Optimization of biaryl Selective HDAC1&2 Inhibitors (SHI-1:2)

Article abstract of DOI:10.1016/j.bmcl.2007.11.047

A class of biaryl benzamides was identified and optimized as selective HDAC1&2 inhibitors (SHI-1:2). These agents exhibit selectivity over class II HDACs 4-7, as well as class I HDACs 3 and 8; providing examples of selective HDAC inhibitors for the HDAC i

Full text of DOI:10.1016/j.bmcl.2007.11.047

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 726–731
Optimization of biaryl Selective HDAC1&2 Inhibitors (SHI-1:2)
David J. Witter,a,* Paul Harrington,a, Kevin J. Wilson,a
Melissa Chenard,c Judith C. Fleming,b Brian Haines,c Astrid M. Kral,b
J. Paul Secristb and Thomas A. Millera
aDepartment of Drug Design and Optimization—Medicinal Chemistry, Merck Research Laboratories,
33 Avenue Louis Pasteur, Boston, MA 02115, USA
bDepartment of Cancer Biology and Therapeutics, Merck Research Laboratories, 33 Avenue Louis Pasteur, Boston, MA 02115, USA
cDepartment of Pharmacology, Merck Research Laboratories, 33 Avenue Louis Pasteur, Boston, MA 02115, USA
Received 18 October 2007; revised 12 November 2007; accepted 14 November 2007
Available online 19 November 2007
Abstract—A class of biaryl benzamides was identified and optimized as selective HDAC1&2 inhibitors (SHI-1:2). These agents exhi-
bit selectivity over class II HDACs 4–7, as well as class I HDACs 3 and 8; providing examples of selective HDAC inhibitors for the
HDAC isoforms most closely associated with cancer. The hypothesis for the increased selectivity is the binding of a pendant aro-
matic group in the internal cavity of the HDAC1&2 enzymes. SAR development based on an initial lead led to a series of potent and
selective inhibitors with reduced off-target activity and tumor growth inhibition activity in a HCT-116 xenograft model.
Ó 2007 Elsevier Ltd. All rights reserved.
HDAC1 and HDAC2 are class I histone deacetylase
(HDAC) enzymes and are members of a broader family
of 11 zinc-dependent HDAC enzymes, which are emerg-
ing therapeutic targets for the treatment of cancer and
other diseases.1 These enzymes, as part of multi-protein
complexes, catalyze the removal of acetyl groups from
lysine residues on proteins, including histones, affecting
gene expression, differentiation, growth arrest, and/or
apoptosis in transformed cell cultures.2
mounting evidence for the involvement of HDAC1
and HDAC2 in cancer suggests that inhibitors selective
for these subtypes may demonstrate an improved thera-
peutic index through enhanced clinical efficacy and/or
better tolerability compared to pan HDAC inhibitors.
Since all current clinical HDAC inhibitors appear to in-
hibit HDAC1, HDAC2, and HDAC3, as well as other
HDAC subtypes,5 our program sought to identify selec-
tive inhibitors that target HDAC1 and HDAC2
preferentially.6
While the biological functions of the various HDAC
subtypes are only partially understood, HDAC1 and
HDAC2 have documented roles in regulating cell prolif-
eration and represent compelling targets for cancer ther-
apeutics.3,4 RNAi-mediated knockdown of HDAC1
expression inhibits proliferation and, importantly, in-
duces apoptosis in several tumor cell lines in vitro.3
Mutations and inactivations of histone acetyltransfer-
ases (HATs) and overexpression of HDAC1 and
HDAC2 have been observed in certain cancers.4 The
A wide range of structures have been shown to inhibit
the activity of class I/II HDAC enzymes and, with few
exceptions, these can be broadly characterized by a com-
mon pharmacophore7 comprised of a metal binding do-
main, a linker domain, and a surface-recognition
domain, exemplified by ZolinzaÒ (SAHA, vorinostat),
which was recently approved for the treatment of cuta-
neous manifestations of cutaneous T-cell lymphoma
(Fig. 1).8 Initial SHI-1:2 design considerations included
minimizing strong metal binding moieties (e.g.,
hydroxamic acids) to decrease the potential for general
metal binding related ‘promiscuity’ and leveraging avail-
able structural data to guide the development of com-
pounds with enhanced potency and selectivity. For
example, the presence of an extensive internal cavity
near the active site of HDAC family members suggests
Keywords: Histone deacetylase; HDAC; Aminobenzamides; Biaryl.
