Structure-based optimization of phenylbutyrate-derived histone deacetylase inhibitors

Article abstract of DOI:10.1021/jm0503749

Previously, we developed a strategy to develop a novel class of histone deacetylase (HDAC) inhibitors by tethering short-chain fatty acids with Zn ²⁺-chelating motifs, which led to N-hydroxy-4-(4-phenylbutyryl-amino) benzamide (HTPB), a hydroxa

Full text of DOI:10.1021/jm0503749

5530
J. Med. Chem. 2005, 48, 5530-5535
Structure-Based Optimization of Phenylbutyrate-Derived Histone Deacetylase
Inhibitors
Qiang Lu, Da-Sheng Wang, Chang-Shi Chen, Yuan-Dong Hu, and Ching-Shih Chen*
Division of Medicinal Chemistry, College of Pharmacy, The Ohio State University, Columbus, Ohio 43210
Received April 20, 2005
Previously, we developed a strategy to develop a novel class of histone deacetylase (HDAC)
inhibitors by tethering short-chain fatty acids with Zn2+-chelating motifs, which led to
N-hydroxy-4-(4-phenylbutyryl-amino)benzamide (HTPB), a hydroxamate-tethered phenylbu-
tyrate derivative with sub-micromolar potency in inhibiting HDAC activity and cancer cell
proliferation. In this study, we carried out structure-based optimization of HTPB by using the
framework generated by the structure of histone deacetylase-like protein (HDLP)-trichostatin
A (TSA) complexes. Docking of HTPB into the HDLP binding domain suggested that the
hydrophobic microenvironment encompassed by Phe-198 and Phe-200 could be exploited for
structural optimization. This premise was corroborated by the greater potency of (S)-(+)-N-
hydroxy-4-(3-methyl-2-phenylbutyrylamino)-benzamide [(S)-11] (IC50 in HDAC inhibition, 16
nM), of which the isopropyl moiety was favorable in interacting with this hydrophobic motif.
(S)-11 at concentrations as low as 0.1 µM was effective in causing histone hyperacetylation
and p21WAF/CIP1 overexpression and suppressing proliferation in cancer cells.
Introduction
hydroxamate group to reach the polar bottom of the
pocket and coordinate with the Zn2+ cation, whereas the
polar cap group at the other end of the spacer makes
contact at the pocket entrance. On the basis of this
three-component working model (i.e., cap group-linker-
Zn2+-chelating motif), we rationalized that the weak
potency of short-chain fatty acids was, in part, attribut-
able to their inability to access the Zn2+ cation in the
active-site pocket. Accordingly, we described a novel
strategy of tethering short-chain fatty acids (valproate,
Histone deacetylase (HDAC) is recognized as a prom-
ising target for cancer treatment because substantial
evidence has linked the dysregulated histone code to the
pathogenesis of many forms of cancer.1,2 This premise
is supported by the ability of HDAC inhibitors to trigger
growth arrest, differentiation, and apoptosis in tumor
cells through the transcriptional activation of a small
set of genes that regulate cell proliferation and cell cycle
progression.3-9 These in vitro findings have also been
confirmed in different xenograft tumor models, indicat-
ing the therapeutic relevance of HDAC inhibitors to
cancer therapy. To date, several structurally distinct
classes of HDAC inhibitors have advanced into phase I
and/or phase II clinical trials in solid tumors and hema-
tological malignancies, including short-chain fatty acids
(e.g., valproate and phenylbutyrate), benzamide deriva-
tives (e.g., MS-275), hydroxamic acids [e.g., suberyoyl
anilide hydroxamic acid (SAHA) and LAQ824], and
cyclic peptides (e.g., depsipeptide). However, because
many of these agents are associated with low potency
or cardiovascular toxicity, there is an urgency to develop
novel HDAC inhibitors to overcome these problems.
Previously, Finnin et al. reported the three-dimen-
sional structure of a histone deacetylase-like protein
(HDLP), which provides insight into the unique mode
of action of SAHA and trichostatin A (TSA) in HDAC
inhibition.10 The HDLP catalytic site consists of a
narrow, tube-like pocket spanning the length equivalent
to four- to six-carbon straight chains. A Zn2+ cation is
positioned near the bottom of this enzyme pocket, which,
in cooperation with two His-Asp charge-relay systems,
facilitates the deacetylation catalysis. Mechanistically,
the long aliphatic chain of TSA and SAHA allows the
butyrate, phenylacetate, and phenylbutyrate) with Zn2+
-
chelating motifs through different aromatic amino acid
linkers to develop a novel class of HDAC inhibitors.11
Among a series of derivatives with varying potency,
N-hydroxy-4-(4-phenylbutyryl-amino)benzamide (HTPB),
a hydroxamate-tethered phenylbutyrate, represented an
optimal agent with an IC50 value of 44 nM, compared
to 0.4 mM for the parent molecule phenylbutyrate.
