Design, synthesis and biological evaluation of novel vicinal diaryl-substituted 1H-Pyrazole analogues of combretastatin A-4 as highly potent tubulin polymerization inhibitors

Article abstract of DOI:10.1016/j.ejmech.2019.111577

A series of 3-(3′,4′,5′-trimethoxyphenyl)-4-substituted 1H-pyrazole and their related 3-aryl-4-(3′,4′,5′-trimethoxyphenyl)-1-H-pyrazole regioisomeric derivatives, prepared as cis-rigidified combretastatin A-4 (CA-4) analogues, were synthesized and evaluated for their in vitro antiproliferative against six different cancer cell lines and, for selected highly active compounds, inhibitory effects on tubulin polymerization, cell cycle effects and in vivo potency. We retained the 3′,4′,5′-trimethoxyphenyl moiety as ring A throughout the present investigation, and a structure-activity relationship (SAR) information was obtained by adding electron-withdrawing (OCF₃, CF₃) or electron-releasing (alkyl and alkoxy) groups on the second aryl ring, corresponding to the B-ring of CA-4, either at the 3- or 4-position of the pyrazole nucleus. In addition, the B-ring was replaced with a benzo[b]thien-2-yl moiety. For many of the compounds, their activity was greater than, or comparable with, that of CA-4. Maximal activity was observed with the two regioisomeric derivatives characterized by the presence of a 4-ethoxyphenyl and a 3′,4′,5′-trimethoxyphenyl group at the C-3 and C-4 positions, and vice versa, of the 1H-pyrazole ring. The data showed that the 3′,4′,5′-trimethoxyphenyl moiety can be moved from the 3- to the 4-position of the 1H-pyrazole ring without significantly affecting antiproliferative activity. The most active derivatives bound to the colchicine site of tubulin and inhibited tubulin polymerization at submicromolar concentrations. In vivo experiments, on an orthotopic murine mammary tumor, revealed that 4c inhibited tumor growth even at low concentrations (5 mg/kg) compared to CA-4P (30 mg/kg).

Full text of DOI:10.1016/j.ejmech.2019.111577

Journal Pre-proof
Design, synthesis and biological evaluation of novel vicinal diaryl-substituted 1H-
Pyrazole analogues of combretastatin A-4 as highly potent tubulin polymerization
inhibitors
Romeo Romagnoli, Paola Oliva, Maria Kimatrai Salvador, Maria Encarnacion
Camacho, Chiara Padroni, Andrea Brancale, Salvatore Ferla, Ernest Hamel,
Roberto Ronca, Elisabetta Grillo, Roberta Bortolozzi, Fatlum Rruga, Elena Mariotto,
Giampietro Viola
PII:
S0223-5234(19)30707-X
DOI:
https://doi.org/10.1016/j.ejmech.2019.111577
EJMECH 111577
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 26 June 2019
Revised Date: 30 July 2019
Accepted Date: 30 July 2019
Please cite this article as: R. Romagnoli, P. Oliva, M.K. Salvador, M.E. Camacho, C. Padroni, A.
Brancale, S. Ferla, E. Hamel, R. Ronca, E. Grillo, R. Bortolozzi, F. Rruga, E. Mariotto, G. Viola,
Design, synthesis and biological evaluation of novel vicinal diaryl-substituted 1H-Pyrazole analogues
of combretastatin A-4 as highly potent tubulin polymerization inhibitors, European Journal of Medicinal
Chemistry (2019), doi: https://doi.org/10.1016/j.ejmech.2019.111577.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
©
2019 Published by Elsevier Masson SAS.

Design, Synthesis and Biological Evaluation of Novel Vicinal Diaryl-Substituted
1
H-Pyrazole Analogues of Combretastatin A-4 as Highly Potent Tubulin
Polymerization Inhibitors
a
a
b
Romeo Romagnoli, * Paola Oliva, Maria Kimatrai Salvador, Maria Encarnacion
b
c
d
d
e
Camacho, Chiara Padroni, Andrea Brancale, Salvatore Ferla, Ernest Hamel,
Roberto Ronca, Elisabetta Grillo, Roberta Bortolozzi, Fatlum Rruga, Elena
f
f
g
g
g
g,h
Mariotto and Giampietro Viola
*
a
Dipartimento di Scienze Chimiche e Farmaceutiche, Università di Ferrara, Ferrara,
b
Italy; Departamento de Química Farmacéutica y Orgánica, Facultad de Farmacia,
Granada, Spain; Aptuit, an Evotec Company, Verona, Italy; School of Pharmacy
and Pharmaceutical Sciences, Cardiff University, Cardiff, UK; National Cancer
Institute, National Institutes of Health, Frederick, Maryland USA; Dipartimento di
c
d
e
f
Medicina Molecolare
e Traslazionale Università di Brescia, Brescia, Italy;
g
Dipartimento di Salute della Donna e del Bambino, Università di Padova, Padova,
h
Italy; Istituto di Ricerca Pediatrica (IRP), Padova, Italy
Comparing couple of regioisomeric derivatives 3p and 4c there was not substantial difference of antiproliferative as well as of tubulin
polymerization inhibitory activity between 3',4',5'-trimethoxyphenyl moiety located at the 3- or 4-position of the pyrazole ring.

