Design, synthesis, and evaluation of biphenyl-4-yl-acrylohydroxamic acid derivatives as histone deacetylase (HDAC) inhibitors

Article abstract of DOI:10.1016/j.ejmech.2008.11.005

A series of hydroxamic acid-based histone deacetylase (HDAC) inhibitors were designed on the basis of a model of the HDAC2 binding site and synthesized. They are characterized by a cinnamic spacer, capped with a substituted phenyl group. Modifications of the spacer are also reported. In an in vitro assay with the isoenzyme HDAC2, a good correlation of the activity with the docking energy was found. In human ovarian carcinoma IGROV-1 cells, selected compounds produced significant acetylation of p53 and α-tubulin. Most compounds showed an antiproliferative activity comparable to that of SAHA. At equitoxic concentrations, the tested compounds were more effective than SAHA in inducing apoptotic cell death. Compounds selected for in vivo evaluation exhibited a significant antitumor activity on three tumor models at well tolerated doses, thus suggesting a good therapeutic index.

Full text of DOI:10.1016/j.ejmech.2008.11.005

European Journal of Medicinal Chemistry 44 (2009) 1900–1912
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Design, synthesis, and evaluation of biphenyl-4-yl-acrylohydroxamic acid
derivatives as histone deacetylase (HDAC) inhibitors
,
a
a
a
b
Sabrina Dallavalle a , Raffaella Cincinelli , Raffaella Nannei , Lucio Merlini , Gabriella Morini ,
*
Sergio Penco c, Claudio Pisano c, Loredana Vesci c, Marcella Barbarino c, Valentina Zuco d,
Michelandrea De Cesare d, Franco Zunino d
a
`
Dipartimento di Scienze Molecolari Agroalimentari, Universita di Milano, Via Celoria 2, 20133 Milano, Italy
b University of Gastronomic Sciences, Pollenzo, Bra, Italy
c Sigma-tau, R&D, Pomezia, Italy
d
`
Unita di Farmacologia Antitumorale Preclinica, Istituto Nazionale per lo Studio e la Cura dei Tumori, Milano, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of hydroxamic acid-based histone deacetylase (HDAC) inhibitors were designed on the basis of
a model of the HDAC2 binding site and synthesized. They are characterized by a cinnamic spacer, capped
with a substituted phenyl group. Modifications of the spacer are also reported. In an in vitro assay with
the isoenzyme HDAC2, a good correlation of the activity with the docking energy was found. In human
Received 5 August 2008
Received in revised form
30 September 2008
Accepted 6 November 2008
Available online 19 November 2008
ovarian carcinoma IGROV-1 cells, selected compounds produced significant acetylation of p53 and a-
tubulin. Most compounds showed an antiproliferative activity comparable to that of SAHA. At equitoxic
concentrations, the tested compounds were more effective than SAHA in inducing apoptotic cell death.
Compounds selected for in vivo evaluation exhibited a significant antitumor activity on three tumor
models at well tolerated doses, thus suggesting a good therapeutic index.
Keywords:
Histone deacetylase
Antiproliferative activity
Synthesis
Ó 2008 Elsevier Masson SAS. All rights reserved.
1. Introduction
indicates a pleiotropic effect on critical pathways involved in cell
proliferation, apoptosis, microtubule function and DNA repair [4,5].
Inappropriate epigenetic modifications of gene expression are
associated with malignant phenotype and tumor progression.
Regulation of gene expression is mediated by several mechanisms
such as DNA methylation, ATP-dependent chromatin remodeling,
and post-translational modifications of histones. The latter mech-
Indeed, HDAC inhibitors have been reported to inhibit cell growth,
induce terminal differentiation in tumor cells and prevent the
formation of malignant tumors in mice [6].
In addition to the antitumor effects seen with HDAC inhibitors
alone, these compounds may also potentiate cytotoxic agents or
synergize with other targeted anticancer agents [7]. The exact
mechanism by which HDAC inhibitors cause cell death is still
unclear and the specific roles of individual HDAC isoenzymes as
therapeutic targets have not been established.
However, in vivo efficacy in tumor xenograft models has further
substantiated HDACs as a valid target for developing novel anti-
cancer agents [1a,8]. Natural and synthetic HDAC inhibitors have
been studied extensively. While suberoylanilide hydroxamic acid
(SAHA; ZolinzaÔ) [9] (Chart 1) has been approved by the FDA for
once-daily oral treatment of advanced cutaneous T-cell lymphoma
(CTCL) [10] many other small molecules that inhibit HDAC proteins
are currently in clinical trials for cancer treatment, e.g. MS-275 and
NVP-LAQ824 [1c,d] (Chart 1).
Several biological studies have indicated that HDACs are
heterogeneous, consisting of 18 isozymes, which can be categorized
into four classes, namely class I (HDAC 1, 2, 3 and 8), class II (HDAC
4, 5, 6, 7, 9 and 10) and classes III and IV, based on their sequence
homology. Class I, II and IV HDACs are zinc-containing amide
anism includes dynamic acetylation and deacetylation of 3-amino
groups of lysine residues present in the tail of the core histones.
Enzymes responsible for the reversible acetylation/deacetylation
processes are histone acetyltransferases (HATs) and histone
deacetylases (HDACs), respectively [1,2]. Inhibitors of histone
deacetylase (HDAC) enzymes have recently gained prominence as
an emerging class of anticancer agents [1].
When HDACs are inhibited, histone hyperacetylation occurs.
The disruption of the chromatin structure by histone hyper-
acetylation leads to the transcriptional activation of a number of
genes [3,4]. In addition to modulation of histone acetylation, HDAC
inhibitors are also involved in acetylation status of non-histone
proteins implicated in regulatory processes (e.g., transcription
factors and tubulins) [4,5]. The response to HDAC inhibitors
* Corresponding author. Tel.: þ39 02 50316818; fax: þ39 02 50316801.
E-mail address: sabrina.dallavalle@unimi.it (S. Dallavalle).
0223-5234/$ – see front matter Ó 2008 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2008.11.005

S. Dallavalle et al. / European Journal of Medicinal Chemistry 44 (2009) 1900–1912
1901
O
O
H
N
OH
O
N
H
NH2
N
H
H
N
O
SAHA
MS-275
N
O
NH
CONHOH
O
OH
N
H
N
1
HO
NVP-LAQ824
Chart 1. Structures of representative HDAC inhibitors.
hydrolases, with a conserved catalytic core but differing in size,
domain structure and tissue expression pattern. Class III consists of
NAD-dependent amide hydrolases, unrelated in sequences and
mechanism to classes I and II [11]. Zinc-dependent HDACs have
received much attention as anticancer drug targets.
inserted in the HDAC2 model, then SAHA was removed), and the
new complex energy minimized. The binding affinity was evaluated
by calculating the intermolecular interaction energy between the
appropriate compound and the HDAC2 model (Docking module in
Insight II). The intermolecular interaction energy of compound 1
docked in HDAC2 appeared more favourable than that of SAHA. In
the crystal structure of HDAC8 complexed with SAHA, the amino
acids lining the binding pocket are hydrophobic residues, G151,
F152, F208, M274 and Y306 and they present multiple favourable
contacts with the aliphatic chain of SAHA. Four of them (G151, F152,
F208 and Y306) are conserved across the class I HDACs, while M274
becomes a leucine in the other members of the family. These amino
acids form the tunnel that the acetyllysine substrate penetrates to
access the catalytic machinery. Our modelling indicated that the
A bacterial HDAC homologue (histone deacetylase-like protein,
HDLP) [12], a human HDAC8 [13] and a bacterial class II histone
deacetylase homologue [14] have been co-crystallized with the
hydroxamic acid-based inhibitor SAHA (Chart 1). The alkyl chain of
SAHA inserts into a tubular pocket, and the hydroxamic group
serves as a bidentate chelator of the catalytic Znþþ ion bound in the
enzyme active site. Based on these crystallographic analyses,
several human class I HDAC inhibitors have been designed. In all of
these designs the common structural motif consists of a metal-
binding head group, which interacts with the Znþþ ion at the
bottom of the active site, a linker domain, which occupies the
narrow tube-like channel, and a cap group, which interacts with
the residues on the rim of the active site. The linker serves to
appropriately position the capping and metal-binding groups for
high-affinity interactions with the target proteins.
In the present study, we have designed novel structurally simple
HDAC inhibitors with a hydroxamic acid as the zinc chelating head
group attached to a cinnamic linker domain and an aromatic group
as a cap structure. The prototype of these new HDAC inhibitors (1) is
shown in Chart 1. Here we report the synthesis, the HDAC2 isoen-
zyme and cancer cell growth inhibition of this novel class of
compounds. HDAC2 was chosen as a representative isoenzyme of
class I HDACs, which are known to have an almost exclusive nuclear
localization and to be essential to cell proliferation [15].
proximal phenyl ring of compound 1 exhibits a
p–p stacking
interaction with the side chain benzyl group of Phe151, Phe206 and
Tyr304 in our HDAC2 model (corresponding to F152, F208 and Y306
in HDAC8). The distal phenyl ring of 1 (cap structure) appears to
accommodate in a large cavity, without a significant steric clash.
However, its environment contains some important amino acid
residues which may be involved in electrostatic interactions, such
as Glu99 and Glu147 (corresponding to Tyr100 and Glu148 in
HDAC8).
Therefore the distal phenyl ring of 1 could be used as a scaffold
to place a variety of substituent groups at appropriate distances for
such interactions (e.g. an OH group, Fig. 1).
3. Chemistry
The prototype 4-phenylcinnamohydroxamic acid
1
was
2. Molecular modelling and design
produced by condensing commercially available 4-phenylcinnamic
acid with hydroxylamine in the presence of HATU and DIPEA.
The substituted aryl or heteroaryl derivatives 3b,e–g,i–l
were obtained by a Suzuki coupling reaction between methyl 4-
bromocinnamate and the appropriate aromatic or heteroaromatic
boronic acids. Conversion of the esters to the corresponding
hydroxamic acids was achieved either by condensing the acids
obtained after hydrolysis with hydroxylamine in the presence of
HATU or HBTU or by coupling the ester with O-THP-protected
hydroxylamine, followed by acidic deprotection. Compounds 3c,d,h
were obtained by a Suzuki coupling reaction between O-THP-pro-
tected 4-bromocinnamohydroxamic acid 5 and the appropriate
aromatic or heteroaromatic boronic acids, followed by acid hydro-
lysis (Scheme 1).
Compound 1 was designed on the basis of the hypothesis that
the proximal phenyl ring, extended from the hydroxamic acid via
a double bond, could be suitable to occupy the narrow tube-like
pocket of the HDAC class I active site. A hydrophobic phenyl ring
was selected because crystallographic analysis of HDAC class I
proteins indicates that the active site residues surrounding the
linker chain are hydrophobic. During the course of this work, other
examples of cinnamyl linkers were reported [1d], but none directly
connected to an aromatic ring.
To understand the binding mode in the catalytic site and derive
structure–activity relationships (SAR) for compound
1 and
analogues, we developed a model of the active site of HDAC2 based
on the homology model derived and validated by Wang et al. [16]
using HDLP as a template. We chose HDAC2 because this isoenzyme
was used for in vitro assays.
Compound 1 was inserted in the obtained HDAC2 site (the
hydroxamic acid moiety was overlapped to that of SAHA previously
Compound 3a was obtained by Heck condensation of 4-(40-bro-
mophenyl)phenol with methyl acrylate, followed by basic hydro-
lysis. The conversion to hydroxamic acid was accomplished using
WSC and HOBT as coupling reagents. A Suzuki one-pot reaction
between 2-chloro-4-bromophenol and methyl 4-bromocinnamate,

