N-Hydroxy-1,2-disubstituted-1H-benzimidazol-5-yl acrylamides as novel histone deacetylase inhibitors: Design, synthesis, SAR studies, and in vivo antitumor activity

Article abstract of DOI:10.1016/j.bmcl.2009.01.041

A series of N-hydroxy-1,2-disubstituted-1H-benzimidazol-5-yl acrylamides were designed and synthesized as novel HDAC inhibitors. General SAR has been established for the substituents at positions 1 and 2, as well as the importance of the ethylene group an

Full text of DOI:10.1016/j.bmcl.2009.01.041

Bioorganic & Medicinal Chemistry Letters 19 (2009) 1403–1408
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
N-Hydroxy-1,2-disubstituted-1H-benzimidazol-5-yl acrylamides as novel
histone deacetylase inhibitors: Design, synthesis, SAR studies,
and in vivo antitumor activity
*
Haishan Wang , Niefang Yu, Hongyan Song, Dizhong Chen, Yong Zou, Weiping Deng, Pek Ling Lye,
Joyce Chang, Melvin Ng, Stéphanie Blanchard, Eric T. Sun, Kanda Sangthongpitag, Xukun Wang,
Kee Chuan Goh, Xiaofeng Wu, Hwee Hoon Khng, Lijuan Fang, Siok Kun Goh, Wai Chung Ong,
Zahid Bonday, Walter Stünkel, Anders Poulsen, Michael Entzeroth
S*BIO Pte Ltd, 1 Science Park Road, #05-09 The Capricorn, Singapore Science Park II, Singapore 117528, Singapore
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of N-hydroxy-1,2-disubstituted-1H-benzimidazol-5-yl acrylamides were designed and synthe-
sized as novel HDAC inhibitors. General SAR has been established for the substituents at positions 1
and 2, as well as the importance of the ethylene group and its attachment to position 5. Optimized com-
pounds are much more potent than SAHA in both enzymatic and cellular assays. A representative com-
pound, 23 (SB639), has demonstrated antitumor activity in a colon cancer xenograft model.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 25 November 2008
Revised 9 January 2009
Accepted 13 January 2009
Available online 19 January 2009
Keywords:
HDAC inhibitor
Benzimidazole
Hydroxamic acid
SAR
Anticancer
The amino termini of histones extend from the nucleosomal
core and are modified by histone acetyltransferases (HATs) and
histone deacetylases (HDACs) during the cell cycle.1 Acetylation
particularly HDACs 1 and 3, are essential to the proliferation and
survival of mammalian carcinoma cells.7
There are a number of HDACi which are currently undergoing
clinical trials either as a single therapy or in combination for
treatment of solid and hematologic malignancies (Fig. 1). The mor-
of the e-amino groups of lysine residues removes positive charges,
thereby reducing the affinity between histones and negatively
charged DNA. This facilitates access of the transcriptional machin-
ery to the DNA template. Therefore, in most cases, histone acetyla-
tion enhances transcription while histone deacetylation represses
transcription.2,3 Inhibition of HDACs results in the accumulation
of acetylated histones, which results in a variety of cell type depen-
dent responses, such as apoptosis, necrosis, differentiation, cell
survival, inhibition of proliferation and cytostasis. HDAC inhibitors
(HDACi) have been studied for their therapeutic effects on cancer
cells.4 Suberoylanilide hydroxamic acid (SAHA, Vorinostat), the
first HDACi approved by the FDA in 2006 for the treatment of cuta-
neous T-cell lymphoma, validates HDAC inhibition as a strategy for
cancer therapy.5
HDAC enzymes are divided into four different classes.6 Class I
and II isozymes, specifically HDAC 1, 2, 3, 6 and 8, have been asso-
ciated with uncontrolled tumor growth. For example, selective
knockdown studies on HDACs suggest that the class I HDACs,
* Corresponding author. Tel.: +65 68275019; fax: +65 68275005.
E-mail address: haishan_wang@sbio.com (H. Wang).
Figure 1. Examples of clinically tested HDAC inhibitors.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.01.041

1404
H. Wang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1403–1408
phology of the HDACi pharmacophore is characterized by three
portions: cap, linker and zinc-binding group (ZBG) as exemplified
by SAHA (Fig. 1). The common linkers are aliphatic chain (e.g.,
six carbon chain in SAHA), aromatic ring (e.g., phenylene in MS-
275)8 and vinyl-aromatic (e.g., styryl in PXD101).9 The most com-
mon ZBGs are hydroxamic acid and benzamide (e.g., MS-275).
