A series of novel, potent, and selective histone deacetylase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2006.09.002

Histone deacetylase (HDAC) inhibitors offer a promising strategy for cancer therapy and the first generation HDAC inhibitors are currently in clinical trials. A structurally novel series of HDAC inhibitors based on the natural cyclic tetrapeptide Apicidin is described. Selected screening of the sample collection looking for L-2-amino-8-oxodecanoic acid (L-Aoda) derivatives identified a small acyclic lead molecule 1 with the unusual ketone zinc binding group. SAR studies around this lead resulted in optimization to potent, low molecular weight, selective, non-hydroxamic acid HDAC inhibitors, equipotent to current clinical candidates.

Full text of DOI:10.1016/j.bmcl.2006.09.002

Bioorganic & Medicinal Chemistry Letters 16 (2006) 5948–5952
A series of novel, potent, and selective histone deacetylase inhibitors
Philip Jones,a,* Sergio Altamura,a Prasun K. Chakravarty,b Ottavia Cecchetti,a
Raffaele De Francesco,a Paola Gallinari,a Raffaele Ingenito,a Peter T. Meinke,b
Alessia Petrocchi,a Michael Rowley,a Rita Scarpelli,a Sergio Serafinia and
Christian Steinkuhlera
¨
aIRBM/Merck Research Laboratories, Via Pontina km 30,600, 00040 Pomezia, Italy
bMerck Research Laboratories, 126 Lincoln Avenue, PO Box 2000, Rahway, NJ 07065, USA
Received 8 August 2006; revised 31 August 2006; accepted 1 September 2006
Available online 20 September 2006
Abstract—Histone deacetylase (HDAC) inhibitors offer a promising strategy for cancer therapy and the first generation HDAC
inhibitors are currently in clinical trials. A structurally novel series of HDAC inhibitors based on the natural cyclic tetrapeptide
Apicidin is described. Selected screening of the sample collection looking for L-2-amino-8-oxodecanoic acid (L-Aoda) derivatives
identified a small acyclic lead molecule 1 with the unusual ketone zinc binding group. SAR studies around this lead resulted in opti-
mization to potent, low molecular weight, selective, non-hydroxamic acid HDAC inhibitors, equipotent to current clinical
candidates.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Aberrant gene regulation plays an important role in tu-
mor initiation and progression, and methods to correct
this dysfunction represent a new anti-cancer strategy.1
The acetylation status of lysine residues in nucleosomal
histone proteins is known to play a crucial role in chro-
matin structure and hence gene transcription.2 The pres-
ence of acetyl groups on these lysine residues neutralizes
the positive charges of the histone tails, therefore
decreasing their interactions with DNA, relaxing the
chromatin, and allowing access to transcription factors.
In contrast, removal of the acetyl groups from these Ac-
lysines facilitates the interactions of e-amino group with
the DNA, condensing the chromatin. This acetylation
status is controlled by the opposing actions of histone
acetyl transferases and HDACs; eleven class I and II
HDACs are known.3 Histone deacetylase inhibitors
(HDACi) are emerging as a new class of anti-cancer
agents and have been shown to alter gene transcription
and exert various anti-tumor effects such as: growth
arrest, differentiation, apoptosis and inhibition of tumor
angiogenesis.4–6 However, the precise anti-tumor mech-
anism of these HDACis remains unclear.4 Currently,
there are several HDACis in clinical trials and showing
efficacy in patients with hematological and solid
malignancies.7,8
HDACis have been widely studied, X-ray crystal struc-
tures with bound inhibitors exist,9 and they can be char-
acterized by a common pharmacophore consisting of: a
zinc binding group (ZBG), a linker domain, and a sur-
face recognition domain. To date they belong to several
distinct structural classes (Fig. 1): hydroxamic acids:
such as SAHA (Suberoyl anilide hydroxamic acid),10
aminobenzamides: for instance, MS-275,11 electrophilic
ketones,12 short chain fatty acids like Valproic acid13
and cyclic peptides: such as FK-22814 and Apicidin.15
Each class has its own limitations: hydroxamic acids
are rapidly metabolized, non-selective inhibitors of most
HDAC isoforms; aminobenzamides and fatty acids have
limited potency; electrophilic ketones are reduced in
plasma; whilst cyclic peptides offer issues of chemical
intractability and FK-228 contains an undesirable thiol
ZBG.