1902
S. Dallavalle et al. / European Journal of Medicinal Chemistry 44 (2009) 1900–1912
activities against a panel of tumor cell lines of different tissue
origin: NB4 (human acute Promyelocytic leukemia), H460 (human
lung carcinoma cell line), HCT116 (human colon cancer cell line),
IGROV-1 and its subline resistant to cisplatinum IGROV-1/Pt1
(human ovarian carcinoma cell lines) (Table 2). The growth-inhib-
itory activities were apparently p53-independent, because the
cellular effects were comparable in cells with functional (IGROV-1)
or defective p53 (IGROV-1/Pt).
Substitution at the distal phenyl ring of compound 1 was first
assessed (compounds 3a–m). Most of these compounds (except
3c,e,i) showed an inhibitory activity in the HDAC2 assay in the same
order of magnitude, or slightly lower than SAHA (Table 1).
Compounds 3a,b,m, among the most potent in this group, were
predicted by the docking to have a strong interaction energy,
similarly to SAHA.
We then investigated the linker domain focusing on altering
saturation and chain length. First, we studied the effect of inserting
an ethyl bridge instead of the cinnamyl double bond, thus allowing
further flexibility of the side chain (compound 4). Also in this case
docking studies and HDAC2 assays are in agreement, indicating that
the unsaturation in the side chain makes the ligand more active
(compare compounds 3a and 4). We then investigated the effect of
the elongation of the linker by two carbon units (10a), finding that
it is detrimental for the activity (as evidenced both by docking
studies and HDAC2 assays). The shift of the position of the side
chain to the adjacent position on the proximal ring (compound 13)
seemed not to affect the interaction with the isoenzyme. In general,
within the compounds with a cinnamic spacer, a pattern emerged
of reasonable correlation of the docking energy with the inhibitory
activity against HDAC2.
Fig. 1. Model of compound 3b in the putative HDAC2 binding site.
followed by hydrolysis with LiOH and condensation with hydrox-
ylamine using HATU and DIPEA, gave compound 3m. Compound 4
was obtained from the corresponding cinnamyl ester by catalytic
hydrogenation, followed by basic hydrolysis and coupling with
hydroxylamine using HBTU and DIPEA.
The results of the antiproliferative activity tests are reported in
Table 2. Generally, it was found that a wide array of functionalities
can be tolerated, particularly as concerns substituents of the distal
phenyl ring, most compounds showing an antiproliferative activity
comparable with that of SAHA.
Synthesis of compound 10a (Scheme 2) was accomplished by
Because we hypothesized that the placement of an appropriate
spacer between the cap group and the zinc binder would modulate
the HDAC inhibitory activity, we varied the chain length of the
spacer. Compounds 10a–c, which contain a dienic system [17],
showed a cytotoxicity of the same order as that of 1, depending on
the cell line. This result indicates that a longer chain is not only
tolerated but in some cases, e.g. 10a, increases the cytotoxicity. Also
insertion of an oxygen atom between the double bond and the
biphenyl system (7) was tolerated, as well as changing the proximal
planar aromatic ring into a cyclohexyl ring (17). On the contrary,
reduction of the double bond in the chain, which in principle
should confer additional flexibility to the spacer, was not, in fact,
beneficial (compounds 3a vs. 4, 10a vs. 11, 17 vs. 18). These results
seem to emphasize the importance of the double bond adjacent to
the hydroxamic group. It is of interest to note that the position of
the chain was important for the antiproliferative activity. In fact
compound 13 showed reduced potency when compared to 3a.
These data show that no precise correlations could be found
between the inhibitory potency against HDAC2 and the anti-
proliferative potency in the tested cell lines. This observation is not
surprising, because a) the antiproliferative potency could reflect
several factors, in particular cellular pharmacokinetics, which could
critically influence cellular response; b) a number of indirect
evidences support the similarity of the profile of HDAC inhibition
by compounds of this series with SAHA known to be a pan-HDAC
inhibitor [18]; c) the cellular effects are expected to be dependent
on the pattern of inhibition among the various HDAC isoenzymes
and on the profile of HDAC expression in the various cell systems.
The most potent compounds, as antiproliferative agents, were
also studied for their ability to induce apoptosis in the ovarian
carcinoma cell line, IGROV-1, carrying a functional wild-type p53
Wadsworth–Horner–Emmons
condensation
of
4-phenyl-
benzaldehyde with ethyl 4-(diethoxyphosphoryl)-but-2-enoate,
followed by hydrolysis and coupling with HBTU and DIPEA.
Compounds 10b and 10c were synthesized by hydrolysis of the
common intermediate 5-(4-bromophenyl)-penta-2,4-dienoic acid
(tetrahydropyran-2-yloxy)-amide 9, on its turn prepared from 4-
bromobenzaldehyde, following the sequencedescribed in Scheme 2.
The saturated derivative 11 was obtained by catalytic hydrogenation
of the dienoic ester and usual hydrolysis and coupling.
Insertion of an oxygen atom between the double bond and the
biphenyl system led to compound 7 whose synthesis started from
addition of 4-phenylphenol to ethyl propiolate, followed by the
aforementioned conventional steps. The synthesis of the analog
(13) of 4a with the phenyl substituent shifted to the meta-position
is shown in Scheme 3. It was obtained from the Suzuki coupling of
methyl 3-bromocinnamate with 4-hydroxyphenylboronic acid,
followed by the usual steps.
A synthetic route towards the derivative with a saturated
proximal ring 17 is outlined in Scheme 4. Acetylation of 4-(4-
hydroxyphenyl)cyclohexanone followed by Wittig reaction with
(methoxymethyl)triphenylphosphonium chloride gave, after
hydrolysis, aldehyde 14. Reaction with methyl(triphenylphosphor-
anylidene)acetate, hydrolysis and coupling with HBTU and DIPEA
furnished 17. Catalytic hydrogenation of the acid 16 gave, after
coupling, compound 18.
4. Results and discussion
The synthesized compounds were tested for their inhibitory
activity towards HDAC2 isoform (Table 1), and for antiproliferative

CONHOH
Br
4
b
HO
R
CONHOH
HO
Br
c,d,e
3
R
COOCH3
COOCH3 f,g
o
a
3a,b,e,i,l,m
3c,d,f,g,h
2a-m
p
CONHO
CONHO
COOH
h
O
O
i
Br
R
Br
6c,d,f,g,h
5
a
i
e
N
N
HO
b
l
f
N
H
OHC
R =
OH
m
c
g
H3CO
HO
Cl
Cl
S
d
h
NC
Scheme 1. Reagents and conditions: a) for 2b–h: RPhB(OH)2, Pdtetrakis(PPh)3, 2 N Na2CO3, toluene, 2–8 h, reflux; for 2i,l,m: RPhBr, bis(pinacolato)diboron, KOAc, PdCl2(dppf), 2 N
Na2CO3, dioxane, reflux, 17–48 h; b) methyl acrylate, Pd(OAc)2, TEA, tri-o-tolylphosphine, 100 ꢀC, 14 h, 85%; c) H2, PtO2, AcOEt, rt, 1 h, 77%; d) LiOH, THF:H2O 1:1, rt, 24 h, dark, 100%;
e) HBTU, DIPEA, DMF, NH2OH$HCl, rt, 4 h, 92%; f) LiOH, THF:H2O 1:1, rt, 24 h, dark; g) for 3a: HOBt, WSC, DMF, rt, 3 h, then NH2OH$HCl, TEA, DMF, rt, overnight, 13%; for 4e,i,l: HBTU,
DIPEA, NH2OH$HCl, DMF, rt, 5–12 h, 36–68%; for 3b,f,m: HATU, DIPEA, NH2OH$HCl, DMF, rt, 3–8 h, 25–65%; h) HOBT, WSC, DMF, rt, 3 h, the THPONH2, 50 ꢀC, 1 h, 84%; i) for 6d:
RPhBr, bis(pinacolato)diboron, KOAc, PdCl2(dppf), dioxane, 100 ꢀC, 1 h, then 8, PdCl2(dppf), 2 N Na2CO3, 100 ꢀC, 4 h, 82%; for 6f,g,h: RPhB(OH)2, Pdtetrakis(PPh)3, 2 N Na2CO3,
toluene, reflux, 3 h, 88–100%; o) for 6c,d,g,h: THPONH2, Li hexamethyldisilazane, THF, ꢁ78 ꢀC, 10 min–1 h, 53–100%; p) for 3c,d,h: MeOH, pTSA, rt, 12–48 h, 46–62%.
O
CONHOH
CHO
CONHO
g,h
O
Br
Br
9
7
i,l
a,b,c
CONHOH
R
COOEt
e,f
d
8
10a R'=H
10b R'=OCH3
10c R'= CN
m,n,o
R'
a R=OH
b R=CHO
CONHOH
11
Scheme 2. Reagents and conditions: a) ethyl propiolate, 4-methylmorpholine, MeCN, rt, 20 min, 91%; b) LiOH, THF:H2O 1:1, rt, 24 h, dark, 53%; c) HBTU, DIPEA, DMF, NH2OH$HCl, rt,
48 h, 19%; d) triethyl 4-phosphonocrotonate, NaH, THF dry, 0 ꢀC, 30 min, then rt 4 h, 61%; e) LiOH, THF:H2O 1:1, rt, 24 h, dark, 95%; f) for 10a: HBTU, DIPEA, DMF, NH2OH$HCl, rt, 12 h,
31%; g) triethyl 4-phosphonocrotonate, NaH, THF dry, 0 ꢀC, 30 min, then rt, 2 h, 58%; h) THPONH2, Li hexamethyldisilazane, THF, ꢁ78 ꢀC, 40 min, 96%; i) RPhB(OH)2, Pdte-
trakis(PPh)3, 2 N Na2CO3, toluene, 2 h, reflux, 34–42%; l) MeOH, pTSA, rt, 12 h, 56–58%; m) H2/Pd/C, MeOH/AcOH 1:1, rt, 12 h, 100%; n) LiOH, THF:H2O 1:1, rt, 24 h, dark, 95%; o)
HBTU, DIPEA, DMF, NH2OH$HCl, rt, 12 h, 62%.