Hydroxamic acid remains among the most potent and popular
ZBGs reported for inhibition of Class I HDACs.
As part of our ongoing effort to discover novel anticancer
agents, we have examined a number of fused heterocyclic rings
such as benzimidazole, benzothiophene, indole and indazole as
part of a novel linker or building block for a low molecular weight
novel HDACi. In this report, we describe the development of N-hy-
droxy-1,2-disubstituted-1H-benzimidazol-5-yl acrylamides (VII)
(Scheme 1), a series of novel HDACi with promising pharmacolog-
ical and pharmacokinetic properties.
The X-ray structures of HDLP (histone deacetylase like protein)
co-crystallized with SAHA and trichostatin A indicate that a linker
incorporating a pi-system is preferred for interaction with two
phenyl groups in the binding pocket (Phe198 and Phe141).10 Dock-
ing studies revealed that vinyl-aromatic rings are suitable link-
ers.11 We envisaged that VII, with appropriately substituted
groups R1 and R2, would have additional interactions with protein
residues in the ‘cap’ region.
An efficient synthesis of a wide range of benzimidazole-based
hydroxamic acids was developed. Scheme 1, Route A, illustrates
the general procedure used for preparing hydroxamic acids (VII),
starting from trans-3-nitro-4-chloro-cinnamic acid (I) in a 4-step
Scheme 2. Reagents and conditions: (a) H2 (1 atm)/10% Pd–C, rt, (b) NH2OHÁHCl
(10 equiv)/NaOMe (20 equiv)/MeOH, 0 °C to rt; (c) (CH3)3S(O)I, NaH, THF, rt, Et2O;
(d) R1X (X = I, Br), K2CO3, CH3CN.
reaction sequence. Chloride displacement by amine R1NH2 (II)
under basic conditions, followed by esterification of the acid (III)
gave methyl ester (IV). Alternatively, I may be esterified to the
methyl ester (Ia) first, followed by installation of amine (II) to give
IV. One-pot reductive cyclization of IV with aldehyde (V) forms the
benzimidazole ring.12 In this key step, the nitro group of IV was
reduced by tin (II) chloride in acetic acid, and in the presence of
aldehyde (V) ring cyclization was achieved to give benzimidazole
(VI) in 40–65% yield. Treatment of the methyl ester (VI) with
excessive hydroxylamine gave hydroxamates (VII) in good yield.13
If the desired aldehyde (V) was not readily available, acid (IX) was
used for formation of amide (X) first, then cyclized to benzimid-
14
azole (VI) under acidic conditions
(Scheme 1, Route B). VII
(R2 = H or Me) can be obtained by refluxing formic acid or acetic
acid, respectively, with diamine (VIII) (Scheme 1, Route C).
To assess the role of the olefinic linker between the zinc-binding
hydroxamic acid and the fused aromatic ring, the saturated alkyl
linker analog (XI) and a cyclopropyl analog (XII) were prepared
as illustrated in Scheme 2. Alkylation of VI (R1 = H) with R1X
(e.g., benzyl bromide, MeI) provided two products after separation:
one was identical to an authentic sample (VII) prepared as in
Scheme 1, the other was the isomer XIII.
Initially synthesized analogs, with R1 fixed as a propanol group,
were profiled against the HDAC1 enzyme and the human colon
cancer cell line COLO205 (Table 1).15,16 The propanol group was
selected as R1 with the expectation of possible interactions
between it and the hydrophilic side-chains in HDAC1, for example,
Asp 99 or Glu 98.10 R2 was optimized with a range of groups:
unsubstituted (1), aromatic and non-aromatic rings (2 and 3),
straight and branched alkyl chains (4, 5 and 6) and arylalkyl chains
(7, 8 and 9). Compounds with a directly attached ring or an
a-
branch (2, 3 and 4) were not potent, but those with -unsubstitut-
a
ed phenylalkyl chains (7, 8 and 9) were not only potent in the
HDAC1 enzyme assay, but also in a cellular anti-proliferation assay
using colon cancer cell line COLO 205.
