Design, synthesis and antiproliferative activities of biarylolefins based on polyhydroxylated and carbohydrate scaffolds

Article abstract of DOI:10.1016/j.ejmech.2011.05.021

A series of diversely substituted biarylolefins based on carbohydrate and dihydroxyethylene scaffolds were synthesized and evaluated for antiproliferative activity against a panel of human tumor cell lines. Among the thirty-five yet unknown biarylolefins

Full text of DOI:10.1016/j.ejmech.2011.05.021

European Journal of Medicinal Chemistry 46 (2011) 3570e3580
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Design, synthesis and antiproliferative activities of biarylolefins based on
polyhydroxylated and carbohydrate scaffolds
Alexandre Novoa a, Nadia Pellegrini-Moïse a, Stéphane Bourg b, Sylviane Thoret c, Joëlle Dubois c,
,
Geneviève Aubert d, Thierry Cresteil d, Yves Chapleur a
*
a UMR 7565 SRSMC, Nancy Université-CNRS, Groupe S.U.C.R.E.S, BP 239, F-54506 Nancy Vandœuvre, France
b Fédération de Recherche Physique et Chimie du Vivant, Université d’Orléans, CNRS, FR 2708, Rue Charles Sadron, 45071 Orléans Cedex 2, France
c Institut de Chimie des Substances Naturelles, UPR2301 CNRS, Avenue de la Terrasse, F-91198 Gif sur Yvette Cedex, France
d Institut de Chimie des Substances Naturelles, Ciblothèque cellulaire, UPR2301 CNRS, Avenue de la terrasse, F-91198 Gif sur Yvette Cedex, France
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of diversely substituted biarylolefins based on carbohydrate and dihydroxyethylene scaffolds
were synthesized and evaluated for antiproliferative activity against a panel of human tumor cell lines.
Among the thirty-five yet unknown biarylolefins prepared, six displayed potent antiproliferative activ-
ities with IC50 values in the micromolar and submicromolar range. As a new type of antiproliferative
agent, the most potent compound 26 showed an IC50 value of 70 nM against SK-OV3 cell line (ovarian
cancer). All the synthesized compounds exhibited a poor or modest tubulin polymerization inhibitory
activity suggesting another mode of action for these compounds. Molecular docking simulations to the
colchicine binding site of tubulin of representative compounds have been used to explain the lack of
activity as inhibitors of tubulin polymerization.
Received 2 March 2011
Received in revised form
6 May 2011
Accepted 8 May 2011
Available online 13 May 2011
Keywords:
Exo-glycals
Carbohydrates
Biarylolefins
Ó 2011 Elsevier Masson SAS. All rights reserved.
Antiproliferative agents
1. Introduction
clinical trials for the treatment of advanced cancers [6e8]. Many
derivatives and analogues have been prepared and evaluated in
Lignans are naturally occurring compounds found in many plants.
These secondary plant metabolites attracted considerable interest
due to their anti-inflammatory, immunosuppresive, cardiovascular
and antitumor activities. The latter activity has been largely inves-
tigated and compounds like podophyllotoxin or its clinically used
derivatives. Etoposide, Etopophos, and combretastatins are among
the most active representatives of lignans. From a structural point of
view, most lignans incorporate two diversely substituted phenyl
rings linked through a simple carbon chain as in combretastatins or
through more complex, often oxygenated, structures like in podo-
phyllotoxin and isotaxiresinol [1] (Chart 1).
structureeactivity relationships. Among all the synthetic
analogues, modifications of ring A, ring B and the double bond have
been investigated [9]. CA-4 analogues containing a variety of
heterocyclic moieties as surrogates of the olefinic bond have been
synthesized and display good cytotoxic activity. The influence of
the bridge structure between the two aromatics rings has been
largely investigated [10e15]. The replacement of the double bond
by a five-membered ring seemed to give the best results, suggesting
that the olefinic bond could be modified or expanded. Interestingly,
the synthesis and biological evaluations of cis-locked vinylogous
CA-4 analogues having a cyclopropylevinyl or a cyclopropyleamide
bridge have been reported [16].
As a result of SAR studies, the 3,4,5-trimethoxyphenyl A-ring
seemed mandatory for antiproliferative activity. The replacement of
the isovanillin (B-ring) moiety by a naphthalene ring results in
naphtylcombretastatins with cytotoxic activities and inhibition
potency of tubulin polymerization comparable to CA-4 [17,18].
Phenstatin, unexpectedly discovered by Petitt’s group in the course
of a SAR study of combretastatin substituents, was described as
a potent cancer cell growth inhibitor with an antitubulin activity
Combretastatin A4 (CA-4) displays a 1,2-diarylethene scaffold
bearing two polyoxygenated aromatic rings (Chart 1). It is highly
cytotoxic toward several cancer cell lines [2e4] and is a strong
inhibitor of tubulin polymerization through binding to the colchi-
cine binding site (IC50 ¼ 1.0
mM) [5]. The corresponding prodrug
water-soluble phosphate (CA4P) is currently in phases II and III
* Corresponding author. Tel.: þ33 383 68 47 43; fax: þ33 383 68 47 80.
E-mail address: yves.chapleur@srsmc.uhp-nancy.fr (Y. Chapleur).
similar to that of combretastatin CA-4 (IC50 ¼ 1.2
mM) (Chart 1) [19].
0223-5234/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2011.05.021

A. Novoa et al. / European Journal of Medicinal Chemistry 46 (2011) 3570e3580
3571
OR
OR
OR
RO
R = protecting group or H
O
MeO
HO
O
O
OH
OH
O
MeO
MeO
MeO
MeO
O
Y
Y
OMe
type II
1,1-diaryl-2,2-dihydroxymethylethylene
Chart 2. Structures of biarylolefin targets.
OMe
Y = OMe, OH, NO2, NH2,
C4H4, -OCH2O-
OH
MeO
OMe
OH
OMe
type I
1,1-diarylexoglycal
R = H Podophyllotoxin
R = 4,6-O-ethylidene-β-D-glucopyranoside
Isotaxiresinol
Etoposide
Our synthetic strategy involved the sequential introduction of the
3,4,5-trimethoxyphenyl moiety and of a second substituted aromatic
or bicyclic aromatic system. The 3,4,5-trimethoxyphenyl ring was
MeO
A
MeO
X
MeO
MeO
OH
A
B
B
first introduced on compound
1
by reaction of 3,4,
OMe
OMe
OH
5-trimethoxyphenyl boronic acid in the presence of tri-(2-furyl)
phosphine (TFP) and PdCl2(PPh3)2. The monosubstituted (Z)-isomer 2
wasformedasthemajorcompound obtainedin47%yield using1.3 eq
of boronic acid. The (E)-isomer 4 was formed in minute amount (7%)
together with the disubstituted compound 3 obtained in 20% yield. A
second palladium cross-coupling reaction using Pd(PPh3)4 was used
to introduce the second substituted aromatic moiety. Using this
procedure with different boronic acids new disubstituted exo-glycals
5, 7, 8 and 9 were obtained in good to excellent yields as single
stereoisomer. Subsequent removal of the hydroxyl protecting group
of 5bytreatmentwithTBAFaffordedthecompound6bearingthetwo
aromatic moieties of phenstatin (3,4,5-trimethoxyphenyl and iso-
vanillin ring). Catalytic hydrogenation on Pd/C of compound 9 gave
the amino derivatives 10 in quantitative yield.
OMe
Phenstatin
X = CH2 Iso-combretastatin isoCA-4
X = CF2 F2isoCA-4
OMe
X = O
Combretastatin CA-4
Chart 1. Representative cytotoxic lignans and derivatives.
In phenstatins, the benzophenone structure maintains the appro-
priate spatial disposition of the two aryl moieties. Several successful
structural modifications of phenstatin have been reported [20]. For
example, aromatic, heteroaromatic or bicyclic systems have been
used to replace the B-ring on the benzophenone scaffold, leading to
naphtylphenstatins, indolphenstatins, thienylphenstatins, and
others [20e26]. The replacement of the carbonyl group by one-atom
or two-atombridgesthatmaintainthe structureof the moleculesled
to potent derivatives [27e31]. The introduction of an amino group at
the C-3 position in B ring of benzophenone strongly enhanced the
cytotoxicity [10]. Iso-combretastatin A based on a 1,1-diarylethylene
scaffold was recently developed and constitutes a novel class of
potent antiproliferative agents clearly demonstrating that the
carbonyl group is not essential for antitubulin activity [32,33] (Chart
1). Many of the SAR requirements described for combretastatins
seemed valuable for phenstatin analogues, in particular the pres-
ence of the 3,4,5-trimethoxyphenyl ring and a non-coplanar rela-
tionship between the two aromatic rings.
We recently described a facile synthesis of biarylolefins from
dibromo-exo-glycals using palladium catalyzed cross-coupling
reactions [34]. These compounds are closely related to the above-
described CA-4 analogues and it seemed interesting to investigate
this new class of compounds (Chart 2, type I compounds, 1,1-diaryl-
exo-glycal) from a biological point of view. The use of carbohydrate
scaffold would provide an easy way to modulate the biological
activity and the water solubility of the resulting compounds by
deprotection of the carbohydrate hydroxyl groups while retaining
the cisoid arrangement of the two aromatic rings which seems
important for the biological activity of CA4. Moreover, the presence
of the sugar ring oxygen would confer some electron-rich character
to the double bond and would modify the electronic distribution of
the system. This methodology also allows the synthesis of more
simple, yet unknown compounds (Chart 2, type II compounds,
1,1-diaryl-2,2-dihydroxymethylethylene) close to isotaxiresinol
a recently investigated lignan which proved more active than taxol
and doxorubicin against colon adenocarcinoma (Chart 1)[35].
Compound 12, a geometrical isomer of 6, was obtained from (E)-
isomer
4 by cross-coupling reaction with 3-[tert-butyldime-
thylsilyl]oxy]-4-methoxyphenyl]boronic acid [37]. Finally, the diols
13e19 were obtained by selective removal of the exocyclic iso-
propylidene acetal on compounds 3, 6e10 and 12 under acidic
conditions.
Simplified analogues, type II compounds (Chart 2), were prepared
using the same methodology. The hydroxyl groups of dihydroxyac-
etone 20 were protected as acetates by treatment with acetic anhy-
dride in pyridine (Scheme 2). Treatment of compound 21 with
triphenylphosphine and carbon tetrabromide in dichloromethane at
0 ꢀC led to the yet unknown olefin 22 in 85% yield. Cross-coupling
reaction of 22 with 3,4,5-trimethoxyphenyl boronic acid was first
investigated using one equivalent of boronic acid in the presence of
TFP/PdCl2(PPh3)2 in dimethoxyethane. The monosubstituted
compound 23 was obtained in less than 20% yield, the starting
dibromo compound 22 being recovered. The use of 1.5 eq of 3,4,
5-trimethoxyphenyl boronic acid allowed the formation of mono-
substituted 23 and disubstituted 24 compounds isolated in 35% and
50% yields respectively. This is likely due to a faster coupling reac-
tion on the monosubstituted derivative 23 as compared to 22.
Commercially available boronic acids and previously synthesized
3-[[tert-butyldimethylsilyl]oxy]-4-methoxyphenylboronic acid were
used for the preparation of disubstituted compounds 25e28.
Removal of tert-butyldimethylsilyl group by treatment with TBAF led
to the phenolic derivative 26. Final deprotection of the hydroxyl
groups was achieved by treatment by a catalytic amount of sodium
methoxide in methanol affording the diols 29e32.