Design, Synthesis and Biological Evaluation of Novel Vicinal Diaryl-
Substituted 1H-Pyrazole Analogues of Combretastatin A-4 as Highly
Potent Tubulin Polymerization Inhibitors
a
a
b
Romeo Romagnoli, * Paola Oliva, Maria Kimatrai Salvador, Maria Encarnacion
b
c
d
d
e
Camacho, Chiara Padroni, Andrea Brancale, Salvatore Ferla, Ernest Hamel, Roberto
f
f
g
g
g
Ronca, Elisabetta Grillo, Roberta Bortolozzi, Fatlum Rruga, Elena Mariotto and
g,h
Giampietro Viola *
a
Dipartimento di Scienze Chimiche e Farmaceutiche, Via Luigi Borsari 46, Università di
Ferrara, 44121 Ferrara, Italy;
b
Departamento de Química Farmacéutica y Orgánica, Facultad de Farmacia, Campus de
Cartuja s/n, 18071, Granada, Spain;
c
Aptuit, an Evotec Company, Via A. Fleming 4, 37135 Verona, Italy;
d
School of Pharmacy and Pharmaceutical Sciences, Cardiff University, King Edward
VII Avenue, Cardiff, CF10 3NB, UK;
e
Screening Technologies Branch, Developmental Therapeutics Program, Division of
Cancer Treatment and Diagnosis, Frederick National Laboratory for Cancer Research,
National Cancer Institute, National Institutes of Health, Frederick, Maryland 21702,
USA;
f
Dipartimento di Medicina Molecolare e Traslazionale Unità di Oncologia Sperimentale
ed Immunologia, Università di Brescia, 25123 Brescia, Italy;
g
Dipartimento di Salute della Donna e del Bambino, Laboratorio di Oncoematologia,
Università di Padova, 35131 Padova, Italy
h
Istituto di Ricerca Pediatrica (IRP), Corso Stati Uniti 4, 35128 Padova, Italy
1

*
To whom correspondence should be addressed. E-mail: rmr@unife.it; Phone: 39-
(0)532-455303. Fax: 39-(0)532-455953. (R.R.); E-mail: giampietro.viola.1@unipd.it
Phone: 39-(0)49-8215485. Fax: 39-(0)49-8211462. (G.V.).
2

Abstract: A series of 3-(3’,4’,5’-trimethoxyphenyl)-4-substituted 1H-pyrazole and their
related 3-aryl-4-(3’,4’,5’-trimethoxyphenyl)-1-H-pyrazole regioisomeric derivatives,
prepared as cis-rigidified combretastatin A-4 (CA-4) analogues, were synthesized and
evaluated for their in vitro antiproliferative against six different cancer cell lines and, for
selected highly active compounds, inhibitory effects on tubulin polymerization, cell
cycle effects and in vivo potency. We retained the 3’,4’,5’-trimethoxyphenyl moiety as
ring A throughout the present investigation, and a structure-activity relationship (SAR)
information was obtained by adding electron-withdrawing (OCF , CF ) or electron-
3
3
releasing (alkyl and alkoxy) groups on the second aryl ring, corresponding to the B-ring
of CA-4, either at the 3- or 4-position of the pyrazole nucleus. In addition, the B-ring
was replaced with a benzo[b]thien-2-yl moiety. For many of the compounds, their
activity was greater than, or comparable with, that of CA-4. Maximal activity was
observed with the two regioisomeric derivatives characterized by the presence of a 4-
ethoxyphenyl and a 3’,4’,5’-trimethoxyphenyl group at the C-3 and C-4 positions, and
vice versa, of the 1H-pyrazole ring. The data showed that the 3’,4’,5’-trimethoxyphenyl
moiety can be moved from the 3- to the 4-position of the 1H-pyrazole ring without
significantly affecting antiproliferative activity. The most active derivatives bound to the
colchicine site of tubulin and inhibited tubulin polymerization at submicromolar
concentrations. In vivo experiments, on an orthotopic murine mammary tumor, revealed
that 4c inhibited tumor growth even at low concentrations (5 mg/kg) compared to CA-
4
P (30 mg/kg).
Keywords. Microtubules, structure-activity relationship, 1H-pyrazole, antiproliferative
activity, tubulin.
3

1
. Introduction
Microtubules are cylindrical structures formed by a polymerization process from αβ-
tubulin heterodimers and are dynamic components of the cytoskeleton of eukaryotic
cells. Their dynamic functions include the formation and disappearance of the mitotic
spindle, which is required for the separation of duplicated chromosomes during cell
division [1]. The microtubule system is involved in a number of other cellular functions
as well, including determination and maintenance of cell shape, organization of
intracellular architecture, secretion, cellular transport, regulation of motility and
intracellular organelle transport [2].
Therefore, especially since tubulin is a validated target for drugs active against cancer,
there has been great interest in identifying and developing new tubulin polymerization
inhibitors. Inhibition of microtubule function using agents that interact specifically with
tubulin, many of which are natural products, remains one approach for the development
of new cytotoxic anticancer drugs [3]. Among the naturally occurring antimitotic agents,
combretastatin A-4 (CA-4, 1a; Figure 1), isolated from the bark of the South African
tree Combretum caffrum [4], is one of the best known inhibitors of tubulin assembly.
CA-4 affects microtubule dynamics by interacting with tubulin at the colchicine site [5].
CA-4 shows potent cytotoxicity against a wide variety of human cancer cell lines,
including those that are multidrug resistant [6]. Due to the low water solubility of CA-4,
the corresponding disodium phosphate salt (CA-4P, 1b) was synthesized as a water
soluble pro-drug. CA-4P causes rapid and selective vascular shutdown of the blood
vessels of tumors, consistent with an anti-vascular mechanism of action as a vascular
disrupting agent (VDA), and CA-4P has shown promising results in phase II and III
clinical trials on advanced and solid tumors [7].
4