To investigate the role of the R1 substituent, a series of
compounds were synthesized in which R2 was held constant as
phenethyl while varying R1. Phenethyl was selected as R1 substitu-
ent because it has reasonable contribution to in vitro potency, is
simpler and achiral compared to 2-phenyl-propyl in 8. A number
of R1 substituents were explored for size (both linear and bulky),
lipophilicity and hydrophilicity (neutral, acidic and basic) and
combinations of the above (Table 2). Hydrophobic and bulky
groups are preferred for R1 with a short flexible linker (m = 1, 13
vs 12 and 11). For R1 substituents with a longer flexible linker
Scheme 1. Reagents and conditions: (a) Et3 N, dioxane, 100 °C; (b) MeOH,H2SO4,
60 °C; (c) SnCl2Á2H2O (5 equiv), AcOH/MeOH (1:9), 40 °C; (d) NH2OHÁHCl
(10 equiv)/NaOMe (20 equiv)/MeOH, 0 °C to rt; (e) SnCl2Á2H2O (5 equiv), AcOH/
MeOH (1:9), 40 °C; (f) coupling reagent (EDCI), HOBt; (g) HOAc, reflux; (h) HCO2H
or CH3CO2H, reflux.

H. Wang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1403–1408
1405
Table 1
Table 3
SAR of R2 for VII with R1 = –(CH2)3–OH fixed
SAR of R1 = amines for VII with R2 = –(CH2)2–Ph fixed
Compound
R1
HDAC1 IC50
(lM)
COLO 205 IC50
(lM)
a
b
R2
H
HDAC1 IC50
1.10 0.14
(
lM)
COLO 205 IC50
Not tested
(lM)
a
b
Compound
22
0.052 0.025
0.035 0.016
0.026 0.014
0.31 0.21
0.22 0.12
1
2
3
3.2c
1.9c
31c
31c
23 (SB639)
0.14 0.05
0.09 0.06
0.61 0.26
>100
24
25
26
27
28
4
5
1.70 0.36
0.10 0.05
Not tested
26 17
0.16 0.06
6
7
8
0.53c
0.17c
3.7 0.1
3.2 0.1
0.19 0.01
2.0 1.1
0.041 0.019
0.16 0.04
0.12 0.07
1.1 0.2
0.08 0.04
0.17c
1.7 0.6
4.1 1.9
9
29
a
Values are expressed as mean standard deviation of at least two independent
duplicate experiments.
a,b
See footnote of Table 1.
b
Values are expressed as mean standard deviation of at least two independent
triplicate experiments.
Hence, a series of compounds with basic R1 side-chains were
synthesized (Table 3). Both enzymatic and cellular activities were
significantly improved with a mono basic amine with a two carbon
linker (22, 23 and 24). Compounds with longer linkers of three car-
bons and more were less potent (23 vs 25, 27 and 29). Increasing
lipophilicity of the amine substituent improved potency slightly
(24 vs 23). Docking compound 23 into the HDAC1 homology mod-
el17,18 revealed a key electrostatic interaction between the R1 basic
center and Asp99 (Fig. 2). There is an additional hydrogen bonding
interaction between the hydroxyl of Tyr204 and benzimidazole N3,
thus both basic centers contribute to the binding of the most po-
tent compounds, such as 22, 23 and 24. Compounds with longer
linkers (25, 26, 27 and 29), due to their increased distance between
R1 basic center and Asp99 according to the binding mode of 23,
have to fold back their R1 side-chains in order to make similar
interactions with Asp99, but the overall bindings become less
effective. However, 28, has a bulky gem-dimethyl group in R1
c
Values obtained from only one duplicate or triplicate experiment.
(m = 2 and 3), none of the acidic (15), cyclic amide (20) or lipophilic
groups (17) enhanced HDAC1 or cell proliferation inhibition. Even
though potencies of compounds with linear chains (7, 14, 18 and
19) were comparable to those with basic chains (16 and 21), the
basic R1 side-chains were chosen for further optimization due to
their higher aqueous solubility.
Table 2
SAR of R1 = –(CH2)m–Y for VII with R2 = –(CH2)2–Ph fixed
a
b
Compound
m
Y
HDAC1 IC50
(
lM)
COLO 205 IC50 (lM)
10
11
12
0
1
1
H
H
Ph
0.17 0.01
0.26 0.14
0.41 0.06
4.4 2.2
1.2 0.6
1.6 0.8
13
1
0.047 0.013
0.32 0.21
14
15
16
17
18
7
2
2
2
2
3
3
3
H
CO2H
Morpholin-4-yl
Ph
H
OH
OCH3
0.22 0.05
2.3 0.4
0.27 0.07
0.88c
1.5 0.5
Not tested
2.8 0.3
8.8 0.8
2.3 0.5
3.2 0.1
2.2 0.3
0.17 0.07
0.17c
19
0.21c
20
3
0.20 0.02
0.25 0.06
11
4
Figure 2. Compound 23 (SB639) docked into the HDAC1 homology model. Key
interactions between the basic centers with Asp99 and Tyr204 contribute to the
potency.
