Combretastatin-like chalcones as inhibitors of microtubule polymerisation. Part 2: Structure-based discovery of alpha-aryl chalcones

Article abstract of DOI:10.1016/j.bmc.2009.09.044

Tubulin is an important molecular target in cancer chemotherapy. Antimitotic agents able to bind to the protein are currently under study, commonly used in the clinic to treat a variety of cancers and/or exploited as probes to investigate the protein's st

Full text of DOI:10.1016/j.bmc.2009.09.044

Bioorganic & Medicinal Chemistry 17 (2009) 7711–7722
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Combretastatin-like chalcones as inhibitors of microtubule polymerisation.
Part 2: Structure-based discovery of alpha-aryl chalcones
c
e
e
c,d
Sylvie Ducki a,b,c,d, , Grant Mackenzie , Ben Greedy , Simon Armitage , Jérémie Fournier Dit Chabert
,
*
Elizabeth Bennett c, Jim Nettles f, James P. Snyder f, Nicholas J. Lawrence a,e,
*
a Department of Chemistry, UMIST, Manchester M60 1QD, UK
b CRC Drug Development Group, Paterson Institute for Cancer Research, Christie Hospital NHS Trust, Manchester M20 4BX, UK
c Centre for Molecular Drug Design, University of Salford, Manchester M5 4WT, UK
d Laboratoire de Chimie des Hétérocycles et des Glucides, ENSCCF, Clermont Université, 63174 Aubière, France
e Drug Discovery Program, H. Lee Moffitt Cancer Center & Research Institute, Department of Interdisciplinary Oncology, University of South Florida, Tampa, FL 33612, USA
f Department of Chemistry, Emory University, 1515 Dickey Drive, Atlanta, GA 30322, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 15 July 2009
Revised 10 September 2009
Accepted 21 September 2009
Available online 2 October 2009
Tubulin is an important molecular target in cancer chemotherapy. Antimitotic agents able to bind to the
protein are currently under study, commonly used in the clinic to treat a variety of cancers and/or
exploited as probes to investigate the protein’s structure and function. Here we report the binding modes
for a series of colchicinoids, combretastatin A4 and chalcones established from docking studies carried
out on the structure of tubulin in complex with colchicine. The proposed models, in agreement with pub-
lished biochemical data, show that combretastatin A4 binds to the colchicine site of b-tubulin and that
chalcones assume an orientation similar to that of podophyllotoxin. The models can be used to design
Keywords:
Chalcone
Tubulin
Combretastatin
Colchicine
Podophyllotoxin
Antivascular
Anticancer
Docking
a new class of podophyllotoxin mimics, the
a
-aryl chalcones, capable of binding to the colchicine-binding
site of b-tubulin with higher affinity.
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
taxanes and vinca alkaloids in clinical oncology. Interest in these
agents has been raised further by the discovery that combretasta-
tin A4 (CA4 2) displays potent and selective toxicity towards tu-
mour vasculature.2,3 This effect is most pronounced for agents
that bind at the colchicine-binding site.4 Combretastatin A4 phos-
phate (CA4P 3), a water soluble prodrug of CA4, is currently in
phase III clinical trials for the treatment of cancer and early results
are very promising.5 Antimitotic agents able to target the colchi-
cine-binding site have regained pharmaceutical interest and
according to the National Cancer Institute (NCI), they are consid-
ered as important lead structures for the development of new anti-
tumour agents.6
Tubulin has become an important target in cancer chemother-
apy.1 The heterodimeric protein plays an important role in the for-
mation of the mitotic spindle, which provides the structural
framework (microtubule) for the physical segregation of the chro-
mosomes during mitosis. Drugs that interfere with the protein’s
dynamic stability are currently under study, commonly used in
the clinic to treat a variety of cancers or exploited as probe to
investigate the protein’s structure and function. Three binding sites
have been identified, named taxane, vinca alkaloid and colchicine
1, after the ligands that occupy them. While agents populating
the taxane pocket, such as paclitaxel, induce the assembly of
microtubules and stabilise them, ligands occupying either the vin-
ca alkaloid or the colchicine-binding sites inhibit the assembly of
the protein into its polymer. These antimitotic agents have gener-
ated considerable interest due to the tremendous success of the
Our group has developed chalcones as an important class of
antimitotic agents.7 The lead compound, 4-hydroxyphenylbute-
none, was isolated from the Chinese mint, Scutellaria barbata.8
A
comprehensive structure–activity relationship (SAR) study was
conducted which led to the discovery of the chalcones which were
found to inhibit tubulin assembly by binding to the colchicine-
binding site of tubulin.9,10
a
-Methyl chalcone SD400 4 was found
to be the most active compound of the series (inhibition of tubulin
assembly, IC50 1.8 M for 4 vs 1.3 M for 2 and 2.6 M for 1).
* Corresponding authors.
E-mail addresses: Sylvie.DUCKI@univ-bpclermont.fr (S. Ducki), Lawrennj@mof
fitt.usf.edu (N.J. Lawrence).
l
l
l
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.09.044

7712
S. Ducki et al. / Bioorg. Med. Chem. 17 (2009) 7711–7722
a
-Methyl chalcone 4 inhibits cancer cell growth at nanomolar con-
with two tubulin heterodimers capped at the top, suggesting that
the microtubule regulatory role of stathmin is achieved by seques-
tering tubulin heterodimers, not allowing their incorporation into
the microtubule. The structure locates the colchicine-binding site
centrations [IC50 (K562) 0.21 nM for 4 compared to 1.4 nM for 2]
and binds to the colchicine-binding site of isolated tubulin ([3H]
colchicine remaining bound after drug competition 92%, same as
CA4 2).10
at the a/b interface on the b-subunit. This structure (1SA0) differs
In 1998, Nogales reported the 3.7 Å structure of the
a
b-tubulin-
substantially from the structure obtained by Nogales (1JFF). The
tubulin dimer is in a straight conformation in 1JFF and a curved
conformation in 1SA0. The colchicine-binding site in 1SA0 is
mostly buried in b-tubulin, in a previously inaccessible domain
in 1JFF. As a consequence this area was not considered in our initial
assessment of 1JFF for potential colchicine-binding sites (results
unpublished). Strands S8 and S9, loop T7 and helices H7 and H8
of b-tubulin surround DAMA-colchicine, which also interacts with
paclitaxel complex, obtained by electron crystallography (EC) of
zinc-induced tubulin sheets, providing a much awaited insight into
the protein’s structure and function.11 Ravelli et al. later deter-
mined the 3.5 Å structure of tubulin in complex with DAMA-col-
chicine and with the stathmin-like domain (SLD) of RB3 clearly
showing the location of the binding site and the conformation
adopted by colchicine.12 This model provides a structural frame-
work for understanding the inhibition of tubulin assembly and
its regulation by antimitotic agents at the molecular level. Our con-
tribution to this understanding recently resulted in a 5D-QSAR
model able to predict the biological activity of agents binding to
the colchicine-binding site.13
loop T5 of the neighbouring a-subunit.
In order to check that the selected docking process was ade-
quate for finding the correct binding mode of ligands binding to
the colchicine-binding site, colchicine was extracted from the ori-
ginal structure (1SA0) and re-docked using AUTODOCK.
18 The binding
mode of the lowest energy complex, post-docking minimisation,
was compared to the crystal structure (Fig. 1). The docking study
using AUTODOCK results in a binding mode for colchicine that is
almost identical to that found in the original X-ray structure
(rms 0.7 Å between the best scored conformers from docking and
X-ray structure 1SA0). The structure correlates well with biochem-
ical data showing that b318 tubulin isoforms have a reduced sen-
sitivity to colchicine and data by Bai showing that the C2-oxygen
atom of colchicine is close to Cysb241 and the C3-oxygen atom is
close to Cysb356.19
OH
O
O
O
A
O
A
7 R
B
D
O
A
B
C
E
3
O
4
O
B
O
3'
C
O
O
OR
O
10
O
O
O
O
4
O
COL 1, R = NHCOCH3
CA4 2, R = H
CA4P 3, R = P(O)O2Na2
PODO 5
O
O
O
O
O
R
A
3' OH
4
2.2. The podophyllotoxin binding site
O
B
O
O
N
Ravelli et al. also determined the 4.2 Å structure of the ternary
tubulin:podophyllotoxin:RB3-SLD complex (PDB 1SA1) and found
that podophyllotoxin (PODO 5) binds to tubulin at the same site
as COL 1 but adopts a different orientation.12 The binding mode
for PODO 5 was also evaluated using the procedure described
above and compared to the binding mode published by Ravelli
et al. (Fig. 2). It is important to note here that the docking was
SD400 4, R = CH3
7, R = H
MDL 6
We now wish to report on the binding mode of a series of mole-
cules able to occupy the colchicine-binding site. Colchicinoids, pod-
ophyllotoxin, combretastatin A4 and a series of chalcones were
docked into the colchicine-binding site of tubulin reported by Ravel-
li et al. (PDB 1SA0). The models were examined and correlated with
observed biochemical evidence to ascertain their reliability. The
models were used to design a new class of podophyllotoxin mimics,
the a-aryl chalcones, capable of binding to the colchicine-binding
site, causing G2/M arrest and inhibiting tubulin assembly.
2. Results and discussion
2.1. The colchicine-binding site
The antimitotic agent, colchicine (COL 1), binds to b-tubulin and
inhibits tubulin assembly. Since this observation in the early
1970s, the mechanistic details of this interaction have often been
addressed. Although the general toxicity of COL has hampered its
use in cancer chemotherapy, a large number of analogues have
been prepared and evaluated for their ability to inhibit tubulin
assembly in vitro, providing a useful insight into the QSAR (quan-
titative structure–activity relationship) of this class of ligands.14
Until recently attempts at localising the colchicine-binding site in
the tubulin dimer had led to conflicting results.15,16
The report of the structure of paclitaxel-stabilised
a,b-tubulin
by electron crystallography in 2001 by Nogales (PDB 1JFF)17 raised
interest in the use of computer modelling techniques to give an
structural insight into the protein. In 2004, Ravelli et al. reported
the structure of tubulin in complex with DAMA-colchicine and
with the stathmin-like domain (SLD) of RB3 (PDB 1SA0), at 3.5 Å
resolution.12 The structure shows the interaction of the RB3-SLD
Figure 1. Comparison of X-ray (PDB 1SA0; red) and docked (blue) COL 1 (rms
0.7 Å).