3. Biological evaluation
2. Chemistry
3.1. Cell culture and cell proliferation assay
Disubstituted exo-glycals [36], i.e.; type I compounds, were
designed from 1,1-dibromo-
D-gulo-hept-1-enitol 1 readily available
The cytotoxicity of some compounds was examined in the
from the corresponding protected
D
-gulono- -lactone (Scheme 1).
g
National Cancer Institute’s Developmental Therapeutics Program

3572
A. Novoa et al. / European Journal of Medicinal Chemistry 46 (2011) 3570e3580
OMe
MeO
OMe
OMe
MeO
OMe
Br
O
Br
O
O
O
O
OMe
OMe
O
O
O
O
O
O
O
O
(a)
Br
Br
OMe
OMe
+
+
O
O
O
O
O
O
O
OMe
MeO
1
2 (Z)-isomer
3
4 (E)-isomer
(e)
(b)
OMe
OMe
MeO
Ar
O
O
Ar
O
O
OMe
OMe
O
O
OR1
OMe
(f)
11 R1 = TBDMS
12 R1 = H
OMe
O
O
O
O
(c)
OMe
5
8, 16
OTBDMS
5
(c)
(d)
6
OMe
7,8
9
(f)
(f)
OMe
OMe
9, 17
OH
O
6, 14
NO2
NH2
MeO
O
10
Ar
HO
O
OMe
OMe
HO
O
HO
HO
O
O
OH
OMe
7, 15
10, 18
OMe
OMe
O
O
13-18
OMe
OMe
13
19
Scheme 1. (a) 3,4,5-Trimethoxyphenylboronic acid (1.3 eq), K2CO3 (2 M), trifuryl phosphine, PdCl2(PPh3)2, DME, 85 ꢀC, 24 h; (b) ArB(OH)2 (2 eq), K2CO3 (2 M), Pd(PPh3)4, DME,
85 ꢀC, 24 h; (c) Bu4NF, CH2Cl2, 0 ꢀC; (d) H2, Pd/C, EtOAc; (e) 3-[tert-butyldimethylsilyl]oxy]-4-methoxyphenyl] boronic acid (2 eq), K2CO3 (2 M), Pd(PPh3)4, DME, 85ꢀ, 24 h; (f) TFA,
EtOH, H2O.
antitumor screening [38]. From our compound library, eleven
compounds were selected for pre-screening by the Biological
Evaluation Committee and granted NCS codes. These compounds
were screened for anti-proliferative activity at a single high dose
listed in Table 1. To be concise, only significant results restricted to
a few cell lines are listed in Table 2.
Concerning the broad spectrum antitumor activity, the results
revealed that the three active compounds 6, 8 and 12 showed
effective growth inhibition GI50 (MG-MID) values of 7.41, 5.50 and
(10 mM) in the full NCI60 cell panel derived from nine cancer cell
types (leukemia, non-small cell lung cancer, colon cancer, CNS
cancer, melanoma, ovarian cancer, renal cancer, prostate cancer and
breast cancer). A sulforhodamine B (SRB) protein assay was used to
estimate cell viability or growth. Results for each compound were
reported as the growth percentage of the treated cells when
compared to that of the untreated control cells. Selected experi-
mental data and the averaged values of mean graph mid-point
(MG-MID) are reported in Supplementary data. Compounds 6, 8
and 12 which exhibited significant growth inhibition were then
selected for re-screening and evaluated against the 60 cell panel at
10.72
m
M respectively, besides cytostatic activity TGI (MG-MID)
values of 49.00, 33.90, 55.00
mM respectively (Table 1). The GI50
(MG-MID) and TGI (MG-MID) of these three compounds were in
the same range of activity against most of the subpanel tumor cell
lines. It should be noted that compounds 6 and 12 were found to be
slightly less active than 8. Compound 6 revealed a relative effec-
tiveness on the leukemia, CNS and prostate subpanel (GI50 (MG-
MID) 4.70, 4.41, 4.20
the highest activity was displayed for CNS subpanel with a TGI
(MG-MID) value of 38.3 M. Excepted against leukemia, compound
mM respectively). Regarding cytostatic effect,
m
five concentration levels (0.01
m
Me100
m
M).
8 showed equipotent activity toward growth inhibition with almost
all the tested subpanels as revealed by GI50 (MG-MID). According to
the TGI (MG-MID) values, this compound exhibited the best cyto-
static activity against CNS cancer, melanoma and breast cancer
Three response parameters (GI50, TGI and LC50) were calculated
for each cell line. The GI50 value (growth inhibition activity)
corresponds to the concentration of compound causing 50%
decrease of cell growth while the TGI (total growth inhibition)
value (cytostatic activity) is the concentration of the compound
resulting in complete growth inhibition. The LC50 value (cytotoxic
activity) is the concentration of compound leading to 50% loss of
initial cells. Table 1 describes the antiproliferative activity using
GI50 and TGI of selected compounds as mean concentration value
(MG-MID) and best activity for each subpanel (IeIX). The GI50 and
TGI full-panel mean graph mid-point (MG-MID) have also been
subpanels (28.7, 29.3 and 34.5
mM respectively). With GI50 (MG-
MID) and TGI (MG-MID) values of 10.72 and 55.0
mM, compound 12
appeared as the less powerful compound. Nevertheless, the results
showed effective growth inhibition against leukemia, CNS cancer
and prostate cancer subpanels (GI50 (MG-MID) 8.88, 7.75 and
4.02
cytostatic activity, the best effect concerning the melanoma
subpanel (TGI (MG-MID) 51.2 M). These compounds showed
mM respectively). Compound 12 was found to have a moderate
m

A. Novoa et al. / European Journal of Medicinal Chemistry 46 (2011) 3570e3580
3573
OMe
OMe
OR
OMe
MeO
MeO
OMe
OMe
OMe
OMe
Br
AcO
Br
Br
AcO
O
(b)
(c)
+
RO
RO
OR
OAc
OAc
24 R = Ac
29 R = H
20 R = H
22
23
(a)
(f)
21 R = Ac
(d)
OMe
OR
Ar
OMe
OMe
Ar
RO
OMe
OMe
OH
OTBDMS
25
26, 30
25
26
R = Ac
R = Ac
(e)
(f)
O
O
27-28 R = Ac
30
R = H
31-32 R = H
(f)
27, 31
28, 32
Scheme 2. (a) Ac2O, DMAP, pyridine; (b) CBr4, PPh3, CH2Cl2; (c) 3,4,5-trimethoxyphenylboronic acid (1.5 eq), K2CO3 (2 M), trifuryl phosphine, PdCl2(PPh3)2, DME, 85 ꢀC;
(d) ArB(OH)2 (2 eq), K2CO3 (2 M), Pd(PPh3)4, DME, 85 ꢀC, 24 h; (e) Bu4NF, CH2Cl2, 0 ꢀC; (f) Na, MeOH, 0 ꢀC.
a broad spectrum of antitumor activity as well a variable pattern of
sensitivity against some individual cell lines (Table 2). With regard
to the sensitivity against some selected individual cell lines
(Table 2), compounds 6, 8 and 12 showed high activity against non-
small cell lung cancer (NSCLC) NCI-H522 cell line (GI50 2.73, 2.26
two concentrations (10ꢁ5 and 10ꢁ6 M). Among the ten compounds
tested in a preliminary assay (see Supplementary data), those
showing some cytotoxicity were further taken for determining their
IC50 values (Table 3). Three of them, exo-glycal derivative 19 and 1,1-
diarylethylene derivatives 26 and 32 displayed a potent activity in
the micromolar and submicromolar range, IC50 values spanning
and 2.95 and TGI 9.08, 6.54 and 7.54
activity against colon cancer HCT-116 cell line (GI50 2.51, TGI 5.70,
LC50 20.6 M) was detected for compound 8 and the analogue 12
also showed a particular effectiveness against colon cancer KM12
cell line (GI50 4.10, TGI 17.8, LC50 72.9 M). Compound 6 exhibited
the highest activity against CNS cancer SNB-75 cell line (GI50 1.80,
TGI 6.31 and LC50 53.5 M). Compound 12 was nearly equipotent
against CNS cancer SF-539 cell line with almost the same range of
activity (GI50 3.15, TGI 8.75 and LC50 86.0 M). Melanoma cell lines
mM respectively). A noticeable
from 70 nM to 6.84 mM. The screening results revealed that these
m
compounds inhibited the growth of all examined tumor cell lines in
the same range and that this effect is cell type dependent. The K562
cells line was the most sensitive to 19 and 26, 32 (IC50 0.45, 0.17,
0.18 mM respectively). It is noteworthy that compound 26 revealed
a potent antiproliferative activity on the SK-OV3 cell line (ovarian
m
m
cancer) (IC50 ¼ 70 nM), compounds 32 and 19 being less potent on
m
this cell line (IC50 SK-OV3 ¼ 6.51 and 6.84
mM respectively).
proved sensitive toward compounds 6, 8 and 12. In particular, the
highest sensitivity is displayed by SK-MEL-5 cell line with GI50, TGI
3.2. Inhibition of tubulin polymerization
and LC50 values in range of 2.69e4.12, 11.2e15.4 and 47.1e51.4 mM,
respectively. The effectiveness of compound 8 against ovarian
cancer SK-MEL-2 should be pointed out (GI50 1.53, TGI 4.12 and LC50
The synthesized derivatives were evaluated for their ability to
inhibit tubulin polymerization at a single concentration, phenstatin
and deoxypodophyllotoxin being used as positive controls. All the
type I and type II compounds (Chart 2) exhibited a modest or poor
tubulin polymerization inhibitory activity (see Supplementary
data).
24.1
tivity against ovarian cancer OVCAR-3 cell line with GI50 value of
2.38, 2.39 and 4.84 M values, respectively. The compounds 6, 8
mM). Compounds 6, 8 and 12 also revealed an obvious sensi-
m
and 12 proved to be moderately sensitive toward renal and breast
cancer cell lines, except for renal RXF 393 cell line with growth
Molecular modelling was used to explain this absence of inter-
action with tubulin. A docking model was constructed starting from
the reported X-ray structure of tubulin cocrystallized with
a colchicine derivative, N-deacetyl-N-(2-mercaptoacetyl)colchicine
(DAMA-colchicine, PDB entry 1SA0) [39,40]. DAMA-colchicine was
first docked to reproduce the co-crystal complex. The docking
inhibition concentrations less than 2.6
2.35 M respectively).
mM (GI50 2.12, 2.54 and
m
Other compounds not selected in the above-described screening
were also evaluated. The cytotoxicity of type II compounds and some
1,1-diaryl-exo-glycals (type I) was evaluated on a cell line panel at
Table 1
Mean concentration values (MG-MID) on subpanel and 60 tumor cell lines (
m
M).
Compounds/NCS code Subpanel tumor cell linesa,b
MG-MIDc
I
II
III
7.88/3.99
IV
V
VI
VII
VIII
4.20/3.29
61.5/5.34 66.90/5.89 66.50/13.3
5.00/2.39 5.24/2.30 5.02/4.33
IX
9.36/2.83
62.10/7.88
4.17/3.22
6/753427/1
8/750870/1
12/750873/1
GI50
TGI
4.70/3.22
15.31/2.73
4.41/1.80 26.50/1.48 22.38/2.38 17.77/2.12
7.41
49.00
5.50
33.90
10.72
87.20/23.30 68.00/9.08 >100/>100 38.3/6.31
4.68/2.26 15.24/2.51 5.47/2.49
>100/>100 57.80/6.54 74.60/5.70
68.9/3.70
3.96/1.53
GI50 60.09/4.02
TGI
GI50
TGI
28.7/7.44 29.30/4.12 49.50/8.02 61.50/4.47 >100/>100 34.50/8.10
7.75/3.15 12.29/2.15 24.47/3.11 18.84/2.35 4.02/3.93 19.78/6.97
8.88/4.82
22.62/2.95 16.72/4.06
81.80/34.0
72.80/7.54 76.60/17.80 58.2/8.75 51.20/5.54 77.70/8.85 71.60/5.07 >100/>100 76.30/50.10 55.00
a
Subpanel tumor cell lines identification. I, leukemia; II, non-small cell lung cancer; III, colon cancer; IV, CNS cancer; V, melanoma; VI, ovarian cancer; VII, renal cancer; VIII,
prostate cancer, IX, breast cancer.
b
Growth inhibition and cytostatic mean concentration values (MG-MID) on subpanel tumor cell lines/best activity value on subpanel tumor cell lines (
Mean graph mid-point MG-MID: the average sensitivity of all cell lines toward the tested agent.
mM).
c

3574
A. Novoa et al. / European Journal of Medicinal Chemistry 46 (2011) 3570e3580
Table 2
Selected biological activities on NCI-tumor cell lines (
mM).
