Tài liệu Báo cáo khóa học: Trichostatin A reduces hormone-induced transcription of the MMTV promoter and has pleiotropic effects on its chromatin structure pptx

We have used Xenopus laevis oocytes to study the effects of TSA on glucocorticoid receptor GR-dependent transcription and we have related these effects to changes in the chromatin structur

Trichostatin A reduces hormone-induced transcription of the MMTV

promoter and has pleiotropic effects on its chromatin structure

Carolina A˚strand1,*, Tomas Klenka1,*, O¨rjan Wrange1and Sergey Belikov1,2

1

Department of Cell and Molecular Biology, Medical Nobel Institute, Karolinska Institute, Stockholm, Sweden;2D.I.Ivanovsky Institute of Virology, Moscow, Russia

The deacetylase inhibitor trichostatin A (TSA) has long

been used to study the relationship between gene

transcrip-tion and the acetylatranscrip-tion status of chromatin We have

used Xenopus laevis oocytes to study the effects of TSA on

glucocorticoid receptor (GR)-dependent transcription and

we have related these effects to changes in the chromatin

structure of a reporter mouse mammary tumor virus

(MMTV) promoter We show that TSA induces a low level

of constitutive transcription This correlates with a change of

acetylation pattern and a more open chromatin structure

over the MMTV chromatin, and with specific acetylation

and remodeling events in the promoter region Specifically, a

repositioning of initially randomly positioned nucleosomes

along the distal MMTV long terminal repeat is seen This

nucleosome rearrangement is similar to the translational nucleosome positioning that occurs upon hormone activa-tion We also note a reduced hormone response in the presence of TSA TSA effects have for a long time been associated with transcriptional activation and chromatin opening through inhibition of the deacetylation of histones However, our results and those of others show that TSA-induced changes in expression and chromatin structure can

be quite different in different promoter contexts and, thus, the effects of TSA are more complex than previously believed

Keywords: MMTV promoter; chromatin structure; tran-scription; Xenopus oocytes; TSA

The role of the nucleosome as the fundamental unit of DNA

packaging has long been accepted, but its purely structural

role has been challenged by an increasing body of

experi-mental data [1] Recent evidence suggests that the

organ-ization of promoters into nucleosome arrays provides

an additional mechanism of gene regulation [2] In this

study, we have used a promoter from the 5¢-long terminal

repeat (LTR) region of the mouse mammary tumor virus

(MMTV) to correlate chromatin structure and gene activity

The MMTV-LTR contains potential regulatory elements

which mediate transcription in the presence of

glucocorti-coid ligands and in the presence of androgen, progesterone

and their respective nuclear receptors (Fig 1A) [3] Six

translationally positioned nucleosomes (A–F) cover this

region [4], one of which, nucleosome B, covers the DNA

segment around position )60 to )240 This segment

contains four glucocorticoid response elements (GREs)

[4–6] This whole DNA segment shows increased

hyper-sensitivity to DNase I upon binding of glucocorticoid receptor (GR) homodimers [4,7,8]

We have used the Xenopus oocyte system to reconstitute chromatin in vivo using single stranded DNA containing the MMTV promoter as a template Single-stranded DNA reconstitutes chromatin more effectively than double-stran-ded DNA as the second-strand synthesis is coupled to chromatin assembly, and thus, seems to mimic the replica-tion coupled chromatin assembly occurring during S phase

of the cell cycle [9]

While the ordered helical domains in the globular body

of the core histones provide a structure for DNA to wrap around [10], the N-terminal histone tails have been shown to protrude through and around the DNA helix in a far less ordered manner [11] They harbor positively charged lysine residues at conserved positions These lysine residues have been shown to act as targets for post-translational modifi-cation [12] Deletion of H3 and H4 N-terminal tails is a lethal event in yeast that significantly alters gene regulation, nucleosome assembly and spacing [13] It is believed that reversible modifications of charged residues can alter chromatin structure by causing changes in the overall charge of the N-terminal tails, and hence their interactions with the negatively charged sugar–phosphate DNA back-bone, or with negatively charged regions located on adjacent nucleosomes [11] An alternative view is that the various chemical modifications of specific amino acids in histones act as a code by serving as binding sites for various effector complexes These complexes can modify the chromatin structure and hence the expression of a gene [14]

The relationship between the histone acetylation status

of chromatin and transcription has been studied in many systems using a variety of promoter constructs and native

Correspondence to S Belikov, Department of Cell and Molecular

Biology, Medical Nobel Institute, Box 285 Karolinska Institute,

SE-171 77 Stockholm, Sweden Fax: + 46 8 31 35 29,

Tel.: + 46 8 52 48 73, E-mail: sergey.belikov@cmb.ki.se

Abbreviations: ChIP, chromatin immunoprecipitation; DMS,

dimethylsulphate methylation; GR, glucocorticoid receptor; GRE,

glucocorticoid response element; HAT, histone acetyltransferase;

HDAC, histone deacetylase; LTR, long terminal repeat; MNase,

micrococcal nuclease; MPE, methidiumpropyl-EDTA–Fe(II); NaBu,

sodium butyrate; TSA, trichostatin A; TA, triamcinolone acetonide.

*Note: Both these authors contributed equally to this work.

(Received 26 August 2003, revised 26 January 2004,

accepted 30 January 2004)

genes The studies have revealed that both histone

acetyl-transferases (HATs) and histone deacetylases (HDACs)

play a vital role in gene regulation by either allowing

transcription or establishing correct repression Blocking

of HDACs using inhibitors such as trichoststin A (TSA),

sodium butyrate (NaBu) and trapoxin [15], has revealed

a complicated picture of the exact role of HDACs in

promoter function TSA has been shown to relieve

repres-sion by non ligand-bound TR/RXR protein bound to

exogenous TRbA promoter constructs in Xenopus oocytes

[16], as well as p53/mSin3A-repressed genes in the

mam-malian cell lines [17] Such effects are due to the inhibition

of HDACs, targeted by specific DNA binding factors to

transcriptionally silent regions as a part of large corepressor

complexes [18] The endogenous Xenopus H1 gene can be

activated in cell lines by TSA, but only after the mid blastula

transition, when histones become hyperacetylated in the

presence of TSA and NaBu [9,19]