S. Ducki et al. / Bioorg. Med. Chem. 17 (2009) 7711–7722
7713
non-biased; no preliminary superposition was carried out. The
docked structure of PODO 5 was found to be very similar to the
original X-ray structure (rms 1.5 Å between the best scored con-
formers from docking and X-ray structure 1SA1).
It was not surprising to find that the trimethoxyphenyl ring of
COL 1 and PODO 5 are both buried in a hydrophobic pocket in
b-tubulin,20 surrounded by Cysb241, Alab250, Leub252, Leub255,
Alab354 and Cysb356. Various biochemical experiments had come
to this conclusion, including Chaudhuri, who found that both COL 1
and PODO 5 could inhibit the formation of crosslinks between
Cysb241 and Cysb356 by EBI.21 Although both COL and PODO pop-
ulate the same pocket on b-tubulin, the molecules do not com-
pletely overlap in the binding site. This had been observed in
competition studies where N-acetyl mescaline (NMA) presented a
high affinity for tubulin binding but was inhibited by both COL
and PODO and tropolone methyl ester (TME) had been shown to
inhibit COL but not PODO binding. These results suggested both li-
gands shared the site where the trimethoxyphenyl ring bound, but
differed in other binding features.22,23
2.3. Docking of combretastatin A4
Since its discovery in 1989,24 CA4 2 and its analogues have had
tremendous coverage in the literature for their ability to inhibit
tubulin assembly by binding to the colchicine-binding site.25–27
Several structure-based models have been developed to elucidate
important structure–activity relationship for the combretasta-
tins.28,29 In 2005, Kong et al. described a molecular model for tubu-
lin–combretastatin interactions and used it to design boronic acid
bioisosteres of CA4.30
CA4 2 was docked into the colchicine-binding site using our
docking methodology. AUTODOCK scores and visual inspection al-
lowed us to select the most probable binding mode (Fig. 3).
Although rings A and B of CA4 and rings A and C of COL occupy
similar orientation, CA4 appears to fit deeper into the b-subunit
of tubulin than in Kong’s proposed binding model. The three meth-
oxy groups in CA4 are 6.12, 3.61 and 3.33 Å (3-Oꢀ ꢀ ꢀS, 4-Oꢀ ꢀ ꢀS and
5-Oꢀ ꢀ ꢀS, respectively) from Cysb241, while the three methoxy
Figure 3. Comparison of CA4 (red) and COL (blue) binding models with tubulin
(PDB 1SA0).
groups of COL in 4.16, 3.25 and 5.44 Å (2-Oꢀ ꢀ ꢀS, 3-Oꢀ ꢀ ꢀS and 4-
Oꢀ ꢀ ꢀS, respectively) away from the same residue. Our model also
shows that the 3-hydroxyl group of the B-ring of CA4 is in close
proximity to the carbonyl group of Thr179 of the
a-tubulin sub-
unit, and that these may be involved in a hydrogen bond (Oꢀ ꢀ ꢀO
3.05 Å). The model also suggests that this hydroxyl moiety might
be involved in a hydrogen bond with the backbone of Lys352 on
b-tubulin (Oꢀ ꢀ ꢀN 4.59 Å), which is consistent with the fact that
CA4 induces GTPase activity.31 The same hydrogen bond is ob-
served with the C-10 carbonyl of COL analogues (Oꢀ ꢀ ꢀN 2.88 Å) that
induce the GTPase activity of the enzyme. Nguyen et al. recently
reported that the binding mode of CA4 was very similar to COL
and that the two ligands belonged the same pharmacophore group,
which is consistent with our finding.32
2.4. Docking of chalcones
Peyrot et al. reported that trans-1-(2,5-dimethoxy)-3-[4-
(dimethylamino)phenyl]-2-methyl-2-propen-1-one (MDL) 6 was
a powerful antimitotic agent, binding rapidly and reversibly to
the colchicine-binding site of tubulin33 while our group recently
reported that both chalcones MDL 6 and SD400 4 were able to bind
to the colchicine-binding site of b-tubulin, inhibiting its assembly
into microtubules.10 MDL and SD400 were docked into the colchi-
cine-binding site using our docking methodology. AUTODOCK scores
and visual inspection allowed us to select the most probable
binding mode. Both chalcones displayed very similar binding
modes (Fig. 4).
The proposed binding site for MDL and SD400 is located on
b-tubulin, at the interface with the
a-subunit. The polymethoxy-
phenyl ring of the three ligands (SD400, PODO and COL) adopt very
similar orientation, close to Cysb241 (SD400 4-Oꢀ ꢀ ꢀS = 3.45 Å,
PODO 4-Oꢀ ꢀ ꢀS = 3.40 Å, COL 3-Oꢀ ꢀ ꢀS = 3.25 Å) which is not surpris-
ing since the binding of chalcones can be inhibited by COL, MTC
and PODO. The importance of this trimethoxyphenyl moiety in
agents interacting with the colchicine-binding site of tubulin has
already been observed.1 In this model, we observe that the enone
bridge between the two aromatic moieties of the chalcone does
Figure 2. Comparison of X-ray (PDB 1SA1; red) and docked (blue) PODO 5 (rms
1.5 Å).

7714
S. Ducki et al. / Bioorg. Med. Chem. 17 (2009) 7711–7722
with the protein through the 30-OH group (ring B). The fact that the
analogues substituted with methoxy-, hydrogen or fluoro- at the
30-position weakly inhibit tubulin assembly reinforces the impor-
tant of this group and its spatial location in the ligand. We have al-
ready demonstrated the importance of this hydroxyl group in our
QSAR study.13
2.5. Structure-based design
The models presented suggest that CA4 binds to b-tubulin in a
similar orientation as COL, whereas the chalcones’ binding mode
was similar to that of PODO. CA4 and COL would therefore belong
to a different pharmacophore group than chalcones and PODO. The
model also suggests that the addition of a planar hydrophobic moi-
ety, such as a phenyl ring C, at the
a-position of the chalcone skel-
eton would mimic ring B of podophyllotoxin very well (Fig. 5).
2.6. Synthesis of a-aryl chalcones
Two routes were used to prepare the alpha-aryl chalcones. In
the first route (Scheme 1), alpha-aryl chalcones 12 were prepared
as single isomers through a Suzuki coupling while the second route
(Scheme 2) provided alpha-aryl chalcones 13 as mixtures of E/Z
isomers.
Chalcones 8a–c were obtained by condensation of 3,4,5-tri-
methoxyacetophenone and substituted benzaldehydes 9a–c in so-
dium hydroxide/methanol in high yields (82–95%). In the case of
chalcone 8c, the hydroxyl group was protected as a silyl ether with
TBDMSCl to give chalcone 8d (85% yield). Bromination (Br2, Et3N)
Figure 4. Comparison of SD400 (red), COL (blue), PODO (yellow) binding models
with tubulin (PDB 1SA0).
not overlap COL or CA4, but has the same orientation as the D-ring
of PODO 5. The C@O of enone 4 even resembles the position of the
C@O of PODO 5 (ring D), and the model suggests that it might be
involved in a hydrogen bond with the backbone of Asp-251 in
b-tubulin (C@OSD400ꢀ ꢀ ꢀH–NAsp-251 = 2.18 Å and C@OPODOꢀ ꢀ ꢀH–NAsp-
251 = 4.07 Å). The importance of this ketone is consistent with our
observation that reduction to its 1-OH analogue resulted in re-
duced cytotoxicity against the human K562 cell line (SD400
4:IC50 of 0.21 nM vs 1-OH analogue:IC50 of 90 nV) and loss of abil-
of chalcones 8a–b,d afforded
a-bromo-chalcones 10a–c (44–
73%), which underwent a Suzuki coupling with substituted phenyl
boronic acids 11a–f to provide alpha-aryl chalcones 12a–h in mod-
erate to good overall yields (26–84%) with high selectivity for the
E-isomer in all cases. The silyl protecting group of 12c was cleaved
with TBAF to afford alpha-aryl chalcone 12i in 95% yield.
The preparation of alpha-aryl chalcones 13a–o was achieved by
Knoevenagel condensations of a range of substituted benzalde-
hydes 9a–d with ketones 14a–c using piperidine as a base in
excellent yields (58–97%). The products from these reactions were
isolated as E/Z mixtures. Ketones 14a–c were prepared by Grignard
addition of substituted benzyl halide 15a–c to 3,4,5-trimethoxy-
benzaldehyde followed by PCC oxidation of the alcohol 16a–c
(53–73% over two steps).
ity of inhibit tubulin assembly (SD400 4:IC50 of 0.46
analogue:IC50 of >10 M).
lM vs 1-OH
l
The arrangement of ring C of COL and ring B of CA4 within the
binding pocket (close to Leu-352 on b-tubulin) has been associated
with induction of GTPase activity, presumably due to a conforma-
tional change caused by interaction with this part of the binding
pocket. Since MDL was shown to induce no GTPase activity, we
were not surprised to observe that neither SD400 nor MDL bound
to this region of the binding site. Instead the chalcones preferred to
adopt a conformation which favoured a hydrogen bond between
OH
O
O
OH
O
O
O
D
O
A
B
C
A
B
O
the 30-hydroxyl moiety (ring B) and the C@O of Thr179 on
a-tubu-
lin (O–Hꢀ ꢀ ꢀO@C 3.00 Å), much like the hydroxyl group on ring C of
PODO (O–Hꢀ ꢀ ꢀO@C 3.54 Å), which was also shown to induce no
GTPase activity.
O
C
O
E
O
O
O
α-aryl chalcone
Podophyllotoxin
The bioactive conformation adopted by the chalcones is worth
further discussion. We previously reported that SD400 4 was more
active than its a
-demethylated analogue 7.9,13 The minimum s-cis
and s-trans conformations for both systems were optimised using
the HF/6-31* level of theory in PC GAMESS version of the GAMESS
(US) QC package. The results of these calculations indicate that the
s-cis conformation (torsion angle O–C1–C2–C3 of the enone system
is 37.6°) is favoured by 2.30 kcal molꢁ1 in the case of chalcone 7,
but that the s-trans conformation (torsion angle O–C1–C2–C3 of
the enone system is 146.1°) is favoured by 1.85 kcal molꢁ1 for
the
a-methyl chalcone 4. We therefore attribute this difference
in activity to the fact that SD400 4 prefers to adopt a bioactive con-
formation closer to a s-trans conformation while chalcone 7 prefers
the energetically less demanding s-cis conformation. This choice in
conformation is ideal for SD400 as it allows additional interaction
Figure 5. Pharmacophoric Points between podophyllotoxin and a-aryl chalcones.

S. Ducki et al. / Bioorg. Med. Chem. 17 (2009) 7711–7722
7715
O
O
O
O
O
R
O
O
O
Br
OHC
R
O
O
O
(a)
(c) (d)
+
R
O
O
O
O
8a, R = H (95%)
8b, R = F (95%)
8c, R = OH (82%)
8d, R = OTBDMS (85%)
9a, R = H
10a, R = H (44%)
10b, R = F (73%)
10c, R = OTBDMS (66%)
9b, R = F
9c, R = OH
(b)
+
R2
R1
R3
O
OH R1
C
B
O
O
B
R2
R3
R4
R
(e)
A
HO
O
O
R4
11a, R2 = R3 = R4 = H, R1 = OH
11b, R1 = R2 = R3 = R4 = H
12a, R = R2 = R3 = R4 = H, R1 = OH (48%)
12b, R = F, R2 = R3 = R4 = H, R1 = OH (61%)
11c, R1 = R2 = R4 = H, R3 = OCH3
11d, R2 = R4 = H, R1 = R3 =OCH3
11e, R1 = R2 = R4 = H, R3 = N(CH3)2
11f, R1 = H, R2 = R3 = R4 = OCH3
12c, R = OTBDMS, R2 = R3 = R4 = H, R1 = OH (84%)
12d, R = R1 = R2 = R3 = R4 = H (43%)
12e, R = R1 = R2 = R4 = H, R3 = OCH3 (30%)
12f, R = R2 = R4 = H, R1 = R3 = OCH3 (7%)
12g, R = R1 = R2 = R4 = H, R3 = N(CH3)2 (26%)
12h, R = R1 = H, R2 = R3 = R4 = OCH3(61%)
12i, R = OH, R2 = R3 = R4 = H, R1 = OH (95%]
(f)
Scheme 1. Reagents and conditions: (a) 50% NaOH, MeOH, rt, 16 h; (b) TBDMSCl, iPr2EtN, DMF, reflux, 1 h; (C) Br2, CHCl3, 0 °C; (d) Et3N, reflux; (e) Ar-B(OH)2 11a–f, Pd2(dba)3,
PPh3, Et2NH, Tol/n-PrOH; (f) TBAF, THF, rt, 30 min.
R2
R2
R1
OH
R3
R4
R1
R3
R4
O
O
O
O
O
O
O
O
(a)
(b)
X
+
R
O
O
O
16a, R1= R2 = R3 = R4 = H (68%)
14a, R1= R2 = R3 = R4 = H (78%)
15a, X= Br, R = H
16b, R1= R2 = R4 = H, R3 = CH3 (91%)
15b, X = Cl, R = CH3
14b, R1= R2 = R4 = H, R3 = CH3 (80%)
14c, R1= R2 = R4 = H, R3 = OCH3 (76%)
15c, X = Cl, R = OCH3
16c, R1= R2 = R4 = H, R3 = OCH3 (75%)
+
R2
R1
R3
O
C
B
O
O
OHC
R5
R6
R4
R5
(c)
A
O
R6
13a, R1= R2 = R3 = R4 = R5 = H, R6 = OCH3
9a, R5 = H, R6 = OCH3
9b, R5 = F, R6 = OCH3
9c, R5 = OH, R6 = OCH3
9d, R5 = NO2, R6 = OCH3
9e, R5 = R6 = H
13b, R1= R2 = R3 = R4 = H, R5 = OH, R6 = OCH3
13c, R1= R2 = R3 = R4 = R5 = R6 = H
13d, R1= R2 = R3 = R4 = H,R5 = NO2, R6 = OCH3
13e, R1= R2 = R3 = R4 = H, R5 = F, R6 = OCH3
13f, R1= R2 = R4 = R5 = R6 = H, R3 = CH3
13g, R1= R2 = R4 = R5 = H, R3 = CH3 , R6 = OCH3
13h, R1= R2 = R4 = H, R3 = CH3, R5 = OH, R6 = OCH3
13i, R1= R2 = R4 = H, R3 = CH3, R5 = NO2, R6 = OCH3
13j, R1= R2 = R4 = H, R3 = CH3, R5 = F, R6 = OCH3
13k, R1= R2 = R4 = R5 = R6 = H, R3 = OCH3
13l, R1= R2 = R4 = R5 = H, R3 = R6 = OCH3
13m, R1= R2 = R4 = H, R5 = OH, R3 = R6 = OCH3
13n, R1= R2 = R4 = H, R5 = NO2, R3 = R6 = OCH3
13o, R1= R2 = R4 = H, R5 = F, R3 = R6 = OCH3
Scheme 2. (a) Mg, THF, reflux, 1 h; (b) PCC, DCM, rt, 3 h; (c) piperidine, benzene, reflux, 6–8 h.
2.7. Biological evaluation
ity followed the same pattern as previously reported with regards
to the substituents at position 3 on ring B, with hydroxylated
analogues fifteen times more active than analogues bearing R = F
or H, highlighting once again the importance of the 3-OH group
on ring B.13 The substitution pattern on ring C had little influence
on the biological activity although less bulky substituents were
preferred, as exemplified by 12f bearing one methoxy group (IC50
K562 = 90 nM) and 12h bearing three methoxy moieties (IC50
Chalcones 12–13 were first assessed in vitro for their ability to
inhibit cell proliferation against the human leukaemia K562 cell
line (Table 1).
As predicted by the model, the addition of the bulky phenyl ring
C had no adverse effect on the biological activities of the chalcones
and all analogues displayed potent bioactivity. The biological activ-

