bipartite nuclear localisation signals.

Increased phosphorylation of histone H1 in mouse fibroblasts transformed with oncogenes or constitutively active mitogen-activated protein kinase kinase

We compared the nucleosomal organization, histone H1 subtypes, and histone H1 phosphorylated isoforms of ras-transformed and parental 10T1/2 mouse fibroblasts. In agreement with previous studies, we found that ras-transformed mouse fibroblasts have a less condensed chromatin structure than normal fibroblasts. ras-transformed and parental 10T1/2 cells had similar amounts of H1 subtypes, proteins that have a key role in the compaction of chromatin. However, labeling studies with 32P and Western blot experiments with an antiphosphorylated H1 antibody show that interphase ras-transformed cells have higher levels of phosphorylated H1 isoforms than parental cells. G1/S phase-arrested ras-transformed cells had higher amounts of phosphorylated H1 than G1/S phase-arrested parental cells. Mouse fibroblasts transformed with fes, mos, raf, myc, or constitutively active mitogen-activated protein (MAP) kinase kinase had increased levels of phosphorylated H1. These observations suggest that increased phosphorylation of H1 is one of the consequences of the persistent activation of the mitogen-activated protein kinase signal transduction pathway. Indirect immunofluorescent studies show that phosphorylated H1b is localized in centers of RNA splicing in the nucleus, suggesting that this modified H1 subtype is complexed to transcriptionally active chromatin.

Phosphorylated and dephosphorylated linker histone H1 reside in distinct chromatin domains in Tetrahymena macronuclei

Phosphorylated and dephosphorylated isoforms of Tetrahymena macronuclear H1 were separated from each other by cation-exchange high performance liquid chromatography and used to generate a pairwise set of antisera that discriminate the phosphorylation state of this linker histone. Affinity-purified antibodies from each sera recognize appropriate H1 isoforms and stain macronuclei under appropriate physiological conditions. Immunogold localizations demonstrate that phosphorylated and dephosphorylated H1 localize nonrandomly in distinct subdomains of macronuclear chromatin. Dephosphorylated H1 is strongly enriched in the electron-dense chromatin bodies that punctuate macronuclear chromatin. In contrast, phosphorylated H1 isoforms, as well as an evolutionarily conserved H2A.F/Z-like variant (hv1) believed to function in the establishment of transcriptionally competent chromatin, are modestly enriched at the periphery of chromatin bodies and in the surrounding euchromatin. Using antibodies against TATA-binding protein, we show that transcriptionally active chromatin lies outside of the chromatin bodies in an area relatively devoid of H1. Antibodies against general core histones are more or less evenly distributed across these domains. Together, these data are consistent with a model in which phosphorylation of H1, perhaps in association with hv1, loosens the binding of H1 in chromatin leading to chromatin decondensation as part of a first-step mechanism in gene activation. In contrast, our data support the view that dephosphorylation of this linker histone facilitates or stabilizes condensed, transcriptionally silent chromatin.

Linker histones are not essential and affect chromatin condensation in vivo

We have (separately) disrupted all of the expressed macronuclear copies of the HHO gene encoding macronuclear histone H1 and of the micronuclear linker histone (MLH) gene encoding the protein MicLH in Tetrahymena thermophila. These disruptions are shown to eliminate completely the expression of each protein. Strains without either linker histone grow at normal rates and reach near-normal cell densities, demonstrating that linker histones are not essential for cell survival. Histone H1 knockout (delta H1) cells have enlarged DAPI-stained macronuclei and normal-sized micronuclei, while MicLH knockout (delta MicLH) cells have enlarged micronuclei and normal-sized macronuclei. delta MicLH cells undergo mitosis normally. However, the micronuclear mitotic chromosome structure is less condensed. These studies provide evidence that linker histones are nonessential and are involved in chromatin packaging and condensation in vivo.

Expression of histone H1 (zero) in transcriptionally activated supraoptic neurons

This study has analysed by immunocytochemistry the expression pattern of histone H1 zero after the osmotically induced activation of transcription in supraoptic nucleus neurons of the rat. In control rats, histone H1 zero was constitutively expressed in neuronal and glial cell nuclei of supraoptic nucleus. After chronic neuronal stimulation by intermittent salt-loading, the majority of neuronal cell nuclei exhibited a marked reduction of immunostaining, which was confirmed by densitometric analysis of immunoreactivity. This effect was reversible, since optical density values returned to control levels when the stimulation of supraoptic neurons was suppressed by rehydration. Ultrastructural immunocytochemistry of histone H1 zero showed that immunogold particles specifically decorated chromatin fibers, with the highest accumulation of particles being on the condensed inactive chromatin. These results indicate that transcriptional activation in supraoptic neurons is accompanied by a depletion of the chromatin-associated histone H1 zero, and also suggest that this transcription-dependent expression of histone H1 zero may be involved in regulating chromatin condensation and gene expression in mature neurons that constitutively express this protein.

