Histone deacetylase (HDAC) inhibitors with a novel connecting unit linker region reveal a selectivity profile for HDAC4 and HDAC5 with improved activity against chemoresistant cancer cells

Article abstract of DOI:10.1021/jm301254q

The synthesis and biological evaluation of new potent hydroxamate-based HDAC inhibitors with a novel alkoxyamide connecting unit linker region are described. Biological evaluation includes MTT and cellular HDAC assays on sensitive and chemoresistant cancer cell lines as well as HDAC profiling of selected compounds. Compound 19i (LMK235) (N-((6-(hydroxyamino)-6-oxohexyl)oxy)- 3,5-dimethylbenzamide) showed similar effects compared to vorinostat on inhibition of cellular HDACs in a pan-HDAC assay but enhanced cytotoxic effects against the human cancer cell lines A2780, Cal27, Kyse510, and MDA-MB231. Subsequent HDAC profiling yielded a novel HDAC isoform selectivity profile of 19i in comparison to vorinostat or trichostatin A (TSA). 19i shows nanomolar inhibition of HDAC4 and HDAC5, whereas vorinostat and TSA inhibit HDAC4 and HDAC5 in the higher micromolar range.

Full text of DOI:10.1021/jm301254q

Article
pubs.acs.org/jmc
Histone Deacetylase (HDAC) Inhibitors with a Novel Connecting Unit
Linker Region Reveal a Selectivity Profile for HDAC4 and HDAC5 with
Improved Activity against Chemoresistant Cancer Cells
Linda Marek,† Alexandra Hamacher,† Finn K. Hansen, Krystina Kuna, Holger Gohlke,
Matthias U. Kassack, and Thomas Kurz*
Institut fur Pharmazeutische und Medizinische Chemie, Heinrich Heine Universitat, Universitatsstrasse 1, 40225 Dusseldorf,
̈
̈
̈
̈
Germany
S
* Supporting Information
ABSTRACT: The synthesis and biological evaluation of new potent
hydroxamate-based HDAC inhibitors with a novel alkoxyamide connecting
unit linker region are described. Biological evaluation includes MTT and
cellular HDAC assays on sensitive and chemoresistant cancer cell lines as
well as HDAC profiling of selected compounds. Compound 19i (LMK235)
(N-((6-(hydroxyamino)-6-oxohexyl)oxy)-3,5-dimethylbenzamide) showed
similar effects compared to vorinostat on inhibition of cellular HDACs in a
pan-HDAC assay but enhanced cytotoxic effects against the human cancer
cell lines A2780, Cal27, Kyse510, and MDA-MB231. Subsequent HDAC
profiling yielded a novel HDAC isoform selectivity profile of 19i in
comparison to vorinostat or trichostatin A (TSA). 19i shows nanomolar
inhibition of HDAC4 and HDAC5, whereas vorinostat and TSA inhibit HDAC4 and HDAC5 in the higher micromolar range.
revealed that the chelator binds to the catalytically important
Zn2+ ion located at the bottom of the active side.7 The linker
mimics the N-acetyllysine side chain within a narrow
hydrophobic channel of 11 Å length and connects the chelating
group via a connecting unit with a hydrophobic cap group.3,7a
Linear, aromatic, and cinnamoyl linkers have been often used,
while the cap groups are mainly of aromatic or heteroaromatic
nature.3 The cap group interacts with amino acid side chains of
the rim region of the active site cavity. Recent results have
shown that the nature of the linker, CU, and cap group has an
impact on the zinc hydroxamate chelation mode and therefore
on the development of subtype selective and/or class selective
HDAC inhibitors.3
The most promising classes of HDAC inhibitors discovered
so far are hydroxamic acids ((trichostatin A (TSA), vorinostat
(SAHA), depsipeptides (romidepsin), short-chain fatty acids
(valproic acid), and o-aminoanilides (mocetinostat and
entinostat); Figure 1). Until now, the pan-HDAC inhibitor
vorinostat and the class I selective prodrug romidepsin have
been approved for treatment of cutaneous T-cell lymphoma
(CTCL).8 In addition, various hydroxamate-based analogues of
vorinostat including panobinostat (phase III) and belinostat
(phase III) as well as the o-aminoanilides (mocetinostat and
etinostat, phase II) are under extensive clinical investigations.5c
INTRODUCTION
■
Histone deacetylases (HDACs) play a key role in the epigenetic
modulation of gene expression by remodeling chromatin
structure.1 Specifically, HDACs catalyze the removal of acetyl
groups from N-acetyllysine residues of histones and other
cellular proteins (e.g., HSP90, tubulin) during post-translational
protein modification.2 Deacetylation causes chromatin con-
densation leading mainly to transcriptional suppression,
whereas acetylation by histone acetyl transferases (HATs)
generally leads to gene activation.2 HDAC classes I (HDACs
1−3, 8), IIa (HDACs 4, 5, 7, 9), IIb (HDACs 6, 10), and IV
(HDAC 11) are Zn2+ dependent enzymes, while class III
HDACs (sirtuins) are NAD+ dependent.3,4 Class I HDACs are
predominantly located in the nucleus, and class II HDACs are
in the nucleus and cytoplasm.3
HDACs are not only regulators of mitosis, cell differ-
entiation, apoptosis, and chromatin organization but are also
involved in the regulation of metabolism, learning, memory,
and immune response.5 As clinically validated cancer targets,
their inhibition has been proven to be a successful strategy for
the development of novel anticancer drugs.6 The HDAC
inhibitor-induced cell death is mediated through several
pathways. Cell growth arrest, differentiation, induction of
apoptosis, antiangiogenic effects, and effects on DNA repair and
on mitosis have been observed in cancer cells after HDAC
inhibition.5c
Received: August 31, 2012
Revised: December 18, 2012
Accepted: December 20, 2012
Structurally, HDAC inhibitors are characterized by a cap
group connecting unit (CU) linker chelator pharmacophore
model.3 Crystal structures of HDAC-Zn2+ inhibitor complexes
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm301254q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
chemoresistant cancer cell lines, and in the case of the most
active derivatives HDAC profiling with selected HDACs.
RESULTS AND DISCUSSION
Chemistry. Potential HDAC inhibitors with a novel linker
region were synthesized as illustrated in Schemes 1−4. The first
■
a
Scheme 1. Synthesis of Carboxylic Acid Derivatives 8−10
Figure 1. Selected HDAC inhibitors.
First phase I and phase II studies demonstrate that pan-
HDAC inhibitors may also cause numerous side effects such as
bone marrow depression, diarrhea, weight loss, taste
disturbances, electrolyte changes, disordered clotting, fatigue,
and cardiac arrhythmias.9 More recent results suggest that
inhibition of specific HDACs might be helpful to discover more
effective inhibitors with reduced side effects.10 However, more
data are required to understand if subtype selective, class
selective, or broad spectrum inhibitors are to be preferred in
cancer treatment. Current clinical data suggest that it depends
on the cancer type and the specific drug combination if subtype
selective, class selective, or broad spectrum inhibitors are to be
preferred in clinical use.8,11
Current clinical and preclinical data suggest that HDAC
inhibitors have potential for the treatment of several cancer
types either alone or in combination with other cytostatic
drugs.12 Notably, the combination of HDAC inhibitors with
various established anticancer drugs (e.g., cisplatin) enhanced
the cytotoxicity of classical chemotherapeutic agents in
hematological and solid tumors.9,13
a
Reagents and conditions: (a) Et3N, CH3CN, reflux, 3−5 h; (b)
methylhydrazine, CH2Cl2, −10 °C, 2 h; (c) R1NCO, CH2Cl2, rt, 18 h;
(d) PhCOOH, EDC, DMAP, rt, 24 h; (e) (i) CDT, CH2Cl2, rt, 2 h;
(ii) phenylhydrazine, Et3N, rt, 18 h; (f) Pd/C, H2 (1 bar), THF, rt, 2
h.
series of compounds was obtained starting from benzyl 6-((1,3-
dioxoisoindolin-2-yl)oxy)hexanoate 3a or benzyl 6-((1,3-
dioxoisoindolin-2-yl)oxy)pentanoate 3b, which were prepared
via O-alkylation of N-hydroxyphthalimide 1 with benzyl ω-
bromocarboxylates 2a,b. Deprotection of the phthaloyl group
of 3a,b utilizing methylhydrazine in ethanol provided the O-
substituted aminoxy derivatives 4a,b. Subsequently, 4a,b were
converted into the alkoxyurea derivatives 5a−f by treatment
with isocyanates. The attempted preparation of the alkoxy-
amino compound 6a using benzoyl chloride provided a
diacylated byproduct. Consequently, we developed a mild and
selective synthesis of 6a utilizing 1-ethyl-3-(3-dimethylamino-
propyl)carbodiimide hydrochloride (EDC) in the presence of a
catalytic amount of 4-dimethylaminopyridine (DMAP) as
coupling agent, which furnished smoothly the desired
intermediate 6a in 88% yield. In order to obtain the
On the basis of a widely accepted HDAC pharmacophore
model, we now report on the development and biological
evaluation of a new class of potent hydroxamate-based HDAC
inhibitors. Cap group, CU, and linker region were modified,
and the resulting scaffolds were combined with different ZBGs
(Figure 2). The biological evaluation of the target compounds
includes MTT assays, pan-HDAC assays on sensitive and
Figure 2. Strategy and target compounds.