1904
S. Dallavalle et al. / European Journal of Medicinal Chemistry 44 (2009) 1900–1912
CONHOH
COOH
COOMe
c
a,b
Br
HO
HO
12
13
Scheme 3. Reagents and conditions: a) 4-HO–PhB(OH)2, Pdtetrakis(PPh)3, 2 N Na2CO3, toluene, 6, reflux, 25%; b) LiOH, THF:H2O 1:1, rt, 24 h, dark, 64%; c) HBTU, DIPEA, DMF,
NH2OH$HCl, rt, 4 h, 23%.
and characterized by an efficient susceptibility to apoptosis under
stress conditions [19,20]. Our analysis also included compounds
equimolar concentrations. Since compounds very effective as
apoptosis inducers (e.g., 3h, 3i and 1b) caused a marginal acetyla-
tion of p53, it is conceivable that modulation of p53 function is not
a primary event in activation of apoptotic pathways.
On the basis of the above results, we thought it worthwhile to
examine the therapeutic potential of some representative deriva-
tives as anticancer agents. Therefore compounds, 3a, 3d and 3g
were selected for in vivo studies (Table 4).
The antitumor efficacy was evaluated in 3 human tumor xeno-
grafts (s.c. implanted) derived from tumor cell lines used in the
antiproliferative assay. Using a daily oral administration (5 day/
week for a total of 10 doses), the methoxyderivative 3g was more
effective than SAHA against the H460 model. The i.p. treatment was
less effective in this preliminary evaluation. All tested compounds
were well tolerated and did not produce toxic deaths. Using a pro-
longed treatment schedule (4 weeks) a transient animal body
weight loss was observed with 3a. Again compound 3g was effec-
tive in the treatment of the colon carcinoma HCT116 and even more
effective against the ovarian carcinoma IGROV-1, producing 71%
growth inhibition at the end of the treatment (Fig. 5). Significant
antitumor activity was observed in a wide range of well tolerated
doses delivered by oral route (20–100 mg/kg) with a prolonged
treatment schedule, thus suggesting a good therapeutic index. The
improved activity of tested compounds over SAHA could reflect
a better bioavailability in oral administration.
exhibiting lower growth-inhibitory activity (IC50 ꢂ 8
mM) in an
attempt to establish possible correlation between apoptosis and
antiproliferative effect. Most of the tested compounds induced
a substantially higher level of apoptotic cell death than SAHA at
concentrations causing equivalent antiproliferative effects (IC80
after 72 h exposure) (Table 3). Indeed, the percentage of drug-
induced apoptotic cells ranged between 40% and 90%. The most
potent inducers of apoptosis were 3c, 3i, 10a which was markedly
effective (80–90% apoptosis) at 2–3 mM. Although these
compounds were among the most potent in inhibiting cell growth,
only a rough correlation was found between antiproliferative and
proapoptotic effects. The proapoptotic activity was also supported
by the presence of sub-G1 peak detected in the cell cycle analysis,
already evident at 24 h following drug exposure (Fig. 2).
This analysis also shows that two representative compounds of
our series (3d and 10a) induced G1 arrest in IGROV-1 cells with
functional wild-type p53. This cellular response was likely mediated
by p21 induction, possibly as a consequence of p53 activation (not
shown). This interpretation was consistent with a different cell cycle
perturbation observed in IGROV-1/Pt1 with mutant p53, exhibiting
a partial accumulation of cells also in S-phase and in G2/M.
In an attempt to understand the cellular response to treatment
we evaluated the acetylation state of histone H4 and a-tubulin after
short-term exposure (4 h) (Fig. 3). The selected compounds, at IC80
and IC50 concentrations respectively, caused a comparable increase
In conclusion, we have designed and synthesized a new class of
hydroxamic acid HDAC inhibitors. In vitro assay with the isoenzyme
HDAC2 and molecular docking with a model of HDAC2 have been
performed. The good correlation between the two validates the use
of the model as a useful tool in the design of further analogues. The
synthesized compound were tested for antiproliferative activities
against a panel of tumor cell lines, from which general features of
structure–activity relationships can be discerned. The 4-phenyl-
cinnamic scaffold appears to be required for a good cytotoxic
activity on different cell lines. Modification of the position of the
side chain, or replacement of the proximal ring with a cyclohexyl,
in the acetylation state of
a-tubulin and of histone H4. As HDAC6 is
known to deacetylate lysine 40 of
a
-tubulin [20] this finding
supports that the selected compounds were also effective HDAC6
inhibitors.
The modulation of the acetylation of
a-tubulin was more
evident after a prolonged exposure (up to 48 h) (Fig. 4). Under these
conditions all examined compounds were more effective than
SAHA in inducing a persistent tubulin acetylation. The same
observation applies to the p53 acetylation analyzed at equitoxic or
COOCH3
COOH
b
CHO
a
15
16
14
HO
HO
HO
c
d,e
CONHOH
CONHOH
17
18
HO
HO
Scheme 4. Reagents and conditions: a) methyl(triphenylphosphoranylidene)acetate, THF, rt then reflux, 4 h, 94%; b) LiOH, THF:H2O 1:1, rt, 24 h, dark, 81%; c) HBTU, DIPEA, DMF,
NH2OH$HCl, rt, 48 h, 13%; d) H2/Pd/C, MeOH, rt, 4 h, 80%; e) HBTU, DIPEA, DMF, NH2OH$HCl, rt, 12 h, 35%.

S. Dallavalle et al. / European Journal of Medicinal Chemistry 44 (2009) 1900–1912
1905
Table 1
Table 3
Selected ligands – HDAC2 model intermolecular interaction energies (kcal/mol) and
HDAC2 inhibitory activity.
Apoptosis induction in IGROV-1 cell by selected HDAC inhibitors.a
Cpd.
IC80
(m
M)
Apoptosis
Entry
HDAC2 assay (IC50
,
m
M)
Interaction energy (kcal/mol)
Untreated
SAHA
SAHA
1b
3a
3b
3c
3d
3g
3h
2 ꢃ 2
15 ꢃ 5
45 ꢃ 5
90 ꢃ 8
40 ꢃ 8
90 ꢃ 2
77 ꢃ 6
90 ꢃ 5
52 ꢃ 3
90 ꢃ 4
89 ꢃ 7
45 ꢃ 6
89 ꢃ 3
65 ꢃ 6
SAHA
1
3a
3b
3m
4
10a
13
0.1 ꢃ 0.02
0.82 ꢃ 0.14
0.59 ꢃ 0.2
0.22 ꢃ 0.06
0.46 ꢃ 0.14
1.7 ꢃ 0.064
>5
ꢁ18.90
ꢁ25.33
ꢁ25.65
ꢁ27.75
ꢁ23.15
ꢁ17.86
ꢁ14.34
ꢁ23.09
5
10
10
20
15
2
10
3
7
0.32 ꢃ 0.024
3i
4
10a
11
3
10
3
led to substantial decrease of the activity. The length and type of the
chain tolerated some changes. A variety of substitutions in the
distal ring were also well tolerated. Selected compounds of this
series exhibited a promising antitumor activity.
12
a
Apoptosis was determined at 72 h following exposure to IC80 by TUNEL assay
and analysis by flow cytometry. SAHA was used as a control at two concentrations,
5 mM (IC80) and 10 mM (IC90). The results are expressed as percentage of apoptotic
5. Experimental
cells vs. total cell number. Values are the mean ꢃ standard deviation of three
independent experiments.
5.1. General chemical methods
of KOAc and 13 mg (0.018 mmol) of PdCl2(dppf) were dissolved in
36 mL of dioxane and the mixture was stirred at 100 ꢀC for 1–2 h
under nitrogen. The solution was cooled at room temperature, and
methyl 4-bromocinnamate (287 mg, 1.19 mmol), PdCl2(dppf)
(13 mg, 0.018 mmol) and 2 N Na2CO3 (1.48 mmol, 0.74 mL) were
added and the mixture was heated at 80–100 ꢀC for 1–4 h. The
solution was cooled at room temperature, diluted with water,
acidified with 2 N HCl (2.4 mL) and extracted with ethyl acetate.
The organic phase was washed with brine, dried over Na2SO4,
filtered and evaporated. The crude product was purified, when
necessary, by flash chromatography.
All reagents and solvents were reagent grade or were purified by
standard methods before use. Melting points were determined in
open capillaries on a Bu¨chi melting point apparatus and are
uncorrected. Column chromatography was carried out on flash
silica gel (Merck 230–400 mesh). TLC analysis was conducted on
silica gel plates (Merck 60F254). NMR spectra were recorded at
300 MHz with a Bruker instrument. Chemical shifts (d values) and
coupling constants (J values) are given in ppm and Hz, respectively.
Analyses indicated by the symbols of the elements or functions
were within ꢃ0.4% of the theoretical values.
5.2. General procedure for one-pot Suzuki condensation (method A)
5.3. General procedure for Suzuki coupling (method B)
A small amount of the appropriate aryl bromide (0.6 mmol),
166 mg (0.65 mmol) of bis(pinacolato)diboron, 174 mg (1.78 mmol)
To a solution of methyl 4-bromocinnamate (883 mg, 3.66 mmol)
in 8 mL of toluene, 4.7 mL of a 2 M solution of Na2CO3, 141 mg
(0.122 mmol) of Pd(tetrakis)triphenylphosphine and 4.1 mmol of
the appropriate boronic acid (previously dissolved in 2 mL of EtOH)
were added. The solution was refluxed for 2 h to 2 days under
nitrogen. After addition of ethyl acetate, the organic phase was
washed with water, brine, dried over Na2SO4 and filtered. The
solvent was removed under reduced pressure to give a crude ester
which was purified, when necessary, by flash chromatography.
Table 2
Inhibition of HDAC2a (IC50 M) and antiproliferative activityb (IC50
, m , mM) of selected
compounds and of reference HDAC inhibitor (SAHA) against a panel of various tumor
cell lines.
Cpd. HDAC2 assay NB4
H460
HCT-116
IGROV-1
IGROV-1/Pt1
SAHA 0.1 ꢃ 0.02
0.7 ꢃ 0.03
3.4 ꢃ 0.8 0.31 ꢃ 0.02 2.2 ꢃ 0.3
5.4 ꢃ 0.1 1.58 ꢃ 0.3 3.5 ꢃ 0.6
6.0 ꢃ 0.9 0.33 ꢃ 0.05 7.6 ꢃ 3.5
3.8 ꢃ 0.2 0.8 ꢃ 0.04 8.0 ꢃ 0.4
2.2 ꢃ 0.2
4.0 ꢃ 0.3
6.5 ꢃ 2.0
5.5 ꢃ 0.4
1
0.82 ꢃ 0.14
0.59 ꢃ 0.2
0.90 ꢃ 0.1
2.3 ꢃ 0.02
3a
3b
3c
3d
3e
3f
0.22 ꢃ 0.06 1.81 ꢃ 0.1
5.4. General procedure for ester hydrolysis (method C)
1.86 ꢃ 0.17
0.76 ꢃ 0.03 nd
0.51 ꢃ 0.1 0.61 ꢃ 0.14 1.36 ꢃ 0.04
0.33 ꢃ 0.001 nd
5.7 ꢃ 0.4 nd
2.7 ꢃ 0.8
3.2 ꢃ 1.6
The appropriate ester (1 mmol) was suspended in a solution of
219 mg (5.2 mmol) of LiOH$H2O in 4 mL of THF/H2O 1:1 and stirred
at room temperature in the dark overnight. After evaporation of
THF the remaining aqueous phase was washed with hexane, diethyl
ether and then acidified with HCl 2 N (0.4 mL). The precipitated
white solid was filtered and dried. No further purification was
necessary.
>5
nd
3.29 ꢃ 0.2
nd
nd
nd
nd
nd
nd
nd
nd
12.1 ꢃ 0.7 20.3 ꢃ 1.5
3g
3h
3i
1.16 ꢃ 0.27
3.7 ꢃ 0.1 0.22 ꢃ 0.01 0.75 ꢃ 0.4 2.3 ꢃ 0.3
0.45 ꢃ 0.04 nd
2.8 ꢃ 0.1 nd
0.86 ꢃ 0.05 nd 0.7 ꢃ 0.4
6.5 ꢃ 0.5 14 ꢃ 2.0 nd
2.8 ꢃ 1.8
1.1 ꢃ 0.5
3.2 ꢃ 1.6
3.5 ꢃ 0.3
>5
3l
3m
4
1.2 ꢃ 0.46
0.46 ꢃ 0.14
1.7 ꢃ 0.064
nd
22.1 ꢃ 0.9 29.3 ꢃ 2.5
0.58 ꢃ 0.09 3.1 ꢃ 0.1 1.46 ꢃ 0.25 4.1 ꢃ 0.5
3.9 ꢃ 0.4
0.95 ꢃ 0.1
5.9 ꢃ 1.3
2.4 ꢃ 1.2
3.9 ꢃ 0.5
15.2 ꢃ 2.0 0.26 ꢃ 0.14 0.8 ꢃ 0.2
nd 5.4 ꢃ 0.7
3.9 ꢃ 0.8
7
nd
10a
1b
10c
11
13
17
18
>5
>5
0.66 ꢃ 0.003 1.9 ꢃ 0.05 0.98 ꢃ 0.03 1.4 ꢃ 0.3
5.5. General procedure for HATU or HBTU condensation (method D)
nd
nd
>20
3.3 ꢃ 0.2 nd
nd
0.57 ꢃ 0.09 2.01 ꢃ 0.91
1.05 ꢃ 0.5 3.1 ꢃ 0.6
1.2 ꢃ 0.02
The appropriate acid (0.89 mmol) was dissolved under nitrogen
in 9 mL of DMF, then 0.89 mmol of HATU or HBTU and 308 mL
0.72 ꢃ 0.12
0.86 ꢃ 0.03 4.7 ꢃ 0.6 2.7 ꢃ 0.4
8.0 ꢃ 1.5
9.1 ꢃ 2.9
>10
0.32 ꢃ 0.024 2.85 ꢃ 0.05 5.6 ꢃ 0.3 1.64 ꢃ 0.2 >10
0.31 ꢃ 0.003 0.9 ꢃ 0.1
nd
6.9 ꢃ 0.1
2.2 ꢃ 0.3 0.37 ꢃ 0.17 2.5 ꢃ 0.5
9.5 ꢃ 0.7 4.27 ꢃ 0.3 31.9
2.6 ꢃ 0.4
(1.78 mmol) of DIPEA were added and the solution was stirred at
room temperature for 2 min. After addition of hydroxylamine
hydrochloride (68 mg, 0.98 mmol), the mixture was stirred at room
temperature for 3–72 h. DMF was removed under reduced pressure
and the residue was washed with water to obtain, after filtration,
a crude product which was purified, when necessary, by flash
chromatography.
23.9
nd ¼ Not determined.
a
HDAC activity determined using HDAC2 from HeLa cells as indicated in
Section 5.
b
Cells were exposed for 72 h. Antiproliferative activity was measured by cell
counting and expressed as concentration required for 50% inhibition of cell growth
(IC50). IC50 values were determined by dose–response curves.