Cell lines
Compounds
6
8
12
GI50
TGI
LC50
GI50
TGI
LC50
GI50
TGI
LC50
NSCLCa
HOP-62
HOP-92
NCI-H522
HCT-116
5.93
5.45
2.73
4.14
25.50
31.80
9.08
/
96.80
9.78
31.40
13.30
6.54
99.80
/
/
10.90
4.00
2.95
NT
34.70
38.30
7.54
NT
/
/
/
/
/
2.67
2.26
2.51
85.60
NT
5.70
20.6
Colon Cancer
CNS Cancer
KM12
SF-539
SNB-19
3.99
2.59
4.28
/
/
/
/
6.00
14.30
2.49
/
/
/
/
4.10
3.15
5.33
17.80
8.75
/
72.90
86.00
/
6.63
/
79.00
6.08
SNB-75
U251
LOX IMVI
MALME-3M
1.80
3.84
5.18
5.46
6.31
53.50
2.89
3.12
3.39
2.07
7.44
11.80
19.80
5.59
/
/
/
/
6.36
5.00
5.93
9.96
/
/
/
48.20
/
/
/
/
/
43.10
22.80
/
74.10
/
Melanoma
MDA-MB-435
SK-MEL-2
SK-MEL-5
1.48
57.90
2.69
3.70
/
11.20
/
/
3.61
1.53
4.07
17.50
4.12
15.40
/
2.15
3.26
4.12
5.54
8.43
15.00
/
/
24.10
51.40
49.30
47.10
Ovarian cancer
OVCAR-3
NCI/ADR-RES
786-0
2.38
2.52
21.20
5.34
6.91
60.60
22.9
/
/
2.39
4.26
2.30
5.67
22.30
4.47
21.70
/
8.60
4.84
3.11
NT
43.30
8.85
NT
/
/
NT
Renal cancer
Breast cancer
RXF 393
2.12
5.89
29.80
2.54
5.99
32.60
2.35
5.07
14.20
MDA-MB-231/ATCC
HS 578T
4.58
2.83
30.70
7.88
/
/
6.28
4.44
69.70
53.40
/
/
9.71
6.97
50.10
67.90
/
/
a
NSCLC ¼ Non-small cell lung cancer; NT: not tested; /: not active, value >100
mM.
mode of the combretastatin A-4 (CA-4), as the lead of numerous
compounds interacting with tubulin was examined [16]. The
docking mode of phenstatin [19] and phenstatin-like compounds,
into the tubulin site was also checked. The best docked poses of
CA-4 and phenstatin did nicely superimpose on the DAMA-
colchicine X-ray crystal structure (see Supplementary data).
Initial conformations of our biaryolefins systems were built and the
docking protocol was applied to compound 30. The best poses
obtained for phenstatin, isoCA-4, F2isoCA-4 and 30 showed
a common binding mode corresponding to that previously deter-
mined. Docking experiments showed that 30 could assume the
same binding mode as CA-4 or phenstatin.
The 3D energy minimized conformation of 6 was superimposed
by hand on docked phenstatin into the colchicine binding site of
tubulin (Fig. 1, red surface). Protein flexibility is supposed to allow
such a conformation. Docking of compound 6 was next carried out.
As seen from Fig. 1, poses fill in the open cavity but in different
orientations. Moreover they are disordered and do not bind in
a coherent mode showing two, one or no hydrogen bond(s) with
4. Discussion
The screening results revealed that four 1,1-diaryl-exo-glycals
(type I compounds) and two derivatives belonging to type II
showed a significant antiproliferative effect in the micromolar and
submicromolar range (Chart 3). From these results it appears that
a set of three compounds 6, 8 and 12 exhibit growth inhibition with
GI50 in the 7e11
mM range associated with cytostatic activities in
the 34e55 M range. All three compounds are active against renal
m
cancer (RXF-393). It is to note that compounds 6 and 12 are
geometric isomers which behave similarly on the RXF-393 cell line
although 12 is less active than 6 against other cell lines. It is worth
to mention that these most active compounds have the same
aromatic substituents than CA-4 or phenstatin. Moreover
compound 8 embodies a napthalene ring which has been shown to
enhance the activity of combretastatin [17,18] or phenstatin
analogues [23]. Three other compounds 19, 26 and 32 revealed
interesting growth inhibition properties on all examined tumor cell
lines in the same range with IC50 below 1 mM but this effect is cell
Cys
b
241 and or Thr
b
353. Compound 6 poses are essentially
type dependent. Examination of Chart 3 clearly shows that the
highest powerful inhibition of cell proliferation was obtained with
yet unknown 1,1-diarylethylene derivatives 26 and 32 (type II
compounds), the 1,1-diaryl-exo-glycal 19 being slightly less active.
However, the latter is the most active compound of the type I class
of compounds. This could be a consequence of the presence of a diol
function instead of the more bulky 5,6-dioxolane ring present in
compounds 6, 8 and 12. It should also be noted that the dihy-
droxylated compound 30, is much less active than its acetylated
derivative 26 whereas 32 is more active than its acetylated deriv-
ative 28 suggesting that, in these compounds, the hydroxyl groups
do not play a key role. Compounds 26 and 32 revealed potent
activity toward SK-OV3 and K563 cell lines. Finally the most active
compound was the biarylolefin 26 with an IC50 ¼ 70 nM against
SK-OV3, ovarian cancer.
oriented toward the solvent, losing the required interactions of
tubulin inhibitors in the binding site.
Table 3
Inhibition of cancer cell proliferation (IC50
, mM).
Cells lines
Compounds
19
26
32
Colon cancer
Breast cancer
Prostate cancer
Glioblastoma
Ovarian cancer
HCT15
MDA231
PC3
SF268
SK-OV3
OVCAR8
HL60
0.63
0.61
0.69
1.10
6.51
2.97
0.82
1.49
0.45
0.68
1.03
0.27
0.97
0.07
1.90
0.68
2.72
0.18
0.39
0.81
0.76
1.25
6.84
4.47
0.78
1.68
0.17
Lymphoblastoma
An important result is the almost complete lack of inhibition of
tubulin polymerization of all compounds of this study. Thus the
observed inhibition of cell proliferation is not due to interaction
HL60R
K562

A. Novoa et al. / European Journal of Medicinal Chemistry 46 (2011) 3570e3580
3575
with tubulin. However,
a
COMPARE analysis [41,42] which
OMe
measures the correlation between two compounds with respect to
their differential antiproliferative activity has been carried out on
the most active compounds but failed to provide any significant
match to a known compound and any associated mechanism of
action.
HO
O
O
O
O
O
O
OMe
OMe
OMe
OMe
Molecular modelling studies revealed that interaction of our
1,1-diarylolefins like 30 with the colchicine binding site to tubulin is
possible. As 30 is not a significant inhibitor of tubulin, this suggests
that the presence of the two hydroxyl groups may induce other
interactions (for example hydrogen bonding) not taken into
account in the docking model, which do not allow a correct
arrangement of 30 in the colchicine site of tubulin. The binding
model of 30 revealed the presence of an open cavity above the two
hydroxyl groups in agreement with previous reports [16,43]. It
should be noted that active iso-combretastatin analogues mostly
include on the double bond a rather flat moiety which could be
accommodated in this putative open space. In our case, the sugar
moiety of compound like 6, is too voluminous for this open space
likely because of the presence of one or two dioxolane rings on the
sugar ring. This precludes our compounds to adopt the correct
orientation of the aromatic rings in the colchicine binding site.
Thus, the introduction of too bulky susbtituents on iso-com-
bretastatin double bond clearly abolishes the antitubulin activity
and the mode of action of our compounds should be different. A
good example of changing the mode of action by introduction of
a rather bulky substituent on a tubulin polymerization inhibitor can
be found in the podophyllotoxin series. Thus, podophyllotoxin is
a tubulin polymerization inhibitor while etoposide, its glycosylated
is a topoisomerase II inhibitor [44e46].
O
O
O
O
MeO
MeO
8
IC50 = 1.53 µM (SK-MEL-2)
IC50 = 2.39 µM (OVCAR-3)
IC50 = 2.54 µM (RXF 393)
6
IC50 = 1.80 µM (SNB-75)
IC50 = 2.38 µM (OVCAR-3)
IC50 = 2.12 µM (RXF 393)
OMe
MeO
OMe
MeO
OMe
OMe
O
HO
O
O
HO
O
OH
OH
O
O
O
O
OMe
OMe
12 IC50 = 2.35 µM (RXF 393)
19 IC50 = 0.45 µM (K562)
IC50 = 0.63 µM (HCT15)
IC50 = 0.45 µM (MDA231)
OH
OMe
OAc
OMe
OMe
MeO
OMe
OMe
5. Conclusion
OMe
We have discovered a new class of antiproliferative agents based
on a biarylolefin motif. Two types of derivatives have been designed
on carbohydrate and dihydroxymethylethylene scaffolds (type I
and II compounds) and were synthesized by a straightforward
approach using sequential palladium cross-coupling reaction. This
flexible strategy allows the sequential addition of different
aromatic rings at a late stage of the synthesis in contrast to phen-
statin analogue syntheses, which start from benzophenone deriv-
atives. Thirty-five new compounds have been prepared among
which six displayed a broad spectrum of antitumor activity in the
micromolar or submicromolar range. A variable pattern of sensi-
tivity against some individual cell lines was observed and the most
potent compound 26 revealed a high activity toward an ovarian
cancer cell line (IC50 ¼ 70 nM, SK-OV3). None of the synthesized
compounds exhibited tubulin polymerization inhibitory activity,
a tentative explanation suggested by molecular modelling could be
a steric hindrance induced by the double bond substituents or by
other interactions due to the presence of free hydroxyl or ester
groups on these compounds. These results suggested that the cell
proliferation inhibition activity of our compounds could be attrib-
uted to another mechanism of action. More biological experiments
are underway to determine a possible mechanism, which would
give some information on how to improve the efficiency of our
leads by chemical structure modulation.
AcO
HO
OH
26 IC50 = 70 nM (SK-OV3)
IC50 = 0.18 µM (K562)
IC50 = 0.27 µM (PC3)
32 IC50 = 0.17 µM (K562)
IC50 = 0.39 µM (HCT15)
Chart 3. Structures and selected activities for the most potent compounds.
6. Experimental section
6.1. Chemistry
Fig. 1. The ten best docked poses of compound 6 superimposed to the “by hand”
docked conformation of 6 (red surface). Poses are represented without hydrogen
atoms for clarity. (For interpretation of the references to colour in this figure legend,
the reader is referred to the web version of this article.)
6.1.1. Material and methods
All reagents were commercially available and were used
without further purification unless otherwise indicated. DME was

3576
A. Novoa et al. / European Journal of Medicinal Chemistry 46 (2011) 3570e3580
ꢀ
stored on activated molecular sieves (4 A) and dichloromethane
was dried by distillation from calcium hydride prior to use. TLC
analyses were performed using standard procedures on Kieselgel
60 F254 plates (Merck). Compounds were visualized using UV light
(254 nm) and 30% methanolic H2SO4/heat as developing agent.
Column chromatography was performed on silica gel SI 60
6.1.2.2. 2,5-Anhydro-1.1-di(3,4,5-trimethoxyphenyl)-1-deoxy-3,4:6,
7-di-O-isopropylidene-
yellow solid; Rf ¼ 0.20 (Hexane/EtOAc 7/3);
(c ¼ 0.93, CHCl3); IR (film)
cmꢁ1; 1H NMR (CDCl3, 250 MHz)
D
-gulo-hept-1-enitol (3). Yield: 20% as a pale
20
[
a]
¼ ꢁ225ꢀ
D
n
2987, 2933, 1645, 1580, 1508 (OMe)
d
1.28 (s, 3H, C(CH3)2), 1.35 (s, 3H,
C(CH3)2), 1.41 (s, 3H, C(CH3)2), 1.48 (s, 3H, C(CH3)2), 3.75 (s, 6H, 2
OCH3), 3.76 (m, 1H, H-7), 3.79 (s, 6H, 2 OCH3), 3.83 (s, 3H, OCH3),
3.87 (s, 3H, OCH3), 4.03 (dd, 1H, H-5, J ¼ 8.5 Hz, J ¼ 4.5 Hz), 4.22 (dd,
1H, H-70, J ¼ 8.5 Hz, J ¼ 7.0 Hz), 4.47 (dt, 1H, H-6, J ¼ 8.5 Hz,
J ¼ 7.0 Hz), 4.70 (dd, 1H, H-4, J ¼ 6.0 Hz, J ¼ 4.5 Hz), 4.94 (d, 1H, H-3,
J ¼ 6.0 Hz), 6.63 (s, 2H, HeAr), 6.72 (s, 2H, HeAr); 13C NMR (CDCl3,
(63e200 mm) (Merck). FTIR spectra were recorded on a Per-
kineElmer spectrum 1000 on film windows. Melting points were
determined in capillary tubes and are uncorrected. Optical rota-
tions were measured on a PerkineElmer 141 polarimeter. 1H and
13C NMR spectra were recorded on a Bruker spectrometer DPX250
(250 MHz and 62.9 MHz, respectively) and DRX400 (400 MHz and
100.6 MHz, respectively). For complete assignment of 1H and 13C
signals, two-dimensional 1H, 1H COSY and 1H, 13C correlation
62.9 MHz)
d 25.1 (C(CH3)2), 25.1 (C(CH3)2), 26.3 (C(CH3)2), 26.8
(C(CH3)2), 55.9 (2 OCH3), 56.0 (2 OCH3), 60.8 (OCH3), 60.8 (OCH3),
65.8 (C-7), 75.5 (C-6), 78.3 (C-4), 80.5 (C-3), 85.2 (C-5), 107.3 (2
CeAr), 108.2 (2 CeAr), 109.7 (C(CH3)2), 112.6 (C(CH3)2), 120.5 (C-1),
133.8 (CeAr), 134.2 (CeAr), 136.9 (CeAr), 137.0 (CeAr), 150.6 (C-2),
152.4 (2 CeAr), 152.5 (2 CeAr); HRMS (ESI) calculated for
C31H40NaO11: 611.2463, [M þ Na]þ found 611.2471.
spectra were recorded. Chemical shifts (d) are given in ppm relative
to the solvent residual peak. The following abbreviations are used
for multiplicity of NMR signals: s ¼ singlet, d ¼ doublet, t ¼ triplet,
q ¼ quadruplet, m ¼ multiplet, br ¼ broad signal. For the sake of
simplicity, nomenclature is inspired from that of carbohydrates.