B
dx.doi.org/10.1021/jm301254q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 2. Synthesis of Target Compounds 16−20
a
Reagents and conditions: (a) NH2OBn, EDC, DMAP, rt, 24 h; (b) CH3NHOBn, EDC, DMAP, rt, 24 h; (c) Pd/C, H2 (1 bar), THF, rt, 2 h.
alkoxysemicarbazide 7, compound 4a was treated with 1,1′-
carbonylditriazole (CDT) to give an N-triazolide intermediate,
which was subsequently reacted with phenylhydrazine to afford
alkoxysemicarbazide 7. The benzyl ester moieties of inter-
mediates 5a−f, 6a, and 7 were conveniently deprotected by
catalytic hydrogenation to provide the free carboxylic acid
derivatives 8a−f, 9a, and 10 (Scheme 1).
Scheme 3. Synthetic Strategy for the Straightforward
Modification of the Cap Group
a
The preparation of target compounds 16−20 is outlined in
Scheme 2. The coupling reactions of 8a−f with O-
benzylhydroxylamine utilizing EDC as coupling agent provided
the O-benzyl protected hydroxamic acids 11a−e and 12, which
were finally transformed via hydrogenolysis into the desired
hydroxamic acid derivatives 16a−e and chain shortened
analogue 17 in yields of 41−67%. The N-methyl-substituted
hydroxamic acid 18 was obtained from 8a and N-methyl-O-
benzylhydroxylamine. In a similar fashion, the alkoxyamido
derivative 9a and the semicarbazide 10 were converted into the
O-benzyl-protected hydroxamic acids 14a and 15 and
subsequently deprotected to yield the target compounds 19a
and 20 in moderate yields (Scheme 2).
In order to efficiently modify the lipophilic cap group while
retaining the hydroxamate moiety as ZBG, we developed an
improved and straightforward synthetic strategy for compounds
of types 16 and 19 utilizing 6-(aminooxy)-N-(benzyloxy)-
hexanamide 23 as key intermediate (Scheme 3). For this
purpose, benzyl 6-((1,3-dioxoisoindolin-2-yl)oxy)hexanoate 3a
was converted into the carboxylic acid 21. Next, 21 was treated
with thionyl chloride to give the corresponding acyl chloride,
which was reacted with O-benzylhydroxylamine to furnish the
O-benzyl protected hydroxamic acid 22. The synthesis of 23
was accomplished by methylhydrazine-mediated cleavage of the
phthaloyl protection group. The key intermediate 23 enabled
the synthesis of target compounds of types 16 and 19 in a
convenient two-step procedure. First, 23 was transformed into
the O-protected alkoxyurea derivatives 11f−k and alkoxya-
mides 14b−i. Finally, the catalytic hydrogenation of inter-
mediates 11f−k and 14b−i afforded the desired HDAC
inhibitors 16f−k and 19b−i in yields of 25−94% and 36−
86%, respectively.
a
Reagents and conditions: (a) Pd/C, H2 (1 bar), THF, rt, 2 h; (b) (i)
SOCl2, CH2Cl2, reflux, 2 h; (ii) NH2OBn, pyridine, CH2Cl2, rt, 18 h;
(c) methylhydrazine, CH2Cl2, −10 °C, 2 h; (d) R1NCO, CH2Cl2, rt,
18 h; (e) (i) CDT, CH2Cl2, rt, 2 h; (ii) R1NH2, Et3N, rt, 18 h; (f)
R1COOH, EDC, DMAP, CH2Cl2, rt, 24 h.
benzyloxyaniline generated anilides 24a,b in overall yields of
82% and 87%, respectively. Cleavage of the phthaloyl
protection group and acylation with phenyl isocyanate yielded
the o-nitro- and o-benzyloxyanilides 26 and 27. Catalytic
hydrogenation of 26 afforded the desired o-aminoanilide 28 in
46% yield, while deprotection of 27 gave o-hydroxyanilide 29
(42% yield, Scheme 4).
Biological Evaluation and Binding Mode. Cellular
Based Activity Assays. All synthesized target compounds
(8a−29) were assessed in a MTT assay for cytotoxicity and in a
pan-HDAC inhibition assay using the human ovarian cancer
cell lines A2780 and its cisplatin resistant subclone A2780CisR.
Results are presented in Table 1.
In order to prepare potential HDAC inhibitors with o-amino-
and o-hydroxyanilide moieties as non-hydroxamate ZGBs,
intermediate 21 was transformed into the corresponding acyl
chloride. The subsequent reaction with 2-nitroaniline and 2-
The carboxylic acid derivatives 8a−e showed weak (8a, 8b)
or no (8c−e) cytotoxic activity up to 100 μM against both cell
lines. Modifications of the lipophilic cap group led to an
C
dx.doi.org/10.1021/jm301254q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
compound of this series, 19i, gave IC50 values of 0.49 μM
(A2780) and 0.32 μM (A2780 CisR), which was even lower
than those of the reference compound vorinostat or the
cytotoxic agent cisplatin. For the majority of compounds, the
cytotoxic activity was similar at the parental and cisplatin
resistant subclone.
Scheme 4. Synthesis of Target Compounds 28 and 29
The carboxylic acid derivatives displayed no inhibitory
activity in the cellular pan-HDAC assay up to 100 μM. Most
probably, this lack of activity is attributed to the weaker zinc
binding properties of the carboxyl group in comparison to the
hydroxamate moiety. Since 8c−e were inactive in the MTT
assay up to 100 μM, their inhibitory activity in the pan-HDAC
assay was not determined in this study. All other compounds
with the exception of 18 showed IC50 values for HDAC
inhibition in the micromolar range which were similar to their
cytotoxic activity.
a
Reagents and conditions: (a) (i) SOCl2, CH2Cl2, reflux, 2 h; (ii) 2-
nitroaniline or 2-benzyloxyaniline, pyridine, CH2Cl2, rt, 18 h; (b)
methylhydrazine, CH2Cl2, −10 °C, 2 h; (c) phenyl isocyanate,
CH2Cl2, rt, 18 h; (d) Pd/C, H2 (1 bar), MeOH, rt, 6 h.
Differences between HDAC inhibition and cytotoxicity
values may be attributed to differences in incubation times
used in these assays. However, we do see a significant
correlation between HDAC inhibition and MTT data, which
supports, but does not prove, the hypothesis that HDAC
inhibition mediates cytotoxicity (Figure S1 of Supporting
Information).
Table 1. Cytotoxic Activity and HDAC Inhibition (IC50
[μM]) of 8a−29, Cisplatin, Vorinostat, and TSA against the
Human Ovarian Cancer Cell Lines A2780 and A2780 CisR
a
cytotoxicity IC50
HDAC inhibition
[μM]
IC50 [μM]
compd
8a
R1
A2780 A2780 CisR A2780 A2780 CisR
Important structural requirements for potent HDAC
inhibition are a hydroxamic acid ZBG, a connecting unit linker
region and a hydrophobic dimethyl substituted phenyl ring as a
suitable CAP group. The most potent compounds in the
cellular HDAC assay were 19d and 19i showing IC50 values of
0.60/0.50 and 0.65/0.32 μM, respectively, which is about
equipotent to the reference compounds vorinostat and TSA.