The activity and structural simplicity of CA-4 have stimulated significant interest in the
development of a wide variety of analogues that mimic the pharmacological properties
of CA-4. These analogues include compounds modified in ring A, ring B and the
ethylene bridge [8]. Previous structure-activity relationship (SAR) studies have
demonstrated that both the 3’,4’,5’-trimethoxy substitution pattern on the A-ring and the
cis-orientation of the vinyl bond were essential for optimal antiproliferative activity. B-
ring structural modifications were tolerated by the target, with the 3’-hydroxy group not
necessary for potent activity while the 4’-methoxy group was crucial for cytoxicity [9].
However, several studies have established that the cis-olefinic double bond of CA-4
tends to isomerize to the thermodynamically more stable trans-form during storage and
in the course of metabolism in liver microsomes, resulting in a dramatic decrease in
both antitubulin and antiproliferative activities.
Thus, with the aim of retaining the cis-olefin configuration of CA-4 required for
bioactivity, a strategy to overcome the problem of cis to trans isomerization developed
in which the ethylene bridge was incorporated into 1,2-diaryl-substituted five-membered
aromatic heterocyclic rings, which can be considered as non-isomerizable and
chemically stable Z-restricted isosteres of CA-4 [10].
In the designed analogues, several research groups have developed a number of
compounds containing pyrazole as the surrogate of the stilbene core of CA-4 and that
maintain the structural elements of CA-4, such as the presence of 3’,4’,5’-
trimethoxyphenyl A-ring, which was considered to be fundamental for the tubulin
binding activity, whereas variations were made in the vicinal substituent, corresponding
to B-ring of CA-4.
5

H CO
3
A
R1
R2
H CO
3
B
OCH3
OCH3
1
1
1
a, R =H, R =OH, Combretastatin A-4, CA-4
1
2
b, R =H, R =OP(O)(ONa) CA-4P
1
2
2,
c, R =R =OH, Combretastatin A-1, CA-1
1
2
1
2
3
N NH
N NH
N NH
5
H3CO
H3CO
H CO
3
4
H CO
3
OCH3
OCH3
2c
NH2
OCH3
H CO
OCH3
OCH3
3
2a
2
b
N NH
OH N NH
OH N NH
H CO
3
HO
NH2
H CO
3
H CO
3
H CO
3
H CO
3
OCH3
H CO
3
OCH3
H CO
3
OCH3
OCH3
2f
OCH3
OCH3
2e
2
d
CH3
NH
N NH
N
N N
H CO
HO
3
H CO
3
H CO
H CO
3
3
H CO
3
OCH3
OCH3
H3CO
OH
OCH3
OCH3
2
i
OCH3
2h
2g
OC H
2 5
Figure 1. Structures of CA-4 (1a), CA-4P (1b) and CA-1 (1c) and molecular formulas
of several synthesized 3,4-disubstituted 1H-pyrazoles (2a-i) previously reported in the
literature.
As far as pyrazole derivatives are concerned, Ohsumi et al. reported the preparation of
3
-(3’,4’,5’-trimethoxyphenyl)-4-(3’-amino-4’-methoxyphenyl)-1H-pyraziole 2a, which
showed potent antimitotic activity in the tubulin polymerization assay, with an IC50
value of 3 µM. Compound 2a was also a potent antiproliferative agent (IC : 8.4 nM)
5
0
6