In this study, we embarked on the structure-based
optimization of HTPB by using the framework gener-
ated by the crystal structure of HDLP-TSA complexes.
On the basis of the hypothesis that the hydrophobic
residues flanking the cap group-binding motif could be
exploited for lead optimization, we have generated (S)-
11, an optically active R-branched phenylbutyryl deriva-
tive, with an IC50 value of 16 nM in HDAC inhibition.
Molecular Docking of HTPB
Despite distinct structural differences, HTPB and
TSA displayed comparable potency in HDAC inhibition,
suggesting a high degree of flexibility in ligand recogni-
tion within the HDAC catalytic domain. To compare the
mode of protein-ligand interactions, TSA and HTPB
were docked into the active-site pocket of HDLP follow-
ing energy minimization (Figure 1).
We rationalized that the high potency of both HTPB
and TSA was attributable to two structural features. A
salt bridge formation between the hydroxamate and the
zinc cation represents a major driving force for ligand
* To whom correspondence should be addressed. Address: Parks
Hall, Rm 336, 500 West 12th Avenue, The Ohio State University,
Columbus, OH 43210-1291. Tel: (614) 688-4008. Fax: (614) 688-8556.
E-mail: chen.844@osu.edu.
10.1021/jm0503749 CCC: $30.25 © 2005 American Chemical Society
Published on Web 07/13/2005

Novel Histone Deacetylase Inhibitors
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 17 5531
Figure 1. Modeled docking of TSA (left panel) and HTPB (right panel) into the active-site pocket of HDLP. Both small molecules
are in green.
binding.12 Equally important, the cap group and linker
play a crucial role in anchoring the ligand inside the
tube-like pocket through the hydrophobic and/or π-π
stacking interactions. As shown, a narrow, hydrophobic
channel, generated by the two juxtaposed phenyl moi-
eties of Phe-141 and Phe-198, was responsible for inter-
acting with the linker, whereas the groove surrounded
by Tyr-91 and Glu-92 was responsible for the cap group.
Conceivably, the higher potency of TSA relative to HTPB
(IC50 ) 5-15 nM vs 44 nM, respectively) might, in part,
be accounted for by hydrogen bonding between the
dimethylamino moiety of TSA and the backbone amide.
Chemistry
In light of the importance of π-π/hydrophobic inter-
actions in ligand anchoring, disruption of any these
interactions might lead to a decrease in inhibitory
potency. To test this premise, we prepared compounds
1-6, which represented hydroxamate-tethered deriva-
Figure 2. Structures of HTPB and compounds 1-11.
tives of HTPB with varying chain lengths in the cap
group/linker, and compound 7, with an additional meth-
These hydroxamate-tethered acyl conjugates were
yl function in the amide linkage of HTPB (Figure 2).
synthesized according to the procedure described in
Scheme 1.
Moreover, we hypothesized that the hydrophobic
microenvironment encompassed by Phe-198 and Phe-
Results
Structural Basis for HDAC-Inhibitory Potency.
Altering the acyl cap group or the ω-amino acid linker
in HTPB resulted in a reduction in HDAC-inhibitory
activity (Figure 3).
Among the five derivatives with varying acyl chain
lengths, the relative potency was in the following
order: HTPB (phenylbutyryl; IC50 ) 44 ( 6 nM) > 3
(phenylpropionyl; IC50 ) 110 ( 37 nM) > 2 (pheny-
lacetyl; IC50 ) 157 ( 26 nM) > 1 (benzoyl; IC50 ) 210
200 could be exploited to enhance the binding potency.
Accordingly, we substituted the phenylbutyryl moiety
of HTPB with different R-branched aromatic acyl groups
(compounds 8-11), of which the rationale was twofold.
First, substitution at the R-position of the acyl cap group
might enhance the metabolic stability by rendering the
amide linkage more sterically hindered. Second, in-
creasing the bulkiness of the acyl function might
enhance binding affinity by augmenting the hydrophobic
interactions with Phe-198 and Phe-200.

5532 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 17
Lu et al.
Scheme 1. General Synthetic Procedures for Compounds 1-11a
a Reagents: (a) (Boc)2O, Et3N, 1,4-dioxane, H2O; (b) NH2OBn‚HCl, BOP-Cl, Et3N; (c) TFA, CH2Cl2; (d) EDC, THF; and (e) 10% Pd/C,
H2, MeOH/THF.
forded the optimal π-π interactions with the Tyr-91
residue. Similarly, extending the ω-amino acid linker,
as in compounds 5 and 6, caused an order of magnitude
increase in IC50 values (0.53 ( 0.10 µM and 1.2 ( 0.2
µM, respectively). Moreover, compound 7, in which the
amide proton of HTPB was substituted with a methyl
function, exhibited a more than 10-fold decrease in po-
tency (IC50 ) 0.52 ( 0.06 µM), which might be attribut-
able to the steric hindrance imposed by the methyl group.