As part of a project to identify a second generation
HDACi the natural product Apicidin was selected as a
suitable starting point. This molecule contains an ethyl
ketone as potential ZBG, a long alkyl chain as linker,
and the cyclic tetrapeptide that interacts with the surface
Keywords: Histone deacetylase inhibitor; HDAC; SAR studies.
*
Corresponding
+00390691093654; e-mail: philip_jones@merck.com
author.
Tel.:
+39
0691093559;
fax:
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.09.002

P. Jones et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5948–5952
5949
O
O
O
O
H
N
H
N
O
N
NH2
OH
OH
H
N
H
N
H
CF3
N
O
O
O
Valproic Acid
Electrophilic Ketone
SAHA
MS-275
S
O
N
O
O
H
O
O
S
NH
O
N
H
H
N
N
H
MeO
H
N
NH
H
N
NH
NH
H
HN
O
N
OMe
NH
H
(S)
N
N
O
O
O
O
O
O
N
H
O
O
O
H
HN
O
O
O
O
Apicidin
Trapoxin B
FK-228
1
Figure 1. Known HDAC inhibitors and the structure of novel lead 1.
Table 1. Activity of 1 and related compounds with alternative ZBGs
of the HDAC. This compound, despite the unusual
ZBG, is a relatively potent HDAC 1 inhibitor,16
IC50 = 44 nM, and displays good anti-proliferation
activity against cervical cancer HeLa cells in a 72 h
growth inhibition assay, IC50 = 290 nM.17 Apicidin is
unique in that it lacks the epoxide whose presence is
integral for biological activity in closely related cyclic
tetrapeptide HDACi such as Trapoxin,18 suggesting that
Apicidin’s macrocycle contributes significantly to
activity.19
O
(S)
HN
N
H
R
O
HN
OMe
N
H
Compound
R
HDAC 1
PRO(HeLa)
a
IC50 (nM) IC50 (nM)
a
Directed screening of the sample collection looking for
compounds containing the unusual L-Aoda amino acid
identified 1 as a more tractable lead. This compound
maintains the ketone motif and pentyl chain, whilst
exerting HDAC inhibitory activity, IC50 = 590 nM
(albeit with a 12-fold loss compared to Apicidin), and
good cellular activity remained as characterized in a
72 h cellular proliferation assay in HeLa cervical cancer
cells,17 PRO(HeLa) IC50 = 730 nM. This loss of enzyme
activity was deemed tolerable as the compound is ame-
nable to rapid analogue synthesis, although it suffers
from: modest potency; high MWt; many rotatable
bonds; and high liver microsome turnover.
SAHA
MS-275
Apicidin
1
27
110
460
1800
44
590
290
730
(R)-1
2
3200
>5000
>25,000
23,700
OH
3
4
5
350
220
150
5700
>25,000
1300
CHO
CO2H
CO2Me
H
N
6
0.8
12
OH
O
NH2
7
8
1760
>25,000
>25,000
Naturally, the SAR focused on three regions of the mol-
ecule: the ZBG, the linker domain, and surface recogni-
tion domain. At an early stage of the program it was
demonstrated that the chirality of these molecules was
critically important as the enantiomer, (R)-1, lost signif-
icant HDAC inhibitory activity. Subsequently all ana-
logues were prepared homochirally in the S-series.
O
>5000
a Values are means of three or more experiments.
Interestingly, when the ketone was replaced by the
strong zinc chelator, a hydroxamic acid, a sub-nanomo-
lar enzyme inhibitor was identified. In fact, compound 6
is one of the most potent HDACis currently known and
shows excellent anti-proliferation activity, PRO(HeLa)
IC50 = 12 nM.
Investigations into the ZBG revealed that the ketone 1
was one of the most promising compounds as some
alternative structures either lost HDAC activity, such
as alcohol 2 or amide 7 (Table 1) or lacked cell-based
activity despite being more potent enzyme inhibitors,
for instance acid 4. The aldehyde 3 and methyl ester 5,
although showing activities comparable to that of the
lead, were not followed up further due to potential lia-
bilities associated with these functionalities in vivo. Fur-
thermore, the necessity for the ketone functionality was
demonstrated by compound 8 lacking the carbonyl. This
compound lost all activity, demonstrating the probable
binding interaction made by this weak zinc chelator.