*
Corresponding author. Tel.: +1 617 992 2067; fax: +1 617 992
2403; e-mail: David_Witter@Merck.com
Present address: Department of Medicinal Chemistry, Amgen Inc.,
One Amgen Center Drive, Thousand Oaks, CA 91320-1799, USA.
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.11.047

D. J. Witter et al. / Bioorg. Med. Chem. Lett. 18 (2008) 726–731
727
O
protected and arylated with phenyl boronic acid using
Suzuki cross-coupling. The resultant nitro-biaryl 4 was
hydrogenated to the mono Boc-protected dianiline 5.
The acid 8 was generated from CDI coupling of 3-pyri-
dinemethanol (6) and 4-(aminomethyl)benzoic acid (7).
Coupling conditions with BOP led to the Boc-protected
9a, which was deprotected with TFA to generate the de-
sired amino-biaryl 2a.10
H
N
OH
N
HDAC1 IC50 (nM) 30
HDAC3 IC50 (nM) 100
H
Zolinza®
O
Surface
Recognition
Metal
Binding
Linker
OH
O
N
H
N
S
O
NH2
H
H
O
N
N
The extent of subtype selectivity of SHI-1:2 2a across
HDAC class I and II family members was measured
in vitro by monitoring the deacetylation of a synthetic
acetyl-lysine bearing peptide by subtype-specific HDAC
complexes purified from mammalian cells overexpress-
ing recombinant enzyme.11 ZolinzaÒ, which is generally
considered a broad spectrum inhibitor, is selective for
the class I and limited class II HDAC enzymes (1–3, 6,
8) with IC50 values ranging from 40 to 410 nM for these
family members, but possesses IC50 values >30,000 nM
for the class II enzymes (4, 5, and 7). MS-275 shares a
similar profile to ZolinzaÒ, with the exception of
HDAC6 and HDAC8 inhibition where greater than
110- and 400-fold selectivities are observed, respectively.
Biaryl lead 1 exhibits HDAC1 potency similar to MS-
275, but possesses greater than 10-fold selectivity over
HDAC3. SHI-1:2 2a is selective over all HDAC iso-
forms tested. The lowest selectivity observed relative to
HDAC1 is 7-fold over HDAC2.
O
R
1
MS-275; R = H
2a; R = Ph
10
6180
HDAC1 IC50 (nM) 210
125
HDAC3 IC50 (nM) 2810
370
Figure 1. ZolinzaÒ, Biaryl 1, MS-275, and SHI-1:2 2a.
that modification of the zinc-binding domain could pro-
vide a rational approach to selective inhibitor design.9
The size and composition of the cavity predicted in
HDAC1 suggests that a large hydrophobic group can
be accommodated in the interior of the enzyme active
site. Gratifyingly, biaryl-phenol 1, which was discovered
as a hit during an HTS campaign, was found to be a po-
tent inhibitor of HDAC1 enzyme preparations and is
10-fold less potent against HDAC3 (Fig. 1 and Table
1). Since biaryl-phenol 1 lacks an extensive linker do-
main and does not possess a surface-recognition do-
main, it represents a novel starting point for SAR
generation. Further, the structural similarity between 1
and the existing class of benzamide HDAC inhibitors,
typified by MS-275 (Fig. 1), suggested that modification
of the linker domain and addition of a more extensive
surface-recognition domain was feasible. To assess the
proposed binding mode for the biaryl HDAC1 selective
inhibitors, the biaryl structural analog of the clinical
agent MS-275 was prepared (2a, Fig. 1). Compound
2a was found to be a potent and selective HDAC1 inhib-
itor, whereas MS-275 is only 3-fold selective for
HDAC1 (125 nM) over HDAC3 (370 nM). In light of
these findings, we undertook an extensive SAR program
to further enhance and characterize the activities of the
biaryl class of selective HDAC1&2 inhibitors (SHI-1:2;
Selective HDAC1&2 Inhibitor(s)).
