Isoxazolo(aza)naphthoquinones: A new class of cytotoxic Hsp90 inhibitors

Article abstract of DOI:10.1016/j.ejmech.2012.03.036

A series of 3-aryl-naphtho[2,3-d]isoxazole-4,9-diones and some of their 6-aza analogues were synthesized and found to inhibit the heat shock protein 90 (Hsp90). The compounds were tested for their binding to Hsp90 and for their effects on Hsp90 client pro

Full text of DOI:10.1016/j.ejmech.2012.03.036

European Journal of Medicinal Chemistry 53 (2012) 64e75
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Isoxazolo(aza)naphthoquinones: A new class of cytotoxic Hsp90 inhibitors
,
a
b
b
Alberto Bargiotti a, Loana Musso a, Sabrina Dallavalle a , Lucio Merlini , Grazia Gallo , Andrea Ciacci ,
Giuseppe Giannini b, Walter Cabri b, Sergio Penco b, Loredana Vesci b, Massimo Castorina b,
Ferdinando Maria Milazzo b, Maria Luisa Cervoni b, Marcella Barbarino b, Claudio Pisano b,
Chiara Giommarelli c, Valentina Zuco c, Michelandrea De Cesare c, Franco Zunino c
*
a Dipartimento di Scienze Molecolari Agroalimentari, Università di Milano, Via Celoria 2, 20133 Milano, Italy
b Sigma-tau, R&D, Via Pontina km 30400, 00040 Pomezia, Italy
c Unità di Chemioterapia e Farmacologia Antitumorale Preclinica, Dipartimento di Oncologia Sperimentale e Laboratori, Fondazione IRCCS Istituto Nazionale dei Tumori,
Via Venezian 1, 20133 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 3-aryl-naphtho[2,3ed]isoxazole-4,9-diones and some of their 6-aza analogues were
Received 2 December 2011
Received in revised form
28 January 2012
Accepted 20 March 2012
Available online 28 March 2012
synthesized and found to inhibit the heat shock protein 90 (Hsp90). The compounds were tested for their
binding to Hsp90 and for their effects on Hsp90 client proteins expression in a series of human tumour
cell lines. Representative compounds (7f, 10c) downregulated the Hsp90 client proteins EGFR, Akt, Cdk4,
Raf-1, and survivin, and upregulated Hsp70. Most of the compounds, in particular the alkylated 3-pyridyl
derivatives, exhibited potent antiproliferative activity, down to two-digit nanomolar range. Preliminary
results indicated in vivo activity of 7f against human epithelial carcinoma A431 model growing as tumour
xenograft in nude mice, thus supporting the therapeutic potential of this novel series of Hsp90 inhibitors.
Ó 2012 Elsevier Masson SAS. All rights reserved.
Dedicated to the memory of Dr. Paolo
Carminati.
Keywords:
Synthesis
Isoxazolonaphthoquinones
Hsp90 inhibitors
Antitumour
Antiproliferative activity
QSAR
1. Introduction
genetic lesions that are present within tumours, Hsp90 allows
mutant proteins to retain or even gain function, while permitting
The interest for the heat shock protein 90 (Hsp90) as a thera-
peutic target is related to its central role in correct folding and
stabilization of proteins involved in malignant behaviour and
tumour progression [1]. Multiple signal transduction pathways
implicated in the regulation of cell proliferation and survival are
dependent on Hsp90 [2]. Several Hsp90 client proteins are involved
in critical processes including cell-cycle regulation and apoptosis
[3]. The heat shock proteins are often overexpressed in tumour
cells, and this supports their ability to survive under unfavourable
stress conditions (e.g. hypoxia and acidosis). The essential chaper-
oning function of Hsp90 is subverted during oncogenesis to make
malignant transformation possible and to facilitate rapid somatic
evolution [3]. Functioning as a biochemical buffer for the numerous
cancer cells to tolerate the imbalanced signalling that such onco-
proteins create. Thus, targeting Hsp90 may have the potential
advantage of simultaneously blocking multiple oncogenic path-
ways [4].
Hsp90 exists as a homodimer made up of three domains [1,2].
The N-terminal domain contains an ATP-binding site that binds the
natural products geldanamycin and radicicol, and the analogue 17-
AAG (Chart 1). The middle domain is highly charged and has high
affinity for cochaperones and client proteins. A second ATP-binding
site is located in the C-terminus of Hsp90. This C-terminal nucle-
otide binding pocket has been shown to bind not only ATP, but
cisplatin, novobiocin, epigallocatechin-3-gallate (EGCG) and
taxol [5].
A number of highly specific Hsp90 inhibitors have been identi-
fied [6]. They redirect Hsp90 chaperoning activity and decrease
cellular levels of its numerous cancer-related client proteins [2].
Such inhibitors exhibit promising antitumour activity as single
* Corresponding author. Tel.: þ39 (2) 50316818; fax: þ39 (2) 50316801.
E-mail address: sabrina.dallavalle@unimi.it (S. Dallavalle).
0223-5234/$ e see front matter Ó 2012 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2012.03.036

A. Bargiotti et al. / European Journal of Medicinal Chemistry 53 (2012) 64e75
65
O
O
O
H
O
O
H
N
OMe
N
O
O
O
N
N
H
N
H
N
H
O
HO
OMe
OMe
OMe
OMe
OMe
OMe
O
O
O
NH2
NH2
NH2
O
O
O
Geldanamycin (GDA)
17-AAG
17-DMAG
OH
OH
Cl
OH
OH
O
N
HO
O
N
N
H
N
H
N
O
O
N
OH
N
O
H
N
N
O
O
O
N
NH
O
O
Sigma-tau compound
Wyeth compound
NVP-AUY922
Chart 1.
agents or in combination with other cytotoxic agents [7e9], and
some of them are in clinical development [10,11]. Geldanamycin,
the first Hsp90 inhibitor discovered, and its derivatives 17-AAG
(tanespimycin) and 17-DMAG (17-dimethylamino-ethylamino-17-
demethoxygeldanamycin, alvespimycin), that have entered clin-
ical trials, are ansa compounds containing a key aminoquinone
moiety [12]. (Chart 1).
Quinones are a class of organic compounds endowed with
a variety of biological activities, mostly connected with their redox
properties. A number of natural and synthetic quinones show
remarkable anticancer activity [12], and a series of synthetic
1,4-naphtoquinones have recently been identified as Hsp90 inhib-
itors [13].
Moreover, among the large number of Hsp90 inhibitors, some
compounds, containing the isoxazole nucleus (Chart 1), have
shown potent and selective inhibition of this molecular chaperone
[14]. The presence of the heterocyclic nucleus seems to exert a role
in the docking of the compounds to the ATP-binding site of
Hsp90 [15].
2 with triethylamine and aqueous NaClO in dichloromethane [19]
(Schemes 1 and 2).
Some compounds (3c,f,h,i) were prepared by demethylation of
the corresponding ethers (3b,e,g). Compounds 6aee were obtained
regioselectively [20] by 1,3-dipolar cycloaddition on the
R'
R
R
R'
O
O
R''
a
+
N
N
R''
O
HO
O
O
2a-b, d-e, g,k
3a-k
1
a R = H R' = H
b R = H R' = OMe
c R = H R' = OH
d R = H R' = F
e R = H R' = H
f R = H R' = H
g R = Cl R' = OMe
h R = Cl R' = OMe
R'' = H
This paper reports the synthesis and the assessment of the
R'' = H
cytotoxic activity of
a series of new isoxazolo-fused naph-
b
c
R'' = H
R'' = H
thoquinones and isoquinolinoquinones, that exert their antitumour
activity via inhibition of the molecular chaperone Hsp90.
R'' = OMe
R'' = OH
R'' = OMe
R'' = OH
R'' = OH
R'' = H
2. Chemistry
d
e
The 1,3-dipolar cycloaddition of benzonitriloxides to quinones is
a welleknown reaction [16] that has been exploited for the
synthesis of simple and substituted isoxazolo-fused quinones [17]
and naphthoquinones [18]. Thus, the first series of compounds
(3aep) was prepared reacting the naphthoquinone 1 with the
nitrile oxides obtained in situ by treating the corresponding oximes
i R = Cl R' = OH
O
N
k R = H R' =
O
HCl
Scheme 1. Reagents and conditions: a) TEA, 15% NaClO, CH2Cl2, 0 ꢀC, 1.5 h; b) HBr 35%,
AcOH, reflux, 5 h; c) BBr3, CH2Cl2, 0 ꢀC, 1 h, then rt, 30 min; d) BBr3, CH2Cl2, 0 ꢀC, 1.5 h,
then rt, 1 h; e) HBr 35%, AcOH, reflux, 48 h.

66
A. Bargiotti et al. / European Journal of Medicinal Chemistry 53 (2012) 64e75
O
O
R
O
O
R
R'
R
a
+
N
N
O
N
O
HO
O
7a-c
O
1
2l-p
3l-p
6c-e
a
H
N
3n-p
7d-i
l
R =
Br-
Ph
N+ CH2 Ph
a R = OMe R' =
b R = OAc R' =
m
n
o
p
R =
R =
R =
R =
Br-
N+ CH2 Ph
N
N
Br-
N+ CH2 Ph
c R = OH
d R = H
e R = H
f R = H
g R = H
h R = H
i R = H
R' =
R' =
R' =
R' =
R' =
R' =
R' =
I-
N+ CH3
Br-
N+ CH2 2-Pyraz
N
Br-
Scheme 2. Reagents and conditions: a) TEA, 15% NaClO, CH2Cl2, 0 ꢀC, 1.5 h.
N+ CH2Ph
Br-
N+ CH2 Ph
bromonaphthoquinones 5 [21], followed in the case of 6b and 6e by
the hydrolysis of the corresponding acetate (6a, 6d) (Scheme 3).
The pyridino derivatives 3nep and 6cee were alkylated with
benzyl bromide or 2-bromomethylpyrazine in DMSO, or by iodo-
methane in methanol, to obtain the quaternary salts 7aei
(Scheme 4).
Reduction of the naphthoquinones to the corresponding
hydroquinones gave unstable compounds, so that two of them (3b,
d) were transformed into the hydroquinone diacetates (8aeb) by
reaction with Zn and acetic acid (Scheme 5).
N+ CH2 Ph
Br-
N+ CH2 2-Pyraz
Br-
Scheme 4. Reagents and conditions: a) For 7aec,feh: benzyl bromide, DMSO, rt,
overnight; for 7d: MeI, MeOH, rt, overnight; for 7e,i : 2-bromomethylpyrazine, DMSO,
rt, 72 h.
As the isoquinolinone nucleus is present in some natural (e.g.
calothryxin [22] and cribrostatin 6 [23]) and synthetic [24] cyto-
toxic compounds, we planned to introduce this moiety in place of
the naphthoquinone nucleus. For this purpose 6-bromo-2-
hydroxyquinoline [25], followed by oxidation of the resulting
dibromoderivative [26]. The 1,3 dipolar cycloaddition was carried
out following the conditions previously described for the synthesis
of the isoxazolonaphthoquinones (Scheme 6). The reaction
azanaphthoquinone
9 was prepared by bromination of 5-
OR
O
OR
O
OH
R'
a,b,c
d
N
O
R'
Br
O
N
O
5a,c
OH
2d,n
HO
6a-e
4
F
F
a
b
c
d
e
R = Ac R' =
R = H R' =
e
N
R = Me R' =
R = Ac R' =
N
N
e
R = H
R' =
Scheme 3. Reagents and conditions: a) Ac2O, Py, rt, 3 h; b) NBS, AcOH/H2O, 50 ꢀC, 2 h; c) for 5c: i) H2SO4 1.5 M, EtOH, reflux, 3 h; ii) MeI, Ag2O, CH2Cl2, rt, overnight; d) TEA, 15%
NaClO, CH2Cl2, 0 ꢀC, 1.5 h; e) K2CO3, H2O/MeOH, rt, 40 min.