1906
S. Dallavalle et al. / European Journal of Medicinal Chemistry 44 (2009) 1900–1912
Fig. 2. Time course of cell cycle perturbations in ovarian carcinoma cells, IGROV-1, and in a cisplatin-resistant subline, IGROV-1/Pt1, following treatment with selected compounds.
The comparison was performed in cells treated with IC80 for each drug and the cell cycle was analyzed by FACScan analysis of PI-stained cells at 24, 48 and 72 h of treatment. In the
case of SAHA a higher concentration (10 mM), corresponding to IC90, was also shown. One experiment representative of three is reported.
5.6. General procedure for the synthesis of hydroxamic acids from
esters via N-(tetrahydropyran-2-yloxy) amides (method E)
the crude N-(tetrahydropyran-2-yloxy) amide, which was purified,
when necessary, by flash chromatography. This product
(0.603 mmol) was suspended in 9 mL of MeOH, then 34 mg
(0.181 mmol) of p-toluenesulfonic acid monohydrate was added
and the mixture was stirred overnight at room temperature. The
solid formed was filtered, washed with Et2O and dried to give the
hydroxamic acid.
The appropriate ester (1.02 mmol) was dissolved in 14 mL of
anhydrous THF, then O-tetrahydropyranylhydroxylamine (120 mg,
1.02 mmol) was added. The solution was cooled to ꢁ78 ꢀC and
treated with 1.07 mL (2.14 mmol) of lithium hexamethyldisilazane.
The mixture was stirred for 1 h under nitrogen, then was quenched
with NH4Cl solution. Once at room temperature, the mixture was
extracted with ethyl acetate and the organic phase was dried over
Na2SO4. The solvent was evaporated under reduced pressure to give
5.6.1. N-Hydroxy-E-3-(biphenyl-4-yl)-acrylamide (1)
E-4-Phenylcinnamic acid (200 mg, 0.89 mmol) was condensed
with hydroxylamine hydrochloride according to procedure D, using

S. Dallavalle et al. / European Journal of Medicinal Chemistry 44 (2009) 1900–1912
1907
IGROV-1
SAHA
5
3a
3b
3c
3d
3g
Conc. (μM)
acetylated
α-tubulin
-
10
3
1
-
20 10
3
-
15
8
-
1
0.6
-
10
3
3
0.75
acetylated
histone H4
β-tubulin
3h
3i
3m
10a
11
1b
-
7
-
3
1.14
-
8
4
-
3
1
12
8
-
10
1
Conc. (μM)
acetylated
α-tubulin
acetylated
histone H4
β-tubulin
Fig. 3. Effects of selected compounds on acetylation of a-tubulin and histone H4 in IGROV-1 cells after 4 h exposure to IC80 and IC50 for each compound. b-tubulin is shown as
a control of protein loading.
HATU, to obtain 220 mg of the title compound as a white solid.
Yield 100%, mp 168–170 ꢀC. 1H NMR (300 MHz, DMSO-d6)
: 10.75
(s, 1H), 9.03 (br s, 1H), 7.79–7.56 (m, 6H), 7.53–7.29 (m, 4H), 6.47 (d,
1H, J ¼ 16 Hz). Anal. C15H13NO2 (C, H, N).
product. Purification by flash chromatography on RP-18 (40–
63 m) reverse phase (CH3OH/H2O 50:50) afforded 34 mg of the
title compound as a white solid. Yield 13%, mp > 300 ꢀC. 1H NMR
(300 MHz, DMSO-d6) : 10.73 (s, 1H), 9.62 (s, 1H), 9.00 (s, 1H), 7.69–
7.48 (m, 6H), 7.43 (d, 1H, J ¼ 16 Hz), 6.82 (d, 2H, J ¼ 8.19 Hz), 6.44 (d,
1H, J ¼ 16 Hz). MS (EI) m/z: 255 (Mþ, 40), 91 (100), 73 (40), 57 (40).
Anal. C15H13NO3 (C, H, N).
d
m
d
5.6.2. N-Hydroxy-E-3-(40-hydroxybiphenyl-4-yl)-acrylamide (3a)
E-4-(40-Hydroxyphenyl)cinnamic acid [21] (250 mg, 1.04 mmol)
was dissolved under nitrogen in 10 mL of DMF, then 169 mg
(1.25 mmol) of HOBt and 259 mg (1.35 mmol) of WSC were added
and the solution stirred at room temperature for 3 h. After addition
of hydroxylamine hydrochloride (361 mg, 5.2 mmol), followed by
0.72 mL (5.2 mmol) of TEA, the mixture was stirred at room
temperature overnight. DMF was removed under reduced pressure
and the residue was washed with water to obtain 263 mg of a crude
5.6.3. N-Hydroxy-E-3-[40-hydroxymethylbiphenyl-4-yl]-
acrylamide (3b)
Methyl 4-bromocinnamate was reacted with 4-hydrox-
ymethylbenzeneboronic acid (method A) to give, after flash
chromatography (hexane/ethyl acetate 80:20) methyl E-3-(40-
hydroxymethylbiphenyl-4-yl)-acrylate as a white solid. Yield 51%,
IGROV-1
SAHA
3a
3b
3c
3d
3g
24h
48h
24h
48h
24h
48h
24h
48h
24h
48h
24h
48h
Conc. (μM)
-
10
5
3
-
10
5
3
-
20 10
3
-
20 10
3
-
15
8
-
15
8
-
1
0.6
-
1
0.6
-
10
3
-
10
3
-
3
0.75 -
3
0.75
acetylated
p53 (lys382)
acetylated
α-tubulin
β-tubulin
3h
3i
3m
4
10a
11
10a 11
1b
24h
48h
24h 48h
24h
48h
24h
48h
24h
48h
24h
48h
Conc. (μM)
acetylated
p53 (lys382)
-
10
-
10
-
3
1.14
-
3
1.14
-
8
4
-
8
4
-
101 0.6 - 10 1 0.6
-
3
1
12
8
-
3
1
12
8
-
10
1
-
10
1
-
acetylated
α-tubulin
β-tubulin
Fig. 4. Effects of selected compounds on acetylation of a-tubulin and p53 in IGROV-1 cells after 24 and 48 h exposure to IC80 and IC50 for each compound. b-tubulin is shown as
a control of protein loading.

