Content uploaded by Anick Auger
Author content
All content in this area was uploaded by Anick Auger on Jan 27, 2016
Content may be subject to copyright.
Opposite Effects of Histone Deacetylase Inhibitors on
Glucocorticoid and Estrogen Signaling in Human Endometrial
Ishikawa Cells
Walter Rocha, Rocio Sanchez, Julie Desch ˆ
enes, Anick Auger, Elise H ´
ebert, John H. White,
and Sylvie Mader
Department of Biochemistry (W.R., R.S., J.D., A.A., E.H., S.M.) and Institute of Research in Immunology and in Cancer (W.R.,
J.D., E.H., S.M.), Universite´ de Montre´ al, Montre´al, Canada; Montre´ al Center for Experimental Therapeutics in Cancer, Jewish
General Hospital, Montre´ al, Canada (W.R., J.D., E.H., J.H.W., S.M.); and Departments of Physiology (J.H.W.) and Medicine
(J.H.W., S.M.), McGill University, Montre´ al, Canada
Received May 3, 2005; accepted September 23, 2005
ABSTRACT
Histone deacetylase inhibitors (HDACi), which have emerged as
a new class of anticancer agents, act by modulating expression
of genes controlling apoptosis or cell proliferation. Here, we
compared the effect of HDACi on transcriptional activation by
estrogen or glucocorticoid receptors (ER and GR, respectively),
two members of the steroid receptor family with cell growth
regulatory properties. Like other transcription factors, steroid
receptors modulate histone acetylation on target promoters.
Using episomal reporter vectors containing minimal promoters
to avoid promoter-specific effects, we observed that long-term
(24-h) incubation with HDACi strongly stimulated GR-depen-
dent but markedly repressed ER-dependent signaling in ER⫹/
GR⫹human endometrial carcinoma Ishikawa cells. These ef-
fects were reproduced on endogenous target genes and
required incubation periods with HDACi substantially longer
than necessary to increase global histone acetylation. Repres-
sion of estrogen signaling was due to direct inhibition of tran-
scription from multiple ER
␣
promoters and correlated with de-
creased histone acetylation of these promoters. In contrast, the
strong HDACi stimulation of GR-dependent gene regulation
was not accounted for by increased GR expression, but it was
mimicked by overexpression of the histone acetyltransferase
complex component transcriptional intermediary factor 2. To-
gether, our results demonstrate striking and opposite effects of
HDACi on ER and GR signaling that involve regulatory events
independent of histone hyperacetylation on receptor target
promoters.
Incorporation of DNA into chromatin plays a major role in
regulating gene expression. Decondensed chromatin (euchro-
matin) is associated with transcriptional activity, whereas
condensed heterochromatin is transcriptionally inactive. N-
terminal tails of histones, which are subject to several post-
translational modifications (Jenuwein and Allis, 2001), play
an important role in regulation of chromatin structure. Pos-
itively charged histone tails of nucleosomes interact with
DNA, other histones, and chromatin components. Acetyla-
tion of lysines by histone acetyl-transferases (HATs) neutral-
izes their positive charges, destabilizes nucleosomes, and can
enhance or block other types of modifications, resulting in
differential binding of many different chromatin proteins
(Jenuwein and Allis, 2001).
Transcriptional activators recruit cofactors, including
HATs. For example, nuclear receptors exhibit hormone-de-
pendent recruitment of HAT complexes composed of p160
(SRC-1/TIF2/AIB1), CBP/p300, and pCAF families of coacti-
This study was supported by operating grants from the Canadian Institutes
for Health Research (CIHR) to S.M. (MT-13147 and IC1–70246) and J.H.W.
(MT-11704). S.M. holds the Canadian Imperial Bank of Commerce Breast
Cancer Research Chair at Universite´ de Montre´ al and is a Senior Scholar of
the Fonds de la Recherche en Sante´ du Que´ bec (FRSQ). W.R. was supported by
fellowships from the Montreal Center for Experimental Therapeutics in Can-
cer-CIHR training program and from the Faculte´ des Etudes Supe´ rieures de
l’Universite´ de Montre´al (FES). J.D. was supported by fellowships from the
FRSQ and FES.
Article, publication date, and citation information can be found at
http://molpharm.aspetjournals.org.
doi:10.1124/mol.105.014514.
ABBREVIATIONS: HAT, histone acetyltransferase; HDAC, histone deacetylase; HDACi, histone deacetylase inhibitor(s); MMTV, mouse mammary
tumor virus; ER, estrogen receptor; GR, glucocorticoid receptor; E2, 17

-estradiol; OHT, 4-hydroxytamoxifen; Dex, dexamethasone; SB, sodium
butyrate; Act-D, actinomycin D; wt. wild-type; CAT, chloramphenicol acetyltransferase; EBV, Epstein Barr virus; ERE, estrogen response element;
GRE, glucocorticoid response element; PBS, phosphate-buffered saline; RT-PCR, reverse transcription-polymerase chain reaction; TIF2, tran-
scriptional intermediary factor 2; TAT, tyrosine aminotransferase; CMV, cytomegalovirus; GAPDH, glyceraldehyde-3-phosphate dehydrogenase;
ICI182,780, faslodex.
0026-895X/05/6806-1852–1862$20.00
MOLECULAR PHARMACOLOGY Vol. 68, No. 6
Copyright © 2005 The American Society for Pharmacology and Experimental Therapeutics 14514/3067617
Mol Pharmacol 68:1852–1862, 2005 Printed in U.S.A.
1852
at ASPET Journals on November 3, 2015molpharm.aspetjournals.orgDownloaded from
vators (Rosenfeld and Glass, 2001). Histone acetylation is
remarkably dynamic on hormone-regulated promoters, be-
cause recruitment of HAT complexes alternates with that of
histone deacetylases (HDACs) on the estrogen target pro-
moter pS2 (Metivier et al., 2003). Nuclear receptor corepres-
sors such as N-CoR and SMRT (Rosenfeld and Glass, 2001) or
NRIP1/RIP140 and LCoR (White et al., 2004) recruit HDACs
in the absence or presence of hormone, respectively. In addi-
tion, HATs and probably also HDACs are active with non-
histone protein substrates, including E2F, pRb, and p53
(McLaughlin and La Thangue, 2004).
HDACi have emerged as a new class of anticancer agents
for treatment of both solid and hematological tumors
(McLaughlin and La Thangue, 2004). The naturally occur-
ring antifungal antibiotic trichostatin A has been invaluable
in validating HDACs as potential anticancer targets. Struc-
turally related inhibitors, including SAHA, PXD101, and
LAQ-824, are currently in clinical trials (Kelly et al., 2003).
Aliphatic acids valproate and butyrate function as less potent
HDACi (McLaughlin and La Thangue, 2004). HDACi induce
apoptosis or differentiation depending on the cell type
(McLaughlin et al., 2003) and, notably, block proliferation of
breast, endometrial, and ovarian cancer cells (Munster et al.,
2001; Strait et al., 2002; Takai et al., 2004). Different HDACi
alter transcription of a common set of genes that control
pathways important for cell survival and proliferation (Gla-
ser et al., 2003; Peart et al., 2005). It is noteworthy that both
enhancement and repression of gene expression were ob-
served in these studies, suggesting more complex mecha-
nisms of action than enhancement of histone acetylation.