Deposition of histone H1 onto reconstituted nucleosome arrays inhibits both initiation and elongation of transcripts by T7 RNA polymerase

The effect of histone H1 on transcription by bacteriophage T7 RNA polymerase was examined using reconstituted chromatin templates. A 3.8 kb linear DNA template consisting of a specific transcription promoter for T7 RNA polymerase placed upstream of 18 tandem repeats of a 207 bp nucleosome positioning sequence derived from the 5S rRNA gene of Lytechinus variegatus was used as a template for chromatin reconstitution. Regularly spaced arrays of nucleosome cores were assembled onto this DNA template from donor histone octamers by salt step dialysis. Histone H1 was incorporated onto free DNA or reconstituted chromatin templates and double label transcription assays were performed. The experiments indicated that histone H1 has a strong inhibitory effect on both transcription initiation and elongation. These effects are especially pronounced on chromatin templates, where both transcription initiation and elongation are virtually halted. The inhibition of transcription elongation appears to result from a dramatic increase in premature termination of transcripts. These experiments indicate that assembly of histone H1 into chromatin can result in structures which are completely repressed with respect to transcription.

The effect of histone H1 and DNA methylation on transcription

We have previously shown that DNA methylation acts as a focus for the formation of inactive chromatin in vivo. We have investigated the mechanism further by in vitro transcription of a template containing two tRNA genes and an extensive (G+C)-rich sequence characteristic of a CpG island. The extent of transcription from the unmethylated or fully methylated template was assayed in the presence of varied levels of histone H1. The transcriptional activity of both templates was inhibited by increasing amounts of histone H1, although inhibition with the methylated template occurs at a lower H1:DNA ratio. The H1c variant shows the greatest preferential inhibition of the methylated template. We demonstrated that histone H1 complexed to DNA is one of the factors that inhibits transcription by preventing the formation of initiation complexes, particularly on methylated template, rather than the formation of disordered H1.DNA aggregates.

Differential compaction of transcriptionally competent and repressed chromatin reconstituted with histone H1 subtypes

Chromatin fragments stripped of H1 histones regain the ability to form higher order structures and aggregates in 0.15 M NaCl following reconstitution with histone H1. However, transcriptionally competent chromatin fragments are resistant to chicken erythrocyte H1/H5 histone-induced 0.15 M NaCl aggregation/precipitation. In this study, we investigated the ability of stripped chromatin fragments reconstituted with one of four histone H1 subtypes (chicken erythrocyte H1, H5, trout liver H1a, H1b) at various stoichiometries to form salt precipitable higher order structures. Our results provide evidence that chicken erythrocyte histone H1 was more effective than histone H5 and trout liver histone H1b better than H1a in forming higher order structures. None of the histone H1 subtypes could render transcriptionally competent chromatin fragments insoluble in 0.15 M NaCl. These results are consistent with the ideas that the histone H1 subtypes differ in their capacities to compact chromatin fiber, and that the alterations in the structure of transcriptionally competent nucleosomes interfere with the capacity of all H1 subtypes to form higher order structures.

Linker histones H1 and H5 prevent the mobility of positioned nucleosomes

We have previously identified a generally occurring short-range mobility of nucleosome cores on DNA in relatively low ionic strength conditions. Here we report that this mobility of histone octamers positioned on constructs of 5S rDNA is suppressed by the binding of histone H1 or H5 to the nucleosome. Histone H5 is the more potent inhibitor of nucleosome mobility, in accordance with its higher affinity for chromatin. We propose that this reversible restraint on chromatin dynamics may play a role in local regulation of processes that require access to the DNA.