In consideration of cytotoxic and HDAC inhibition activity,
19e, 19h, and 19i were the most active compounds of this
series. Interestingly, all three compounds showed a comparable
cytotoxic activity at A2780 and its cisplatin resistant subclone
A2780 CisR with a slightly higher pan-HDAC inhibition at
A2780 CisR (19e, 1.20-fold; 19h, 1.83-fold; 19i, 2.03-fold).
This effect was similar for TSA and vorinostat. To further
investigate the activity of 19e, 19h, and 19i in chemoresistant
cell lines, the cytotoxicity and HDAC inhibitory activities of
19e, 19h, and 19i were tested at the human triple negative
breast cancer cell line MDA-MB-231, the human tongue cancer
cell line Cal27, the human esophagus cell line Kyse510, and
their cisplatin resistant subclones. The results are shown in
Table 2. 19e, 19h, 19i, and vorinostat showed similar inhibition
in the parental and the cisplatin-resistant subclones in MTT
and HDAC assays except for a 2.3- to 2.8-fold higher HDAC
inhibition in Kyse510 CisR compared to Kyse sens. However,
the slightly higher inhibition of HDAC in Kyse CisR is not
leading to a higher cytotoxicity in Kyse CisR. Again, 19i was the
most potent compound and displayed a higher cytotoxicity and
HDAC inhibitory activity than the reference compound
vorinostat.
Ph
23.5
67.8
>100
>100
>100
8.47
9.05
8.05
10.3
7.63
10.8
27.1
6.34
23.0
8.33
2.43
62.6
87.9
8.33
16.4
9.53
4.13
2.10
4.09
25.6
1.54
0.49
34.9
62.5
35.0
1.56
2.42
0.29
72.5
73.1
>100
>100
>100
5.47
9.34
10.2
18.2
7.87
11.6
58.8
8.80
30.1
8.12
5.11
35.2
>100
5.72
60.3
30.8
4.96
2.25
5.00
33.8
1.78
0.32
42.4
86.9
84.6
13.4
3.12
0.22
>100
>100
nd
>100
>100
nd
8b
4-F-Ph
8c
4-CH3-Ph
4-CH3O-Ph
1-naphthyl
Ph
8d
nd
nd
8e
nd
nd
16a
16b
16c
16d
16e
16f
1.70
2.04
1.45
6.64
1.06
1.07
2.86
1.41
11.7
7.63
5.27
15.2
>100
1.58
2.57
2.09
0.60
1.58
4.21
17.1
1.61
0.65
5.86
34.0
34.0
ne
1.71
1.66
1.43
5.73
0.90
0.85
2.05
1.32
8.86
5.38
4.64
16.6
>100
1.26
2.00
1.95
0.50
0.90
3.46
15.4
0.88
0.32
4.95
43.9
43.9
ne
4-F-Ph
4-CH3-Ph
4-CH3O-Ph
1-naphthyl
3-chinolyl
4-i-Pr-Ph
4-t-Bu-Ph
2-CH3-Ph
3-CH3-Ph
3,5-CH3-Ph
Ph
16g
16h
16i
16j
16k
17
18
Ph
19a
19b
19c
19d
19e
19f
Ph
4-F-Ph
4-CF3-Ph
4-i-Pr-Ph
3-CH3-Ph
4-CH3-Ph
2-CH3-Ph
3,4-CH3-Ph
3,5-CH3-Ph
Ph
19g
19h
19i
20
28
HDAC IC50 Profiling. The three most cytotoxic compounds
19e, 19h, and 19i were selected for a HDAC profiling against
seven HDAC isoforms (1, 2, 4, 5, 6, 8, and 11). Vorinostat and
TSA were used as reference compounds (Table 3).
The isoform assays show that compounds 19e, 19h, and 19i
display a profile against HDACs 1, 2, 8, 11 similar to vorinostat.
However, 19e, 19h, and 19i showed increased inhibitory
activity at HDACs 4, 5, and 6. 19h displayed the highest activity
at HDAC6 with an IC50 of 18.1 nM, whereas 19e and 19i
showed selectivity for HDAC isoforms 4 and 5. 19i exhibited
the highest activity on HDAC4 with an IC50 of 11.9 nM. 19e
29
cisplatin
vorinostat
TSA
0.53
0.43
0.44
0.25
a
nd = not determined. ne = no effect up to 316 μM. Values are the
mean of three experiments. The standard deviations are <10% of the
mean.
increase in cytotoxic activity. The majority of the compounds
showed IC50 values in the micromolar range. The most active
D
dx.doi.org/10.1021/jm301254q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
When 19i was docked, the compound showed more
favorable interactions with the catalytic zinc ion of HDAC4
than of HDAC8, as demonstrated by distances between the
oxygen of the hydroxy group of the hydroxamic acid and the
zinc ion of <2 Å (<3 Å) found in 6% (31%) of the dockings to
HDAC4 and in 0% (4%) of the dockings to HDAC8. These
differences arise from differences in the binding modes of the
ligand (Figure 3). In the case of HDAC4, 19i orients its phenyl
Table 2. Cytotoxicity (MTT Assay) (A) and HDAC
Inhibition Activity (B) of 19e, 19h, 19i, Cisplatin, and
Vorinostat against Human Cancer Cell Lines with Different
a
Sensitivity Towards Cisplatin
(A) MTT Assay
IC50 [μM] of cell line
MDA-MB-231
Cal27
Kyse510
compd
19e
sens
CisR
sens
CisR
sens
CisR
2.79
2.77
1.37
1.66
35.7
3.81
3.77
1.68
2.69
96.5
2.97
2.90
1.03
2.64
19.2
3.72
4.31
1.81
2.08
43.9
8.19
10.0
2.96
4.62
2.49
8.21
9.46
2.48
4.66
8.44
19h
19i
vorinostat
cisplatin
(B) HDAC Assay
IC50 [μM] of cell line
MDA-MB-231
Cal27
Kyse510
compd
sens
CisR
sens
CisR
sens
CisR
Figure 3. Binding mode of 19i in HDAC4 (A, PDB code 2VQW) and
HDAC8 (B, PDB code 2V5W) as predicted by docking. Amino acids
discussed in the text are labeled and shown in stick representation. The
zinc ion is shown as orange sphere, and the residues coordinating it are
in stick representation.
19e
1.41
1.57
0.46
0.61
ne
1.28
1.66
0.41
0.61
ne
1.00
1.10
0.36
0.36
ne
0.88
0.91
0.36
0.61
ne
2.62
2.82
1.00
0.70
ne
1.13
1.10
0.35
0.59
ne
19h
19i
vorinostat
cisplatin
a
ne = no effect up to a concentration of 316 μM. Data are IC50 values
in μM. Values are the mean of three experiments. The standard
deviations are <10% of the mean.
moiety into a groove formed by loop α1−α2, loop β4−α8, and
loop β5−α9. In the case of HDAC8, 19i orients this moiety
into a shallow indentation formed by loop β7−β8 and loop
β9−α15. The respective other binding region is sterically
occupied in both cases.
showed the highest activity on HDAC5 with an IC50 of 2.31
nM. 19e and 19i showed a 11- to 13-fold selectivity for
HDAC5 over HDAC6.
These differences in the binding modes provide an
explanation for the observed selectivity of 19i (Figure 3). In
HDAC4, a charge-assisted hydrogen bond between the
alkoxyamide nitrogen and the carboxylate group of Asp115 is
formed (O−N distance, 2.6 Å), which presumably is enforced
by the additional polarization of the N−H bond due to the
presence of the N-alkoxy moiety. In contrast, in HDAC8, a
hydrogen bond is formed between the alkoxyamide nitrogen
and the sulfur atom of Met274 (S−N distance, 3.5 Å). The
latter is a much weaker hydrogen bond acceptor than the
carboxylate group. These findings suggest that the novel
alkoxyamide connecting unit linker region contributes to the
selectivity of 19i. The binding mode in HDAC4 also explains
why the inhibition activity of 19h against this isoform is lower
by factors of ∼5 and ∼8 than those of 19e and 19i (Table 3).
In the case of 19h, the 4-methyl group at the phenyl ring would
sterically interfere with Pro106, which would lead to a shift of
the alkoxyamide moiety and, hence, likely to a breaking of the
charge-assisted hydrogen bond with Asp115. In contrast, 19e
and 19i show a similar inhibition activity against HDAC4,
which is in agreement with the fact that one of the two methyl
Binding Mode of 19i in HDAC4 and HDAC8. In order to
understand differences in the inhibition activities of compounds
19e, 19h, and 19i with respect to HDAC4 and HDAC8, 19i,
which shows the highest selectivity for HDAC4 over HDAC8,
was docked into the crystal structures of HDAC4 (PDB code
2VQW) and HDAC8 (PDB code 2V5W) using AutoDock314
as a docking engine and DrugScore15 as an objective function.