7716
S. Ducki et al. / Bioorg. Med. Chem. 17 (2009) 7711–7722
Table 1
is caused by their ability to bind to tubulin and inhibit its assembly
into microtubules. It also shows that the addition of a phenyl ring
alpha to the enone does not prevent the chalcone from binding to
tubulin.
Chalcones 12b, 12i and 13m were selected for further evalua-
tion. The compounds were evaluated against a series of cell lines
(murine leukaemia P388, human lymphoma L1210, human lung
A549, human ovarian A2780), in which they displayed very potent
activity, in the nanomolar range (Table 2).
Flow cytometry was then used to evaluate the effects of these
agents on the growth and division of cells, by measuring the
DNA content of eukaryotic cells (Table 2). It was evident that the
cell population of the untreated cells predominantly resided in
the G0–G1/S-phases of the cycle (77%). This contrasted significantly
with cells treated with chalcones 12b, 12i and 13m which ap-
peared to act by blocking the cells in mitosis, illustrated through
an accumulation of cells in the G2/M phase (46–75% compared to
23% for untreated cells). Such cells appeared to have the capacity
to replicate their DNA but were incapable to proceed through the
cell cycle to cell division. These results are consistent with those
obtained for classical tubulin-binding drugs. We noted that the
newly designed alpha-aryl chalcone 12i was able to block the cell
cycle in the G2/M phase more extensively than SD400 4 and CA4 2
(75% vs 56% and 63%).
Inhibition of cancer cell proliferation against the K562 human leukaemia cell line
(IC50, concentration required to inhibit 50% of cell proliferation), inhibition of tubulin
assembly (IC50, concentration required to inhibit 50% of tubulin assembly)
Compd
Yield (%)
E/Z ratio
IC50 (K562) nM
IC50 (tubulin) lM
12a
12b
12c
12d
12e
12f
12g
12h
12i
13a
13b
13c
13d
13e
13f
13g
13h
13i
13j
13k
13l
13m
13n
13o
4
48
61
84
43
30
37
26
61
1:0
1:0
1:0
1:0
1:0
1:0
1::0
1:0
1:0
370
60
380
20
40
90
>10
7.2
>10
N.D.
N.D.
N.D.
N.D.
N.D.
5.5
150
2200
49
240
35
6300
>10,000
790
2970
760
32
2430
570
4920
500
12
95 from 12c
91
74
80
90
87
93
77
95
58
72
97
88
90
67
67
1.7:1
1:0
9.5
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
2.6
N.D.
4.4
N.D.
N.D.
2.1
1.4:1
2:1
1.5:1
2:1
2:1
2:1
7:6
11:4
1:1
2:1
5:4
4:1
10:7
1:0
2660
450
4.4
N.D.
4.2
3.3
3. Conclusion
2
0:1
3.2
3.7
Using an experimentally determined structure for the colchi-
cine-binding site of tubulin as a template, we established the bind-
ing modes for a series of colchicinoinds, podophyllotoxin 5,
combretastatin A4 2 and chalcone SD400 4 deduced from docking
studies. Our models suggest that combretastatin A4 adopts a sim-
ilar orientation to that of colchicine when bound to b-tubulin. Our
modelling studies imply that the chalcones’ binding mode is very
similar to that of podophyllotoxin and that the two ligands would
therefore belong to the same pharmacophore group. The model
pointed toward the addition of a planar hydrophobic moiety, such
as a phenyl ring, at the alpha position of the chalcone skeleton
which would mimic ring B of podophyllotoxin very well.
We designed and synthesised a series of alpha-aryl chalcones
12–13 as mimics of podophyllotoxin to validate this model. During
our modelling studies, we noted that the B-rings of both chalcones
and combretastatin A4 did not occupy the same pocket in tubulin
and belonged to different pharmacophore groups. We imagined
that the alpha-aryl chalcones 12–13 could occupy both pockets
and display synergistic activity. We found that chalcone 13m
displayed the highest bioactivity against cancer cells in vitro
(IC50 K562 12 nM). Also chalcone 13m was able to inhibit tubulin
assembly at a higher degree than the reference compounds 2
K562 = 2.2 nM). Since two routes were taken to prepare the alpha-
aryl chalcones, we were also able to look at the influence of the
configuration of the double bond. For example, chalcone 12d
(IC50 K562 = 20 nM) was synthesised as a single E-isomer while
chalcone 13a (IC50 K562 = 240 nM) was obtained as a 10:17 mix-
ture of two isomers. It is clear, in this case, that the E-configuration
of the double bond is preferred for high activity.
The conformation of the enone system has also been found to be
an important factor for good activity. Here we designed 2-hydrox-
ylated chalcone 12i, where the 2-OH would be able to form an
intramolecular hydrogen bond with the C@O of the enone hence
favouring the S-trans configuration. However, compared to chal-
cone 13b (IC50 K562 = 35 nM), chalcone 12i (IC50 K562 = 49nM)
did not present any advantage in vitro. The most active chalcone
of the series was found to be chalcone 13m (IC50 K562 = 12 nM).
This activity represents that of the mixture of isomers since we
did not manage to separate them. This chalcone is only three times
less active than the most active chalcone reported so far, chalcone
2 (IC50 K562 = 4.4 nM) and bears the 3-OH and 4-OCH3 substitu-
ents on its B-ring and a 4-OCH3 on its C-ring.
(IC50 tubulin 3.3 lM) and 4 (IC50 tubulin 3.7 lM). This chalcone
Selected a-aryl chalcones were evaluated for their ability to in-
has the 3-OH and 4-OCH3 substituent on the B-ring, just like com-
bretastatin A4 and chalcone 4, while the C ring is substituted with
a 4-OCH3 and can be considered as chalcone:combretastatin hybrid
compound.
hibit tubulin assembly in vitro (Table 1). Here again the analogues
bearing 3-OH on ring B displayed the highest activity, with two
analogues 13h and 13m more effective than SD400 4 and CA4 2.
These results prove that the cytotoxic effect of the a-aryl chalcones
Table 2
Inhibition of cancer cell proliferation (IC50, concentration required to inhibit 50% of cell proliferation), flow cytometry (cell distribution,% cells in G0/G1, S, G2/M phases)
Compd
Inhibition of cell proliferation IC50 (nM)
Flow cytometry
S
P388
L1210
A549
A2780
G0/G1
G2/M
12b
12i
13m
4
530
28
20
26
3.1
60
36
160
11
140
50
680
10
40
34
10
1.8
1.9
10.6
6.5
10.8
10.7
14.0
47.9
43.6
18.1
34.6
32.8
23.3
28.8
45.7
75.4
54.6
56.4
62.7
23.3
2
2.3
7.8
Control

S. Ducki et al. / Bioorg. Med. Chem. 17 (2009) 7711–7722
7717
ulation was run for 1000 fs within the NTV ensemble at a temper-
ature of 20 K and a structure was extracted every 10 fs. The 140 fs
structure was chosen to take to the next stage of the structure
preparation. The energy had dropped from an initial potential en-
ergy of 17189.09 kcal molꢁ1 using the current force field setup to
13955.73 kcal molꢁ1, while the RMSD of the atoms in the backbone
is only 0.24 Å.
The next stage of the preparation of the deposited crystal struc-
ture was the relaxation of whole molecule when none of the atoms
are held fixed. The dynamics simulation was run using the
MMFF94 force field and charges with a distant dependant dielec-
tric of 3. The simulation was run for 5000 fs within the NTV ensem-
ble at a temperature of 20 K and a structure was extracted every
10 fs. The 240 fs structure was chosen to take to the next stage of
the structure preparation. The energy had dropped from an initial
potential energy of 13955.73 kcal molꢁ1 using the current force
field setup to 11520.24 kcal molꢁ1, while the RMSD of the atoms
is 0.5 Å.
4. Experimental section
4.1. Molecular modelling
The BIOPOLYMER and SYBYL (Tripos, St. Louis) programs were used to
build and visualise the models. All calculations in this work were
performed on a 2.4 GHz Pentium IV processor running Windows
2000. The docking calculations were carried out using AUTODOCK
3.0.5.18 Molecular mechanics calculations were performed using
the Molecular Operating Environment (MOE) 2004.03 molecular
modelling software. Quantum mechanical calculations were
performed using the PC GAMESS version of the GAMESS (US) QC
package.
4.1.1. Preparation of the structure
The 3.58 Å structure of tubulin in complex with DAMA-colchi-
cine and with the stathmin-like domain (SLD) of RB3 was down-
loaded (PDB 1SA0). Tubulin dimer 1 (chains A and B) was
extracted along with the associated GTP, GDP, DAMA-colchicine
and Mg. This was as the starting point for structural relaxation
using the MMFF94 force field. There were several residues missing
from the crystal structure due to weakly defined electron densities.
The final step of the preparation process was a full minimisation
of the system. The MMFF94s force field and charges was used with
a distant dependant dielectric of 3. The minimisation completed
when the energy converged to a RMSD of 0.05 kcal molꢁ1. This
minimised structure was chosen to be the final relaxed structure.
These were residues 38–46 and 438–451 of a-tubulin, and residues
278–285 and 439–455 of b-tubulin. Five percentage of the side
chains were poorly defined and modelled as alanines in the down-
loaded structure. DAMA-colchicine, GTP and GDP were given the
appropriate atom types; the Tripos force field atom types allocated
to the protein by SYBYL were recognised and converted to MMFF94
atom types. The hydrogens were added using SYBYL’s BIOPOLYMER. The
missing residues were added to the structure by homology to the
equivalent residues in the minimised 1JFF structure. The mini-
mised structure derived from 1JFF was aligned to the end of the
chains of the missing residues to obtain the correct orientation.
The required residues were extracted from the 1JFF structure and
merged into the 1SA0 structure. These added residues were mini-
mised to a RMS of 0.5 kcal molꢁ1 using the MMFF94 force field and
a distant dependant dielectric of 3. During the minimisation runs
the rest of the system was held fixed as an aggregate. The missing
side chains were then added by mutating the alanine residues to
the required residue using the SYBYL BIOPOLYMER module. These were
then minimised to a RMS of 0.05 kcal molꢁ1 using the MMFF94
force field and a distant dependant dielectric of 3 while the rest
of the system was held fixed as an aggregate. The relaxation of
the crystal structure was approached in a step-wise manner using
low-temperature dynamics to relieve bad contacts from the depos-
ited crystal structure but to retain the secondary structure of the
protein. The side chains were relaxed first with the rest of the
atoms fixed as an aggregate. A constant dielectric of 78 is used to
stop the electrostatic interaction between the charged side chains
being exaggerated. The simulation of the protein is performed in a
vacuum so this dielectric prevents the unrealistic movement of the
side chains due to unshielded electrostatic interaction. The dynam-
ics simulation was run in SYBYL 6.9 using the MMFF94 force field
and charges. The simulation was run for 1000 fs within the NTV
ensemble at a temperature of 20 K and a structure was extracted
every 10 fs. The 140 fs structure was chosen to take to the next
stage of the structure preparation. The energy had dropped from
an initial potential energy of 37299.56 kcal molꢁ1 using the current
force field setup to 19200.40 kcal molꢁ1, while the RMSD of the
side-chain atoms is only of 0.38 Å.
4.1.2. Docking
The minimised structure was used in a series of docking calcu-
lations using the AUTODOCK 3.0.5 program. The coordinates of the
minimised structure were used but the molecular partial charges
were calculated with a different method. The scoring function
used in AUTODOCK 3.0.5 is based on the Amber force field. Because
of this it was decided that the use of Kollman ESP derived partial
charges would be more appropriate. MOE was used to allocate the
Kollman charges (AMBER 94 in MOE). AUTODOCK tools were then
used to produce the required files for grid generation and running
the AUTODOCK calculation. A grid of 60 ꢂ 60 ꢂ 60 was used to ensure
that the area probed was adequate for the ligands to explore all
possible binding modes. The AUTODOCK results were clustered using
a RMS tolerance of 2.0 Å to distinguish between similar binding
modes. Low energy binding modes and the lowest energy confor-
mation from clusters of 5 or more occurrences were selected for
further refinement. These were then refined using the MMFF94
force field, with the binding mode that had the lowest complex
energy after minimisation being selected as the final binding mode
taken forward for analysis and comparison to experimentally ob-
served properties.
4.1.3. Conformational search
Conformational searches for the ligands under investigation
were performed with the stochastic search method, as imple-
mented in MOE. In these calculations all rotational bond were ro-
tated to a random dihedral angle with a bias added to select
angles with a sum-of-gaussians distribution with peaks at multi-
ples of 30°. Following the rotation of the bonds a dihedral minimi-
sation was performed, to an RMS gradient of 100 kcal/mol Å, to
relieve bad non-bonded contacts. Subsequent Cartesian minimisa-
tion, to an RMS gradient of 0.001 kcal/mol Å, generates the final
conformation, which is checked to determine if it is unique by
comparing all atom positions using an RMS tolerance of 0.1 Å.
The procedure was terminated when the number of failures to find
a new conformation exceeded 10,000 in a row and conformations
with energies over 20 kcal/mol above the minima were rejected.
The quantum mechanical optimisation was performed in PC
GAMESS using the restricted Hartree–Fock (RHF) Hamiltonian
and Pople’s 6-31G* split valence basis set with polarisation func-
tions on heavy atoms. Input coordinates were taken from the sto-
chastic conformational searches.
Using the structure from the last dynamics calculation as input,
the next stage was to relieve bad contacts in the backbone of the
protein. In this simulation the backbone atoms and all atoms with-
in 2 Å of these were allowed to move with all others being held
fixed. The dynamics simulation was run using the MMFF94 force
field and charges with a distant dependant dielectric of 3. The sim-