Mapping of linear epitopes of human histone H1 recognized by rabbit anti-H1/H5 antisera and antibodies from autoimmune patients

Seventeen synthetic peptides of 15-16 residues, covering the complete sequence of the major human H1b variant, were tested for their capacity to bind serum IgG antibodies from 128 patients with systemic lupus erythematosus (SLE), rheumatoid arthritis (RA) and Sjögren's syndrome (pSS). One peptide (residues 111-127) of the human H1 degree variant and six synthetic and natural fragments of H5 were also tested. Results were compared to those obtained with antibodies from 11 rabbits immunized against chicken H1 and H5, and calf H1. The activity of peptides was tested in direct ELISA and in inhibition assays with free peptides in solution. A major epitope recognized by antibodies from SLE, RA and pSS patients as well as by rabbit antibodies was identified in the C-terminus of H1b (residues 204-218). Other peptides in the globular (residues 79-94) and C-terminal domains of H1b and peptide 111-127 of H1 degree were also recognized, albeit at a lower level and frequency, and some of them contain sequence homologies with peptide 204-218. Patients' antibodies and rabbit antisera were tested with complete H1 proteins from HeLa cells, calf thymus and chicken erythrocytes and with chicken H5. Less than 25% of autoimmune sera contained IgG antibodies reacting with H1/H5 in a direct ELISA. In dot-immunoassay, antigenic activity with intact H1/H5 proteins was detected in a larger number of sera. Using antibodies raised in rabbits against peptides 1-16 and 204-218 of H1b, we found no reaction with H1 immobilized on a solid-phase. In contrast, peptides 144-159, 170-185 and 204-218, which contain identical structural domains, compete with H1 in solution indicating that any of these three regions are accessible at the surface of free H1 and may be involved in the induction of specific antibodies in autoimmune patients.

Multiple functions of dynamic histone acetylation

Besides its role in organizing nuclear DNA, the nuclear matrix is involved in specific nuclear functions, including replication, transcription, and RNA splicing. It is becoming increasingly evident that nuclear processes are localized to distinct regions in the nucleus. For example, transcriptionally active genes and RNA transcripts are found in discrete transcription foci. Current evidence suggests that nuclear matrix-bound transcriptionally active DNA sequences are in nucleosomes with dynamically acetylated histones. Histone acetylation, which precedes transcription, alters nucleosome and chromatin structure, decondensing the chromatin fibre and making the nucleosomal DNA accessible to transcription factors. Histone acetyltransferase and histone deacetylase, which catalyze this rapid acetylation and deacetylation, are associated with the internal nuclear matrix. We hypothesize that these enzymes play a role in maintaining the association of the active chromatin domains with the internal nuclear matrix at sites of ongoing transcription.

Distribution of globin genes and histone variants in micrococcal nuclease-generated subfractions of chromatin from Friend erythroleukemia cells at different malignant states

The distribution of the alpha and beta-globin genes and histone variants was examined in micrococcal nuclease-generated chromatin fractions of three Friend murine erythroleukemia cell types differing in malignant potential and inducibility to erythroid differentiation. A preferential concentration of globin gene sequences, as compared to satellite DNA, was noted in a physiological salt-soluble, histone H1-depleted, mononucleosomal chromatin fraction (Sup 120) in all Friend cell types, even those in which the globin gene was not capable of transcriptional activation by chemical induction. The level of globin gene enrichment in the Sup 120 fraction was highest in the most malignant and inducible cell type. The chemical induction of erythroid differentiation in this cell line did not change the distribution of globin genes in the chromatin fractions. The Sup 120 chromatin fraction prepared from mouse brain nuclei was not enriched in globin genes. Besides the previously reported low H2A. 1/H2A.2 ratio [Blankstein and Levy: Nature 260:638-640, 1976], chromatin from the most tumorigenic cell type showed the lowest H2B.2 to H2B.1 ratio, highest levels of histone H4 acetylation, and the most pronounced change in relative amounts of two major electrophoretic bands of histone H1 variants as compared to the less malignant cell types. The histone variant content of the micrococcal nuclease-generated chromatin fractions from the three Friend cell types reflected the core histone variant differences for the respective intact nuclei. However, the electrophoretic separation of mononucleosomes by size revealed several classes with different H2A variant ratios. The results demonstrate the existence of structural differences in globin gene and histone variants in erythroleukemia cell chromatin associated with distinguishable phenotypes during malignant cell progression.