1908
S. Dallavalle et al. / European Journal of Medicinal Chemistry 44 (2009) 1900–1912
Table 4
(d, 1H, J ¼ 16.00 Hz), 7.63–7.47 (m, 6H), 7.41 (d, 2H, J ¼ 8.34 Hz), 6.47
Antitumor activity of selected HDAC inhibitors.
(d, 1H, J ¼ 16.00 Hz), 3.81 (s, 3H). This ester was reacted with O-tet-
rahydropyranylhydroxylamine (method E) to give E-3-(40-chlor-
Tumor model
Compound Schedulea
Dose
(mg/
kg)
Route TVI%b BWL%c
obiphenyl-4-yl)-N-(tetrahydropyran-2-yloxy)-acrylamide.
Yield
100%, mp 120–122 ꢀC. 1H NMR (300 MHz, CDCl3)
d: 7.73 (d, 1H,
H460 (nonSCLC)
SAHA
qd ꢄ 5 d/
week
100
p.o. 34
p.o. 57
2
J ¼ 15.63 Hz), 7.62–7.46 (m, 6H), 7.37 (d, 2H, J ¼ 8.34 Hz), 6.43 (br s,
1H), 5.12(brs,1H), 4.02–3.87(m,1H), 3.71–3.57(m,1H),1.92–1.73(m,
4H), 1.70–1.31 (m, 2H). This compound was treated according to
method E to give the title compound as a white solid. Yield 46%, mp
3g
3g
ꢄ2 weeks
100
100
1
4
i.p.
39
HCT116 (colon
carcinoma)
SAHA
qd ꢄ 5 d/
week
100
p.o. 48
4
200–202 ꢀC. 1H NMR (300 MHz, DMSO-d6)
d: 10.50 (s, 1H), 9.10 (s,
1H), 7.75–7.50 (m, 8H), 7.49 (s, 1H, J ¼ 16.00 Hz), 6.50 (s, 1H,
3a
3g
3g
ꢄ4 weeks
40
20
20
p.o. 65
p.o. 35
8
1
2
J ¼ 16.00 Hz). Anal. C15H12ClNO2 (C, H, N).
i.p.
50
5.6.5. N-Hydroxy-E-3-(40-cyanobiphenyl-4-yl)-acrylamide (3d)
Methyl 4-bromocinnamate was reacted with 4-cyanobenzene-
boronic acid (method B) to give, after flash chromatography
(hexane/ethyl acetate 90:10) methyl E-3-(40-cyanobiphenyl-4-yl)-
acrylate as a white solid. Yield 35%, mp 150–152 ꢀC. 1H NMR
IGROV-1 (ovarian
carcinoma)
3a
qd ꢄ 5 d/
week
ꢄ4 weeks
20
p.o. 44
2
3a
3g
3d
40
20
20
p.o. 49
p.o. 71
p.o. 37
6
0
0
a
qd ꢄ 5 d/week, daily treatment for 5 consecutive days in each week. The treat-
(300 MHz, CDCl3)
d: 7.81–7.57 (m, 9H), 6.50 (d, 1H, J ¼ 16.00 Hz),
ment was repeated for the indicated weeks.
3.82 (s, 3H). The above ester was reacted with O-tetrahydropyr-
anylhydroxylamine (method E) to give, after flash chromatography
(hexane/ethyl acetate 60:40) E-3-(40-cyanobiphenyl-4-yl)-N-(tetra-
hydropyran-2-yloxy)-acrylamide. Yield 53%, mp 211–213 ꢀC. 1H
b
Tumor volume inhibition (TVI%) was determined within the first week after
treatment.
c
Body weight loss.
NMR (300 MHz, DMSO-d6) d: 7.83–7.62 (m, 5H), 7.61–7.50 (m, 4H),
mp 180–182 ꢀC. 1H NMR (300 MHz, CDCl3)
d: 7.72 (d, 1H,
6.49 (br s, 1H), 5.12–4.93 (m, 1H), 4.05–3.90 (m, 1H), 3.73–3.58 (m,
1H), 1.97–1.74 (m, 3H), 1.72–1.52 (m, 3H). Alternatively, this inter-
mediate could be obtained by reacting 4-bromobenzonitrile
J ¼ 16.00 Hz), 7.66–7.53 (m, 6H), 7.45 (d, 2H, J ¼ 7.82 Hz), 6.45 (d,1H,
J ¼ 16.00 Hz), 4.75 (s, 2H), 3.81 (s, 3H). The above ester was
hydrolyzed (method C) to the corresponding acid. Yield 85%, mp
with
E-3-(40-bromophenyl)-N-(tetrahydropyran-2-yloxy)-acryl-
265–266 ꢀC. 1H NMR (300 MHz, DMSO-d6)
d: 7.77–7.58 (m, 7H),
amide (method A). Yield 82%. The product was treated accord-
ing to method E to give the title compound. Yield 62%, mp
7.37 (d, 2H, J ¼ 7.82 Hz), 6.52 (d, 1H, J ¼ 16.00 Hz), 5.20 (t, 1H,
J ¼ 5.58 Hz), 4.50 (d, 2H, J ¼ 5.58 Hz). This acid was condensed with
hydroxylamine hydrochloride (method D, HATU) to give the title
compound as a white solid. Yield 65%, mp 220–223 ꢀC dec. 1H NMR
212–214 ꢀC. 1H NMR (300 MHz, DMSO-d6)
d: 10.80 (s, 1H), 9.05
(s, 1H), 8.00–7.80 (m, 4H), 7.82 (d, 2H, J ¼ 8.20 Hz), 7.69 (d, 2H,
J ¼ 8.20 Hz), 7.51 (d, 1H, J ¼ 15.30 Hz), 6.54 (d, 1H, J ¼ 15.30 Hz).
Anal. C16H12N2O2 (C, H, N).
(300 MHz, DMSO-d6) d: 10.75 (s,1H), 9.03 (s,1H), 7.77–7.54 (m, 6H),
7.47 (d, 1H, J ¼ 16.00 Hz), 7.38 (d, 2H, J ¼ 7.82 Hz), 6.47 (d, 1H,
J ¼ 16.00 Hz), 5.20 (t, 1H, J ¼ 5.58 Hz), 4.51 (2H, d, J ¼ 5.58 Hz). Anal.
C16H15NO3 (C, H, N).
5.6.6. N-Hydroxy-E-3-(4-pyridin-4-yl-phenyl)-acrylamide (3e)
Methyl 4-bromocinnamate was reacted with pyridine-4-
boronic acid (method B) to give, after flash chromatography
(hexane/ethyl acetate 50:50) methyl E-3-(4-pyridin-4-yl-phenyl)-
5.6.4. N-Hydroxy-E-3-(40-chlorobiphenyl-4-yl)-acrylamide (3c)
Methyl 4-bromocinnamate was reacted with 4-chlor-
obenzeneboronicacid(methodB)togive, afterflashchromatography
(hexane/ethyl acetate 92:8) methyl E-3-(40-chlorobiphenyl-4-yl)-
acrylate. Yield 63%, mp133–134 ꢀC. 1H NMR (300 MHz, CDCl3)
d:
8.70 (d, 2H, J ¼ 4.84 Hz), 7.80–7.58 (m, 7H), 6.51 (d, 1H,
J ¼ 16.00 Hz), 3.82 (s, 3H). This ester was hydrolysed (method C) to
give the corresponding acid as a white solid. Yield 85%, mp 270 ꢀC.
acrylate. Yield 46%, mp 155–156 ꢀC.1H NMR (300 MHz, CDCl3)
d: 7.71
1H NMR (300 MHz, DMSO-d6)
d
: 8.87 (d, 2H, J ¼ 4.84 Hz), 8.21 (d,
2H, J ¼ 4.84 Hz), 8.02 (d, 2H, J ¼ 8.19 Hz), 7.90 (d, 2H, J ¼ 8.19 Hz),
7.66 (d, 1H, J ¼ 16.00 Hz), 6.71 (d, 1H, J ¼ 16.00 Hz). This acid was
condensed with hydroxylamine hydrochloride (method D, HBTU)
to obtain, after flash chromatography on reverse phase (CH2Cl2/
MeOH 95:5), the title compound. Yield 36%, mp 180 ꢀC dec. 1H NMR
2000
1500
1000
500
0
(300 MHz, DMSO-d6) d: 10.82 (s, 1H), 9.11 (br s, 1H), 8.64 (d, 2H,
J ¼ 4.84 Hz), 7.86 (d, 2H, J ¼ 8.19 Hz), 7.79–7.66 (m, 4H), 7.51 (d, 1H,
J ¼ 16.00 Hz), 6.54 (d, 1H, J ¼ 16.00 Hz). MS (EI) m/z: 241 (Mþ, 100).
Anal. C14H12N2O2 (C, H, N).
5.6.7. N-Hydroxy-E-3-(40-formylbiphenyl-4-yl)-acrylamide (3f)
Methyl 4-bromocinnamate was reacted with 4-formylben-
zeneboronic acid (method B) to give, after flash chromatography
(hexane/ethyl acetate 90:10), methyl E-3-(40-formylbiphenyl-4-yl)-
acrylate as a yellow solid. Yield 32%, mp 160–162 ꢀC. 1H NMR
(300 MHz, CDCl3)
d
: 10.06 (s, 1H), 7.96 (d, 2H, J ¼ 8.19 Hz), 7.76 (d,
0
10
20
Days after tumor implant
30
40
50
2H, J ¼ 8.19 Hz), 7.73 (d, 1H, J ¼ 16.00 Hz), 7.67–7.62 (m, 4H), 6.50 (d,
1H, J ¼ 16.00 Hz), 3.82 (s, 3H). This ester was hydrolyzed (method C)
to the corresponding acid. Yield 22%, mp 209–212 ꢀC. 1H NMR
(300 MHz, DMSO-d6) d: 10.06 (s, 1H), 7.94–7.74 (m, 9H). This acid
was condensed (method D, HATU) with hydroxylamine hydro-
chloride to afford 8 mg of the title compound. Yield 25%, mp
Fig. 5. Antitumor activity of compound 3g against the ovarian carcinoma model
IGROV-1. Animals were treated by oral route with 20 mg/kg daily (5 daily adminis-
tration for week for a total of 20 doses). (B) control (solvent-treated animals); (-)
drug-treated animals. Drug treatment started when tumors were just measurable (w
100 mm3).