Both HDAC structures have a “closed” conformation of the
structural zinc-binding domain16 and have been used
successfully for docking previously.17 For evaluating the applied
docking approach, we initially performed cross-docking of the
hydroxamic acid ligands taken from PDB code 2VQM
(HDAC4) (1W22 (HDAC8)) to the structure of 2VQW
(2V5W). Despite the fact that cross-docking is a more stringent
test than re-docking,18 in both cases good docking solutions
(root-mean-square deviation of the ligand atoms after super-
imposing the respective protein structures was 2.71 Å (2.42 Å))
were identified in the second largest (largest) cluster, validating
the combination AutoDock3/DrugScore as a suitable docking
approach.
Table 3. Inhibition Activities (IC50 [nM]) of Compounds 19e, 19h, 19i, Vorinostat, and TSA against HDAC Isoforms 1, 2, 4, 5,
6, 8, and 11
IC50 [nM] of HDAC isoforms
compd
19e
1
2
4
5
6
8
11
341
320
320
320
10.9
1392
944
881
920
32.1
16.3
2.31
26.9
18.1
55.7
160
1043
833
2135
997
852
480
31.0
19h
82.6
33.4
19i
11.9
4.22
1278
960
vorinostat
TSA
48300
8439
20000
6155
1.50
196
E
dx.doi.org/10.1021/jm301254q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
groups of 19i is oriented toward the solvent; hence, removing
it, as done in 19e, has a small effect only on the inhibition
activity. In contrast, because of the more solvent-exposed
binding mode of 19i in HDAC8, neither removing the second
methyl group (19e) nor shifting one group into the 4-position
(19h) is expected to have a profound effect on the inhibition
activity, in line with observed differences of a factor of <1.6
(Table 3).
In summary, docking studies suggest distinct binding modes
of 19i in HDAC4 and HDAC8. These binding modes are in
line with and provide structural explanations for the observed
selectivity of 19i for HDAC4 over HDAC8 as well as the
differences in the inhibition activities of 19e, 19h, and 19i with
respect to HDAC4 and HDAC8, respectively.
Table 5. Synergism Studies between Cisplatin and 19i and
Vorinostat
a
(A) MDA-MB-231 CisR
19i [μM]
0.5
vorinostat [μM]
0.5
cisplatin 0.25
1
1.5
0.25
1
1.5
5
∗
∗
∗
∗
∗
∗
0.79
0.88
0.81
∗
0.82
0.79
0.73
0.84
0.80
∗
∗
∗
∗
∗
∗
∗
∗
∗
∗
0.88
1.00
0.84
∗
∗
0.83
0.91
0.88
10
25
40
50
0.85
0.72
0.78
0.75
0.90
0.92
0.93
(B) Cal27 CisR
19i [μM]
0.5
vorinostat [μM]
0.5
cisplatin 0.25
1
1.5
0.25
1
1.5
Enhancement of Cisplatin-Induced Toxicity by 19i.
Vorinostat and 19i alone showed a concentration-dependent
cytotoxic effect on different cancer cell lines (Table 1 and Table
2A). On the basis of their potent cytotoxic and HDAC
inhibitory activity, we investigated a possible enhancement of
the cytotoxic effect of cisplatin in combination with 19i in four
different cisplatin resistant cell lines (Cal27 CisR, A2780 CisR,
Kyse510 CisR, and MDA-MB231 CisR). Vorinostat was used
as reference compound. The cell lines were pretreated with
vorinostat or 19i for 48 h followed by a treatment with cisplatin
for 72 h.
1
∗
∗
∗
0.85
∗
∗
0.45
0.41
0.31
∗
0.48
0.34
0.22
0.20
0.23
∗
∗
∗
∗
∗
∗
∗
∗
∗
∗
0.65
0.53
0.29
∗
2.5
5
0.38
0.26
0.24
0.27
0.93
0.37
0.30
0.31
10
45
0.27
0.39
0.38
(C) Kyse510 CisR
19i [μM]
vorinostat [μM]
cisplatin 0.25
0.5
1
1.5
0.25
0.5
1
1.5
0.5
0.75
1
0.18
0.10
0.13
0.18
0.18
0.13
0.07
0.08
0.09
0.16
0.08
0.05
0.05
0.09
0.18
0.06
0.05
0.05
0.12
0.16
0.71
0.51
0.47
0.46
0.24
0.52
0.44
0.42
0.50
0.23
0.66
0.67
0.56
0.55
0.27
0.75
0.68
0.60
0.51
0.42
The combination of cisplatin and vorinostat or 19i markedly
enhanced the sensitivity of platinum-resistant cell lines to
cisplatin (Table 4). IC50 values for cisplatin in combination
3.16
8
(D) A2780 CisR
19i [μM]
Table 4. IC50 [μM] of Cisplatin in Cisplatin Resistant A2780
CisR, Cal27 CisR, Kyse510 CisR, and MDA-MB231 CisR
Cell Lines after Treatment with Cisplatin or in Combination
vorinostat [μM]
cisplatin 0.25
0.5
1
1.5
0.25
0.5
1
1.5
a
1
∗
0.85
0.73
0.66
0.69
0.58
0.82
0.64
0.59
0.61
0.49
0.97
0.81
0.74
0.68
0.64
∗
∗
∗
∗
∗
0.86
0.74
0.74
0.77
0.62
0.86
0.73
0.69
0.67
0.56
with 1 μM Vorinostat or 1 μM 19i
5
0.85
0.78
0.78
0.76
0.91
0.79
0.72
0.68
cisplatin + vori-
8
nostat
cisplatin + 19i
10
20
cell line
cisplatin, IC50
IC50
SF
IC50
SF
0.79
a
A2780 CisR
13.4
43.9
8.44
96.5
5.80
2.53
0.78
65.8
2.31
17.4
10.8
1.47
3.33
2.03
0.45
27.3
4.02
21.6
18.8
3.53
The cisplatin-resistant tumor cell lines MDA-MB-231 (A), Cal27
Cal27 CisR
CisR (B), Kyse510 CisR (C), and A2780 CisR (D) were treated with
different combinations of cisplatin and 19i/vorinostat. CI (combina-
tion index) was calculated using CalcuSyn 2.1 based on the Chou−
Talalay method. CI > 1.0 indicates antagonism. CI = 1 indicates an
additive effect, and CI < 0.9 indicates synergism. ∗ = fraction affected
less than 0.20. Values are the mean of three experiments. Standard
deviation is <10% of the mean.
Kyse510 CisR
MDA-MB231 CisR
a
SF means shift factor and was calculated by dividing the IC50 of
cisplatin alone by the IC50 of the corresponding drug combination.
Values are the mean of three experiments. The standard deviations are
<10% of the mean.
with vorinostat or 19i (Table 4) were even lower than IC50
values for the corresponding parental cell lines (Table 2A). The
combination of cisplatin and 19i was more effective than the
cisplatin−vorinostat combination, as can be seen from higher
shift factors (SF) for the cisplatin/19i combination (Table 4).
A remarkably high chemosensitization against cisplatin was
observed for the HNSCC cell lines Cal27 CisR and Kyse510
CisR with shift factors of 21.6 and 18.8, respectively.
Furthermore, the combination of cisplatin and 19i produced
a higher reduction in cell viability in comparison to the
combination of cisplatin and vorinostat in all cell lines.
Accordingly, 19i is a novel nanomolar HDAC inhibitor that
reverts cisplatin resistance in different human cancer cell lines.
In order to gather additional information about the
interaction between cisplatin and 19i as well as between
cisplatin and vorinostat, concentration−effect analyses were
performed (Table 5). The concentrations of 19i and vorinostat
used were 0.25, 0.5, 1.0, and 1.5 μM for all cell lines, while the
concentrations of cisplatin were selected for each tumor cell
line from the cisplatin-induced growth inhibition curve. The
highest cisplatin concentration represents the IC50 value,
whereas the other four used concentrations were below the
IC50 values of cisplatin.
