Vol. 265, No.

28, Issue of October 5, PI,, 17174-17179,199O

(0 1990 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in Ii S. A.

Potent and Specific Inhibition of Mammalian Histone Deacetylase Both

in Vivo and in Vitro by Trichostatin A*
(Received for publication, June 7, 1990)

Minoru YoshidaS, Masako Kijima, Mitsuru Akita, and Teruhiko Beppu

From the Department of Agricultural Chemistry, Faculty of Agriculture, The University of Tokyo, Yayoi I-l-l, Bunkyo-ku,
Tokyo 113, Japan

(R)-Trichostatin A (TSA) is a Streptomyces product provided by Riggs et al. (22), who showed that the exposure
which causes the induction of Friend cell differentia- of cultured cells to a high concentration of sodium n-butyrate
tion and specific inhibition of the cell cycle of normal causes a reversible accumulation of highly acetylated histones
rat fibroblasts in the Gl and G2 phases at the very low in their nuclei. The effect of n-butyrate as a noncompetitive
concentrations. We found that TSA caused an accu- inhibitor on the histone deacetylase was proved with the
mulation of acetylated histone species in a vareity of partially purified preparation (23). n-Butyrate has also been
mammalian cell lines. Pulse-labeling experiments in- known to have various biological effects on cultured mam-
dicated that TSA markedly prolonged the in uiuo half- malian cells, such as suppression of the characteristic phe-
life of the labile acetyl groups on histones in mouse
mammary gland tumor cells, FMSA. The partially pu- notypes of transformed cells, induction of differentiation in
rified histone deacetylase from wild-type FM3A cells many tumor cell lines, as well as specific inhibition of the cell
was effectively inhibited by TSA in a noncompetitive cycle at the Gl and G2 phases (24-28). However, whether n-
manner with Ki = 3.4 nM. A newly isolated mutant cell butyrate treatment leads to an activation of some specific
line of FM3A resistant to TSA did not show the accu- genes through the effect on histone deacetylases is still un-
mulation of the acetylated histones in the presence of clear. Experiments in vivo using n-butyrate at a high concen-
a higher concentration of TSA. The histone deacetylase tration have to be interpreted with caution because of its
preparation from the mutant showed decreased sensi- pleiotropic effects on other enzymes, cytoskeleton, cell mem-
tivity to TSA (Ri = 31 nM, noncompetitive). These branes, etc. (26,27). Therefore, the use of a more specific and
results clearly indicate that TSA is a potent and spe- potent inhibitor of histone deacetylase seems to be necessary
cific inhibitor of the histone deacetylase and that the to carry out further more refined analyses.
in uiuo effect of TSA on cell proliferation and differ- (R)-Trichostatin A (TSA)’ was originally reported as a
entiation can be attributed to the inhibition of the fungistatic antibiotic by Tsuji et al. (29). Recently, we dem-
enzyme. onstrated that the extremely low concentrations of TSA
caused induction of Friend leukemia cell differentiation (30)
as well as inhibition of the cell cycle of normal rat fibroblasts
Reversible histone acetylation, which occurs in the t-amino in both the Gl and G2 phases (31). The fact that an unnatural
group of specific internal lysine residues located at the highly (S)-enantiomer of TSA was totally inactive suggests the in-
basic N-terminal domains of core histones, was discovered by volvement of a target molecule with strict stereospecificity in
Allfrey and colleagues (1, 2). Turnover of the acetyl groups these biological effects (32). In the present paper, we show
on histone molecules occurs rapidly in the cells (3,4), and the that nanomolar concentrations of (R)-TSA cause a marked
level of the acetylation is controlled by equilibrium of the accumulation of highly acetylated histones in vivo and
acetylating and deacetylating activities (5-8). Its possible strongly inhibit the activity of the partially purified histone
roles in DNA replication (9, lo), transcription (g-13), and deacetylase in vitro. In addition, experiments using a TSA-
spermiogenesis (14, 15) have been postulated. In general, resistant mutant suggest that inhibition of the enzyme is the
hyperacetylation of histones is assumed to cause relaxation primary reason for the inhibitory effect of TSA in vivo. TSA
of the chromatin structure to make a variety of factors acces- thus seems to be a very useful tool for analyzing the multiple
sible to DNA (8, 16). Recently, highly acetylated histones functions of histone acetylation in regulatory mechanisms of
were found to be co-purified with actively transcribed genes cell proliferation and differentiation.
but not with inactive genes (12, 17-19), suggesting that his-
EXPERIMENTAL PROCEDURES
tones in the nucleosomes containing active and potentially
active genes are highly acetylated. Although these observation Materials-(R)-TSA prepared from a culture broth of Streptomyces
suggest involvement of histone acetylation in controlling the dater&s No. 145 (30) was used for most of the experiments. Chemi-
cally synthesized ‘(R)- and (S)-TSA and (R)- and (S)-trichostatic
chromatin functions in eukaryotic cells, numerous subsequent acids (Fig. 1) were supnlied by Mori and Koseki (33). Leptomvcin B.
studies are still too contradictory to give a definitive conclu- which’we also found-to be the Gl- and GP-specific inhibitor-of the
sion (20, 21). cell cycle of both mammalian and fission yeast cells (34), was purified
An important tool for studies of histone acetylation was from a culture broth of Streptomyces sp. ATS 1287 as described
previously (35). Cycloheximide, actinomycin D, adriamycin, Col-
* This work was supported by a grant-in-aid for Cancer Research cemid, and vinblastine were purchased from Sigma. Sodium butyrate
from the Ministry of Education. Science. and Culture. Janan. The and peplomycin were from Wako Pure Chemical Industries Ltd. [3H]
costs of publication of this article were defrayed in.part by the Acetate (2.7 Ci/mmol) was purchased from Amersham Corp. All the
payment of page charges. This article must therefore be hereby other chemicals were of reagent grade.
marked “advertisement” in accordance with 18 U.S.C. Section 1734 Cells and Culture Conditiorz-A mouse mammary gland tumor cell
solely to indicate this fact.
$ To whom correspondence should be addressed. i The abbreviation used is: TSA, trichostatin A.