In this study, we analyze TSA-treated chromatin in

Xenopus oocytes, and relate its structure to the function

of an MMTV-LTR reporter construct We show that TSA

increases the acetylation of bulk histone H3 as well as

H3 acetylation over the MMTV–LTR Furthermore, TSA

treatment causes a generally more open chromatin

struc-ture, and increases DNA-accessibility to micrococcal

nuc-lease (MNase) in the MMTV promoter It also triggers a

nucleosome repositioning in the distal part of the MMTV

LTR, similar to the nucleosome rearrangement that occurs

during hormone activation [6] Our results, and the

results of others, highlight the pleiotropic effects that TSA

administration has on chromatin structure and on gene

expression

Materials and methods

DNA and plasmids

Construction of the MMTV reporter and the plasmid for

in vitrotranscription of rat GR mRNA has been described

previously [6]

Culture and injection ofXenopus oocytes

Xenopus laevis oocytes were prepared and injected as

described previously [20]

Transcription analysis

Quantification of MMTV transcription by S1-nuclease and

DNA analysis was performed as described previously [21],

with one difference A synthetic oligonucleotide identical

to the lower strand of the )8/+64 DNA segment of the

MMTV-LTR was labeled using [32P]ATP[cP] (Amersham

Biosciences) and T4-polynucleotide kinase, and used as

probe

MMTV transcription was also quantified by primer

extension using the following procedure Homogenate

equivalent to eight oocytes was first treated with

0.5 mgÆmL)1 proteinase K for 2 h at 37C and then

RNA extracted with Trizol (GibcoBRL) and chloroform

according to the manufacturer’s instructions, precipitated

with 0.7 vol of isopropanol One oocyte equivalent was

used for primer extension The primers were32P end-labeled oligonucleotides with the following sequences: 5¢-GC GGGAGTTTCACGCCACCAAGATCC-3¢ (MMTV, 10 pmol) and 5¢-GGCTTGGTGATGCCCTGGATGTTAT CC-3¢ (H4, loading control, 20 pmol) Primer extension was performed according to a protocol modified from [22]: dry RNA pellets were resuspended in 4 lL of each primer dilution and 2 lL 5· First Strand buffer (GibcoBRL), primers were then annealed at 95C for 10 min, 55 C for

25 min, 45C for 10 min Extension was performed in

20 lL at 45C with 1 lL Superscript II (–RNaseH) RT (GibcoBRL), in 10 mMdithiothreitol, 0.5 mMdNTP for a further 40 min Samples were diluted 1 : 1 (v/v) with denaturing loading buffer and run on 6% polyacrylamide sequencing gels: extension products were analyzed and quantified on a Fuji Bio-Imaging analyzer BAS-2500 using IMAGE GAUGEV3.3 software

Chromatin and protein–DNA analysis Micrococcal nuclease (MNase) digestion and in situ cleavage

by methidiumpropyl-EDTA–Fe(II) (MPE) was performed

as described previously [6] as was the supercoiling assay [23] that used a chloroquine concentration of 60 lgÆmL)1 Radioactivity scans and quantifications were performed using a Fuji Bio-Imaging analyzer BAS-2500 withIMAGE GAUGEV3.3 software

Analysis of proteins extracted fromXenopus oocytes Nuclear proteins were isolated by physical separation of the germinal vesicle from the cytoplasm of injected oocytes using fine forceps in a pool of isolation buffer (20 mMTris/ HCl, pH 7.5, 0.5 mMMgSO4, 140 mMKCl) Both fractions were homogenized in the same buffer with 1% SDS, boiled for 5 min, and samples run on 10% or 15% SDS/PAGE gels (for GR and histone analysis, respectively) in Tris/ glycine buffer with 0.1% SDS [24] Proteins were transferred onto poly(vinylidene difluoride) (PVDF) membranes (Mil-lipore) in Tris/glycine buffer containing 20% (v/v) methanol and 0.037% (w/v) SDS [24] at 20 V for 1 h Filters were probed with antibodies to acetylated histones H3 and H4 (Upstate Biotechnology) and acetylated H3 (Upstate Bio-technology), and antibodies against specific modifications such as acetylated H3-K9 (Cell Signaling Technology) and H3-K14 (Abcam) and anti-H3 C-terminal (Abcam) Ana-lysis of GR was performed at 1 : 1000 dilution of primary antibody in Tris-buffered saline with 0.05% (v/v) Tween 20 (TTBS) and 5% (w/v) dried milk powder Secondary antibody HRP conjugates were used at 1 : 1000 dilution in TTBS Protein bands were visualized by chemiluminescence (GibcoBRL) Quantification was via IMAGE GAUGE V3.3 software For an internal standard and loading control, the oocytes were incubated in oocyte medium also containing [35S]methionine (Amersham Biosciences) for 5 h After Western blotting, the filters were analyzed for radioactivity using a Fuji Bio-Imaging analyzer BAS-2500 as above Chromatin immunoprecipitation

Pools of oocytes were injected with 4.5 ng sspMMTV [6] and treated with or without TSA prior to fixation with 1% (v/v)

formaldehyde for 10 min at ambient temperature

Chro-matin immunoprecipitation (ChIP) was performed according

to a protocol described previously with some modifications

[25] Cells were washed and 12 nuclei per pool were dissected

and collected in sonication buffer (20 mMTris/HCl, pH 7.2,

60 mMKCl, 15 mMNaCl, 1 mMEDTA, 1 mM

dithiothre-itol and 1· protease inhibitor cocktail (Sigma), 50 lL per

nucleus) After sonication on ice (4· 20 s), samples were

diluted with 1 vol of buffer I (0.1% sodium deoxycholate,

1% Triton X-100, 2 mM EDTA, 50 mM Hepes pH 7.2,

150 mMNaCl, 1 mMdithithreitol and 1· protease inhibitor

cocktail (Sigma) and centrifuged at 13 000 g, 4C, for

10 min to remove insoluble debris Supernatant,

equival-ent to one nucleus, was used for immunoprecipitation

Acetylated histones were immunoprecipitated for 4 h with

anti-AcH3 (Upstate Biotechnology), H3 C-terminal and

AcH3-K14 (Abcam) on protein A sepharose beads

(Amer-sham) precoated with calf thymus DNA in 5% (w/v) dry

milk Complexes were washed (15 min wash) with buffers:

buffer I described above; buffer II [0.1% (w/v) sodium

deoxycholate, 1% (v/v) Triton X-100, 2 mMEDTA, 50 mM

Hepes, pH 7.2, 500 mM NaCl, 1 mM dithiothreitol and

1· protease inhibitor cocktail]; buffer III [0.25MLiCl, 0.5%

(v/v) NP-40, 0.5% (w/v) sodium deoxycholate, 1 mM

EDTA, 10 mMTris/HCl, pH 8.0, 1 mMdithiothreitol and

1· protease inhibitor cocktail] and TE, pH 8.0 (1 mM

dithiothreitol, 1· protease inhibitor cocktail) Bound

mater-ial was eluted in elution buffer [0.5% (w/v) SDS, 0.1M

NaHCO3, 0.5 lgÆlL)1 proteinase K] Crosslinking was

reversed at 65C overnight and DNA was purified by

extraction with phenol/chloroform and isopropanol

preci-pitation PCR was performed in 21 cycles with primers

covering the nucleosome B ()291/+42), the

nucleo-some F ()1044/)732) and the M13 vector (2699/2990), and

products were analyzed on a 6% (w/v) polyacrylamide

sequencing gel Radioactivity scans and quantifications were

performed using a Fuji Bio-Imaging analyzer BAS-2500

usingIMAGE GAUGEV3.3 software

Results

Trichostatin A has pleiotropic effects on MMTV

transcription

Pools of oocytes were injected with 5 ng GR mRNA into the

cytoplasm, followed by intranuclear injection of 1 ng of

sspMMTV:M13 DNA TSA (16 nM) was added to some of

the pools immediately after DNA injection to assemble

chromatin in the presence of TSA, this is referred to as early

TSA (E) Hormone induction of half of these pools was

performed 18 h later by addition of the synthetic

glucocor-ticoid hormone, triamcinolone acetonide (TA) at a

concen-tration of 10)6M TSA was added, at the same time, to a

pool of injected oocytes, these are referred to as late TSA

(L) Transcription was allowed to continue for 6 h Oocytes

were harvested and total RNA was extracted from all pools

(Fig 1B) S1 nuclease or primer extension analysis of the

injected MMTV reporter transcripts was performed; the

latter using a primer for endogenous histone H4 RNA as an

internal control, followed by quantification on a

phosphor-imager (eight experiments) Both assays generated identical

results

Hormone induction of the MMTV promoter caused a dramatic increase in transcription over the basal level This basal level was
0.5% of full induction (Fig 1C, compare lanes 1, 2 and 3, 4) However, TSA alone tended to induce transcription at a weak level, also in the absence of hormone (Fig 1C, compare lanes 1, 2 and 5, 6 and 9, 10) A further effect of TSA treatment was a significant reduction in hormone-induced transcription (
50% of full induction when added early, compare lanes 3, 4 and 11, 12) Neither of these effects depended on the TSA concentration over the range used in these experiments, i.e 16 nMand 64 nM(data not shown) This agrees with similar studies that used TSA

to affect gene transcription [9,16,26] A TSA concentration

of 16 nMwas used for subsequent experiments as this was enough to elicit a reproducible response

The TSA-induced, hormone-independent, or leaky, transcription was clearly seen only in the case of early TSA-treated oocyte pools (Fig 1D), [1.74 ± 0.4% (E) vs 0.60 ± 0.5% (L); n¼ 8] Similarly, the TSA-mediated effect of reducing the response to hormone was less evident when added late [52.5 ± 17.1% (E) of full hormone induction compared to 76.2 ± 21.6% (L) (n¼ 8)] We conclude that TSA causes a weak hormone-independent transcription of the MMTV promoter and partly inhibits hormone-inducible transcription

TSA affects acetylation levels of endogenous histone pools as well as histones in MMTV containing minichromosomes

Treatment of oocytes with deacetylase inhibitors such as TSA may change the bulk acetylation pattern of histones, and in this way alter the structure of chromatin incorpor-ating them To see whether the increased transcription leakage observed at early addition of TSA can be explained

by an accumulation of acetylated forms of histones in

a time-dependent manner, we looked at the pattern of TSA-induced histone acetylation

Using antibodies specific to acetylated forms of H3 and H4, we endeavored to look at the level of histone hyper-acetylation 12 h (E) and 18 h (L) before the harvest of noninjected oocytes Nuclei isolated from nontreated oocytes showed significant levels of nuclear AcH4, which did not increase upon late addition nor on early addition of TSA (Fig 2A) A striking response to TSA was seen in the levels of AcH3: almost no AcH3 was found in nontreated oocytes while oocytes treated with late TSA showed a significant increase in AcH3 This increase was even more pronounced when TSA was added early (Fig 2A) We conclude that TSA induces a time-dependent increase in the level of AcH3 We also looked for specific histone modifications (Fig 2B) and noted an increased acetylation of histone H3 lysine residues

14 and 9 upon early addition of TSA

To monitor the acetylation status of the MMTV promoter subjected to TSA treatment, we used a chromatin immunoprecipitation assay (ChIP) and evaluated the acetylation status of the so-called B- and nucleosome F [4] and compared these patterns with the M13 vector (Fig 2C)

As a control for the potential loss of histone–DNA contacts during treatment, an antibody against the carboxyterminal segment of histone H3, which is not subjected to any known modifications, was also included [27] The ChIP analysis

demonstrated TSA-dependent five- to 10-fold increase in

histone H3 acetylation which involved both the MMTV

promoter, the nucleosome B, the distal MMTV LTR, here

presented by the nucleosome F, and the M13 vector DNA

(Fig 2C)

We conclude that early TSA addition increases the

acetylation status of bulk histones as well as the histones

organizing the minichromosomes

Structural alterations in nucleosomal organization

caused by TSA treatment

Changes in the acetylation status of histones by TSA

treatment may cause changes in the organization of

chromatin Such altered chromatin may no longer be able

to repress transcription from inducible promoters and it may have less capacity to organize effective transcription in the induced state We used several methods to look at chromatin structure and chromatin remodeling within the MMTV Chromatin remodeling can be followed by in situ chromatin digestion with appropriate restriction enzymes [8,28] A restriction enzyme accessibility assay utilizing a SacI or HinfI restriction site revealed a hormone-dependent remodeling of the chromatin in this region [6] However, similar experiments failed to show any significant effect of TSA on SacI or HinfI accessibility (data not shown) We therefore used other approaches that were more sensitive to the small changes in the chromatin structure over a wide area of the nucleosome B: MNase digestion assay, topology assay and MPE chemical cleavage together with indirect end-labeling