HDACi influence steroid receptor gene regulation in a cell-,
promoter- and receptor-dependent manner. HDACi pre-
vented activation of transiently transfected, episomal, or
chromosomal MMTV promoters by glucocorticoids (Mulhol-
land et al., 2003; Kinyamu and Archer, 2004). Although
sodium butyrate inhibited glucocorticoid induction of the ty-
rosine aminotransferase gene in rat HTC cells (Plesko et al.,
1983), it enhanced glucocorticoid induction of alkaline phos-
phatase in HeLa S3 cells (Littlefield and Cidlowski, 1984).
Finally, trichostatin A induced estrogen-dependent tran-
scription in MCF-7 cells (Ruh et al., 1999) and in stably
transfected HepG2 cells (Mao and Shapiro, 2000).
Some of the effects of HDACi on estrogen target genes
seem to be mediated by modulation of estrogen receptor (ER)
expression. Inhibition of ER
␣
expression by HDAC1 in
MCF-7 breast cancer cells was reversed by trichostatin A
(Kawai et al., 2003). Trichostatin A induced ER
␣
expression
in ER-negative breast cancer cells (Yang et al., 2001),
whereas another study found that trichostatin A induced
ER

but not ER
␣
expression in MDA-MB-231 cells (Jang et
al., 2004). Finally, valproic acid induced ER
␣
expression in
endometrial carcinoma Ishikawa and in MCF-7 cells (Grazi-
ani et al., 2003). Conversely, others reported inhibition of
ER
␣
expression by HDACi, which may explain the increased
sensitivity of ER⫹breast cancer cell lines to HDACi (Alao et
al., 2004; Margueron et al., 2004b; Reid et al., 2005). Finally,
HDACi may induce hyperacetylation of nuclear receptors by
associated HAT complexes, altering their function. Indeed,
acetylation of ER
␣
modulated sensitivity to hormone (Fu et
al., 2004).
Variations in cell lines and/or target promoters, which can
be regulated by steroid receptors through different mecha-
nisms (Sanchez et al., 2002), probably account for the vari-
ability in the reported effects of HDACi on steroid-mediated
transcription. Here, we compared the effects of HDACi on
ER
␣
and glucocorticoid receptor (GR)-dependent transcrip-
tion on reporter vectors containing minimal estrogen- or glu-
cocorticoid-responsive promoters propagated as episomes in
human endometrial carcinoma Ishikawa cells, which express
both receptors. Using this system, modulation by HDACi of
receptor-dependent transcription can be monitored in the
absence of a confounding influence of other transcription
factors or of variable sites of chromosomal integration. Our
results indicate striking and opposite effects of HDACi on
estrogen and glucocorticoid signaling, leading us to explore
the mechanisms underlying this differential regulation of
two closely related steroid receptors in Ishikawa cells.
Materials and Methods
Plasmids and Reagents. 17

-Estradiol (E2), 4-hydroxytamox-
ifen (OHT), dexamethasone (Dex), sodium butyrate (SB), cyclohexi-
mide, anisomycin, puromycin, and actinomycin D (Act-D) were pur-
chased from Sigma Diagnostics (Oakville, ON, Canada), ICI182,780
(faslodex) was purchased from Tocris Cookson Inc. (Ellisville, MO),
and trichostatin A was procured from Wako Pure Chemicals (Osaka,
Japan). pSG5-hER
␣
and pSG5-TIF2.1 were kind gifts from Prof.
Pierre Chambon (Institut de Ge´ne´tique et de Biologie Moleculaire
et Cellulaire, Illkirch, France). pCDNA3.1-ER
␣
and pCDNA3.1-
ER
␣
(K302A/K303A) were constructed as follows. cDNAs for the wt
ER
␣
cDNA and the ER
␣
(K302A/K303A) mutant were released from
pSG5-hER
␣
and pCI-neo-ER
␣
(K302A/K303A) (a kind gift from Dr.
Richard G. Pestell, Georgetown University School of Medicine,
Washington, DC), respectively, by EcoRI digest (MBI Fermentas,
Burlington, ON, Canada), and ligated into the EcoRI site of
pCDNA3.1 (Invitrogen Burlington, ON, Canada). Reporter vectors
GRE5-TATA-CAT/EBV, ERE3-TATA-CAT/EBV, and ERE3-TATA-
LUC have been described previously (Barsalou et al., 2002; Fer-
nandes et al., 2003).
Cell Lines and Reporter Assays. MCF-7 breast carcinoma and
endometrial carcinoma Ishikawa cells were maintained in
␣
-minimal
Eagle’s medium (Wisent, St-Bruno, QC, Canada) supplemented with
10 and 5% fetal bovine serum, respectively (Sigma Diagnostics)
supplemented with 1% penicillin/streptomycin (Wisent). Stable re-
porter cell lines Ishikawa-GRE5/EBV and Ishikawa-ERE3/EBV
(Barsalou et al., 2002) were maintained in the same medium as the
parental cells supplemented with 50
g/ml hygromycin B.
Three days before experiments, Ishikawa cells were switched to
phenol red-free Dulbecco’s modified Eagle’s medium containing 5%
charcoal-stripped serum, 1% sodium pyruvate (Wisent), 1% penicil-
lin/streptomycin, and 1% L-glutamine (Wisent). For CAT assays,
cells were stimulated 24 h after seeding with 25 nM E2 or 25 nM Dex
and either vehicle (ethanol), trichostatin A, or sodium butyrate (vari-
able concentrations) for another 24 h. Whole cell extracts were pre-
pared in 0.25 M Tris-HCl, pH 7.5, by three cycles of freeze-thawing
and were standardized for protein amount. CAT assays were per-
formed as described previously (Barsalou et al., 2002). Each assay
included triplicates for each condition and was repeated at least
three times. A typical experiment is shown.
For luciferase assays, Ishikawa cells were transfected with the
calcium-phosphate method (Barsalou et al., 2002) in six-well plates
(2 ⫻10
6
cells/well). Typically, a DNA mix contained 150 ng of
expression vector, 350 ng of ERE3-TATA-Luc reporter vector, and 2
g of pBlueScript as carrier; after 24 h, cells were washed with fresh
medium and stimulated for another 24 h with 25 nM E2 and/or 300
nM trichostatin A or vehicle (ethanol). Cells were washed two times
with 1⫻PBS and harvested in lysis buffer (100 mM Tris-HCl, pH
7.9, 0.5% Nonidet P-40, and 1 mM dithiothreitol). Luciferase activity
Effects of HDAC Inhibitors on Steroid Receptor Signaling 1853
at ASPET Journals on November 3, 2015molpharm.aspetjournals.orgDownloaded from
was measured in the presence of luciferin with a Fusion universal
microplate analyser (PerkinElmer Life and Analytical Sciences,
Woodbridge, ON, Canada). Each transfection was carried out in
triplicate and repeated at least three times. Proteins were quantified
by BioRad reagent (Bio-Rad, Mississauga, ON, Canada).
Alkaline Phosphatase Assays. Alkaline phosphatase assays
were conducted as described previously (Barsalou et al., 2002).
Treatments were performed in triplicates for 24 h, after which cells
were washed in PBS twice, frozen at ⫺80°C for 15 min, and incu-
bated with 50
l of reaction buffer (5 mM p-nitrophenyl phosphate,
0.24 mM MgCl
2
, and 1 M diethanolamine, pH 9.8). Plates were
incubated at room temperature until production of a yellow color,
and levels of p-nitrophenyl were quantified by measuring absorption
at 410 nm.