S. Dallavalle et al. / European Journal of Medicinal Chemistry 44 (2009) 1900–1912
1909
188–190 ꢀC dec. 1H NMR (300 MHz, DMSO-d6)
10.78 (s, 1H), 9.08 (br s, 1H), 8.10–7.37 (m, 9H), 6.52 (d, 1H,
J ¼ 15.26 Hz). Anal. C16H13NO3 (C, H, N).
d
: 11.30 (br s, 1H),
phenyl]-acrylate. Yield 26%, mp 162–164 ꢀC. 1H NMR (300 MHz,
CDCl3)
(m, 2H), 7.63–7.54 (m, 3H), 7.46 (br s, 2H), 6.64–6.57 (m, 1H), 6.46
(d, 1H, J ¼ 16 Hz), 3.81 (s, 3H). The above ester was hydrolysed
(method C) to give the corresponding acid. Yield 70%, mp 230 ꢀC. 1H
d: 8.21 (br s,1H), 7.88 (s,1H), 7.73 (d,1H, J ¼ 16 Hz), 7.70–7.64
5.6.8. N-Hydroxy-E-3-(40-methoxybiphenyl-4-yl)-acrylamide (3g)
Methyl 4-bromocinnamate was reacted with 4-methox-
ybenzeneboronicacid(methodB)togive, afterflashchromatography
(hexane/ethyl acetate 95:5 then hexane/ethyl acetate 80:20), methyl
E-3-(40-methoxybiphenyl-4-yl)-acrylate as a pale yellow solid. Yield
NMR (300 MHz, DMSO-d6) d: 11.18 (s,1H), 7.88 (s,1H), 7.78–7.70 (m,
5H), 7.62 (d, 1H, J ¼ 16.00 Hz), 7.49–7.43 (m, 2H), 7.39 (d, 1H,
J ¼ 2.61 Hz), 6.58–6.47 (m, 2H). This product was condensed with
hydroxylamine hydrochloride (method D, HBTU) to give the title
compound as a brown solid. Yield 55%, mp 215 ꢀC dec. 1H NMR
51%. 1H NMR (300 MHz, CDCl3)
d
: 7.71 (d,1H, J ¼ 16.00 Hz), 7.65–7.50
(m, 6H), 6.97 (d, 2H, J ¼ 8.39 Hz), 6.46 (d, 1H, J ¼ 16.00 Hz), 3.85 (s,
3H), 3.81 (s, 3H). This ester was reacted with O-tetrahydropyr-
anylhydroxylamine (method E) to give E-3-(40-methoxybiphenyl-4-
yl)-N-(tetrahydropyran-2-yloxy)-acrylamide as a yellow solid. Yield
(300 MHz, DMSO-d6) d: 11.18 (s, 1H), 10.75 (br s, 1H), 9.04 (br s, 1H),
7.87 (s, 1H), 7.72 (d, 2H, J ¼ 7.82 Hz), 7.61 (d, 2H, J ¼ 7.82 Hz), 7.48–
7.34 (m, 4H), 6.55–6.42 (m, 2H). Anal. C17H14N2O2 (C, H, N).
100%.1HNMR(300 MHz, CDCl3)
d
:7.74(d,1H,J ¼ 15.64 Hz), 7.64–7.47
5.6.12. N-Hydroxy-E-3-[30-chloro-40-hydroxybiphenyl-4-yl]-
acrylamide (3m)
(m, 6H), 6.97 (d, 2H, J ¼ 8.39 Hz), 6.44 (br s, 1H), 5.13–4.93 (m, 1H),
4.06–3.90 (m, 1H), 3.84 (s, 3H), 3.74–3.61 (m, 1H), 1.99–1.75 (m, 3H),
1.74–1.45 (m, 3H). Alternatively, this compound could be prepared by
reacting E-3-(40-bromophenyl)-N-(tetrahydropyran-2-yloxy)-acryl-
amide with 4-methoxybenzeneboronic acid (method B). Yield 100%.
The product was treated according to method E to give the title
compound as a white solid. Yield 88%. 1H NMR (300 MHz, DMSO-d6)
Condensation of methyl 4-bromocinnamate (463 mg,
1.92 mmol) with 2-chloro-4-bromophenol (200 mg, 0.96 mmol)
was accomplished according to procedure A to give ester 2m as
a white solid (40%) after flash chromatography (hexane/ethyl
acetate 8:2), mp 153–154 ꢀC. 1H NMR (300 MHz, CDCl3)
d: 7.71 (d,
1H, J ¼ 16 Hz), 7.50–7.65 (m, 5H), 7.43 (dd, 1H, J ¼ 2.23, 8.56 Hz),
7.09 (d, 1H, J ¼ 8.56 Hz), 6.46 (d, 1H, J ¼ 16 Hz), 3.81 (3H, s). The
ester was hydrolyzed (procedure C) to obtain the corresponding
d: 10.76 (s, 1H), 9.05 (s, 1H), 7.73–7.56 (m, 6H), 7.48 (d, 1H,
J ¼ 15.60 Hz), 7.03 (d, 2H, J ¼ 8.90 Hz), 6.47 (d, 1H, J ¼ 15.60 Hz), 3.80
(s, 3H). Anal. C16H15NO3 (C, H, N).
acid in 94% yield, mp 233–234 ꢀC. 1H NMR (300 MHz, DMSO-d6)
d:
12.45 (br s, 1H), 10.45 (br s, 1H), 7.60–7.80 (m, 5H), 7.69 (d, 1H,
J ¼ 16 Hz), 7.53 (dd, 1H, J ¼ 1.86, 8.56 Hz), 7.05 (d, 1H, J ¼ 8.56 Hz),
6.55 (d, 1H, J ¼ 16 Hz). The acid was condensed with hydroxylamine
hydrochloride according to procedure D (HATU). Purification by
5.6.9. N-Hydroxy-E-3-(40-thiophen-2-yl-phenyl)-acrylamide (3h)
Methyl 4-bromocinnamate was reacted with thiophene-2-
boronic acid (method B) to give methyl-E-3-(40-thiophen-2-yl-
phenyl)-acrylate as a pale yellow solid. Yield 43%, mp 158–162 ꢀC. 1H
flash chromatography RP-18 (40–63 mm) reverse phase (CH3OH/
NMR (300 MHz, CDCl3)
d
: 7.68 (d, 1H, J ¼ 16.00 Hz), 7.62 (d, 2H,
H2O 50:50) and crystallization from diethyl ether afforded the title
J ¼ 8.19 Hz), 7.53 (d, 2H, J ¼ 8.19 Hz), 7.42–7.28 (m, 2H), 7.15–7.05 (m,
1H), 6.43(d,1H,J ¼ 16.00 Hz), 3.83(s, 3H). Thisesterwasreactedwith
O-tetrahydropyranylhydroxylamine (method E) to give E-3-(40-thi-
ophen-2-yl)phenyl)-N-(tetrahydropyran-2-yloxy)-acrylamide. Yield
compound as a white solid. Yield 24%, mp 172–175 ꢀC dec. 1H NMR
(300 MHz, DMSO-d6) d: 10.77 (s, 1H), 10.42 (s, 1H), 9.06 (br s, 1H),
7.77–7.40 (m, 7H), 7.05 (d,1H, J ¼ 8.56 Hz), 6.47 (d, 1H, J ¼ 15.63 Hz).
Anal. C15H12ClNO3 (C, H, N).
100%, mp120–122 ꢀC.1HNMR (300 MHz, CDCl3)
d: 7.77–7.57 (m, 2H),
7.55–7.44 (m, 3H), 7.41–7.27 (m, 2H), 7.13–7.04 (m,1H), 6.39 (br s,1H),
5.08–4.94(m,1H), 4.03–3.89(m,1H), 3.72–3.60(m,1H),1.94–1.77 (m,
3H), 1.72–1.52 (m, 3H). Alternatively, this intermediate could be
prepared by reacting E-3-(40-bromophenyl)-N-(tetrahydropyran-2-
yloxy)-acrylamide with thiophene-2-boronic acid (method B). Yield
88%. The product was treated according to method E to give the title
compound. Yield 60%, mp 182–184 ꢀC dec. 1H NMR (300 MHz,
5.6.13. N-Hydroxy-3-(40-hydroxybiphenyl-4-yl)-propionamide (4)
3-(40-Hydroxybiphenyl-4-yl)-acrylic acid methyl ester [21]
(500 mg, 1.97 mmol) was dissolved in 250 mL of ethyl acetate, then
PtO2 (291 mg, 1.3 mmol) was added and the mixture was hydroge-
nated for 1 h. After filtration of the catalyst, the solvent was evapo-
rated to obtain 390 mg of 3-(40-hydroxybiphenyl-4-yl)-propionic
acid methyl ester. Yield 77%, mp 127–128 ꢀC. 1H NMR (300 MHz,
DMSO-d6)
d
: 10.77 (s, 1H), 9.07 (br s, 1H), 7.70 (d, 2H, J ¼ 8.19 Hz),
CDCl3) d: 7.56–7.37 (m, 4H), 7.32–7.17 (m, 2H), 6.87 (d, 2H,
7.65–7.55 (m, 4H), 7.46 (d, 1H, J ¼ 16.00 Hz), 7.15 (t, 1H, J ¼ 4.47 Hz),
J ¼ 8.56 Hz), 3.67 (s, 3H), 2.97 (t, 2H, J ¼ 7.44 Hz), 2.68 (t, 2H,
J ¼ 7.44 Hz). The above ester (390 mg, 1.52 mmol) was hydrolyzed
according to method C. The solid formed was filtered and dried to
give 368 mg of 3-(40-hydroxybiphenyl-4-yl)-propionic acid. Yield
6.48 (d, 1H, J ¼ 16.00 Hz). Anal. C13H11NO2S (C, H, N).
5.6.10. N-Hydroxy-E-3-[40-dimethylaminobiphenyl-4-yl]
acrylamide (3i)
100%, mp 207–208 ꢀC. 1H NMR (300 MHz, DMSO-d6)
d: 12.13 (s, 1H),
4-Bromo-N,N-dimethylaniline was reacted with methyl 4-bro-
mocinnamate according to method A to give methyl E-3-(40-dime-
thylaminobiphenyl-4-yl)-acrylate. Yield 8%. 1H NMR (300 MHz,
9.62 (s, 1H), 7.50–7.40 (m, 4H), 7.23 (d, 2H, J ¼ 7.82 Hz), 6.81 (d, 2H,
J ¼ 8.56 Hz), 2.80 (t, 2H, J ¼ 7.44 Hz), 2.52 (t, 2H, J ¼ 7.44 Hz). The
above acid was condensed according to method D (HBTU), to give the
title compound as a white solid. Yield 92%, mp 180–182 ꢀC. 1H NMR
CDCl3)
d
: 7.73 (d, 1H, J ¼ 16.00 Hz), 7.70–7.53 (m, 8H), 6.48 (d, 1H,
J ¼ 16.00 Hz), 3.83 (s, 3H), 3.16 (s, 3H), 2.97 (s, 3H). The above ester
(300 MHz, DMSO-d6) d: 10.38 (s, 1H), 9.49 (s, 1H), 8.71 (s, 1H), 7.49–
was hydrolyzed (method C) to give the corresponding acid. Yield
7.39 (m, 4H), 7.21 (d, 2H, J ¼ 7.82 Hz), 6.82 (d, 2H, J ¼ 8.56 Hz), 2.81 (t,
77%, mp 314–315 ꢀC. 1H NMR (300 MHz, DMSO-d6)
d: 7.75–7.55 (m,
2H, J ¼ 7.44 Hz), 2.26 (t, 2H, J ¼ 7.44 Hz). Anal. C15H15NO3 (C, H, N).
7H), 7.10–6.90 (m, 2H), 6.54 (d, 1H, J ¼ 16 Hz), 3.00 (s, 6H). This acid
was condensed with hydroxylamine hydrochloride (method D,
HBTU) to give the title compound. Yield 68%, mp 260–263 ꢀC dec. 1H
5.6.14. E-3-(4-Bromophenyl)-N-(tetrahydropyran-2-yloxy)-
acrylamide (5)
NMR (300 MHz, DMSO-d6)
d
: 10.73 (s, 1H), 9.03 (s, 1H), 7.70–7.52
E-4-Bromocinnamic acid (1.28 g, 5.68 mmol) was dissolved
under nitrogen in 100 mL of anhydrous DMF, then 1.15 g
(8.53 mmol) of HOBt and 1.30 g (6.82 mmol) of WSC were added and
the solution was stirred at room temperature for 3 h. After addition
of O-tetrahydropyranylhydroxylamine (1,00 g, 8.53 mmol), the
mixture was heated at 50 ꢀC for 1 h. DMF was evaporated under
reduced pressure and the residue was taken up with CH2Cl2. The
(m, 6H), 7.46 (d, 1H, J ¼ 16.38 Hz), 6.80 (d, 2H, J ¼ 8.93 Hz), 6.44 (d,
1H, J ¼ 16.38 Hz), 2.95 (s, 6H). Anal. C17H18N2O2 (C, H, N).
5.6.11. N-Hydroxy-E-3-[4-(1H-indol-5-yl)-phenyl]-acrylamide (3l)
4-Bromoindole was reacted with methyl 4-bromocinnamate
according to procedure A to give methyl E-3-[4-(1H-indol-5-yl)-