A. Bargiotti et al. / European Journal of Medicinal Chemistry 53 (2012) 64e75
67
Table 1
R
R
Binding affinity to Hsp90 and antiproliferative activity for the prepared compounds.
Cpd. Binding Hsp90 NCI-H460
STO IC50ꢁSD
A 431 IC50ꢁSD
[mM] 72 h
(FP) IC50ꢁSD
M]
IC50ꢁSD [
m
M]
[m
M] 72 h
O
O
[m
72 h
OAc
OAc
17-AAG 1.095 ꢁ 0.05
0.010 ꢁ 0.002
0.12 ꢁ 0.03
0.65 ꢁ 0.03
0.51 ꢁ 0.004
0.35 ꢁ 0.02
0.29 ꢁ 0.001
>1
e
0.069 ꢁ 0.005
0.12 ꢁ 0.008
0.35 ꢁ 0.03
0.21 ꢁ 0.01
0.26 ꢁ 0.011
0.13 ꢁ 0.01
3.4 ꢁ 0.7
a
3a
3b
3c
7.85 ꢁ 0.04
5.84 ꢁ 0.04
>100
0.02 ꢁ 0.004
0.16 ꢁ 0.04
0.46 ꢁ 0.17
0.02 ꢁ 0.02
0.66 ꢁ 0.04
3.4 ꢁ 0.1
N
N
O
O
3d
3e
3f
3g
3h
3i
>100
>100
1.55 ꢁ 0.024
>100
0.88 ꢁ 0.07
0.135 ꢁ 0.06
0.30 ꢁ 0.08
1.0 ꢁ 0.1
1.4 ꢁ 0.3
8a R = OCH3
8b R = F
3b R = OCH3
3d R = F
>100
>1
2.98 ꢁ 0.40
23.0 ꢁ 7.2
4.2 ꢁ 0.01
1.2 ꢁ 0.1
>100
>1
3k
3l
0.866 ꢁ 0.031
0.30 ꢁ 0.01
>1 (40%)
0.14 ꢁ 0.02
0.062 ꢁ 0.003
1.0 ꢁ 0.05
0.078 ꢁ 0.005
0.124 ꢁ 0.009
0.88 ꢁ 0.07
1.13 ꢁ 0.1
0.44 ꢁ 0.21
0.32 ꢁ 0.08
0.32 ꢁ 0.12
0.056 ꢁ 0.005
0.043 ꢁ 0.001
0.34 ꢁ 0.09
0.27 ꢁ 0.01
0.29 ꢁ 0.22
0.33 ꢁ 0.03
0.06 ꢁ 0.01
0.068 ꢁ 0.010
0.168 ꢁ 0.01
3.96 ꢁ 0.6
0.32 ꢁ 0.04
0.12 ꢁ 0.1
Scheme 5. Reagents and conditions: a) Zn powder, TEA, acetic anhydride, reflux, 3 h.
3m
3n
3o
3p
6a
6b
6c
6d
6e
7a
7b
7c
7d
7e
7f
7g
7h
7i
>100
0.39 ꢁ 0.01
0.54 ꢁ 0.009
1.06 ꢁ 0.01
1.94 ꢁ 0.01
0.29 ꢁ 0.01
1.9 ꢁ 0.1
0.011 ꢁ 0.04
0.04 ꢁ 0.02
0.02 ꢁ 0.01
1.05 ꢁ 0.68
0.61 ꢁ 0.05
0.261 ꢁ 0.060
0.150 ꢁ 0.080
0.103 ꢁ 0.120
0.168 ꢁ 0.090
3.96 ꢁ 0.20
21.56 ꢁ 0.90
1.20 ꢁ 0.05
0.22 ꢁ 0.05
appeared regioselective, giving only the isomers with structure
10aed, as confirmed by NMR spectra.
0.084 ꢁ 0.001 0.15 ꢁ 0.04
>100
0.88 ꢁ 0.07
0.18 ꢁ 0.01
3. Results and discussion
0.34 ꢁ 0.01
0.031 ꢁ 0.001 >1 (32%)
0.098 ꢁ 0.001 0.88 ꢁ 0.07
The prepared compounds were tested in a binding assay for
their affinity to Hsp90 and for their antiproliferative activity on
a number of human tumour cell lines (non small-cell lung cancer
NCI-H460, squamous-cell carcinoma A431, peritoneal mesothe-
lioma STO) (Table 1).
In general, some of the compounds tested showed strong anti-
proliferative activity, comparable to or even higher than the activity
of 17-AAG, especially on the carcinoma cells A431. Among the
naphthoisoxazolediones, a number of them showed strong binding
to Hsp90 (3n,o, 6b,d, 7aef,h,i), with some correlation with the
antiproliferative activity. The presence of a substituent in the ortho
position on the phenyl ring had a deleterious effect (3fei), possibly
due to steric hindrance. The pyridine derivatives (3nep, 6ced)
showed a better binding and stronger activity than the correspond-
ingphenylderivatives(3a,b,6a,b), althoughtheintroductionofanOH
group on the naphthoquinone ring has opposite effects in the two
series (6e vs. 6b).
21.6 ꢁ 0.7
0.63 ꢁ 0.01
0.68 ꢁ 0.01
0.33 ꢁ 0.01
0.48 ꢁ 0.06
0.052 ꢁ 0.001
0.017 ꢁ 0.001
0.034 ꢁ 0.002 0.018 ꢁ 0.0007 0.0022 ꢁ 0.0010 0.0016 ꢁ 0.0010
>100
>1
2.24 ꢁ 0.80
0.01 ꢁ 0.06
0.11 ꢁ 0.05
1.00 ꢁ 0.04
0.67 ꢁ 0.21
0.004 ꢁ 0.001
0.092 ꢁ 0.022
0.054 ꢁ 0.01
2.24 ꢁ 0.07
0.020 ꢁ 0.0009
0.042 ꢁ 0.002
1.30 ꢁ 0.03
0.77 ꢁ 0.009
0.51 ꢁ 0.01
>100
0.021 ꢁ 0.001
0.046 ꢁ 0.004
>1 (39%)
8a
8b
10a
10b
10c
10d
>100
>1
0.69 ꢁ 0.22
5.6 ꢁ 0.1
1.6 ꢁ 0.1
<0.005a
0.12 ꢁ 0.01
0.183 ꢁ 0.01
0.446 ꢁ 0.042
0.041 ꢁ 0.004
0.085 ꢁ 0.0009
0.004 ꢁ 0.002
0.098 ꢁ 0.036
0.02 ꢁ 0.01
0.06 ꢁ 0.018 0.0085 ꢁ 0.0033
a
For the detection limits of the assay, see Chiosis et al. [27].
To increase the polarity and possibly the water solubility, the
pyridine derivatives 3nep and 6cee were alkylated with benzyl or
methyl or 2-pyrazinylmethyl halides, to give the corresponding
quaternary salts 7aei. This modification led to a general increase in
the binding affinity, especially for 7b,c,f, all exhibiting sub-
micromolar values. With the exception of 7b,7c and of 7g (where
again the alkylation on the ortho position was detrimental), these
compounds were characterized by
activity with nanomolar IC50 values.
The most potent compounds of the whole series, 7f and 10c,
were tested for their ability to downregulate the expression of
Hsp90 client proteins in the squamous cell carcinoma A431 and
the mesothelioma STO cells (Fig. 1). After 24eh exposure to
The two hydroquinone acetates 8a,b showed a very weak
binding affinity, and low antiproliferative activity.
Change of the naphthoquinone to an isoquinolinone nucleus led
to an improvement both of binding affinity and of antiproliferative
activity (10aed) with respect to the corresponding naph-
thoquinone derivatives (see for ex. 10c vs 3k). Compound 10c, that
contains a polar, hydrophilic morpholino group, was the most
active compound of the series in terms of Hsp90 binding. Relevant
to this point is the observation that the morpholine moiety is
present in all the most active isoxazole-containing compounds
reported in the literature [14] (see chart 1).
a potent antiproliferative
R
X
R
O
O
X
O
N
a
a X = C R = F
N
+
N
b X = C R = OCH3
N
Br
O
HO
c X = C R =
O
N
d X = N
HCl
O
O
9
10a-d
2b,d,k,n
Scheme 6. Reagents and conditions: a) TEA, 15% NaClO, CH2Cl2, 0 ꢀC, 1.5 h.