against the colon-26 adenocarcinoma cell line [11]. In a previous study, Medarde and
co-workers also described two 3,4-disubstituted pyrazole derivatives (2b and 2c),
obtained by reversing the positions of the 2’-naphthyl and 3’,4’,5’-trimethoxyphenyl
ring moieties. Both these compounds showed potent antiproliferative activity against
HCT-116 (human colon carcinoma) and A549 (human lung adenocarcinoma) cancer
cell lines, with IC values ranging from 6 to 37 nM. Both these compounds had 10-fold
5
0
reduced activity relative to that of CA-4 as inhibitors of tubulin polymerization [12].
Pirali and co-workers also reported a concise synthesis of 3,4-diarylpyrazole derivatives
as rigidified analogues of CA-1 (1c), and among these compounds 2d and 2e were the
most potent of the series with low nanomolar potencies against the SH-SY5Y
neuroblastoma cell line [13]. Several research groups have also reported that the
introduction of the small hydrophilic amino group (2f) was tolerated at the 5-position of
the 3,4-disubstituted pyrazole system [14, 15]. Among the 3,4-diaryl-1-substituted
pyrazoles recently synthesized, the 3’,4’,5’-trimethoxyphenyl)-4-(4’-ethoxyphenyl)-1-
methylpyrazole analogue 2g emerged as a new potent, chemically stable cis-restricted
derivative of CA-4 [16]. Compound 2g showed antiproliferative activity of 17 and 31
nM against the human ovarian SKOV and human breast MDA-MD231 cancer cell lines,
respectively. This compound was comparable to CA-4 in the tubulin polymerization
assay when the two derivatives were tested at the same concentration (10 µM). Harrity
et al. have reported two isomer pyrazole-based analogues of CA-4 with the same
substitution pattern, corresponding to compounds 2h and 2i, with derivative 2h
characterized by a high antiproliferative activity (IC : 5 nM) against human umbilical
5
0
vein endothelial cells (HUVECs), while isomer 2i was inactive [17]. Derivatives 2h and
2
i were also recently synthesized by Zhang et al , who found similar activities for the
7

two compounds against three human carcinoma cell lines (gastric adenocarcinoma SGC-
901 cells, mouth epidermal carcinoma KB cells and brosarcoma HT-1080 cells), with
IC values ranging from 26 to 60 nM. [18]
7
5
0
In our ongoing efforts to discover new potent antimitotic agents and to extend our
knowledge of the pyrazole ring as a suitable mimic for the cis-olefin bridge present in
CA-4, we report here a pharmacophore exploration and optimization effort around the
pyrazole ring. We prepared two different regioisomeric series of 3,4-diarylsubstituted
1
H-pyrazole derivatives with general structures 3 and 4 (Figure 2). In these two series of
designed analogues, obtained by interchanging the substitution pattern of rings A and B,
we fixed one of the aryl groups as the 3’,4’,5’-trimethoxyphenyl motif, identical with
the A-ring of CA-4, and the modifications were focused on variation of the substituents
at the para- position on the second phenyl ring, corresponding to the B-ring of CA-4,
with electron-withdrawing (OCF , CF ) or electron-releasing (Me, Et, n-Pr, MeO, EtO,
3
3
MeS, EtS) groups (EWG and ERG, respectively). Since the methyl, methoxy and ethoxy
groups proved to be favorable for bioactivity, we introduced an additional substituent
(F, Cl, Me or MeO) at the meta-position of the para-methyl/methoxy/ethoxyphenyl ring.
Medarde et al have previously reported that the bulky and lipophilic 2-naphthyl moiety
can replace the 3’-hydroxy-4’-methoxyphenyl ring B of combretastatin A-4 without
significant loss of potency [19]. For this reason, in compound 2c we evaluated the
bioisosteric replacement of the 2’-naphthyl with the 2’-benzo[b]thienyl moiety, to
obtain derivative 3a.
The first large series of compounds with general structure 3 was characterized by the
presence of a 3’,4’,5’-trimethoxyphenyl residue at the 3-position of the pyrazole ring
combined with a benzo[b]thien-2-yl or a phenyl ring with different ERGs or EWGs at
the vicinal 4-position of the pyrazole core.
8

In order to understand whether the substituents at the 3- and 4-positions on the 1H-
pyrazole nucleus can be interchanged without affecting biological activity, starting from
the 4’-tolyl (3b), 4’-methoxyphenyl (3c) and 4’-ethoxyphenyl (3p) derivatives,
switching the aryl and 3’,4’,5’-trimethoxyphenyl moieties, the corresponding
regioisomeric derivatives 4a, 4b and 4c, respectively, were synthesized.
A noteworthy point was that the first series of 3-(3’,4’,5’-trimethoxyphenyl)-4-
substituted 1H-pyrazoles 3a-r was concisely synthesized via an efficient and versatile
convergent procedure starting from a common key intermediate represented by a 1-
tosyl-3-(3’,4’,5’-trimethoxyphenyl)-4-bromo-1H-pyrazole [20].
1
2
NH
N
NH
N
3
5
H CO
3
R
4
R
H CO
3
OCH3
H CO
3
OCH3
3
a-r
OCH3
4
a-c
a, R=4'-CH -C H
6 4
4
4
4
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
a, R=benzo[b]thien-2'-yl
b, R=4'-CH -C H
c, R=4'-CH , 3'-F-C H
3
b, R=4'-OCH -C H
3
6 4
3
6 4
c, R=4'-OCH CH -C H
2
3
6 4
3
6
4
d, R=3',4'-(CH ) -C H
3
2
6
3
e, R=4'-CH , 3'-OCH -C H
3
3
6
3
f, R=4'-CF -C H
6 4
3
g, R=4'-CH CH -C H
6 4
3
2
h, R=4'-CH (CH ) -C H
6 4
3
2 2
i, R=4'-OCH -C H
6 4
3
j, R=4'-SCH -C H
6 4
3
k, R=4'-OCH , 3'-F-C H
6 3
3
l, R=4'-OCH , 3'-Cl-C H
6 3
3
m, R=4'-OCH , 3'-CH -C H
6 3
3
3
n, R=4'-OCH , 3',5'-(CH ) -C H
3
3 2
6
2
o, R=4'-OCF -C H
6 4
3
p, R=4'-OCH CH -C H
6 4
2
3
q, R=4'-SCH CH -C H
6 4
2
3
r, R=4'-OCH CH 3'-Cl-C H
6 3
2
3,
Figure 2. Structures of 3-(3’,4’,5’-trimethoxyphenyl)-4-substituted-1H-pyrazoles 3a-r
and their isomeric 3-aryl-4-(3’,4’,5’-trimethoxyphenyl)-1H-pyrazole analogues 4a-c
9