19i and vorinostat enhanced in a concentration-dependent
manner the cytotoxic effect of cisplatin. The combination index
analysis based on the Chou−Talalay method revealed a
synergistic interaction with cisplatin (CI < 0.9) for both
compounds. All four cisplatin-resistant tumor cell lines showed
higher synergistic effects (=lower CIs) for the combination of
cisplatin and 19i in comparison to the cisplatin vorinostat
combination, except for 1.5 μM vorinostat in A2780 CisR.
Especially in the cisplatin-resistant HNSCC cell lines Cal27
CisR and Kyse510, 19i enhanced markedly the cytotoxicity of
cisplatin. The CI for 19i in Kyse510 CisR is less than 0.2, which
F
dx.doi.org/10.1021/jm301254q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
analysis was performed on a UHR-TOF maXis 4G, Bruker Daltonics,
Bremen instrument. If necessary, the purity was determined by HPLC:
instrument, Elite LaChrom system [Hitachi L-2130 (pump) and L-
2400 (UV detector)]; column, Phenomenex Luna C-18(2) 1.8 μm
particle (250 mm × 4.6 mm), supported by Phenomenex Security
Guard cartridge kit C18 (4.0 mm × 3.0 mm). The purity of all final
compounds determined by HPLC was 95% or higher. Compounds
2a,b and 21 have been prepared according to known procedures.20
Experimental Data. General procedures for the synthesis of
HDAC inhibitors and compound characterization data for selected
compounds 16c, 17, 18, 19i, 20, 28, and 29 are given below. The
synthesis of all other compounds is reported in the Supporting
Information.
indicates a very strong synergism, whereas the CI for vorinostat
indicates a moderate synergism. This is in accordance with the
observed higher shift factors for 19i in HNSCC in comparison
to vorinostat (Table 4). The synergistic effect of 19i was not
only observed for the combination with cisplatin in a
concentration of 1 μM (a concentration near the IC50 of
19i), it was also detected for lower concentrations (0.25 and 0.5
μM). Even in combinations of 19i with low-dose cisplatin a
highly synergistic interaction was detected. On the basis of the
results, we draw the conclusion that 19i acts with cisplatin in a
synergistic manner and enhances the cytotoxicity of cisplatin to
a greater extent than vorinostat, especially in HNSCC cell lines.
Synthesis of HDAC Inhibitors. General Procedure for the
Preparation of Target Compounds 16a−k, 17, 18, 19a−i, 20,
28, and 29. A solution of the respective O-benzyl-protected
hydroxamic acid or anilide intermediate (2 mmol) in dry THF (150
mL) or methanol (150 mL, in the cases of compounds 28 and 29) was
hydrogenated (1 bar) at room temperature in the presence of a
catalytic amount of Pd−C (10 wt %). Upon completion, the crude
mixture was filtered through Celite to remove the catalyst and the
filtrate was concentrated under reduced pressure. The residue
crystallized upon storage in a freezer. The residue was purified by
column chromatography using ethyl acetate or ethyl acetate/methanol
(9:1) as eluent.
CONCLUSION
■
On the basis of a widely accepted HDAC pharmacophore
model, we developed new potent hydroxamate-based HDAC
inhibitors with a novel alkoxyamide connecting unit linker
region. Their cytotoxicity and HDAC inhibitory activity were
determined in different human cisplatin sensitive and resistant
cancer cell lines by MTT and HDAC inhibition assays. 19i
(LMK235) (N-((6-(hydroxyamino)-6-oxohexyl)oxy)-3,5-dime-
thylbenzamide) was identified as the most cytotoxic compound
and displayed equipotent HDAC inhibition in the pan-HDAC
assay compared to vorinostat. In contrast to vorinostat, 19i
showed a novel HDAC isoform selectivity profile with
preference for HDAC4 and HDAC5, which are inhibited
with low nanomolar IC50 values. A binding mode of 19i in
HDAC4 predicted by docking provides structural explanations
for the observed selectivity of 19i for HDAC4. The enhanced
cytotoxicity of 19i in the investigated cancer cell lines may be
related to this novel selectivity profile. The combination of 19i
with cisplatin enhanced markedly the cisplatin sensitivity of our
cisplatin resistant sublines. Pretreatment with 19i 48 h prior to
cisplatin led to a complete resensitization of the platinum
resistant cancer cell lines. Most notably, the combination of
cisplatin and 19i was more effective than the cisplatin
vorinostat combination. Therefore, it can be assumed that
HDAC4 and HDAC5 inhibition may contribute to this
chemosensitizing effect of 19i. Our findings are in agreement
with recent results of Stronach et al. who showed that silencing
of HDAC4 led to an increase in cisplatin sensitivity.19 We
believe that especially compound 19i is a promising starting
point for further structural optimization. As a class IIa selective
HDAC inhibitor, 19i could be a promising tool to investigate
the function of these isoforms concerning the molecular
mechanism of cancer development and treatment.
N-Hydroxy-6-((3-(p-tolyl)ureido)oxy)hexanamide (16c). Re-
1
action time: 12 h. White solid; yield 67%; mp 121 °C. H NMR
(500.13 MHz, DMSO-d6): δ = 1.26−1.35 (m, 2H), 1.46−1.65 (m,
4H), 1.96 (t, J = 7.3 Hz, 2H), 2.24 (s, 3H), 3.74 (t, J = 6.7 Hz, 2H),
7.06 (d, J = 8.3 Hz, 2H), 7.43 (d, J = 8.3 Hz, 2H), 8.56 (s, 1H), 8.70
́
(s, 1H), 9.36 (s, 1H), 10.35 (s, 1H) ppm. 13C NMR (125.76 MHz,
DMSO-d6): δ = 20.25, 24.83, 27.15, 32.09, 75.54, 119.54, 128.74,
̃
136.30, 157.06, 168.92 ppm. IR (KBr): ν = 3323, 3208 (NH), 2937,
2869 (CH2), 1659 (CO) cm−1. Anal. Calcd for C14H21N3O4: C
56.94, H 7.17, N 14.23. Found: C 57.00, H 7.35, N 13.95.
N-Hydroxy-5-((3-phenylureido)oxy)pentanamide (17). Reac-
1
tion time: 4 h. White solid; yield 41%; mp 83 °C. H NMR (500.13
MHz, DMSO-d6): δ = 1.51−1.68 (m, 4H), 1.99 (t, J = 6.7 Hz, 2H),
3.76 (t, J = 5.9 Hz, 2H), 6.99 (t, J = 7.3 Hz, 1H), 7.26 (t, J = 7.8 Hz,
2H), 7.56 (d, J = 7.8 Hz, 2H), 8.69 (s, 1H), 8.70 (s, 1H), 9.43 (s, 1H),
10.37 (s, 1H) ppm. 13C NMR (125.76 MHz, DMSO-d6): δ = 21.36,
26.96, 31.85, 75.30, 119.45, 122.32, 128.33, 138.91, 157.01, 168.84
ppm. IR (KBr): ν
̃
= 3214 (NH), 2924, 2877 (CH2), 1662 (CO)
cm−1. Anal. Calcd for C12H17N3O4: C 53.92, H 6.41, N 15.72. Found:
C 53.72, H 6.69, N 15.42.
N-Hydroxy-N-methyl-6-((3-phenylureido)oxy)hexanamide
(18). Reaction time: 4 h. White solid; yield 63%; mp 93 °C. 1H NMR
(500.13 MHz, DMSO-d6): δ = 1.28−1.41 (m, 2H), 1.46−1.57 (m,
2H), 1.58−1.68 (m, 2H), 2.35 (t, J = 7.1 Hz, 2H), 3.08 (s, 3H), 3.76
(t, J = 6.7 Hz, 2H), 6.95−7.03 (m, 1H), 7.26 (t, J = 7.9 Hz, 2H), 7.54−
7.59 (m, 2H), 8.66 (s, 1H), 9.43 (s, 1H), 9.76 (s, 1H) ppm. 13C NMR
(125.76 MHz, DMSO-d6): δ = 23.96, 25.03, 27.27, 31.34, 35.60, 75.63,
̃
119.47, 122.33, 128.34, 138.91, 157.00, 172.77 ppm. IR (KBr): ν =
EXPERIMENTAL SECTION
■
3222 (NH), 2934, 2865 (CH2), 1660 (CO) cm−1. Anal. Calcd for
C14H21N3O4: C 56.94, H 7.17, N 14.23. Found: C 56.79, H 7.29, N
14.02.
1. Chemistry. General. All solvents and chemicals were used as
purchased without further purification. The progress of all reactions
was monitored on Merck precoated silica gel plates (with fluorescence
indicator UV254) using ethyl acetate/n-hexane as solvent system.