17174

This is an Open Access article under the CC BY license.

 Trichostatin
R=NHOH Trlch0stm A (TSA)
R=OH Trlchostatlc acid
FIG. 1. Structures of trichostatin A and its
pounds.
line, FMBA (36), was obtained from D. Ayusawa of the University of
Tokyo. A rat embryonic fibroblast cell line 3Yl-B (37) and its SV40-
transformed derivative, W3Y (38), a mouse Friend leukemia cell line
DS19 (39), and a mouse teratocarcinoma cell line F9 (40) were
obtained from M. Oishi of the University of Tokyo. Cells were
cultured in a humidified atmosphere of 5% Con, 95% air at 37 “C in
ES medium supplemented with 2% calf serum for FM3A or minimal
essential medium supplemented with 12% fetal bovine serum for the
other cell lines. In all experiments, media and sera from the same lot
number and source were used.
Isolation of TSA-resistant Mutant-A TSA-resistant mutant cell
line TR303 was isolated as follows: exponentially growing FMBA cells
in ES medium were treated with 0.25 wg/ml of N-methyl-N-nitro-
N-nitrosoguanidine at 33.5 “C for 2 h. After washing out the N-
methyl-N’-nitro-N-nitrosoguanidine, the cells were incubated at
33.5 “C for 3 days. Then the cells were further inoculated in ES soft
agar plates containing 50 rig/ml TSA and cultured at 33.5 “C for 3
weeks. The surviving colonies were tested for their TSA sensitivities
at 37 “C. A mutant clone obtained was designated as TR303.
Assay for Cytotoxicity-Exponentially growing cultures of wild-
type FM3A cells and mutant TR303 cells were suspended in the
medium containing various concentrations of drugs at an initial cell
density of 5 x 10’ cells/ml in a 60-mm Petri dish. After incubation
for various lengths of time, the viable cell number of each dish was
scored in a Burker-Turk cell counter using the trypan blue exclusion
method. The cell number was determined in duplicated cultures.
Pulse-Chase Cultivation with [‘HIAcetate-FM3A cells grown in
suspension in ES medium at a density of 1 x lo6 cells/ml were
exposed to 0.5 mCi/ml of [‘HIacetate for 30 min. Then the radioactive
medium was removed, and the cells were washed three times with
fresh medium with or without 100 rig/ml TSA. The washed cells were
finally resuspended in the fresh medium with or without TSA at a
cell density of 1 x 10’ cells/ml and cultivation was continued in the
presence or absence of TSA at 37 “C.
Extraction of Cellular Histones-Histones of cultured cells were
extracted according to the procedure of Cousens et al. (23). About 2
x 10’ cells cultured for 6 h with or without TSA were harvested using
a rubber policeman, collected by centrifugation at 700 X g for 10 min,
and washed once with phosphate-buffered saline. The washed cells
were suspended in 1 ml of ice-cold lysis buffer (10 mM Tris-HCl, 50
mM sodium bisulfite, 1% Triton X-100, 10 mM MgCl,, 8.6% sucrose,
pH 6.5). After Dounce homogenization, the nuclei were collected by
centrifugation at 1,000 X g fir 10 min, washed three times with the
lvsis buffer. and once with 10 mM Tris-HCl. 13 mM EDTA. DH 7.4.
&ccessively. The pellet was suspended in 0.1 ml of ice-cold H,b using
a Vortex mixer, and concentrated H2S04 was added to the suspension
to give a concentration of 0.4 N. After incubation at 4 “C for at least
1 h, the suspension was centrifuged for 5 min at 15,000 rpm using a
microcentrifuge, and the supernatant was taken and mixed with 1 ml
of acetone. After overnight incubation at -20 “C, the coagulated
material was collected by microcentrifugation and air-dried. This acid
soluble histone fraction was dissolved in 50 ~1 of HZO. Protein was
quantitated using a protein assay kit (Bio-Rad).
Acid Urea Triton Gel Electrophoresis-Acetylation of cellular his-
tones during the treatment with TSA was analyzed by slab gel
electrophoresis using acid/urea/Triton (AUT) gel (1 M acetic acid, 8
M urea, 0.5% Triton X-100, 45 mM NH, 16% acrylamide) basically
as according to Cohen et al. (41). The procedure was modified by
incorporating 3 cm of an upper gel (1 M acetic acid, 6.3 M urea, 4.4%
acrylamide) onto 10 cm of the separating gel. The extracted cellular
histones were incubated with the same volume of loading buffer (7.4