Further profiling of compound 6 demonstrated excellent
inhibitory activity on HDACs 1–3 whilst the other
HDAC isoforms were inhibited at higher concentrations
(Table 2).16 Potent anti-proliferative activity was also
seen against a wide panel of cancer cell lines and the
compound showed good selectivity over normal cells,
such as renal epithelial cells.17 This anti-proliferative
activity correlated an increased level of histone H3 acet-
ylation. Despite the potency of this compound work
continued to focus on the ketone series.

5950
P. Jones et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5948–5952
Table 2. Activity of 6 on the HDAC isoforms and various cell lines
Table 3. SAR of replacements of the 2-methyl-5-methoxyindole group
HDAC
1
2
4
3
4
5
6
7
8
a
IC50 (nM)
O
0.8
1.6
260
2500
45
4200
200
(S)
HN
N
H
a
IC50 (nM)
Cell line
HeLa
HN
O
O
12
64
R
Colon-HCT116
Lung-A549
U937
Compound
R
HDAC 1
a
IC50 (nM)
PRO(HeLa)
a
IC50 (nM)
87
<8
A2780
G401
110
100
130
930
9
Me
930
590
7600
3000
S
S
MCF-7
Human renal epithelial cells
a Values are means of three or more experiments.
10
N
N
N
11
480
3400
Investigations into the pentyl chain and around the two
amide groups (b-amino acids and alkylated derivatives)
revealed these to be preferred (data not shown), accord-
ingly a robust synthesis of the key amino acid was re-
quired. This was achieved using Scho¨llkopf chemistry
(Scheme 1).20 Alkylation of the protected alkyl iodide
32, readily synthesized without chromatography, gave
homo-chiral amino acid 35 in 58% yield after hydrolysis
and Boc-protection. SAR into the capping groups was
then achieved by standard coupling methodology.
Me
N
12
13
14
15
540
200
220
190
2500
1200
2000
<390
NMe
NMe
N
a Values are means of three or more experiments.
Extensive SAR work to find a replacement for the 2-
methyl-5-methoxyindole group of 1 identified small het-
erocyclic derivatives, such as 5-carboxythiazole 10 or the
thiadiazole 11, as suitable alternatives (Table 3). These
groups maintained HDAC activity but lost a few fold
in the anti-proliferation assay. Small cyclic amino
groups, such as N-methyl piperidines 12 + 13 and pyr-
rolidine 14, were also well tolerated, resulting in modest
improvements in HDAC inhibition. These substantially
reduced MWt and furthermore improved microsomal
stability. Pendant amines such as the 3-piperid-1-ylpro-
panamide 15 also behave similarly, improving both
enzyme and cellular activities.
Fortunately, anilide 18 gave rise to a 3-fold improve-
ment in enzymatic activity compared to 1 whilst reduc-
ing the MWt by 140 amu (Table 4). Further SAR
around this group revealed a more dramatic improve-
ment in enzymatic activity could be achieved by substi-
tution in the meta-position. Chlorine (20), methoxy (22),
and phenyl (23) groups were well tolerated at these posi-
tions resulting in inhibitors with double-digit nanomolar
potencies and sub-micromolar anti-proliferation activi-
ties. Likewise, heteroaromatic amines were also tolerat-
ed, especially those bearing groups in the distal position,
such as phenyl thiazole 24.
Finding alternatives to the large 2-(2-phenyl-1H-indol-
3-yl)ethanamine moiety was more problematic and most
replacements, with the exception of (hetero)anilides,
resulted in a dramatic loss of activity (data not shown).
After this initial SAR the most interesting substituents
were combined and incorporated for a more extensive
SAR around this series in a 2-D array. From this work
small, low molecular weight, potent HDACis were iden-
tified with sub-micromolar activities against HeLa cell
proliferation (Table 5). The best of these derivatives
show activities comparable to those of SAHA and
MS-275. Interestingly, the results from this work re-
vealed that the SAR into the two separate amide groups
was not additive and binding to one part of the surface
clearly influenced binding of the second amide group.