Parental nucleosomes segregated to newly replicated chromatin are underacetylated relative to those assembled de novo

Antibodies specific for acetylated histone H4 were used to examine the acetylation state of parental histones that segregate to newly replicated DNA. To generate newly replicated chromatin containing only segregated parental nucleosomes, isolated nuclei were labeled with [3H]TTP in vitro; alternatively, whole cells were labeled with [3H]thymidine in the presence of cycloheximide. Soluble chromatin was prepared by micrococcal nuclease digestion, and subjected to immunoprecipitation with "penta" antibodies (Lin et al., 1989). In sharp contrast to nucleosomes containing newly synthesized, diacetylated H4 (Perry et al., 1993), chromatin replicated in vitro was only marginally susceptible to immunoprecipitation. Control experiments established that bona fide acetylated chromatin was selectively immunoprecipitated by the same techniques and that segregated nucleosomes were not disassembled during treatment with "penta" antibodies. When replication was coupled to an in vitro histone acetylation system, the enrichment for segregated nucleosomes in the immunopellet increased approximately 3-fold, demonstrating that changes in the acetylation state of segregated histones can be detected immunologically and that parental histones on new DNA are accessible to acetyltransferases during, or immediately after, DNA replication. In vivo pulse-chase experiments, performed in the presence of cycloheximide, confirmed these results. Uptake experiments further established that concurrent histone acetylation did not alter the rate of DNA synthesis in vitro. Our results provide evidence that replication-competent chromatin is not obligatorily acetylated, and indicate that the acetylation status of segregated histones may be maintained during chromatin replication. The possible significance of this, with respect to the regulation of chromatin higher order structures during DNA replication, and the propagation of transcriptionally active vs inactive chromatin structures, is discussed.

SAR-dependent mobilization of histone H1 by HMG-I/Y in vitro: HMG-I/Y is enriched in H1-depleted chromatin

An experimental assay was developed to search for proteins capable of antagonizing histone H1-mediated general repression of transcription. T7 RNA polymerase templates containing an upstream scaffold-associated region (SAR) were highly selectively repressed by H1 relative to non-SAR control templates. This is due to the nucleation of H1 assembly into flanking DNA brought about by the numerous A-tracts (AT-rich sequences containing short homopolymeric runs of dA.dT base pairs) of the SAR. Partial, selective titration of these A-tracts by the high mobility group (HMG) protein HMG-I/Y led to the complete derepression of transcription from the SAR template by inducing the redistribution of H1 on to non-SAR templates. SARs are associated with many highly transcribed regulated genes where they may serve to facilitate the HMG-I/Y-mediated displacement of histone H1 in chromatin. Indeed, HMG-I/Y was found to be strongly enriched in the H1-depleted subfraction which can be isolated from chromatin.

Two non-histone proteins are associated with the promoter region and histone HI with the transcribed region of active hsp-70 genes as revealed by UV-induced DNA-protein crosslinking in vivo

We described here an approach for mapping proteins on any sequence of genomic DNA. UV-induced DNA-protein crosslinking within whole cells and the 'protein image' hybridization technique (1) were applied to test the proteins bound to different regions of the D. melanogaster hsp-70 gene. The histone H1-DNA association with the coding region is shown to be maintained, even during very intensive transcription, but is absent in the promoter. Two non-histone proteins with apparent molecular masses of 50 kD (p50) and 100 kD (p100) are crosslinked only to the active hsp-70 gene regulatory region and preferentially bind to its complementary and coding DNA strands, respectively.

Specific non-enzymatic glycation of the rat histone H1 nucleotide binding site in vitro in the presence of AlF4-. A putative mechanism for impaired chromatin function

We show here that an aluminium derivative, AlF4-, stimulates glycation of histone H1 selectively in the proximity of its nucleotide-binding site. This adduct formation interferes with nucleoside triphosphate hydrolysis by H1 and with nucleotide modulation of H1 DNA binding. The present mode of aluminium action may in part be responsible for its effects on the chromatin structure and expression of tissue-specific genes, and may constitute a mechanism in the pathogenesis of aluminium-induced encephalopathy and in that of Alzheimer's disease, for example.

DNA binding of histone H1 is modulated by nucleotides

Histone H1 acts as a general repressor of transcription in eukaryotes by organizing nucleosomes into inaccessible condensed forms of chromatin. The capability of H1 to bind to DNA with some sequence specificity is likely to be critical in the control of these processes. We show here that ATP and several other nucleotides, including non-hydrolyzable derivatives, can inhibit DNA binding of H1. The results also show that ATP differentially affects binding of H1 to DNA in a fashion enhancing nucleotide sequence specificity of the binding. The study suggests a novel mechanism of modulation of H1 activity that has important implications for the role of H1 as a transcriptional regulator.