68
A. Bargiotti et al. / European Journal of Medicinal Chemistry 53 (2012) 64e75
Fig. 1. Analysis of Hsp90 client protein levels in A431 and STO cells after 24 h of
treatment with 7f. EGFR, survivin, Cdk4, Akt and Raf-1 levels. Total cellular extracts
were obtained 24 h after cell treatment (7f, 22 nM, IC80). Actin is shown as a control for
protein loading. Lane 1: control; lane 2: 7f.
a cytotoxic concentration of 7f, Western blot analysis indicated
depletion of Cdk4, Raf and survivin in both cell lines. Moreover,
the treatment with 7f caused an almost complete depletion of
EGFR protein in A431 cells, which exhibited overexpression of the
receptor.
Fig. 3. Analysis of Hsp70 expression in A431 tumour cells treated with compounds 7f
(a) and 10c (b). Western blot analysis was performed on total cellular extracts after
24 h of treatment. Results of densitometry analysis were reported as normalized (to
actin) ratios.
b-
biochemical analysis of Hsp90 client expression modulation is
consistent with this interpretation.
Analogously, 10c caused
a dramatic and dose-dependent
depletion of Akt, CDK-4, survivin and EGFR in A431 cells (Fig. 2).
Thus, the modulation of Hsp90 client proteins was consistent with
inhibition of Hsp90 function. Moreover, Fig. 3 reports Hsp70
protein expression modulation in A431 cells following exposure to
7f or of 10c. This dose-dependent upregulation of Hsp70 cocha-
perone is consistent with the target Hsp90 modulation [28].
The observation that compounds which exhibited high binding
affinity (e.g. 7f, 6d, 10c) are also potent as cytotoxic agents supports
that Hsp90 is a cellular target implicated in the antiproliferative/
antitumour effects of some compounds of this series. The
However, some compounds, although strongly binding the
enzyme, show low antiproliferative activity. In this case the
chemico-physical features of the compounds may influence the
cellular pharmacokinetic behaviour and therefore cellular response
(compare 7b with 7f).
On the contrary, the appreciable cell growth inhibition by
compounds (e.g. 3cee, 6e) which did not exhibit a strong Hsp90
interaction suggests that other off-target effects are likely involved
in the mechanism of action of compounds of this class.
By virtue of their quinone moiety, the benzoquinone ansamycin
class of Hsp90 inhibitors can undergo bioreduction to hydroqui-
none forms by action of NAD(P)H:quinone oxidoreductase 1
(NQO1), a FAD-dependent direct two-electron reductase that can
use either NADH or NADPH as reducing cofactor and reduces
quinones directly to hydroquinones [29]. Relatively high levels of
NQO1 protein have been detected in many human solid tumours,
including lung, breast, colon, ovary, and pancreas [30]. Hydroqui-
none forms of 17AAG and 17DMAG are relatively stable, but they
have been shown to be sensitive to copper-mediated reoxidation
[31]. Hydroquinone forms of 17-AAG and 17-DMAG have also been
shown to be more potent inhibitors of Hsp90 in in vitro studies
using purified Hsp90 compared with their parent quinones [32,33].
It appears thus established that NQO1 acts as an important factor to
modulate the Hsp90 inhibitory activity of anticancer quinones
[1,33]. Other data also illustrate the close relationship between
Hsp90 regulation and cellular redox homeostasis, pointing to the
role of oxidative stress, such as induced by quinones, in the
disruption of the Hsp90 signalling complex [34]. Therefore it
Fig. 2. Analysis of Hsp90 client protein levels in A431 cells after treatment with
different concentrations of 10c. Actin is shown as a control for protein loading.

A. Bargiotti et al. / European Journal of Medicinal Chemistry 53 (2012) 64e75
69
cannot be excluded that also our compounds may modulate Hsp90
function via NQO1.
On the basis of the cellular/biochemical profile of activity, the
therapeuticeffectof7fwasdeterminedintheA431modelgrowingas
tumour xenograft in nude mice. Using oral administration, daily
treatment for 3 weeks with 7f produced an appreciable inhibition of
tumour growth, i.e. 50e65%in two independentexperiments(Fig. 4).
4. Molecular modelling and QSAR study
Docking studies were performed using three different confor-
mations/crystal structures of Hsp90 (PDB code: 1UY6 for helical
conformation; 2BYH for open conformation and 2VCI for closed
conformation). The best solutions were found using Hsp90 helical
conformation.
Thus, we compared the orientation of our most active
compounds in the Hsp90 active site with the orientation of
SNX2112 [35], a compound in clinical development, co-crystallized
in the helical conformation. Fig. 5 shows the superimposition of 7f
and SNX2112 where the common interactions are represented by
quinone group vs. carbonyl function of the amide group and by the
heteroaromatic substituents of the two structures filling the
internal pocket behind the helix.
Additional interaction for 7f is found between the isoxazole
group and a conserved water molecule, while the interaction with
Asp93 is not retrieved due to the absence of H-bond donor groups.
Comparison with 17-DMAG and NVP-AUY922 is shown in the SI.
An attempt to correlate the structural relationships of the
compounds described above with the binding to Hsp90 was made
Fig. 5. Docking pose of compound 7f (green) and SNX2112 (orange) in Hsp90 active
site (helical conformation). H-bonds are shown as dotted lines. (For interpretation of
the references to colour in this figure legend, the reader is referred to the web version
of this article.)
A comprehensive conformational search was generated for each
of the 27 molecules, sampling the conformational space that may
represent possible active structures. Conformers beyond 8.3 kJ/mol
(2 kcal/mol) of the global minimum were automatically discarded.
Visual inspection and careful analysis of chemical features were
conducted to guide the pharmacophore development process.
Particular attention was paid to compound 10c, as it represents
the most active derivative. Five-membered and six-membered
heterocyclic rings were replaced with a single, custom group
feature (X) representing this functional group. Pharmacophore
variants considered for Hsp90 inhibition were composed of four
chemical features: hydrogen-bond acceptor (A, pink sphere),
aromatic ring (R, ring), hydrophobic (H, green sphere), and
heterocyclic ring (X, blue sphere in order to avoid that aromatic
using
pharmacophore modelling and
3D quantitative
structureeactivity relationships (3D-QSAR). Then, in this study, the
generation of a pharmacophore for the Hsp90 binding allowed us to
determine the molecular components required for activity and 3D-
QSAR allowed us to predict Hsp90 activity values.
Table 2 shows the compounds (27) tested for the generation of
the three-dimensional (3D) pharmacophore model in order to
rationalize chemical structure with the observed binding affinities
of the 4,9-dioxo-4,9-dihydronaphtho[2,3eb]isoxazoles. Compounds
with pIC50 FP > 7.3 (an arbitrarily chosen activity threshold of 50 nM)
were assigned to the active class (3 compounds) and used in the
development of the pharmacophore.
Table 2
Binding to Hsp90 (Hsp90 (FP) IC50
selected compounds used for QSAR analysis.
, mM), pIC50 (FP), predicted pIC50 (FP) and set of
Cpd.
Binding Hsp90
(FP) IC50[mM]
pIC50 (FP)
Pred. pIC50
QSAR set
3a
3d
3f
7.85
200
1.55
4.2
5.11
3.7
4.87
4.87
5.29
5.09
5.85
6.30
5.73
6.02
6.47
7.11
6.23
7.38
3.8
training
training
training
training
training
training
training
training
training
training
training
training
training
training
training
training
training
training
test
test
test
test
test
test
test
test
test
5.81
5.38
5.92
6.41
5.97
6.54
5.72
7.08
6.2
3i
3k
3n
3p
6b
6c
6d
7d
7f
1.2
0.39
1.06
0.29
1.9
0.084
0.63
0.034
200
7.47
3.7
7g
7i
0.51
5.6
6.29
5.25
5.61
8.47
6.92
5.23
3.7
5.71
6.27
6.47
7.51
7.01
6.17
6.11
6.25
5.7
10a
10b
10c
10d
3b
3e
3l
3o
7a
7b
7c
2.45
<0.005
0.12
5.84
200
1.94
0.54
0.34
0.031
0.098
0.68
0.77
5.78
8.69
6.36
5.49
5.13
5.76
5.3
7.28
7.51
6.64
6.98
6.56
Fig. 4. Antitumor activity of 7f against the squamous cell carcinoma A431 xenograft in
nude mice. The compound, 7f, dissolved in physiologic saline was administered p.o.
(10 mg/kg) for four consecutive days/week (C). The treatment started when the
tumours were just palpable, was repeated for three weeks (i.e., qdx4d/week x 3
weeks). (B), untreated controls.
7e
7h

70
A. Bargiotti et al. / European Journal of Medicinal Chemistry 53 (2012) 64e75
rings were misplaced) with at least six sites. Common pharmaco-
phores exhibiting AAHRRX were searched in all actives with a final
box size and minimum intersite distance of 1.0 Å. Two hypotheses
were generated, and that with the best score (Survival ¼ 3.163
based on the PHASE scoring functions [36,37]) was chosen as the
final Hsp90 pharmacophore for use in subsequent 3D-QSAR
investigation. The identified variant assigned a single site, X, to the
isoxazole ring; the hydrogen bond acceptor sites (A), corresponded
to the 4,9-dioxo groups; aromatic features (R) was placed on the
naphthoquinone aromatic ring and on the first ring of the isoxazole
substituent; finally, a hydrophobic feature (H) corresponded to
aromatic or alkyl group at the end of substituent. While not
essential for activity, molecules lacking hydrophobic features
generally show a decrease in pIC50. Apart from correctly identifying
the active compounds, the pharmacophore was used to prealign
the molecules for QSAR analysis.
Attempts to statistically analyse three-dimensional molecular
fields by means of a QSAR analysis were made: molecules reported
in Table 2 were placed into training and test sets once aligned onto
the pharmacophore (see Fig. 6) with an effort to minimize struc-
tural redundancy while maximizing coverage of the experimental
activities. The training and test sets were composed of 18 and 9
molecules respectively. Atom based QSAR model was built using
a grid spacing of 1.0 Å, two partial least-squares (PLS) factor and
eliminating variables with jt-valuej < 2.0.
The performance of the two PLS factors atom-based QSAR model
on the training and test set molecules is illustrated in Fig. 7. The
scatter plot indicates good correlation coefficients (Stability ¼ 0.64;
R2 ¼ 0.81 and Q2 ¼ 0.64), indicative of a model with strong
predictive power and significance.
Table 2 also compares experimental and predicted pIC50 values
for both the training and test set molecules, showing that activity
was effectively predicted. This observation further supports the
validity of the pharmacophore model, suggesting that the spatial
arrangement of chemical features, when aligned by the pharma-
cophore, is indicative of the probable active conformation of the
molecule. Although good predictive power is evident, the low
number of training/test set compounds warrants caution when
using this model for this purpose.
Fig. 8 provides a visual representation of the atom-based 3D-
QSAR model (global, hydrophobic/non polar, positive ionic, and
electronegative) represented by colour codes.
The global regions of favourable and unfavourable interactions
are represented in Fig. 8a. Hydrophobic contributions are shown in
Fig. 8b; the map showed a big green coloured region indicating that
an increase in the hydrophobicity in this region is expected to
improve the activity of the dioxodihydronaphtho[2,3eb]isoxazole
Fig. 7. Scatter plot of predicted vs actual Hsp90 binding activity applied to 27
compounds.
derivatives (see for example compound 3n vs 7f). A purple col-
oured region corresponding to the ortho-substituent of 3-phenyl
ring disfavours the placement of hydrophobic groups (see for
example compound 7g vs 7f,7h). Positive ionic contributions are
shown in Fig. 7c; the map showed that the 4-pyridinium group is
favoured vs. 2-pyridinium one (see for example compound 7g vs 7f).
Electronegative contributions are shown in Fig. 7d. The map
describes the spatial arrangement of favourable electron-
withdrawing atoms: the atom 6 (see compound 3n vs 10d) and C5
position (see compound 3n vs. 6d) of dihydronaphtho[2,3eb]
isoxazole, the para position of the aromatic ring in C3 (see
compound 3a vs 3n) and the region occupied by morpholine oxygen
atom in the most active molecule (see compound 7f vs 10c).
Fig. 8. Visual representation of atom-based PHASE QSAR model: a) Global model
(colour code: blue, positive; red, negative); b) Hydrophobic/non-polar contributions
(colour code: green, positive; purple, negative); c) Positive ionic (colour code: blue,
positive; orange-yellow, negative); d) Electronegative (colour code: pale red, positive;
light green, negative). (For interpretation of the references to colour in this figure
legend, the reader is referred to the web version of this article.)
Fig. 6. Pharmacophore for Hsp90 inhibition with overlay of the whole set of
compounds.