HRMS spectra were recorded on a Bruker MicroTOFQ apparatus.
Microwave syntheses were performed in a mono-mode oven, the
temperature being controlled with a standard external IR-sensor.
Analytical high-performance liquid chromatography (HPLC) was
performed using a Spectra P1000XR apparatus with an Interchim
UP50DB-25QS C18 reverse phase 250 mm ꢂ 4.6 mm column.
Samples were detected using an absorbance detector UV150
(wavelength of 254 nm). The flow rate of the mobile phase (water/
acetonitrile 3/7) was of 1 mL/min. The purity of the tested
compounds was determined by HPLC and unless otherwise
stated, the purity level was ꢃ95%. Details of the synthesis and
characterizations of intermediate compounds are given in the
Supplementary data.
6.1.2.3. (E)-2,5-Anhydro-1-bromo-1-(3,4,5-trimethoxyphenyl)-1-deoxy-
3,4:6,7-di-O-isopropylidene- -gulo-hept-1-enitol (4). Yield: 7% as
D
a pale yellow solid; Rf ¼ 0.44 (Hexane/EtOAc 7/3); 1H NMR (CDCl3,
250 MHz)
d 1.36 (s, 3H, C(CH3)2), 1.42 (s, 3H, C(CH3)2), 1.43 (s, 3H,
C(CH3)2), 1.50 (s, 3H, C(CH3)2), 3.79 (dd, 1H, H-7, J ¼ 8.5 Hz,
J ¼ 6.5 Hz), 3.86 (s, 3H, OCH3), 3.88 (s, 6H, 2 OCH3), 4.18 (dd,1H, H-5,
J ¼ 8.5 Hz, J ¼ 4.0 Hz), 4.23 (dd, 1H, H-70, J ¼ 8.5 Hz, J ¼ 6.5 Hz), 4.49
(dt, 1H, H-6, J ¼ 8.5 Hz, J ¼ 6.5 Hz), 4.74 (dd, 1H, H-4, J ¼ 6.0 Hz,
J ¼ 4.0 Hz), 5.51 (d, 1H, H-3, J ¼ 6.0 Hz), 7.16 (s, 2H, HeAr); 13C NMR
(CDCl3, 62.9 MHz)
d 25.2 (C(CH3)2), 25.7 (C(CH3)2), 26.5 (C(CH3)2),
26.8 (C(CH3)2), 55.9 (2 OCH3), 60.8 (OCH3), 65.8 (C-7), 75.4 (C-6),
78.0 (C-4), 82.9 (C-3), 86.1 (C-5), 102.1 (C-1), 106.3 (2 CeAr), 109.8
(C(CH3)2), 113.6 (C(CH3)2), 131.1 (CeAr), 137.4 (CeAr), 152.0 (C-2),
152.4 (2 CeAr); HRMS (ESI) calculated for C22H29BrNaO8: 523.0936,
[M þ Na]þ found 523.0938.
6.1.2. General procedure for first cross-coupling reaction
In a dry round-bottomed flask equipped with a condenser
flushed with argon, 5 mL of dry DME, dibromo-exo-glycal 1
(150 mg, 0.362 mmol, 1 eq), 3,4,5-trimethoxyphenylboronic acid
(100 mg, 0.47 mmol, 1.3 eq), 2 M K2CO3 aqueous solution (2.6 eq)
were stirred for 10 min. Tris-(2-furyl)phosphine (26 mg,
0.109 mmol, 0.3 eq) and PdCl2(PPh3)2 (13 mg, 0.018 mmol, 0.05 eq)
were added as solids. The reaction mixture was stirred and heated
at 85 ꢀC during 24 h. The reaction was then allowed to cool to room
temperature and was concentrated under reduced pressure. The
residue was diluted with ethyl acetate (50 mL) and washed with
NH4Cl (2 ꢂ 5 mL). The dried organic layer was evaporated under
reduced pressure and the residue was purified by chromatography
(n-hexane/EtOAc).
6.1.3. General procedure for second cross-coupling reaction
In a dry round-bottomed flask equipped with a condenser
flushed with argon, 5 mL of dry DME, bromo-exo-glycals 2 or 4
(0.362 mmol), boronic acid (2 eq), 2 M K2CO3 aqueous solution
(4 eq) were stirred for 10 min. Solid Pd(PPh3)4 (20 mg, 0.018 mmol,
0.05 eq) was then added. The reaction mixture was stirred and
heated at 85 ꢀC during 24 h. The reaction was then allowed to cool
to room temperature and was evaporated to dryness under reduced
pressure. The residue was diluted with ethyl acetate (50 mL) and
washed with NH4Cl (2 ꢂ 5 mL). The dried organic layer was evap-
orated under reduced pressure and the residue was purified by
chromatography (n-hexane/EtOAc).
6.1.2.1. (Z)-2,5-Anhydro-1-bromo-1-(3,4,5-trimethoxyphenyl)-1-deoxy-
6.1.3.1. (Z)-2,5-Anhydro-1-(3-tertbutyldimethylsilyloxy-4-methoxy-
phenyl)-1-(3,4,5-trimethoxyphenyl)-1-deoxy-3,4:6,7-di-O-isopropy-
3,4:6,7-di-O-isopropylidene-
a pale yellow solid; Rf ¼ 0.33 (Hexane/EtOAc 7/3); [
(c ¼ 0.95, CHCl3); IR (film)
NMR (CDCl3, 400 MHz)
D
-gulo-hept-1-enitol (2). Yield: 47% as
20
a]
¼ ꢁ228ꢀ
lidene-
Rf ¼ 0.36 (Hexane/EtOAc 7/3); [
(film)
(CDCl3, 400 MHz) d 0.16 (s, 3H, Si(CH3)2), 0.17 (s, 3H, Si(CH3)2), 0.99
D
-gulo-hept-1-enitol (5). Yield: 70% as a pale yellow solid;
D
20
n
2987, 2933, 1655, 1580, 1504 cmꢁ1; 1H
a
]
¼ ꢁ184ꢀ (c ¼ 0.57, CHCl3); IR
D
d
1.33 (s, 3H, C(CH3)2), 1.44 (s, 3H, C(CH3)2),
n ;
2987, 2954, 2927, 2851, 1645, 1583, 1508 cmꢁ1 1H NMR
1.51 (s, 3H, C(CH3)2), 1.55 (s, 3H, C(CH3)2), 3.77 (dd, 1H, H-7,
J ¼ 8.5 Hz, J ¼ 7.0 Hz), 3.88 (s, 9H, 3 OCH3), 4.16 (dd, 1H, H-5,
J ¼ 8.0 Hz, J ¼ 5.0 Hz), 4.26 (dd, 1H, H-70, J ¼ 8.5 Hz, J ¼ 7.0 Hz), 4.51
(dt, 1H, H-6, J ¼ 8.0 Hz, J ¼ 7.0 Hz), 4.82 (dd, 1H, H-4, J ¼ 6.0 Hz,
J ¼ 5.0 Hz), 4.99 (d, 1H, H-3, J ¼ 6.0 Hz), 6.93 (s, 2H, HeAr); 13C NMR
(s, 9H, SiC(CH3)3), 1.29 (s, 3H, C(CH3)2), 1.40 (s, 3H, C(CH3)2), 1.49 (s,
3H, C(CH3)2), 1.51 (s, 3H, C(CH3)2), 3.78 (m, 1H, H-7), 3.80 (s, 9H, 3
OCH3), 3.89 (s, 3H, OCH3), 4.12 (dd, 1H, H-5, J ¼ 8.5 Hz, J ¼ 5.0 Hz),
4.26 (dd, 1H, H-70, J ¼ 8.5 Hz, J ¼ 7.0 Hz), 4.47 (dt, 1H, H-6, J ¼ 8.5 Hz,
J ¼ 7.0 Hz), 4.71 (dd, 1H, H-4, J ¼ 6.0 Hz, J ¼ 5.0 Hz), 4.96 (d, 1H, H-3,
J ¼ 6.0 Hz), 6.60 (s, 2H, HeAr), 6.77 (d, 1H, HeAr, J ¼ 8.5 Hz), 6.90
(dd, 1H, HeAr, J ¼ 8.5 Hz, J ¼ 2.0 Hz), 7.09 (d, 1H, HeAr, J ¼ 2.0 Hz);
(CDCl3, 62.9 MHz)
d 25.1 (C(CH3)2), 25.3 (C(CH3)2), 26.2 (C(CH3)2),
26.8 (C(CH3)2), 56.0 (2 OCH3), 60.9 (OCH3), 65.9 (C-7), 75.8 (C-6),
79.0 (C-4), 79.4 (C-3), 85.4 (C-5), 101.0 (C-1), 107.0 (2 CeAr), 110.0
(C(CH3)2), 113.1 (C(CH3)2), 132.4 (CeAr), 137.9 (CeAr), 152.1 (C-2),
152.5 (2 CeAr); HRMS (ESI) calculated for C22H29BrNaO8: 523.0938,
[M þ Na]þ found 523.0933.
13C NMR (CDCl3, 62.9 MHz)
(SiC(CH3)3), 25.2 (C(CH3)2), 25.3 (C(CH3)2), 25.8 (3 SiC(CH3)3), 26.2
(C(CH3)2), 26.9 (C(CH3)2), 55.5 (OCH3), 55.9 (2 OCH3), 60.9 (OCH3),
d
ꢁ4.6 (Si(CH3)2), ꢁ4.5 (Si(CH3)2), 18.4

A. Novoa et al. / European Journal of Medicinal Chemistry 46 (2011) 3570e3580
3577
65.9 (C-7), 76.1 (C-6), 78.3 (C-4), 80.4 (C-3), 85.4 (C-5), 108.1 (2
CeAr), 109.7 (C(CH3)2), 111.5 (CeAr), 112.4 (C(CH3)2), 119.6 (C-1),
122.3 (CeAr), 123.2 (CeAr), 131.4 (CeAr), 135.1 (CeAr), 136.8
(CeAr), 144.1 (CeAr), 149.7 (CeAr), 149.8 (C-2), 152.4 (2 CeAr);
HRMS (ESI) calculated for C35H50LiO10Si: 665.3329, [M þ Li]þ found
665.3355.
1.49 (s, 3H, C(CH3)2), 1.50 (s, 3H, C(CH3)2), 3.78 (s, 6H, 2 OCH3), 3.79
(m, 1H, H-70), 3.86 (s, 3H, OCH3), 3.87 (s, 3H, OCH3), 4.06 (dd, 1H, H-
5, J ¼ 8.0 Hz, J ¼ 5.0 Hz), 4.23 (dd, 1H, H-7, J ¼ 8.5 Hz, J ¼ 6.5 Hz),
4.46 (dt, 1H, H-6, J ¼ 8.0 Hz, J ¼ 6.5 Hz), 4.69 (dd, 1H, H-4, J ¼ 6.0 Hz,
J ¼ 5.0 Hz), 4.93 (d, 1H, H-3, J ¼ 6.0 Hz), 5.64 (br s, 1H, OH), 6.59 (s,
2H, HeAr), 6.78 (d, 1H, HeAr, J ¼ 8.5 Hz), 6.95 (dd, 1H, HeAr,
J ¼ 8.5 Hz, J ¼ 2.0 Hz), 6.99 (d, 1H, HeAr, J ¼ 2.0 Hz); 13C NMR
6.1.3.2. (Z)-2,5-Anhydro-1-(2-naphtyl)-1-(3,4,5-trimethoxyphenyl)-
(CDCl3, 62.9 MHz) d 25.2 (C(CH3)2), 25.3 (C(CH3)2), 26.2 (C(CH3)2),
1-deoxy-3,4:6,7-di-O-isopropylidene-
D
-gulo-hept-1-enitol (8). Yield:
26.9 (C(CH3)2), 55.8 (OCH3), 55.9 (2 OCH3), 60.9 (OCH3), 66.0 (C-7),
75.8 (C-6), 78.3 (C-4), 80.4 (C-3), 85.1 (C-5), 108.1 (2 CeAr), 109.8
(C(CH3)2), 110.2 (CeAr), 112.5 (C(CH3)2), 116.0 (CeAr), 120.0 (C-1),
121.7 (CeAr), 132.0 (CeAr), 134.9 (CeAr), 136.8 (CeAr), 144.8
(CeAr), 145.4 (CeAr), 150.1 (C-2), 152.4 (2 CeAr); HRMS (ESI)
calculated for C29H36NaO10: 567.2201, [M þ Na]þ found 567.2252.