Structural Optimization of HTPB: Steric and
Stereochemical Effects. To exploit the hydrophobic
microdomain nearby Phe-198 and Phe-200 for structural
optimization, we substituted the phenylbutyryl scaffold
of HTPB with a series of R-branched derivatives, includ-
ing 2-methyl-2-phenylpropionyl (8), 2,2-dimethylphe-
nylbutyryl (9), 2-phenylbutyryl (10), and 3-methyl-2-
phenylbutyryl (11). It is noteworthy that except for
compound 8, these derivatives exhibited potency similar
to or even higher than that of HTPB. The IC50 values
of these analogues in inhibiting HDAC activity were as
Figure 3. HDAC-inhibitory potency of compounds 1-7
compared to that of HTPB. In vitro HDAC assay was carried
out using a commercial enzyme assay kit as described in
Experimental Section. Data are presented as the means ( SD
(n ) 3).
nM ( 52 nM) > 4 (phenylpentanoyl; IC50 ) 250 ( 64
nM), indicating that the phenylbutyryl side chain af-
Figure 4. Modeled docking of (S)-11 (left panel; in green) and (R)-11 (right panel; in yellow) into the active-site pocket of HDLP.

Novel Histone Deacetylase Inhibitors
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 17 5533
follows: 8, 98 ( 16 nM; 9, 25 ( 6 nM; rac-10, 54 ( 13
nM; and rac-11, 30 ( 8 nM.
In light of the high potency of rac-10 and rac-11, we
further examined the activity of the optically active
isomers of both agents to investigate the effect of stereo-
chemistry on ligand binding. Our data indicated that
stereoisomeric preference did exist in enzyme inhibition,
that is, the potency of the (S) isomers of both compounds
was significantly higher than that of their (R) counter-
parts. The IC50 values were as follows: (S)-10, 34 ( 7
nM versus (R)-10, 68 ( 16 nM; (S)-11, 16 ( 4 nM versus
(R)-11, 84 ( 21 nM. To shed light on this stereochemical
discrimination, (S)- and (R)-11 were docked into the
binding motif after energy minimization (Figure 4).
This docking analysis revealed a major difference in
the relative locality of the R-isopropyl group inside the
pocket, which might account for the 5-fold difference in
the respective HDAC-inhibitory potency. For (S)-11, the
isopropyl moiety was located in the proximity of Phe-
198 and Phe-200, which favored hydrophobic bonding
with the phenyl rings. In contrast, the orientation of
the isopropyl moiety of the (R) counterpart prohibited
this hydrophobic interaction.
Mechanistic Validation and Stereoselective Ef-
fect on Cell Viability. As part of the target validation,
we assessed the differential effects of these two enan-
tiomers on the status of histone acetylation and the cyc-
lin-dependent kinase inhibitor p21WAF/CIP1 expression in
DU-145 prostate cancer cells, which represent two im-
portant biomarkers associated with intracellular HDAC
inhibition.2 DU-145 cells were exposed to (S)- or (R)-11
at 0.1, 0.25, 0.5, 1, and 2.5 µM in RPMI 1640 medium
supplemented with 10% fetal bovine serum (FBS) for
48 h. Western blot analysis of the cell lysates indicated
that (S)-11 was effective in perturbing both biomarkers
at concentrations as low as 0.1 µM, whereas (R)-11 re-
quired a concentration of at least 0.5 µM to achieve an
appreciable effect (Figure 5A). The preferential impact
of (S)-11 on these intracellular biomarkers also correla-
ted with its higher efficacy in inhibiting the growth of
DU-145 cells in 10% FBS-containing medium. The IC50
values for (S)- and (R)-11 were 0.11 ( 0.01 and 0.82 (
0.13 µM, respectively, after a 96-h exposure (Figure 5B).
Figure 5. Cellular effects of (S)- and (R)-11 in DU-145 pros-
tate cancer cells. (A) Dose-dependent effect of (S)- and (R)-11
on histone H3 acetylation and p21WAF/CIP1 expression. DU-145
cells were exposed to (S)- or (R)-11 at the indicated concentra-
tions in 10% FBS-supplemented RPMI 1640 medium for 48
h. An equivalent amount of proteins from individual lysates
was eletrophoresed and probed by Western blotting with
respective antibodies. Actin was used as an internal reference
protein. (B) Dose-dependent effect of (S)- and (R)-11 on cell
viability. DU-145 cells were exposed to (S)- or (R)-11 at the
indicated concentrations in 10% FBS-supplemented RPMI
1640 medium for 96 h. Viable cells were examined by the MTT
assay. Data are presented as the means ( SD (n ) 6).
optimal access of the hydroxamate group to the zinc
cation. Moreover, we proposed that the hydrophobic
microenvironment encompassed by Phe-198 and Phe-
200 could be exploited for enhancing the inhibitory
potency. This premise was corroborated by the greater
potency of the hydrophobic motif (S)-11, which displayed
a favorable interaction with the R-isopropyl moiety.