Having demonstrated that the biaryl structural class of
SHI-1:2 possesses a high degree of HDAC selectivity,
we wanted to explore additional analogs and test the
anti-proliferative activities of these agents in a cell-based
assay.12
Additional SHI-1:2 2a analogs were prepared via the
synthetic pathway depicted in Scheme 2. Amide forma-
tion with chloromethyl benzoyl chloride and the Boc-
protected biaryl benzamide generates the chloromethyl
intermediate 11a, which is converted to free amine 12a
via phthalimide displacement of the chloride and subse-
quent deprotection. Acylation of the resulting amine
generates the desired amide or carbamate analogs
14a–i after TFA deprotection.10
We initially explored subtle changes to the carbamate
core of 2a to determine the effects on HDAC1 and
cell-based activity (Table 2). Analog 14a, which is the
corresponding amide of carbamate 2a, exhibited similar
enzymatic and cellular activity to the parent 2a. How-
ever, replacement of the 3-pyridyl with 2-pyridyl (14b)
or phenyl (14c) resulted in reduced anti-proliferative po-
tency. Moreover, removal of the methylene spacer in 2a,
generating phenyl carbamate 14d, also decreased activity
in the HCT-116 assay.
The synthesis of SHI-1:2 2a is shown in Scheme 1. Com-
mercially available 4-bromo-2-nitroaniline (3) was Boc-
Table 1. HDAC selectivity of ZolinzaÒ, 1, MS-275, and SHI-1:2 2a
a
IC50 (nM)
HDAC
ZolinzaTM
1
MS-275
2a
1
2
3
4
5
6
7
8
30
170
210
ND
125
340
10
72
The ancillary pharmacology profiles of the SHI-1:2
carbamate analogs were also gathered to investigate
off-target activities (Table 2). CYP inhibition potential
was determined against a number of human isoforms,
and ion channel affinity in a displacement binding as-
say using radio-labeled MK-499.13 SHI-1:2 2a was
found to be an inhibitor of CYP isoforms and it
100
2810
ND
370
6180
>50,000
>30,000
38
>50,000
>50,000
>50,000
>50,000
14,120
>50,000
>50,000
>50,000
>30,000
>20,000
ND
>50,000
ND
ND
>30,000
410
a Values are means of three experiments. ND = not determined.

728
D. J. Witter et al. / Bioorg. Med. Chem. Lett. 18 (2008) 726–731
Br
Ar
Br
Ar
a
b
c
O2N
H2N
O2N
O2N
HN
HN
NH2
HN
Boc
Boc
Boc
3
5a; Ar = Ph
5b; Ar = 2-thienyl
4a; Ar = Ph
4b; Ar = 2-thienyl
O
H2N
OH
e
d
O
N
H
+
OH
OH
O
N
N
O
6
7
8
O
O
NH2
Ar
O
N
Boc
f
H
HN
O
N
H
H
N
H
N
N
N
O
O
Ar
9a; Ar = Ph
9b; Ar = 2-thienyl
2a; Ar = Ph
2b; Ar = 2-thienyl
Scheme 1. Reagents: (a) Boc2O, Et3N, DCM (62%); (b) Ar-B(OH)2, K2CO3, Pd(PPh3)4, THF (89–95%); (c) H2, Pd/C, MeOH (72–80%); (d) CDI,
DBU, THF (80%); (e) 5a or 5b, EDCI, HOBt, DMF (75–85%); (f) TFA, DCM (60–70%).
Boc
Boc
H2N
HN
Cl
HN
H
H
N
Cl
N
a
b
c
Cl
O
O
O
O
Ar
12a; Ar = Ph
Ar
11a; Ar = Ph
10
11b; Ar = 2-thienyl
12b; Ar = 2-thienyl
O
Boc
R
N
H
HN
H
R
N
NH2
H
N
H
d
N
O
O
Ar
Ar
13a-i; Ar = Ph
15e-i; Ar = 2-thienyl
14a-i; Ar = Ph
15e-i; Ar = 2-thienyl
Scheme 2. Reagents: (a) 5a or 5b, i-PrEt2N, THF (90–94%); (b) phthalimide, K2CO3, KI, Tol (68–75%); (ii) NH2NH2, EtOH (67–77%); (c) RC(O)Cl,
i-PrEt2N, THF or ROH, CDI, THF or ROC(O)Cl, i-PrEt2N, THF or RCO2H, HATU; (d) TFA, DCM.