RNA Extraction and RT-PCR Assays. Ishikawa cells were
seeded in six-well plates (2.5 ⫻10
5
cells/well) and treated with
300 nM trichostatin A or 5 mM sodium butyrate, with or without
25 nM E2 or 25 nM Dex for different times (as indicated in the
figure legends). For treatments with 2
g/ml actinomycin D, 10
g/ml cycloheximide, 5
M anisomycin, or 5
M puromycin, incu-
bation was initiated 1 h before HDACi addition and continued for
6 h thereafter. The medium was then removed, and total RNA was
extracted in 1 ml of TRIzol reagent (Invitrogen) and quantified by
UV absorption. RNAs (2
g) were reverse transcribed using the
RevertAid H first minus strand cDNA synthesis kit (MBI Fermen-
tas) as recommended by the manufacturer. Sequences of oligonu-
cleotides used for polymerase chain reaction amplification are
available upon request. Primers used for alternative ER
␣
5⬘exons
were designed according to published GenBank references (Kos et
al., 2001). Polymerase chain reaction was performed using TAQ
polymerase (MBI Fermentas). Amplified cDNA fragments were
resolved on 2% agarose gels and stained with ethidium bromide.
Each assay was reproduced at least three times. A typical exper-
iment is shown.
Fig. 1. HDAC inhibitors trichostatin A and sodium butyrate have opposite effects on GR and ER transcriptional activity with episomal reporter vectors
containing minimal steroid-responsive promoters. Ishikawa-GRE5/EBV (A and C) or Ishikawa-ERE3/EBV (B and D) cells, which propagate the
GRE
5
-TATA-CAT/EBV or ERE
3
-TATA-CAT/EBV episomal reporter vectors, respectively, were treated for 24 h with or without 25 nM Dex (A and C)
or 25 nM E2 (B and D) and either SB (A and B) or TSA (C and D) at the indicated concentrations. CAT activity was assayed in whole cell extracts and
normalized on protein concentration.
1854 Rocha et al.
at ASPET Journals on November 3, 2015molpharm.aspetjournals.orgDownloaded from
Western Analysis. Ishikawa cells were treated with 25 nM E2,
25 nM Dex, 100 nM OHT, 100 nM ICI182,780, or vehicle for 24 h
with or without HDACi (300 nM trichostatin A or 5 mM sodium
butyrate. Cells were harvested in ice-cold PBS, and whole cell ex-
tracts were prepared by three freeze-thaw cycles in high salt buffer
(25 mM Tris-HCl, pH 7.4, 0.1 mM EDTA, pH 8.0, 400 mM NaCl, 10%
glycerol, 1 mM dithiothreitol; 1 mM phenylmethylsulfonyl fluoride,
and protease inhibitors). After electrophoresis on an SDS-polyacryl-
amide gel (7.5% acrylamide), proteins were transferred onto polyvi-
nylidene difluoride membranes (Hybond P; GE Healthcare, Little
Chalfont, Buckinghamshire, UK). Blots were incubated with anti-
ER
␣
mouse monoclonal or anti-TIF2 mouse monoclonal antibodies
(B10 and 3Ti-3F1, respectively; both kind gifts from Prof. P. Cham-
bon), anti-GR rabbit polyclonal antibody (PA1–511; ABR Affinity
BioReagents, Golden, CO), anti-acetylated-H3, anti-acetylated-H4
(Upstate Biotechnology, Lake Placid, NY), or anti-

-actin mouse
monoclonal antibody (Sigma Diagnostics). Immunodetection was
performed using enhanced chemiluminescence (PerkinElmer Life
and Analytical Sciences, Boston, MA) as recommended by the man-
ufacturer. Each result was reproduced at least three times. A typical
experiment is shown.
Chromatin Immunoprecipitation Assays. Ishikawa cells were
treated with 1.5% formaldehyde for 10 min at room temperature and
fragmented by sonication as reported previously (Bourdeau et al.,
2004), yielding fragments of approximately 350 base pairs, average
size. Antibodies against acetylated H3 and acetylated H4 were pur-
chased from Upstate Biotechnology. The sequences of the primers
used in chromatin immunoprecipitation assays are available upon
request. Chromatin immunoprecipitation experiments were per-
formed twice with similar results. A representative set of results is
shown.
Results
To investigate the effect of HDACi on steroid receptor-
mediated transcription, we used stably transfected Ishikawa
cell lines carrying Epstein Barr virus episomal reporter vec-
tors sensitive to either glucocorticoids or estrogens. The re-
porter vectors contain a CAT reporter gene under control of
minimal promoters composed of a TATA box placed down-
stream of either five glucocorticoid response elements
(GRE5-TATA-CAT/EBV; Mader and White, 1993) or three
estrogen response elements (ERE3-TATA-CAT/EBV; Bar-
salou et al., 2002). The use of minimal promoters and the
absence of integration into the host cell chromosomes enable
monitoring the effects of HDACi on transcriptional activation
by GR or ER without confounding cooperative effects of tran-
scription factors. Surprisingly, in contrast to the reported
repressive effects of HDACi on glucocorticoid stimulation of
the MMTV promoter (Mulholland et al., 2003; Kinyamu and
Archer, 2004; and references therein), a marked and dose-
dependent increase in the GRE5-TATA reporter activity was
observed with increasing concentrations of the HDACi SB
(0.5–5 mM) in the presence of 25 nM Dex but not in its
absence, in the Ishikawa-GRE5/EBV cell line. The maximal
stimulation by sodium butyrate, obtained at the highest con-
centration tested, was ⬃10-fold (Fig. 1A).
The effects of sodium butyrate on reporter expression from
Ishikawa-ERE3/EBV cells were opposite to those observed
from GRE5/EBV in that a dose-dependent decrease in re-
porter activity was noted in the presence of 25 nM E2, reach-
ing more than 4-fold repression at 5 mM (Fig. 1B). The
differential effects observed here with the two reporter cell
lines suggest that sodium butyrate has a differential func-
tional impact on estrogen and glucocorticoid signaling path-
ways rather than a general effect on global transcription, or
on the stability of the CAT enzyme.
To verify that the effects of sodium butyrate on both sig-
naling pathways are related to its HDAC inhibitory proper-
ties, we incubated the two reporter cell populations with
trichostatin A, a structurally unrelated HDACi. Similar to
results described above, glucocorticoid-stimulated reporter
gene expression was markedly enhanced by increasing con-
centrations of trichostatin A (Fig. 1C), and estrogen-induced
expression was repressed at the highest dose assayed, 300
nM (Fig. 1D). Note that the apparent increase in E2-regu-
lated expression at lower trichostatin A concentrations was
not statistically significant. The comparable actions of so-
dium butyrate and trichostatin A on estrogen- and glucocor-
ticoid-driven reporter gene expression suggest that they are
acting through a common mechanism (i.e., the inhibition of
one or several of the HDACs expressed in Ishikawa cells)
(HDACs 1–10; Fig. 2A).
To confirm that HDACi treatment increases global histone
acetylation in the two episomal cell populations, we per-
formed Western analysis of Ishikawa cell nuclear extracts
using antibodies specific for acetylated H3 and H4 (Fig. 2B).