Moreover, the stereochemical preference of (S)-11 over
the (R) counterpart in HDAC inhibition is noteworthy.
(S)-11 at concentrations as low as 0.1 µM was effective
in causing histone hyperacetylation and p21WAF/CIP1
overexpression, whereas at least a 5-fold higher con-
centration was required to achieve the same effect for
(R)-11. Conceivably, this stereoisomer strategy might
also be applied to the optimization of other established
HDAC inhibitors.
Discussion
Previously, we described a strategy that tethered
short-chain fatty acids with Zn2+-chelating motifs to
generate a novel class of HDAC inhibitors.11 This effort
has led to HTPB, an optimal agent with sub-micromolar
potency in inhibiting HDAC activity and cancer cell
proliferation. Because HTPB is structurally distinct
from other potent HDAC inhibitors such as TSA and
SAHA, we rationalized that a high degree of flexibility
exists in the active-site pocket for accommodating cap
groups with different stereoelectronic properties. On the
basis of this premise, we carried out the structural
optimization of HTPB by using the framework gener-
ated by the X-ray structure of HDLP-TSA complexes.
Comparing the docking of HTPB versus TSA into the
HDLP binding domain revealed unique structural fea-
tures that underlie each molecule’s ability to attain
optimal inhibitory potency. For both molecules, hydro-
phobic and/or π-π interactions of its scaffold with the
Phe-141/Phe-198 and Tyr-91/Glu-92 subdomains play
an important role in anchoring the ligand to allow
Conclusion
Because many of the established HDAC inhibitors are
associated with certain drawbacks, such as moderate
potency or cardiotoxicity, there is an urgent need to
develop novel agents for clinical testing. (S)-11 provides
a proof of concept that a high degree of flexibility exists
in the active-site pocket to allow for the design of novel
inhibitors with distinct stereoelectronic properties. In
addition to high potency, simple chemical structure and
oral bioavailability are among the unique features of
this novel agent. Testing of the in vivo efficacy of this
agent in a prostate tumor xenograft model is currently
under way.
Experimental Section
Chemical reagents and organic solvents were purchased
from Aldrich unless otherwise mentioned. Nuclear magnetic

5534 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 17
Lu et al.
resonance spectra (1H NMR) were measured on a Bruker 250
MHz. Chemical shifts (δ) were reported in parts per million
(ppm) relative to the TMS peak. Electrospray ionization (ESI)
mass spectrometry analyses were performed with a 3-Tesla
Finnigan FTMS-2000 Fourier Transform mass spectrometer.
Elemental analyses were within ( 0.4% of calculated values.
Flash column chromatography was performed with silica gel
(230-400 mesh). Rabbit anti-acetyl-histone H3 polyclonal
antibodies were purchased from Upstate Biotechnology (Lake
Placid, NY), Rabbit anti-p21WAF/CIP1 antibodies were obtained
from Santa Cruz Biotechnology (Santa Cruz, CA). Mouse
monoclonal anti-actin was from ICN Biomedicals (Irvine, CA).
HRP-conjugated goat anti-rabbit IgG and HRP-conjugated goat
anti-mouse IgG were obtained from Jackson ImmunoResearch
(West Grove, PA).
mL of water was added to the reaction mixture at a temper-
ature under 15 °C. The aqueous layer was separated, and the
organic layer was washed with a mixture of 20 mL of water
and 30 mL of ethyl ether. Aqueous layers were combined, back-
extracted with 20 mL of ethyl ether, and acidified with 6 N
hydrochloric acid, and the product was extracted with 30 mL
of ethyl ether twice. The combined organic layer was washed
with 20 mL of saturated brine, dried with Na2SO4, and
concentrated under vacuum. Hexane was added to the result-
ing colorless oil to yield 1.1 g of white solid.
Resolution of (()-2-Isopropylphenylacetic Acid. 2-Iso-
propylphenylacetic acid (2.3 g), (+)-R-phenylethylamine (1.64
mL), and 63% aqueous ethanol (50 mL) were refluxed and
allowed to cool slowly. The resulting salt was recrystallized
four times from 63% aqueous ethanol. The salt was treated
with 10% sulfuric acid at 0 °C, extracted with ethyl acetate,
dried, and concentrated to yield 340 mg of (R)-(-)-2-isopropy-
Compounds 1-11 were synthesized according to the five-
step procedure described in Scheme 1, the general methods of
which are described as follows.
lphenylacetic acid, [R]25 -59.5° (CHCl3, c 2.27). The mother
D
liquids from the above four recrystalizations were combined,
concentrated, acidified with 10% sulfuric acid at 0 °C, extracted
with ethyl acetate, dried and concentrated, and then dissolved
in 50 mL of ethyl alcohol. (-)-R-Phenylethylamine (1.64 mL)
was added to this solution. The process described above was
repeated to obtain 320 mg of (S)-(+)-2-isopropylphenylacetic
Method a [tert-butyl carbonate (tBoc-) protection]. Triethy-
lamine (TEA, 1.5 equiv) was added to a solution of individual
ω-amino acids in aqueous dioxane (1:1), followed by the
addition of di-tert-butyl dicarbonate (1.5 equiv). After being
stirred overnight, solvents were removed under vacuum, and
3 N aqueous hydrochloric acid (4.5 equiv) was added dropwise
to the residue. The resulting precipitate was collected, washed
with water, and dried.