M urea, 1.4 M NH:, 10 mM dithiothreitol) for 5 min and were then
added with a l/s volume of 1% pyronine G in glacial acetic acid. The
mixture was applied onto of the upper gel and electrophoresed in 0.2
A, an Inhibitor of Histone Deacetylase 17175
M glycine, 1 M acetic acid. Gels were stained with Coomassie Brilliant
Blue R-250 (Nacalai tesque), dried, and photographed. When fluo-
rography became necessary, the gels were treated with EN”HANCE
(Du Pont-New England Nuclear) before drying, and dried gels were
exposed at -80 “C to FUJI RXO-H film for 3 weeks.
Partial Purification of Mouse Histone Deacetylase-The mouse
histone deacetylase was partially purified from both wild-type FM3A
and mutant TR303 cells. Cells in 4 liters of the suspension culture (1
x 10” cells/ml in ES supplemented with 2% calf serum) in an 8-liter
spinner culture bottle were centrifuged and suspended in 40 ml of the
$IDA buffer consisting of 15 mM potassium phosphate, pH 7.5, 5%
related com- plvcerol. and 0.2 mM EDTA. After homogenization, nuclei were
I _
collected by centrifugation (35,000 x g, 10 min) and homogenated in
20 ml of the buffer containing 1 M (NH&SO,. The viscous homoge-
nate was sonicated and clarified by centrifugation, and the deacetylase
was precipitated by raising the concentration of (NH&SO, to 3.5 M.
The precipitated protein was dissolved in 10 ml of the HDA buffer
and dialyzed against 4 liters of the same buffer. The dialyzate was
then loaded onto a DEAE-cellulose (Whatman DE52) column (25 X
85 mm) equilibrated with the same buffer and eluted with 300 ml of
a linear gradient (O-O.6 M) of NaCl. A single peak of histone dea-
cetylase activity was eluted between 0.2 and 0.3 M NaCl, which was
estimated to be purified go-fold in the specific activity.
Preparation of [‘H]Acetyl H&ones-To obtain [“Hlacetyl-labeled
histones as the substrate for the histone deacetylase assay, 1 X 10’
cells of FMIA in 50 ml of ES medium were incubated with 0.5 mCi/
ml [“HIacetate in the presence of 5 mM sodium butyrate for 30 min,
and the labeled histone fraction was immediately extracted as de-
scribed above. Radioactivity was determined using a liquid scintilla-
tion spectrometer (Aloka Co.). The specific activity was determined
to be 0.45 rCi/mg histones.
Assay for Historze Deacetylase Actiuity-The assay for in uitro
deacetylation of histones was basically according to that as proposed
by Inoue and Fujimoto (5). For the standard assay, 4 ~1 of [“Hlacetyl-
labeled histones-(2500 cpm/ 5pg) was added to 96 ~1 of the enzyme
fraction. and the mixture was incubated at 37 “C for 10 min. The
reaction was stopped by the addition of 10 pl of concentrated HCl.
The released [:‘H]acetic acid was extracted with 1 ml of ethyl acetate,
and 0.9 ml of the solvent layer was taken into 5 ml of toluene
scintillation solution for determination of radioactivity.
RESULTS
TSA Causes the Accumulation of Acetylated Hi&ones in
Mammalian Cells-In the course of analyzing the effects of
TSA on histone modifications such as phosphorylation, acet-
ylation, poly(ADP)-ribosylation, and ubiquitination, we found
that TSA caused a dramatic increase in histone acetylation
in various mammalian cell lines at very low concentrations.
AUT gel electrophoresis allows separation of each cellular
histone (Hl, H2A, H2B, H3, and H4) with acetylation differ-
ing extents due to the slower migration rates of the acetylated
species. Fig. 2 shows the profiles of histones extracted from
various cultured cell lines separated on AUT gel electropho-
resis. The cell lines that we tested for response to TSA
treatment were; 3Yl cells, a normal fibroblast cell line cloned
from Fisher rat embryo showing strong contact-inhibition
3Yl W3Y DS19 HeLa F9
-+- +-+-+-+
FIG. 2. Effect of TSA on histone acetylation in various cell
lines. Cells (see the text) were incubated without (-) or with (+) 100
rig/ml of TSA for 6 h. Each histone fraction was extracted and
analyzed on an AUT gel as described under “Experimental Proce-
dures.” The gel was stained with Coomassie Blue.
17176 Trichostatin A, an Inhibitor of Histone Deacetylase