Marked increases in acetylation were observed in both cell
populations at 1 h after treatment using 300 nM trichostatin
A (Fig. 2B). Furthermore, experiments using extracts from
cells harvested at different times after trichostatin A treat-
ment indicate that elevated histone acetylation was detect-
able for at least 16 h after incubation with 300 nM tricho-
statin A, but it was much more transient after treatment
with a 30 nM dose (Fig. 2C). The high levels of trichostatin A
necessary to obtain maximum alteration of dexamethasone-
and estradiol-mediated expression at 24 h (Fig. 1) suggest
that prolonged exposure to trichostatin A is necessary to
induce the observed changes in gene expression.
Time-course experiments of treatment with dexametha-
sone or estradiol indicate that increases in levels of the CAT
Fig. 2. Expression of HDACs and dose- and time-dependent effects of
trichostatin A on histone acetylation in Ishikawa cells. A, expression of
HDAC mRNAs in Ishikawa cells was monitored by RT-PCR from total
RNAs using primers specific for each HDAC. B, TSA treatment (300 nM;
1 h) increases global acetylation levels of histone H3 and H4 in Ishikawa-
ERE3/EBV and Ishikawa-GRE5/EBV cells. Levels of acetyl-H3 (Ac-H3)
and Acetyl-H4 (Ac-H4) were detected by Western analysis as described
under Materials and Methods. C, hyperacetylation of histone H3 by 30
or 300 nM TSA is reversible in a time- and concentration-dependent
manner.
Effects of HDAC Inhibitors on Steroid Receptor Signaling 1855
at ASPET Journals on November 3, 2015molpharm.aspetjournals.orgDownloaded from
enzyme were gradual, being detectable at 6 to 8 h and rising
through 24 h (Fig. 3, A and B). Trichostatin A (300 nM) had
little effect at8honreporter gene expression, whereas its
effects became pronounced at 24 h. Trichostatin A only af-
fected minimally dexamethasone- or estradiol-dependent ex-
pression if added during the last8hofa24-h exposure to
either hormone (Fig. 3, C and D). Finally, addition of an 18-h
pretreatment with trichostatin A before treatment with
dexamethasone and trichostatin A boosted the stimulatory
effect of trichostatin A on GR-dependent expression (from 20-
to 30-fold; Fig. 3E) and its repressive effect on ER-dependent
expression (from 2.5- to 10-fold; Fig. 3F). Together, these
results indicate that the effects of trichostatin A on steroid-
induced expression from our minimal promoters require
higher concentrations and are much slower than its effects on
global histone acetylation levels.
To verify that HDACi have global effects on ER- and GR-
mediated pathways as suggested by experiments using min-
imal reporter vectors, we examined the effect of HDACi on
expression of endogenous estrogen and glucocorticoid target
genes. The ALPPL2 alkaline phosphatase gene is strongly
induced by estrogen at the transcriptional level in Ishikawa
cells, and to a lesser extent by the partial antiestrogen OHT,
but not by the full antiestrogen ICI182,780 (Fig. 4A). Alka-
line phosphatase activity is also markedly induced by estro-
gen at 24 h (Fig. 4B), whereas induction by 4-hydroxytamox-
Fig. 3. Long-term coincubation
with trichostatin A is necessary
for effects on GR- and ER-de-
pendent expression in Ishikawa
reporter cell lines. Ishikawa-
GRE5/EBV (A, C, and E) or Ish-
ikawa-ERE3/EBV (B, D, and F)
cells were treated for different
times with or without 25 nM
Dex (A, C, and E) or 25 nM E2
(B, D, and F) and TSA at the
indicated concentrations and
times. A time course of cotreat-
ment with hormones and tri-
chostatin A indicates that a
long-term incubation (24 h) is
required for marked effects (A
and B). Addition of trichostatin
A during the last8hofhormone
treatment (16 to 24 h after hor-
mone addition) does not lead to
significant effects (C and D). In
contrast, pretreatment with tri-
chostatin A for 18 h (⫹) before
hormone addition further in-
creased the magnitude of these
effects (E and F).
1856 Rocha et al.
at ASPET Journals on November 3, 2015molpharm.aspetjournals.orgDownloaded from
ifen is detectable only at later times. Treatment with
trichostatin A had a slight stimulatory effect on basal levels
of ALPPL2 activity, an effect that was independent of ER
function because it was not repressed by treatment with the
full antiestrogen ICI182,780. However, the stimulatory ef-
fects of estrogen on transcription of ALPPL2 (Fig. 4A, ar-
rows) and on alkaline phosphatase activity (Fig. 4B) were
both lost upon HDACi treatment. The weak stimulation of
alkaline phosphatase activity by 4-hydroxytamoxifen in the
presence of trichostatin A, although consistently observed in
three experiments, was not statistically significant in a Stu-
dent’s ttest analysis.
The human tyrosine aminotransferase (TAT) gene is a
strongly induced glucocorticoid target gene in fetal liver (Na-
gao et al., 1987). Dexamethasone treatment was found to
stimulate the expression of TAT transcripts in Ishikawa cells
(Fig. 4C). Trichostatin A treatment alone did not affect TAT
expression, but cotreatment with dexamethasone and tricho-
statin A markedly augmented the effect of dexamethasone
alone. Thus, the effects of HDACi on expression of endoge-
nous estrogen and glucocorticoid target genes in parental
Ishikawa cells were similar to those observed with our re-
porter cell lines, supporting the notion that components es-
sential to ER and GR signaling are regulated by HDACi, with
opposite effects on the activities of these two pathways.
HDACi could potentially affect the ER signaling pathway
at several levels. Because the delayed kinetics of HDACi
effects on ER-dependent transcription are compatible with
modulation of receptor expression, we assessed mRNA levels
of ER
␣
and ER

in Ishikawa cells treated with sodium bu-
tyrate or trichostatin A. Although no significant effects were
observed on ER

expression (data not shown), ER
␣
expres-
sion was strongly repressed by 5 mM sodium butyrate and by
300 nM trichostatin A at 16 h, irrespective of ligand treat-
ment (Fig. 5A). At 24 h, receptor levels were returned to
near-untreated levels in the presence of trichostatin A but
not of sodium butyrate (Fig. 5A), consistent with the stronger
repression of estrogen reporter gene expression observed
with sodium butyrate (Fig. 1). Sodium butyrate also re-
pressed ER
␣
protein levels to a greater extent than tricho-
statin A over a 24-h treatment period (Fig. 5B).
If inhibition of ER
␣
expression by HDACi is the main basis
for their repressive effects on ER target genes, then expres-
sion of exogenous ER
␣
should reverse this repression. In-
deed, although estradiol-induced expression from a trans-
fected ERE-TATA-Luc reporter vector was repressed by
trichostatin A in Ishikawa cells, cotransfection of the
pCDNA3.1-ER
␣
expression vector led to a marked synergism
between trichostatin A and estradiol for reporter gene ex-
pression (Fig. 5C). This synergism was due in part to a
stimulatory effect of trichostatin A on ER
␣
expression from
the pCDNA3.1 vector (Fig. 5D) and was colinear with the
concentration of exogenous expression vector cotransfected
(data not shown). Finally, similar results were obtained
when an expression vector for ER
␣
(K302A/K303A) was co-
transfected instead of the vector expressing wild-type ER
␣
.