Previous MNase experiments revealed changes in the canonical nucleosomal ladder over the nucleosome B region

in response to hormone-dependent GR binding [6,20] As seen in Fig 3, increasing amounts of MNase reduced the nucleosome B region of the hormone-activated promoter

to predominantly mono- and subnucleosomal fragments Compare lanes 4–6 (nucleosome B probe) with lanes 1–3 (nucleosome B probe) and also Fig 3B (left panel) This hormone-dependent appearance of a subnucleosome is specific for the nucleosome B area as it is not seen while reprobing the filter with the vector probe Compare lanes 4–6 (nucleosome B probe) with lanes 4–6 (M13 vector probe) The nucleosome B is further affected by the early addition of TSA In this case, the nucleosome B area is distinctly hypersensitive to MNase action, especially at high concentrations Late addition of TSA, on the other hand, had virtually no effect (compare lanes 4–6 with 10–12 and 16–18, nucleosome B probe) Thus, the observed reduction

in hormone response of the system in the presence of TSA correlates with detectable changes in the local chromatin architecture of the MMTV promoter The TSA-induced leaky transcription also seems to correlate with a loss in chromatin structure regularity of the bulk chromatin; at lower MNase concentrations this is seen as increased

Fig 1 TSA decreases hormone-induced transcription of the MMTV promoter, and increases basal transcription in the absence of hormone (A) The reporter DNA construct, the pMMTV:M13 used for injection with the primer used for primer extension analysis of DMS methyla-tion protecmethyla-tion (solid black arrow), and the restricmethyla-tion enzyme cleavage sites that are referred to in the text White boxes designate GRE hexanucleotide elements numbers I to IV, the black box shows the NF1 site, dark gray boxes show the Oct 1 sites, and light gray box shows the TATA sequence The nucleosome B probe used in the MNase experiments is shown below (B) Time-course of the oocyte injection experiment Collagenased oocytes were allowed to recover for

18 h prior to injection of GR mRNA and DNA, TSA addition [early (E) or late (L)] and hormone induction RNA and DNA were extracted from pools of eight oocytes each (C) Representative dena-turing acrylamide gel showing analysis in duplicate of the MMTV transcription in the presence of hormone and TSA (D) Phosphor-imager analysis of MMTV transcription assayed by primer extension normalized to H4 The lower panel shows a smaller scale graph highlighting the increase in basal transcription Error bars signify SD (n ¼ 8).

smearing of the nucleosomal pattern both in the promoter

and vector sequences (Fig 3A, both panels, compare lanes

1, 7 and 13 and Fig 3B, right panel)

Topological changes in chromatin induced by TSA

treatment and GR binding

We have demonstrated previously that hormone-dependent

activation of the MMTV promoter is associated with

alterations in the chromatin structure that can be detected in

a DNA topology assay as the loss of negative superhelical turns [20] These alterations in DNA topology take place even if histones are not physically disrupted from the chromatin template [29]

Oocytes were injected with sspBSLSwt, a construct containing the same MMTV-TK fusion used in other experiments, but cloned into a Bluescript vector The size of the injected DNA was smaller in these experiments, and the resolution of the topoisomers was thus improved Following treatment, oocyte pools were extracted and the DNA resolved on an agarose gel containing 60 lgÆmL)1 chloro-quine to visualize any changes in superhelical density arising from TSA/hormone treatment Treatment with TSA decreased the negative superhelicity by 1.5 superhelical turns, equal to 1.5 nucleosomes (Fig 4 lanes/scans 1, 3 and 5) This indicates a more open conformation of the chromatin, which correlates with a loss of chromatin regularity and is consistent with increased smearing observed in the MNase-digested DNA Interestingly, this phenomenon was as evident in the presence of early TSA

as it was with late TSA, suggesting that changes in the topology may occur quickly The overall change in the topology caused by GR binding and the MMTV induction results in an overall loss of about 7 negative supercoils, an effect which was decreased by TSA treatment by two and one superhelical turns for early and late TSA, respectively (Fig 4 lanes/scans 2, 4 and 6) This indicates that in contrast

to the uninduced promoter, TSA treatment during hormone activation leads to a less open chromatin structure This observation agrees with the reduced hormone-dependent transcription from the MMTV promoter in the presence of TSA (Fig 1D)

TSA treatment causes nucleosome repositioning within the MMTV LTR

For mapping of the translational nucleosome positioning along the MMTV LTR, we have used the chemical nuclease MPE, which has a strong preference for internucleosomal regions and shows significantly less sequence bias in cleaving DNA than MNase [6,30] The MPE cleavage data suppor-ted the previous finding [6] that hormone induction causes

a dramatic remodeling event within the MMTV LTR, resulting in hypercutting over the nucleosome B area, protection of the nucleosome C area and repositioning of initially randomly positioned nucleosomes (Fig 5A and B, compare lanes 1, 2, and 3, 4 and corresponding scans) Quite unexpectedly, we observed a hormone-independent remodeling event within the MMTV-LTR after the addition

of TSA On early addition of TSA alone, the pattern

of remodeling was seen over the region covered by nucleosomes C–F, which resembles the pattern obtained

by hormone treatment in the absence of TSA (Fig 5A and

B, compare lanes 5, 6 to lanes 1, 2 and lanes 3, 4 and corresponding scans) This effect was detectable but less evident when TSA was added after chromatin assembly, i.e late TSA treatment (compare lanes 5, 6 and 9, 10) No significant effects of TSA treatment were detected on nucleosome B On the other hand, simultaneous addition

of TA and TSA resulted in a digestion pattern indistin-guishable from that observed after treatment with TA alone

Fig 2 TSA treatment causes acetylation of bulk histones and

acetyla-tion of histones in MMTV-containing minichromosomes Pools of

dis-sected nuclei from noninjected oocytes were analyzed by SDS/PAGE.

(A) Western blot probed with anti-acetylated H3 (upper panel) and

anti-acetylated H4 (lower panel) In vivo [ 35 S]methionine-labeled

pro-teins in the nuclear extract were detected on the filter after blotting and

were used as an internal standard This showed that equal amounts of

protein were loaded in each lane (not shown) (B) Western blot probed

with antibodies against AcH3-K14, AcH3-K9 and H3 C-terminal.

TSA was either not added (-), added early (E) or added late (L),

according to the schedule in Fig 1B (C) Effects of TSA treatment on

H3 acetylation at different regions of the MMTV promoter and vector

sequences The DNA-injected oocytes were treated (early) or not

treated with TSA The ChIP assay was performed as described in

Materials and methods Radioactively labeled PCR fragments from

nucleosome B, nucleosome F and vector are shown to the left, and

corresponding bars to the right Bars represent the intensity of the

bands, normalization was performed according to actual histone

H3-DNA binding (H3 C-term) in TSA-treated and nontreated

oocytes Dark bars show TSA-treated cells.