exhibits IKr ion channel binding activity (IC50
3.0 lM). The CYP inhibition for 2a was greatest for
CYP3A4 with 70% at 1 lM. Both CYP2D6 and
CYP2C9 were also affected but to a lesser extent,
69% and 71% at 10 lM, respectively. The 3-pyridyl
amide 14a faired better than the parent with respect
to both IKr binding as well as CYP inhibition. For
2-pyridyl analog 14a, the IKr binding and CYP3A4
inhibition were also attenuated. The benzyl and phe-
nyl analogs, 14c and 14d, respectively, did not appre-
ciably inhibit CYP enzymes or display IKr binding.
Based on these results, additional analogs were ex-
plored to maintain the clean off-target activity, but
improve biochemical and anti-proliferation activities
of the initial carbamate analogs.
Based on the off-target activity issues noted with the 3-
pyridylmethyl carbamate, additional analogs were ex-
plored focusing on alkyl substituents. Accordingly, the
ethyl and methyl carbamate analogs of 2a were pre-
pared, affording 14e and 14f, respectively, which main-
tain potent enzymatic and cellular activity (Table 3).
Gratifyingly, CYP inhibition and IKr ion channel bind-
ing activities were greatly attenuated. The desired trend
of reduced off-target activity continued with amide ana-
logs 14g and 14h. The HDAC1 enzymatic inhibitory
activities for 14g and 14h were IC50 = 10 nM for both
analogs and the anti-proliferation activities were
GI50 = 220 and 150 nM, respectively, indicating toler-
ance for various moieties in the recognition domain.
Larger alkyl groups are acceptable in the recognition

D. J. Witter et al. / Bioorg. Med. Chem. Lett. 18 (2008) 726–731
Table 2. The initial series of biaryl SHI-1:2
729
Ph
O
N
H
H
R
N
NH2
O
R
HDAC1
HCT-116
% MK-499 Binding
% CYP Inhibition at 10 lM
a
IC50 (nM)
a
GI50 (nM)
Inhibition at 10 lM
3A4
2C9
2D6
2a
3-Pyridyl-CH2O
3-Pyridyl-(CH2)2
2-Pyridyl-CH2O
BnO
10
11
8
225
337
IC50 = 3.0 lM
IC50 = 3.7 lM
51
30
15
70 at 1 lM
69
73
67
27
21
71
25
29
16
12
14a
14b
14c
14d
65
38
28
22
435
820
17
16
PhO
1440
a Values are means of n P 2 experiments.
Table 3. The alkyl carbamate and amide series of biaryl SHI-1:2
Ph
O
N
H
H
R
N
NH2
O
R
HDAC1
HCT-116
% MK-499 Binding
% CYP Inhibition at 10 lM
a
IC50 (nM)
a
GI50 (nM)
Inhibition at 10 lM
3A4
2C9
2D6
2a
3-Pyridyl-CH2O
EtO
10
13
10
10
10
20
225
230
230
220
150
335
IC50 = 3.0 lM
70 at 1 lM
69
35
46
27
28
32
71
26
49
28
38
17
14e
14f
14g
14h
14i
40
28
38
24
30
22
23
33
30
12
MeO
n-Pr
Et
Cyclohexyl-
a Values are means of n P 2 experiments.
domain. Cyclohexyl amide 14i, for example, has slightly
reduced enzyme and anti-proliferative activities with
minimal off-target activity.
were tested in vivo for the ability to attenuate tumor
growth in an HCT-116 colon xenograft model in nude
mice (Fig. 2). The methyl carbamate demonstrated
marked dose dependent efficacy (50, 100, and 150 mg/
kg po qd) at tolerated doses (<10% body weight loss)
To further investigate the SAR of the pendant phenyl
group of the biaryl SHI-1:2, a series of 2-thienyl substi-
tuted analogs (15e–i) was prepared. These agents, which
were prepared in a manner analogous to that used for
the phenyl analogs (Schemes 1 and 2), demonstrated
similar, although slightly greater, activity relative to
the phenyl derivatives 14e–i (Table 4).