2
. Chemistry
3
,4-Disubstituted 1H-pyrazole derivatives 3a-r and 4a-c were synthesized by the
general six-step procedure described in Scheme 1. Condensation of 3’,4’,5’-
trimethoxyacetophenone 5 and appropriately substituted acetophenones 6a-c with
dimethylformamide dimethyl acetal (DMF-DMA) provided enaminones 7 and 8a-c,
respectively, in good yields, which were subsequently reacted with hydrazine hydrate in
refluxing ethanol to provide the corresponding 3-aryl-1H-pyrazole derivatives 9 and
1
0a-c. The subsequent regioselective monobromination at the 4-position of the pyrazole
nucleus of compounds 9 and 10a-c with N-bromosuccinimide (NBS) in DMF furnished
the corresponding 3-aryl-4-bromo-1H-pyrazole analogues 11 and 12a-c, respectively.
The pyrazole hydrogen of these latter compounds was temporarily replaced with a tosyl
group by treatment with p-toluenesulfonyl chloride (TsCl) in a mixture of
dichloromethane and pyridine to afford the formation of the key N-tosyl-4-
bromopyrazole intermediates 13 and 14a-c, respectively. The brominated products were
subjected to a Suzuki cross-coupling process in the presence of the appropriate
commercially available arylboronic acid under heterogeneous conditions [PdCl (DPPF),
2
CsF] in 1,4-dioxane at 65 °C, giving rise to the corresponding 3,4-diaryl pyrazole
derivatives 15a-r and 16a-c. The tosyl protecting group was then removed by alkaline
hydrolysis using a mixture of 1 M aqueous solution of NaOH and ethanol to afford the
target 3-(3’,4’,5’-trimethoxyphenyl)-4-substituted and isomeric 3-aryl-4-(3’,4’,5’-
trimethoxyphenyl)-1H-pyrazole derivatives 3a-r and 4a-c, respectively.
1
0

O
O
H CO
H3CO
H3CO
N NH
3
a
b
N
H CO
3
H CO
3
OCH3
OCH3
H CO
3
9
5
7
OCH3
c
Ts
N N
Ts
N NH
N N
H CO
3
H CO
3
H CO
3
e
d
Br
Br
R
H CO
3
H CO
3
H CO
3
11
13
OCH3
OCH3
OCH3
f
15a-r
N NH
R
N NH
H CO
3
R
4a-c
H CO
3
OCH3
3a-r
H CO
3
OCH3
OCH3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
a, 15a, R=benzo[b]thien-2'-yl
b, 15b, R=4'-CH -C H
3
6 4
f
c, 15c, R=4'-CH , 3'-F-C H
6 4
3
Ts
d, 15d, R=3',4'-(CH ) -C H
6 3
3
2
N N
e, 15e, R=4'-CH , 3'-OCH -C H
3
3
6
3
f, 15f, R=4'-CF -C H
6 4
R
3
g, 15g, R=4'-CH CH -C H
6 4
3
2
h, 15h, R=4'-CH (CH ) -C H
6 4
3
2 2
i, 15i, R=4'-OCH -C H
6 4
3
H CO
OCH3
3
j, 15j, R=4'-SCH -C H
6 4
3
OCH3
16a-c
k, 15k, R=4'-OCH , 3'-F-C H
6 3
3
e
l, 15l, R=4'-OCH , 3'-Cl-C H
6 3
3
m, 15m, R=4'-OCH , 3'-CH -C H
6 3
Ts
3
3
N N
n, 15n, R=4'-OCH , 3',5'-(CH ) -C H
3
3 2
6
2
o, 15o, R=4'-OCF -C H
6 4
14a-c
3
R
p, 15p, R=4'-OCH CH -C H
6 4
2
3
Br
q, 15q, R=4'-SCH CH -C H
6 4
2
3
r, 15r, R=4'-OCH CH 3'-Cl-C H3
d
2
3,
6
O
O
NH
N
N
NH
a
b
c
R
N
R
R
R
6
a-c
10a-c
a, 6a, 8a, 10a, 12a, 14a, 16a, R=4'-CH -C H
6 4
b, 6b, 8b, 10b, 12b, 14b, 16b, R=4'-OCH -C H
c, 6c, 8c, 10c, 12c, 14c, 16c, R=4'-OCH CH -C H
12a-c
8
a-c
Br
4
4
4
3
3
6
4
2
3
6 4
.
Scheme 1. Reagents: a: DMF-DMA, DMF, reflux; b: NH NH H O, EtOH, reflux; c:
2
2
2
NBS, DMF, room temperature; d: TsCl, Py, CH Cl , room temperature, 72 h; e:
2
2
PdCl (DPPF), ArB(OH) , CsF, 1,4-dioxane, 65 °C; f: 1 M aqueous NaOH, EtOH, 50
2
2
°
C.
1
1