(Fig 5A and B, compare lane 4 and 8, lower scan) We

conclude that TSA can cause specific nucleosome

rear-rangements in the distal MMTV-LTR similar to the

hormone-induced rearrangements in this DNA segment

TSA treatment does not affect GR binding

to the chromatin template

Graphical calculation of the total GR expressed in an

oocyte following injection of 5 ng GR mRNA, and

com-parison to a standard dilution curve (Fig 6A) allowed us to

estimate an average of 67 ng of GR protein is present in

each oocyte under the injection conditions used This is

equivalent to 0.76 pmol per oocyte (relative molecular mass

of GR¼ 87 500) We also analyzed the nuclear localization

of GR in nuclei microdissected from TSA treated/untreated

oocytes and found no difference in localization patterns

between the oocyte pools (data not shown) To find out

whether the reduced hormone response of the MMTV

promoter after addition of TSA could be a result of

compromised binding of GR to GREs in a hyperacetylated

chromatin context, we analyzed GR–DNA interactions by

dimethylsulphate (DMS) methylation [8], and the cleavage

of DNA by alkali [31] The method allows easy detection of

DNA–protein interactions via the N7 position of guanines

in the major groove and via the N3 position of adenines in

the minor groove The DMS cleavage pattern was

devel-oped by primer extension (Fig 6B) The pattern of the

nonhormone-induced MMTV promoter is virtually

identi-cal to that obtained for naked DNA (data not shown)

Hence, there is no protein binding detected by the DMS

methylation assay in the MMTV promoter in the

noninduced state Addition of hormone resulted in a drastic reduction in DMS methylation (protection) over the glucocorticoid response elements In agreement with our previous results [20], we observed
40% DMS methylation

of the corresponding guanines over the GREs 1–4 (Fig 6B, compare lanes 1 and 2 and corresponding radioactivity scans) Addition of TSA to the oocyte media had virtually

no effect on the DMS digestion pattern (Fig 6B, compare lanes 1, 2 and 3, 4) This shows that TSA has no effect on GR-DNA binding

Discussion

We have shown that TSA treatment of oocytes alters the MMTV transcription profile and causes changes in the bulk chromatin structure, as well as specific changes in the promoter region To the best of our knowledge this is the first report to demonstrate a specific translational reposi-tioning of nucleosomes induced by TSA or any other HDAC inhibitor

The fact that we see effects more clearly when TSA is present during chromatin assembly (early addition) suggests that the deacetylation step of chromatin maturation is being blocked Pools of newly synthesized histone H4 diacetylated

at the evolutionarily conserved K5 and K12 residues are known to exist in a variety of organisms [32], and diAcH4 is also the main form of stored H4 in Xenopus oocytes [19] Newly synthesized H3 has also been found in a diacetylated form in Drosophila and Tetrahymena, although this appears

to be more transient, and the pattern of lysine residues acetylated in this manner is far less well conserved between species [32] We were able to detect only tiny amounts of

Fig 3 Nucleosomal organization of the MMTV promoter (A) Autoradiogram of a Southern blot of chromatin digested with increasing amounts of MNase, probed with vector M13 DNA (left) and reprobed with nucleosome B probe (right) Positions of bands corresponding to tri-, di-, mono-and subnucleosomal bmono-ands are indicated (B) Phosphorimager profiles of individual lanes from (A), indicating changes in MNase digestion on treatment of oocytes with TA

or TSA Positions of tri-, di-, and the mono-nucleosomal bands are indicated, as is the hormone-induced subnucleosomal fragment.

AcH3 in resting oocytes in our experiments, and the other

researchers have not detected AcH3 in HeLa cells [32] 2D

PAGE analysis of Xenopus oocytes treated with NaBu or

TSA have revealed little change in the overall H4

acetyla-tion, and inconclusive changes to AcH3 until the

mid-blastula transition [19] In agreement with this, our

experiments did not show any change in the level of AcH4

following treatment with TSA, but we did see a distinct and

time-dependent increase in the hyperacetylation of histone

H3 as well as an increased acetylation at specific sites, i.e

lysines 14 and 9 (Fig 2B)

There are a number of reasons to suspect, a priori, that

changes in the acetylation status of histones result in

alterations in chromatin structure and DNA–protein

inter-actions We have used several methods to address this issue

We have not found any differences in GR binding to GREs

with or without TSA (Fig 6) At the same time, histone

acetylation facilitates the binding of TFIIIA, GAL4 and

USF to nucleosomal DNA in vitro [33–35] The restriction

enzyme accessibility assay, which is a sensitive method that

is capable of detecting even subtle changes in chromatin

structure, also failed to reveal any difference between

chromatin treated with TSA and untreated chromatin (data

not shown)

Fig 4 Changes in DNA topology upon hormone induction of the

MMTV promoter and addition of TSA Southern blot of pBSLSwt

minichromosomes extracted from GR-injected oocyte pools treated

with TSA and/or hormone, followed by separation on an agarose gel

containing 60 lgÆmL)1chloroquine to reveal the superhelical density;

probed with radiolabeled DNA fragment )103/+431 One oocyte

equivalent per lane Phosphorimager profiles of scanned lanes showing

the distributions of superhelical species are shown below The circle

indicates the most frequent topoisomer(s).

Fig 5 Nucleosome positioning analysis in situ by MPE digestion (A) Southern blot of MPE-cleaved DNA from treated oocyte pools, probed with radiolabeled 513 bp probe (EcoRV-SacI fragment, +425/ )108) The diagram to the left shows the positions of nucleosomes on the MMTV promoter [6], the transcriptional start site and the cleavage sites for SacI (S), HinfI (H) and BamHI (B) (B) Radioactivity scans from selected lanes.