Method b [bis(2-oxo-3-oxazolidinyl)phosphoramidic chloride
(BOP-Cl) coupling]. TEA (1 equiv) was added to a solution of
the resulting acid in dry THF (5-10 mmol/mL) under N2. The
mixture was stirred for 10 min, and BOP-Cl (1.1 equiv),
O-benzylhydroxylamine hydrochloride (1 equiv), and TEA (3
equiv) were added. After being stirred overnight, the solution
was concentrated under vacuum, and ethyl acetate (100 mL)
was added, followed by the addition of 3% NaHCO3 (50 mL).
The organic phase was separated, washed consecutively with
water and saturated brine (50 mL each), dried over Na2SO4,
and concentrated under vacuum. The resulting residue was
purified by silica gel flash chromatography.
Method c (tBoc-deprotection). A solution of the above pro-
duct (1 g) in a mixture of dichloromethane and trifluoroacetic
acid (6:1, 35 mL) was stirred for 2 h. The solvent was evap-
orated under vacuum, and the residue was dissolved in ethyl
acetate (100 mL). The mixture was washed consecutively with
50 mL of water (twice) and 50 mL of saturated brine, dried
over Na2SO4, and concentrated under vacuum. The resulting
residue was purified by silica gel flash chromatography.
Method d [1-(3-dimethylaminopropyl)-3-ethyl carbodiimide
hydrochloride (EDC) coupling]. Various amines generated from
method c (1 equiv) were added to a solution of individual short-
chain fatty acids in dry THF (5-10 mmol/mL) under N2,
followed by the addition of EDC (1.3 equiv). After the solution
was stirred overnight, THF was removed under vacuum, and
the residue was dissolved in ethyl acetate (100 mL). The
mixture was washed consecutively with 50 mL of water (twice)
and 50 mL of saturated brine. The organic layer was dried
over Na2SO4 and concentrated under vacuum. The resulting
residue was purified by silica gel flash chromatography.
Method e (hydrogenolysis). The N-benzyloxy derivative re-
sulting from method d was dissolved in 1:1 methanol/THF (5-10
mmol/mL), and 10% palladium on charcoal (10% w/w) was add-
ed. The mixture was treated with hydrogen under atmospher-
ic pressure for 2 h and was filtered. The solvent was evapo-
rated, and the residue was recrystallized with ethyl acetate.
2,2,-Dimethyl-4-phenylbutyric acid. Isobutyric acid (1.4
mL) was added dropwise to a mixture of diisopropylamine (2.2
mL, 0.015 mol) and 54% sodium hydride in mineral oil (0.74
g, 0.0165 mol) in THF (40 mL) and refluxed for 15 min. After
the solution was cooled to 0 °C, a standard solution of
n-butyllithium in heptane (1.45 mmol/mL; 9.4 mL) was added.
After 20 min at 0 °C, the mixture was heated to 30-35 °C for
30 min. After the solution was cooled to 0 °C again, (2-bromo-
ethyl)-benzene (2.8 mL, 15 mmol) was added to the reaction
mixture over 20 min. The ice-bath was retained for 30 min,
the mixture was then heated to 30-35 °C for 1 h, and then 40
acid, [R]25 -59.5° (CHCl3, c 2.13).
D
HTPB: 1H NMR (DMSO-d6) δ 11.02 (s, 1H), 10.1 (s, 1H),
8.94 (s, 1H), 7.67 (m, 4H), 7.27 (m, 5H), 2.63 (t, J ) 7.5 Hz,
2H), 2.35 (t, J ) 7.4 Hz, 2H), 1.87 (m, 2H); HRMS exact mass
of (M + Na)+, 321.120961 amu; observed mass of (M + Na)+,
321.11940 amu. Anal. (C17H18N2O3) C, H, N.
4-Benzoylamino-N-hydroxybenzamide (1): 1H NMR
(DMSO-d6) δ 11.1 (s, 1H), 10.51 (s, 1H), 8.97 (s, 1H), 7.85 (m,
4H), 7.61 (m, 5H); HRMS exact mass of (M + Na)+, 279.074011
amu; observed mass of (M + Na)+, 279.07370 amu. Anal.
(C14H12N2O3) C, H, N.
N-Hydroxy-4-(phenylacetylamino)benzamide (2): 1H
NMR (DMSO-d6) δ 11.1 (s, 1H), 10.40 (s, 1H), 8.94 (s, 1H),
7.67 (m, 4H), 7.33 (m, 5H), 3.67 (s, 2H); HRMS exact mass of
(M + Na)+, 293.089661 amu; observed mass of (M + Na)+,
293.08957 amu. Anal. (C15H14N2O3) C, H, N.