3
. Biological Results and Discussion
3
.1. In vitro antiproliferative activities.
The synthesized 3-(3’,4’,5’-trimethoxyphenyl)-4-substituted 1H-pyrazoles 3a-r and the
corresponding regioisomeric 3-(4’-methyl/methoxy/ethoxyphenyl)-4-(3’,4’,5’-
trimethoxyphenyl)-1H-pyrazole analogues 4a, 4b and 4c were evaluated for their ability
to inhibit the growth of a panel of six different human cancer cell lines and compared
with the previously published 3-(3’,4’,5’-trimethoxyphenyl)-4-(2’-naphthyl)-1H-
pyrazole derivative 2c and the reference compound CA-4 (1a) (Table 1).
CA-4 had nanomolar activity against the HeLa, MDA-MB-231, HL-60 and SEM cancer
cell lines, while HT-29 and MCF-7 cells were more resistant to CA-4, with IC values
5
0
of 3100 and 370 nM, respectively. As shown in Table 1, the antiproliferative activities
of most of the tested compounds were less pronounced against MCF-7 cells as
compared with the other cell lines. Nevertheless, compounds 3a, 3i, 3p, 3r, 4b and,
especially, 4c had excellent activity against MCF-7 cells (the IC values ranged from
5
0
0
.2 to17 nM). Excluding compounds 3e, 3h, 3n and 3o, the tested compounds showed
excellent antiproliferative activity against CA-4 resistant HT-29 cells, with IC values
5
0
ranging from double-digit to subnanomolar concentrations.
In particular, two of the synthesized compounds, 3p, bearing a 4’-ethoxyphenyl at the C-
position of the 1H-pyrazole ring, and its isomeric analogue 4c, exhibited the best
antiproliferative activity among the tested derivatives, with IC values of 0.05-4.5 and
4
5
0
0
.06-0.7 nM, respectively, in the six cell lines, as compared with a range of 1-3100 and
3
.3-28.4 nM obtained with the reference compounds CA-4 (1a) and 2c. However,
compound 4c was the only compound more active than CA-4 against all cell lines, while
the isomeric derivative 3p was less active than CA-4 against HL-60 cells only. In
1
2

addition to highly active compounds 3p and 4c, derivatives 3a and 4b were more active
than CA-4 in five of the six cell lines, again with the exception of the HL-60 cells.
Table 1. In vitro inhibitory effects of compounds 2c, 3a-r, 4a-c and CA-4 (1a)
Compound
IC (nM)
5
0
HeLa
1.7±0.3
HT-29
7.4±1.9
MCF-7
1.4±0.4
MDA-MB-231
2.5±0.5
HL-60
2.3±0.3
SEM
0.5±0.05
3
a
3
b
11.1±1.7
26.2±3.7
35.9±4.5
286.7±53.2
227.4±45.1
33.5±4.0
361.3±35.1
6.0±1.2
28.2±9.1
67.4±14.1
68.4±9.5
316.1±42.5
67.4±14.1
41.3±7.1
540.9±55.2
10.5±3.3
51.3±8.7
15.8±2.6
0.2±0.05
37.3±5.2
>10,000
2416±382
0.3±0.1
127.0±34.2
1695±450
769±181
>10000
10.6±1.6
481±26.3
2606±124
5283±365
523±45.6
3271±326
1553±101
8.6±2.1
11.9±1.1
31.0±4.5
51.3±6.2
160.2±22.3
179.0±18.4
23.3±2.5
553±49.1
4.4±0.7
2.6±0.2
23.5±2.0
28.5±3.8
305.3±24.8
187.2±26.1
18.7±1.5
406.1±56.2
1.0±0.1
3
c
3
d
3
e
3
f
4375±672
>10,000
2790±535
16.6±4.5
>10,000
805.3±230
1389±385
>10000
3
g
3
h
3
i
3
j
28.7±3.7
7.3±1.1
3650±254
327±45
40.9±7.6
5.1±0.4
30.1±3.2
2.7±0.25
2.4±0.2
3
k
3
l
1.4±0.2
21.5±1.6
2680±125
>10,000
5740±435
4.5±0.9
17.0±1.9
17.7±1.8
>10,000
1748±270
3.2±0.4
3m
32.9±4.0
>10,000
3376±647
0.05±0.0
12.5±1.5
1.6±0.25
25.2±3.8
1.9±0.3
17.2±1.4
3340±275
2470±181
0.2±0.05
13.9±1.8
1.4±0.2
3
n
>10,000
>10000
3
o
p
q
3
1.5±0.6
3
5.4±1.3
492±110
4.5±0.08
700±165
2.6±0.8
31.4±5.6
6.9±1.1
29.7±3.8
1.5±0.2
3
r
4.2±0.2
4
a
44.4±6.3
0.8±0.2
262±15.8
0.9±0.2
31.4±4.2
2.1±0.25
0.25±0.03
9.4±0.9
14.7±1.2
0.4±0.05
0.06±0.0
3.3±0.7
4
b
4
c
c
0.07±0.0
6.6±1.4
0.06±0.0
8.1±2.3
0.21±0.07
21.4±6.5
370±100
0.7±0.1
2
28.4±1.2
4.5±0.2
CA-4 (1a)
4.0±1.1
3100±100
1.0±0.2
5.1±0.1
a
IC = compound concentration required to inhibit tumor cell proliferation by 50%. Values are the mean ±
50
SE from the dose-response curves of at least three independent experiments carried out in triplicate.
1
3