A. Bargiotti et al. / European Journal of Medicinal Chemistry 53 (2012) 64e75
71
5. Conclusions
organic phase was washed with sodium bicarbonate, water and
dried over Na2SO4. The solvent was evaporated under reduce
pressure to give a crude that was purified by flash chromatography
(CH2Cl2/CH3OH, 97.5:2.5) to obtain 80 mg of 3c as an orange sticky
The synthesis, inhibition of Hsp90 and assessment of the cyto-
toxicity of a series of new isoxazolo-fused naphthoquinones and
isoquinolinoquinones was carried out.
solid (83%). 1H NMR (DMSO-d6)
d: 10.15 (brs, 1H), 8.25e8.12 (m,
The available evidence based on the biochemical and cellular
effects indicates that the compounds described in this paper repre-
sent a novel series of Hsp90 inhibitors. Specific variations of the
structure improved the Hsp90 inhibitory effect. In particular,
compounds containing an alkylated pyridinium moieties showed
high affinityand strong antiproliferative effects. The invivo activity of
a representative compound (7f) against human epithelial carcinoma
A431 model, growing astumourxenograft innudemice, supportsthe
therapeutic potential of this novel series of Hsp90 inhibitors.
2H), 8.05e7.89 (m, 4H), 7.00e6.90 (m, 2H). Anal calcd for C17H9NO4
(291.26): C 70.10, H 3.11, N 4.81. Found: C 70.35, H 3.01, N 4.59.
6.2.4. 3-(4-Fluorophenyl)-naphtho[2,3ed]isoxazole-4,9-dione (3d)
Purified by flash chromatography (Eluent CH2Cl2). Pale yellow
solid (31%). Mp.: 166e167 ꢀC. 1H NMR (DMSO-d6)
d: 8.27e8.08 (m,
4H), 8.02e7.83 (m, 2H), 7.56e7.35 (m, 2H). Anal calcd for
C17H8FNO3 (293.25): C 69.63, H 2.75, N 4.78. Found: C 69.32, H 2.98,
N 4.55.
6. Experimental
6.2.5. 3-(2-Methoxyphenyl)-naphtho[2,3ed]isoxazole-4,9-dione
(3e)
6.1. General
Purified by flash chromatography (Eluent CH2Cl2). Bright yellow
solid (57%). Mp.: 212e214 ꢀC. 1H NMR (DMSO-d6)
d: 8.23e8.14 (m,
All reagents and solvents were reagent grade or were purified by
standard methods before use. Melting points were determined in
open capillaries. NMR spectra were recorded at 300 MHz. Solvents
were routinely distilled prior to use; anhydrous tetrahydrofuran
(THF) and ether (Et2O) were obtained by distillation from
sodiumebenzophenone ketyl; dry dichloromethane was obtained
by distillation from phosphorus pentoxide. All reactions requiring
anhydrous conditions were performed under a positive nitrogen
flow, and all glassware were oven dried and/or flame dried. Isola-
tion and purification of the compounds were performed by flash
column chromatography on silica gel 60 (230e400 mesh).
Analytical thin-layer chromatography (TLC) was conducted on
TLC plates (silica gel 60 F254, aluminium foil) visualized by UV light.
1H), 8.13e8.03 (m, 1H), 7.99e7.90 (m, 2H), 7.66e7.56 (m, 1H), 7.48
(d, 1H, J ¼ 8.0 Hz), 7.26 (d, 1H, J ¼ 8.4 Hz), 7.16e7.07 (m, 1H), 3.76 (s,
3H). Anal calcd for C18H11NO4 (305.28): C 70.82, H 3.63, N 4.59.
Found: C 70.66, H 3.91, N 4.39.
6.2.6. 3-(2-Hydroxyphenyl)-naphtho[2,3ed]isoxazole-4,9-dione
(3f)
Compound 3e (50 mg, 0.16 mmol) was dissolved in 4 mL of dry
dichloromethane at 0 ꢀC and 1 M boron tribromide solution in
dichloromethane (200 mL) was added under nitrogen. The mixture
was stirred at 0 ꢀC for 1.5 h then at room temperature for 30 min.
The solution was diluted with dichloromethane, washed with
sodium bicarbonate, dried over Na2SO4 and evaporated under
reduced pressure. The residue was purified by flash chromatog-
raphy (CH2Cl2) to give 45 mg of 3f as a yellow solid (93%). Mp.:
6.2. General procedure for the cycloaddition reaction
190e191 ꢀC. 1H NMR (CDCl3)
1.9 Hz), 8.37e8.30 (m, 2H), 7.96e7.85 (m, 2H), 7.53e7.46 (m, 1H),
7.18e7.11 (m, 2H). Anal calcd for C17H9NO4 (291.26): C 70.10, H 3.11,
N 4.81. Found: C 70.29, H 2.98, N 4.67.
d
: 9.19 (s, 1H), 8.74 (dd, 1H, J ¼ 8.4,
The appropriate quinone (10.5 mmol), triethylamine (100 mg,
0.98 mmol) and 15% aqueous sodium hypochlorite (15 mL) were
added to dichloromethane (10 mL), then the appropriate oxime
(10.5 mmol) was added dropwise at 0 ꢀC over 15 min. After stirring
for 1.5 h, the reaction phases were separated and the aqueous phase
was extracted with dichloromethane. The organic phase was dried
with Na2SO4 and the solvent was removed under reduced pressure.
The residue was purified by flash chromatography to give the
desired product.
6.2.7. 3-(5-Chloro-2,4-dimethoxyphenyl)-naphtho[2,3ed]
isoxazole-4,9-dione (3g)
Purified by flash chromatography (Eluent hexane/AcOEt 3:2).
Dark yellow sticky solid (75%). 1H NMR (CDCl3)
d: 8.34e8.25 (m,
1H), 8.24e8.12 (m, 1H), 7.92e7.73 (m, 2H), 7.56 (s, 1H), 6.66 (s, 1H),
4.03 (s, 3H), 3.85 (s, 3H). Anal calcd for C19H12ClNO5 (369.76): C
61.72, H, 3.27, N 3.79. Found: C 61.94, H, 3.08, N 3.85.
Thus, when not stated otherwise, were prepared the following
products:
6.2.1. 3-Phenyl-naphtho[2,3ed]isoxazole-4,9-dione (3a)
6.2.8. 3-(5-Chloro-2-hydroxy-4-methoxyphenyl)-naphtho[2,3ed]
isoxazole-4,9-dione (3h)
Purified by flash chromatography. (Eluent CH2Cl2). Sticky pale
yellow solid. (55%). 1H NMR (DMSO-d6)
d: 8.27e8.14 (m, 2H),
Compound 3g (100 mg, 0.27 mmol) was dissolved in 8 mL of dry
dichloromethane at 0 ꢀC and 1 M boron tribromide in dichloro-
methane (0.5 mL) was added under nitrogen. The mixture was
stirred at 0 ꢀC for 1.5 h then at room temperature for 1 h. The
solution was diluted with dichloromethane, washed with sodium
bicarbonate, dried over Na2SO4 and evaporated under reduced
pressure to give 90 mg of 3h as a yellow solid (94%). Mp.:
8.12e8.03 (m, 2H), 7.98e7.88 (m, 2H), 7.69e7.52 (m, 3H). Anal calcd
for C17H9NO3 (275.26): C 74.18, H 3.30, N 5.09. Found: C 74.39, H
3.12, N 5.19.
6.2.2. 3-(4-Methoxyphenyl)-naphtho[2,3ed]isoxazole-4,9-dione (3b)
Purified by flash chromatography (Eluent hexane/AcOEt 3:2).
Brown solid. (60%). Mp.: 163e164 ꢀC. 1H NMR (DMSO-d6)
d:
205e206 ꢀC. 1H NMR (DMSO-d6)
d: 10.48 (brs, 1H), 8.21e8.15 (m,
8.28e8.17 (m, 2H), 8.11 (d, 2H, J ¼ 8.5 Hz), 8.03e7.89 (m, 2H), 7.18 (d,
2H, J ¼ 8.5 Hz), 3.88 (s, 3H). Anal calcd for C18H11NO4 (305.28): C
70.82, H 3.63, N 4.59. Found: C 70.56, H 3.79, N 4.41.
1H), 8.12e8.05 (m, 1H), 7.98e7.89 (m, 2H), 7.49 (s, 1H), 6.74 (s, 1H),
3.90 (s, 3H). Anal calcd for C18H10ClNO5 (355.73): C 60.77, H 2.83, N
3.94. Found: C 60.58, H 2.99, N 3.79.
6.2.3. 3-(4-Hydroxyphenyl)-naphtho[2,3ed]isoxazole-4,9-dione (3c)
Compound 3b (100 mg, 0.33 mmol) was dissolved in 2 mL of HBr
35% in acetic acid and the solution was refluxed for 5 h. After
cooling at room temperature, dichloromethane was added and the
6.2.9. 3-(5-Chloro-2,4-dihydroxyphenyl)-naphtho[2,3ed]isoxazole-
4,9-dione (3i)
Compound 3h (140 mg, 0.39 mmol) was dissolved in 15 mL of
HBr 35% in acetic acid and the solution was refluxed for 48 h. After