Table 4. The 2-thienyl series of biaryl SHI-1:2
S
O
N
To verify that HDAC1&2 selectivity was maintained
during the lead optimization process, representative
examples were tested against a panel of HDAC isoforms
(Table 5). The HDAC selectivity data reveal that the
biaryl benzamide moiety confers selectivity for HDAC1
and HDAC2 over the other HDACs tested.14 The
HDAC activity was greatly attenuated for not only the
class II HDACs 4–7, but also the other members of class
I HDACs, 3 and 8. Moreover, there was a slight prefer-
ence for HDAC1 over HDAC2.
H
H
R
N
NH2
O
R
HDAC1
a
IC50 (nM)
HCT-116
a
GI50 (nM)
2b
3-Pyridyl-CH2O
EtO
6
7
155
130
100
135
146
185
15e
15f
15g
15h
15i
MeO
n-Pr
Et
6
9
9
13
Cyclohexyl-
Based on the desirable off-target profile of methyl carba-
mate 14f, the pharmacologic properties of this analog
a Values are means of n P 2 experiments.

730
D. J. Witter et al. / Bioorg. Med. Chem. Lett. 18 (2008) 726–731
Table 5. HDAC selectivity of SHI-1:2 analogs, 2b, 14f, 15e, and 15f
B
A
a
IC50 (nM)
HDAC
O
Ph
R
2b
14f
15e
15f
NH
HN
1
2
3
4
5
6
7
8
6
190
13
115
7
71
10
105
O
11Å Channel
27,710
>50,000
ND
22,450
>50,000
>50,000
>50,000
>50,000
>50,000
>50,000
>50,000
>50,000
>50,000
>50,000
>50,000
>50,000
>50,000
ND
O
HN
O
Zn+2
NH2
>50,000
ND
31,738
>50,000
ND
>50,000
Zn+2
HN
O
H
a Values are means of three experiments. ND = not determined.
C
O
350
N
R
H
N
Vehicle, IP
H
Zolinza, 150 mg/kg IP
Vehicle, PO
Internal
Cavity
300
O
14f; 50 mg/kg PO
14f; 100 mg/kg PO
H2N
14f; 150 mg/kg PO
250
Surface
Recognition
Metal
Binding
Linker
2-way ANOVA
** : P < 0.005
*** : P < 0.001
Figure 3. Schematic representation of ZolinzaÒ and a biaryl SHI-1:2
analog bound in the HDAC1 active site. (A) ZolinzaÒ bound in HDLP
occupying the hydrophobic channel depicted in yellow. (B) Biaryl SHI-
1:2 occupying both the channel (yellow) and internal cavity depicted in
blue. (C) New pharmacophore model for selective HDAC inhibition
(SHI-1:2).
200
**
**
150
100
50
1
4
7
10
13
16
19
22
the HDAC1 internal cavity, we propose refinement
of the current HDAC pharmacophore model to include
the internal cavity as depicted in Figure 3. In-house
docking studies using a homology model of hHDAC117
support this hypothesis.
Days after beginning treatment
Figure 2. HCT-116 mouse xenograft model with ZolinzaÒ and 14f.
when orally dosed once daily for 21 days. The tumor
growth inhibition at 50 mg/kg qd was equivalent to Zo-
linzaÒ (150 mg/kg ip qd).
In conclusion, we have identified a series of biaryl benz-
amides that are selective HDAC1&2 inhibitors (SHI-
1:2), exhibiting selectivity over class II HDACs 4–7, as
well as class I HDACs 3 and 8. The biaryl class of
SHI-1:2 analogs provide an example of selective HDAC
inhibitors for the HDAC isoforms most closely associ-
ated with cancer. The hypothesis for the increased
HDAC selectivity is the binding of the pendant aromatic
group in the internal cavity of the HDAC isoform. SAR
development based on initial lead 1 led to a series of po-
tent and selective inhibitors with in-vivo xenograft activ-
ity and acceptable off-target activity. The therapeutic
advantages that greater selectivity offers with regard to
an improved tolerability and efficacy compared to Zolin-
zaÒ and other less selective HDAC inhibitors require
further preclinical and clinical examination. Additional
studies of SHI-1:2 2a and related analogs will be re-
ported in due course.