76% as
a
pale yellow solid; Rf ¼ 0.26 (Hexane/EtOAc 7/3);
20
[
a]
¼ ꢁ269ꢀ (c ¼ 0.37, CHCl3); IR (film)
n
2986, 2938, 1645, 1580,
D
1505 cmꢁ1 1H NMR (CDCl3, 400 MHz)
; d 1.32 (s, 3H, C(CH3)2), 1.38
(s, 3H, C(CH3)2), 1.52 (s, 3H, C(CH3)2), 1.54 (s, 3H, C(CH3)2), 3.78 (s,
6H, 2 OCH3), 3.84 (dd, 1H, H-70, J ¼ 8.5 Hz, J ¼ 6.5 Hz), 3.91 (s, 3H,
OCH3), 4.15 (dd, 1H, H-5, J ¼ 8.5 Hz, J ¼ 5.0 Hz), 4.24 (dd, 1H, H-7,
J ¼ 8.5 Hz, J ¼ 6.5 Hz), 4.50 (dt, 1H, H-6, J ¼ 8.5 Hz, J ¼ 6.5 Hz), 4.77
(dd, 1H, H-4, J ¼ 6.0 Hz, J ¼ 5.0 Hz), 5.04 (d, 1H, H-3, J ¼ 6.0 Hz), 6.67
(s, 2H, HeAr), 7.40e7.43 (m, 2H, HeAr), 7.59 (dd, 1H, HeAr,
J ¼ 8.5 Hz, J ¼ 1.5 Hz), 7.70e7.82 (m, 4H, HeAr); 13C NMR (CDCl3,
6.1.3.5. (E)-2,5-Anhydro-1-(3-hydroxy-4-methoxyhenyl)-1-(3,4,5-tri-
methoxyphenyl)-1-deoxy-3,4:6,7-di-O-isopropylidene-D-gulo-hept-1-
enitol (12). To a solution of silyl ether 11 (0.14 mmol) in dry CH2Cl2
(14 mL) was added slowly via syringe at 0 ꢀC under argon tetra-
62.9 MHz)
d
25.2 (C(CH3)2), 25.3 (C(CH3)2), 26.2 (C(CH3)2), 27.0
butylammonium fluoride (1 M solution THF, 160 mL, 0.16 mmol,
(C(CH3)2), 56.0 (2 OCH3), 60.9 (OCH3), 66.0 (C-7), 75.7 (C-6), 78.4 (C-
4), 80.4 (C-3), 85.1 (C-5), 108.2 (2 CeAr), 109.8 (C(CH3)2), 112.6
(C(CH3)2), 120.7 (C-1), 125.6 (CeAr), 125.7 (CeAr), 127.1 (CeAr),
127.4 (CeAr), 127.9 (CeAr), 128.2 (CeAr), 128.8 (CeAr), 132.4
(CeAr), 133.3 (CeAr), 134.8 (CeAr), 136.2 (CeAr), 137.0 (CeAr), 151.1
(C-2), 152.6 (2 CeAr); HRMS (ESI) calculated for C32H36NaO8:
571.2302, [M þ Na]þ found 571.2306.
1.1 eq). After 30 min, water (10 mL) was added. The organic layer
was separated and the aqueous layer was extracted with CH2Cl2
(2 ꢂ10 mL). The combined organic layers were washed with water
(10 mL), dried (MgSO4) and filtered. The filtrate was concentrated
under reduced pressure and the residue was purified by chroma-
tography (silica gel, Hexane/EtOAc). Yield: 55% as a pale yellow
20
solid; Rf ¼ 0.27 (Hexane/EtOAc 6/4);
CHCl3); IR (film)
NMR (CDCl3, 250 MHz) d 1.29 (s, 3H, C(CH3)2), 1.37 (s, 3H, C(CH3)2),
[
a
]
¼ ꢁ247ꢀ (c ¼ 0.34,
D
n
2987, 2938, 2836, 1642, 1580, 1508 cmꢁ1 1H
;
6.1.3.3. (E)-2,5-Anhydro-1-(3-tertbutyldimethylsilyloxy-4-methoxy-
phenyl)-1-(3,4,5-trimethoxyphenyl)-1-deoxy-3,4:6,7-di-O-isopropyli-
1.43 (s, 3H, C(CH3)2), 1.50 (s, 3H, C(CH3)2), 3.76 (s, 6H, 2 OCH3), 3.78
(m, 1H, H-70), 3.84 (s, 3H, OCH3), 3.92 (s, 3H, OCH3), 4.04 (dd, 1H,
H-5, J ¼ 8.5 Hz, J ¼ 4.5 Hz), 4.24 (dd, 1H, H-7, J ¼ 8.5 Hz, J ¼ 7.0 Hz),
4.50 (dt, 1H, H-6, J ¼ 8.5 Hz, J ¼ 7.0 Hz), 4.67 (dd, 1H, H-4, J ¼ 6.0 Hz,
J ¼ 4.5 Hz), 4.99 (d, 1H, H-3, J ¼ 6.0 Hz), 5.58 (br s, 1H, OH), 6.72 (s,
2H, HeAr), 6.83 (d, 1H, HeAr, J ¼ 8.0 Hz), 6.87 (dd, 1H, HeAr,
J ¼ 8.0 Hz, J ¼ 2.0 Hz), 6.91 (d, 1H, HeAr, J ¼ 2.0 Hz); 13C NMR
1-enitol (11). Yield: 45% as a pale yellow solid; Rf ¼ 0.44 (Hexane/
20
EtOAc 7/3); [
a
]
D
¼ ꢁ240ꢀ (c ¼ 0.48, CHCl3); IR (film)
n
2987, 2938,
d 0.15 (s, 3H,
2857,1654,1578,1511 cmꢁ1; 1H NMR (CDCl3, 400 MHz)
Si(CH3)2), 0.16 (s, 3H, Si(CH3)2), 0.99 (s, 9H, SiC(CH3)3), 1.27 (s, 3H,
C(CH3)2), 1.37 (s, 3H, C(CH3)2), 1.43 (s, 3H, C(CH3)2), 1.48 (s, 3H,
C(CH3)2), 3.75 (s, 9H, 3 OCH3), 3.78 (dd, 1H, H-70, J ¼ 8.5 Hz,
J ¼ 7.0 Hz), 3.83 (s, 3H, OCH3), 4.04 (dd, 1H, H-5, J ¼ 8.5 Hz,
J ¼ 4.5 Hz), 4.24 (dd, 1H, H-7, J ¼ 8.5 Hz, J ¼ 7.0 Hz), 4.50 (dt, 1H, H-6,
J ¼ 8.5 Hz, J ¼ 7.0 Hz), 4.67 (dd, 1H, H-4, J ¼ 6.0 Hz, J ¼ 4.5 Hz), 4.95
(d, 1H, H-3, J ¼ 6.0 Hz), 6.72 (s, 2H, HeAr), 6.81 (d, 1H, HeAr,
J ¼ 8.5 Hz), 6.85 (dd, 1H, HeAr, J ¼ 8.5 Hz, J ¼ 2.0 Hz), 6.95 (d, 1H,
(CDCl3, 62.9 MHz)
d 25.2 (C(CH3)2), 25.4 (C(CH3)2), 26.4 (C(CH3)2),
26.8 (C(CH3)2), 55.8 (OCH3), 55.9 (2 OCH3), 60.8 (OCH3), 65.8 (C-7),
75.7 (C-6), 78.3 (C-4), 80.5 (C-3), 85.0 (C-5), 107.1 (2 CeAr), 109.7
(C(CH3)2), 110.0 (CeAr), 112.8 (C(CH3)2), 117.0 (CeAr), 119.5 (C-1),
122.9 (CeAr), 132.0 (CeAr), 134.4 (CeAr), 136.8 (CeAr), 144.9
(CeAr), 145.7 (CeAr), 150.5 (C-2), 152.4 (2 CeAr); HRMS (ESI)
calculated for C29H36NaO10: 567.2201, [M þ Na]þ found 567.2250.
HeAr, J ¼ 2.0 Hz); 13C NMR (CDCl3, 62.9 MHz)
d
ꢁ4.7 (2 Si(CH3)2),
18.4 (SiC(CH3)3), 25.7 (C(CH3)2), 25.2 (C(CH3)2), 25.3 (SiC(CH3)3),
26.3 (C(CH3)2), 26.8 (C(CH3)2), 55.4 (OCH3), 55.8 (2 OCH3), 60.8
(OCH3), 65.8 (C-7), 75.6 (C-6), 78.3 (C-4), 80.5 (C-3), 85.1 (C-5), 107.0
(2 CeAr), 109.6 (C(CH3)2), 111.4 (CeAr), 112.8 (C(CH3)2), 119.4 (C-1),
124.0 (CeAr), 124.2 (CeAr), 131.4 (CeAr), 134.5 (CeAr), 136.6
(CeAr), 144.2 (CeAr), 150.2 (CeAr), 150.5 (C-2), 152.4 (2 CeAr);
HRMS (ESI) calculated for C35H50NaO10Si: 681.3065, [M þ Na]þ
found 681.3082.
6.1.3.6. (E)-2,5-Anhydro-1-(3-hydroxy-4-methoxyphenyl)-1-(3,4,5-
trimethoxyphenyl)-1-deoxy-3,4-O-isopropylidene-D-gulo-hept-1-enitol
(19). To a stirred solution of the disubstituted exo-glycal 12
(0.34 mmol) in ethanol (2 mL) and water (2 mL) was added tri-
fluoroacetic acid (2 mL), at 0 ꢀC. After stirring 1 h at room
temperature CH2Cl2 (60 mL) was added. The organic layer was
washed with an aqueous solution of NaHCO3 until pH ¼ 7 and then
with water (10 mL). The organic layer was dried over MgSO4 and
filtered. The solvent was removed under reduced pressure and the
resulting product was purified by chromatography (silica gel,
EtOAc). Yield: 62% as a pale yellow solid; Rf ¼ 0.44 (EtOAc);
6.1.3.4. (Z)-2,5-Anhydro-1-(3-hydroxy-4-methoxyphenyl)-1-(3,4,5-
trimethoxyphenyl)-1-deoxy-3,4:6,7-di-O-isopropylidene-
1-enitol (6). The silyl ether 5 (0.14 mmol) was dissolved in dry
CH2Cl2 (14 mL), tetrabutylammonium fluoride (156 L, 0.16 mmol,
D-gulo-hept-
m
1.1 eq) was added slowly via syringe at 0 ꢀC under argon. After
30 min, water (10 mL) was added. The organic layer was separated
and the aqueous layer was extracted with CH2Cl2 (2 ꢂ 10 mL). Then
the combined organic layers were washed with water (10 mL),
dried (MgSO4) and filtered. The filtrate was evaporated under
reduced pressure and the residue was purified by chromatography
[a
]
¼ ꢁ242ꢀ (c ¼ 0.25, CHCl3); IR (Film)
n
3432, 2986, 2936, 2837,
D
1645, 1585, 1510 cmꢁ1 1H NMR (CDCl3, 400 MHz)
; d 1.33 (s, 3H,
C(CH3)2),1.52 (s, 3H, C(CH3)2), 2.20 (br s,1H, OH), 2.95 (br s,1H, OH),
3.76 (s, 6H, 2 OCH3), 3.76e3.90 (m, 2H, H-7, H-70), 3.84 (s, 3H, OCH3),
3.92 (s, 3H, OCH3), 4.10 (m, 1H, H-5), 4.19 (m, 1H, H-6), 4.79 (dd, 1H,
H-4, J ¼ 6.0 Hz, J ¼ 4.0 Hz), 5.07 (d,1H, H-3, J ¼ 6.0 Hz), 5.61 (br s,1H,
OH), 6.59 (s, 2H, HeAr), 6.83 (d, 1H, HeAr, J ¼ 8.0 Hz), 6.88 (dd, 1H,
HeAr, J ¼ 8.0 Hz, J ¼ 2.0 Hz), 6.91 (d, 1H, HeAr, J ¼ 2.0 Hz); 13C NMR
(silica gel, Hexane/EtOAc). Yield: 70% as a pale yellow solid;
20
Rf ¼ 0.19 (Hexane/EtOAc 4/6); [
(film)
NMR (CDCl3, 400 MHz)
a]
¼ ꢁ276ꢀ (c ¼ 0.64, CHCl3); IR
D
n
3439 (OH), 2997, 2938 (CeH), 1656, 1580, 1508 cmꢁ1
;
1H
(CDCl3, 100 MHz) d 25.7 (C(CH3)2), 26.8 (C(CH3)2), 56.2 (OCH3), 56.5
d
1.27 (s, 3H, C(CH3)2), 1.38 (s, 3H, C(CH3)2),
(2 OCH3), 61.3 (OCH3), 63.6 (C-7), 71.3 (C-6), 79.1 (C-4), 80.9 (C-3),

3578
A. Novoa et al. / European Journal of Medicinal Chemistry 46 (2011) 3570e3580
82.8 (C-5), 107.4 (2 CeAr), 110.4 (CeAr), 113.2 (C(CH3)2), 117.1
(CeAr), 120.4 (C-1), 123.0 (CeAr), 132.4 (CeAr), 135.0 (CeAr), 137.2
(CeAr), 145.3 (CeAr), 146.2 (CeAr), 150.3 (C-2), 153.0 (2 CeAr);
HRMS (ESI) calculated for C26H32NaO10 [M þ Na]þ: 527.1888, found:
527.1897.