72
A. Bargiotti et al. / European Journal of Medicinal Chemistry 53 (2012) 64e75
cooling at room temperature, dichloromethane was added and the
organic phase was washed with sodium bicarbonate, water and
dried over Na2SO4. The solvent was evaporated under reduced
pressure to give a crude that was purified by flash chromatography
(CH2Cl2/acetone 99:1) to obtain 30 mg of 3i as a yellow solid (23%).
6.2.17. 3-(4-Fluorophenyl)-5-hydroxynaphtho[2,3ed]isoxazole-
4,9-dione (6b)
Compound 6a (50 mg, 0.14 mmol) was dissolved in 7 mL of
methanol and K2CO3 (50 mg, 0.36 mmol) in 3 mL of water was
added under nitrogen. After stirring for 40 min at room temper-
ature, the solution was poured into water and acidified with
diluted acetic acid. The solution was extracted with dichloro-
methane and the organic phase was washed with water, dried
over Na2SO4 and evaporated under reduced pressure. The residue
was purified by flash chromatography (CH2Cl2) to give 35 mg of 6b
Mp.: 199e200 ꢀC 1H NMR (CDCl3)
d: 10.07 (brs, 1H), 9.05 (brs, 1H),
8.44e8.25 (m, 2H), 8.00e7.84 (m, 3H), 6.36 (s, 1H). Anal calcd for
C17H8ClNO5 (341.70): C 59.75, H 2.36, N 4.10. Found C 59.53, H 2.56,
N 4.21.
6.2.10. 4-{2-[4-(4,9-dioxo-4,9-dihydronaptho[2,3ed]isoxazol-3-
yl)-phenoxy]-ethyl}-morpholin-4-ium chloride (3k)
(80%). Orange solid Mp.: 208e209 ꢀC. 1H NMR (CDCl3)
d: 12.20
(brs, 1H), 8.20e8.12 (m, 2H), 7.87 (d, 1H, J ¼ 8.2 Hz), 7.75e7.68 (m,
1H), 7.40 (d, 1H, J ¼ 8.2 Hz), 7.30e7.21 (m, 2H). Anal calcd for
C17H8FNO4 (309.25): C 66.03, H 2.61, N 4.53. Found: C 66.24, H
2.41, N 4.29.
Purified by flash chromatography (Eluent CH2Cl2/CH3OH 95:5).
The product was dissolved in dichloromethane (5 mL) and treated
with conc. HCl in 1,4-dioxane until pH ¼ 2. The solid was filtered
and washed with diethyl ether to give a yellow solid (52%). Mp.:
240 ꢀC. 1H NMR (DMSO-d6)
d: 10.70 (brs, 1H), 8.28e8.09 (m, 4H),
6.2.18. 5-Methoxy-3-pyridin-4-yl-naphtho[2,3ed]isoxazole-4,9-
dione (6c)
8.07e7.91 (m, 2H), 7.25 (d, 2H, J ¼ 8.5 Hz), 4.65e4.48 (m, 2H),
4.12e3.95 (m, 2H), 3.90e3.74 (m, 2H), 3.70e3.49 (m, 4H),
3.31e3.02 (m, 2H). Anal calcd for C23H21ClN2O5 (440.88): C 62.66, H
4.80, N 6.35. Found: C 62.79, H 4.65, N 6.21.
Purified by flash chromatography (Eluent CH2Cl2/CH3OH 96:4).
Yellow solid (35%). Mp.: 214e215 ꢀC. 1H NMR (CDCl3)
d: 8.81 (d, 2H,
J ¼ 6.1 Hz), 8.07 (d, 2H, J ¼ 6.1 Hz), 7.96 (d, 1H, J ¼ 8.0 Hz), 7.78 (dd,
1H, J ¼ 8.4 and 8.0 Hz), 7.46 (d, 1H, J ¼ 8.4 Hz), 4.06 (s, 3H). Anal.
Calcd. for C17H10N2O4 (306.27): C 66.67, H 3.29, N 9.15. Found: C
66.99, H 3.01, N 9.15.
6.2.11. 3-(1H-indol-5-yl)-naphtho[2,3ed]isoxazole-4,9-dione (3l)
Purified by flash chromatography (Eluent hexane/AcOEt 2:1).
Sticky solid (35%). 1H NMR (CD3)2CO
d: 8.58 (s, 1H), 8.35e8.21 (m,
2H), 8.06e7.87 (m, 3H), 7.61 (d, 1H, J ¼ 8.6 Hz), 7.51e7.44 (m, 1H),
6.72e6.63 (m, 1H). Anal calcd for C19H10N2O3 (314.29): C 72.61, H
3.21, N 8.91. Found: C 72.87, H 3.05, N 8.65.
6.2.19. Acetic acid 4,9-dioxo-3-pyridin-4-yl-4,9-dihydro-naphtho
[2,3ed]isoxazol-5-yl ester (6d)
Purified by flash chromatography (Eluent CH2Cl2/CH3OH 97:3).
Brown solid (29%). Mp.: 151e152 ꢀC. 1H NMR (CDCl3)
d: 8.95e8.74
6.2.12. 3-Styryl-naphtho[2,3ed]isoxazole-4,9-dione (3m)
(m, 2H), 8.30 (d, 1H, J ¼ 8.0 Hz), 8.03e7.96 (m, 2H), 7.89e7.80 (m,
1H), 7.50 (d, 1H, J ¼ 8.4 Hz), 2.46 (s, 3H). Anal calcd for C18H10N2O5
(334.28): C 64.67, H 3.02, N 8.38. Found C 64.98, H 2.98, N 8.49.
Purified by flash chromatography (Eluent hexane/AcOEt 4:1).
Bright yellow sticky solid (40%). 1H NMR (CDCl3)
d: 8.38e8.28 (m,
2H), 8.24 (d, 1H, J ¼ 16.2 Hz), 7.95e7.80 (m, 2H), 7.73e7.62 (m, 2H),
7.52e7.36 (m, 4H). Anal calcd for C19H11NO3 (301.30): C 75.74, H
3.68, N 4.65. Found: C 75.48, H 3.89, N 4.45.
6.2.20. 5-Hydroxy-3-pyridin-4-yl-naphtho[2,3ed]isoxazole-4,9-
dione (6e)
Compound 6d (120 mg, 0.36 mmol) was dissolved in 10 mL of
methanol and K2CO3 (360 mg, 2.6 mmol) in 7 mL of water was
added under nitrogen. After stirring for 30 min at room tempera-
ture, the solution was poured into water and acidified with diluted
acetic acid. The solution was extracted with dichloromethane and
the organic phase was washed with water, dried over Na2SO4 and
evaporated under reduced pressure. The residue was purified by
flash chromatography (CH2Cl2/hexane, 50:50) to give 100 mg of 6e
6.2.13. 3-Pyridin-4-yl-naphtho[2,3ed]isoxazole-4,9-dione (3n)
Purified by flash chromatography (Eluent CH2Cl2/CH3OH 20:1).
Yellow solid (26%). Mp.: 220 ꢀC. 1H NMR (DMSO-d6)
d: 8.88 (d, 2H,
J ¼ 6.1 Hz), 8.32e8.17 (m, 2H), 8.08 (d, 2H, J ¼ 6.1 Hz), 8.06e7.93 (m,
2H). Anal calcd for C16H8N2O3 (276.25): C 69.57, H 2.92, N 10.14.
Found: C 69.39, H 3.09, N 10.34.
6.2.14. 3-Pyridin-3-yl-naphtho[2,3ed]isoxazole-4,9-dione (3o)
Purified by flash chromatography (Eluent CH2Cl2/CH3OH 97.5:
2.5). Light yellow solid (91%). Mp.: 188e189 ꢀC. 1H NMR (DMSO-d6)
as an orange solid (95%). Mp.: 210e211 ꢀC. 1H NMR (CDCl3)
d: 12.16
(brs, 1H), 8.84(d, 2H, J ¼ 6.1 Hz), 8.05 (d, 2H, J ¼ 6.1 Hz), 7.86 (d, 1H,
J ¼ 7.8 Hz), 7.72 (dd, 1H, J ¼ 8.2, 7.8 Hz), 7.40 (d, 1H, J ¼ 8.2 Hz). Anal
calcd for C16H8N2O4 (292.25): C 65.76, H 2.76, N 9.59. Found: C
65.56, H 2.98, N 9.72.
d: 9.21 (s, 1H), 8.89e8.80 (m, 1H), 8.51e8.40 (m, 1H), 8.29e8.02 (m,
2H), 8.08e7.91 (m, 2H), 7.73e7.62 (m, 1H). Anal calcd for C16H8N2O3
(276.25): C 69.57, H 2.92, N 10.14. Found: C 69.69, H 2.75, N 10.27.
6.2.21. 3-(4-Fluorophenyl)-1-oxa-2,6-diazacyclopenta[b]
naphthalene-4,9-dione (10a)
6.2.15. 3-Pyridin-2-yl-naphtho[2,3ed]isoxazole-4,9-dione (3p)
Purified by flash chromatography (Eluent CH2Cl2/CH3OH 97.5:
Purified by flash chromatography (Eluent CH2Cl2/CH3OH 96:4).
2.5). Yellow solid (28%). Mp.: 170e171 ꢀC. 1H NMR (DMSO-d6)
d:
(27%). 1H NMR (CDCl3)
8.30e8.22 (m, 2H), 8.09 (d, 1H, J ¼ 4.8 Hz), 7.35e7.27 (m, 2H). Anal
calcd for C16H7FN2O3 (294.24): C 65.31, H 2.40, N 9.52. Found: C
65.54, H 2.63, N 9.28.
d
: 9.56 (s, 1H), 9.22 (d, 1H, J ¼ 4.8 Hz),
8.88e8.79 (m, 1H), 8.28e8.07 (m, 4H), 8.03e7.92 (m, 2H), 7.76e7.60
(m, 1H). Anal calcd for C16H8N2O3 (276.25): C 69.57, H 2.92, N 10.14.
Found: C 69.78, H 2.72, N 10.29.
6.2.16. Acetic acid 3-(4-fluorophenyl)-4,9-dioxo-4,9-
dihydronaphtho[2,3ed]isoxazol-5-yl ester (6a)
6.2.22. 3-(4-Methoxyphenyl)-1-oxa-2,6-diaza-cyclopenta[b]
naphthalene-4,9-dione (10b)
Purified by flash chromatography (Eluent CH2Cl2/CH3OH 40:1)
Purified by flash chromatography (Eluent CH2Cl2/CH3OH
Yellow solid (57%). Mp.: 191e192 ꢀC. 1H NMR (CDCl3)
d
: 8.27 (dd,1H,
97.5:2.5) Yellow solid (30%). Mp.: 182e183 ꢀC. 1H NMR (CDCl3)
d:
J ¼ 7.8, 1.1 Hz), 8.14e8.05 (m, 2H), 7.85 (dd, 1H, J ¼ 7.8, 7.8 Hz), 7.50
(dd, 1H, J ¼ 7.82, 1.12 Hz), 7.28e7.18 (m, 2H), 2.47 (s, 3H). Anal calcd
for C19H10FNO5 (351.28): C 64.96, H 2.87, N 3.99. Found: C 65.09, H
2.67, N 4.11.
9.56 (s, 1H), 9.22 (d, 1H, J ¼ 8.3 Hz), 8.27e8.12 (m, 2H), 8.10e8.03
(m, 1H), 7.05 (d, 2H, J ¼ 8.3 Hz), 3.91 (s, 3H). Anal calcd for
C17H10N2O4 (306.27): C 66.67, H 3.29, N 9.15. Found: C 66.41, H 3.50,
N 9.28.