7718
S. Ducki et al. / Bioorg. Med. Chem. 17 (2009) 7711–7722
4.2. Chemistry
4.2.2. General procedure for a-bromination of chalcones
A standardised solution of bromine in chloroform (17 mL,
1 mol Lꢁ1) was added dropwise to the chalcone 8a, 8b or 8d
(17 mmol) in chloroform (90 mL) at 0 °C while being vigorously
stirred under a nitrogen atmosphere. The solution turns from yel-
low to red/brown. After 0.5 h at 0 °C, chloroform was removed in
400 MHz 1H and 100 MHz 13C NMR spectra were recorded
using a Bruker AC 400 Spectrometer. 13C NMR spectra were re-
corded using Distortionless Enhancement by Polarisation Transfer.
Both 1H and 13C spectra were recorded using CDCl3 as internal
standard unless otherwise indicated. Chemical ionisation (CI)
mass spectra were recorded using a Kratos MS25 mass spectrom-
eter. Accurate mass determinations were carried out on a Kratos
Concept IS spectrometer. Melting points were determined using
an electrothermal melting point apparatus and were uncorrected.
Column chromatography was conducted using Silica Gel 60 230–
400 mesh (Merck & Co.). Silica TLC was conducted on pre-coated
aluminium sheets (60 F254) with a 0.2 mm thickness (Aldrich
Chemical Co.).
vacuo leaving
a red solid (2,3-dibrominated). Triethylamine
(50 mL) was added slowly to a stirred solution of 2,3-dibromo-
chalcone (5 mmol) in chloroform (5 mL) at 0 °C. The mixture was
heated under reflux for 1 h and solvents were removed in vacuo.
The resulting cream powder was dissolved in chloroform (ca.
30 mL) and washed with water (3 ꢂ 20 mL) and aqueous solution
of saturated sodium bicarbonate (2 ꢂ 20 mL). The organic layer
was dried over anhydrous MgSO4, filtered, evaporated and recrys-
tallised from ethyl acetate and hexane to afford the a-bromo-chal-
cones. (NB: These products are light-sensitive and isomerise
4.2.1. General procedure for the preparation of chalcones 8a–c
A stirred solution of 3,4,5-trimethoxyacetophenone (12 mmol)
and the substituted benzaldehyde 9a–c (12 mmol) in methanol
(30 mL) was treated at room temperature with a 50% methanol
NaOH solution (1 mL) and the resulting solution stirred at room
temperature for 24 h. The mixture was then acidified by addition
of 1 N HCl and the crude product extracted into dichloromethane
(3 ꢂ 20 mL). The organic phases were dried (MgSO4) and evapo-
rated in vacuo. The resulting solid was recrystallised from metha-
nol affording chalcones 8a–c in excellent yields (82–95%).
readily.)
4.2.2.1. (Z)-2-Bromo-3-(4-methoxy-phenyl)-1-(3,4,5-trimeth-
oxy-phenyl)-prop-2-en-1-one 10a. Yellow solid (44%); mp 113–
114 °C; dH/ppm (400 MHz, CDCl3) 3.87 (3H, s), 3.89 (6H, s), 3.94
(3H, s), 6.97 (2H, d, J 8.9 Hz), 7.03 (2H, s), 7.69 (1H, s), 7.91 (2H,
d, J 8.9 Hz); dc/ppm (100 MHz, CDCl3) 55.8 (CH3), 56.8 (2 ꢂ CH3),
61.4 (CH3), 107.7 (2 ꢂ CH), 114.4 (2 ꢂ CH), 120.1 (C), 126.4 (C),
132.2 (C), 132.9 (2 ꢂ CH), 142.3 (C), 142.7 (CH), 153.5 (2 ꢂ C),
81
Br
161.9 (C), 191.2 (C@O); m/z (CI) 409 (100%, M +H+), 407 (95,
M7B9r +H+).
4.2.1.1. (E)-3-(4-Methoxyphenyl)-1-(3,4,5-trimethoxyphenyl)-
prop-2-en-1-one 8a. Pale yellow crystals (95%); mp 101–103 °C;
dH/ppm (300 MHz, CDCl3) 3.89 (3H, s), 3.96 (3H, s), 3.98 (6H, s),
6.97 (2H, d, J 9.0 Hz), 7.30 (2H, s), 7.41 (1H, d, J 15.6 Hz), 7.65
(2H, d, J 9.0 Hz), 7.84 (1H, d, J 15.1 Hz).
4.2.2.2. (Z)-2-Bromo-3-(3-fluoro-4-methoxy-phenyl)-1-(3,4,5-
trimethoxy-phenyl)-prop-2-en-1-one 10b. White powder (73%);
mp 116–117 °C; dH/ppm (400 MHz, CDCl3) 3.90 (6H, s), 3.94 (6H, s),
7.01 (1H, m), 7.05 (2H, s), 7.56 (1H, d, J 12.6 Hz), 7.63 (1H, s), 7.86
(1H, dd, J 2.0; 12.6 Hz); dc/ppm (100 MHz, CDCl3) 56.6 (CH3), 56.7
(2 ꢂ CH3), 61.3 (CH3), 107.6 (2 ꢂ CH), 113.2 (CH, d, J 1 Hz), 117.7
(CH, d, J 20 Hz), 121.0 (C), 126.7 (C, d, J 7 Hz), 128.4 (CH, d, J
3 Hz), 131.7 (C), 141.0 (CH), 142.5 (C), 149.9 (C, d, J 11 Hz), 152.0
(C, d, J 247 Hz), 153.5 (2 ꢂ C), 190.8 (C@O); m/z (CI) 427 (98%,
M8B1r +H+), 425 (100, M7B9r +H+).
4.2.1.2. (E)-3-(3-Fluoro-4-methoxyphenyl)-1-(3,4,5-trimethoxy-
phenyl)prop-2-en-1-one 8b. Yellow solid (95%); mp 103–105 °C;
dH/ppm (300 MHz, CDCl3) 3.94 (6H, s), 3.96 (6H, s), 6.99 (1H, t, J
8.5 Hz), 7.27 (2H, s), 7.33–7.46 (2H, m), 7.35 (1H, J 15.8 Hz), 7.74
(1H, d, J 15.8 Hz).
4.2.1.3. (E)-3-(3-Hydroxy-4-methoxyphenyl)-1-(3,4,5-trimeth-
oxyphenyl)prop-2-en-1-one 8c. Yellow powder (82%); mp 144–
146 °C; dH/ppm (300 MHz, CDCl3) 3.93 (3H, s), 3.95 (9H, s), 5.78
(1H, s), 6.87 (1H, d, J 8.3 Hz), 7.13 (1H, dd, J 2.0, 8.3 Hz), 7.25–
7.30 (3H, m), 7.35 (1H, d, J 15.5 Hz), 7.75 (1H, d, J 15.5 Hz).
4.2.2.3. (Z)-2-Bromo-3-[3-(tert-butyl-dimethyl-silanyloxy)-4-
methoxy-phenyl]-1-(3,4,5-trimethoxy-phenyl)-prop-2-en-1-
one 10c. White powder (66%); mp 107–108 °C; dH/ppm (400 MHz,
CDCl3) 0.18 (6H, s), 1.00 (9H, s), 3.85 (3H, s), 3.88 (6H, s), 3.93 (3H,
s), 6.88 (1H, d, J 8.5 Hz), 7.02 (2H, s), 7.41 (1H, dd, J 2.2; 8.5 Hz),
7.62 (1H, s), 7.64 (1H, d, J 2.2 Hz); dc/ppm (100 MHz, CDCl3) ꢁ4.2
(2 ꢂ CH3), 18.8 (C), 26.1 (3 ꢂ CH3), 55.9 (CH3), 56.8 (2 ꢂ CH3),
61.4 (CH3), 107.6 (2 ꢂ CH), 111.8 (CH), 120.0 (C), 122.8 (CH),
4.2.1.4. (E)-3-[3-(tert-Butyl-dimethyl-silanyloxy)-4-methoxy-
phenyl)-phenyl]-1-(3,4,5-trimethoxy-phenyl)-prop-2-en-1-one
8d. To a solution of chalcone 8c (1.03 g, 3.0 mmol) in DMF
(10 mL), was added diisopropylethylamine (1.44 g, 3.0 mmol) and
tert-butyl-dimethylsilyl chloride (TBDMS-Cl) (1.36 g, 9.0 mmol).
The mixture was heated under reflux for 1 h and cooled down to
room temperature. Water (10 mL) was added to the mixture and
extracted with DCM (3 ꢂ 20 mL). The combined organic extracts
were washed with brine (30 mL), dried over anhydrous magne-
sium sulfate, filtered and evaporated in vacuo. The resulting yellow
solid was purified by flash chromatography (SiO2, ethyl ace-
tate:cyclohexane, 1:1) to afford chalcone 8d as a pale yellow pow-
der (1.17 g, 85% yield); mp 114–115 °C; dH/ppm (400 MHz, CDCl3)
0.21 (6H, s), 1.04 (9H, s), 3.86 (3H, s), 3.95 (6H, s), 3.96 (3H, s), 6.89
(1H, d, J 8.4 Hz), 7.19 (1H, d, J 2.1 Hz), 7.27 (1H, dd, J 2.1; 8.4 Hz),
7.29 (2H, s), 7.36 (1H, d, J 15.6 Hz), 7.74 (1H, d, J 15.6 Hz); dc/
ppm (100 MHz, CDCl3) ꢁ4.4 (2 ꢂ CH3), 18.6 (C), 26.0 (3 ꢂ CH3),
55.7 (CH3), 56.5 (2 ꢂ CH3), 61.1 (CH3), 106.3 (2 ꢂ CH), 112.1 (CH),
120.0 (CH), 120.6 (CH), 123.7 (CH), 128.3 (C), 134.1 (C), 142.5 (C),
144.8 (CH), 145.5 (C), 153.4 (2 ꢂ C), 153.7 (C), 189.3 (C@O); m/z
(CI) 459 (100%, M+H+), 401 (50, MꢁtBu).
126.4 (CH), 126.6 (C), 132.2 (C), 142.3 (C), 142.8 (CH), 145.1 (C),
81
153.3 (2 ꢂ C), 153.7 (C), 191.1 (C@O); m/z (CI) 539 (95%, M +H+),
Br
537 (100, M79+H+), 481 (50, MB81r ꢁtBu), 479 (55, M79ꢁtBu).
Br
Br
4.2.3. General procedure for the Suzuki coupling
A mixture of -bromo-chalcone 10a–c (1.0 mmol), substituted
a
phenyl boronic acid 11a–f (1.2 mmol), tris(dibenzylideneac-
tone)dipalladium(0) (23 mg, 0.022 mmol) and triphenylphosphine
(11 mg, 0.042 mmol) was dissolved in toluene (6 mL) and n-PrOH
(2 mL) and degassed. After stirring for 10 min, diethylamine
(0.13 mL, 1.3 mmol) and H2O (1.6 mL) were added. The mixture
was degassed a second time, heated under reflux for a 1 h, then
cooled down to room temperature and poured into ethyl acetate
(100 mL). The aqueous layer was separated, and the remaining or-
ganic layer was washed with 0.2 M NaOH (30 mL), 0.05 M HCl
(30 mL) and H2O (30 mL), dried over anhydrous MgSO4, and fil-
tered. The solvent was removed in vacuo, and the product was
purified by column chromatography (SiO2, ethyl acetate/hexane
1:3) and recrystallised from ethyl acetate/hexane to afford the
a-