In comparing the IC50 values, we found that the position of the 3’,4’,5’-
trimethoxyphenyl moiety at either the 3- or 4-position of the 1H-pyrazole ring had little
effect on antiproliferative activity, since the potency of 3-(3’,4’,5’-trimethoxyphenyl)-
1
H-pyrazole derivatives 3b, 3i and 3p was comparable and not dramatically reduced as
compared with their regioisomers 4a, 4b and 4c, respectively. This is not observed for
each couple of regioisomeric heterocyclic mimic cis-stilbene the double bond of CA-4,
such as oxazole [21], thiadiazole [22] and imidazole [23]. In examining the effect of
switching the position of the two aromatic rings at the 3- and 4-positions on the 1H-
pyrazole system (3b vs. 4a, 3i vs. 4b and 3p vs. 4c), with the exception of the two
regioisomeric 4’-tolyl derivatives 3b and 4a (the latter was from 1.5 to 25-fold less
active than the former), the 3-(4’-alkoxyphenyl)-1H-pyrazole derivatives 4b (methoxy)
and 4c (ethoxy) were slightly more active (2-10-fold) than their corresponding 4-(4’-
alkoxyphenyl)-1H-pyrazole counterparts 3i and 3p, respectively, against each cancer cell
line. Specifically, for the 4’-ethoxyphenyl isomeric derivatives 3p and 4c, this latter
compound was from 4- to 6-fold more active than 3p in five of the six cancer cell lines,
the exception being the HeLa cells, in which the two compounds were equipotent.
Comparing 3i and 4b, which shared a common 4’-methoxyphenyl moiety at the 3- and
4
-position of the 1H-pyrazole ring, respectively, 4b was 2- to 13-fold more active than
its regioisomeric counterpart 3i.
Replacement of the 2’-naphthyl by the bioisosteric 2’-benzo[b]thienyl ring (compounds
2
c and 3a, respectively) resulted in a 4- to 15-fold increase in antiproliferative activity
against five of the six cell lines, while the difference between these two derivatives was
minimal in HT-29 cells. Except in HL-60 cells, 3a had greater antiproliferative activity
1
4

than CA-4, indicating that 2’-benzo[b]thienyl moiety, like the 2-naphthyl ring, was a
good surrogate for the CA-4 B-ring
The para-tolyl derivative 3b showed interesting antiproliferative activities, with IC50
values ranging from 2.6 to 127 nM, with the SEM and MCF-7 cells as the most
sensitive and resistant, respectively, to this compound.
Replacement of methyl with the more electron-withdrawing trifluoromethyl moiety
(compounds 3b and 3f, respectively) resulted in a 2.4-72-fold reduction in
antiproliferative activity, which was least pronounced (2-fold) with the HT-29 cells.
Encouraged by the activity obtained with compound 3b, we then synthesized
compounds 3c-e to determine whether various electron-withdrawing or electron-
releasing substituents on the phenyl ring would further enhance activity. However,
without exception 3c-e were less active than 3b in all cell lines.
In an effort to further understand the steric effect of the alkyl substituent at the para-
position, the ethyl and n-propyl derivatives (3g and 3h, respectively) were prepared,
resulting in over a 1.5-7-fold loss of activity in four of the six cell lines for the p-
ethylphenyl homologue 3g relative to the p-tolyl derivative 3b, with a sharp drop of
potency against MCF-7 and MDA-MB-231 cells (IC >10 and 3.3 µM, respectively).
5
0
The reduction of activity was more pronounced (20-156-fold) against all cell lines by
replacing the 4’-methyl group of 3b with the n-propyl group (3h), indicating that
lengthening the 4’-alkyl chain was not tolerated and caused a significant reduction in
antiproliferative activity in all cell lines.
Replacement of the methyl group both of 3b and its regioisomeric derivative 4a with a
more electron-releasing methoxy moiety (to furnish 3i and 4b, respectively) resulted in
enhanced antiproliferative activity for 3i, which was much more pronounced (13-291-
1
5