N-Hydroxy-4-(3-phenylpropionylamino)-benzamide (3):
1H NMR (DMSO-d6) δ 11.09 (s, 1 H), 10.17 (s, 1H), 8.94 (s,
1H), 7.67 (m, 4H), 7.27 (m, 5H), 2.92 (t, J ) 7.5 Hz, 2H), 2.63
(t, J ) 7.4 Hz, 2H); HRMS exact mass of (M + Na)+,
307.105311 amu; observed mass of (M + Na)+, 321.10579 amu.
Anal. (C16H16N2O3) C, H, N.
N-Hydroxy-4-(5-phenylpentanoylamino)benzamide (4):
1H NMR (DMSO-d6) δ 11.04 (s, 1H), 10.14 (s, 1H), 8.94 (s, 1H),
7.67 (m, 4H), 7.27 (m, 5H), 2.51 (m, 4H), 1.61 (m, 4H); HRMS
exact mass of (M + Na)+, 321.120961 amu; observed mass of
(M + Na)+, 335.136611 amu. Anal. (C18H20N2O3) C, H, N.
N-Hydroxy-4-[(4-phenylbutyrylamino)methyl]benz-
amide (5): 1H NMR (DMSO-d6) δ 11.2 (s, 1H), 9.0 (s, 1H),
8.4 (t, J ) 5.8 Hz, 1H), 7.7 (d, J ) 8.0 Hz, 2H), 7.21 (m, 7H),
4.3 (d, J ) 5.8 Hz, 2H), 2.58 (t, J ) 7.3 Hz, 2H), 2.18 (t, J )
7.3 Hz, 2H), 1.83 (m, 2H); HRMS exact mass of (M + Na)+,
335.136611 amu; observed mass of (M + Na)+, 335.13716 amu.
Anal. (C18H20N2O3) C, H, N.
[4-(2-Hydroxycarbamoylethyl)phenyl]-4-phenylbutyra-
mide (6): 1H NMR (DMSO-d6) δ 10.36 (s, 1H), 9.80 (s, 1H),
8.70 (s, 1H), 7.52 (d, J ) 8.5 Hz, 2H), 7.2-7.4 (m, 5H), 7.10 (d,
J ) 8.4 Hz, 2H), 2.75 (t, J ) 7.4 Hz, 2H), 2.62 (t, J ) 7.5 Hz,
2H), 2.15-2.4 (m, 4H), 1.8-2.0 (m, 2H); HRMS exact mass of
(M + Na)+, 349.152261 amu; observed mass of (M + Na)+,
349.15223 amu. Anal. (C19H22N2O3) C, H, N.
N-Hydroxy-4-[methyl(4-phenylbutyryl)amino]benza-
1
mide (7): H NMR (DMSO-d6) δ 11.18 (s, 1H), 9.11 (s, 1H),
7.80 (d, J ) 7.5 Hz, 2H), 7.38 (d, J ) 7.5 Hz, 2H), 7.21 (m,
5H), 3.18 (s, 3H), 2.48 (t, J ) 7.5 Hz, 2H), 2.11 (t, J ) 7.4 Hz,
2H), 1.78 (m, 2H); HRMS exact mass of (M + Na)+, 335.136611
amu; observed mass of (M + Na)+, 335.13859 amu. Anal.
(C18H20N2O3) C, H, N.
N-Hydroxy-4-(2-methyl-2-phenylpropionylamino)ben-
zamide (8): 1H NMR (DMSO-d6) δ 11.05 (s, 1H), 9.34 (s, 1H),

Novel Histone Deacetylase Inhibitors
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 17 5535
8.93 (s, 1H), 7.68 (m, 4H), 7.28 (m, 5H), 3.33 (s, 6H); HRMS
exact mass of (M + Na)+, 321.120961 amu; observed mass of
(M + Na)+, 321.12255 amu. Anal. (C17H18N2O3) C, H, N.
4-(2,2-Dimethyl-4-phenylbutyrylamino)-N-hydroxyben-
zamide (9): 1H NMR (DMSO-d6) δ 11.04 (s, 1H), 9.45 (s, 1H),
8.94 (s, 1H), 7.69 (m, 4H), 7.27 (m, 5H), 1.92 (m, 2H), 1.28 (s,
6H); HRMS exact mass of (M + Na)+, 349.152261 amu;
observed mass of (M + Na)+, 349.15187 amu. Anal. (C19H22N2O3)
C, H, N.
the primary antibody, the membrane was washed three times
with TBST for a total of 15 min, washed with goat anti-rabbit
or anti-mouse IgG-horseradish peroxidase conjugates (diluted
1:3000) for 1 h at room temperature, and washed three more
times with TBST for a total of 1 h. The immunoblots were
visualized by enhanced chemiluminescence.