A. Bargiotti et al. / European Journal of Medicinal Chemistry 53 (2012) 64e75
73
6.2.23. 4-{2-[4-(4,9-dioxo-4,9-dihydro-1-oxa-2,6-diazacyclopenta
[b]naphthalen-3-yl)-phenoxy]-ethyl}-morpholin-4-ium chloride
(10c)
8.10e7.94 (m, 2H), 4.49 (s, 3H). Anal calcd for C17H11IN2O3 (418.19):
C 48.83, H 2.65, N 6.70. Found: C 48.55, H 2.91, N 6.52.
Purified by flash chromatography (Eluent CH2Cl2/CH3OH 95:5).
The product was dissolved in dichloromethane (5 mL) and treated
with conc. HCl in 1,4-dioxane until pH ¼ 2. The solid was filtered
and washed with diethyl ether (19%). Mp.: 194e195 ꢀC. 1H NMR
6.3.5. 4-(4,9-Dioxo-4,9-dihydro-naphtho[2,3ed]isoxazol-3-yl)-1-
(pyrazin-2-yl)methyl-pyridinium bromide (7e)
Compound 3n (50 mg, 0.18 mmol) was dissolved in 1.5 ml of
DMSO and 2-bromomethylpyrazine (50 mg, 0.28 mmol) was added.
After having stirred for 72 h at room temperature under nitrogen,
the solution was poured into water and extracted with dichloro-
methane. The organic phase was dried and evaporated under
reduced pressure. The residue was purified by crystallization with
methanol/diethyl ether to give 30 mg of 7e. as a yellow solid (37%)
(DMSO-d6)
d
: 11.20 (brs, 1H), 9.41 (s, 1H), 9.24 (d, 1H, J ¼ 4.8 Hz),
8.21e8.02 (m, 3H), 7.29 (d, 2H, J ¼ 8.5 Hz), 4.69e4.52 (m, 2H),
4.20e3.20 (m, 10H). Anal calcd for C22H20ClN3O5 (441.86): C 59.80,
H 4.56, N 9.51. Found: C 59.59, H 4.68, N 9.66.
6.2.24. 3-Pyridin-4-yl-1-oxa-2,6-diaza-cyclopenta[b]naphthalene-
4,9-dione (10d)
Mp.: 170e171 ꢀC. 1H NMR (DMSO-d6)
d
: 9.45 (d, 2H, J ¼ 5.9 Hz), 9.05
(s, 1H), 8.92 (d, 2H, J ¼ 5.9 Hz), 8.86 (s, 1H), 8.81 (s, 1H), 8.31e8.20
(m, 2H), 8.12e8.00 (m, 2H), 6.28 (s, 2H). Anal calcd for
C21H13BrN4O3 (449.26): C 56.14, H 2.92, N 12.47. Found: C 56.38, H
2.69, N 12.30.
Purified by flash chromatography (Eluent CH2Cl2/CH3OH 20:1)
(20%). 1H NMR (CDCl3)
2H, J ¼ 5.1 Hz), 8.27e8.17 (d, 2H, J ¼ 5.1 Hz), 8.08 (d, 1H, J ¼ 4.8 Hz).
Anal calcd for C15H7N3O3 (277.23): C 64.98, H 2.54, N 15.16. Found:
C 64.71, H 2.76, N 15.28.
d
: 9.58 (s,1H), 9.23 (d,1H, J ¼ 4.8 Hz), 8.89 (d,
6.3.6. 1-Benzyl-4-(4,9-dioxo-4,9-dihydro-naphtho[2,3ed]isoxazol-
3-yl)-pyridinium bromide (7f)
6.3. General procedure for the alkylation reaction
Yellow solid (56%). Mp.: 206 ꢀC. 1H NMR (DMSO-d6)
d: 9.51 (d,
2H, J ¼ 6.1 Hz), 8.89 (d, 2H, J ¼ 6.1 Hz), 8.35e8.18 (m, 2H), 8.11e7.96
(m, 2H), 7.78e7.59 (m, 2H), 7.55e7.40 (m, 3H), 6.01 (s, 2H). Anal
calcd for C23H15BrN2O3 (447.28): C 61.76, H 3.38, N. 6.26. Found: C
61.57, H 3.61, N. 6.01.
The appropriate 3-pyridyl derivative (0.1 mmol) was dissolved
in 1 mL of DMSO and benzyl bromide (0.33 mmol) was added. The
mixture was stirred at room temperature overnight under nitrogen.
After cooling diethyl ether was added, the solid was filtered,
washed with diethyl ether and crystallized.
6.3.7. 1-Benzyl-2-(4,9-dioxo-4,9-dihydro-naphtho[2,3ed]isoxazol-
3-yl)-pyridinium bromide (7g)
Thus, when not stated otherwise, were prepared the following
products:
Purified by flash chromatography (CH2Cl2/CH3OH, 4:1). Sticky
solid (37%). 1H NMR (DMSO-d6)
d: 8.79e8.70 (m, 1H), 8.57e8.42 (m,
6.3.1. 1-Benzyl-4-(5-methoxy-4,9-dioxo-4,9-dihydronaphtho
[2,3ed]isoxazol-3-yl)-pyridinium bromide (7a)
1H), 8.03e7.95 (m, 1H), 7.93e7.81 (m, 3H), 8.80e7.70 (m, 1H),
7.69e7.60 (m, 1H), 7.46e7.25 (m, 5H), 5.69 (s, 2H). Anal calcd for
C23H15BrN2O3 (447.28): C 61.76, H 3.38, N. 6.26. Found: C 61.56, H
3.65, N. 6.45.
Yellow solid (64%). Mp.: 216e218 ꢀC (from methanol/diethyl
ether). 1H NMR (DMSO-d6)
d: 9.44 (d, 2H, J ¼ 5.9 Hz), 8.80 (d, 2H,
J ¼ 5.9 Hz), 7.99e7.88 (m, 2H), 7.78e7.69 (m, 1H), 7.67e7.57 (m, 2H),
7.56e7.42 (m, 3H), 5.99 (s, 2H), 3.98 (s, 3H). Anal calcd for
C24H17BrN2O4 (477.31): C 60.39, H 3.59, N 5.87. Found: C 60.61, H
3.32, N 5.65.
6.3.8. 1-Benzyl-3-(4,9-dioxo-4,9-dihydro-naphtho[2,3ed]isoxazol-
3-yl)-pyridinium bromide (7h)
Yellow solid. (60%). Mp.: 184e185 ꢀC (from methanol/diethyl
ether). 1H NMR (DMSO-d6)
d: 9.98 (s, 1H), 9.52e9.41 (m, 1H),
6.3.2. 4-(5-Acetoxy-4,9-dioxo-4,9-dihydronaphtho[2,3ed]isoxazol-
3-yl)-1-benzyl-pyridinium bromide (7b)
9.35e9.24 (m, 1H), 8.53e8.39 (m, 1H), 8.30e8.15 (m, 2H), 8.09e7.98
(m, 2H), 7.75e7.62 (m, 2H), 7.58e7.40 (m, 3H), 6.65 (s, 2H). Anal
calcd for C23H15BrN2O3 (447.28): C 61.76, H 3.38, N. 6.26. Found: C
61.91, H 3.11, N. 6.13.
Yellow solid (97%). Mp.: 203e204 ꢀC (from methanol). 1H NMR
(DMSO-d6)
d
: 9.45 (d, 2H, J ¼ 5.9 Hz), 8.77 (d, 2H, J ¼ 5.9 Hz),
8.25e8.17 (m, 1H), 8.09e7.99 (m, 1H), 7.79e7.70 (m, 1H), 7.67e7.59
(m, 2H), 7.54e7.43 (m, 3H), 5.98 (s, 2H), 2.39 (s, 3H). Anal calcd for
C25H17BrN2O5 (505.32): C 59.42, H 3.39, N. 5.54. Found: C 59.63, H
3.54, N. 5.40.
6.3.9. 3-(4,9-Dioxo-4,9-dihydro-naphtho[2,3ed]isoxazol-3-yl)-1-
pyrazin-2-ylmethyl-pyridinium bromide (7i)
Compound 3o (50 mg, 0.18 mmol) was dissolved in 1 mL of
DMSO and 2-bromomethylpyrazine (50 mg, 0.29 mmol) was added.
The mixture was stirred at room temperature for 72 h under
nitrogen. After cooling diethyl ether was added, the solid was
filtered and purified by crystallization from methanol/diethyl ether
to give 58 mg of 7i. as a brown solid (72%). Mp.: 178e179 ꢀC. 1H
6.3.3. 1-Benzyl-4-(5-hydroxy-4,9-dioxo-4,9-dihydronaphtho
[2,3ed]isoxazol-3-yl)-pyridinium bromide (7c)
Yellow solid (65%). Mp.: 198e199 ꢀC (from methanol/diethyl
ether). 1H NMR (DMSO-d6)
8.83 (d, 2H, J ¼ 6.5 Hz), 7.90e7.77 (m, 2H), 7.67e7.58 (m, 2H),
7.56e7.43 (m, 4H), 6.01 (s, 2H). Anal calcd for C23H15BrN2O4
(463.28): C 59.63, H 3.26, N 6.05. Found: C 59.38, H 3.48, N 6.25.
d
: 11.79 (s, 1H), 9.47 (d, 2H, J ¼ 6.5 Hz),
NMR (DMSO-d6) d: 9.94 (s, 1H), 9.51e9.31 (m, 2H), 9.02 (s, 1H),
8.80e8.63 (m, 2H), 8.60e8.47 (m, 1H), 8.33e8.16 (m, 2H), 8.10e7.91
(m, 2H), 6.29 (s, 2H). Anal calcd for C21H13BrN4O3 (449.26): C 56.14,
H 2.92, N 12.47. Found: C 55.97, H 3.09, N 12.11.
6.3.4. 4-(4,9-Dioxo-4,9-dihydronaphtho[2,3ed]isoxazol-3-yl)-1-
methylpyridinium iodide (7d)
6.4. Acetic acid 9-acetoxy-3-(4-methoxyphenyl)-4,9-
Compound 3n (30 mg, 0.1 mmol) was dissolved in 1 mL of
methanol and iodomethane (1 mL) was added. The mixture was
stirred at room temperature overnight then the solvent was
evaporated under reduce pressure. The crude was purified by
crystallization with methanol/dichloromethane to give 15 mg of 7d
dihydronaphtho[2,3ed]isoxazol-4-yl ester (8a)
Compound 3b (90 mg, 0.29 mmol), zinc powder (30 mg,
0.45 mmol) and triethylamine (1.4 g, 14 mmol) in 4 mL of acetic
anhydride were refluxed for 3 h. After cooling to room temperature,
the solution was poured into water and extracted with CH2Cl2. The
solvent was evaporated under reduced pressure and the residue
as a yellow solid (36%). Mp.: 210e211 ꢀC. 1H NMR (DMSO-d6)
d
: 9.25
(d, 2H, J ¼ 6.0 Hz), 8.81 (d, 2H, J ¼ 6.0 Hz), 8.31e8.17 (m, 2H),