1910
S. Dallavalle et al. / European Journal of Medicinal Chemistry 44 (2009) 1900–1912
organic phase was washed with satd. aq. NaHCO3, dried over Na2SO4
and filtered. The solvent was evaporated under reduced pressure to
give a crude product which was purified by flash chromatography
(CH2Cl2/CH3OH 99:1) to afford 1.56 g of the title compound as
compound was treated according to method E to afford the title
compound. Yield 58%, mp 200–202 ꢀC dec. 1H NMR (300 MHz,
DMSO-d6) d: 10.73 (s, 1H), 8.96 (s, 1H), 7.73–7.53 (m, 6H), 7.25 (dd,
1H, J ¼ 14.89, 10.42 Hz), 7.02–6.97 (m, 4H), 6.01 (d,1H, J ¼ 14.89 Hz),
a white solid. Yield 84%, mp 119–120 ꢀC.1H NMR (300 MHz, CDCl3)
d:
3.80 (s, 3H). Anal. C18H17NO3 (C, H, N).
8.31 (br s,1H), 7.66 (d,1H, J ¼ 16.00 Hz), 7.50 (d, 2H, J ¼ 7.82 Hz), 7.36
(d, 2H, J ¼ 7.82 Hz), 6.37 (br s, 1H), 4.99 (br s, 1H), 4.01–3.88 (m, 1H),
3.71–3.59 (m, 1H), 1.92–1.77 (m, 3H), 1.70–1.57 (m, 3H).
5.6.19. E,E-5-(40-Cyano-biphenyl-4-yl)-penta-2,4-dienoic acid N-
hydroxyamide (10c)
5-(4-Bromophenyl)-penta-2,4-dienoic acid (tetrahydropyran-2-
yloxy)-amide 9 was reacted with 4-cyanobenzeneboronic acid
(method B) to give, after flash chromatography (ethyl acetate/
hexane 60:40), 5-(40-cyanobiphenyl-4-yl)-penta-2,4-dienoic acid
(tetrahydropyran-2-yloxy)-amide as a yellow solid. Yield 42%, mp
5.6.15. E-3-(Biphenyl-4-yloxy)-N-hydroxy-acrylamide (7)
4-Phenylphenol (160 mg, 0.94 mmol) was dissolved in 2 mL of
CH3CN and treated with 6
mL (0.05 mmol) of 4-methylmorpholine.
After stirring for 20 min at room temperature, an amount of 115
mL
(1.13 mmol) of ethyl propiolate was dropped. Dilution with water,
extraction with ethyl acetate, washing with brine, drying over
Na2SO4, filtering, evaporating, and flash chromatography (hexane/
ethyl acetate 80:20) gave 230 mg of ethyl E-3-(biphenyl-4-yloxy)-
193–195 ꢀC. 1H NMR (300 MHz, CDCl3)
d: 8.22 (s, 1H), 7.79–7.64 (m,
4H), 7.63–7.46 (m, 5H), 6.97–6.88 (m, 2H), 6.05 (br s, 1H), 5.04–4.90
(m, 1H), 4.02–4.89 (m, 1H), 3.71–3.60 (m, 1H), 1.93–1.77 (m, 3H),
1.71–1.57 (m, 3H). This compound was treated according to method
E to give the title compound. Yield 56%, mp 202–205 ꢀC dec. 1H
acrylate. Yield 91%, mp 74–76 ꢀC. 1H NMR (300 MHz, CDCl3)
d: 7.82
(d, 1H, J ¼ 12.28 Hz), 7.63–7.50 (m, 5H), 7.48–7.38 (m, 2H), 7.13 (d,
2H, J ¼ 8.56 Hz), 5.58 (d, 1H, J ¼ 12.28 Hz), 4.19 (q, 2H, J ¼ 7.07 Hz),
1.28 (t, 3H, J ¼ 7.07 Hz). The above ester was hydrolysed (method C)
to give the corresponding acid. Yield 53%, mp 161–162 ꢀC. 1H NMR
NMR (300 MHz, DMSO-d6) d: 10.75 (s, 1H), 8.99 (s, 1H), 8.05–7.86
(m, 4H), 7.79 (d, 2H, J ¼ 8.19 Hz), 7.69 (d, 2H, J ¼ 8.19 Hz), 7.35–7.10
(m, 2H), 7.01 (d, 1H, J ¼ 14.51 Hz), 6.05 (d, 1H, J ¼ 14.14 Hz). Anal.
C18H14N2O2 (C, H, N).
(300 MHz, acetone-d6)
d
: 7.87 (d, 1H, J ¼ 12.29 Hz), 7.73 (d, 2H,
J ¼ 9.00 Hz), 7.68–7.61 (m, 2H), 7.50–7.41 (m, 2H), 7.39–7.34 (m,1H),
7.28 (d, 2H, J ¼ 9.00 Hz), 5.54 (d, 1H, J ¼ 12.29 Hz). This acid was
condensed with hydroxylamine hydrochloride (method D, HBTU)
to afford the title compound. Yield 19%, mp 123–125 ꢀC. 1H NMR
5.6.20. 5-(Biphenyl-4-yl)-pentanoic acid N-hydroxyamide (11)
5-Biphenyl-4-yl-pentanoic acid [24] was condensed with of
hydroxylamine hydrochloride (method D, HBTU) to afford the title
compound. Yield 62%, mp 148–151 ꢀC. 1H NMR (300 MHz, DMSO-
(300 MHz, DMSO-d6)
(m, 5H), 7.55–7.19 (m, 5H), 5.56 (d, 1H, J ¼ 12.28 Hz).
d
: 10.44 (br s, 1H), 8.87 (br s, 1H), 7.85–7.57
d6)
d
: 10.37 (s, 1H), 8.70 (s, 1H), 7.65 (d, 2H, J ¼ 8.34 Hz), 7.57 (d, 2H,
J ¼ 7.44 Hz), 7.47 (t, 2H, J ¼ 7.44 Hz), 7.36 (d, 1H, J ¼ 7.44 Hz), 7.27 (d,
2H, J ¼ 8.34 Hz), 2.65–2.55 (m, 2H), 2.02–1.90 (m, 2H),1.66–1.45 (m,
4H). Anal. C17H19NO2 (C, H, N).
5.6.16. 5-(4-Bromophenyl)-penta-2,4-dienoic acid
(tetrahydropyran-2-yloxy)-amide (9)
A solution of 700 mg (2.5 mmol) of 5-(4-bromophenyl)-penta-
2,4-dienoic acid ethyl ester [22] and 293 mg (2.5 mmol) of O-tet-
rahydropyranylhydroxylamine in 33 mL of anhydrous THF was
cooled to ꢁ78 ꢀC, then treated with 5.12 mL (5.25 mmol) of lithium
hexamethyldisilazane. The mixture was stirred for 40 min under
nitrogen, then was quenched with NH4Cl saturated solution. Once
at room temperature, the product was extracted with ethyl acetate
(3 ꢄ 30 mL) and the organic phase was dried over Na2SO4. The
solvent was evaporated under reduced pressure to afford the title
5.6.21. N-hydroxy-E-3-[40-Hydroxybiphenyl-3-yl]-acrylamide (13)
Methyl E-3-bromophenylcinnamate was reacted with 4-
hydroxybenzeneboronic acid according to method B and the
product was purified by flash chromatography (hexane/ethyl
acetate 85:15) to afford methyl E-3-(40-hydroxybiphenyl-3-yl)-
acrylate as a pale yellow solid. Yield 25%, mp 128–129 ꢀC. 1H NMR
(300 MHz, CDCl3)
d
: 7.73 (d, 1H, J ¼ 15.6 Hz), 7.66 (s, 1H), 7.57–7.38
(m, 5H), 6.91 (d, 2H, J ¼ 8.2 Hz), 6.48 (d,1H, J ¼ 15.6 Hz), 3.81 (s, 3H).
The above ester was hydrolyzed according to method C to give E-3-
(40-hydroxybiphenyl-3-yl)-acrylic acid. Yield 64%, mp 229–230 ꢀC.
compound as a yellow solid. Yield 96%. 1H NMR (300 MHz, CDCl3)
d:
11.23 (s, 1H), 7.60–7.47 (m, 4H), 7.33–7.04 (m, 2H), 6.96 (d, 1H,
J ¼ 15.26 Hz), 6.06 (d,1H, J ¼ 15.26 Hz), 4.87 (br s,1H), 4.01–3.86 (m,
1H), 3.58–3.47 (m, 1H), 1.75–1.62 (m, 3H), 1.60–1.46 (m, 3H).
1H NMR (300 MHz, DMSO-d6)
d: 7.86 (s, 1H), 7.70–7.62 (m, 5H), 7.45
(t, 1H, J ¼ 8.56 Hz), 6.86 (d, 2H, J ¼ 8.19 Hz), 6.63 (d, 1H,
J ¼ 15.63 Hz). The acid was condensed with hydroxylamine
hydrochloride according to method D (HBTU) to give, after flash
chromatography on reverse phase (CH2Cl2/MeOH 95:5), the title
compound as a white solid. Yield 23%, mp 127–128 ꢀC. 1H NMR
5.6.17. E,E-5-Biphenyl-4-yl-penta-2,4-dienoic acid N-
hydroxyamide (10a)
5-Biphenyl-4-yl-penta-2,4-dienoic acid [23] was condensed
with hydroxylamine hydrochloride (method D, HBTU) to afford the
title compound. Yield 31%, mp 178–180 ꢀC dec. 1H NMR (300 MHz,
(300 MHz, DMSO-d6) d: 10.73 (s, 1H), 9.59 (s, 1H), 9.05 (s, 1H), 7.74
(s, 1H), 7.65–7.37 (m, 6H), 6.85 (d, 2H, J ¼ 8.19 Hz), 6.53 (d, 1H,
J ¼ 15.63 Hz). Anal. C15H13NO3 (C, H, N).
DMSO-d6) d: 10.75 (s, 1H), 9.00 (s, 1H), 7.75–7.60 (m, 6H), 7.50–7.45
(m, 2H), 7.40 (m, 1H), 7.40–6.90 (m, 3H), 6.02 (s, 1H, J ¼ 14.89 Hz).
5.6.22. N-Hydroxy-E-3-[4-(4-hydroxyphenyl)-cyclohexyl]-
acrylamide (17)
Anal. C17H15NO2 (C, H, N).
To a solution of 4-(4-hydroxyphenyl)-cyclohexylaldehyde [25]
(204 mg, 1 mmol) in 10 mL of THF was added 501 mg (1.5 mmol) of
methyl(triphenylphosphoranylidene)acetate, and the mixture
refluxed 4 h. Adding water (1 mL), extracting with AcOEt (5 mL)
and flash chromatography with hexane/AcOEt 85:15 gave 214 mg
(82%) of ethyl E-3-[4-(4-hydroxyphenyl)-cyclohexyl]-acrylate (15).
5.6.18. E,E-5-(40-Methoxybiphenyl-4-yl)-penta-2,4-dienoic acid N-
hydroxyamide (10b)
5-(4-Bromophenyl)-penta-2,4-dienoic acid (tetrahydropyran-2-
yloxy)-amide 9 was reacted with 4-methoxybenzeneboronic acid
(method B) to give, after flash chromatography (ethyl acetate/
hexane 70:30) 5-(40-methoxybiphenyl-4-yl)-penta-2,4-dienoic
acid (tetrahydropyran-2-yloxy)-amide as a yellow solid. Yield 34%,
1H NMR (300 MHz, CDCl3)
d
: 7.10 (d, 2H, J ¼ 8.3 Hz), 6.85 (dd, 1H,
J ¼ 7.0,15.6 Hz), 6.73 (d, 2H, J ¼ 8.3 Hz), 5.77 (d,1H, J ¼ 15.6 Hz), 3.72
(s, 3H, OMe), 2.5–1.2 (10H), and 5% of the Z diastereoisomer. This
ester was hydrolyzed (method C) to give the corresponding acid
mp 137–139 ꢀC. 1H NMR (300 MHz, CDCl3)
d: 8.41 (br s, 1H), 7.59–
7.41 (m, 6H), 7.29 (d, 1H, J ¼ 8.56 Hz), 6.96–6.79 (m, 4H), 6.04 (d, 1H,
J ¼ 15.26 Hz), 5.08–4.87 (m, 1H), 4.05–3.90 (m, 1H), 3.84 (s, 3H),
3.71–3.56 (m, 1H), 1.94–1.75 (m, 3H), 1.71–1.50 (m, 3H). This
(16). Yield 81%, mp 222–225 ꢀC. 1H NMR (300 MHz, acetone-d6)
d:
7.04 (d, 2H, J ¼ 8.34 Hz), 6.91 (dd, 1H, J ¼ 7.03, 15.59 Hz), 6.74 (d, 2H,