(37); W3Y cells, an SV40-transformed 3Yl cell line showing A B

no contact-inhibition (38); HeLa cells, an epithelia-like car- 1234567 1234567

cinoma cell line from human cervix; DS19 cells, a murine

Friend erythroleukemia cell line (39); and F9 cells, a murine
teratocarcinoma cell line (40). It is evident that the core H2A 1 *++tb2+ ; ‘; *

histones in all cell types derived from human, rat, and mouse
were highly acetylated when they were incubated with 100
rig/ml TSA for 6 h. In particular, the H4 band was followed
by four additional bands like a ladder, which corresponded to
mono-, di-, tri-, and tetra-acetylated forms, respectively.
The dose response of TSA on histone acetylation was
examined using the mouse mammary gland tumor cell, FM3A, FIG. 4. Effect of TSA on turnover of [3H]acetate incorpo-
by AUT gel electrophoresis (Fig. 3). The obvious increases in rated into FMBA cells were pulse-labeled with [“HIacetate
histones.
the amount of di- and tri-acetylated species of H4 along with for 30 min. After washing with an unlabeled medium, the cells were
chased for 0 h (lane I). 0.5 h (lanes 2 and 3). 2 h (lanes 4 and 5), and
concomitant decrease in nonacetylated form were observed in 4 h (lanes 6 anh 7) without (i) or with (+). 100 rig/ml of TSA. The
the cells exposed to 2 rig/ml TSA (6.7 nM). Accumulations of histones were isolated and electrophoresed on an AUT gel. The gel
tri- and tetra-acetylated forms of H4 and the acetylation of was stained with Coomassie Blue (A) and then fluorographed (B).
other core histones were more remarkable in the cultures with
TSA at concentrations of more than 10 rig/ml (33 nM).
Although this effect of TSA on histone acetylation is very 400
06 106-f .L.