K302 and K303 are tandem lysine residues that are acety-
lated by p300 (Wang et al., 2001). This suggests that acety-
lation of the receptor does not play a major role in the effects
of HDACi under our experimental conditions (Fig. 5, C and
D). Effects of HDACi on exogenous ER
␣
expression are prob-
ably due to a stimulation of the CMV promoter of the expres-
sion vector, as expression from a CMV-

Gal reporter vector
was also markedly stimulated (data not shown).
Expression of ER
␣
is driven from several promoters that
function in a tissue-specific manner (Kos et al., 2001). In
Ishikawa cells, we detected ER
␣
transcripts expressed from
promoters A, B, and C (Fig. 6A). Expression from promoter F
was detectable only in MCF-7 cells (Fig. 6A). In Ishikawa
cells, levels of transcripts originating from promoters A, B,
and C were reduced by trichostatin A or sodium butyrate,
whereas expression of GAPDH was not affected (Fig. 6B). In
MCF7 cells, trichostatin A also reduced levels of transcripts
originating from promoters A, B, F, and to a lower extent C,
whereas expression of GAPDH was not affected (Fig. 6C).
Western blot analysis confirmed that both the 66-kDa form of
ER
␣
, originating from promoters A, B, and C, and the 46-kDa
form originating from promoter F were less abundant in
MCF7 after treatment with trichostatin A (Fig. 6D).
Trichostatin A could exert its effects through regulation of
transcript stability by a mechanism involving regulatory se-
quences common to all repressed RNA isoforms. Therefore,
we assessed whether repression by HDACi would be ob-
served in the presence of the transcriptional inhibitor Act-D.
Although basal ER
␣
transcript levels were repressed by ac-
tinomycin D treatment at 6 h, as expected, no further repres-
sion by sodium butyrate was observed (Fig. 6E). This sug-
gests that HDACi repress transcription from the ER
␣
promoters rather than mRNA stability. We then investigated
Fig. 4. HDACi repress estradiol-stimulated expression of ALPPL2 and
stimulate induction of the TAT gene by dexamethasone. A and B, Ish-
ikawa cells were treated with 25 nM E2 or with antiestrogens OHT (100
nM) or ICI182,780 (ICI; 100 nM) in the absence or presence of 5 mM SB
or of 300 nM TSA for 24 h. mRNA levels of the ALPPL2 gene were
assessed by RT-PCR (A), and alkaline phosphatase activity was assayed
by p-nitrophenyl hydrolysis (B). C, Ishikawa cells were treated with 25
nM Dex or 300 nM trichostatin A or both for 24 h. mRNA levels of the
human TAT gene and of the control housekeeping GAPDH gene were
monitored by RT-PCR.
Effects of HDAC Inhibitors on Steroid Receptor Signaling 1857
at ASPET Journals on November 3, 2015molpharm.aspetjournals.orgDownloaded from
the levels of acetylated histones H3 and H4 on ER
␣
promot-
ers A, B, and C in Ishikawa cells in the presence or absence
of HDACi. Treatment with trichostatin A for 6 h led to a
reduction in the levels of acetylated H3 or H4 associated with
these promoters (Fig. 6F), despite the large increase in over-
all acetylated histone levels in the cell at this time (Fig. 2C).
These results suggest that these promoters are in a tran-
scriptionally less active state in the presence of HDACi.
Finally, we investigated whether the transcriptional repres-
sion of ER
␣
by HDACi is independent of protein synthesis.
The repressive effects of trichostatin A on promoters A, B,
and C were maintained, although attenuated, in the presence
of protein synthesis inhibitors cycloheximide (10
g/ml), ani-
somycin (5
M), or puromycin (5
M), indicating that de novo
protein synthesis was not required for at least part of the
repressive effect (Fig. 6G). Likewise, repression by sodium
butyrate was also still observed in the presence of cyclohex-
imide (data not shown).
We then examined whether sodium butyrate or trichosta-
tin A increased endogenous GR mRNA levels in Ishikawa
cells, which would provide a mechanism for the observed
stimulation of GR-dependent expression. Trichostatin A did
not alter GR mRNA expression at 8 or 24 h (Fig. 7A) or GR
protein levels at 1, 8, or 24 h (Fig. 7B). Treatment with
sodium butyrate also did not change GR mRNA or protein
levels at 24 h (data not shown). To test whether the effects of
HDACi on glucocorticoid signaling can be mimicked by in-
creased HAT activity, we transiently transfected a truncated
form of the p160 coactivator TIF2/SRC2, TIF2.1, which con-
tains the receptor interaction domain and activation domains
(Voegel et al., 1998). TIF2.1 increased GR-dependent expres-
sion by 10-fold, but it attenuated the effects of HDAC inhib-
itors from ⬃10- to ⬃2-fold (Fig. 8). Thus, increased expres-
sion of TIF2.1, which can recruit HAT activities such as
CBP/p300 and PCAF, has the same effect as global suppres-
sion of HDAC activity. This suggests that a substrate com-
Fig. 5. Repressive effects of HDACi on estrogen signaling are due to repression of ER
␣
expression in parental Ishikawa cells. A and B, ER
␣
expression
is inhibited by HDACi at the mRNA and protein levels. Ishikawa cells were treated with 25 nM E2, 100 nM 4-hydroxytamoxifen, or 100 nM ICI182,780
in the absence or presence of 5 mM SB or of 300 nM trichostatin A for 16 or 24 h. mRNA levels of the human ER
␣
and of the control

-actin gene were
monitored by RT-PCR. Primers for ER
␣
were chosen in the coding region common to all transcripts. C and D, reexpression of ER
␣
by transient
transfection prevents the repressive effects of trichostatin A independently from acetylation of the receptor. Ishikawa cells were transiently
transfected with an ERE3-TATA-Luc reporter vector with expression vectors for wt ER
␣
or for the ER
␣
(K302A/K303A) mutant affected in the
acetylation sites or with the parental pCDNA3.1 expression vector. Cells were treated with 25 nM E2 and/or 300 nM TSA as indicated for 24 h (C).
Western analysis of ER
␣
expression levels was performed in parallel and indicates that trichostatin A increases expression directed by the CMV
promoter in the pCDNA3.1 vector (D).
1858 Rocha et al.
at ASPET Journals on November 3, 2015molpharm.aspetjournals.orgDownloaded from
mon to the type I/II HDACs expressed in Ishikawa cells and
to the HAT activities in the p160-CBP/p300-PCAF complex
stimulates GR signaling in these cells in an acetylation-
dependent manner.