Column chromatography was performed with Fluka silica gel 60
(230−400 mesh ASTM) with the solvent mixtures specified in the
corresponding experiment. Spots were visualized by irradiation with
ultraviolet light (254 nm). Melting points were taken in open
capillaries on a Mettler FP 5 melting point apparatus and are
uncorrected. Proton (1H) and carbon (13C) NMR spectra were
N-((6-(Hydroxyamino)-6-oxohexyl)oxy)-3,5-dimethylbenza-
mide (19i). Reaction time: 4 h. Purple solid; yield 58%; mp 132 °C.
1H NMR (500.13 MHz, DMSO-d6): δ = 1.29−1.39 (m, 2H), 1.48−
1.64 (m, 4H), 1.96 (t, J = 7.3 Hz, 2H), 2.30 (s, 6H), 3.84 (t, J = 6.4
Hz, 2H), 7.17 (s, 1H), 7.35 (s, 2H), 8.67 (s, 1H), 10.35 (s, 1H), 11.50
(s, 1H) ppm. 13C NMR (125.76 MHz, DMSO-d6): δ = 20.70, 24.81,
24.98, 27.34, 32.10, 74.94, 124.66, 132.36, 132.62, 137.44, 164.37,
̃
168.92 ppm. IR (KBr): ν = 3204 (NH), 2941, 2866 (CH2), 1671,
1
1625 (CO) cm−1. Anal. Calcd for C15H22N2O4: C 61.21, H 7.53, N
recorded on a Bruker Avance 500 (500.13 MHz for H; 125.76 MHz
for 13C) using DMSO-d6 and CDCl3 as solvents. Chemical shifts are
given in parts per million (ppm) (δ relative to residual solvent peak for
1H and 13C and to external tetramethylsilane). Elemental analysis was
performed on a Perkin Elmer PE 2400 CHN elemental analyzer. IR
spectra were recorded on a Varian 800 FT-IR Scimitar series. HRMS
9.52. Found: C 61.47, H 7.71, N 9.73.
N-((6-(Hydroxyamino)-6-oxohexyl)oxy)-2-phenylhydrazine-
carboxamide (20). Reaction time: 6 h. Yellow solid; yield 12%; mp
100 °C. 1H NMR (500.13 MHz, DMSO-d6): δ = 1.24−1.34 (m, 2H),
1.46−1.65 (m, 4H), 1.95 (t, J = 7.3 Hz, 2H), 3.72 (t, J = 6.6 Hz, 2H),
G
dx.doi.org/10.1021/jm301254q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
6.65−6.73 (m, 3H), 7.13 (t, J = 7.7 Hz, 2H), 7.47 (s, 1H), 8.60 (s,
1H), 8.66 (s, 1H), 9.38 (s, 1H), 10.34 (s, 1H) ppm. 13C NMR (125.76
MHz, DMSO-d6): δ = 24.84, 24.88, 27.20, 32.10, 75.46, 111.91,
Combination Experiments. For the investigation of the effect of
19i or vorinostat on the cisplatin induced cytotoxicity, 19i or
vorinostat was added 48 h before cisplatin. After 72 h, the cytotoxic
effect was determined with a MTT cell viability assay.
Whole-Cell HDAC Inhibition Assay. The cellular HDAC assay
was based on an assay published by Ciossek et al.25 and Bonfils et al.26
with minor modifications.
̃
118.14, 128.50, 149.77, 159.70, 168.95 ppm. IR (KBr): ν = 3237
(NH), 2937, 2866 (CH2), 1658 (CO) cm−1. Anal. Calcd for
C13H20N4O4: C 52.69, H 6.80, N 18.91. Found: C 52.49, H 7.04, N
18.76.
Briefly, human cancer cell lines Cal27sens/Cal27 CisR, Kyse510-
sens/Kyse510 CisR, A2780/A2780 CisR, and MDA-MB231sens/CisR
were seeded in 96-well tissue culture plates (Corning, Germany) at a
density of 1.5 × 104 cells/well in a total volume of 90 μL of culture
medium. After 24 h, cells were incubated for 18 h with increasing
concentrations of test compounds. The reaction was started by adding
10 μL of 3 mM Boc-Lys(ε-Ac)-AMC (Bachem, Germany) to reach a
final concentration of 0.3 mM. The cells were incubated with the Boc-
Lys(ε-Ac)-AMC for 3 h under cell culture conditions. After this
incubation, 100 μL/well stop solution (25 mM Tris-HCl (pH 8), 137
mM NaCl, 2.7 mM KCl, 1 mM MgCl2, 1% NP40, 2.0 mg/mL trypsin,
10 μM vorinostat) was added and the mixture was developed for 3 h
under cell culture conditions. Fluorescence intensity was measured at
an excitation of 320 nm and emission of 520 nm in a NOVOstar
microplate reader (BMG LabTech, Offenburg, Germany).
HDAC IC50 Profiling. The in vitro inhibitory activity of compounds
19e, 19h, and 19i against seven human HDAC isoforms (1, 2, 4 C2A,
5 C2A, 6, 8, and 11) were performed at Reaction Biology Corp.
(Malvern, PA) with a fluorescent based assay according to the
company’s standard operating procedure. The IC50 values were
determined using 10 different concentrations with 3-fold serial dilution
starting at 10 μM. TSA and vorinostat were used as reference
compounds.
Combination Index Analysis. The combined effect of cisplatin
and 19i or cisplatin and vorinostat was analyzed using the MTT assay.
Cell viability was determined from each well relative to the average
absorbance of control wells. The combination indexes (CIs) were
calculated using CalcuSyn 2.1 software (Biosoft, Cambridge, U.K.)
based on the Chou−Talalay method.27 CI > 1 indicates antagonism.
CI = 1 indicates an additive effect, and CI < 1 indicates synergism.
Data Analysis. Concentration−effect curves were constructed with
Prism 4.0 (GraphPad, San Diego, CA) by fitting the pooled data of at
least three experiments performed in triplicate to the four-parameter
logistic equation.
3. Docking Studies. The coordinates of the protein structures
were retrieved from the Protein Data Bank (PDB). For cross-docking,
the HDAC4 and HDAC8 structures were superimposed based on their
Cα atoms. In all cases, all water molecules were removed as were all
ions except for the catalytic zinc ion. In the case of multiple protein
chains in the PDB file, only the first one was retained. Ligand
coordinates were either extracted from the PDB files (for the cross-
docking) or generated by Moloc28 (19i). Sybyl atom types were
manually assigned to all ligands. The ligands were further prepared by
autotors for assigning rotatable bonds. As a docking engine,
AutoDock3.0.514 was used together with the distance-dependent pair
potentials of DrugScore15a as an objective function as described in ref
15b. The grids were dimensioned sufficiently large to extend at least 7
Å beyond any atom of the ligands used for cross-docking in their
crystallographic binding modes. The same grids were then used for
docking 19i.
N-(2-Aminophenyl)-6-((3-phenylureido)oxy)hexanamide
1
(28). Reaction time: 6 h. Yellow solid; yield 46%; mp 128 °C. H
NMR (500.13 MHz, DMSO-d6): δ = 1.35−1.44 (m, 2H), 1.58−1.72
(m, 4H), 2.33 (t, J = 7.4 Hz, 2H), 3.79 (t, J = 6.6 Hz, 2H), 4.82 (s,
2H), 6.51−6.56 (m, 1H), 6.71 (d, J = 7.9 Hz, 1H), 6.86−6.92 (m,
1H), 6.98 (t, J = 7.4 Hz, 1H), 7.16 (d, J = 7.8, 1H), 7.25 (t, J = 7.9 Hz,
2H), 7.56 (d, J = 8.2 Hz, 2H), 8.67 (s, 1H), 9.11 (s, 1H), 9.45 (s, 1H)
ppm. 13C NMR (125.76 MHz, DMSO-d6): δ = 24.92, 25.01, 27.23,
35.58, 75.62, 115.77, 116.05, 119.44, 122.21, 123.44, 125.18, 125.58,
̃
128.33, 138.91, 141.77, 156.99, 170.96 ppm. IR (KBr): ν = 3331, 3223
(NH), 2939, 2864 (CH2), 1658 (CO) cm−1. Anal. Calcd for
C19H24N4O3: C 64.03, H 6.79, N 15.72. Found: C 63.96, H 6.69, N
15.82.