similar to that of sodium n-butyrate (22, 23, 42-44), the

effective concentration of TSA is lower than that of n-butyr-
ate (5 mM) by 5 orders of magnitude.
Turnover of [‘H]Acetyl Histones in Vivo Is Blocked by
TSA-The level of histone acetylation is mainly controlled
by the acetyltransferase-deacetylase equilibrium (8). To test
whether the effect of TSA is due to increased histone acety-
lation or decreased histone deacetylation, we conducted a
pulse-chase experiment using [“HIacetate. FM3A cells were
labeled with [“HIacetate for 30 min and then chased in the
presence or absence of 100 rig/ml TSA in the nonradioactive
medium for 0.5, 2, and 4 h, respectively. After purification of
the histones from each chased culture, they were electropho-
resed on an AUT gel (Fig. 4A), and the distribution of the
radioactivity in the various histone species was analyzed by
fluorography (Fig. 4B). In the control experiment without
chasing, four acetylated forms of H4 (mono-, di-, tri-, and
tetra-acetylated H4) were clearly separated. All these bands
of acetylated H4 rapidly disappeared during the chase culture
in the absence of TSA, while radioactivity in the tri- and
tetra-acetylated H4 had distinctly increased in the presence
of TSA. Decrease in the total radioactivity in the acetylated
H4 histones was blocked by TSA. A similar effect of TSA was
also observed in H3 and other core histones. These results
clearly showed that the release of the incorporated [“Hlacetyl
groups in histones was inhibited by the TSA treatment in
vivo.
H&one Deacetylase Activity in Vitro Is Strongly Inhibited
by TSA-The results obtained by the pulse-chase experiments
TSA, rig/ml
FIG. 3. Dose response of TSA on histone acetylation
FM3A cells.FM3A cells were treated with various concentrations
of TSA for 6 h. Histones were extracted from the treated cells and
were then analyzed on an AUT gel.
300
o,41 - OS g
I
200 2 - 0.4 a,
0
0.2 2 i
100 - 0.2 g
01
0 -0 9
0
FIG. 5. Elution of histone deacetylase activity from a
DEAE-cellulose column. Nuclear histone
deacetylase was precipi-
tated with 3.5 M (NH&SO, and then dialyzed and applied to a
DEAE-cellulose column. The proteins were eluted with a linear
gradient of NaCl, and each 6 ml fraction was collected. A small
aliquot (0.1 ml) was taken from each fraction and diluted 1:9 with
HDA buffer for the standard assay of the histone deacetylase activity
as described under “Experimental Procedures.” Active fractions (frac-
tions 25-29) were collected as the partially purified enzyme prepara-
tion.
suggested that the most likely target of TSA is the histone
deacetylase. This possibility was tested by analyzing the in
vitro effects of TSA on the partially purified histone dea-
cetylase from mouse FM3A cells. The enzymatic histone dea-
cetylase activity was monitored by measuring the amounts of
[“HIacetic acid released from the [“Hlacetylated histones as
substrate. The chromatin-bound histone deacetylase was par-
tially purified from nuclei of FMBA cells as described under
“Experimental Procedures.” The DEAE-cellulose column
chromatography at the final purification step gave a single
peak of the histone deacetylase (Fig. 5), whose specific activity
had increased 60-fold from the initial crude nuclear extract.
The radioactivity was linearly released from the labeled his-
tones during incubation within 10 min and the released activ-
ity increased proportionally to the increased amounts of the
enzyme preparation added (data not shown). The results of
kinetic analysis in the presence of various concentrations of
TSA are shown in Fig. 6. The Lineweaver-Burk plots of l/v
versus 1/S in the absence and presence of TSA crossed at a
fixed point on the l/S axis suggesting that TSA had no effect
on the apparent K,,, but decreased the V,,,,, value. Thus TSA
is concluded to be a noncompetitive-type inhibitor of histone
deacetylase. The apparent K,,, value of the enzyme for acety-
lated histones was calculated to be 10 pg/ml, while the Ki
value for TSA was determined to be 3.4 nM (Fig. 6, inset).
in
This low Ki value almost corresponds to the effective concen-
trations of TSA to induce hyperacetylation in vivo (below
lo-’ M).
We have observed that chemically synthesized (R)-TSA has
 Trichostatin A, an Inhibitor of Histone Deacetylase 17177
inhibitor of histone deacetylase, it is still not clear whether
the inhibition is a primary reason for the biological effects of
TSA in vivo. This may be concluded if the enzyme of a TSA-
resistant cell line is confirmed to be TSA-resistant. Therefore,
we tried to isolate such a mutant clone from the mouse FM3A
cell line. A TSA-resistant mutant, designated TR303, was
isolated after mutagenesis with N-methyl-N’-nitro-N-nitro-
soguanidine and selection with 50 rig/ml TSA as described
under “Experimental Procedures.” Both wild-type FMBA cells
and TR303 cells grew in suspension with almost the same
doubling time of 10 h at 37 “C in the absence of TSA (Fig. 8).
The presence of TSA at the concentration of 10 rig/ml (33
nM) prolonged the doubling time of wild-type cells to 40 h,