S. Ducki et al. / Bioorg. Med. Chem. 17 (2009) 7711–7722
7719
aryl-chalcones 12a–h. Chalcones 12 were obtained exclusively as
E-isomers as ascertained by 1H NMR and X-ray crystallography.
The samples were stored away from the light to avoid
isomerisation.
s); dC/ppm (100 MHz; CDCl3) 55.27 (CH3), 55.28 (CH3), 56.21
(2 ꢂ CH3), 60.96 (CH3), 107.29 (2 ꢂ CH), 113.76 (2 ꢂ CH), 114.40
(2 ꢂ CH), 127.56 (C), 129.40 (C), 130.95 (2 ꢂ CH), 132.11
(2 ꢂ CH), 133.40 (C), 138.26 (C), 139.43 (CH), 141.32 (C), 152.70
(C), 159.19 (C), 160.14 (C), 196.61 (C@O); m/z (AP+) 435 (MH+,
100); HRMS found [MH]+ 435.1799, C26H27O6 requires [MH]+
435.1802.
4.2.3.1. (E)-2-(2-Hydroxy-phenyl)-3-(4-methoxy-phenyl)-1-
(3,4,5-trimethoxy-phenyl)-prop-2-en-1-one 12a. Yellow solid
(48%); mp 69–70 °C; dH/ppm (400 MHz, CDCl3) 3.67 (3H, s), 3.79
(6H, s), 3.85 (3H, s), 6.62 (2H, d, J 8.8 Hz), 6.78 (1H, ddd, J 0.9;
7.5; 7.5 Hz), 6.94 (2H, d, J 8.8 Hz), 6.97 (1H, dd, J 0.9; 7.5 Hz),
7.01 (1H, m), 7.12 (2H, s), 7.14 (1H, s), 7.18 (1H, m); dc/ppm
(100 MHz, CDCl3) 55.6 (CH3), 56.7 (2 ꢂ CH3), 61.3 (CH3), 108.2
(2 ꢂ CH), 114.3 (2 ꢂ CH), 118.5 (CH), 121.3 (CH), 124.3 (C), 127.1
(C), 130.6 (CH), 131.1 (CH), 132.3 (C), 132.4 (2 ꢂ CH), 135.4 (C),
141.4 (CH), 153.3 (2 ꢂ C), 153.3 (C), 155.0 (C), 160.9 (C), 199.3
(C@O); mmax (KBr) 3416 (OH), 1582 (C@O); m/z (CI) 421 (40,
M+H+), 128 (100); HRMS found [MH]+ 421.1649, C25H24O6 requires
[MH]+ 421.1646.
4.2.3.6. (E)-2-(2,4-Dimethoxyphenyl)-1-(3,4,5-trimethoxy-
phenyl)-3-(4-methoxyphenyl)prop-2-en-1-one 12f. Cream crys-
tals (37%); mp 124–126 °C; dH/ppm (400 MHz; CDCl3) 3.59 (3H, s),
3.78 (3H, s), 3.81 (6H, s), 3.83 (3H, s), 3.89 (3H, s), 6.44–6.51 (2H,
m), 6.73 (2H, d, J 8.6 Hz), 7.07 (2H, s), 7.10 (1H, d, J 8.0 Hz), 7.14
(2H, d, J 8.6 Hz), 7.23 (H, s); dC/ppm (100 MHz; CDCl3) 55.25
(CH3), 55.39 (CH3), 55.49 (CH3), 56.15 (2 ꢂ CH3), 60.93 (CH3),
99.07 (CH), 105.23 (CH), 106.96 (2 ꢂ CH), 113.63 (2 ꢂ CH),
119.44 (C), 128.04 (C), 131.69 (CH), 131.75 (2 ꢂ CH), 133.87 (C),
135.42 (C), 139.68 (CH), 141.00 (C), 152.52 (C), 158.05 (C),
159.98 (C), 161.08 (C), 197.09 (C@O); m/z (AP+) 465 (MH+, 100);
HRMS found [MH]+ 465.1901, C27H29O7 requires [MH]+ 465.1908.
4.2.3.2. (E)-3-(3-Fluoro-4-methoxy-phenyl)-2-(2-hydroxy-phe-
nyl)-1-(3,4,5-trimethoxy-phenyl)-prop-2-en-1-one 12b. Yellow
powder (61%); mp 96–97 °C; dH/ppm (400 MHz, CDCl3) 3.84 (3H,
s), 3.88 (6H, s), 3.95 (3H, s), 6.75–6.82 (2H, m), 6.89 (1H, m), 6.90
(1H, m), 6.97 (1H, m), 7.11 (1H, m), 7.13 (1H, s), 7.24 (2H, s),
7.29 (1H, m); dc/ppm (100 MHz, CDCl3) 56.5 (CH3), 56.7
(2 ꢂ CH3), 61.4 (CH3), 108.2 (2 ꢂ CH), 113.1 (CH), 117.6 (d, CH, J
19 Hz), 118.6 (CH), 121.5 (CH), 123.7 (C), 127.6 (d, C, J 9 Hz),
127.7 (CH), 130.8 (CH), 130.9 (CH), 131.8 (C), 136.7 (C), 139.6
(CH), 143.2 (C), 149.0 (d, C, J 11 Hz), 152.2 (d, C, J 251 Hz), 153.3
(2 ꢂ C), 154.9 (C), 200.2 (C@O); mmax (KBr) 3417 (OH), 1581
(C@O); m/z (EI) 438 (10%, M+), 195 (100); HRMS found [MH]+
439.1554, C25H24O6F requires [MH]+ 439.1551.
4.2.3.7. (E)-2-(4-(Dimethylamino)phenyl)-1-(3,4,5-trimethoxy-
phenyl)-3-(4-methoxyphenyl)prop-2-en-1-one 12g. Yellow crys-
tals (26%); mp 129–131 °C; dH/ppm (400 MHz; CDCl3) 2.97 (6H, s),
3.78 (3H, s), 3.80 (6H, s), 3.89 (3H, s), 6.68 (2H, d, J 8.8 Hz), 6.74
(2H, d, J 8.8 Hz), 7.07 (2H, s), 7.10–7.14 (3H, m), 7.18 (2H, d, J
8.8 Hz); dC/ppm (100 MHz; CDCl3) 40.79 (2 ꢂ CH3), 55.62 (CH3),
56.55 (2 ꢂ CH3), 61.30 (CH3), 107.76 (2 ꢂ CH), 112.97 (2 ꢂ CH),
114.04 (2 ꢂ CH), 125.07 (C), 128.52 (C), 130.89 (2 ꢂ CH), 132.32
(2 ꢂ CH), 133.82 (C), 137.82 (CH), 139.41 (C), 150.32 (C), 152.98
(C), 153.43 (C–N), 160.19 (C), 197.36 (C@O); m/z (AP+) 448 (MH+,
30), 447 (M, 100); HRMS found [MH]+ 448.2112, C27H30NO5 requires
[MH]+ 448.2118.
4.2.3.3. (E)-3-[3-(tert-Butyl-dimethyl-silanyloxy)-4-methoxy-
phenyl]-2-(2-hydroxy-phenyl)-1-(3,4,5-trimethoxy-phenyl)-
prop-2-en-1-one 12c. Yellow powder (84%); mp 122–123 °C; dH/
ppm (400 MHz, CDCl3) 0.01 (6H, s), 0.90 (9H, s), 3.78 (3H, s), 3.90
(6H, s), 3.96 (3H, s), 6.56 (1H, d, J 1.9 Hz), 6.70 (1H, d, J 8.5 Hz),
6.76 (1H, dd, J 1.9; 8.5 Hz), 6.89 (1H, dd, J 7.3; 7.6 Hz), 7.07 (1H,
d, J 7.9 Hz), 7.12 (1H, dd, J 1.3; 7.6 Hz), 7.14 (1H, s), 7.24 (2H, s),
4.2.3.8. (E)-2-(2,3,4-Trimethoxyphenyl)-1-(3,4,5-trimethoxy-
phenyl)-3-(4-methoxyphenyl)prop-2-en-1-one 12h. Brown oil
(61%); dH/ppm (400 MHz; CDCl3) 3.63 (3H, s), 3.71 (3H, s), 3.78
(6H, s), 3.78 (3H, s), 3.82 (3H, s), 3.84 (3H, s), 6.59 (H, d, J 8.6 Hz),
6.67 (2H, d, J 8.9 Hz), 6.82 (H, d, J 8.6 Hz), 7.04 (2H, d, J 8.9 Hz),
7.05 (2H, s), 7.15 (H, s); dC/ppm (100 MHz; CDCl3) 55.26 (CH3),
55.98 (CH3), 56.21 (2 ꢂ CH3), 60.69 (CH3), 60.91 (CH3), 60.93
(CH3), 107.15 (CH), 107.61 (2 ꢂ CH), 113.61 (2 ꢂ CH), 124.43 (C),
125.45 (CH), 127.56 (C), 131.86 (2 ꢂ CH), 133.82 (C), 135.38 (C),
140.39 (CH), 141.20 (C), 142.42 (C), 151.45 (C), 152.64 (C),
153.94 (C), 160.17 (C), 197.07 (C@O); m/z (AP+) 495 (MH+, 50),
117 (100); HRMS found [MH]+ 495.2008, C19H18O6 requires
[MH]+ 495.2013.
7.28 (1H, ddd,
J 1.3; 7.3; 7.9 Hz), 7.99 (1H, br s); dc/ppm
(100 MHz, CDCl3) ꢁ4.4 (2 ꢂ CH3), 18.7 (C), 26.0 (3 ꢂ CH3), 55.7
(CH3), 56.6 (2 ꢂ CH3), 61.3 (CH3), 108.1 (2 ꢂ CH), 111.7 (CH),
118.4 (CH), 121.3 (CH), 122.3 (CH), 124.3 (C), 126.1 (CH), 127.4
(C), 130.6 (CH), 131.1 (CH), 132.4 (C), 135.2 (C), 141.4 (CH), 142.8
(C), 145.0 (C), 152.8 (C), 153.2 (2 ꢂ C), 154.9 (C), 199.1 (C@O); m/
z (EI) 550 (10%, M+), 493 (40, MꢁtBu), 195 (100), HRMS found
[MH]+ 551.2466, C25H25O7 requires [MH]+ 551.2460.
4.2.3.9. (E)-3-(3-Hydroxy-4-methoxy-phenyl)-2-(2-hydroxy-
phenyl)-1-(3,4,5-trimethoxy-phenyl)-prop-2-en-1-one 12i. To a
solution of 12c (0.56 mmol) in dry THF (10 mL) was added a 1 M
solution of TBAF in THF (2 mmol). The yellow solution turned red
and was left to stir for 30 min, then neutralised with water
(20 mL). The aqueous solution was extracted with DCM
(3 ꢂ 20 mL) and the combined organic layers were washed with
water (20 mL)then brine (20 mL). Theorganic extract was dried over
anhydrous MgSO4, filtered and evaporated in vacuo. Purification by
flash chromatography (SiO2, ethyl acetate:cyclohexane, 1:3) affor-
ded chalcone 12i as a yellow foamy solid (95%); mp 87–88 °C; dH/
ppm (400 MHz, CDCl3) 3.75 (3H, s), 3.80 (6H, s), 3.86 (3H, s), 6.53–
6.59 (2H, m), 6.55 (1H, s), 6.79 (1H, ddd, J 1.1; 7.5; 7.6 Hz), 6.97
(1H, d, J 8.0 Hz), 7.03 (1H, d, J 7.6 Hz), 7.06 (1H, s), 7.15 (2H, s),
7.18 (1H, ddd, J 1.8; 7.5; 8.0 Hz); dc/ppm (100 MHz, CDCl3) 56.2
(CH3), 56.7 (2 ꢂ CH3), 61.4 (CH3), 108.2 (2 ꢂ CH), 110.7 (CH), 116.5
(CH), 118.6 (CH), 121.3 (CH), 123.9 (CH), 124.1 (C), 127.9 (C), 130.7
(CH), 131.0 (CH), 132.0 (C), 135.8 (C), 141.2 (CH), 143.1 (C), 145.7
4.2.3.4. (E)-1-(3,4,5-Trimethoxyphenyl)-3-(4-methoxyphenyl)-
2-phenylprop-2-en-1-one 12d. Creamy-coloured fine needles
(43%); mp 136–137 °C; dH/ppm (400 MHz; CDCl3) 3.70 (3H, s),
3.75 (6H, s), 3.83 (3H, s), 6.65 (2H, d, J 8.8 Hz), 6.98 (2H, s), 6.99
(2H, d, J 8.8 Hz), 7.19–7.31 (6H, m); dC/ppm (100 MHz; CDCl3)
55.28 (CH3), 56.21 (2 ꢂ CH3), 60.97 (CH3), 107.31 (2 ꢂ CH),
113.77 (2 ꢂ CH), 127.33 (C), 127.86 (2 ꢂ CH), 128.98 (CH), 129.74
(2 ꢂ CH), 132.22 (2 ꢂ CH), 133.31 (C), 137.33 (C), 138.58 (C),
140.02 (CH), 141.37 (C), 152.71 (C), 160.27 (C), 196.17 (C@O); m/
z (AP+) 405 (MH+, 100), HRMS found [MH]+ 405.1697, C25H25O5 re-
quires [MH]+ 405.1697.
4.2.3.5. (E)-1-(3,4,5-Trimethoxyphenyl)-2,3-bis(4-methoxy-
phenyl)prop-2-en-1-one 12e. Yellow solid (30%); mp 112–
114 °C; dH/ppm (400 MHz; CDCl3) 3.71 (3H, s), 3.75 (6H, s), 3.76
(3H, s), 3.83 (3H, s), 6.66 (2H, d, J 8.8 Hz), 6.83 (2H, d, J 8.7 Hz),
6.98 (2H, s), 7.03 (2H, d, J 8.8 Hz), 7.13 (2H, d, J 8.7 Hz), 7.16 (H,