21
3
Morpholin-4-yl
1.9 0.9
a,b,c
See footnote of Table 1.

1406
H. Wang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1403–1408
Table 4
SAR of Z for XIV
R1
R2
Z
HDAC1 IC50
(l
M)
COLO 205 IC50 (lM)
a
b
Compound
5
30
–CH@CH–
–CH2CH2–
0.10 0.05
>10
26 17
Not tested
7
31
–CH@CH–
– (bond)
0.17c
3.3c
3.2 0.1
26
1
13
32
–CH@CH–
–CH2CH2–
0.047 0.013
0.17 0.00
0.32 0.21
2.4 0.5
11
33
CH3
–CH@CH–
0.26 0.14
1.3 1.0
1.2 0.6
Not tested
a,b,c
See footnote of Table 1.
side-chain which can force it to adopt favorable conformation(s) to
make effective interaction with Asp99 explaining its high potency.
Assessment of the role of the linker between the zinc-binding
hydroxamic acid and the benzimidazole ring was explored with
compounds made according to Scheme 2 (biological data listed in
Table 4). Saturation of the double bond either reduced HDAC1
potency 3.6-fold (32 vs 13) or destroyed it completely (30 vs 5).
Clearly the nature of specific side-chains in combination with the
Z linker is crucially important. The cyclopropyl moiety, as a
replacement for ethylene (33) lost 5-fold potency perhaps due to
an unfavorable hydrophobic interaction or conformational change.
When the hydroxamic acid is directly attached to the benzimid-
azole ring (31, which was made by using method analogous to
Route A, Scheme 1 and 4-chloro-3-nitro-benzoic acid methyl ester
as starting material instead of I), inhibition of HDAC1 enzyme
activity was reduced by over an order of magnitude (31 vs 7).
The position of attachment of the alkenoyl hydroxamate to the
benzimidazole ring at either position 5 or 6 was explored with the
pair of compounds 34 (Table 5) and 12. The 5-substituted com-
pound 12 is more potent against HDAC1. Illustrating the superior
Table 5
Isomeric benzimidazole and benzamides
a
a
Compound
Structures
HDAC1 IC50
0.41 0.06
(
lM)
COLO 205 IC50
1.6 0.8
(lM)
12
34
35
1.2 0.2
0.25 0.25
>100
5.4 0.5
0.88 0.46
5.1 2.6
36
a,b
See footnote of Table 1.

H. Wang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1403–1408
1407
Table 6
Cellular IC50
(lM) of selected compounds
Compound
HCT116
A2780
PC3
11
22
23
24
28
1.3 0.3
0.73 0.04
0.10 0.00
0.19 0.14
0.031 0.000
0.11 0.00
0.89 0.23
0.20 0.09
0.15 0.08
0.12 0.04
0.18 0.08
0.24 0.09
0.20 0.10
0.10 0.03
0.16 0.10
Table 7
Half-life T1/2 (min) in liver microsomal stability assay19
Compound
Human
>30
Mouse
11
22
23
24
28
23
3
16
5
2.8 1.7
2.7 2.0
1.3 0.5
1.0 0.0
Figure 3. Antitumor activity of 23 in mouse HCT116 xenograft model
>30
10
12
1
1
Table 8
SB639 (23) is a pan-HDAC inhibitor
HDACa
Kib (nM)
HDACa
Kib (nM)
potency of the hydroxamate ZBG, 35 is a much more potent HDAC1
inhibitor than the corresponding benzamide (36). The above re-
sults clearly demonstrate that the properties of the side-chains
and their positions of substitution on the benzimidazole ring are
crucial for optimal enzyme and cellular potency.
1
2
3
4
5
11.6 0.3
36.2 5.3
12.5 1.7
9.6 0.4
6
8
9
10
11
20.6 0.8
82
9
19.2 1.1
18.9 1.1
12.8 2.5
The most promising compounds representing different R1 side-
chains were selected for further evaluation in cell lines and in vivo.