Nevertheless, 29 displayed IC50 = 55 nM against HDAC
1 with HeLa anti-proliferation activity of 430 nM and
improved microsomal stability (human Clint = 54 lL/
min/mg compared to >300 lL/min/mg for 1) so warrant-
ed further profiling on the other HDAC isoforms and a
panel of cancer cell lines (Table 6).
Extensive profiling revealed that 29 was indeed a iso-
form selective HDAC inhibitor displaying potent inhibi-
Scheme 1. Synthesis of key amino acid building block.

P. Jones et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5948–5952
5951
Table 4. SAR of replacements of the 2-phenylindole group
Table 6. Activity of 29 on the HDAC isoforms and various cell lines
HDAC
1
2
3
4
5
6
7
8
O
(S)
a
R
N
H
IC50
(nM)
55 170 14 NA at 1900 13 NA at 3000
5 lM 5 lM
HN
O
O
OMe
a
IC50 (nM)
Cell line
N
Cervical–HeLa
Colon-HCT116
Lung-A549
430
670
H
2000
450
Compound
R
HDAC 1
a
IC50 (nM)
PRO(HeLa)
a
IC50 (nM)
Myeloid-U937
Ovarian-A2780
Kidney-G401
1700
250
Breast-MCF7
Human renal epithelial
230
>20,000
16
560
3800
HN
a Values are means of three or more experiments.
17
18
19
20
21
22
Et
Ph
>5000
180
3300
44
>25,000
1500
o-Cl–Ph–
m-Cl–Ph–
p-Cl–Ph–
27,700
520
2000
concentrations of 20 lM were well tolerated without
significant growth inhibition.
230
23
m-MeO–Ph–
110
In summary, a novel class of potent, low molecular
weight, non-hydroxamic acid histone deacetylase inhib-
itors has been developed. These HDACis are equipotent
with current clinical candidates.
23
24
18
25
230
Ph
S
<390
N
Ph
a Values are means of three or more experiments.
Acknowledgments
Thanks to Amy Cheung and Michael J. Symonifka for
their work in the initial preparation of compound 1
and related derivatives.
Table 5. Simultaneous SAR into both amide groups
O
(S)
R
N
H
HN
O
O
R'
References and notes
Compound
R
R0
HDAC1
PRO(HeLa)
a
IC50 (nM) IC50 (nM)
a
1. Johnstone, R. W. Nat. Rev. Drug Discov. 2002, 1, 287.
2. Cheung, W. L.; Briggs, S. D.; Allis, C. D. Curr. Opin. Cell
Biol. 2000, 12, 326.
3. De Ruijter, A. J. M.; van Gennip, A. H.; Caron, H. N.;
Kemp, S.; van Kuilenburg, A. B. P. J. Biochem. 2003, 370,
737.
4. Minucci, S.; Pelicci, P. G. Nat. Rev. Cancer 2006, 6, 38.
5. Miller, T. A.; Witter, D. J.; Belvedere, S. J. Med. Chem.
2003, 46, 5097.
6. Marks, P. A.; Rifkind, R. A.; Richon, V. M.; Breslow, R.;
Miller, T.; Kelly, W. K. Nat. Rev. Cancer 2001, 1, 194.
7. Rosato, R. R.; Grant, S. Expert Opin. Investig. Drugs
2004, 13, 21.
8. Olsen, E.; Kim, Y. H.; Kuzel, T.; Pacheco, T. R.; Foss, F.;
Parker, S.; Wang, J. G.; Frankel, S.R.; Lis, J.; Duvic, M.
Vorinostat (SAHA) is clinically active in advanced cuta-
neous T-cell lymphoma (CTCL): Results of a phase IIb
trial. Abstract 7500, Annual Meeting of the American
Society of Clinical Oncology, Atlanta, Ga., June 2–6,
2006.
S
N
N
25
26
27
28
29
m-Cl–Ph–
100
33
79
55
55
1600
880
500
720
430
S
m-MeO–Ph–
N
Me
N
Ph
S
N
Ph
S
Me
N
N
Ph
a Values are means of three or more experiments.
9. 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.
10. Richon, V. M.; Webb, Y.; Merger, R.; Sheppard, T.;
Jursic, B.; Ngo, L.; Civoli, F.; Breslow, R.; Rifkind, R. A.;
Marks, P. A. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 5705.