Histone H1a subtype presents structural differences compared to other histone H1 subtypes. Evidence for a specific motif in the C-terminal domain

Following a previous isolation by reverse-phase HPLC of five histone H1 subtypes from adult rat liver, purity of three of them, H1a, H1b and H1d (according to Lennox's nomenclature), was achieved. Structural features of these three subtypes were investigated. Partial cleavage of these subtypes by endoproteinase Glu-C showed a different behavior of the H1a subtype when compared to the H1b and H1d subtypes. Under the conditions used in this work, the H1b and H1d subtypes present three major sites accessible to the endoproteinase Glu-C, while the H1a subtype presents only one major site accessible to the proteinase. Partial N-terminal sequence of the different fragments obtained after proteolysis indicated that the two H1b and H1d subtypes were cleaved inside the globular domain (Glu-54,-75) and between the globular domain and the C-terminal one (Glu-116). The H1a subtype was only cleaved between the globular domain and the C-terminal tail (Glu-116), though Glu-54 and Glu-75 sites were present. These results would suggest some differences in the conformation of these proteins. Furthermore, the partial determined sequences of H1b and H1d showed 85% similarity to each other (the main differences were threonine residues instead of alanine residues in the C-terminal domain) while H1a was only 60% similar to H1b and H1d, for the sequences which aligned. The strongest differences between the H1a subtype and the two other subtypes were observed in the first amino acid residues of the C-terminal domain. The 117-126 amino acid residues (SKASTTKVTV) of H1a were quite different from those of H1b and H1d. This sequence, which showed a number of serine and threonine residues, was not found in any other histone sequence, after consultation with data bases. This H1a subtype was a minor component in adult liver (2.4%). As it was described in testis as a major component, testis histone H1 proteins were fractionated onto reverse-phase HPLC under the same conditions as those used for histone H1 proteins from liver. The pure testis H1a fraction was submitted to the endoproteinase Glu-C digestion. The pattern digestion was the same as that observed for liver H1a. The two 44-76 and 117-126 determined amino acid residues of H1a from testis were strictly identical to those of liver H1a. We demonstrate that H1a is the same protein in liver and testis and we give evidence for a specific motif SKASTTKVTV (117-126 residues) in the sequence of the C-terminal domain.(ABSTRACT TRUNCATED AT 400 WORDS)

The lysine-rich H1 histones from the slime moulds, Physarum polycephalum and Dictyostelium discoideum lack phosphorylation sites recognised by cyclic AMP-dependent protein kinase in vitro

Calcium chloride-extracted histones were prepared from nuclei of the slime moulds, Physarum polycephalum and Dictyostelium discoideum, and phosphorylation by purified preparations of cyclic AMP-dependent protein kinase (cAMP-d PK) and growth-associated H1 histone kinase (HKG) examined and compared. Among the major histone fractions and other proteins in the two preparations, the H1 histones from both organisms were found to be effective and exclusive substrates for HKG. cAMP-d PK, which phosphorylates mammalian H1 histone and certain, in particular H2B, of the mammalian core histones, phosphorylated several of the core histones from both slime moulds but did not phosphorylate H1 histone from either. The slime mould H1s remained ineffective substrates for cAMP-d PK even after extensive alkaline phosphatase treatment of the histone preparations. Additional studies demonstrated that the lack of slime mould H1 phosphorylation by cAMP-d PK was not due to competition of the H1 molecules with the core histones for the kinase. Our studies suggest that H1 histones from these organisms, whilst clearly containing sites for phosphorylation by HKG, apparently lack phosphorylation sites recognised by cAMP-d PK. Thus, the mediation of specific nuclear functions by cAMP-dependent phosphorylation of H1 in higher organisms may not occur or be required in these lower eukaryotes.

Chromatin dynamics and the modulation of genetic activity

Chromatin, the genetic material of eukaryotes, is a dynamic macromolecular assembly that continuously changes its composition and conformation to accommodate different stages of genetic activity, e.g. transcription and replication. Evidence is accumulating that the dynamic behavior of chromatin has important functional roles in the modulation of genetic activity, largely due to the intrinsic properties of arrays of nucleosome cores.

Nucleoside triphosphate binding and hydrolysis by histone H1

We present here further evidence supporting that histone H1 contains a nucleotide binding site interacting e.g. with ADP, ATP, GDP and GTP. The finding is in accordance with the previous observation that nucleotides modulate recognition of DNA by H1. Most interestingly, H1 appears to be capable of hydrolyzing NTPs and incorporating phosphate to exogenous proteins. The mode of nucleotide action on H1 may be considered highly analogous to that of GTPases. Nuclear receptors may thus act through mechanisms similar to those for receptors on the plasma membrane.