Biosynthesis and posttranslational acetylation of histones during spherulation of Physarum polycephalum

Histone H4 acetylation during interleukin-2 stimulation of mouse T cells

Proliferation and cell cycle progression of eukaryotic cells are closely linked to changes in chromatin structure and gene expression. By reversible histone acetylation the cell is able to modulate chromatin condensation and accessibility of specific regions within the chromatin. Here, we examined histone H4 acetylation patterns during growth induction of the murine interleukin-2 dependent T cell line B6.1. In order to detect acetylation on each of the four potential target residues we produced a set of antibodies recognizing specifically acetylated lysine 5, 8, 12 and 16 in the N-terminal tail of histone H4. Acetylation was generally low in resting T cells, but increased after stimulation with a specific kinetics for each lysine. Lysine 16 was acetylated during the G1 phase and deacetylated during S phase. H4 acetylation on lysine 5, 8 and 12, in contrast, was induced before cells started to replicate, and persisted until cells entered mitosis. Treatment of resting B6.1 cells with the specific deacetylase inhibitor trichostatin A (TSA) led to H4 hyperacetylation at all four lysine residues indicating that the histone modification can occur in the absence of replication. After release from TSA treatment normal H4 acetylation levels were reestablished by extremely rapid deacetylation of lysines 5, 8, 12 and 16. The deacetylation step was 60-100 times faster than TSA induced acetylation and equally efficient in resting and exponentially growing T cells. Our results indicate the presence of cell cycle regulated lysine specific acetylating and deacetylating activities in mouse T cells.

Apoptotic death in adenocarcinoma cell lines induced by butyrate and other histone deacetylase inhibitors

n-Butyrate inhibits the growth of colon cancer cell lines. In the HCT 116 cell line, butyrate-induced growth inhibition is almost fully reversible, whereas in the VACO 5 cell line, a subpopulation undergoes apoptosis within 30 hr of treatment with butyrate. Concurrent treatment of VACO 5 cells with butyrate and the phorbol ester 12-O-tetradecanoylphorbol 13-acetate (TPA) accelerates and increases the incidence of cell death to nearly 100% of the population, whereas HCT 116 cells largely remain alive during treatment with this combination. The action of butyrate as an inhibitor of histone deacetylase was assessed in these cell lines by examining extracted core histones for their electrophoretic mobility in Triton/acid/urea gels. The concentrations of butyrate that were effective for inducing apoptosis were similar to the concentrations that caused hyperacetylation of core histones in the VACO 5 cell line. Furthermore, an examination of other carboxylic acids for induction of apoptosis revealed a rank order that corresponded to the order of potency in causing hyperacetylation of core histones. Specifically, the active acids were 3-5 carbons in length and lacked substitution at the 2-position. Isovaleric and propionic acids, in particular, proved to be effective inducers of both hyperacetylation and apoptosis at 5 mM concentrations, a finding of potential relevance to the unusual pancytopenia occurring after acidotic episodes in isovaleric and propionic acidemias. The duration of butyrate treatment required for chromatin fragmentation (10-20 hr) corresponded to the time required for histone H4 to become predominantly tetraacetylated. Furthermore, trichostatin A, a structurally dissimilar inhibitor of histone deacetylase, mimicked butyrate-induced apoptosis of VACO 5 cells and growth inhibition of HCT 116 cells. The dramatic enhancement of VACO 5 cell death by TPA, and the high level resistance of HCT 116 cells to butyrate were not evident from histone acetylation determinations. Thus, applications of butyrate for cytoreduction therapy will benefit from pharmacodynamic assessment of histone acetylation, but will require additional work to predict susceptibility to butyrate-induced death.

Purification of histone deacetylase HD1-A of germinating maize embryos

We have purified the soluble nuclear histone deacetylase HD1-A of germinating maize embryos. By a combination of 6 chromatographic steps we achieved a 77,000-fold purification of an enzymatically active protein. Gel filtration chromatography revealed a molecular weight of 45 kDa of the native enzyme and electrophoretic analysis of the purified enzyme by SDS-PAGE resulted in a single band at a molecular weight of 48 kDa, indicating that the enzyme is a monomer protein. When fractions with enzyme activity of different stages of chromatographic purification were subjected to isoelectric focusing, enzyme activity focused at a pH of around 6.4 as measured in an activity gel assay; second dimension SDS-PAGE again revealed a protein spot at a molecular weight of 48 kDa.