7720
S. Ducki et al. / Bioorg. Med. Chem. 17 (2009) 7711–7722
(C), 148.2 (C), 153.3 (2 ꢂ C), 154.9 (C), 199.3 (C@O);
m
max (KBr) 3415
s), 6.85 (2H, d, J 8.6 Hz), 7.17 (2H, d, J 8.6 Hz), 7.24 (2H, s); dC/
ppm (100 MHz, CDCl3) 44.9, 55.4, 56.4, 61.1, 106.4, 114.4, 127.0,
130.5, 131.9, 142.7, 153.2, 158.8, 196.9; MS (ES+) 317.1 (100%,
[M+H]+); HRMS found [MH]+ 317.1384, C18H20O5 requires [MH]+
317.1384.
(OH), 1582 (C@O); m/z (EI) 434 (60%, M+), 195 (100); HRMS found
[MH]+ 437.1590, C25H24O7 requires [MH]+ 495.1595.
4.2.4. General procedure for the preparation of alcohols 16a–c
The substituted benzyl halide 15a–c (38 mmol) was added to a
mixture of magnesium turnings (38 mmol) and one crystal of iodine
in anhydrous diethyl ether (40 mL) and stirred under reflux for
30 min. The resulting grey solution was cooled to ꢁ78 °C and a solu-
tion of 3,4,5-trimethoxybenzaldehyde 9f (5 g, 25.5 mmol) in anhy-
drous THF (10 mL) was added over 10 min. The reaction was
allowed to warm to 0 °C and subsequently poured into a separating
funnel containing 1 N HCl (10 mL) and ice (10 g). The product was
extracted into ethyl acetate (3 ꢂ 20 mL) and the organic phases
dried over MgSO4. Removal of the solvent by rotary evaporation
afforded the crude product which was subjected to column chroma-
tography over silica gel. Further purification of the resulting solid by
recrystallisation afforded the alcohol in good yields (68–91%)
2-Phenyl-1-(3,4,5-trimethoxyphenyl)ethanol 16a as white crys-
tals (68%). Mp: 83–84 °C; dH/ppm (400 MHz, CDCl3) 2.97–3.00 (2H,
m), 3.83 (6H, s), 3.83 (3H, s), 4.81 (1H, dd, J 5.6; 7.5 Hz), 6.54 (2H,
s), 7.18–7.32 (5H, m); dC/ppm (100 MHz, CDCl3) 46.3, 56.3, 61.1,
75.7, 103.0, 126.9, 128.7, 129.8, 137.4, 138.2, 139.8, 153.4; MS
(ES+) 289.1 (30%, [M+H]+), 271.1 (100%, [MꢁH2O]+.
4.2.6. General procedure for the preparation of alpha-aryl
chalcones 13
A mixture of ketone 14a–c (0.7 mmol) and the appropriate
benzaldehyde 9a–e (0.7 mmol) in anhydrous benzene (2 mL) was
treated at room temperature with piperidine (5 lL). The reaction
was brought to reflux and water removed by passing the vapours
through a condenser packed with molecular sieves (4 Å, held in
place with glass wool). The reaction was followed by TLC for com-
pletion (6–8 h). On completion, the reaction mixture was poured
into water and the phases separated. The aqueous phase was ex-
tracted with ethyl acetate (3 ꢂ 20 mL) and the combined organic
phases dried over MgSO4. After filtration and evaporation of the
solvent in vacuo, the crude product was purified by column chro-
matography. Chalcones 13 were obtained as mixtures of E- and Z
-isomers. The ratio of each isomer was determined by NMR (1H
and HMBC and NOE) and X-ray crystallography. The samples were
stored away from the light to avoid isomerisation.
4.2.6.1. 3-(4-Methoxyphenyl)-2-phenyl-1-(3,4,5-trimethoxy-
phenyl)prop-2-en-1-one 13a. Yellow oil which solidified on
standing (91%, two isomers, ratio 10:17); dH/ppm (400 MHz, CDCl3)
3.61 (isomer 2, 3H, s), 3.63 (isomer 1, 3H, s), 3.67 (isomer 2, 6H, s),
3.70 (isomer 1, 6H, s), 3.77 (isomer 2, 3H, s), 3.80 (isomer 1, 3H, s),
6.59–6.61 (isomer 1, 2H, m), 6.62–6.65 (isomer 2, 2H, m), 6.97 (iso-
mer 2, 2H, s), 7.04 (isomer 1, 2H, s), 7.13–7.34 (isomer 1, 8H, m;
isomer 2, 8H, m); MS (ES+) 405.1 (100%, [M+H]+); HRMS found
[MH]+ 405.1706, C25H24O5 requires [MH]+ 405.1697.
2-(4-Methoxyphenyl)-1-(3,4,5-trimethoxyphenyl)ethanol 16c
as white crystals (75%). Mp: 120–121 °C; dH/ppm (400 MHz, CDCl3)
2.05 (1H, br s), 2.87–2.98 (2H, m), 3.79 (3H, s), 3.84 (9H, s), 4.77
(1H, dd, J 5.0; 8.0 Hz), 6.55 (2H, s), 6.85 (2H, d, J 8.5 Hz), 7.11
(2H, d, J 8,5 Hz); dC/ppm (100 MHz, CDCl3) 45.4, 55.5, 56.3, 61.1,
75.8, 103.0, 114.1, 130.1, 130.7, 137.4, 139.9, 153.4, 158.6; MS
(ES+) 319.1 (20%, [M+H]+), 301.1 (100%, [MꢁH2O]+).
4.2.5. General procedure for the preparation of ketones 14a–c
Pyridinium chlorochromate (11 mmol) was suspended in anhy-
drous DCM (10 mL) and a solution of alcohol 16a–c (7.3 mmol) in
DCM (10 mL) was rapidly added at ambient temperature. The solu-
tion became briefly homogeneous before depositing the reduced
reagent as a black insoluble solid around the walls of the reaction
flask. The reaction was stirred at room temperature and monitored
for completion by TLC (3–5 h). The black reaction mixture was di-
luted with DCM and filtered through Celite. The Celite was washed
several times with DCM and the filtrate concentrated under re-
duced pressure. The product was purified by column chromatogra-
phy over silica gel and the resulting solid recrystallised to afford
ketones 14a–c in good yields (76–80%).
4.2.6.2. 3-(3-Hydroxy-4-methoxyphenyl)-2-phenyl-1-(3,4,5-tri-
methoxyphenyl)prop-2-en-1-one 13b. Yellow oil which solidi-
fied on standing, further purified by recrystallisation from
hexane/ethyl acetate (74%, single isomer); dH/ppm (400 MHz,
CDCl3) 3.82 (6H, s), 3.86 (3H, s), 3.90 (3H, s), 5.46 (1H, s), 6.67
(1H, s), 6.68 (2H, s), 7.06 (2H, s), 7.23–7.39 (6H, m); MS (ES+) 421
(100%, [M+H]+); HRMS found [MH]+ 421.1649, C25H24O6 requires
[MH]+ 421.1646.
4.2.6.3. 2,3-diphenyl-1-(3,4,5-trimethoxyphenyl)prop-2-en-1-
one 13c. Sticky yellow oil which partially solidified on standing
(80%, two isomers, ratio 10:14); dH/ppm (400 MHz, CDCl3) 3.77
(isomer 2, 6H, s), 3.81 (isomer 1, 6H, s), 3.86 (isomer 2, 3H, s),
3.90 (isomer 1, 3H, s), 7.12–7.37 (isomer 1, 12H, m; isomer 2,
12H, m), 7.44–7.47 (isomer 1, 1H, m; isomer 2, 1H, m); MS (ES+)
375.1 (100%, [M+H]+); HRMS found [MH]+ 375.1597, C24H22O4 re-
quires [MH]+ 375.1591.
4.2.5.1. 2-Phenyl-1-(3,4,5-trimethoxyphenyl)ethanone
14a. Colourless crystals (78%). Mp: 95–96 °C; dH/ppm (400 MHz,
CDCl3) 3.88 (6H, s), 3.90 (3H, s), 4.25 (2H, s), 7.21–7.38 (7H, m);
dC/ppm (100 MHz, CDCl3) 45.9, 56.5, 61.2, 106.5, 127.2, 129.0,
129.5, 131.9, 135.1, 142.8, 153.2, 196.7; MS (ES+) 287.1 (100%,
[M+H]+); HRMS found [MH]+ 287.1285, C17H18O4 requires [MH]+
287.1278.
4.2.6.4. 3-(4-Methoxy-3-nitrophenyl)-2-phenyl-1-(3,4,5-trime-
thoxyphenyl)prop-2-en-1-one 13d. Thick yellow oil (90%, two
isomers, ratio 1:2); dH/ppm (400 MHz, CDCl3) 3.79 (isomer 2, 6H,
s), 3.80 (isomer 1, 6H, s), 3.88 (isomer 2, 3H, d, J 0.8 Hz), 3.90 (iso-
mer 1, 3H, s; isomer 2, 3H, s), 3.91 (isomer 1, 3H, s), 6.88 (isomer 1,
1H, d, J 8.8 Hz), 6.92 (isomer 2, 1H, d, J 8.8 Hz), 7.06–7.47 (isomer 1,
9H, m; isomer 2, 9H, m), 7.61 (isomer 1, 1H, d, J 2.0 Hz), 7.81 (iso-
mer 2, 1H, d, J 2.4 Hz); MS (ES+) 450.1 (100%, [M+H]+); HRMS found
[MH]+ 450.1553, C25H23NO7 requires [MH]+ 450.1547.
4.2.5.2. 2-p-Tolyl-1-(3,4,5-trimethoxyphenyl)ethanone
14b. Pale yellow powder (80%). Mp: 73–74 °C; dH/ppm (400 MHz,
CDCl3) 2.30 (3H, s), 3.86 (6H, s), 3.88 (3H, s), 4.19 (2H, s), 7.10 (4H,
m), 7.25 (2H, s); dC/ppm (100 MHz, CDCl3) 21.3, 45.5, 56.4, 61.1,
106.5, 129.3, 129.7, 131.9, 136.7, 142.7, 153.2, 153.9, 196.8; MS
(ES+) 301.1 (100%, [M+H]+); HRMS found [MH]+ 301.1443,
C18H20O4 requires [MH]+ 301.1434.
4.2.6.5. 3-(3-fluoro-4-methoxyphenyl)-2-phenyl-1-(3,4,5-tri-
methoxyphenyl)prop-2-en-1-one 13e. Thick yellow oil (87%,
two isomers, ratio 2:3); dH/ppm (400 MHz, CDCl3) 3.79 (isomer 2,
6H, s), 3.81 (isomer 1, 6H, s), 3.84 (isomer 2, 3H, s), 3.86 (isomer
4.2.5.3. 2-(4-Methoxyphenyl)-1-(3,4,5-trimethoxyphenyl)etha-
none 14c. Pale yellow powder (76%). Mp: 81–82 °C; dH/ppm
(400 MHz, CDCl3) 3.76 (3H, s), 3.87 (6H, s), 3.89 (3H, s), 4.17 (2H,