Broad activity against other human tumor cell lines, such as
HCT116, A2780 and PC3 was encouraging (Table 6).16
17.2 2.9
a
HDAC isoenzymes (HDAC7 not tested).
Dissociation constant.15
b
Given the good cellular activity in vivo xenograft studies were
initiated in HCT116 tumor-bearing nude mice (10 mice per dose
group). Considering their low metabolic stability in mouse (Table
7), compounds were given daily ip at doses up to MTD (maximum
tolerated dose).20 22, 23, 24 and 28 were dosed as the di-hydro-
chloride salt and 11 as the mono-hydrochloride. 23 demonstrated
significant antitumor activities compared to vehicle control, tumor
growth inhibition (TGI) on day 22 was 89% (130 mg/kg, MTD,
p < 0.01, maximum body weight loss (MBWL): 8.7% on day 19)
and 59% (70 mg/kg, p < 0.01, MBWL: 0.5% on day 15) (Fig. 3). SAHA
potent pan-HDAC inhibitor, it inhibits class I, II and IV HDAC isoen-
zymes (Table 8). The basic nitrogens in 23 are not only essential for
HDAC inhibition, but also for physicochemical properties. The
hydrochloride salt of 23 has good solubility in mildly acidic media
(1.6 mg/mL at pH 6.5) and lipophilicity at neutral pH in the ideal
range (logD = 2.1 at pH 7.3).
In conclusion, we have designed and synthesized a series of
N-hydroxy-1,2-disubstituted-1H-benzimidazol-5-yl acrylamides,
which exhibit potent enzymatic and cellular activity. A representa-
tive compound, 23 (SB639), exhibited excellent antitumor efficacy
in an HCT116 xenograft model. Compound 23 was also orally
active demonstrating the good drug-like properties from this tem-
plate. Future reports will build on the SAR of the benzimidazole
core structure with a view to delivering a clinical candidate.
(IC50 = 0.12 0.04 and 2.9 1.6 lM for HDAC1 and HCT116,
respectively) was used as a positive control and showed moderate
activity at 200 mg/kg (MTD) with TGI = 53% (p < 0.01).
The antitumor activities of 11, 22 and 24 were not significantly
different from vehicle controls (p > 0.05) in other experiments. The
TGIs were 23% (50 mg/kg, qdx21, MTD) and 32% (75 mg/kg, qdx21,
MTD) on day 21 for 11 and 24, respectively. The TGI on day 22 was
12% for 22 (60 mg/kg, qdx21, MTD). Compound 28 was not toler-
ated in nude mice even at the lowest dose tested (100 mg/kg),
treatment ended on day 12, 2/10 and 1/10 treatment-related
deaths occurred on days 13 and 14, respectively. The TGI on day
11 was 60% (100 mg/kg, p > 0.05), no further evaluation was done
at lower doses.
Acknowledgments
We thank the S*BIO sample management team for preparing
sample solutions for assays, the PKDM department for DMPK stud-
ies on 23,19 and Dr. Brian W. Dymock for critical review of the
manuscript.
References and notes
Compound 23 was also given orally to tumor-bearing mice,
with TGI on day 21 of 78% (200 mg/kg, qdx21, p < 0.001, MBWL:
5.7% on day 7) and 43% (100 mg/kg, qdx21, p < 0.05, no body
weight loss). Pharmacokinetic studies have determined that com-
pound 23 has an oral bioavailability in mouse of 13%.19 When
1. Grunstein, M. Nature 1997, 389, 349.
2. Struhl, K. Genes Dev. 1998, 12, 599.
3. Wade, P. A. Hum. Mol. Genet. 2001, 10, 693.
4. Butler, L. M.; Agus, D. B.; Scher, H. I.; Higgins, B.; Rose, A.; Cordon-Cardo, C.;
Thaler, H. T.; Rifkind, R. A.; Marks, P. A.; Richon, V. M. Cancer Res. 2000, 60,
5165.
dosed at 50 mg/kg (po), the Cmax (0.92 0.05
above the IC50 levels for HDAC1 (0.035 0.016
l
M),19 was well
lM) and HCT116
5. ZolinzaTM (SAHA, Vorinostat) was approved for the treatment of patients with
cutaneous T-cell lymphoma who have progressive, persistent or recurrent
disease on or following two systemic therapies, see FDA approval documents:
http://www.fda.gov/cder/foi/nda/2006/021991s000_ZolinzaTOC.htm.