11. Suzuki, T.; Ando, T.; Tsuchiya, K.; Fukazawa, N.; Saito,
A.; Mariko, Y.; Yamashita, T.; Nakanishi, O. J. Med.
Chem. 1999, 42, 3001.
tion of HDACs 1, 2, and 3 + 6, whilst only weak or no
activity on HDACs 4, 5, 7 + 8. Furthermore, 29 main-
tained activity on a wide range of cancer cell lines and
showed sub-micromolar activity against: colon
HCT116, myeloid U937, kidney G401, and breast
MCF7 lines. Moreover, this and other of these novel
HDACis show good selectivity over normal cells as
demonstrated in human renal epithelial cells where

5952
P. Jones et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5948–5952
12. Frey, R. R.; Wada, C. K.; Garland, R. B.; Curtin, M. L.;
Michaelides, M. R.; Li, J.; Pease, L. J.; Glaser, K. B.;
Marcotte, P. A.; Bouska, J. J.; Murphy, S. S.; Davidsen, S.
K. Bioorg. Med. Chem. Lett. 2002, 12, 3443.
affinity then anion-exchange (MonoQ) chromatographies,
and finally gel filtration. The IC50 were determined using
the HDAC Fluorescent Activity Assay, BioMol Research
Laboratories (Plymouth Meeting, PA).
13. Chen, J. S.; Faller, D. V.; Spanjaard, R. A. Curr. Cancer
Drug Targets 2003, 3, 219.
17. Anti-proliferation assays were conducted for 72 h using
the Promega (Madison, WI) CellTiter-BlueTM Cell Viability
Assay. Cells + 10% FCS were grown in 96-well plates,
5000–15,000 cells/well, in the presence of HDACi for 72 h.
CellTiter-BlueTM was added, cells were incubated for an
hour and read (ex. 550 nm, em. 590 nm).
18. Itazaki, H.; Nagashima, K.; Sugita, K.; Yoshida, H.;
Kawamura, Y.; Yasuda, Y.; Matsumoto, K.; Ishii, K.;
Uotani, N.; Nakai, H.; Terui, A.; Yoshimatsu, S.; Ikeni-
shi, Y.; Nakawaga, Y. J. Antibiot. 1990, 43, 1524.
14. Ueda, H.; Nakajima, H.; Hori, Y.; Fujita, T.; Nishimura,
M.; Goto, T.; Okuhara, M. J. Antibiot 1994, 47, 301.
15. Darkin-Rattray, S. J.; Gurnett, A. M.; Myers, R. W.;
Dulski, P. M.; Crumley, T. M.; Allocco, J. J.; Cannova,
C.; Meinke, P. T.; Colletti, S. L.; Bednarek, M. A.; Singh,
S. B.; Goetz, M. A.; Dombrowski, A. W.; Polishook, J.
D.; Schmatz, D. M. Proc. Natl. Acad. Sci. U.S.A. 1996, 93,
13143.
16. C-terminally Flag tagged HDACs 1, 2, and 3 + 6 were
expressed in HEK293 cells and affinity purified using anti-
FLAG antibody affinity resin and eluted with a 3· FLAG
peptide. The HDACs 4, 5, and 7 + 8 were expressed in
Escherichia coli (an organism devoid of endogenous
HDACs to overcome the problem of co-purification of
endogenous HDACs). Enzymes were purified to apparent
homogeneity (Coomassie-stained SDS–PAGE) by nickel
19. Incorporating a terminal epoxide into Apicidin’s L-Aoda
side chain resulted in a dramatic increase in HDAC
activity. Colletti, S. L.; Myers, R. W.; Darkin-Rattray, S.
J.; Gurnett, A. M.; Dulski, P. M.; Galuska, S.; Allocco, J.
J.; Ayer, M. B.; Li, C.; Lim, J.; Crumley, T. M.; Cannova,
C.; Schmatz, D. M.; Wyvratt, M. J.; Fisher, M. H.;
Meinke, P. T. Bioorg. Med. Chem. Lett. 2001, 11, 107.
20. Scho¨llkopf, U. Tetrahedron 1983, 39, 2085.