S. Ducki et al. / Bioorg. Med. Chem. 17 (2009) 7711–7722
7721
1, 3H, s), 3.89 (isomer 2, 3H, d, J 0.4 Hz), 3.90 (isomer 1, 3H, d, J
0.4 Hz), 6.76–7.43 (isomer 1, 11H, m; isomer 2, 11H, m); MS
(ES+) 423.1 (100%, [M+H]+); HRMS found [MH]+ 423.1610,
C25H23FO5 requires [MH]+ 423.1602.
3.82 (6H, s, isomer 2), 3.87 (3H, s, isomer 1), 3.88 (3H, s, isomer
2), 3.90 (3H, s, isomer 2), 6.73 (2H, d, J 8.8 Hz, isomer 1; 2H, d, J
8.8 Hz, isomer 2), 6.87 (2H, d, J 9.2 Hz, isomer 1), 6.90 (2H, d, J
8.6 Hz, isomer 2), 7.05 (2H, s, isomer 2), 7.06 (H, s, isomer 1),
7.10 (2H, d, J 8.6 Hz, isomer 2), 7.21 (2H, d, J 9.2 Hz, isomer 1),
7.19–7.26 (3H, m, isomer 2), 7.28 (2H, s, isomer 1), 7.35 (2H, d, J
8.8 Hz, isomer 1); MS (ES+) 435.1 (100%, [M+H]+); HRMS found
[MH]+ 435.1812, C26H26FO6 requires [MH]+ 435.1802.
4.2.6.6. 3-Phenyl-2-p-tolyl-1-(3,4,5-trimethoxyphenyl)prop-2-
en-1-one 13f. Thick yellow oil (93%, two isomers, ratio ꢃ2:1); dH/
ppm (400 MHz, CDCl3) 2.34 (isomer 1, 3H, s), 2.35 (isomer 2, 3H, s),
3.77 (isomer 1, 6H, s), 3.81 (isomer 1, 3H, s), 3.86 (isomer 2, 6H, s),
3.90 (isomer 2, 3H, s), 7.11–7.35 (isomer 1, 12H, m; isomer 2, 12H,
m); MS (ES+) 389.1 (100%, [M+H]+); HRMS found [MH]+ 389.1749,
C25H24O4 requires [MH]+ 389.1747.
4.2.6.13. 3-(3-Hydroxy-4-methoxyphenyl)-3-methoxyphenyl-1-
(3,4,5-trimethoxyphenyl)prop-2-en-1-one 13m. Thick yellow
oil (90%, two isomers, ratio ꢃ5:4); dH/ppm (400 MHz, CDCl3) 3.78
(6H, s, isomer 1), 3.79 (6H, s, isomer 2), 3.79 (3H, s, isomer 1),
3.80 (3H, s, isomer 2), 3.82 (3H, s, isomer 1), 3.83 (3H, s, isomer
2), 3.87 (3H, s, isomer 1), 3.90 (3H, s, isomer 2), 6.64–6.90 (2H, iso-
mer 1; 5H, isomer 2), 6.78 (1H, dd, J 1.8; 8.6 Hz isomer 1), 6.86 (2H,
d, J 8.8 Hz, isomer 1), 7.00 (1H, s, isomer 1), 7.06 (2H, s, isomer 2),
7.16 (1H, s, isomer 2), 7.20 (2H, d, J 9.2 Hz, isomer 2), 7.27 (2H, s,
isomer 1), 7.34 (2H, d, J 8.8 Hz, isomer 1); MS (ES+) 451.1 (100%,
[M+H]+); HRMS found [MH]+ 451.1755, C26H26O7 requires [MH]+
451.1751.
4.2.6.7. 3-(4-Methoxyphenyl)-2-p-tolyl-1-(3,4,5-trimethoxy-
phenyl)prop-2-en-1-one 13g. Thick yellow oil (77%, two isomers,
ratio ꢃ2:1); dH/ppm (400 MHz, CDCl3) 2.33 (3H, s isomer 1), 2.36
(3H, s, isomer 2), 3.73 (6H, s, isomer 2), 3.76 (3H, s, isomer 2),
3.78 (6H, s, isomer 1), 3.81 (3H, s, isomer 1), 3.87 (3H, s, isomer
1), 3.90 (3H, s, isomer 2), 6.71–6.74 (2H, m, isomer 2), 6.73 (2H,
d, J 7.6 Hz, isomer 1), 7.06–7.33 (9H, m, isomer 1; 9H, m, isomer
2); MS (ES+) 419.2 (100%, [M+H]+); HRMS found [MH]+ 419.1858,
C26H26O5 requires [MH]+ 419.1853.
4.2.6.14. 3-(4-Methoxy-3-nitrophenyl)-2-(4-methoxyphenyl)-1-
(3,4,5-trimethoxyphenyl)prop-2-en-1-one 13n. Thick yellow oil
(67%, two isomers, ratio ꢃ4:1); dH/ppm (400 MHz, CDCl3) 3.77–
3.90 (15H, isomer 1; 15H, isomer 2), 6.86–6.90 (3H, m, isomer 1;
3H, m, isomer 2), 6.98 (1H, s, isomer 2), 7.04 (2H, s, isomer 1),
7.08 (1H, s, isomer 1), 7.16 (2H, d, J 9.2 Hz, isomer 1), 7.22 (2H, s,
isomer 2), 7.27 (1H, dd, J 1.6; 9.2 Hz, isomer 1), 7.35 (2H, d, J
8.8 Hz, isomer 2), 7.42 (1H, dd, J 1.6; 8.6 Hz, isomer 2), 7.64 (1H,
d, J 1.6 Hz, isomer 1), 7.76 (1H, d, J 1.6 Hz, isomer 2); MS (ES+)
480.1 (55%, [M+H]+); HRMS found [MH]+ 480.1652, C26H25NO8 re-
quires [MH]+ 480.1653.
4.2.6.8. 3-(3-Hydroxy-4-methoxyphenyl)-2-p-tolyl-1-(3,4,5-tri-
methoxyphenyl)prop-2-en-1-one 13h. Thick yellow oil (138 mg,
95%); dH/ppm (400 MHz, CDCl3) 2.34 (3H, s, isomer 1), 2.36 (3H, s,
isomer 2), 3.76 (6H, s, isomer 2), 3.76 (3H, s, isomer 2), 3.79 (6H, s,
isomer 1), 3.81 (3H, s, isomer 1), 3.87 (3H, s, isomer 1), 3.91 (3H, s,
isomer 2), 6.71–6.74 (4H, m, isomers 1 and 2), 7.05–7.35 (8H, m,
isomer 1; 8H, m, isomer 2); MS (ES+) 435.1 (100%, [M+H]+); HRMS
found [MH]+ 435.1805, C26H26O6 requires [MH]+ 435.1802.
4.2.6.9. 3-(4-Methoxy-3-nitrophenyl)-2-p-tolyl-1-(3,4,5-trime-
thoxyphenyl)prop-2-en-1-one 13i. Thick yellow oil (58%, two
isomers, ratio 6:7); dH/ppm (400 MHz, CDCl3) 2.33 (isomer 2, 3H,
s), 2.35 (isomer 1, 3H, s), 3.78 and 3.79 (isomer 1, 6H, s; isomer
2, 6H, s), 3.87 (isomer 2, 3H, d, J 1.2 Hz), 3.88 (isomer 1, 3H, s; iso-
mer 2, 3H, s), 3.90 (isomer 1, 3H, s), 6.87–7.45 (isomer 1, 9H, m, H-
3; isomer 2, 9H, m), 7.63 (isomer 1, 1H, br s), 7.78 (isomer 2, 1H, br
s); MS (ES+) 464.1 (100%, [M+H]+); HRMS found [MH]+ 464.1711,
C26H25NO7 requires [MH]+ 464.1704.
4.2.6.15. 3-(3-Fluoro-4-methoxyphenyl)-2-(4-methoxyphenyl)-
1-(3,4,5-trimethoxyphenyl)prop-2-en-1-one 13o. Thick yellow
oil (67%, two isomers, ratio ꢃ10:7); dH/ppm (400 MHz, CDCl3)
3.78–3.80 (12H, isomer 1; 9H, isomer 2), 3.84 (3H, s, isomer 2),
3.87 (3H, s, isomer 1), 3.88 (3H, s, isomer 2), 6.74–7.01 (4H, m, iso-
mer 1; 5H, m, isomer 2), 6.97 (2H, d, J 7.4 Hz, isomer 1), 7.04 (2H, s,
isomer 2), 7.12 (1H, s, isomer 2), 7.17 (2H, d, J 7.2 Hz, isomer 2),
7.25 (2H, s, isomer 1), 7.33 (2H, d, J 7.4 Hz, isomer 1); MS (ES+)
453.1 (100%, [M+H]+); HRMS found [MH]+ 453.1716, C26H25FO6 re-
quires [MH]+ 453.1708.
4.2.6.10. 3-(3-Fluoro-4-methoxyphenyl)-2-p-tolyl-1-(3,4,5-tri-
methoxyphenyl)prop-2-en-1-one 13j. Thick yellow oil (72%, two
isomers, ratio 4:11); dH/ppm (400 MHz, CDCl3) 2.34 (isomer 2, 3H,
s), 2.37 (isomer 1, 3H, s), 3.79 (isomer 2, 6H, s), 3.81 (isomer 1, 6H,
s), 3.82 (isomer 2, 3H, s), 3.85 (isomer 1, 3H, s), 3.88 (isomer 2, 3H,
s), 3.92 (isomer 1, 3H, s), 6.76–7.32 (isomer 1, 10H, m; isomer 2,
10H, m); MS (ES+) 437.1 (100%, [M+H]+); HRMS found [MH]+
437.1769, C26H25FO5 requires [MH]+ 437.1759.
4.3. Biological evaluation
4.3.1. Cell inhibitory properties
Cells were cultured in RMPI medium, free of antibiotics and
containing 2-mercaptoethanol (2 lM) and L-glutamine (2 lM),
supplemented with foetal calf serum (FCS) (10% v/v). At monthly
intervals all cell lines were inspected for mycoplasma contamina-
tion by the staff of the Paterson Institute. The cells were adjusted
to a concentration depending on their observed doubling time
(ca. 2000 cells/mL), in RPMI medium supplemented with FCS
(10% v/v). The candidate drug was dissolved in DMSO. To 4 mL of
4.2.6.11. 2-(4-methoxyphenyl)-3-phenyl-1-(3,4,5-trimethoxy-
phenyl)prop-2-en-1-one 13k. Thick yellow oil (97%, two isomers,
ratio 1:1); dH/ppm (400 MHz, CDCl3) 3.78 (6H, s, isomer 1 or 2),
3.79 (3H, s, isomer 1 or 2), 3.80 (3H, s, isomer 1 or 2), 3.82 (6H,
s, isomer 1 or 2), 3.86 (3H, s, isomer 1 or 2), 3.90 (3H, s, isomer 1
or 2), 6.86 (2H, d, J 8.8 Hz, isomer 1 or 2), 6.88 (2H, d, J 8.4 Hz, iso-
mer 1 or 2), 7.10 (3H, s, isomer 1 or 2), 7.15–7.22 (11H, m, isomer 1
and 2), 7.27–7.29 (4H, m, isomer 1 or 2), 7.37 (1H, s, isomer 1 or 2),
7.39 (1H, s, isomer 1 or 2); MS (ES+) 405.1 (100%, [M+H]+); HRMS
found [MH]+ 405.1702, C25H24O5 requires [MH]+ 405.1697.
cell solution was added 4 lL of the drug solution and 1 mL of this
solution was added to 1 mL of cell solution in the adjacent tube,
giving a drug concentration half that of the first dilution. This series
of dilutions was continued to afford seven samples at different
concentrations leaving one cell solution free of drug acting as a
control. Aliquots of 200 lL of the treated cells were then pipetted
4.2.6.12. 2,3-Bis(4-methoxyphenyl)-1-(3,4,5-trimethoxy-
phenyl)prop-2-en-1-one 13l. Thick yellow oil (88%, two isomers,
ratio ꢃ2:1); dH/ppm (400 MHz, CDCl3) 3.73 (3H, s, isomer 1), 3.77
(3H, s, isomer 2), 3.78 (6H, s, isomer 1), 3.79 (3H, s, isomer 1),
in triplicate into a 96-well microtitre testplate and incubated
(37 °C, 5% CO2 in air) for 5 days. After this time, the plate was re-
moved from the incubator and 50
lL of a solution of MTT (3 mg/
mL in PBS) was added to each well. After incubation (37 °C, 5%