The structural changes over the whole minichromosome

upon TSA treatment, as indicated by the increased

acces-sibility of MNase, have been reported previously [16] Our

MNase digestion results, when coupled with the topology

changes resulting from the loss of 1.5 superhelical turns

of the DNA over the whole construct, suggest that the

chromatin is indeed more relaxed in the absence of

hormone Interestingly, this phenomenon is as evident in

the presence of early TSA as it is with late, suggesting that

changes in the topology may occur quickly, and that these

changes may be a sensitive readout of histone

hyperacety-lation The MMTV-LTR specific structural changes in

chromatin, detected by MPE, require a longer exposure to

TSA to develop and they require exposure to TSA during

second-strand synthesis and chromatin assembly On the

other hand, hormone activation results in an overall loss of

about seven negative supercoils This effect was decreased

by TSA treatment by one or two superhelical turns

following TSA treatment, which indicates that addition of TSA leads to formation of less open structure This is in striking correlation with a reduced hormone response in TSA-treated oocytes Our results are in good agreement with those previously published from studies in vitro [29] and in vivo [36] on the effects of TSA on DNA topology However, changes in the DNA topology of minichromo-somes assembled with acetylated/nonacetylated histones were not significant in other studies [16,37] Experimental data are consistent with an idea that in a hyperacetylating environment, the net charge over positively charged lysine residues on histone tails will be neutralized, thus altering histone–DNA [38] and, possibly, histone–histone [11] inter-actions This alteration would eventually lead to a decrease

in chromatin compaction [39]

Previously, we have shown that hormone induction in Xenopusoocytes results in the establishment of a specific nucleosome positioning pattern over initially randomly organized nucleosomes in the MMTV promoter [6] After treatment of injected oocytes with TSA alone, an altered pattern of MPE cleavage over nucleosomes C and D is seen (Fig 5), reflecting a partial, hormone-independent chroma-tin remodeling event This mimics the situation occurring during hormone induction, where GR bound to the nucleosome B region of the MMTV promoter renders an array of six positioned nucleosomes (A–F) A TSA-induced chromatin remodeling event occurs in the HIV promoter This promoter harbors a nucleosome positioned at the initiation site that is disrupted upon TNF-a-induced expression Treatment with TSA results in a similar chromatin remodeling of the HIV promoter and in consti-tutive transcription [40] However, we are not aware of any previous report of a specific translational positioning being induced by TSA Our observations suggest that nucleosome repositioning during hormone induction [6] is not solely explained by GR binding to the GRE sequences within the nucleosome B region, but also depends on the subsequent histone acetylation in the surrounding area Indeed, it was shown that histone H4 acetylation at Lys8 is responsible directly for the recruitment of the SWI/SNF complex to IFN-b gene during activation [41] Another possibility is that the TSA-induced remodeling of the MMTV promoter that we have observed is triggered by acetylation of a transactive factor(s), which might lead to specific binding to the distal part of the MMTV-LTR and thereby direct the translational nucleosome positioning

The acetylation status of histones H3 and H4 associated with different parts of the MMTV promoter has been studied recently using ChIP [42–44] Rather unexpectedly,

it was shown that upon activation, promoter-proximal histones (the nucleosome B area) become deacetylated whereas the acetylation of both H3 and H4 of nuclesome

F was increased [42] Addition of TSA resulted in only an insignificant increase of the acetylation level of histone H4 in the nucleosome B region [43] These conclusions were made assuming that the overall amount of histone–DNA cross-links induced by formaldehyde in the nucleosome B and F areas are the same However, this might not be the case, given the strong remodeling of nucleosome B that occurs during transcription activation [6,7] This remodeling might result in the partial loss of histone–DNA contacts in the nucleosome B area Thus, the decrease of the acetylated

Fig 6 DMS methylation protection over the nucleosome B segment.

(A) SDS/PAGE and Western blot of total oocyte protein extracts

following injection of 5 ng GR mRNA, probed with GR polyclonal

antibodies 0.5 and 0.25 oocyte equivalent was compared to a standard

curve of GR protein of known concentration purified from rat liver

[56] (B) DMS methylation protection over the nucleosome B segment

in the presence/absence of TA and TSA Oocytes in groups of five were

treated with DMS, see Materials and methods The methylation

pat-tern was developed by primer (+42/+15) extension Corresponding

guanidine residues that are protected after hormone induction are

indicated with arrows Radioactivity scans of corresponding lanes are

shown to the right.

signal in ChIP experiments might indicate a loss of histone

in the respective area [27] Our results and the results of

others [25] show an increase in acetylation status of histone

H3 over the MMTV-LTR upon TSA treatment This

hyperacetylation was also seen in the vector sequences

Interestingly, our results show that a small but clearly

detectable increase in the level of basal transcription occurs

upon the TSA treatment This shows that TSA can only

insignificantly overcome the repressive nature of the

chro-matin We also discovered a reduction of
50% in the

hormone response of the system in the presence of TSA,

whereas previous studies on MMTV have reported

aug-mentation of hormone induction by TSA [26,45,46]

However, addition of HDAC inhibitors resulted in the

down-regulation of the MMTV transcription to various

extents in several studies [43,47,48] The inhibitory effect in

our experiments was more evident in the case of early TSA

treatment, suggesting that HDAC activity during chromatin

maturation may not only help to establish sufficiently

repressive chromatin, but may also be necessary for the

formation of transcriptionally competent chromatin [49]

Deacetylase activity has for a long time been associated

with transcriptional repression through the deacetylation of

histones [50] However, several studies have shown that

HDACs are required for both transcriptional activation and

repression [51] One recent example of the down-regulation

by HDAC inhibitors is that of the STAT 5 target genes [52],

where transcription involves recruitment of HDAC1 [53]

Chromatin remodeling of these genes is not affected by

TSA, but recruitment of the components of the basal

transcription machinery is blocked [52] Interestingly, GR is

able to recruit HDAC activity and thereby deacetylate

histone H4, and in this way also repress the expression of

IL-1b -stimulated granulocyte-macrophage

colony-stimula-ting factor [54] One may speculate that a specific pattern of

modified histone tails is required to recruit the basal

transcription machinery, and that TSA can distort this

pattern and thus reduce the hormone-induced

transcrip-tional response [41,52] To understand these events, it will be

essential to map the detailed pattern of histone

modifica-tions that occurs in the MMTV promoter during

transcrip-tion activatranscrip-tion, as has recently been done in the PHO5

promoter [27] This work is in progress in our laboratory

HDAC inhibitors are exciting and promising anticancer

drugs [55], not only for their ability to inhibit histone

deacetylases but also due to their strong potency to induce

growth arrest, to promote differentiation and to induce

apoptosis It is believed that they exert their effects via

up-regulation of gene expression [55] However, our results

and the results of others [43,48] suggest that the

down-regulation of viral tumor promoters may be equally

important for the clinical effects HDAC inhibitors, and

thus also for their possible future use as pharmaceuticals

Acknowledgements

We are grateful to Ulla Bjo¨rk for skilful technical assistance and

Dr Birgitta Gelius for skilful nuclear dissections and for performing the

GR localization experiment We thank Dr Jiemin Wong for kindly

sharing the ChIP protocol for Xenopus oocytes, and Dr Ola

Hermanson for providing the antibody against AcH3-K9 This

work was supported by the Swedish Cancer Foundation (project

2222-BOZ-18XBC) and the Royal Swedish Academy of Sciences (12682) Project support was also provided by the European Commis-sion, TMR, to O¨ W (Network Contract ERBFMRXCT-98–0191).