The structural details of HDAC inhibitor/enzyme inter-
actions were first elucidated by Finnin et al. in 1999.9
The crystal structure of histone deacetylase-like protein
(HDLP), a homolog of mammalian HDAC1 with ꢀ35%
sequence identity, was solved with ZolinzaÒ bound to
the active site. The structural findings revealed that
˚
HDLP has a tubelike 11 A deep channel wherein the al-
kyl chain of ZolinzaÒ resides and the hydroxamate che-
lates with the catalytic zinc (Fig. 3A). Immediately
˚
adjacent to the active site zinc is 14 A long internal cav-
ity lined primarily with hydrophobic residues. The inter-
nal cavity’s function is postulated to be a diffusion
pathway for the acetate by-product after enzymatic
hydrolysis.15 We hypothesize that the novel class of biar-
yl SHI-1:2 bind to this internal cavity via the pendant
aryl moiety of the biaryl benzamide headgroup
(Fig. 3B) and that differences in the various HDAC fam-
ily member’s active sites and internal cavities lead to a
preference for the biaryl headgroup with the HDAC1
isoform. The crystal structure of the hHDAC8 supports
this hypothesis, since the presence of the Trp-141 at the
bottom of the active site channel could block the zinc
chelation with the aminobenzamide.16 Moreover, the
shape of the internal cavity of hHDAC8 differs from
that proposed for HDAC1. Based on the utilization of
References and notes
1. (a) Huang, L. J. Cell. Physiol. 2006, 209, 611; (b) Liu, T.;
Kuljaca, S.; Tee, A.; Marshall, G. M. Cancer Treatment
Rev. 2006, 32, 157; (c) Sadri-Vakili, G.; Cha, J.-H. J. Curr.
Alzh. Res. 2006, 3, 403.
2. (a) Rodriquez, M.; Aquino, M.; Bruno, I.; De Martino,
G.; Taddei, M.; Gomez-Paloma, L. Curr. Med. Chem.

D. J. Witter et al. / Bioorg. Med. Chem. Lett. 18 (2008) 726–731
731
2006, 13, 1119; (b) Dokmanovic, M.; Marks, P. A. J. Cell.
Biochem. 2005, 96, 293.
3. Glaser, K. B.; Li, J.; Staver, M. J.; Wei, R.-Q.; Albert, D.
A.; Davidsen, S. K. Biochem. Biophys. Res. Commun.
2003, 310, 529.
(d, J = 7.8 Hz, 2H), 7.50 (d, J = 1.8 Hz, 1H), 7.37 (m, 5H),
7.31 (dd, J = 8.4 and 1.8 Hz, 1H), 7.22 (t, J = 7.8 Hz, 1H),
6.84 (d, J = 8.4 Hz, 1H), 5.08 (br s, 4H), 4.26 (d, J = 6.0 Hz,
2H); MS: Calcd 453 (MH+), exp 453 (MH+). 14f;1H NMR
(d6-DMSO, 600 MHz) d 9.86 (s, 1H), 8.73 (s, 1H), 7.90 (d,
J = 8.4 Hz, 2H), 7.81 (d, J = 1.8 Hz, 1H), 7.75 (t,
J = 6.0 Hz, 1H), 7.61 (m, 3H), 7.48 (dd, J = 9.0 and
3.6 Hz, 1H), 7.43 (t, J = 8.1 Hz, 2H), 7.38 (d, J = 8.4 Hz,
2H), 7.32 (t, J = 7.5 Hz, 1H), 4.23 (d, J = 6.0 Hz, 2H), 3.53
(s, 3H), 1.42 (s, 9H); MS: Calcd 376 (MH+), exp 376 (MH+).
11. See reference within Jones, P.; Altamura, S.; Chakravarty,
P. K.; Cecchetti, O.; De Francesco, R.; Gallinari, P.;
Ingenito, R.; Meinke, P. T.; Petrocchi, A.; Rowley, M.;
4. (a) Kishimoto, M.; Kohno, T.; Okudela, K.; Otsuka, A.;
Saski, H.; Tanabe, C.; Sakiyama, T.; Hirama, C.; Kita-
bayashi, I.; Minna, J. D.; Takenoshita, S.; Yokota, J. Clin.