Discussion
In this study, we have used minimal reporter vectors to
assess the overall effects of HDACi on two steroid receptor
genomic pathways. Estrogen and glucocorticoid receptors are
closely related and share similar functional properties, but
they have distinct DNA binding specificities. Synthetic pro-
moters composed of their respective binding sites inserted
upstream of a TATA box thus allow easy monitoring of the
activity of the corresponding signaling pathways, with min-
imal influence from other transcription factors. Our reporter
vectors are propagated as episomes, which are stably main-
tained at moderate copy number in the form of chromatin,
circumventing variations in promoter activity caused by dif-
ferent sites of chromosomal integration (Mader and White,
1993). Because both receptors recruit coactivators with HAT
activity to remodel chromatin at target promoters, it might
be expected that HDACi treatment would enhance both ER-
and GR-mediated transcription. Acetylation of steroid recep-
tors themselves has also been shown to occur in a dynamic
Fig. 6. HDACi decrease ER
␣
transcription from promoters A, B, and C in Ishikawa cells in the absence of de novo translation. A, promoters A, B, and
C drive expression of ER
␣
in Ishikawa cells, as demonstrated by detection of the corresponding transcripts with alternative 5⬘exons. Note that
promoter A and promoter F are more active in MCF-7 cells. B, treatment with 300 nM TSA or 5 mM SB for 6 h represses expression from all active
promoters (A, B, and C) in Ishikawa cells, whereas GAPDH expression is not affected. C, treatment with 300 nM TSA for 6 h represses expression from
all active promoters (A, B, C, and F) in MCF7 cells. D, expression of both the 66- and 46-kDa isoforms of ER
␣
is inhibited by TSA treatment (300 nM;
6 h) in MCF7 cells. E, Act-D treatment (2
g/ml) prevents repression of ER
␣
expression by sodium butyrate (5 mM; 6 h). F, chromatin immunopre-
cipitation experiments indicate that treatment of Ishikawa cells with TSA (300 nM; 6 h) leads to hypoacetylation of histones H3 and H4 on promoters
A, B, and C. G, treatment with translation inhibitors cycloheximide (CHX; 10
g/ml), puromycin (Puro; 5
M), and anisomycin (Aniso; 5
M) does not
prevent repression of ER
␣
transcription by 300 nM trichostatin A (6 h).
Fig. 7. HDACi do not increase glucocorticoid recep-
tor expression in Ishikawa cells. A, treatment of
Ishikawa cells with 5 mM SB or 300 nM TSA does
not lead to increases in GR mRNA in the absence or
presence of 25 nM Dex at 8 or 24 h. Expression of
the GAPDH mRNA is shown as a control. B, GR
protein levels, detected by Western analysis using
the polyclonal rabbit PA1–511 antibody, were not
up-regulated at 1, 8, or 24 h. Expression of

-actin is
shown as a control.
Effects of HDAC Inhibitors on Steroid Receptor Signaling 1859
at ASPET Journals on November 3, 2015molpharm.aspetjournals.orgDownloaded from
manner and may impact their transcriptional activation
properties (Fu et al., 2004).
Remarkably, our results indicate that HDACi had opposite
effects on estrogen and glucocorticoid genomic signaling in
Ishikawa cells. Effects on endogenous target genes were sim-
ilar to those obtained with our reporter vectors. Dose-depen-
dent stimulation of glucocorticoid signaling by HDACi was
unexpected because studies of integrated or episomal MMTV
reporter vectors in different cell lines have reported repres-
sive effects of HDACi on glucocorticoid-mediated transcrip-
tion at the concentrations used in this study. Also unexpected
was the requirement for high doses of HDACi and long incu-
bation periods to observe these effects. Both factors are in
fact intricately linked, because our observations indicate that
trichostatin A has a relatively transient effect on histone
acetylation in Ishikawa cells, which can be prolonged by use
of higher doses of this inhibitor. These requirements suggest
that the observed effects of HDACi may involve long-term
effects on components of the receptor signaling pathways
rather than immediate modulation of target promoter his-
tone acetylation.
Modulatory effects of HDAC inhibitors on ER expression
have been described in the literature, although with variable
end results. Although ER
␣
expression was found to be re-
pressed in breast cancer cells in several studies (Alao et al.,
2004; Margueron et al., 2004a; Reid et al., 2005), induction of
ER
␣
expression has also been reported in breast cancer cells
by HDACi (Keen et al., 2003; Yang et al., 2001) and in
Ishikawa cells by the HDACi valproate (Graziani et al.,
2003). Induction of ER

by trichostatin A was also observed
in MDA-MB-231 cells (Jang et al., 2004). We did not observe
significant effects on ER

expression in Ishikawa cells, but
we detected a strong reduction in ER
␣
transcription. It is
unclear whether the difference between these repressive ef-
fects of trichostatin A or butyrate and the previously reported
induction of ER
␣
by valproate in Ishikawa cells results from
use of different HDACi or from different isolates of the Ish-
ikawa cell line. Note, however, that valproate stimulated
growth of Ishikawa cells (Graziani et al., 2003), whereas
sodium butyrate and trichostatin A inhibited proliferation
under our experimental conditions (data not shown). Further
analysis confirmed that reduction in ER
␣
mRNA levels re-
quires transcription; i.e., mRNA destabilization by HDACi is
not involved. Several alternative promoters control ER
␣
ex-
pression in Ishikawa and in MCF7 cells. The various tran-
script isoforms encode the same 66-kDa protein, except for
transcripts originating from promoter F. In MCF7 cells, 10%
of these transcripts give rise through alternative splicing to a
truncated 46-kDa form (Kos et al., 2001). Interestingly, re-
pression of transcripts originating from all active promoters
was observed both in Ishikawa and in MCF7 cell types. Note
that promoter F is located ⬃115-kilobase upstream of pro-
moter C, indicating either long-range or multiple sites of
transcriptional shut-off. It is unlikely that induced expres-
sion of a repressor is involved, because the effects of HDACi
were also observed in the presence of three different protein
synthesis inhibitors. Our results differ in this respect from
those of Reid et al. (2005), who reported that the repressive
effects of valproate or trichostatin A on ER
␣
expression in
MCF7 cells are abolished by cycloheximide, but are compat-
ible with the lack of effect of cycloheximide on repression of
ER
␣
expression observed with trichostatin A by Alao et al.
(2004). Potential mechanisms may be activation of a tran-
scriptional repressor or loss/repression of a transcriptional
activator by acetylation, both being compatible with the ob-
served decrease in histone acetylation on the repressed pro-
moters. Of note, Reid et al. (2005) reported recruitment of the
methyl binding protein MeCP2 on the ER
␣
A promoter in the
presence of valproate, suggesting induction of promoter
methylation by this HDACi, an event often associated with
decreased histone acetylation.
The strong dose-dependent stimulatory effects of HDACi
on GRE5-TATA-CAT and endogenous TAT gene expression
differ markedly from previously reported results demonstrat-
ing down-regulation of the stimulatory effect of glucocorti-
coids on the MMTV promoter in various cell types (Mulhol-
land et al., 2003; Kinyamu and Archer, 2004) or on the TAT
gene in rat hepatoma cells (Plesko et al., 1983). Our results,
in contrast, are compatible with earlier observations that
sodium butyrate enhances dexamethasone responsiveness of
the alkaline phosphatase gene in HeLa S3 cells (Littlefield
and Cidlowski, 1984). Although the long time course of in-
duction of glucocorticoid reporter vectors in Ishikawa cells
may suggest indirect effects mediated by the altered expres-
sion of a component of the glucocorticoid signaling pathway,
our assays for GR mRNA and protein levels are not consis-
tent with an induction in GR expression. Interestingly, tran-
sient expression of the p160 coactivator derivative TIF2.1,
which is highly expressed and contains all domains of TIF2
required for coactivation of nuclear receptors (Voegel et al.,
1998), mimicked the effect of HDACi. TIF2, like other p160
members, is a component of HAT complexes containing co-
factors CBP/p300 and PCAF (Rosenfeld and Glass, 2001).