62.9 MHz)
d
ꢁ4.7 (2 Si(CH3)2), 18.4 (OCOCH3), 20.9 (OCOCH3), 21.0
(SiC(CH3)3), 25.6 (3 SiC(CH3)3), 55.4 (OCH3), 56.0 (2 OCH3),
60.9 (OCH3), 63.3 (CH2), 63.4 (CH2), 106.7 (2 CeAr), 111.5 (CeAr),
122.2 (CeAr), 123.0 (CeAr), 126.0 (C]C), 132.7 (CeAr), 136.2
(CeAr), 137.7 (CeAr), 144.4 (CeAr), 149.0 (C]C), 150.9 (CeAr), 152.8
(2 CeAr), 170.7 (OCOCH3), 170.8 (OCOCH3); HRMS (ESI) calculated
for C30H42NaO9Si: 597.2490, [M þ H]þ found 597.2492.
6.1.3.7. 2-(1,1-Dibromomethylene)-1,3-diacetate-propan-1,3-diol (22).
To a stirred solution of CBr4 (11.44 g, 34.5 mmol, 3 eq) in CH2Cl2
(60 mL) was added dropwise a solution of PPh3 (18.10 g, 69 mmol,
6 eq) in CH2Cl2 (45 mL) at 0 ꢀC, under argon and with exclusion of
light (aluminium foil). Then a solution of 21 (2.0 g, 11.5 mmol) in
CH2Cl2 (60 mL) was added dropwise. After stirring at room
temperature for 2 h, hexane (800 mL) was added. The precipitate
was removed by filtration and the filtrate was concentrated under
reduced pressure. The residue was purified by column chroma-
tography (silica gel, Hexane/EtOAc ¼ 8/2). Yield: 85% as a white
6.1.3.10. 2-(1-(3-Hydroxy-4-methoxyphenyl)-1-(3,4,5-trimethoxy-
phenyl)methylene)-1,3-diacetate-propan-1,3-diol (26). To a solu-
tion of silyl ether 25 (0.14 mmol) in dry CH2Cl2 (14 mL) was added
slowly via syringe at 0 ꢀC under argon tetrabutylammonium fluo-
ride (1 M solution in THF, 160 mL, 0.16 mmol, 1.1 eq) After 30 min,
water (10 mL) was added. The organic layer was separated and the
aqueous layer was extracted with CH2Cl2 (2 ꢂ 10 mL). Then the
combined organic layers were washed with water (10 mL), dried
(MgSO4) and filtered. The solvent was evaporated under reduced
pressure and the residue was purified by chromatography (silica gel,
Hexane/EtOAc). Yield: 90% as a pale yellow solid; Rf ¼ 0.25 (Hexane/
solid; Rf ¼ 0.57 (Hexane/EtOAc 7/3); IR (film)
NMR (CDCl3, 250 MHz)
2.08 (s, 6H, OCOCH3), 4.79 (s, 4H, CH2); 13
NMR (CDCl3, 62.9 MHz) 20.6 (2 OCOCH3), 63.9 (2 CH2), 98.0 (C]
n
2959, 1744 cmꢁ1; 1H
d
C
d
C), 135.5 (C]C), 170.3 (2 OCOCH3); HRMS (ESI) calculated for
EtOAc 1/1); IR (film)
n
3418, 2997, 2943, 2836, 1736, 1650, 1580,
2.07 (s, 3H, OCOCH3), 2.09
C8H10Br2NaO4: 352.8838, [M þ Na]þ found 352.8889.
1508 cmꢁ1; 1H NMR (CDCl3, 250 MHz)
d
(s, 3H, OCOCH3), 3.78 (s, 6H, OCH3), 3.84 (s, 3H, OCH3), 3.88 (s, 3H,
OCH3), 4.67 (s, 2H, CH2), 4.68 (s, 2H, CH2), 5.65 (br s, 1H, OH), 6.36 (s,
2H, HeAr), 6.67 (dd, 1H, HeAr, J ¼ 8.0 Hz, J ¼ 2.0 Hz), 6.72 (d, 1H,
HeAr, J ¼ 2.0 Hz), 6.79 (d, 1H, HeAr, J ¼ 8.0 Hz); 13C NMR (CDCl3,
6.1.3.8. 2-(1-Bromo-1-(3,4,5-trimethoxyphenyl)methylene)-1,3-diace-
tate-propan-1,3-diol (23). In a dry round-bottomed flask equipped
with a condenser flushed with argon, 5 mL of dry DME, compound
22 (150 mg, 0.362 mmol), 3,4,5-trimethoxyphenylboronic acid
(115 mg, 0.543 mmol, 1.5 eq), 2 M K2CO3 aqueous solution (3 eq)
were stirred during 10 min, then tris(2-furyl)phosphine (26 mg,
0.109 mmol, 0.3 eq) and PdCl2(PPh3)2 (13 mg, 0.018 mmol, 0.05 eq)
were added as solids. The reaction mixture was stirred and heated at
85 ꢀC for 24 h. The reaction was then allowed to cool to room
temperature and was evaporated to dryness under reduced pres-
sure. The residue was diluted with ethyl acetate (50 mL) and washed
with NH4Cl (2 ꢂ 5 mL). The solvent was removed under reduced
pressure and the residue was purified by column chromatography
(n-hexane/EtOAc). Yield: 35% as a pale yellow solid; Rf ¼ 0.58
62.9 MHz)
d 20.9 (OCOCH3), 21.0 (OCOCH3), 55.9 (OCH3), 56.1 (2
OCH3), 60.8 (OCH3), 63.1 (CH2), 63.2 (CH2), 106.6 (2 CeAr), 110.1
(CeAr), 115.6 (CeAr), 121.1 (CeAr), 126.5 (C]C), 133.3 (CeAr), 136.0
(CeAr), 137.7 (CeAr), 145.2 (CeAr), 146.4 (CeAr), 148.8 (C]C), 152.8
(2 CeAr), 170.7 (OCOCH3), 170.8 (OCOCH3); HRMS (ESI) calculated
for C24H28NaO9 [M þ H]þ: 483.1626, found: 483.1625.
6.1.3.11. 2-(1-(2-Naphtyl)-1-(3,4,5-trimethoxyphenyl)methylene)-
propan-1,3-diol (32). To a solution of 28 (0.24 mmol) in methanol
(5 mL) was added at 0 ꢀC under argon a freshly prepared solution of
sodium methanolate in methanol (1 mL). After 30 min, Amberlite
IR-120 (Hþ) was added until pH ¼ 7. After filtration, the solvent was
removed under reduced pressure and the residue was purified by
chromatography (Hexane/EtOAc). Yield: 83% as a pale yellow solid;
(Hexane/EtOAc 1/1); IR (film)
n 2991, 2938, 2830, 1741, 1647, 1577,
1502 cmꢁ1; 1H NMR (CDCl3, 250 MHz) 2.05 (s, 3H, OCOCH3), 2.12 (s,
3H, OCOCH3), 3.85 (s, 6H, OCH3), 3.87 (s, 3H, OCH3), 4.57 (s, 2H, CH2),
4.97 (s, 2H, CH2), 6.56 (s, 2H, HeAr); 13C NMR (CDCl3, 62.9 MHz)
Rf ¼ 0.20 (Hexane/EtOAc 1/2); IR (film)
n
3402, 3051, 2933, 2851,
3.77 (s,
d
20.7 (2 OCOCH3), 56.2 (OCH3), 56.3 (OCH3), 60.9 (OCH3), 61.9 (CH2),
1650, 1580 (OMe), 1503 cmꢁ1; 1H NMR (CDCl3, 400 MHz)
d
65.1 (CH2), 106.1 (2 CeAr), 129.3 (C]C), 130.8 (CeAr), 134.0 (CeAr),
137.5 (C]C),152.9 (2 CeAr),170.2 (OCOCH3),170.6 (OCOCH3); HRMS
(ESI) calculated for C17H21BrNaO7: 439.0363, [M þ Na]þ found
441.0350.
6H, OCH3), 3.86 (s, 3H, OCH3), 4.45 (br s, 4H, CH2), 6.46 (s, 2H,
HeAr), 7.46e7.52 (m, 3H, HeAr), 7.70 (m, 1H, HeAr), 7.75e7.85 (m,
3H, HeAr); 13C NMR (CDCl3, 62.9 MHz)
d 56.1 (2 OCH3), 60.9 (OCH3),
63.7 (CH2), 63.8 (CH2), 107.1 (2 CeAr), 126.2 (CeAr), 126.3 (CeAr),
127.5 (CeAr), 127.6 (CeAr), 127.7 (CeAr), 128.1 (CeAr), 128.3
(CeAr), 132.6 (CeAr), 132.9 (CeAr), 135.3 (C]C), 136.3 (CeAr), 137.4
(CeAr), 138.0 (CeAr), 143.5 (C]C), 152.8 (2 CeAr); HRMS (ESI)
calculated for C23H24NaO5: 403.1516, [M þ H]þ found: 403.1497.
6.1.3.9. 2-(1-(3-Tert-butyldimethylsilyloxy-4-methoxyphenyl)-1-(3,4,
5-trimethoxyphenyl)methylene)-1,3-diacetate-propan-1,3-diol (25).
In a dry round-bottomed flask equipped with a condenser flushed
with argon, 5 mL of dry DME, compound 23 (151 mg, 0.362 mmol,
1 eq), boronic acid (2 eq), 2 M K2CO3 aqueous solution (4 eq) were
stirred during 10 min. Pd(PPh3)4 (20 mg, 0.018 mmol, 0.05 eq) was
added as solids. The reaction mixture was stirred and heated at
85 ꢀC during 24 h. The reaction was then allowed to cool to room
temperature and was evaporated to dryness under reduced pres-
sure. The residue was diluted with ethyl acetate (50 mL) and
washed with NH4Cl (2 ꢂ 5 mL). The dried organic layer was
concentrated under reduced pressure and the residue was purified
by chromatography (n-hexane/EtOAc). Yield: 53% as a pale yellow
6.2. Antiproliferative assays (NCI60 cell lines panel)
The human tumor cell lines of the cancer screening panel are
grown in RPMI 1640 medium containing 5% fetal bovine serum
and 2 mM
L
-glutamine. For a typical screening experiment, cells
L at plating
are inoculated into 96-well microtiter plates in 100
m
densities ranging from 5000 to 40,000 cells/well depending on
the doubling time of individual cell lines. After cell inoculation,
the microtiter plates are incubated at 37 ꢀC, 5% CO2, 95% air and
100% relative humidity for 24 h prior to addition of experimental
drugs. After 24 h, two plates of each cell line are fixed in situ with
TCA, to represent a measurement of the cell population for each
cell line at the time of drug addition (Tz). Experimental drugs are
solubilised in dimethyl sulfoxide at 400-fold the desired final
maximum test concentration and stored frozen prior to use. At
solid; Rf ¼ 0.62 (Hexane/EtOAc 1/1); IR (film)
n
2954, 2933, 2857,
0.11 (s,
1742, 1642, 1580, 1508 cmꢁ1 1H NMR (CDCl3, 400 MHz)
;
d
6H, Si(CH3)2), 0.96 (s, 9H, SiC(CH3)3), 2.08 (s, 3H, OCOCH3), 2.09 (s,
3H, OCOCH3), 3.77 (s, 6H, OCH3), 3.81 (s, 3H, OCH3), 3.85 (s, 3H,
OCH3), 4.68 (s, 2H, CH2), 4.70 (s, 2H, CH2), 6.36 (s, 2H, HeAr), 6.69 (s,
1H, HeAr), 6.70 (m, 1H, HeAr), 6.79 (m, 1H, HeAr); 13C NMR (CDCl3,

A. Novoa et al. / European Journal of Medicinal Chemistry 46 (2011) 3570e3580
3579
the time of drug addition, an aliquot of frozen concentrate is
thawed and diluted to twice the desired final maximum test
concentration with complete medium containing 50 g/mL
gentamicin. Additional four, 10-fold or 1/2 log serial dilutions are
made to provide a total of five drug concentrations plus control.