7722
S. Ducki et al. / Bioorg. Med. Chem. 17 (2009) 7711–7722
CO2 in air, 3 h) the medium was carefully removed from each well
by suction and the resulting formazan precipitate was dissolved in
grateful to Prof. Dennis C. Liotta (Emory University) for generous
assistance and hospitality during the course of the work.
200 lL of DMSO. The optical density of each well was then read at
two wavelengths (k 540 and 690 nm) using a Titretek Multiscan
MCC/340 platereader. After processing and analysis through the
application of an ‘in-house’ software package, the results obtained
enabled the calculation of the drug dose required to inhibit cell
growth by 50% (IC50 value), determined by graphical means as a
percentage of the control growth.
References and notes
1. Hadfield, J. A.; Ducki, S.; Hirst, N.; McGown, A. T. Tubulin and Microtubules as
target for Anticancer Drugs. In Progress in Cell Cycle Research; Meijer, L., Jezequel,
A., Roberge, M., Eds.; Kluwer Academic Plenum Publishers, 2003; Vol. 5, pp 309–
325.
2. Chaplin, D. J.; Hill, S. A.; Dougherty, G. J.; Prise, V. E.; Tozer, G. M.; Pettit, R.
Human Gene Ther. 1999, 10, 840.
3. Hill, S. A.; Tozer, G. M.; Pettit, G. R.; Chaplin, D. J. Anticancer Res. 2002, 22,
14538.
4. Marx, M. A. Exp. Opin. Ther. Pat. 2002, 12, 769.
5. Siemann, D. W.; Chaplin, D. J.; Walicke, P. A. Exp. Opin. Invest. Drugs 2009, 18,
189.
6. Leoni, L. M.; Hamel, E.; Genini, D.; Shih, H.; Carrera, C. J.; Cottam, H. B.; Carson,
D. A. J. Natl. Cancer. Inst. 2000, 92, 217.
7. Ducki, S. Idrugs 2007, 10, 42–46.
8. Ducki, S.; Hadfield, J. A.; Lawrence, N. J.; Liu, C. Y.; McGown, A. T.; Zhang, X. G.
Planta Med. 1996, 62, 185.
9. Ducki, S.; Forrest, R.; Hadfield, J. A.; Kendall, A.; Lawrence, N. J.; McGown, A. T.;
Rennison, D. Bioorg. Med. Chem. Lett. 1998, 8, 1051.
10. Lawrence, N. J.; McGown, A. T.; Ducki, S.; Hadfield, J. A. Anti-Cancer Drug Des.
2000, 15, 135.
4.3.2. Tubulin assembly assay
Six samples were prepared directly in quartz cuvettes at 0 °C
and contained Mes buffer [740
0.5 mM MgCl2, distilled water, pH 6.6)], GTP (10
Mes buffer), tubulin (150 L, 10–15 M in Mes buffer) and the can-
didate drug (10 L in DMSO). The tubulin/drug samples were
l
L (0.1 M Mes, 1 mM EGTA,
l
L, 100 mM in
l
l
l
immediately placed in a Varian Cary Bio 300 UV/visible spectro-
photometer, preheated at 37 °C, alongside six blank samples con-
taining Mes buffer (890
lL), GTP (100 lL) and the drug in DMSO
(10 L). Recording the absorbance (k 350 nm) for a period of
l
11. Nogales, E.; Wolf, S. G.; Downing, K. H. Nature 1998, 391, 199.
12. Ravelli, R. B. G.; Gigant, B.; Curmi, P. A.; Jourdain, I.; Lachkar, S.; Sobel, A. Nature
2004, 428, 198.
30 min, the results were compared to the untreated control cells
to evaluate the relative degree of change in optical density.
13. Ducki, S.; Mackenzie, G.; Lawrence, N. J.; Snyder, J. P. J. Med. Chem. 2005, 48,
457.
14. Polanski, J. Acta Biochim. Pol. 2000, 47, 37–45.
4.3.3. Flow cytometry
To K562 cells in RPMI medium supplemented with FCS (10% v/
15. Uppuluri, S.; Knipling, L.; Sackett, D. L.; Wolff, J. Proc. Natl. Acad. Sci. U.S.A. 1993,
90, 11598.
16. ter Haar, E.; Rosenkranz, H. S.; Hamel, E.; Day, B. W. Bioorg. Med. Chem. 1996, 4,
v) (10 mL, 2 ꢂ 105 cells/mL) was added the candidate drug (10
lL,
5000 ꢂ IC50 in DMSO). After incubation (37 °C, 5% CO2 in air, over-
night) the cells were centrifuged (1800 RPM, 10 min) and the
supernatant discarded. The resulting pellet was re-suspended in
a cold 1:1 mixture of acetone:ethanol (10 mL) and the mixture
was stored at 4 °C until analysis was performed. The fixed cells
were centrifuged (1800 rpm, 10 min) and the supernatant dis-
carded. The resulting cell pellet was re-suspended in ice-cold PBS
(10 mL) and centrifuged (1800 rpm, 10 min), again the supernatant
discarded. The next stage involved the incubation (37 °C, 15 min)
1659.
17. Lowe, J.; Li, H.; Downing, K. H.; Nogales, E. J. Mol. Biol. 2001, 313, 1045.
18. Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart, W. E.; Belew, R. K.;
Olson, A. J. J. Comput. Chem. 1998, 19, 1639.
19. Cerbai, G.; Turbanti, L.; Bianchini, P.; Bramanti, G.; Tellini, N. Boll. Chim. Farm.
1967, 106, 837.
20. Andreu, J. M.; Perez-Ramirez, B.; Gorbunoff, M. J.; Ayala, D.; Timasheff, S. N.
Biochemistry 1998, 37, 8356.
21. Chaudhuri, A. R.; Seetharamalu, P.; Schwarz, P. M.; Hausheer, F. H.; Luduena, R.
F. J. Mol. Biol. 2000, 303, 679.
22. Cortese, F.; Bhattacharyya, B.; Wolff, J. J. Biol. Chem. 1977, 252, 1134.
23. Dumortier, C.; Yan, Q. Y.; Bane, S.; Engelborghs, Y. Biochem. J. 1997, 327,
685.
24. Pettit, G. R.; Singh, S. B.; Hamel, E.; Lin, C. M.; Alberts, D. S.; Garciakendall, D.
Experientia 1989, 45, 209.
25. Tron, G. C.; Pirali, T.; Sorba, G.; Pagliai, F.; Busacca, S.; Genazzani, A. A. J. Med.
Chem. 2006, 49, 3033.
26. Woods, J. A.; Hadfield, J. A.; Pettit, G. R.; Fox, B. W.; McGown, A. T. Br. J. Cancer
1995, 71, 705.
of the cells with RNAse (150
from bovine pancreas). The cell pellet was dissolved a solution of
propidium iodide (0.25 mL, 50 g/mL in PBS) and transferred to
lL, 100 lg/mL in PBS, DNAse free,
l
flow cytometry tubes before analysis by flow cytometry. The intra-
cellular fluorescence was measured in a Coulter Epics V flow
cytometer, using an argon laser (kex 488 nm).
27. Simoni, D.; Romagnoli, R.; Baruchello, R.; Rondanin, R.; Rizzi, M.; Pavani, M. G.;
Alloatti, D.; Giannini, G.; Marcellini, M.; Riccioni, T.; Castorina, M.; Guglielmi,
M. B.; Bucci, F.; Carminati, P.; Pisano, C. J. Med. Chem. 2006, 49, 3143.
28. Brown, R. T.; Fox, B. W.; Hadfield, J. A.; McGown, A. T.; Mayalarp, S. P.; Pettit, G.
R.; Woods, J. A. J. Chem. Soc., Perkin Trans. 1 1995, 577.
29. Nandy, P.; Banerjee, S.; Gao, H.; Hui, M. B. V.; Lien, E. J. Pharm. Res. 1991, 8, 776.
30. Kong, Y.; Grembecka, J.; Edler, M. C.; Hamel, E.; Mooberry, S. L.; Sabat, M.;
Rieger, J.; Brown, M. L. Chem. Biol. 2005, 12, 1007.
31. Pandit, B.; Sun, Y. J.; Chen, P.; Sackett, D. L.; Hu, Z. G.; Rich, W.; Li, C. L.; Lewis,
A.; Schaefer, K.; Li, P. K. Bioorg. Med. Chem. 2006, 14, 6492.
32. Nguyen, T. L.; McGrath, C.; Hermone, A. R.; Burnett, J. C.; Zaharevitz, D. W.; Day,
B. W.; Wipf, P.; Hamel, E.; Gussio, R. J. Med. Chem. 2005, 48, 6107.
33. Peyrot, V.; Leynadier, D.; Sarrazin, M.; Briand, C.; Menendez, M.; Laynez, J.;
Andreu, J. M. Biochemistry 1992, 31, 11125.
Acknowledgments
The authors would like to acknowledge financial support from
BBSRC (106/B18017, B18009 and BB/C524389/1) for postdoctoral
fellowships for G.M., B.G. and J.F., a GTA award from UoS for E.B.,
a Bourse Innovation from the Conseil Régional d’Auvergne and
Fonds européen de développement regional (FEDER) for J.F. and
the Emerson Center for Scientific Computing (Emory University)
for a visiting fellowship award for S.D. We thank the EPSRC Na-
tional Mass Spectrometry Service Centre for HRMS. We are also