Molecular Docking. The crystal structure of HDLP in
complex with TSA10 was obtained from the Brookhaven Protein
Data Bank (http://www.rcsb.org/pdb/; entry code, 1c3r), the A
form structure of which was used in this study. The three-
dimensional structures of small molecules were generated
using the software SYBYL 6.9 (Tripos Associates, St. Louis,
MO; 2002). Geiger charges were computed, and energy mini-
mization was carried out with default parameters. The
AutoDock Tools software (http://www.scripps.edu/pub/olson-
web/doc/autodock/) was used to designate the rotatable bonds
and generate grid parameter file and docking parameter file
using a default parameter. Zinc ion was modeled by Amber
force field potentials with parameters defined by Stone and
Karplus (r ) 1.1 Å, ꢀ ) 0.25 kcal/mol) and a charge of +2.0 e
as the initial parameters.13 Docking was performed with
AutoDock 3.05 (http://www.scripps.edu/pub/olson-web/doc/
autodock/) which predicts the bound conformations of a small,
flexible ligand to a nonflexible macromolecular target of known
structure. This software is an automated docking package that
combines a rapid grid-based method for energy evaluation with
a Lamarckian genetic algorithm method for conformation
searches.14 The docked complexes for each inhibitor were
selected according to the criteria of interaction energy in
combination with geometrical matching quality. All of the
molecular modeling calculations and manipulations were
performed on Silicon Graphics O2 (Silicon Graphics Inc.,
Mountain View, CA).
(S)-(+)-N-Hydroxy-4-(2-phenylbutyrylamino)benza-
1
mide [(S)-10]: H NMR (DMSO-d6) δ 11.09 (s, 1H), 10.3 (s,
1H), 8.93 (s, 1H), 7.67 (m, 4H), 7.34 (m, 5H), 3.58 (t, J ) 6.5
Hz, 2H), 2.10 (m, 1H), 1.74 (m, 1H), 1.18 (t, J ) 7.3 Hz, 3H);
[R]25D +91.1° (c 1.4, CH3OH); HRMS exact mass of (M + Na)+,
321.120961 amu; observed mass of (M + Na)+, 321.11940 amu.
Anal. (C17H18N2O3) C, H, N.
(R)-(-)-N-Hydroxy-4-(2-phenylbutyrylamino)benza-
1
mide [(R)-10]: H NMR (DMSO-d6) δ 11.09 (s, 1H), 10.3 (s,
1H), 8.93 (s, 1H), 7.67 (m, 4H), 7.34 (m, 5H), 3.58 (t, J ) 6.5
Hz, 2H), 2.10 (m, 1H), 1.74 (m, 1H), 1.18 (t, J ) 7.3 Hz, 3H);
[R]25D -90.9° (c 1.21, CH3OH); HRMS exact mass of (M + Na)+,
321.120961 amu; observed mass of (M + Na)+, 321.11940 amu.
Anal. (C17H18N2O3) C, H, N.
(S)-(+)-N-Hydroxy-4-(3-methyl-2-phenylbutyrylamino)-
benzamide [(S)-11]: 1H NMR (DMSO-d6) δ 11.09 (s, 1H), 10.3
(s, 1H), 8.93 (s, 1H), 7.67 (m, 4H), 7.34 (m, 5H), 2.38 (m, 1H),
1.74 (m, 1H), 1.03 (d, J ) 6.5 Hz, 3H), 0.69 (d, J ) 6.5 Hz,
3H); [R]25D +96.8° (c 1.0, CH3OH); HRMS exact mass of (M +
Na)+, 335.136611 amu; observed mass of (M + Na)+, 335.13710
amu. Anal. (C18H20N2O3) C, H, N.
(R)-(-)-N-Hydroxy-4-(3-methyl-2-phenyl-butyrylamino)-
benzamide [(R)-11]: 1H NMR (DMSO-d6) δ 11.09 (s, 1H), 10.3
(s, 1H), 8.93 (s, 1H), 7.67 (m, 4H), 7.34 (m, 5H), 2.38 (m, 1H),
1.74 (m, 1H), 1.03 (d, J ) 6.5 Hz, 3H), 0.69 (d, J ) 6.5 Hz,
3H); [R]25D -90.6° (c 1.0, CH3OH); HRMS exact mass of (M +
Na)+, 335.136611 amu; observed mass of (M + Na)+, 335.13604
amu. Anal. (C18H20N2O3) C, H, N.
Acknowledgment. This investigation was sup-
ported by the National Institutes of Health Grant
CA94829.
References
In Vitro HDAC Assay. HDAC activity was analyzed by
using a HDAC assay kit (Upstate Biotechnology, Lake Placid,
NY), following the manufacturer’s instruction with slight
modifications. This assay was based on the ability of DU-145
nuclear extract, which is rich in HDAC activity, to mediate
the deacetylation of the biotinylated [3H]-acetyl histone H4
peptide that was bound to streptavidin agarose beads. The
release of [3H]-acetate into the supernatant was measured to
calculate the HDAC activity. Sodium butyrate (0.25-1 mM)
was used as a positive control.