Although overexpression of a HAT coactivator is thus a plau-
sible hypothesis, no increases in the mRNA levels of the HAT
Fig. 8. Overexpression of the p160 coactivator derivative TIF2.1 attenu-
ates the trichostatin A stimulation of GR-dependent transcription. A,
Ishikawa cells were transiently transfected with 2
gofGRE
5
-TATA-
CAT/EBV and 2
g of pSG5-TIF2.1 by the calcium phosphate method.
After 18 h, cells were treated with 25 nM Dex in the presence of 30 or 300
nM TSA for 24 h. Expression of the transfected TIF2.1 is confirmed by
Western blot analysis using the mouse monoclonal antibody 3Ti-3F1
(inset).
1860 Rocha et al.
at ASPET Journals on November 3, 2015molpharm.aspetjournals.orgDownloaded from
coactivators of steroid receptors were detected by RT-PCR in
the presence of HDACi (data not shown). It remains possible
that expression of a HAT coactivator may be affected at the
post-transcriptional level or alternatively that HAT/HDAC
activities may affect the expression of a common substrate
that plays an important role in glucocorticoid signaling.
We have also considered two other potential mechanisms
by which HDACi could synergize with glucocorticoids for
GR-mediated transcription. Decreased expression/activity of
an enzyme involved in degradation of glucocorticoids might
in theory explain the observed effects of HDACi on increased
GR activity, but this is unlikely to be the case in our exper-
imental system because dose-response curves of dexametha-
sone stimulation at 24 h did not reveal a shift in the exoge-
nous hormone concentrations required for the response (data
not shown). In addition, RT-PCR amplification of the 11

-
HSD type II enzyme, which is responsible for limiting the
antiproliferating activity of glucocorticoids in breast cancer
cells (Lipka et al., 2004) did not reveal differences in expres-
sion in the absence or presence of HDACi (data not shown).
Another potential mechanism may be effects of HDACi on the
cell cycle, because most HDACi induce a block at the G
1
/S
transition in different cell lines. The GR has been reported to
have differential transcription activity in G
1
and S phases
(permissive) and in G
2
/M phases (nonpermissive) (Hsu and
DeFranco, 1995; King and Cidlowski, 1998). Long-term (3
day) effects of sodium butyrate on GR activation of the alka-
line phosphatase gene in HeLa S3 cells were attributed to
synchronization of the cells in the permissive G
1
phase
(Littlefield and Cidlowski, 1984). Note, however, that a re-
cent study reported that treatment with 300 nM trichostatin
A for 3 days is accompanied by a decrease in the proportion of
cells in both the G
0
/G
1
and S phases and an increase in cells
in G
2
/M (Takai et al., 2004). Thus, effects on the cell cycle
seem unlikely to explain the synergism observed in Ishikawa
cells between glucocorticoids and HDACi at the level of GR
transcription.
Although additional experiments will be needed to further
pinpoint the exact mechanisms of action of HDACi in Ish-
ikawa cells, including an assessment of whether distinct
subsets of the HDAC expressed in Ishikawa cells are involved
in the effects of HDAC on estrogen or glucocorticoid signal-
ing, it is of interest that signaling pathways involving differ-
ent nuclear receptors can be modulated differentially by
HDACi, whose use in cancer treatment seems promising.
Inhibition of ER
␣
expression would be of benefit in the treat-
ment of ER
␣
positive breast tumors if it entails repression of
growth-stimulatory ER
␣
target genes, although the revers-
ible character of this inhibition may require repeated admin-
istration of high doses of HDAC. In addition, glucocorticoid
receptors have been reported to have growth inhibitory prop-
erties in several hematological and solid tumor cells, includ-
ing in Ishikawa cells (King and Cidlowski, 1998). It will be of
interest in the future to assess whether HDACi also have a
stimulatory effect on the glucocorticoid target genes that
mediate these antiproliferative activities.
Acknowledgments
We are grateful to Drs. Pierre Chambon and Richard Pestell for
kind gifts of reagents.
References
Alao JP, Lam EW, Ali S, Buluwela L, Bordogna W, Lockey P, Varshochi R, Stavro-
poulou AV, Coombes RC, and Vigushin DM (2004) Histone deacetylase inhibitor
trichostatin A represses estrogen receptor alpha-dependent transcription and pro-
motes proteasomal degradation of cyclin D1 in human breast carcinoma cell lines.
Clin Cancer Res 10:8094–8104.
Barsalou A, Dayan G, Anghel SI, Alaoui-Jamali M, Van de Velde P, and Mader S
(2002) Growth-stimulatory and transcriptional activation properties of raloxifene
in human endometrial Ishikawa cells. Mol Cell Endocrinol 190:65–73.
Bourdeau V, Deschenes J, Metivier R, Nagai Y, Nguyen D, Bretschneider N, Gannon
F, White JH, and Mader S (2004) Genome-wide identification of high-affinity
estrogen response elements in human and mouse. Mol Endocrinol 18:1411–1427.
Fernandes I, Bastien Y, Wai T, Nygard K, Lin R, Cormier O, Lee HS, Eng F, Bertos
NR, Pelletier N, et al. (2003) Ligand-dependent nuclear receptor corepressor LCoR
functions by histone deacetylase-dependent and -independent mechanisms. Mol
Cell 11:139–150.
Fu M, Wang C, Zhang X, and Pestell RG (2004) Acetylation of nuclear receptors in
cellular growth and apoptosis. Biochem Pharmacol 68:1199–1208.
Glaser KB, Staver MJ, Waring JF, Stender J, Ulrich RG, and Davidsen SK (2003)
Gene expression profiling of multiple histone deacetylase (HDAC) inhibitors: de-
fining a common gene set produced by HDAC inhibition in T24 and MDA carci-
noma cell lines. Mol Cancer Ther 2:151–163.
Graziani G, Tentori L, Portarena I, Vergati M, and Navarra P (2003) Valproic acid
increases the stimulatory effect of estrogens on proliferation of human endometrial
adenocarcinoma cells. Endocrinology 144:2822–2828.
Hsu SC and DeFranco DB (1995) Selectivity of cell cycle regulation of glucocorticoid
receptor function. J Biol Chem 270:3359–3364.
Jang ER, Lim SJ, Lee ES, Jeong G, Kim TY, Bang YJ, and Lee JS (2004) The histone
deacetylase inhibitor trichostatin A sensitizes estrogen receptor alpha-negative
breast cancer cells to tamoxifen. Oncogene 23:1724–1736.
Jenuwein T and Allis CD (2001) Translating the histone code. Science (Wash DC)
293:1074–1080.
Kawai H, Li H, Avraham S, Jiang S, and Avraham HK (2003) Overexpression of
histone deacetylase HDAC1 modulates breast cancer progression by negative
regulation of estrogen receptor alpha. Int J Cancer 107:353–358.
Keen JC, Yan L, Mack KM, Pettit C, Smith D, Sharma D, and Davidson NE (2003)
A novel histone deacetylase inhibitor, scriptaid, enhances expression of functional
estrogen receptor alpha (ER) in ER negative human breast cancer cells in combi-
nation with 5-aza 2⬘-deoxycytidine. Breast Cancer Res Treat 81:177–186.