74
A. Bargiotti et al. / European Journal of Medicinal Chemistry 53 (2012) 64e75
was purified by flash chromatography (CH2Cl2) to give 60 mg of 8a
(53%). Mp.: 154e155 ꢀC. 1H NMR (CDCl3)
7.99 (d, 1H, J ¼ 8.9 Hz), 7.77 (d, 2H, J ¼ 8.2 Hz), 7.70e7.66 (m, 1H),
7.56e7.53 (m, 1H), 7.09 (d, 2H, J ¼ 8.2 Hz), 3.93 (s, 3H), 2.62 (s, 3H),
2.10 (s, 3H). Anal calcd for C22H19NO6 (393.39): C 67.17, H 4.87, N
3.56. Found: C 67.39, H 4.65, N 3.40.
kindly provided by Dr. N. Zaffaroni (IRCCS Istituto Nazionale
Tumori, Milan) and grown in 50/50 DMEM-Ham’s F12 medium
(Lonza, Vierviers, Belgium), supplemented with 10% (v/v) heat-
d
: 8.05 (d, 1H, J ¼ 9.3 Hz),
inactivated foetal bovine serum and 2 mM L-glutamine (L-Gln).
All cell lines were maintained at 37 ꢀC in a 5%/95% CO2/air
atmosphere.
6.5. Acetic acid 9-acetoxy-3-(4-fluorophenyl)-4,9-dihydro-naphtho
[2,3ed]isoxazol-4-yl ester (8b)
6.8.2. Drugs and treatment conditions
All the compounds tested were freshly prepared dissolving
powder in dimethylsulfoxide (DMSO) (BDH Prolabo, Milan, Italy)
and following dilution in culture medium. Cells were incubated
with drugs for 6, 24 or 72 h (depending on the assay), at 37 ꢀC in
culture medium supplemented with 10% (v/v) heat-inactivated
foetal bovine serum; drug concentrations used for each assay are
reported in Legends to Figures.
Compound 3d (90 mg, 0.25 mmol), zinc powder (30 mg,
0.45 mmol) and triethylamine (1.4 g, 14 mmol) in 4 mL of acetic
anhydride were refluxed for 2 h. After cooling to room temperature,
the solution was poured into water and extracted with CH2Cl2. The
solvent was evaporated under reduced pressure and the residue
was purified by flash chromatography (CH2Cl2) to give 95 mg of 8b.
(100%). 1H NMR (CDCl3)
7.71e7.61 (m, 1H), 7.59e7.48 (m, 1H), 7.32e7.20 (m, 2H), 2.60 (s,
d: 8.09e7.94 (m, 2H), 7.88e7.78 (m, 2H),
6.8.3. Fluorescence polarization assay
d
GM-FITC Geldanamycin- fluorescein-5-isothiocyanate (working
concentration: 1e400 nM), supplied by Invivogen (06C23-MT, Cal-
ifornia 92192, USA), was previously dissolved in DMSO to obtain
10 mM stock solutions and kept at ꢂ20 ꢀC until use. Hsp90,
purchased from (Stressgen cat. No. SPP-776, Victoria BC, Canada),
3H), 2.09 (s, 3H). Anal calcd for C21H16FNO5 (381.35): C 66.14, H
4.23, N 3.67. Found C 66.34, H 4.01, N 3.51.
6.6. Quality of the pharmacophoric hypotheses
was previously dissolved in assay buffer (HFB) to form 2.2 mM stock
The quality of each pharmacophore was first measured in three
ways based on the alignments to the input structures: (1) the
alignment score, which is the root-mean-squared deviation (rmsd)
in the site-point positions; (2) the vector score, which is the average
cosine of the angles formed by corresponding pairs of vector
features (acceptors, donors and aromatic rings) in the aligned
structures; (3) a volume score based on the overlap of van der
Waals models of the non-hydrogen atoms in each pair of structures.
For details, see Dixon et al. [36].
solutions and kept at ꢂ80 ꢀC until use.
The compounds were previously dissolved in DMSO to obtain
stock solutions and kept at ꢂ20 ꢀC. The day of experiment, the
compounds were prepared by serial dilutions in assay buffer
(HFB) containing 20 mM HEPES (K) pH 7.3, 50 mM KCl, 5 mM
MgCl2, 20 mM Na2MoO4 and 0.01% NP40. Before each use,
0.1 mg/ml Bovine Gamma globulin and 2 mM DTT were freshly
added.
Fluorescence Polarization (FP) was performed in Opti-PlateÔ-
96F well plates (Perkin Elmer, Zaventem, Belgium) using a plate
reader (Wallac Envision 2101 multilabel reader, Perkin Elmer, Zaven-
tem, Belgium). To evaluate the binding affinity of the molecules,
6.7. Computational details
All molecules investigated in this study were built with the
Maestro 9.1 graphical interface [38] Each structure was subse-
quently minimized with the OPLS_2005 force field, a constant
dielectric of 1.0 and an implicit solvation treatment using Macro-
Model 9.8 [39]. Conformational sampling was carried out using the
same settings of minimization with minimum atom deviation of
0.5 Å. The maximum relative energy difference was set to 2.0 kcal/
mol. Pharmacophore development, site-point superpositions,
vector characteristics alignment, molecular volume overlap,
penalty scoring, and QSAR model building were carried out using
Phase 3.2 [40]. All calculations were performed in Maestro 9.1.
Phase QSAR models do not use internal cross-validation tech-
niques, but rather use distinct training and test sets. However, the
use of leave-n-out techniques are useful for assessing the stability
of the model to changes in the training set. In Phase QSAR models,
leave-n-out models were built, and the.
50
m
l of the GM-FTC solution (5 nM) were added to 30 nM of Hsp90
in the presence of 5
ml of the test compounds at increasing
concentrations. The plate was mixed on a shaker at 4 ꢀC for 4 h, and
the FP values in mP (millipolarization units) were recorded. The
IC50 values were calculated as the inhibitor concentration where
50% of the tracer is displaced; each data point is the result of the
average of triplicate wells, and was determined from a plot using
nonlinear least-squares analysis. Curve fitting was performed using
Prism GraphPad software program (GraphPad software, Inc., San
Diego, CA).
6.8.4. Protein expression analysis
Total cell lysates were prepared rinsing cells twice with ice-cold
PBS supplemented with 0.1 mM sodium orthovanadate, then lysing
them in hot sample buffer as previously described [41]. After
determination of protein concentration (BCA Protein Assay
Reagent, Pierce, Thermo Fisher Scientific Inc., Rockford, IL, USA),
whole-cell extracts were separated by SDS polyacrylamide gel
electrophoresis (PAGE) and transferred onto nitrocellulose
membranes. Detection of proteins was accomplished using horse-
radish peroxidase-conjugated secondary antibodies and a chem-
iluminescence reagent purchased through Amersham Biosciences
(Rockford, IL). For quantification of signals, blots were also sub-
jected to densitometry analysis.
R2 value was computed between the leave-n-out predictions
and the predictions from the model built on the full training set.
This value was reported as the stability value. Models with high
stability (maximum value ¼ 1) were preferred because they were
not overly dependent on any particular training set.
6.8. Biological activity
6.8.1. Cell lines and culture conditions
Primary antibodies used in this study are: anti-Raf-1 and anti-
Cdk4 (Santa Cruz Biotechnology Inc., CA, USA), anti-EGFR
(Upstate Biotechnology, Millipore Corporate, Billerica, MA, USA),
anti-Survivin (Abcam, Cambridge, UK), anti-Akt (Transduction
Laboratories, Lexington, USA), anti-HSP70 (BRM-22) and anti-Actin
(Sigma Chemical Co., St. Louis, MO, USA).
The epithelial carcinoma cell line A431 and the non-small cell
lung carcinoma cell line NCI-H460 were routinely grown in RPMI
1640 (Lonza, Vierviers, Belgium), supplemented with 10% (v/v)
heat-inactivated foetal bovine serum (GIBCO, Invitrogen, Paisley,
UK). The human peritoneal mesothelioma cell line STO was