N-(2-Hydroxyphenyl)-6-((3-phenylureido)oxy)hexanamide
(29). Reaction time: 6 h. White solid; yield 42%; mp 80 °C. 1H NMR
(500.13 MHz, DMSO-d6): δ = 1.29−1.43 (m, 2H), 1.56−1.74 (m,
4H), 2.41 (t, J = 7.3 Hz, 2H), 3.78 (t, J = 6.6 Hz, 2H), 6.75 (t, J = 7.6
Hz, 1H), 6.85 (d, J = 7.9 Hz, 1H), 6.90−7.01 (m, 2H), 7.25 (t, J = 7.9
Hz, 2H), 7.56 (d, J = 8.2 Hz, 2H), 7.68 (d, J = 7.8 Hz, 1H), 8.66 (s,
1H), 9.25 (s, 1H), 9.44 (s, 1H), 9.74 (s, 1H) ppm. 13C NMR (125.76
MHz, DMSO-d6): δ = 24.89, 24.98, 27.22, 35.73, 75.62, 115.81,
118.86, 119.44, 122.22, 122.32, 124.49, 126.29, 128.33, 138.90, 147.74,
̃
156.99, 171.73 ppm. IR (KBr): ν = 3349, 3176 (NH), 2951, 2862
(CH2), 1661 (CO) cm−1. Anal. Calcd for C19H23N3O4: C 63.85, H
6.49, N 11.76. Found: C 63.70, H 6.61, N 11.75.
2. Biological Evaluation. Reagents. Trichostatin A and cisplatin
were purchased from Invivogen (France) and Sigma (Germany).
Vorinostat was synthesized according to known procedures.21 All
other reagents were supplied by PAN Biotech (Germany) unless
otherwise stated.
Cell Lines and Cell Culture. The human ovarian carcinoma cell
line A2780 was obtained from European Collection of Cell Cultures
(ECACC, Salisbury, U.K.). The human tongue cell line Cal27 and the
human esophagus cell line Kyse510 were obtained from the German
Collection of Microorganisms and Cell Cultures (DSMZ, Germany).
The human triple negative breast cancer cell line MDA-MB-231 was
obtained from American Type Culture Collection (ATCC, U.S.). The
cisplatin resistant CisR cell lines were generated by exposing the
parental cell lines to weekly cycles of cisplatin in a IC50 concentration
over a period of 24−30 weeks as described by Gosepath et al.22 and
Eckstein et al.23
All cancer cell lines were grown at 37 °C under humidified air
supplemented with 5% CO2 in RPMI 1640 (A2780, Kyse510) or
DMEM (Cal27, MDA-MB-231) containing 10% (MDA-MB-231,
15%) fetal calf serum, 120 IU/mL penicillin, and 120 μg/mL
streptomycin. The cells were grown to 80% confluency before using
them for the appropriate assays.
MTT Cell Viability Assay. The rate of cell survival under the
action of test substances was evaluated by an improved MTT assay as
previously described.24 The assay is based on the ability of viable cells
to metabolize yellow 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazo-
lium bromide (MTT, Applichem, Germany) to violet formazan that
can be detected spectrophotometrically. In brief, A2780, Cal27,
Kyse510, and MDA-MB-231 cell lines were seeded at a density of
5000, 7000, 8000, and 10 000 cells/well in 96-well plates (Corning,
Germany). After 24 h, cells were exposed to increased concentrations
of the test compounds. Incubation was ended after 72 h, and cell
survival was determined by addition of MTT solution (5 mg/mL in
phosphate buffered saline). The formazan precipitate was dissolved in
DMSO (VWR, Germany). Absorbance was measured at 544 and 690
nm in a FLUOstar microplate reader (BMG LabTech, Offenburg,
Germany).
Docking was performed using the Lamarkian genetic algorithm.
Each docking experiment was repeated 100 times, yielding 100 docked
protein−ligand configurations. The following parameters were used for
docking: a population size of 100, a random starting position and
conformation, a maximal effect of mutation of 2 Å for translation and
50° for rotation, a mutation rate of 0.02, and a crossover rate of 0.80.
All docking simulations were performed with a maximum of 50 000
generations. The number of energy evaluations was set to 1.0 × 106.
Finally, all docked conformations were clustered using a tolerance of
1.0 Å rmsd.
For the cross-docking, the docking accuracy was evaluated by
calculating the rmsd between docked ligand poses and the crystallo-
graphic binding modes. The ligand pose found on the first scoring
H
dx.doi.org/10.1021/jm301254q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
rank of the largest or second largest cluster was chosen for this. For
docking of 19i, the first pose of the largest or second largest cluster
was chosen that showed a distance between the hydroxyl oxygen and
the zinc ion below 3 Å. These poses are depicted in Figure 3.
Cellamare, S.; Carotti, A.; Altucci, L.; Jung, M.; Sippl, W. Carbamate
prodrug concept for hydroxamate HDAC inhibitors. ChemMedChem
2011, 6, 1193−1198. (c) Schafer, S.; Saunders, L.; Schlimme, S.;
̈
Valkov, V.; Wagner, J. M.; Kratz, F.; Sippl, W.; Verdin, E.; Jung, M.
Pyridylalanine-containing hydroxamic acids as selective HDAC6
inhibitors. ChemMedChem 2009, 4, 283−290. (d) Stolfa, D. A;
Stefanachi, A.; Gajer, J. M.; Nebbioso, A.; Altucci, L.; Cellamare, S.;
Jung, M.; Carotti, A. Design, synthesis, and biological evaluation of 2-
aminobenzanilide derivatives as potent and selective HDAC inhibitors.
ChemMedChem 2012, 7, 1256−1266.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures and compound characterization data
for all intermediates and target compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
(7) (a) 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. (b) Vannini, A.; Volpari, C.;
Filocamo, G.; Casavola, E. C.; Brunetti, M.; Renzoni, D.; Chakravarty,
AUTHOR INFORMATION
Corresponding Author
*Phone: (+49) 21181-14985. Fax: (+49) 21181-13847. E-mail:
thomas.kurz@uni-duesseldorf.de.
■
P.; Paolini, C.; De Francesco, R.; Gallinari, P.; Steinkuhler, C.; Di
̈
Marco, S. Crystal structure of a eukaryotic zinc-dependent histone
deacetylase, human HDAC8, complexed with a hydroxamic acid
inhibitor. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 15064−15069.
(c) Vannini, A.; Volpari, C.; Gallinari, P.; Jones, P.; Mattu, M.; Carfí,
Author Contributions
†Linda Marek and Alexandra Hamacher contributed equally to
this publication.
Notes
A.; De Francesco, R.; Steinkuhler, C.; Di Marco, S. Substrate binding
̈
The authors declare no competing financial interest.
to histone deacetylases as shown by the crystal structure of the
HDAC8-substrate complex. EMBO Rep. 2007, 8, 879−884.
(8) Balasubramanian, S.; Verner, E.; Buggy, J. J. Isoform-specific
histone deacetylase inhibitors: the next step? Cancer Lett. 2009, 280,
211−221.
ACKNOWLEDGMENTS
■
The Deutsche Forschungsgemeinschaft (DFG) is acknowl-
edged for funds used to purchase the UHR-TOF maXis 4G,
Bruker Daltonics HRMS instrument used in this research.
(9) Thurn, K. T.; Thomas, S.; Moore, A.; Munster, P. N. Rational
therapeutic combinations with histone deacetylase inhibitors for the
treatment of cancer. Future Oncol. 2011, 7, 263−283.
(10) Noureen, N.; Rashid, H.; Kalsoom, S. Identification of type-
specific anticancer histone deacetylase inhibitors: road to success.
Cancer Chemother. Pharmacol. 2010, 66, 625−633.
(11) Bieliauskas, A. V.; Pflum, M. K. H. Isoform-selective histone
deacetylase inhibitors. Chem. Soc. Rev. 2008, 37, 1402−1413.
(12) Bots, M.; Johnstone, R. W. Rational combinations using HDAC
inhibitors. Clin. Cancer Res. 2009, 15, 3970−3977.
(13) Ong, P.-S.; Wang, X.-Q.; Lin, H.-S.; Chan, S.-Y.; Ho, P. C.
Synergistic effects of suberoylanilide hydroxamic acid combined with
cisplatin causing cell cycle arrest independent apoptosis in platinum-
resistant ovarian cancer cells. Int. J. Oncol. 2012, 40, 1705−1713.