6. (a) De Ruijter, A. J. M.; van Gennip, A. H.; Caron, H. N.; Kemp, S.; van Kuilenburg,
A. B. P. Biochem. J. 2003, 370, 737; (b) Michan, S.; Sinclair, D. Biochem. J. 2007,
404, 1.
(0.20 0.10 lM). Plasma concentration was maintained above
the HDAC IC50 for about 4 h. Given that the doses in the HCT116
xenograft experiment were much higher than this PK study it is
not surprising that such good efficacy was seen.
The mechanism of action of 23 was confirmed both in vitro and
in vivo as evidenced by hyperacetylation of H3 in HCT116 cells and
tumor tissues after treatment with 23 (data not shown). 23 is also a
7. (a) Curtin, M.; Glaser, K. Curr. Med. Chem. 2003, 10, 2373; (b) Glaser, K. B.; Li, J.;
Staver, M. J.; Wei, R.-Q.; Albert, D. H.; Davidsen, S. K. Biochem. Biophys. Res.
Commun. 2003, 310, 529.

1408
H. Wang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1403–1408
8. (a) Saito, A.; Yamashita, T.; Mariko, Y.; Nosaka, Y.; Tsuchiya, K.; Ando, T.;
Suzuki, T.; Tsuruo, T.; Nakanishi, O. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 4592;
(b) Suzuki, T.; Ando, T.; Tsuchiya, K.; Fukasawa, N.; Saito, A.; Mariko, Y.;
Yamashita, T.; Nakanishi, O. J. Med. Chem. 1999, 42, 3001.
proliferation assay (Promega Pte Ltd, Singapore). Dose–response curves were
plotted to determine the IC50 values using XL-fit (ID business solution Ltd,
USA). IC50 is the concentration needed for inhibition of 50% cell proliferation of
tumor cells.
9. Plumb, J. A.; Finn, P. W.; Williams, R. J.; Bandara, M. J.; Romero, M. R.; Watkins,
C. J.; La Tangue, N. B.; Brown, R. Mol. Cancer Ther. 2003, 2, 721.
10. Finnin, M. S.; Donigian, J. R.; Cohen, A.; Richon, V. M.; Rifkind, R. A.; Marks, P.
A.; Breslow, R.; Pavletich, N. P. Nature 1999, 401, 188.
17. Homology Modeling. The human HDAC1 protein sequence was downloaded
from ExPASy (http://expasy.org). The Aquifex aeolicus Histone Deacetylase Like
Protein (HDLP) sequence was downloaded from the Protein Data Bank (PDB,
http://rcsb.org). The proteins were aligned using clustal W 1.8318a and the
alignment manually edited. The secondary structure was predicted by
Psipred18b and SSPro.18c The HDLP X-ray structure 1C3R10 was downloaded
from the PDB and used as a template. The homology model was built using
Prime18d with standard settings. The homology model was finally subjected to
5000 steps of Steepest Decent minimization using MacroModel18d with OPLS-
AA force field18e and the GB/SA solvation model.18f The HDAC1 homology
model were prepared using the protein preparation wizard in Maestro18d with
standard settings. Grids were generated using Glide version 4.0.21718d
11. Preliminary docking studies were done using HDLP X-ray structures.10 See
Refs. 17 and 18 for details of similar software and procedure used for HDAC1
homology modeling.
12. Wu, Z.; Rea, P.; Wickham, G. Tetrahedron Lett. 2000, 41, 9871.
13. Remiszewski, S. W.; Sambucetti, L. C.; Atadja, P.; Bair, K. W.; Cornell, W. D.;
Green, M. A.; Howell, K. L.; Jung, M.; Kwon, P.; Trogani, N.; Walker, H. J. Med.
Chem. 2002, 45, 753.
14. Baudy, R. B.; Fletcher, H., III; Yardley, J. P.; Zaleska, M. M.; Bramlett, D. R.; Tasse,
R. P.; Kowal, D. M.; Katz, A. H.; Moyer, J. A.; Abou-Gharbia, M. J. Med. Chem.
2001, 44, 1516.