but that of TR303 cells to 13 h. Although wild-type cells could
1011 l/S,
0 100
(me Histone/ml)e’
200 300
not grow at concentrations of more than 20 rig/ml, TR303
cells could proliferate even in the presence of 200 rig/ml TSA
(670 nM) with a doubling time of 19 h.
FIG. 6. Kinetic analysis of inhibition by TSA of mouse his- The sensitivity of TR303 cells to other metabolic inhibitors
tone deacetylase. Radioactivities released from various concentra-
tions of the labeled histone substrate in the presence or absence of is summarized in Table I. No marked cross-resistance was
TSA were determined. The data are presented as a Lineweaver-Burk observed, suggesting that known multiple drug resistance
plot at several different TSA concentrations. The concentrations of mechanisms such as P-glycoprotein (45) are not responsible
TSA added were 0 (O), 2 (A), and 5 (D) nM. The V,,,., values obtained for the TSA-resistance of TR303 cells.
from the Lineweaver-Burk plot were replotted for the determination Histone Deacetylase in TR303 Cells Is Resistant to TSA-
of the K, value (inset). We then examined the effects of TSA on histone deacetylation
in TR303 cells in vivo and in vitro. Histones extracted from
TR303 cells treated with various concentrations of TSA for 6
h were analyzed by AUT gel electrophoresis. Densitometric
analysis of the H4 region in AUT gels stained with Coomassie
Blue is shown in Fig. 9. It is evident that the level of acetylated
species of H4 is very low in TR303 cells even in the presence
of high concentrations of TSA.
In order to analyze the sensitivity to TSA of the in vitro
histone deacetylase activity, we prepared the partially purified
enzyme from TR303 cells as described under “Experimental
Procedures.” The specific activity and the apparent K,,, (10
S, jog Histondml pg/ml) of the enzyme from the mutant cells were almost
identical to those of the enzyme from wild-type cells. A kinetic
FIG. 7. Comparison of inhibitory effects of chemically syn- analysis of the mutant enzyme indicated that TSA also acted
thesized trichostatin derivatives on histone deacetylase. Ra-
dioactivities released from various concentrations of the labeled his- as a noncompetitive inhibitor, and the K, value was estimated
tone substrate in the presence or absence of trichostatin derivatives to be 31 nM (Fig. 10). This Ki value was B-fold higher than
were determined. The data are presented as an S uwsus u plot. Added that of the wild-type enzyme, suggesting significant alteration
derivatives and their concentrations are as follows: no drug (0), 4 nM of the enzyme structure by mutation. Decreased sensitivity to
(R)-TSA (A), 40 nM (R)-TSA (A), 4 nM (S)-TSA (U), 40 nM (S)- TSA of the histone deacetylase in TR303 cells both in vivo
TSA (m), 4 nM (R)-trichostatic acid (V), 40 nM (R)-trichostatic acid and in vitro indicates that the intracellular target of TSA is
(v), 4 nM (S)-trichostatic acid (0), 40 nM (S)-trichostatic acid (+).
the histone deacetylase itself.
essentially the same activity as that of natural TSA but
synthetic (S)-TSA has no effect on both the induction of
Friend cell differentiation and the inhibition of the cell cycle
progression (32). In addition, trichostatic acid, which has a
carboxyl group in place of the hydroxamate group of TSA
(Fig. l), is also biologically inactive (32). To analyze the
relationship between the in vivo biological effects and the in
vitro effects on the histone deacetylase, we compared the
inhibitory activities of these compounds on histone deacety-
lase activity in vitro. As shown in Fig. 7, 4 and 40 nM of
synthetic (R)-TSA inhibited about 60 and 90%, respectively,
of the enzyme activity, which is coincident with the activity
of natural TSA. On the other hand, synthetic (S)-TSA and
(R)- and (S)-trichostatic acids had no effect on the histone
deacetylase activity in vitro. These results indicate that both 0 24 48 72 0 24 48 72
the R configuration at C-6 position and the presence of the Hours in culture
hydroxamate group (Fig. 1) are essential for the inhibition of
FIG. 8. Effect of TSA on growth curves of FM3A and
the histone deacetylase as observed in the biological activity
TR303 cells. Exponentially growing FM3A or TR303 cells were
in vivo. inoculated in the ES medium containing various concentrations of
Isolation of TSA-resistant Mutant of FM3A Cell-Although TSA at the initial density of 5 X lo* cells/ml. Viable cell numbers of
the above mentioned results showed that (R)-TSA is a potent FM3A (0) and TR303 (0) were determined at indicated times.
17178 Trichostatin A, an Inhibitor of Histone Deacetylase
TABLE I
Effect of metabolic inhibitors on the growth of TR303
FMXA cells and TR303 cells were inoculated at an initial density
of 5 x 10’ cells/ml and cultured in the presence of various concentra-
tions of the metabolic inhibitors. The final cell densities were deter-
mined after 3 days of cultivation.
Viable cell number
Drugs DOS2
FMBA TR303
kc/ml cells/ml X IO-’
None 225 278
TSA 0.01 26 220
0.02 Dead 260
0.1 Dead 128
0.2 Dead 100
Cycloheximide 0.02 120 110
0.1 14 14 -100 0 100 200