6.5. Computational procedure
m
All molecular modelling studies were performed with Schro-
dinger Molecular Modelling Suite 2010 [48]. Maestro is the inter-
face piloting the diverse modules. Macromodel was used to prepare
ligands and Glide is the docking software. Calculations were run on
a Linux station: IntelÒ CoreÔ i7 CPU 950 @ 3.07 GHz. Tubulin
complex with DAMA-colchicine was retrieved from the protein
data bank [40]. Subunits C, D and E were removed. Only subunits A,
B (colchicine binding site) and small molecules DAMA-colchicine
(CN2), GTP and Mg2þ ion in this site were conserved. Structures
were next prepared using the workflow Protein Preparation
Wizard. Structures were preprocessed (hydrogen atoms added,
incomplete residues filled .), bond orders and connections of
ligands were manually corrected. An exhaustive sampling was done
regarding hydrogen bond assignment and the complex was finally
refined by a minimization stage with a constraint to converge to
Aliquots of 100
the appropriate microtiter wells already containing 100
m
L of these different drug dilutions are added to
L of
m
medium, resulting in the required final drug concentrations.
Following drug addition, the plates are incubated for an addi-
tional 48 h at 37 ꢀC, 5% CO2, 95% air, and 100% relative humidity.
For adherent cells, the assay is terminated by the addition of cold
TCA. Cells are fixed in situ by the gentle addition of 50 mL of cold
50% (w/v) TCA (final concentration, 10% TCA) and incubated for
60 min at 4 ꢀC. The supernatant is discarded, and the plates are
washed five times with tap water and air dried. Sulforhodamine B
(SRB) solution (100 mL) at 0.4% (w/v) in 1% acetic acid is added to
each well, and plates are incubated for 10 min at room temper-
ature. After staining, unbound dye is removed by washing five
times with 1% acetic acid and the plates are air dried. Bound stain
is subsequently solubilised with 10 mM trizma base, and the
absorbance is read on an automated plate reader at a wavelength
of 515 nm. For suspension cells, the methodology is the same
except that the assay is terminated by fixing settled cells at the
ꢀ
a structure with an RMSD of 0.3 A (OPLS2005 force field), essen-
tially in order to remove steric clashes. Docking calculations were
performed with standard precision. Ligand flexibility is taken into
account and the option of sampling of ring conformation was
activated. CA-4 was built within the builder module of Maestro [40]
and was submitted to Corina a 3D structure generator [49,50]. This
conformation was next minimized within Macromodel using the
OPLS 2005 force field [42]. Next the optimized energy minimized
conformations have been submitted to the docking software Glide
[40]. Ten poses at most were written out per ligand and a post-
docking minimization stage was performed with 1000 poses per
ligand to be included.
bottom of the wells by gently adding 50
mL of 80% TCA (final
concentration, 16% TCA). Three dose response parameters (GI50
,
TGI and LC50) were calculated for each experimental agent and
each cell line [38].
6.3. Antiproliferative assays (ICSN 9 cell lines panel)
The human cell lines were obtained from ATCC, except when
otherwise stated: HCT15 (colon adenocarcinoma), SK-OV3 (ovary
adenocarcinoma from NCI), OVCAR8 (ovary adenocarcinoma from
Dr Liscovitch, Rehovot, Israel), SF268 (glioblastoma from NCI), PC-3
(prostate adenocarcinoma), HL60 (promyeocytic leukemia), K562
(chronic myelogenous leukemia) and MDA-MB-231 (breast
adenocarcinoma) were grown in RPMI medium supplemented with
10% fetal calf serum, in the presence of penicillin, streptomycin and
fungizone in 75 cm2 flask under 5% CO2. HL60R/R10 was originated
from Oncodesign and maintained in complete RPMI medium con-
taining 100 nM adriamycin.
Acknowledgments
We thank F. Dupire, B. Fernette and P. Lemière for technical
assistance.
Appendix. Supplementary data
Experimental procedures and spectroscopic data for interme-
diate and final compounds, additional cytotoxicity data in the
NCI60 cell lines and in the ciblothèque-ICSN cell lines.
Supplementary data associated with this article can be found in
the online version, at doi:10.1016/j.ejmech.2011.05.021.
Cells were plated in 96-well tissue culture plates in 200
mL
medium and treated 24 h later with 2 stock solution of
mL
compounds dissolved in DMSO using a Biomek 3000 (Beck-
maneCoulter). Controls received the same volume of DMSO (1%
final volume). After 72 h exposure, MTS reagent (Promega) was
added and incubated for 3 h at 37 ꢀC: the absorbance was moni-
tored at 490 nm and results expressed as the inhibition of cell
proliferation calculated as the ratio [(1 ꢁ (OD490 treated/OD490
control)) ꢂ 100] in triplicate experiments. For IC50 determination
[50% inhibition of cell proliferation], cells were incubated for 72 h
following the same protocol with compound concentrations ranged
References
[1] M. Saleem, H.J. Kim, M.S. Ali, Y.S. Lee, An update on bioactive plant lignans,
Nat. Prod. Rep. 22 (2005) 696e716.
[2] C.M. Lin, S.B. Singh, P.S. Chu, R.O. Dempcy, J.M. Schmidt, G.R. Pettit, E. Hamel,
Interactions of tubulin with potent natural and synthetic analogs of the
antimitotic agent combretastatin: a structureeactivity study, Mol. Pharmacol.
34 (1988) 200e208.
[3] G.R. Pettit, S.B. Singh, M.R. Boyd, E. Hamel, R.K. Pettit, J.M. Schmidt, F. Hogan,
Antineoplastic agents. 291. Isolation and synthesis of combretastatins A-4,
A-5, and A-6(1a), J. Med. Chem. 38 (1995) 1666e1672.
[4] G.R. Pettit, M.P. Grealish, D.L. Herald, M.R. Boyd, E. Hamel, R.K. Pettit, Antineo-
plastic agents. 443. Synthesis of the cancer cell growth inhibitor hydroxyphen-
statin and its sodium diphosphate prodrug, J. Med. Chem. 43 (2000) 2731e2737.
[5] C.M. Lin, H.H. Ho, G.R. Pettit, E. Hamel, Antimitotic natural products com-
bretastatin A-4 and combretastatin A-2: studies on the mechanism of their
inhibition of the binding of colchicine to tubulin, Biochemistry 28 (1989)
6984e6991.
[6] P. Hinnen, F.A.L.M. Eskens, Vascular disrupting agents in clinical development,
Br. J. Cancer 96 (2007) 1159e1165.
[7] D.J. Chaplin, G.R. Pettit, S.A. Hill, Anti-vascular approaches to solid tumor
therapy: evaluation of combretastatin A4 phosphate, Anticancer Res. 19
(1999) 189e196.
[8] C.J. Mooney, G. Nagaiah, P. Fu, J.K. Wasman, M.M. Cooney, P.S. Savvides,
J.A. Bokar, A. Dowlati, D. Wang, S.S. Agarwala, S.M. Flick, P.H. Hartman,
J.D. Ortiz, P. Lavertu, S.C. Remick, A phase II trial of fosbretabulin in advanced
anaplastic thyroid carcinoma and correlation of baseline serum-soluble
intracellular adhesion molecule-1 with outcome, Thyroid 19 (2009) 233e240.
5 nM to 100 mM in separate duplicate experiments.
6.4. Inhibition of tubulin assembly
Sheep brain microtubule proteins were purified by two cycles of
assembly/disassembly at 37 ꢀC/0 ꢀC and then dissolved in MES
buffer: 100 mM MES (2-[N-morpholino]-ethanesulfonic acid, pH
6.6), 1 mM EGTA (ethyleneglycol-bis[ß-aminoethyl ether]-N,N,N0,N0-
tetraacetic acid), 0.5 mMMgCl2. Allsamples were dissolved in DMSO,
incubated at 37 ꢀC for 10 min and at 0 ꢀC for 5 min before evaluation
of the tubulin assembly rate. The tubulin assemblyassay was realized
according to a slightly modified Guénard protocol using deoxy-
podophyllotoxin as reference compound [47].

3580
A. Novoa et al. / European Journal of Medicinal Chemistry 46 (2011) 3570e3580
[9] G.C. Tron, T. Pirali, G. Sorba, F. Pagliai, S. Busacca, A.A. Genazzani, Medicinal
chemistry of combretastatin A4: present and future directions, J. Med. Chem.
49 (2006) 3033e3044.
[10] J.-P. Liou, J.-Y. Chang, C.-W. Chang, C.-Y. Chang, N. Mahindroo, F.-M. Kuo,
H.-P. Hsieh, Synthesis and structureeactivity relationships of 3-amino-
benzophenones as antimitotic agents, J. Med. Chem. 47 (2004) 2897e2905.
[11] M.G. Banwell, E. Hamel, D.C.R. Hockless, P. Verdier-Pinard, A.C. Willis,
D.J. Wong, 4,5-Diaryl-1H-pyrrole-2-carboxylates as combretastatin A-4/
lamellarin T hybrids: synthesis and evaluation as anti-mitotic and cytotoxic
agents, Bioorg. Med. Chem. 14 (2006) 4627e4638.
[12] E.G. Barbosa, L.A.S. Bega, A. Beatriz, T. Sarkar, E. Hamel, M.S. do Amaral,
D. Pires de Lima, A diaryl sulfide, sulfoxide, and sulfone bearing structural
similarities to combretastatin A-4, Eur. J. Med. Chem. 44 (2009) 2685e2688.
[13] (a) R. Romagnoli, P.G. Baraldi, M.D. Carrion, O. Cruz-Lopez, C. Lopez Cara,
G. Basso, G. Viola, M. Khedr, J. Balzarini, S. Mahboobi, A. Sellmer, A. Brancale,
E. Hamel, 2-Arylamino-4-amino-5-aroylthiazoles. One-pot synthesis and
biological evaluation of a new class of inhibitors of tubulin polymerization,
J. Med. Chem. 52 (2009) 5551e5555;
potent tubulin polymerization inhibitors, Bioorg. Med. Chem. 18 (2010)
5114e5122.
[27] J.-P. Liou, Y.-L. Chang, F.-M. Kuo, C.-W. Chang, H.-Y. Tseng, C.-C. Wang,
Y.-N. Yang, J.-Y. Chang, S.-J. Lee, H.-P. Hsieh, Concise synthesis and structur-
eeactivity relationships of combretastatin A-4 analogues, 1-aroylindoles and
3-aroylindoles, as novel classes of potent antitubulin agents, J. Med. Chem. 47
(2004) 4247e4257.
[28] G. Dupeyre, G.G. Chabot, S. Thoret, X. Cachet, J. Seguin, D. Guenard,
F. Tillequin, D. Scherman, M. Koch, S. Michel, Synthesis and biological evalu-
ation of (3,4,5-trimethoxyphenyl)indol-3-ylmethane derivatives as potential
antivascular agents, Bioorg. Med. Chem. 14 (2006) 4410e4426.
[29] J.-P. Liou, N. Mahindroo, C.-W. Chang, F.-M. Guo, S.W.-H. Lee, U.-K. Tan,
T.-K. Yeh, C.-C. Kuo, Y.-W. Chang, P.-H. Lu, Y.-S. Tung, K.-T. Lin, J.-Y. Chang, H.-
P. Hsieh, Structureeactivity relationship studies of 3-aroylindoles as potent
antimitotic agents, ChemMedChem 1 (2006) 1106e1118.