Histone acetylation: facts and questions

The DNA of eukaryotic cells is organized in a complex with proteins, either as interphase chromatin or mitotic chromosomes. Nucleosomes, the structural subunits of chromatin, have long been considered as static structures, incompatible with processes occurring in chromatin. During the past few years it has become evident that the histone part of the nucleosome has important regulatory functions. Some of these functions are mediated by the N-terminal core histone domains which contain sites for posttranslational modifications, among them lysine residues for reversible acetylation. Recent results indicate that acetylation and deacetylation of N-terminal lysines of nucleosomal core histones represent a means of molecular communication between chromatin and the cellular signal transduction network, resulting in heritable epigenetic information. Data on enzymes involved in acetylation and the pattern of acetylated lysine sites on chromosomes, as well as genetic data on yeast transcriptional repression, suggest that acetylation may lead to structural transitions as well as specific signalling within distinct chromatin domains.

Properties of the yeast nuclear histone deacetylase

A nuclear histone deacetylase from yeast was partially purified and some of its characteristics were studied. Histone deacetylase activity was stimulated in vitro by high-mobility-group nonhistone chromatin proteins 1 and 2 and ubiquitin and inhibited by spermine and spermidine, whereas n-butyrate had no significant inhibitory effect. Like the mammalian enzyme, partially purified histone deacetylase from yeast was strongly inhibited by trichostatin A. However, in crude extract preparations the yeast enzyme was not inhibited and treatment with trichostatin in vivo did not show any effect, either on the histone acetylation level or on cell viability. At low ionic strength, the enzyme can be isolated as a complex of high molecular mass that is much less inhibited by trichostatin A than is partially purified histone deacetylase activity. Furthermore, radiolabelled oligonucleosomes were more efficiently deacetylated by the complex than by the low-molecular-mass form of the enzyme. The histone deacetylase activity was separated from a polyamine deacetylase activity and its specificity studied. Using h.p.l.c.-purified core histone species as substrate, histone deacetylase from yeast is able to deacetylate all core histones with a slight preference for H3. Our results support the idea that the yeast histone deacetylase may act as a high-molecular-mass complex in vivo.

Change in chromatin organization related to in vivo transcriptional activity and histone synthesis independent of DNA replication during differentiation (germination) of Physarum spherules

During the germination of Physarum spherules, increases have been observed, at the same moment, in the level of in vivo transcriptional activity as measured by [³H] uridine incorporation, and the accessibility of DNA for ethidium bromide staining as shown by flow cytometric measurements. We suppose that the changes observed in these two processes are due to a difference in chromatin organization between the first and the second period of the premitotic germination stage. In the second period, the four nucleosome core histories are synthesized in the absence of DNA replication and may correspond to a replacement of spherulation histone variants by plasmodial histone types in nucleosomes. The synthesis of historic H₄ clearly distinguishes the second period of the premitotic germination stage from a growing plasmodium G₂ phase, though nuclei exhibit a G₂ phase DNA content. The same pattern of histone synthesis has been found during the cell cycle following the first mitosis after germination and the growing plasmodium cell cycle, with a synthesis of two histories H₂B and H₂A and the high mobility group (HMG)-like protein AS₃ during the G₂ phase, i.e. in the absence of DNA synthesis.

Enzymes involved in the dynamic equilibrium of core histone acetylation of Physarum polycephalum

DEAE-Sepharose chromatography of extracts from plasmodia of the myxomycete Physarum polycephalum revealed the presence of multiple histone acetyltransferases and histone deacetylases. A cytoplasmic histone acetyltransferase B, specific for histone H4, and two nuclear acetyltransferases A1 and A2 were identified; A1 acetylates all core histones with a preference for H3 and H2A, whereas A2 is specific for H3 and also slightly for H2B. Two histone deacetylases, HD1 and HD2, could be discriminated. They differ with respect to substrate specificity and pH dependence. For the first time the substrate specificity of histone deacetylases was determined using HPLC-purified individual core histone species. The order of acetylated substrate preference is H2A much greater than H3 greater than or equal to H4 greater than H2B for HD1 and H3 greater than H2A greater than H4 for HD2, respectively; HD2 is inactive with H2B as substrate. Moreover histone deacetylases are very sensitive to butyrate, since 2 mM butyrate leads to more than 50% inhibition of enzyme activity.