0.2 5 3 l/S, (mg Histone/ml)K’
Actinomycin D 0.0005 94 63
0.001 32 33 FIG. 10. Kinetic analysis of inhibition by TSA of histone
0.005 3 Dead deacetylase from TR303 cells. Kinetic analysis similar to that in
Peplomycin 1 94 99 Fig. 6 was carried out using the mutant enzyme preparation. The
2 79 70 concentrations of TSA added were 0 (O), 4 (A), 8 (W), and 12 (+) nM.
5 Dead Dead
Adriamycin 0.2 100 79
0.5 46 33 n-Butyrate has also been reported to have a similar inhib-
1 Dead Dead itory effect on the histone deacetylation in vivo (43, 44) and
Colcemid 0.002 121 107 in vitro (23,42). However, it requires far higher concentrations
0.005 6 15 for both inducing hyperacetylation in viva (5 mM) and inhib-
0.01 Dead Dead iting the enzyme activity in vitro (& = 60 PM), and also other
Vinblastine 0.002 102 114 short chain fatty acids such as n-propionate and n-pentan-
0.005 23 1
0.01 Dead Dead onate are active, suggesting that n-butyrate acts as a weak
Leptomycin B 0.0005 72 12 and less specific inhibitor of the enzyme (23). In contrast,
0.001 Dead Dead only TSA with the R configuration can inhibit the enzyme
activity at the nanomolar concentrations among its deriva-
tives tested in this study. These results suggest that TSA
binds the enzyme at a specific binding site, while n-butyrate
may interact in a nonspecific manner like a detergent (23).
We confirmed that TSA has no effect on the other enzyme
activities such as protein kinases, protein phosphatases, DNA
topoisomerases, and calmodulin in uitro.’ Therefore, the use
of TSA will be expected to be useful in analyzing the biological
roles of histone acetylation without giving adverse side effects.
In fact, there is no report on the isolation of a n-butyrate-
resistant mutant, probably due to its detergent-like inhibition
as well as pleiotropic effects on the other cellular functions.
In the present paper, we showed that a TSA-resistant mutant
could be isolated from FM3A cells and that the histone
deacetylase activity of the mutant was actually resistant to
TSA. Our preliminary data indicated that the resistance of
TR303 was dominant,3 suggesting the possibility of cloning
and subsequent genetic analyses of the histone deacetylase
gene.
TSA reversibly blocks the cell cycle of normal rat diploid
fibroblasts (3Yl cells) at both the Gl and G2 phases and
removal of the drug from the G2-arrested culture induces the
formation of proliferative tetraploid cells (31). Interestingly,
a higher concentration of n-butyrate is also reported to show
FIG. 9. Densitometric analysis of histone H4 region on AUT
gels. The cells were exposed to various concentrations of TSA for 6 similar effects on the cell cycle of 3Yl cells and the production
h, and their histones extracted were analyzed by AUT gel electropho- of tetraploid cells (28). However, the relationship between the
resis. The level of histone H4 acetylation was compared between inhibitory effects on the cell cycle or cell growth and hy-pera-
FM3A (A) and TR303 (B) using a dual wavelength flying spot scanner cetylation of histones induced by n-butyrate is not clear. For
(Shimazu). example, Rubenstein et al. (46) observed that growth of hep-
atoma tissue culture cells continues even when 70% of the H3
DISCUSSION and H4 histones are acetylated in the presence of 6 mM n-
The results presented in this paper clearly show that TSA propionate. In our present work, we also observed that both
is a potent inhibitor of the mammalian histone deacetylating wild-type FM3A and mutant TR303 continued to grow even
enzyme. Its extremely low Ki value for the histone deacetylase
can fully account for the accumulation of highly acetylated ’ M. Yoshida and T. Beppu, unpublished results.
histones by the low concentrations of the drug in uiuo. 3 M. Kijima, M. Yoshida, and T. Beppu, unpublished results.
 Trichostatin A, an Inhibitor of Histone Deacetylase 17179
when approximately 70% of H4 was acetylated in the presence 10. Waterborg, J. M., and Matthews, H. R. (1984) Eur. J. Biochem.
142,329-335
of 5 rig/ml and 100 rig/ml, respectively, of TSA (Figs. 8 and
11. Chahal, S., Matthews, H. R., and Bradbury, E. M. (1980) Nature
9). However, growth inhibition at the higher concentrations 287, 76-79
of TSA was found to correlate with a significant accumulation 12. Allegra, P., Sterner, R., Clayton, D. F., and Allfrey, V. G. (1987)
of highly acetylated species such as tetra-acetylated H4 (Fig. J. Mol. Biol. 196, 379-388
9). These observations, together with the fact that the histone 13. Reeves, R. (1984) Biochim. Biophys. Acta 782,343-394
deacetylase from the TSA-resistant mutant cells was resistant 14. Christensen, M. E., and Dixon, G. H. (1982) Deu. Biol. 93, 404-
to TSA, suggest that histone hyperacetylation of at least some 415
15. Gatewood, J. M., Cook, G. R., Balhorn, R., Bradbury, E. M., and
specific parts of chromatin domains is responsible for the Schmid, C. W. (1988) Science 236,962-964
inhibitory effect on cell growth and the cell cycle. 16. Norton, V. G., Imai, B. S., Yau, P., and Bradbury, E. M. (1989)
Another intriguing activity commonly observed with both Cell 57,449-457