15. The recombinant HDAC enzymes, including HDAC1, 2, 3, 4, 5, 6, 7, 8, 9, 10 and
11, were produced by the Protein Biochemistry Group in S*BIO Pte Ltd. The
assay was performed in a 96-well format using the BIOMOL fluorescent-based
HDAC activity assay (BioMol International, L.P). The reaction mixture was
composed of assay buffer, containing 25 mM Tris, pH 7.5, 137 mM NaCl,
2.7 mM KCl, 1 mM MgCl2, 1 mg/ml BSA, tested compounds, an appropriate
following the standard procedure recommended by Schrödinger.
A metal
constraint was included in the grid files in order to force the docked poses to
form an interaction with the catalytic zinc.
18. (a) Chenna, R.; Sugawara, H.; Koike, T.; Lopez, R.; Gibson, T. J.; Higgins, D. G.;
Thompson, J. D. Nucleic Acids Res. 2003, 31, 3497; (b) Jones, D. T. J. Mol. Biol.
1999, 292, 195; (c) Pollastri, G.; Przybylski, D.; Rost, B.; Baldi, P. Proteins 2002,
47, 228; (d) Prime, MacroModel, Glide and Maestro are products of
Schrödinger, LLC, New York, NY (www.schrodinger.com).; (e) Kaminski, G. A.;
Friesner, R. A.; Tirado-Rives, J.; Jorgensen, W. L. J. Phys. Chem. B 2001, 105, 6474;
(f) Still, W. C.; Tempczyk, A.; Hawley, R. C.; Hendrickson, T. J. Am. Chem. Soc.
1990, 112, 6127.
19. For in vitro metabolic studies and pharmacokinetic data of compound 23
(SB639), see Venkatesh, P. R.; Goh, E.; Zeng, P.; New, L. S.; Xin, L.; Pasha,
M. K.; Sangthongpitag, K.; Yeo, P.; Kantharaj, E. Biol. Pharm. Bull. 2007, 30,
1021.
20. Female athymic nude mice, 10–12 weeks of age, were implanted
subcutaneously in the flank with 5 Â 106 HCT116 cells. When the tumor
reached a size of 100 mm3, the mice were pair-matched prior to treatment.
Tumor size was measured every second day and the tumor volume calculated
as follows: tumor volume (mm3) = (width)2 Â length/2. Compounds were
dissolved in 10% N,N-dimethylacetamide (DMA)/10% Cremophore/80% water
for ip route or 0.5% methylcellulose/0.1%Tween 80 for oral administration.
Drugs were injected ip or orally administered via gavage everyday for a period
of 21 days. Tumor growth inhibition (TGI) was calculated according to %
TGI = (Ct À Tt) * 100/(Ct À Ct1) where Ct, the mean tumor size of the vehicle
control group at time t; Tt, the mean tumor size of the treatment group at the
time t; Ct1 = the mean tumor size of the vehicle control group on the first day of
treatment. Statistic analysis was done with one-way ANOVA followed by
Dunnett’s Multiple Comparison test (Prism 4.0).
concentration of HDAC1 and 250 lM Flur de lys generic substrate. The
fluorescence was detected at the excitation wavelength 360 nm and emission
wavelength 460 nm using Tecan Ultra Microplate Detection System (Tecan
Group Ltd, Switzerland). The analytical software, Prism 4.0 (GraphPad
Software, Inc.) was used to generate IC50 values from the data. IC50s for
HDACs 2–11 were obtained by using analogous protocols but with appropriate
adjustment of protein and/or substrate concentrations. The dissociation
constant for binding of inhibitor to enzyme, the Ki, was calculated using the
Cheng–Prusoff equation: Ki = IC50/(1 + ([substrate]/Km)).
16. Human colon cancer cell lines COLO 205, HCT116 and prostate cancer cell line
PC3 were obtained from the American Type Culture Collection (ATCC; Virginia,
USA). Human ovarian cancer cell line A2780 was obtained from the European
Collection of Cell Culture (ECACC, Wilshire, UK). They were cultivated
according to the supplier’s instructions. After sub-confluent growth, cells
were seeded in 96-wells plate at log growth phase. Before treatment, the plates
seeded with solid tumor cells were incubated at 37 °C, 5% CO2 for 24 h, while
the plates seeded with liquid tumor cell lines were incubated at 37 °C, 5% CO2
for 2 h. Treatment with compounds was carried out in triplicate wells for 96 h.
Proliferation of solid tumor cells was monitored using CyquantTM cell
proliferation assay (Invitrogen Pte Ltd, Singapore), while proliferation of
liquid tumor cells was monitored using CellTiter96 Aqueous one solution cell