References

1 van Holde, K.E (1988) Chromatin, Springer Verlag, New-York.

2 Wolffe, A.P (2001) Transcriptional regulation in the context of chromatin structure Essays Biochem 37, 45–57.

3 Beato, M (1996) Chromatin structure and the regulation of gene expression: remodeling at the MMTV promoter J.Mol.Med.74, 711–724.

4 Richard-Foy, H & Hager, G.L (1987) Sequence-specific posi-tioning of nucleosomes over the steroid-inducible MMTV pro-moter EMBO J 6, 2321–2328.

5 Fragoso, G., John, S., Roberts, M.S & Hager, G.L (1995) Nucleosome positioning on the MMTV LTR results from the frequency-biased occupancy of multiple frames Genes Dev 9, 1933–1947.

6 Belikov, S., Gelius, B., Almouzni, G & Wrange, O¨ (2000) Hor-mone activation induces nucleosome positioning in vivo EMBO J.

19, 1023–1033.

7 Zaret, K.S & Yamamoto, K.R (1984) Reversible and persistent changes in chromatin structure accompany activation of a gluco-corticoid-dependent enhancer element Cell 38, 29–38.

8 Truss, M., Bartsch, J., Schulbert, A., Hache, R.J.G & Beato, M (1995) Hormone induces binding of receptors and transcription factors to a rearranged nucleosome on the MMTV promoter

in vivo EMBO J 14, 1737–1751.

9 Almouzni, G., Khochbin, S., Dimitrov, S & Wolffe, A.P (1994) Histone acetylation influences both gene expression and develop-ment of Xenopus laevis Dev.Biol.165, 654–669.

10 Arents, G., Burlingame, R.W., Wang, B.-C., Love, W.E & Moudrianakis, E.N (1991) The nucleosomal core histone octamer

at 3.1 A˚ resolution: a tripartite protein assembly and a left-handed superhelix Proc.Natl Acad.Sci.USA 88, 10148–10152.

11 Luger, K., Mader, A.W., Richmond, R.K., Sargent, D.F & Richmond, T.J (1997) Crystal structure of the nucleosome core particle at 2.8 A˚ resolution Nature 389, 251–260.

12 Neely, K.E & Workman, J.L (2002) Histone acetylation and chromatin remodeling: which comes first? Mol.Genet.Metab.76, 1–5.

13 Ling, X., Harkness, T.A., Schultz, M.C., Fisher-Adams, G & Grunstein, M (1996) Yeast histone H3 and H4 amino termini are important for nucleosome assembly in vivo and in vitro: redundant and position-independent functions in assembly but not in gene regulation Genes Dev 10, 686–699.

14 Jenuwein, T & Allis, C.D (2001) Translating the histone code Science 293, 1074–1080.

15 Yoshida, M & Horinouchi, S (1999) Trichostatin and lepto-mycin Inhibition of histone deacetylation and signal-dependent nuclear export Ann.NY Acad.Sci.886, 23–36.

16 Wong, J., Patterton, D., Imhof, A., Gushin, D., Shi, Y.-B & Wolffe, A.P (1998) Distinct requirements for chromatin assembly

in transcriptional repression by thyroid hormone receptor and histone deacetylase EMBO J 17, 520–534.

17 Murphy, M., Ahn, J., Walker, K.K., Hoff man, W.H., Evans, R.M., Levine, A.J & George, D.L (1999) Transcriptional repression by wild-type p53 utilizes histone deacetylases, mediated

by interaction with msin3a Genes Dev 13, 2490–2501.

18 Maldonado, E., Hampsey, M & Reinberg, D (1999) Repression: targeting the heart of the matter Cell 99, 455–458.

19 Dimitrov, S., Almouzni, G., Dasso, M & Wolffe, A.P (1993) Chromatin transitions during early Xenopus embryogenesis: changes in histone H4 acetylation and in linker histone type Dev Biol.(Orlando).160, 214–227.

20 Belikov, S., Gelius, B & Wrange, O (2001) Hormone-induced

nucleosome positioning in the MMTV promoter is reversible.

EMBO J 20, 2802–2811.

21 Gelius, B., Wade, P., Wolffe, A.P., Wrange, O¨ & O¨stlund

Farrants, A.-K (1999) Characterization of a chromatin

remodel-ing activity in Xenopus oocytes Eur.J.Biochem.262, 426–434.

22 Li, Q., Herrler, M., Landsberger, N., Kaludov, N., Ogryzko, V.V.,

Nakatani, Y & Wolffe, A.P (1998) Xenopus NF-Y pre-sets

chromatin to potentiate p300 and acetylation-responsive

tran-scription from the Xenopus hsp 70 promoter in vivo EMBO J 17,

6300–6315.

23 Clark, J.D & Wolffe, A.P (1991) Superhelical stress and

nucleo-some-mediated repression of 5S RNA transcription in vitro.

EMBO J 10, 3419–3428.

24 Sambrook, J., Fritsch, E.F & Maniatis, T (1989) Molecular

Cloning.A Laboratory Manual, 2nd edn Cold Spring Harbor

Laboratory Press, NY.

25 Li, J., Lin, Q., Wang, W., Wade, P & Wong, J (2002) Specific

targeting and constitutive association of histone deacetylase

complexes during transcriptional repression Genes Dev 16, 687–

692.

26 Bartsch, J., Truss, M., Bode, J & Beato, M (1996) Moderate

increase in histone acetylation activates the mouse mammary

tumor virus promoter and remodels its nucleosome structure.

Proc.Natl Acad.Sci.USA 93, 10741–10746.

27 Reinke, H & Horz, W (2003) Histones are first hyperacetylated

and then lose contact with the activated PHO5 promoter Mol.

Cell 11, 1599–1607.