Cell Viability Assay. The effect of the test agents on cell
viability was assessed by the 3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyl-2H-tetrazolium bromide (MTT) assay in 96-well, flat-
bottomed plates, in which 4,000 DU-145 cells/well were seeded.
Cells were exposed to the test agent at the indicated concen-
trations in 10% FBS-supplemented RPMI 1640 medium at 37
°C in 5% CO2 for the indicated time. The medium was removed
and replaced by 150 µL of 0.5 mg/mL of MTT in RPMI 1640
medium, and the cells were incubated in the CO2 incubator
at 37 °C for 2 h. Supernatants were removed from the wells,
and the reduced MTT dye was solubilized with 200 µL/well of
DMSO. Absorbance was determined on a plate reader at 570
nm. Each treatment was repeated in six wells.
(1) Jenuwein, T.; Allis, C. D. Translating the histone code. Science
2001, 293, 1074-1080.
(2) Marks, P.; Rifkind, R. A.; Richon, V. M.; Breslow, R.; Miller, T.;
Kelly, W. K. Histone deacetylases and cancer: causes and
therapies. Nat. Rev. Cancer 2001, 1, 194-202.
(3) Marks, P. A.; Richon, V. M.; Rifkind, R. A. Histone deacetylase
inhibitors: inducers of differentiation or apoptosis of trans-
formed cells. J. Natl. Cancer Inst. 2000, 92, 1210-1216.
(4) Johnstone, R. W. Histone-deacetylase inhibitors: novel drugs
for the treatment of cancer. Nat. Rev. Drug Discovery 2002, 1,
287-299.
(5) Jung, M. Inhibitors of histone deacetylase as new anticancer
agents. Curr. Med. Chem. 2001, 8, 1505-1511.
(6) Grozinger, C. M.; Schreiber, S. L. Deacetylase enzymes: biologi-
cal functions and the use of small-molecule inhibitors. Chem.
Biol. 2002, 9, 3-16.
(7) Miller, T. A.; Witter, D. J.; Belvedere, S. Histone deacetylase
inhibitors. J. Med. Chem. 2003, 46, 5097-5116.
(8) Rosato, R. R.; Grant, S. Histone deacetylase inhibitors in clinical
development. Expert Opin. Invest. Drugs 2004, 13, 21-38.
(9) Marks, P. A.; Miller, T.; Richon, V. M. Histone deacetylases.
Curr. Opin. Pharmacol. 2003, 3, 344-351.
(10) Finnin, M. S.; Donigian, J. R.; Cohen, A.; Richon, V. M.; Rifkind,
R. A.; Marks, P. A.; Breslow, R.; Pavletich, N. P. Structures of
a histone deacetylase homologue bound to the TSA and SAHA
inhibitors. Nature 1999, 401, 188-193.
(11) Lu, Q.; Yang, Y. T.; Chen, C. S.; Davis, M.; Byrd, J. C.; Etherton,
M.; Umar, A.; Chen, C. S. Zn2+-chelating motif-tethered short-
chain fatty acids as a novel class of histone deacetylase inhibi-
tors. J. Med. Chem. 2004, 47, 467-474.
(12) Vanommeslaeghe, K.; Van Alsenoy, C.; De Profit, F.; Martins,
J. C.; Tourwe, D.; Geerlings, P. Ab intio study of the binding of
Trichostatin A (TSA) in the active site of histone deacetylase
like protein (HDLP). Org. Biomol. Chem. 2003, 1, 2951-2957.
(13) Stote, R. H.; Karplus, M. Zinc binding in proteins and solution:
a simple but accurate nonbonded representation. Proteins:
Struct., Funct., Genet. 1995, 23, 12-31.
(14) Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart,
W. E.; Belew, R. K.; Olson, A. L. Automated docking using a
Lamarckian genetic algorithm and empirical binding free energy
function. J. Comput. Chem. 1998, 19, 1639-1662.
Western Blot Analysis. DU-145 cells (1 × 106) treated with
the test agent at the indicated concentrations in 10% FBS-
supplemented RPMI 1640 medium for 48 h were collected and
sonicated. Protein concentrations of the lysates were deter-
mined by using a Bradford protein assay kit (Bio-Rad, Her-
cules, CA); equivalent amounts of proteins from each lysate
were resolved in 10% SDS-polyacrylamide gel, and then
transferred onto Immobilon-nitrocellulose membranes (Mil-
lipore, Bellerica, MA) in a semi-dry transfer cell. The trans-
blotted membrane was washed twice with tris-buffered saline
containing 0.1% Tween 20 (TBST). After being blocked with
TBST containing 5% nonfat milk for 40 min, the membrane
was incubated with the primary antibody (1:3000 dilution) in
TBST/1% nonfat milk at 4 °C overnight. After treatment with
JM0503749