(14) Morris, G.; Goodsell, D. Automated docking using a Lamarckian
genetic algorithm and an empirical binding free energy function. J.
Comput. Chem. 1998, 19, 1639−1662.
(15) (a) Gohlke, H.; Hendlich, M.; Klebe, G. Knowledge-based
scoring function to predict protein−ligand interactions. J. Mol. Biol.
2000, 295, 337−356. (b) Sotriffer, C. A.; Gohlke, H.; Klebe, G.
Docking into knowledge-based potential fields: a comparative
evaluation of DrugScore. J. Med. Chem. 2002, 45, 1967−1970.
(16) Bottomley, M. J.; Lo Surdo, P.; Di Giovine, P.; Cirillo, A.;
Scarpelli, R.; Ferrigno, F.; Jones, P.; Neddermann, P.; De Francesco,
DEDICATION
Dedicated to Linda Marek, in memoriam.
■
ABBREVIATIONS USED
■
CDT, 1,1′-carbonyldi(1,2,4-triazole); CisR, cisplatin resistant
subclone; DMAP, 4-dimethylaminopyridine; EDC, 1-ethyl-3-
(3-dimethylaminopropyl)carbodiimide hydrochloride; Et3N,
triethylamine; HDAC, histone deacetylase; HDACi, histone
deacetylase inhibitor; HSP90, heat shock protein 90; MTT, 3-
(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide;
Pd/C, palladium on activated carbon; rt, room temperature;
SAHA, suberoylanilide hydroxamic acid; sens, parental cell line
of the CisR subclone; THF, tetrahydrofuran; TSA, trichostatin
A
REFERENCES
■
(1) Biel, M.; Wascholowski, V.; Giannis, A. Epigeneticsan
epicenter of gene regulation: histones and histone-modifying enzymes.
Angew. Chem., Int. Ed. 2005, 44, 3186−3216.
(2) (a) Paris, M.; Porcelloni, M.; Binaschi, M.; Fattori, D. Histone
deacetylase inhibitors: from bench to clinic. J. Med. Chem. 2008, 51,
1505−1529. (b) Schemies, J.; Sippl, W.; Jung, M. Histone deacetylase
inhibitors that target tubulin. Cancer lett. 2009, 280, 222−232.
(3) Bertrand, P. Inside HDAC with HDAC inhibitors. Eur. J. Med.
Chem. 2010, 45, 2095−2116.
R.; Steinkuhler, C.; Gallinari, P.; Carfí, A. Structural and functional
̈
analysis of the human HDAC4 catalytic domain reveals a regulatory
structural zinc-binding domain. J. Biol. Chem. 2008, 283, 26694−
26704.
(17) (a) Oger, F.; Lecorgne, A.; Sala, E.; Nardese, V.; Demay, F.;
Chevance, S.; Desravines, D. C.; Aleksandrova, N.; Le Guevel, R.;
́
(4) Witt, O.; Deubzer, H. E.; Milde, T.; Oehme, I. HDAC family:
What are the cancer relevant targets? Cancer Lett. 2009, 277, 8−21.
(5) (a) Agis-Balboa, R. C.; Pavelka, Z.; Kerimoglu, C.; Fischer, A.
Loss of HDAC5 impairs memory function: implications for
Alzheimer’s disease. J. Alzheimer’s Dis. 2013, 33, 35−44. (b) Zhang,
H.; Xiao, Y.; Zhu, Z.; Li, B.; Greene, M. I. Immune regulation by
histone deacetylases: a focus on the alteration of FOXP3 activity.
Immunol. Cell. Biol. 2012, 90, 95−100. (c) Khan, O.; La Thangue, N.
B. HDAC inhibitors in cancer biology: emerging mechanisms and
clinical applications. Immunol. Cell Biol. 2012, 90, 85−94.
Lorenzi, S.; Beccari, A. R.; Barath, P.; Hart, D. J.; Bondon, A.;
Carettoni, D.; Simonneaux, G.; Salbert, G. Biological and biophysical
properties of the histone deacetylase inhibitor suberoylanilide
hydroxamic acid are affected by the presence of short alkyl groups
on the phenyl ring. J. Med. Chem. 2010, 53, 1937−1950.
(b) Brunsteiner, M.; Petukhov, P. A. Insights from comprehensive
multiple receptor docking to HDAC8. J. Mol. Model. 2012, 18, 3927−
3939.
(18) Kazemi, S.; Kruger, D. M.; Sirockin, F.; Gohlke, H. Elastic
̈
potential grids: accurate and efficient representation of intermolecular
interactions for fully flexible docking. ChemMedChem 2009, 4, 1264−
1268.
(6) (a) Miller, T. A.; Witter, D. J.; Belvedere, S. Histone deacetylase
inhibitors. J. Med. Chem. 2003, 46, 5097−5116. (b) Schlimme, S.;
Hauser, A.-T.; Carafa, V.; Heinke, R.; Kannan, S.; Stolfa, D. A.;
I
dx.doi.org/10.1021/jm301254q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(19) Stronach, E. A.; Alfraidi, A.; Rama, N.; Datler, C.; Studd, J.;
Agarwal, R.; Guney, T. G.; Gourley, C.; Hennessy, B. T.; Mills, G. B.;
Mai, A.; Brown, R.; Dina, R.; Gabra, H. HDAC4-regulated STAT1
activation mediates platinum resistance in ovarian cancer. Cancer Res.
2011, 71, 4412−4422.
(20) (a) Hager, K.; Franz, A.; Hirsch, A. Self-assembly of chiral
depsipeptide dendrimers. Chem.Eur. J. 2006, 12, 2663−2679. (b) Li,
J.; Yao, S. Q. “Singapore Green”: a new fluorescent dye for microarray
and bioimaging applications. Org. Lett. 2009, 11, 405−408.
(21) Gediya, L. K.; Chopra, P.; Purushottamachar, P.; Maheshwari,
N.; Njar, V. C. O. A new simple and high-yield synthesis of
suberoylanilide hydroxamic acid and its inhibitory effect alone or in
combination with retinoids on proliferation of human prostate cancer
cells. J. Med. Chem. 2005, 48, 5047−5051.
(22) Gosepath, E. M.; Eckstein, N.; Hamacher, A.; Servan, K.; von
Jonquieres, G.; Lage, H.; Gyorffy, B.; Royer, H. D.; Kassack, M. U.
̈
Acquired cisplatin resistance in the head−neck cancer cell line Cal27 is
associated with decreased DKK1 expression and can partially be
reversed by overexpression of DKK1. Int. J. Cancer 2008, 123, 2013−
2019.
(23) Eckstein, N.; Servan, K.; Girard, L.; Cai, D.; von Jonquieres, G.;
Jaehde, U.; Kassack, M. U.; Gazdar, A. F.; Minna, J. D.; Royer, H. D.
Epidermal growth factor receptor pathway analysis identifies
amphiregulin as a key factor for cisplatin resistance of human breast
cancer cells. J. Biol. Chem. 2008, 283, 739−750.
(24) Mueller, H.; Kassack, M. U.; Wiese, M. Comparison of the
usefulness of the MTT, ATP, and calcein assays to predict the potency
of cytotoxic agents in various human cancer cell lines. J. Biomol.
Screening 2004, 9, 506−515.
(25) Ciossek, T.; Julius, H.; Wieland, H.; Maier, T.; Beckers, T. A
homogeneous cellular histone deacetylase assay suitable for compound
profiling and robotic screening. Anal. Biochem. 2008, 372, 72−81.
(26) Bonfils, C.; Kalita, A.; Dubay, M.; Siu, L. L.; Carducci, M. A.;
Reid, G.; Martell, R. E.; Besterman, J. M.; Li, Z. Evaluation of the
pharmacodynamic effects of MGCD0103 from preclinical models to
human using a novel HDAC enzyme assay. Clin. Cancer Res. 2008, 14,
3441−3449.
(27) Chou, T. C. Drug combination studies and their synergy
quantification using the Chou−Talalay method. Cancer Res. 2010, 70,
440−446.
(28) Gerber, P. R.; Muller, K. MAB, a generally applicable molecular
̈
force field for structure modelling in medicinal chemistry. J. Comput.-
Aided Mol. Des. 1995, 9, 251−268.
J
dx.doi.org/10.1021/jm301254q | J. Med. Chem. XXXX, XXX, XXX−XXX