28 Archer, T.K., Lefebvre, P., Wolford, R.G & Hager, G.L (1992)

Transcription factor loading on the MMTV promoter: a bimodal

mechanism for promoter activation Science 255, 1573–1576.

29 Norton, V.G., Imai, B.S., Yau, P & Bradbury, E.M (1989)

Histone acetylation reduces nucleosome core particle linking

number change Cell 57, 449–457.

30 Cartwright, I.L., Hertzberg, R.P., Dervan, P.B & Elgin, S.C.

(1983) Cleavage of chromatin with methidiumpropyl-EDTA iron

(II) Proc.Natl Acad.Sci.USA 80, 3213–3217.

31 Maxam, A.M & Gilbert, W (1977) A new method for sequencing

DNA Proc.Natl Acad.Sci.USA 74, 560–564.

32 Sobel, R.E., Cook, R.G & Allis, C.D (1994) Non-random

acetylation of histone H4 by a cytoplasmic histone

acetyl-transferase as determined by novel methodology J.Biol.Chem.

269, 18576–18582.

33 Lee, D.Y., Hayes, J.J., Pruss, D & Wolffe, A.P (1993) A positive

role for histone acetylation in transcription factor access to

nucleosomal DNA Cell 72, 73–84.

34 Vitolo, J.M., Thiriet, C & Hayes, J.J (2000) The H3–H4

N-ter-minal tail domains are the primary mediators of transcription

factor IIIA access to 5S DNA within a nucleosome Mol.Cell Biol.

20, 2167–2175.

35 Vettese-Dadey, M., Grant, P.A., Hebbes, T.R., Crane-Robinson,

C., Allis, C.D & Workman, J.L (1996) Acetylation of histone H4

plays a primary role in enhancing transcription factor binding to

nucleosomal DNA in vitro EMBO J 15, 2508–2518.

36 Thomsen, B., Bendixen, C & Westergaard, O (1991) Histone

hyperacetylation is accompanied by changes in DNA topology

in vivo Eur J.Biochem.201, 107–111.

37 Lutter, L.C., Judis, L & Paretti, R.F (1992) Effects of histone

acetylation on chromatin topology in vivo Mol.Cell Biol.12,

5004–5014.

38 Ebralidse, K.K., Hebbes, T.R., Clayton, A.L., Thorne, A.W &

Crane-Robinson, C (1993) Nucleosomal structure at

hyper-acetylated loci probed in nuclei by DNA-histone crosslinking.

Nucleic Acids Res 21, 4734–4738.

39 Zheng, C & Hayes, J.J (2003) Structures and interactions of the core histone tail domains Biopolymers 68, 539–546.

40 Van Lint, C., Emiliani, S., Ott, M & Verdin, E (1996) Tran-scriptional activation and chromatin remodeling of the HIV-1 promoter in response to histone acetylation EMBO J 15, 1112– 1120.

41 Agalioti, T., Chen, G & Thanos, D (2002) Deciphering the transcriptional histone acetylation code for a human gene Cell

111, 381–392.

42 Sheldon, L.A., Becker, M & Smith, C.L (2001) Steroid hormone receptor-mediated histone deacetylation and transcription at the mouse mammary tumor virus promoter J.Biol.Chem.276, 32423–32426.

43 Wilson, M.A., Ricci, A.R., Deroo, B.J & Archer, T.K (2002) The histone deacetylase inhibitor trichostatin A blocks progesterone receptor-mediated transactivation of the mouse mammary tumor virus promoter in vivo J.Biol.Chem.277, 15171–15181.

44 Mulholland, N.M., Soeth, E & Smith, C.L (2003) Inhibition of MMTV transcription by HDAC inhibitors occurs independent of changes in chromatin remodeling and increased histone acetyla-tion Oncogene 22, 4807–4818.

45 List, H.J., Smith, C.L., Rodriguez, O., Danielsen, M & Riegel, A.T (1999) Inhibition of histone deacetylation augments dihydro-testosterone induction of androgen receptor levels: an explanation for trichostatin A effects on androgen-induced chromatin remodeling and transcription of the mouse mammary tumor virus promoter Exp.Cell Res.252, 471–478.

46 Gunjan, A & Brown, D.T (1999) Overproduction of histone H1 variants in vivo increases basal and induced activity of the mouse mammary tumor virus promoter Nucleic Acids Res 27, 3355– 3363.

47 Bresnick, E.H., John, S., Berard, D.S., LeFebvre, P & Hager, G.L (1990) Glucocorticoid receptor-dependent disruption of a specific nucleosome on the mouse mammary tumor virus pro-moter is prevented by sodium butyrate Proc.Natl Acad.Sci.USA

87, 3977–3981.

48 Lambert, J.R & Nordeen, S.K (1998) Steroid-selective initiation

of chromatin remodeling and transcriptional activation of the mouse mammary tumor virus promoter is controlled by the site of promoter integration J.Biol.Chem.273, 32708–32714.

49 Bourachot, B., Yaniv, M & Muchardt, C (2003) Growth inhibition by the mammalian SWI-SNF subunit Brm is regulated

by acetylation EMBO J 22, 6505–6515.

50 Cress, W.D & Seto, E (2000) Histone deacetylases, transcrip-tional control, and cancer, J.Cell Physiol.184, 1–16.

51 Bernstein, B.E., Tong, J.K & Schreiber, S.L (2000) Genomewide studies of histone deacetylase function in yeast Proc.Natl Acad Sci.USA 97, 13708–13713.

52 Rascle, A., Johnston, J.A & Amati, B (2003) Deacetylase activity

is required for recruitment of the basal transcription machinery and transactivation by STAT5 Mol.Cell Biol.23, 4162–4173.

53 Xu, M., Nie, L., Kim, S.H & Sun, X.H (2003) STAT5-induced Id-1 transcription involves recruitment of HDAC1 and deacetyl-ation of C/EBPbeta EMBO J 22, 893–904.

54 Ito, K., Barnes, P.J & Adcock, I.M (2000) Glucocorticoid receptor recruitment of histone deacetylase 2 inhibits interleukin-1beta-induced histone H4 acetylation on lysines 8 and 12 Mol Cell Biol 20, 6891–6903.

55 Pandolfi, P.P (2001) Histone deacetylases and transcriptional therapy with their inhibitors Cancer Chemother.Pharmacol.48, S17–S19.

56 Perlmann, T & Wrange, O¨ (1988) Specific glucocorticoid receptor binding to DNA reconstituted in a nucleosome EMBO J 7, 3073– 3079.