ADP-ribosylation of core histones and their acetylated subspecies

ADP-ribosylation of core histones was investigated in isolated nuclei of Physarum polycephalum. Core histone species differed in the mode of modification. Whereas ADP-ribosylation of H2A and H2B is sensitive to inhibition by 3-methoxybenzamide, as with most other nuclear acceptor proteins, the modification of H3 and H4 is not inhibited. Cleavage experiments with hydroxylamine indicate a carboxylate ester type ADP-ribose-protein bond for H2A and H2B and arginine-linked ADP-ribose residues for H3 and H4. ADP-ribosylation preferentially occurs on acetylated histone subspecies, as shown for H4. These data are substantiated by the use of n-butyrate, which induces hyperacetylation of core histones; the butyrate-induced shift towards more acetylated H4 subspecies is accompanied by an increase of ADP-ribose incorporation into highly acetylated H4 subspecies.

Differential synthesis of histone H1 during early spherulation in Physarum polycephalum

H1 and P2 (an H1 degree/HMG-like protein) accumulate during exponential growth of Physarum microplasmodia (unpublished results), indicating that these proteins may play a role in differentiation (spherulation). To test this hypothesis, pulse labelling using [14C]lysine was used to determine whether any differential histone synthesis occurs during salts-induced spherulation. A peak in the uptake of [14C]lysine into microplasmodia was detected between 12 and 24 h following salts-induction. During the same interval, incorporation of label into the CaCl2-extracted histones occurred, with H1 being synthesised at approx. 3 times the level of the core histones and P2. Densitometry of SDS-PAGE gels showed that high levels of H1 were maintained up to 40 h in salts medium, beyond the observed peak in synthesis. The synthesis and accumulation of high levels of H1 during early spherulation indicates a role for this histone in the initiation and maintenance of a transcriptionally inactive differentiated state.

Histone hyperacetylation can induce unfolding of the nucleosome core particle

A direct correlation exists between the level of histone H4 hyperacetylation induced by sodium butyrate and the extent to which nucleosomes lose their compact shape and become elongated (62.0% of the particles have a length/width ratio over 1.6; overall mean in the length/width ratio = 1.83 +/- 0.48) when bound to electron microscope specimen grids at low ionic strength (1mM EDTA, 10mM Tris, pH 8.0). A marked proportion of elongated core particles is also observed in the naturally occurring hyperacetylated chicken testis chromatin undergoing spermatogenesis when analyzed at low ionic strength (36.8% of the particles have a length/width ratio over 1.6). Core particles of elongated shape (length/width ratio over 1.6) generated under low ionic strength conditions are absent in the hypoacetylated chicken erythrocyte chromatin and represent only 2.3% of the untreated Hela S3 cell core particles containing a low proportion of hyperacetylated histones. The marked differences between control and hyperacetylated core particles are absent if the particles are bound to the carbon support film in the presence of 0.2 M NaCl, 6mM MgCl2 and 10mM Tris pH 8.0, conditions known to stabilize nucleosomes. A survey of the published work on histone hyperacetylation together with the present results indicate that histone hyperacetylation does not produce any marked disruption of the core particle 'per se', but that it decreases intranucleosomal stabilizing forces as judged by the lowered stability of the hyperacetylated core particle under conditions of shearing stress such as cationic competition by the carbon support film of the EM grid for DNA binding.

Metabolic behaviors of the core histones in proliferating Friend cells

This study examines the turnover of the core histones in proliferating Friend cells. It was calculated that these proteins turn over with half-lives of 21.6 days for H2A, 13.8 days for H2B, 43.3 days for H3, and 138.6 days for H4. The significant differences in the half-lives of the four core histones indicate that the protein moiety of the nucleosome is not replaced as one entire unit but as a "mosaic" in which each component follows its own rate of replacement. In some experiments the turnover rates of the variants of H2A, H2B, and H3 were compared. The results did not indicate any differences among these histone variants, suggesting that they are not excluded from the mechanisms controlling histone turnover. Metabolic heterogeneity was discovered, however, when the turnover rates of the acetylated and nonacetylated molecules of histone H4 were followed: it appeared that the acetylated molecules are replaced 2.5 times faster. The comparison of the rate of replacement of the histones in proliferating and differentiated cells from one site and their level of acetylation from another suggests that this postsynthetic modification might be involved in the control of histone metabolism. Such a conclusion is supported also by a number of model experiments.