[30] Y. Jin, Z.Y. Zhou, W. Tian, Q. Yu, Q.Y. Longa, 40-Alkoxyl substitution enhancing
the anti-mitotic effect of 5-(30,40,50-substituted)anilino-4-hydroxy-8-
nitroquinazolines as a novel class of anti-microtubule agents, Bioorg. Med.
Chem. Lett. 16 (2006) 5864e5869.
[31] C. Alvarez, R. Alvarez, P. Corchete, J.L. Lopez, C. Perez-Melero, R. Pelaez,
M. Medarde, Diarylmethyloxime and hydrazone derivatives with 5-indolyl
moieties as potent inhibitors of tubulin polymerization, Bioorg. Med. Chem.
16 (2008) 5952e5961.
[32] S. Messaoudi, B. Treguier, A. Hamze, O. Provot, J.-F. Peyrat, J.R. De Losada,
J.-M. Liu, J. Bignon, J. Wdzieczak-Bakala, S. Thoret, J. Dubois, J.-D. Brion,
M. Alami, Isocombretastatins A versus combretastatins A: the forgotten isoCA-
4 isomer as a highly promising cytotoxic and antitubulin agent, J. Med. Chem.
52 (2009) 4538e4542.
[33] R. Alvarez, C. Alvarez, F. Mollinedo, B.G. Sierra, M. Medarde, R. Pelaez, Iso-
combretastatins A: 1,1-diarylethenes as potent inhibitors of tubulin poly-
merization and cytotoxic compounds, Bioorg. Med. Chem. 17 (2009)
6422e6431.
(b) R. Romagnoli, P.G. Baraldi, O. Cruz-Lopez, C. Lopez Cara, M.D. Carrion,
A. Brancale, E. Hamel, L. Chen, R. Bortolozzi, G. Basso, G. Viola, Synthesis and
antitumor activity of 1,5-disubstituted 1,2,4-triazoles as cis-restricted com-
bretastatin analogues, J. Med. Chem. 53 (2010) 4248e4258.
[14] B. Burja, T. Cimbora-Zovko, S. Tomic, T. Jelusic, M. Kocevar, S. Polanc,
M. Osmak, Pyrazolone-fused combretastatins and their precursors: synthesis,
cytotoxicity, antitubulin activity and molecular modeling studies, Bioorg.
Med. Chem. 18 (2010) 2375e2387.
[15] (a) K. Odlo, J. Fournier-Dit-Chabert, S. Ducki, O.A.B.S.M. Gani, I. Sylte,
T.V. Hansen, 1,2,3-Triazole analogs of combretastatin A-4 as potential
microtubule-binding agents, Bioorg. Med. Chem. 18 (2010) 6874e6885;
(b) C. Bailly, C. Bal, P. Barbier, S. Combes, J.-P. Finet, M.-P. Hildebrand,
V. Peyrot, N. Wattez, Synthesis and biological evaluation of 4-arylcoumarin
analogues of combretastatins, J. Med. Chem. 46 (2003) 5437e5444.
[16] N. Ty, J. Kaffy, A. Arrault, S. Thoret, R. Pontikis, J. Dubois, L. Morin-Allory,
J.-C. Florent, Synthesis and biological evaluation of cis-locked vinylogous
[34] A. Novoa, N. Pellegrini-Moïse, S. Lamandé-Langle, Y. Chapleur, Efficient access
to disubstituted exo-glycals. Application to the preparation of C-glycosyl
compounds, Tetrahedron Lett. 50 (2009) 6484e6487.
combretastatin-A4 analogues: derivatives with
a cyclopropyl-vinyl or
a cyclopropyl-amide bridge, Bioorg. Med. Chem. Lett. 19 (2009) 1318e1322.
[17] A.B.S. Maya, C. Perez-Melero, C. Mateo, D. Alonso, J.L. Fernandez, C. Gajate,
F. Mollinedo, R. Pelaez, E. Caballero, M. Medarde, Further naphthylcom-
bretastatins. An investigation on the role of the naphthalene moiety, J. Med.
Chem. 48 (2005) 556e568.
[18] A.B. Sanchez Maya, C. Perez-Melero, N. Salvador, R. Pelaez, E. Caballero,
M. Medarde, New naphthylcombretastatins. Modifications on the ethylene
bridge, Bioorg. Med. Chem. 13 (2005) 2097e2107.
[35] S.K. Chattopadhyay, T.R.S. Kumar, P.R. Maulik, S. Srivastava, A. Garg, A. Sharon,
A.S. Negi, S.P.S. Khanuja, Absolute configuration and anticancer activity of
taxiresinol and related lignans of Taxus wallichiana, Bioorg. Med. Chem. 11
(2003) 4945e4948.
[36] (a) C. Taillefumier, Y. Chapleur, Synthesis and uses of exo-glycals, Chem. Rev.
104 (2004) 263e292;
(b) Y. Lakhrissi, C. Taillefumier, F. Chrétien, Y. Chapleur, Facile dibromoolefi-
nation of lactones using bromomethylene triphenylphosphorane, Tetrahe-
dron Lett. 42 (2001) 7265e7268.
[19] G.R. Pettit, B. Toki, D.L. Herald, P. Verdier-Pinard, M.R. Boyd, E. Hamel,
R.K. Pettit, Antineoplastic agents. 379. Synthesis of phenstatin phosphate,
J. Med. Chem. 41 (1998) 1688e1695.
[37] (a) K.S. Feldman, Cyclization pathways of a (Z)-stilbene-derived bis(ortho
quinone monoketal), J. Org. Chem. 62 (1997) 4983e4990;
[20] C. Alvarez, R. Alvarez, P. Corchete, C. Perez-Melero, R. Pelaez, M. Medarde,
Exploring the effect of 2,3,4-trimethoxyphenyl moiety as a component of
indolephenstatins, Eur. J. Med. Chem. 45 (2010) 588e597.
(b) C.A. Ocasio, T.S. Scanlan, Characterization of thyroid hormone receptor
alpha (TR
a)-specific analogs with varying inner- and outer-ring substituents,
Bioorg. Med. Chem. 16 (2008) 762e770.
[21] G.R. Pettit, M.P. Grealish, M.K. Jung, E. Hamel, R.K. Pettit, J.C. Chapuis,
J.M. Schmidt, Antineoplastic agents. 465. Structural modification of resvera-
trol: sodium resverastatin phosphate, J. Med. Chem. 45 (2002) 2534e2542.
[22] (a) R. Romagnoli, P.G. Baraldi, M.K. Jung, M.A. Iaconinoto, M.D. Carrion,
V. Remusat, D. Preti, M.A. Tabrizi, F. Francesca, E. De Clercq, J. Balzarini,
E. Hamel, Synthesis and preliminary biological evaluation of new anti-tubulin
agents containing different benzoheterocycles, Bioorg. Med. Chem. Lett. 15
(2005) 4048e4052;
[38] National Cancer Institute (NCI), Bethesda. Available from: http://dtp.nci.nih.
gov.
[39] R.B.G. Ravelli, B. Gigant, P.A. Curmi, I. Jourdain, S. Lachkar, A. Sobel,
M. Knossow, Insight into tubulin regulation from a complex with colchicine
and a stathmin-like domain, Nature (London, U.K 428 (2004) 198e202.
[40] 3D structure of phenstatin and phenstatin-like compounds were next built
from the best CA-4 docked pose and then optimized again by minimization.
Available from: http://www.rcsb.org/pdb/home/home.do.
(b) R. Romagnoli, P.G. Baraldi, M.G. Pavani, M.A. Tabrizi, D. Preti, F. Fruttarolo,
L. Piccagli, M.K. Jung, E. Hamel, M. Borgatti, R. Gambari, Synthesis and bio-
logical evaluation of 2-amino-3-(30,40,50-trimethoxybenzoyl)-5-aryl thio-
phenes as a new class of potent antitubulin agents, J. Med. Chem. 49 (2006)
3906e3915;
(c) R. Romagnoli, P.G. Baraldi, V. Remusat, M.D. Carrion, C.L. Cara, D. Preti,
F. Fruttarolo, M.G. Pavani, M.A. Tabrizi, M. Tolomeo, S. Grimaudo, J. Balzarini,
M.A. Jordan, E. Hamel, Synthesis and biological evaluation of 2-(30,40,50-tri-
[41] R. Bai, K.D. Paull, C.L. Herald, L. Malspeis, G.R. Pettit, E. Hamel, Halichondrin B
and homohalichondrin B, marine natural products binding in the vinca
domain of tubulin. Discovery of tubulin-based mechanism of action by anal-
ysis of differential cytotoxicity data, J. Biol. Chem. 266 (1991) 15882e15889.
[42] K.D. Paull, C.M. Lin, L. Malspeis, E. Hamel, Identification of novel antimitotic
agents acting at the tubulin level by computer-assisted evaluation of differ-
ential cytotoxicity data, Cancer Res. 52 (1992) 3892e3900.
[43] F. Bellina, S. Cauteruccio, S. Montib, R. Rossia, Novel imidazole-based com-
bretastatin A-4 analogues: evaluation of their in vitro antitumor activity and
molecular modeling study of their binding to the colchicine site of tubulin,
Bioorg. Med. Chem. Lett. 16 (2006) 5757e5762.
[44] A. Minocha, B.H. Long, Inhibition of the DNA catenation activity of type II
topoisomerase by VP16-213 and VM26, Biochem. Biophys. Res. Commun. 122
(1984) 165e170.
[45] P. Meresse, E. Dechaux, C. Monneret, E. Bertounesque, Etoposide: discovery
and medicinal chemistry, Curr. Med. Chem. 11 (2004) 2443e2466.
[46] H. Xu, M. Lv, X. Tian, Review on hemisynthesis, biosynthesis, biological
activities, mode of action, and structure-activity relationship of podophyllo-
toxins: 2003e2007, Curr. Med. Chem. 16 (2009) 327e349.
methoxybenzoyl)-3-amino 5-aryl thiophenes as
a new class of tubulin
inhibitors, J. Med. Chem. 49 (2006) 6425e6428.
[23] (a) C. Alvarez, R. Alvarez, P. Corchete, C. Perez-Melero, R. Pelaez, M. Medarde,
Synthesis and biological activity of naphthalene analogues of phenstatins:
naphthylphenstatins, Bioorg. Med. Chem. Lett. 17 (2007) 3417e3420;
(b) C. Alvarez, R. Alvarez, P. Corchete, C. Perez-Melero, R. Pelaez, M. Medarde,
Naphthylphenstatins as tubulin ligands: synthesis and biological evaluation,
Bioorg. Med. Chem. 16 (2008) 8999e9008.
[24] S. Jiang, C. Crogan-Grundy, J. Drewe, B. Tseng, S.X. Cai, Discovery of (naph-
thalen-4-yl)(phenyl)methanones
amines as new apoptosis inducers using
and
N-methyl-N-phenylnaphthalen-1-
cell- and caspase-based HTS
a
assay, Bioorg. Med. Chem. Lett. 18 (2008) 5725e5728.
[47] F. Zavala, D. Guenard, J.-P. Robin, E. Brown, Structureeantitubulin activity rela-
tionships in steganacin congeners and analogues. Inhibition of tubulin polymer-
ization in vitro by isodeoxypodophyllotoxin, J. Med. Chem. 23 (1980) 546e549.
[48] Maestro, version 9.1, Glide, version 5.6, MacroModel, version 9.8, Schrödinger,
LLC, New York, NY, 2010.
[25] S. Ducki, G. MacKenzie, B. Greedy, S. Armitage, J. Fournier Dit Chabert,
E. Bennett, J. Nettles, J.P. Snyder, N.J. Lawrence, Combretastatin-like chalcones
as inhibitors of microtubule polymerisation. Part 2: structure-based discovery
of alpha-aryl chalcones, Bioorg. Med. Chem. 17 (2009) 7711e7722.
[26] R. Romagnoli, P.G. Baraldi, M.D. Carrion, O. Cruz-Lopez, M. Tolomeo,
S. Grimaudo, A. Di Cristina, M.R. Pipitone, J. Balzarini, A. Brancale, E. Hamel,
Substituted 2-(30,40,50-trimethoxybenzoyl)-benzo[b]thiophene derivatives as
[49] Available from: http://www.molecular-networks.com/products/corina.
[50] J. Sadowski, J. Gasteiger, From atoms and bonds to three-dimensional atomic
coordinates: automatic model builders, Chem. Rev. 93 (1993) 2567e2581.