Cancer Res. 2005, 11, 512; (b) Ishihama, K.; Yamakawa,
M.; Semba, S.; Takeda, H.; Kawata, S.; Kimura, S.;
Kimura, W. J. Clin. Pathol. 2007, 24, 1; (c) Zhu, P.;
Martin, E.; Mengwasser, J.; Schlag, P.; Janssen, K.-P.;
Go¨ttlicher, M. Cancer Cell 2004, 5, 455; (d) Senese, S.;
Zaragoza, K.; Minardi, S.; Muradore, I.; Ronzoni, S.;
Passafaro, A.; Bernard, L.; Draetta, G. F.; Alcalay, M.;
Seiser, C.; Chiocca, S. Mol. Cell. Biol. 2007, 27, 4784; (e)
Ropero, S.; Fraga, M. F.; Ballestar, E.; Hamelin, R.;
Yamamoto, H.; Boix-Chornet, M.; Caballero, R.; Alami-
nos, M.; Setien, F.; Paz, M. F.; Herranz, M.; Palacios, J.;
Arango, D.; Orntoft, T. F.; Aaltonen, L. A.; Schwartz, S.,
Jr.; Esteller, M. Nat. Genet. 2006, 38, 566.
5. Glaser, K. B. Biochem. Pharm. 2007, 74, 659.
6. During this manuscript submission, a reference was released
via JMC web-site. Moradei, O. M.; Mallais, T. C.; Frechette,
S.; Paquin, I.; Tessier, P. E.; Leit, S. M.; Fournal, M.;
Bonfils, C.; Trachy-Bourget, M.-C.; Liu, J.; Yan, T. P.; Lu,
A.-H.; Rahil, J.; Wang, J.; Lefebvre, S.; Li, Z.; Vaisburg, A.
F.; Besterman, J. M. J. Med. Chem. 2007, ASAP.
7. Miller, T. A.; Witter, D. J.; Belvedere, S. J. Med. Chem.
2003, 46, 5097.
Scarpelli, R.; Serafini, S.; Steinkuhler, C. Bioorg. Med.
Chem. Lett. 2007, 16, 5948.
¨
12. HDAC cell-based proliferation results were determined by
the following procedure: HCT116 cells were plated in 96-
well plates at a density of 1000 cells/well. The next day,
cells were treated with either 0.2% DMSO or increasing
concentrations of 7i dissolved in DMSO (final concentra-
tion DMSO 0.2%). After 72 h incubation at 37 °C with 5%
CO2, viable cells were quantitated using Vialight Plus
(Cambrex) according to manufacturer’s instructions.
13. Wang, J.; Della Penna, K.; Wang, H.; Karczewski, J.;
Connolly, T. M.; Koblan, K. S.; Bennett, P. B.; Salata, J.
J. Am. J. Physiol. 2003, 284, H256.
14. HDAC3 IC50 biochemical data for all compounds in the
manuscript are at least >10 lM, where the value is not
shown within.
15. Wang, D.-F.; Wiest, O.; Helquist, P.; Lan-Hargest, H. Y.;
Wiesch, N. L. J. Med. Chem. 2004, 47, 3409.
16. Vannini, A.; Volpari, C.; Filocamo, G.; Casavola, E. C.;
Brunetti, M.; Renzoni, D.; Chakravarty, P.; Paolini, C.;
De Francesco, R.; Gallinari, P.; Steinkuhler, C.; Di
Marco, S. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 15064.
17. Witter, D. J.; Belvedere, S.; Chen, L.; Secrist, J. P.;
Mosley, R. T.; Miller, T. A. Bioorg. Med. Chem. Lett.
2007, 17, 4562.
8. Grant, S.; Easley, C.; Kirkpatrick, P. Nature Rev. Drug
Discov. 2007, 6, 21.
9. 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.
10. Spectral data for 2a/14f: 2a; 1H NMR (d6-DMSO,
600 MHz) d 9.69 (s, 1H), 8.57 (s, 1H), 8.51 (dd, J = 4.8
and 1.8 Hz, 1H), 7.94 (m, 3H), 7.77 (d, J = 7.8 Hz, 1H), 7.53