TSA and n-butyrate is the induction of differentiation in 17. Sterner, R., Boffa, L., Chen, T. A., and Allfrey, V. G. (1987)
Friend erythroleukemia cells (25, 30). Again relation between Nucleic Acids Res. 4, 3769-3786
the induction of differentiation and hyperacetylation of his- 18. Chen, T. A., and Allfrev. V. G. (1987) Proc. Natl. Acad. Sci. U. S.
A. 84,5252-5256 -
tones is doubted, since dimethyl sulfoxide, a potent differen- 19. Hebbes. T. R.. Thorne. A. W.. and Crane-Robinson. I C. (1988)
tiation inducer, did not cause such an increase in acetylated EMBb J. 7,‘1395-1462
histones (42). However, our previous studies have revealed 20. Ausio, J., and Van Holde, K. E. (1986) Biochemistry 25, 1421-
that nanomolar concentrations of TSA are sufficient to induce 1428
erythroid differentiation via a process similar to that of di- 21. Imai, B. S., Yau, P., Baldwin, J. P., Ibel, K., May, R. P., and
Bradbury, E. M. (1986) J. Biol. Chem. 261,8784-8792
methyl sulfoxide (30), and the effect is limited only to the
22. Riggs, M. G., Whittaker, R. G., Neumann, J. R., and Ingram, V.
@)-isomer of TSA (32). These facts suggest a direct linkage M. (1977) Nature 268, 462-464
between Friend cell differentiation and core histone acetyla- 23. Cousens, L. S., Gallwitz, D., and Alberts, B. M. (1979) J. Biol.
tion, although at least bulk hyperacetylation is unnecessary. Chem. 254,1716-1723
It seems probable that the transient increase in the site- 24. Ginsburg, G., Salomon, D., Sreevalsan, T., and Freese, E. (1973)
specific or phase-dependent acetylation may be essential for Proc. Natl. Acad. Sci. U. S. A. 70, 2457-2461
25. Leder, A., and Leder, P. (1975) Cell 5, 319-322
the early stage of differentiation induction. Recent studies
26. Prasad, K. N., and Sinha, K. (1976) In Vitro 12, 125-132
using mercury-agarose affinity chromatography suggest that 27. Kruh, J. (1982) Mol. Cell..Biochem. 42, 65-82
a transient wave of histone acetylation passes through the 28. Yamada, K., and Kimura. G. (1985) J. Cell. Phvsiol. 122.59-63
chromatin structures, which may cause sequential activation 29. Tsuji, N., Kobayashi, M., Nagashima, K., Wakisaka, k., and
of the essential genes for differentiation and progression of Koizumi, K. (1976) J. Antibiot. (Tokyo) 29, 1-6
the cell cycle (12, 17-19). TSA will be an important tool in 30. Yoshida, M., Nomura, S., and Beppu, T. (1987) Cancer Res. 47,
3688-3691
the analysis of the role of histone acetylation in the regulation 31. Yoshida, M., and Beppu, T. (1988) Exp. Cell Res. 177, 122-131
of the chromatin structure, differentiation, and the cell cycle. 32. Yoshida, M., Hoshikawa, Y., Koseki, K., Mori, K., and Beppu, T.
(1990) J. Antibiot. (Tokyo), in press
Acknowledgments-We are grateful to D. Ayusawa and M. Oishi, 33. Mori, K., and Koseki, K. (1988) Tetrahedron 44.6013-6020
Institute of Applied Microbiology, the University of Tokyo, for gen- 34. Yoshida, M., Nishikawa, M., Nishi, K., Abe, K.,~Horinouchi, S.,
erously supplying the cell lines. We also thank K. Mori, Department and Bepnu. T. (1990) EXD. Cell Res. 187. 150-156
of Agricultural Chemistry, the University of Tokyo, for his gift of the 35. Hamamoto, T., Gunji, S.,’ Tsuji, H., and’Beppu, T. (1983) J.
chemically synthesized trichostatin derivatives. Antibiot. (Tokyo) 36, 639-645
36. Nakano, N. (1966) Tohoku J. Exp. Med. 88,69-84
REFERENCES 37. Kimura, G., Itagaki, A., and Summers, J. (1975) Int. J. Cancer
15,694-706
1. Allfrey, V. G., Faulkner, R., and Mirsky, A. E. (1964) Proc. Nutl. 38. Maruyama, K., Hiwasa, T., and Oda, K. (1981) J. Virol. 37,
Acad. Sci. U. S. A. 51, 786-794 1028-1043
2. Allfrey, V. G. (1970) in Histone and Nucleohistones (Phillips, D. 39. Friend, C., Patuleia, M. C., and deHarven, E. (1966) N&l. Cancer
M. P., ed) pp. 241-294, Plenum Publishing Corp., London Inst. Monogr. 228, 505-520
3. Sanders, L. A., Schechter, N. M., and McCarty, K. S. (1973) 40. Bernstine, E. G., Hooper, M. L., Grandchamp, S., and Ephrussi,
Biochemistry 12,783-791 B. (1973) Proc. N&l. Acad. Sci. U. S. A. 70, 3899-3903
4. Jackson, V., Shires, A., Chalkley, R., and Granner, D. K. (1975) 41. Chohen, L. H., Newcock, K. M., and Zweidler A. (1975) Science
J. Biol. Chem. 250, 4856-4863 190,994-997
5. Inoue, A., and Fujimoto, D. (1970) Biochim. Biophys. Acta 220, 42. Candido, E. P. M., Reeves, R., and Davie, J. R. (1978) Cell 14,
307-316 105-113
6. Vidali, G., Boffa, L. C., and Allfrey, V. G. (1972) J. Biol. Chem. 43. Sealy, L., and Chalkley, R. (1978) Cell 14, 115-121
247,7365-7373 44. Vidali, G., Boffa, L. C., Bradbury, E. M., and Allfrey, V. G. (1978)
7. Kikuchi, H., and Fujimoto, D. (1973) FEBS Lett. 29, 280-282 Proc. Natl. Acad. Sci. U. S. A. 75, 2239-2243
8. Csordas, A. (1990) Biochem. J. 265,23-38 45. Riordan, J. R., Deuchars, K., Kartner, N., Alon, N., Trent, J.,
9. Allfrey, V. G. (1980) in Cell Biology: A Comprehensive Treatise and Ling, V. (1985) Nature 316,817-819
(Goldstein, L., and Prescott, D. M., eds) Vol. 3, pp. 347-437, 46. Rubenstein, P., Sealy, L., Marshall, S., and Chalkley, R. (1979)
Academic Press, New York Nature 280,692-693