S. Dallavalle et al. / European Journal of Medicinal Chemistry 44 (2009) 1900–1912
1911
J ¼ 8.34 Hz), 5.80 (d, 1H, J ¼ 15.59 Hz), 2.51–2.33 (m, 1H), 2.32–2.14
(m, 1H), 1.93–1.75 (m, 4H), 1.60–1.41 (m, 2H), 1.40–1.23 (m, 2H). The
above acid was condensed with hydroxylamine hydrochloride
(method D, HBTU) to give the title compound. Yield 13%, mp 182–
components were incubated at 37 ꢀC for 40 min. The reaction was
quenched with the addition of 50 L HDAC-FDL Developer (KI-105,
M TSA. The
plates were incubated 30 min at room temperature to allow the
fluorescent signal to develop. The fluorescence generated was
monitored at 355/460 nm (excitation/emission) wavelength.
m
Biomol) 20ꢄ stock diluted in KI-143 buffer with 2
m
185 ꢀC. 1H NMR (300 MHz, acetone-d6)
d: 10.10 (br s, 1H), 8.04 (br s,
1H), 7.05 (d, 2H, J ¼ 7.72 Hz), 6.84–6.69 (m, 3H), 5.84 (d, 1H,
J ¼ 15.44 Hz), 2.50–2.33 (m, 1H), 2.28–2.11 (m, 1H), 1.96–1.76 (m,
4H), 1.61–1.41 (m, 2H), 1.40–1.31 (m, 2H). Anal. C15H19NO3 (C, H, N).
5.9. Cellular sensitivity to drugs
5.6.23. N-Hydroxy-3-[4-(4-hydroxyphenyl)-cyclohexyl]-
propionamide (18)
Cellular sensitivity to drugs was evaluated by growth-inhibition
assay after 72-h drug exposure. Cells in the logarithmic phase of
growth were seeded in duplicate into 6-well plates. Twenty-four
hours after seeding, the drug was added to the medium. Cells were
harvested 72 h after drug exposure and counted with a cell counter.
IC50 is defined as the drug concentration causing a 50% reduction of
cell number compared with that of untreated control.
3-[4-(4-Hydroxyphenyl)-cyclohexyl]-acrylic
acid
(38 mg,
0.157 mmol) was dissolved in 1.5 mL of MeOH, then a catalytic
amount of Pd/C 10% was added and the mixture was hydrogenated
for 40 min. After filtration of the catalyst, the solvent was evaporated
to obtain 30 mg of 3-[4-(4-hydroxyphenyl)-cyclohexyl]-propionic
acid. Yield 80%, mp 178–183 ꢀC. 1H NMR (300 MHz, acetone-d6)
d:
7.01 (d, 2H, J ¼ 8.54 Hz), 6.71 (d, 2H, J ¼ 8.54 Hz), 2.45–2.17 (m, 5H),
1.92–1.67 (m, 5H), 1.67–0.95 (m, 4H). The above acid was condensed
with hydroxylamine hydrochloride (method D, HBTU) to give the
title compound. Yield 35%, mp 149–152 ꢀC. 1H NMR (300 MHz,
5.10. Cell cycle analysis
The cell cycle distribution was analyzed in propidium iodide-
stained cells by FACScan flow cytometry, as described [27].
acetone-d6)
d
: 9.99 (br s, 1H), 8.20 (br s, 1H), 7.01 (d, 2H, J ¼ 8.54 Hz),
6.71 (d, 2H, J ¼ 8.54 Hz), 2.44–2.27 (m, 1H), 2.18–2.05 (m, 2H), 1.91–
5.11. TUNEL assay
1.66 (m, 5H), 1.58–0.97 (m, 4H). Anal. C15H21NO3 (C, H, N).
Apoptosis was determined in ovarian carcinoma IGROV-1 cells
by TUNEL assay following 72-h exposure to the drug and fixed in 4%
paraformaldehyde for 45 min, at room temperature. The in situ cell
death detection kit fluorescein (Roche, Mannheim, Germany) was
used according to manufacturers instructions. Samples were
analyzed by flow cytometry (Becton Dickinson).
5.7. Molecular modelling
Three-dimensional molecular models of the designed
compounds were built on a Silicon Graphics O2, using the programs
Insight II and Discover (Accelrys Inc., San Diego, CA). Minimizations
were performed with the AMBER all-atom forcefield [23] and the
conjugate gradients algorithm. For atomic partial charges of the
ligand atoms Mulliken charges calculated on the minimised struc-
ture using the MOPAC program [26] with the MNDO Hamiltonian
were used. The model of the active site of HDAC2 was based on the
homology model derived and validated by Wang et al. [16] using
HDLP as a template. The model consists essentially of all amino acid
residues within a radius of 12 Å around SAHA, inserted overlapping
the histidines 142 and 143 in the catalytic pocket of the crystal
structure of HDAC8 complexed with SAHA (PDB entry 1T69) [13a]
with the corresponding histidines 141 and 142 in the HDAC2 model,
and then deleting the structure of HDAC8. During energy minimi-
zation the backbone of the amino acids of the site model was frozen,
while the rest was allowed to relax. The intermolecular interaction
energy between the appropriate compound and the HDAC2 model,
after energy minimization of the complex, was obtained using the
docking module in Insight II, where the nonbonding energy,
including Van der Waals and electrostatic energy, is calculated.
5.12. Western blotting
Cells treated with various concentrations of HDAC inhibitors for
4 h were collected and lysed as previously described [19].
5.13. Tumor models and evaluation of antitumor activity
The experiments were performed using female athymic Swiss
nude mice. Mice were maintained in laminar flow rooms keeping
temperature and humidity constant. Mice had free access to food
and water. Experiments were approved by the Ethics Committee for
Animal Experimentation of the Istituto Nazionale Tumori of Milan
according to institutional guidelines.
Exponentially growing tumor cells (107 cells/mouse) were s.c.
injected into the right flank of athymic nude mice. Tumor lines
were achieved by serial s.c. passages of fragments (about
2 ꢄ 2 ꢄ 6 mm) from growing tumors into healthy mice. Animals
were treated 3 days after tumor implantation. Compounds were
dissolved in DMSO and diluted in PBS containing 5% Cremophor to
a final concentration of 10% DMSO.
5.8. HDAC2 assay
HDAC2 was immunoprecipitated from HeLa cells. Whole cell
lysates were obtained by lysing the cells in a buffer containing
20 mM Tris–HCl pH 7.5, 150 mM NaCl, 1 mM EDTA, 10% glycerol,
0.5% NP-40, 1 mM PMSF, protease inhibitors cocktail (Roche).
Lysates were clarified by centrifugation (12,000 ꢄ g) for 10 min at
4 ꢀC and were diluted with TBST (20 mM Tris–HCl pH 7.5, 150 mM
NaCl and 0.1% Tween 20) containing 1 mM PMSF. Purified IgG from
rabbit antisera to HDAC2 (Santa Cruz Biotechnology sc-7899) were
then added and immune complexes allowed to form for 1 h at 4 ꢀC.
Acknowledgements
This work was partially supported by Associazione Italiana per la
Ricerca sul Cancro, Ministero della Salute, and by Fondazione
CARIPLO, Milan, Italy.
References
Protein A-Sepharose (10 mL of settled beads) were added and the
[1] (a) G. Elaut, V. Rogiers, T. Vanhaecke, Curr. Pharm. Des. 13 (2007) 2584–2620;
(b) K.V. Balakin, Y.A. Ivanenkov, A.S. Kiselyov, S.E. Tkachenko, Anticancer
Agents Med. Chem. 7 (2007) 576–592;
(c) V.S. Xu, R.B. Parmigiani, P.A. Marks, Oncogene 26 (2007) 5541–5552;
(d) M. Paris, M. Porcelloni, M. Binaschi, D. Fattori, J. Med. Chem. 51 (2008) 1505–1529.
[2] (a) M. Grunstein, Nature 389 (1997) 349–352;
components mixed on a rotor overnight at 4 ꢀC. Immune complexes
were collected by centrifugation and washed with cold TBST.
HDAC2 activity was assayed with a pan-HDAC substrate (KI-104;
Biomol Research Laboratories Inc., PA, USA) and the reaction was
carried out in half-volume white 96-well plates. The assay
(b) W.L. Cheung, S.D. Briggs, C.D. Allis, Curr. Opin. Cell Biol. 12 (2000) 326–333.

1912
S. Dallavalle et al. / European Journal of Medicinal Chemistry 44 (2009) 1900–1912
[3] J. Taunton, C.A. Hassig, S.L. Schreiber, Science 272 (1996) 408–411.
[4] P.A. Marks, T. Miller, V.M. Richon, Curr. Opin. Pharmacol. 3 (2003) 344–351.
[5] F. McLaughlin, N.B. La Thangue, Biochem. Pharmacol. 68 (2004) 1139–1144.
[6] (a) P.A. Marks, R.A. Rifkind, V.M. Richon, R. Breslow, T. Miller, W.K. Kelly, Nat.
Rev. Cancer 1 (2001) 194–202;
(b) W.K. Kelly, O.A. O’Connor, P.A. Marks, Expert Opin. Investig. Drugs 11
(2002) 1695–1713;
(c) J.E.Bolden,M.J. Peart, R.W. Johnstone, Nat.Rev. DrugDiscov. 5(2006)769–784.
[7] (a) L. Sigalotti, E. Fratta, S. Coral, E. Cortini, A. Covre, H.J. Nicolay, L. Anzalone,
L. Pezzani, A.M. Di Giacomo, E. Fonsatti, F. Coalizzi, M. Altomonte, L. Calabro`,
M. Maio, J. Cell. Physiol. 212 (2007) 330–344;
[14] T.K. Nielsen, C. Hildmann, A. Dickmanns, A. Schwienhorst, R. Ficner, J. Mol.
Biol. 354 (2005) 107–120.
[15] A.J.M. De Ruijter, A.H. van Gennip, H.N. Caron, S. Kemp, A.B.P. van Kuilenburg,
Biochem. J. 370 (2003) 737–749.
[16] D.F. Wang, P. Helquist, N.L. Wiech, O. Wiest, J. Med. Chem. 48 (2005) 6934–6947.
[17] For another example of a dienic spacer, see C.M. Marson, N. Serradji, A.S. Rioja,
S.P. Gastaud, J.P. Alao, R.C. Coombes, D.M. Vigushin, Bioorg. Med. Chem. Lett. 14
(2004) 2477–2481.
[18] W.K. Rasheed, R.W. Johnstone, H.M. Prince, Expert Opin. Investig. Drugs 16
(2007) 659–678.
[19] V. Zuco, F. Zunino, Neoplasia 10 (2008) 587–596.
(b) D. Marchion, P. Mu¨nster, Expert Rev. Anticancer Ther. 7 (2007) 583–598.
[8] H. Hess-Stumpp, Eur. J. Cell Biol. 84 (2005) 109–121.
[20] V. Zuco, C. Zanchi, G. Cassinelli, C. Lanzi, R. Supino, C. Pisano, R. Zanier,
V. Giordano, E. Garattini, F. Zunino, Cell Death Differ. 11 (2004) 280–289.
[21] R. Cincinelli, S. Dallavalle, R. Nannei, S. Carella, D. De Zani, L. Merlini, S. Penco,
E. Garattini, G. Giannini, C. Pisano, L. Vesci, P. Carminati, V. Zuco, C. Zanchi,
F. Zunino, J. Med. Chem. 48 (2005) 4931–4946.
[22] L.M. Werbel, N. Headen, E.F. Elslager, J. Med. Chem. 10 (1967) 366–370.
[23] E. Grovenstein Jr., S. Akabori, J. Am. Chem. Soc. 97 (1975) 4620–4626.
[24] S.M. Kelly, A. Germann, M. Schadt, Liq. Cryst. 16 (1994) 491–507.
[25] (a) P. Weiner, J. Comput. Chem. 2 (1981) 287–303;
[9] W.K. Kelly, P.A. Marks, Nat. Clin. Pract. Oncol. 2 (2005) 150–157.
[10] C. S.Grant, P. EasleyKirkpatrick, Nat. Rev. Drug Discov. 6 (2007) 21–22.
[11] S.C. Hodawadekar, R. Marmorstein, Oncogene 26 (2007) 5528–5540.
[12] M.S. Finnin, J.R. Donigian, A. Cohen, V.M. Richon, R.A. Rifkind, P.A. Marks,
R. Breslow, N.P. Pavletich, Nature 401 (1999) 188–193.
[13] (a) J.R. Somoza, R.J. Skene, B.A. Kats, C. Mol, J.D. Ho, A.J. Jennings, C. Luong,
A. Arvai, J.J. Buggy, E. Chi, C. Tang, B.C. Sang, E. Verner, R. Wynands, E.M. Leahy,
D.R. Dougan, G. Snell, G. Navre, M.V. Knuth, R.V. Swanson, D.E. McRee,
L.W. Tari, Structure 12 (2004) 1325–1334;
(b) S.J.K. Weiner, D.A. Case, U.C. Singh, C. Ghio, G. Alagona, S. Profeta, P. Weiner,
J. Am. Chem. Soc. 106 (1984) 765–784.
(b) A. Vannini, C. Volpari, G. Filocamo, E. Caroli Casavola, M. Brunetti,
D. Renzoni, P. Chakravarty, C. Paolini, R. De Francesco, P. Gallinari, C. Steinku¨hler,
S. Di Marco, Proc. Natl. Acad. Sci. U.S.A. 101 (2004) 15064–15069.
[26] J.J.P. Stewart, J. Comput. Aided Mol. Des. 4 (1990) 1–105.
[27] C. Hubbert, A. Guardiola, R. Shao, Y. Kawaguchi, A. Ito, A. Nixon, M. Yoshida,
X.F. Wang, T.P. Yao, Nature 417 (2002) 455–458.