Flow cytometric determination of nuclear DNA content during differentiation (spherulation and germination) of the myxomycete Physarum polycephalum

Using flow cytometry, spherulating nuclei of Physarum isolated at the beginning of spherule wall formation were found to exhibit a DNA content corresponding to the G₂ phase of the cell cycle, although 8% lower. Before the first mitosis after spherule germination, a very slight incorporation of ³H thymidine into DNA was observed that was too weak to correspond to S phase, strongly suggesting that nuclei are stopped in G₂ phase inside the spherules. The lower value of nuclear DNA content found using flow cytometry of germinating spherules may not be related to DNA quantity, but may be due to a difference in chromatin organization during growth or spherulation, resulting in interference with the staining.

Histone acetylation in chicken erythrocytes. Rates of acetylation and evidence that histones in both active and potentially active chromatin are rapidly modified

Of the modifiable histone lysine sites, 2-4% participate in dynamic acetylation in chicken erythrocytes, suggesting the involvement of no more than 1-2% of the total genome. The rates and chromatin locality of this dynamic acetylation were studied in both chicken mature and immature red blood cells. In mature erythrocytes, two rates of acetylation of radiolabelled, monoacetylated H4 are observed, with half-lives of approximately 12 and approximately 300 min. In contrast, only one rate with a half-life (t1/2) of 12 min is observed in immature cells, and further experiments rule out the possibility of a slow rate of acetylation (with a t1/2 of approximately 300 min) for any form of H4 in this cell type. The simplest interpretation of these quantitative results, taken together with the behaviour of H3, H2B and H4 observed on the fluorograms used for rate analysis, is that a portion of the rapidly acetylated histone is converted to a more slowly acetylated form during erythrocyte maturation. The transcriptionally active adult beta-globin and H5 nucleohistone, which are presumably converted to potentially active chromatin during the maturation process, remain of the rapidly acetylated form in the mature cell.

Towards an understanding of the biological function of histone acetylation

A model is presented which explains the biological function of posttranslational acetylation of core histones in chromatin. Along the lines of this model histone acetylation serves as a general mechanism to destabilize nucleosome core particles during various processes occurring in chromatin. Acetylation acts as a signal that modulates histone-protein and histone-DNA interactions and finally leads to the displacement of particular histones from nucleosome cores. The high specificity of the acetylation signal for different processes (DNA replication, transcription, differentiation-specific histone replacement) is achieved by site specificity and asymmetry of acetylation in nucleosomes. The essential features of this model are in accord with the more recent results on histone acetylation.

Postsynthetic acetylation of histones during the cell cycle: a general function for the displacement of histones during chromatin rearrangements

Postsynthetic acetylation of core histones exhibits a peak during S-phase of the Physarum cell cycle. The maximum 3H-acetate incorporation precedes the maximum of histone synthesis. Acetate is incorporated into all core histones during S-phase, but only into H2A and H2B during G2-period. Resolution of acetylated H4-subspecies reveals acetate incorporation into preexisting H4, but not into newly synthesized molecules during mitosis and early S-phase. In a protamine competition assay histones from S-phase chromatin are released at lower protamine concentrations as compared to the lower acetylated G2-chromatin. We demonstrate a preferential release of highly acetylated H4-subspecies at low protamine concentrations. Our results fit into a general model of the relationship between histone acetylation and chromatin assembly. According to this model acetylation of core histones would serve as a signal for displacement of histones from nucleosomes by modulating histone-protein or histone-DNA interactions. We propose that this mechanism operates during DNA-replication and transcription, as well as during other chromatin rearrangements.

Histone acetyltransferase activity during the cell cycle

Histone acetyltransferase activity was measured in isolated nuclei during the synchronous cell cycle of the myxomycete Physarum polycephalum. Nuclei were incubated with [14C]acetyl-coenzyme A and an excess of exogenous calf thymus histones. The activity is periodic during the cell cycle; it rises during the S-phase to reach a maximum in the early G2-period with a decline in mid and late G2. Comparison of the pattern of enzyme activity with the in vivo acetylation of histones during the cell cycle reveals that the enzyme activity does not wholly determine the acetylation state, indicating that other factors, including possibly the structural state of chromatin, are responsible for the observed cell cycle pattern of in vivo histone acetylation.