Kelly WK, Richon VM, O’Connor O, Curley T, MacGregor-Curtelli B, Tong W, Klang
M, Schwartz L, Richardson S, Rosa E, et al. (2003) Phase I clinical trial of histone
deacetylase inhibitor: suberoylanilide hydroxamic acid administered intrave-
nously. Clin Cancer Res 9:3578–3588.
King KL and Cidlowski JA (1998) Cell cycle regulation and apoptosis. Annu Rev
Physiol 60:601–617.
Kinyamu HK and Archer TK (2004) Modifying chromatin to permit steroid hormone
receptor-dependent transcription. Biochim Biophys Acta 1677:30–45.
Kos M, Reid G, Denger S, and Gannon F (2001) Minireview: genomic organization of
the human ERalpha gene promoter region. Mol Endocrinol 15:2057–2063.
Lipka C, Mankertz J, Fromm M, Lubbert H, Buhler H, Kuhn W, Ragosch V, and
Hundertmark S (2004) Impairment of the antiproliferative effect of glucocortico-
steroids by 11beta-hydroxysteroid dehydrogenase type 2 overexpression in MCF-7
breast-cancer cells. Horm Metab Res 36:437–444.
Littlefield BA and Cidlowski JA (1984) Increased steroid responsiveness during
sodium butyrate-induced “differentiation” of HeLa S3 cells. Endocrinology 114:
566–575.
Mader S and White JH (1993) A steroid-inducible promoter for the controlled over-
expression of cloned genes in eukaryotic cells. Proc Natl Acad Sci USA 90:5603–
5607.
Mao C and Shapiro DJ (2000) A histone deacetylase inhibitor potentiates estrogen
receptor activation of a stably integrated vitellogenin promoter in HepG2 cells.
Endocrinology 141:2361–2369.
Margueron R, Duong V, Bonnet S, Escande A, Vignon F, Balaguer P, and Cavailles
V (2004a) Histone deacetylase inhibition and estrogen receptor alpha levels mod-
ulate the transcriptional activity of partial antiestrogens. J Mol Endocrinol 32:
583–594.
Margueron R, Duong V, Castet A, and Cavailles V (2004b) Histone deacetylase
inhibition and estrogen signalling in human breast cancer cells. Biochem Phar-
macol 68:1239–1246.
McLaughlin F, Finn P, and La Thangue NB (2003) The cell cycle, chromatin and
cancer: mechanism-based therapeutics come of age. Drug Discov Today 8:793–802.
McLaughlin F and La Thangue NB (2004) Histone deacetylase inhibitors open new
doors in cancer therapy. Biochem Pharmacol 68:1139–1144.
Metivier R, Penot G, Hubner MR, Reid G, Brand H, Kos M, and Gannon F (2003)
Estrogen receptor-alpha directs ordered, cyclical and combinatorial recruitment of
cofactors on a natural target promoter. Cell 115:751–763.
Mulholland NM, Soeth E, and Smith CL (2003) Inhibition of MMTV transcription by
HDAC inhibitors occurs independent of changes in chromatin remodeling and
increased histone acetylation. Oncogene 22:4807–4818.
Munster PN, Troso-Sandoval T, Rosen N, Rifkind R, Marks PA, and Richon VM
(2001) The histone deacetylase inhibitor suberoylanilide hydroxamic acid induces
differentiation of human breast cancer cells. Cancer Res 61:8492–8497.
Nagao M, Oyanagi K, Tsuchiyama A, Aoyama T, and Nakao T (1987) Studies on the
expression of liver-specific functions of human fetal hepatocytes in primary cul-
ture. Tohoku J Exp Med 152:23–29.
Peart MJ, Smyth GK, van Laar RK, Bowtell DD, Richon VM, Marks PA, Holloway
AJ, and Johnstone RW (2005) Identification and functional significance of genes
regulated by structurally different histone deacetylase inhibitors. Proc Natl Acad
Sci USA 102:3697–3702.
Plesko MM, Hargrove JL, Granner DK, and Chalkley R (1983) Inhibition by sodium
Effects of HDAC Inhibitors on Steroid Receptor Signaling 1861
at ASPET Journals on November 3, 2015molpharm.aspetjournals.orgDownloaded from
butyrate of enzyme induction by glucocorticoids and dibutyryl cyclic AMP. A role
for the rapid form of histone acetylation. J Biol Chem 258:13738–13744.
Reid G, Metivier R, Lin CY, Denger S, Ibberson D, Ivacevic T, Brand H, Benes V, Liu
ET, and Gannon F (2005) Multiple mechanisms induce transcriptional silencing of
a subset of genes, including oestrogen receptor alpha, in response to deacetylase
inhibition by valproic acid and trichostatin A. Oncogene 24:4894–4907.
Rosenfeld MG and Glass CK (2001) Coregulator codes of transcriptional regulation
by nuclear receptors. J Biol Chem 276:36865–36868.
Ruh MF, Tian S, Cox LK, and Ruh TS (1999) The effects of histone acetylation on
estrogen responsiveness in MCF-7 cells. Endocrine 11:157–164.
Sanchez R, Nguyen D, Rocha W, White JH, and Mader S (2002) Diversity in the
mechanisms of gene regulation by estrogen receptors. Bioessays 24:244–254.
Strait KA, Dabbas B, Hammond EH, Warnick CT, Iistrup SJ, and Ford CD (2002)
Cell cycle blockade and differentiation of ovarian cancer cells by the histone
deacetylase inhibitor trichostatin A are associated with changes in p21, Rb and Id
proteins. Mol Cancer Ther 1:1181–1190.
Takai N, Desmond JC, Kumagai T, Gui D, Said JW, Whittaker S, Miyakawa I, and
Koeffler HP (2004) Histone deacetylase inhibitors have a profound antigrowth
activity in endometrial cancer cells. Clin Cancer Res 10:1141–1149.
Voegel JJ, Heine MJ, Tini M, Vivat V, Chambon P, and Gronemeyer H (1998) The
coactivator TIF2 contains three nuclear receptor-binding motifs and mediates
transactivation through CBP binding-dependent and -independent pathways.
EMBO (Eur Mol Biol Organ) J 17:507–519.
Wang C, Fu M, Angeletti RH, Siconolfi-Baez L, Reutens AT, Albanese C, Lisanti MP,
Katzenellenbogen BS, Kato S, Hopp T, Fuqua SA, et al. (2001) Direct acetylation
of the estrogen receptor alpha hinge region by p300 regulates transactivation and
hormone sensitivity. J Biol Chem 276:18375–18383.
White JH, Fernandes I, Mader S, and Yang XJ (2004) Corepressor recruitment by
agonist-bound nuclear receptors. Vitam Horm 68:123–143.
Yang X, Phillips DL, Ferguson AT, Nelson WG, Herman JG, and Davidson NE (2001)
Synergistic activation of functional estrogen receptor (ER)-alpha by DNA methyl-
transferase and histone deacetylase inhibition in human ER-alpha-negative
breast cancer cells. Cancer Res 61:7025–7029.
Address correspondence to: Dr. Sylvie Mader, De´ partement de Biochimie,
Universite´ de Montre´ al, CP 6128 Succursale Centre-Ville, Montre´al, Que´bec
H3C 3J7, Canada. E-mail: sylvie.mader@umontreal.ca
1862 Rocha et al.
at ASPET Journals on November 3, 2015molpharm.aspetjournals.orgDownloaded from