A. Bargiotti et al. / European Journal of Medicinal Chemistry 53 (2012) 64e75
[15] T. Taldone, W. Sun, G. Chiosis, Bioorg. Med. Chem. 17 (2009) 2225e2235.
75
6.8.5. Animals
[16] C. Grundmann, P. Grünanger, Nitrile Oxides, Springer Verlag, Berlin, 1971.
Experiments were carried out using female athymic Swiss nude
mice, 8e10 weeks old (Charles River, Calco, Italy). Mice were
maintained in laminar flow rooms keeping temperature and
humidity constant. Mice had free access to food and water. Exper-
iments were approved by the Ethics Committee for Animal Exper-
imentation of the Istituto Nazionale Tumori of Milan according to
institutional guidelines.
[17] (a) A. Quilico, G. Stagno D’Alcontres, Gazz. Chim. Ital 80 (1950) 140e150;
(b) S. Morrocchi, A. Quilico, A. Ricca, A. Selva, Gazz. Chim. Ital 98 (1968)
891e906;
(c) T. Matsuura, J.W. Bode, Y. Hachisu, K. Suzuki, Synlett (2003) 1746e1748;
(d) N. Ben Hamadi, M. Msaddek, Heterocycl. Comm. 12 (2006) 457e462.
[18] (a) R. Fusco, S. Maiorana, P. Del Buttero, E. Licandro, A. Alemagna, Chim. e Ind.
(Milan) 63 (1981) 401e404;
(b) M.P. Fernandez, B. Gonzalez, M. Pardo, J.L. Soto, Anal. Quim 90 (1994)
477e482;
(c) G. Brahmeshwari, V.R. Rao, T.V.P. Rao, Indian J. Chem. 34B (1995) 139e140;
(d) F. Farina, M.V. Martin, M. Munoz, M.C. Paredes, R. Rodriguez, Heterocycles
40 (1995) 413e424.
Acknowledgements
[19] G.A. Lee, Synthesis (1982) 508e509.
We thank Mr. Silvio Zavatto for the excellent technical assis-
tance. This work was supported by grants from Sigma-Tau
Research, Switzerland S.A., P.O. Box 1823 Via Motta 2a - CH-6850
Mendrisio, Switzerland, Tel.: (þ41)091640.4050.
[20] J.L. Stevens, T.D. Welton, J.P. Deville, V. Behar, Tetr. Lett. 44 (2003) 8901e8903.
[21] Y. Kitani, A. Morita, T. Kumamoto, T. Ishikawa, Helv. Chim. Acta 85 (2002)
1186e1195.
[22] (a) R.W. Rickards, J.M. Rothschild, A.C. Willis, M.M. de Chazal, J. Kirk, K. Kirk,
K.J. Saliba, G.D. Smith, Tetrahedron 55 (1999) 13513e13520;
(b) P.H. Bernardo, C.L.L. Chai, M. Le Guen, G.D. Smith, P.H. Waring, Bioorg.
Med. Chem. Lett. 17 (2007) 82e85.
Appendix A. Supplementary data
[23] G.R. Pettit, J.C. Collins, J.C. Knight, B.C. Herald, R.A. Nieman, M.D. Williams,
R.K. Pettit, J. Nat. Prod. 66 (2003) 544e547.
[24] (a) A.P. Krapcho, M.E. Petry, Z. Getahun, J.J. Landi Jr., J. Stallman,
J.F. Polsenberg, C.E. Gallagher, M.J. Maresch, M.P. Hacker, J. Med. Chem. 37
(1994) 828e837;
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.ejmech.2012.03.036.
(b) J.A. Valderrana, M.F. González, D. Pessoa-Mahana, R.A. Tapia, H. Fillion,
F. Pautet, J.A. Rodriguez, C. Theoduloz, G. Schmeda-Hirschmann, Bioorg. Med.
Chem. 14 (2006) 5003e5011.
References
[25] B.A. Czeskis, J. Label Compd. Radiopharm. 39 (1997) 49e58.
[26] H. Young Choi, D. Yoon Chi, Tetr. Lett. 60 (2004) 4945e4951.
[27] J. Kim, S. Felts, L. Llauger, H. He, H. Huezo, N. Rosen, G. Chiosis, J. Biomol.
Screen. 9 (2004) 375e381.
[28] D.B. Solit, G. Chiosis, Drug Discov. Today 13 (2008) 38e43.
[29] (a) L.R. Kelland, S.Y. Sharp, P.M. Rogers, T.G. Myers, P. Workman, J. Natl. Cancer
Inst. 91 (1999) 1940e1949;
[1] C. Prodromou, S.M. Roe, R. O’Brien, J.E. Ladbury, P.W. Piper, L.H. Pearl, Cell 90
(1997) 65e75.
[2] L. Whitesell, S.L. Lindquist, Nat. Rev. Cancer 5 (2005) 761e772.
[3] (a) K. Richter, J. Buchner, Structure 19 (2011) 445e446;
(b) S.K. Calderwood, J. Gong, J. Cell. Biochem. (2011). doi:10.1002/jcb.23461.
[4] A.E. Kabakov, V.A. Kudryavtsev, Curr. Top. Pharmacol. 14 (2010) 89e102.
[5] A. Donnelly, B.S.J. Blagg, Curr. Med. Chem. 15 (2008) 2702e2717.
[6] Y.L. Janin, Drug Discov. Today 15 (2010) 342e343.
(b) W. Guo, P. Reigan, D. Siegel, J. Zirrolli, D. Gustafson, D. Ross, Cancer Res. 65
(2005) 10006e10015;
(c) W. Guo, P. Reigan, D. Siegel, J. Zirrolli, D. Gustafson, D. Ross, Mol. Phar-
macol. 70 (2006) 1194e1203.
[7] L. Neckers, Trends Mol. Med. 8 (2002) S55eS61.
[8] Y.L. Janin, J. Med. Chem. 48 (2005) 7503e7512.
[9] M.J. Drysdale, P.A. Brough, Drug Discov. Today 13 (2008) 38e43.
[10] T. Taldone, A. Gozman, R. Maharaj, G. Chiosis, Curr. Opin. Pharmacol. 8 (2008)
370e374.
[11] (a) M.A. Biamonte, R. Van de Water, J.W. Arndt, R.H. Scannevin, D. Perret, W.-
C. Lee, J. Med. Chem. 53 (2010) 3e17;
(b) S. Messaoudi, J.F. Peyrat, J.D. Brion, M. Alami, Expert Opin. Ther. Pat 21
(2011) 1501e1542.
[12] M.A. Colucci, C.J. Moody, G.D. Couch, Org. Biomol. Chem. 6 (2008) 637e656.
[13] M.K. Hadden, S.A. Hill, J. Davenport, R.L. Matts, B.S.J. Blagg, Bioorg. Med. Chem.
17 (2009) 634e640.
[30] A.M. Lewis, M. Ough, M.M. Hinkhouse, M.S. Tsao, L.W. Oberley, J.J. Cullen, Mol.
Carcinog 43 (2006) 215e224 (and refs. quoted therein).
[31 W. Guo, D. Siegel, D. Ross, J. Pharm. Sci. 97 (2008) 5147e5157.
[32] A.C. Maroney, J.J. Marugan, T.M. Mezzasalma, A.N. Barnakov, T.A. Garrabrant,
L.E. Weaner, W.J. Jones, L.A. Barnakova, H.K. Koblish, M.J. Todd, Biochemistry
45 (2006) 5678e5685.
[33] D. Siegel, B. Shieh, C. Yan, J.K. Kepa, D. Ross, J. Pharm. Exp. Ther. 336 (2011)
874e880.
[34] For a recent review see R. Beck, N. Dejeans, C. Glorieux, R.C. Pedrosa, D. Vás-
quez, J.A. Valderrama, P.B. Calderon, J. Verrax, Curr. Med. Chem. 18 (2011)
2816e2825.
[35] T. Taldone, D. Zatorska, P.D. Patel, H. Zong, A. Rodina, J.H. Ahn, K. Moulick,
M.L. Guzman, G. Chiosis, Bioorg. Med. Chem. 19 (2011) 2603e2614.
[36] S.L. Dixon, A.M. Smondyrev, E.H. Knoll, S.N. Rao, D.E. Shaw, R.A. Friesner,
J. Comput.-Aided Mol. Des 20 (2006) 647e671.
[37] Phase, Version 3.2. User Manual, Schrodinger, LLC, New York, 2005.
[38] Maestro, Version 9.1, Schrodinger, LLC, New York, 2005.
[39] MacroModel, Version 9.8, Schrodinger, LLC, New York, 2005.
[40] Phase, Version 3.2, Schrödinger, LLC, New York, 2005.
[41] 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)
280e289.
[14] (a) A. Gopalsamy, M. Shi, J. Golas, E. Vogan, J. Jacob, M. Johnson, F. Lee, R. Nilakantan,
R. Petersen, K. Svenson, R. Chopra, M.S. Tam, Y. Wen, J. Ellingboe, K. Arndt,
F. Boschelli, J. Med. Chem. 51 (2008) 373e375;
(b) P.A. Brough, W. Aherne, X. Barril, J. Borgognoni, K. Boxall, J.E. Cansfield, K.-
M.J. Cheung, I. Collins, N.G.M. Davies, M.J. Drysdale, B. Dymock, S.A. Eccles, H. Finch,
A.Fink,A.Hayes,R.Howes,R.E.Hubbard,K. James,A.M.Jordan,A.Lockie, V.Martins,
A. Massey, T.P. Matthews, E. McDonald, C.J. Northfield, L.H. Pearl, C. Prodromou,
S. Ray, F.I. Raynaud, S.D. Roughley, S.Y. Sharp, A. Surgenor, D.L. Walmsley, P. Webb,
M. Wood, P. Workman, L. Wright, J. Med. Chem. 51 (2008) 196e218;
(c) R. Baruchello, D. Simoni, G. Grisolia, G. Barbato, P. Marchetti, R. Rondanin,
S. Mangiola, G. Giannini, T. Brunetti, D. Alloatti, G. Gallo, A. Ciacci, L. Vesci,
M. Castorina, F.M. Milazzo, M.L. Cervoni, M.B. Guglielmi, M. Barbarino, R. Fodera,
C. Pisano, W. Cabri, J. Med. Chem. 54 (2011) 8592e8604.
̀