fold) for 4b against all six cell lines relative to 3b and 4a, respectively. For the two
derivatives 4a and 4b, the greatest differences in activity were 269- and 291-fold in the
MCF-7 and MDA-MB-231 cells, respectively.
Replacement of the para-methoxy group of 3i with a weak electron-releasing thiomethyl
group, to furnish derivative 3j, led to a dramatic drop in potency against MCF-7 and
MDA-MB-231 cells (IC >10 and 3.6 µM, respectively), while a 5-30-fold reduction in
5
0
activity was observed against the other four cancer cell lines. A substantial reduction of
activity was also observed when the methoxy group was replaced with the strong
electron-withdrawing and bulkier trifluoromethoxy moiety (3o).
Relative to the activity of 3i, the insertion of an additional electron-withdrawing (F or
Cl) or electron-releasing methyl group at the 3’-position of the 4’-methoxyphenyl ring
had varying effects on antiproliferative activity. Introduction of a fluorine atom, to
furnish the p-OMe, m-F derivative 3k, had little overall effect on antiproliferative
activity against HeLa, HL-60 and HT-29 cells, while a 3-, 38- and 50-fold reduction of
potency with respect to compound 3i was observed against SEM, MDA-MB-231 and
MCF-7 cells, respectively. The addition of a meta-chlorine atom in 3i, to yield 3l, had
contrasting effects, with a 2-84-fold reduction in activity in four cell lines (most
pronounced in MCF-7 cells), while 3l was 4- and 50-fold more active than 3i in the
HeLa and HT-29 cells, respectively. The cell growth inhibitory activity against four of
the six cancer cells was 4-17-fold reduced (and strongly reduced against MCF-7 and
MDA-MB-231 cells, with IC >10 and 2.7 µM, respectively) by the introduction of
5
0
methyl group at the meta-position of 3i (to yield 3m). Adding a second methyl group, to
furnish the p-OMe-m,m’-diMe phenyl derivative 3o, caused a dramatic decrease in
potency relative to 3m against all cell lines.
1
6

Comparing the para-alkoxyphenyl derivatives 3i and 3p with the corresponding
isomeric cognates 4b and 4c against each cancer cell line, replacing the methoxy (3i and
4
b) with an ethoxy (3p and 4c) moiety resulted in enhanced antiproliferative activities
in all cell lines, confirming that the p-ethoxyphenyl ring is a good surrogate for the B-
ring of CA-4. The enhanced effect on activity resulting from replacement of the
methoxy group with an ethoxy moiety in colchicine site was previously observed by us
and by others [24].
The para-ethoxyphenyl derivative 3p was 2- to 120-fold more potent than its methoxy
counterpart 3i in five of the six cancer cell lines, while the two compounds had similar
activity against HL-60 cells. For the two isomeric derivatives 4b and 4c, we found that
replacement of methoxy with ethoxy, to furnish derivative 4c, improved
antiproliferative activity 7-27-fold relative to 4b against five of the six cell lines, with
the two compounds being equipotent against MDA-MB-231 cells. Replacement of
ethoxy with thioethyl (3q) caused a decrease in activity in all cell lines, which was
moderate (7-18-fold) against MDA-MB-231, HL-60 and HT-29 cells, and much more
pronounced (69-328-fold) against HeLa, MCF-7 and SEM cells.
While the 4’-ethoxy group was favorable for potency, the introduction of an additional
meta-EWG chlorine group in compound 3p, resulting in compound 3r, produced a 1.5-
3
2-fold reduction in antiproliferative activity against five of the six cell lines, the
exception being the HL-60 cells, in which 3r was 2-fold more potent than 3p.
Compound 3r, as with the corresponding 3-Cl, 4-OMe congener 3l, was more active
than CA-4 in four of the six cell lines, the exceptions being HL-60 and MDA-MB-231
cells.
1
7

3
.2. Effects of compounds 3p and 4c in nontumor cells.
To obtain a preliminary indication of the cytotoxic potential of these derivatives in
normal human cells, the most active compounds, 3p and 4c, were evaluated in vitro
against peripheral blood lymphocytes (PBL) from healthy donors. Both compounds
showed a GI greater than 10 µM, both in quiescent lymphocytes and in lymphocytes in
5
0
an active phase of proliferation induced by phytohematoagglutinin (PHA), a mitogenic
stimulus (Table 2).
Moreover, since one of the major adverse effects of antimitotics is neurotoxicity, we
wanted to evaluate the cytotoxicity of these compounds in normal human astrocytes
(
NHA). Both compounds again, as shown in Table 2, showed a GI >10 µM, indicating
50
that in this cell line, too, they are not toxic. Altogether, these results suggest that these
compounds have very low toxicity in normal cells in comparison to tumor cells,
suggesting potential for an excellent therapeutic index.
Table 2. Cytotoxicity of compounds 3p and 4c for human peripheral blood lymphocytes
(PBL) and normal human astrocytes (NHA)
a
Compound
GI (µM)
50
3
p
4c
b
PBLresting
8.7±0.7
>10
>10
>10
>10
c
PBLPHA
NHA
>10
a
Compound concentration required to inhibit cell growth by 50%.
b
PBL not stimulated with PHA.
c
PBL stimulated with PHA.
Values are the mean ± SEM for three separate experiments.
1
8

3
.3. Effect of compounds 3p and 4c in multidrug-resistant cells.
To evaluate if compounds 3p and 4c are substrates of drug efflux pumps, they were
Vbl-100
tested against the CEM
cell line that is a multidrug-resistant line selected against
vinblastine and that overexpress P-glycoprotein (P-gp) [25].
This membrane protein acts as a drug efflux pump and exhibits resistance to a wide
variety of structurally unrelated anticancer drugs and other compounds. As shown in
Vbl100
Table 3, both compounds showed cytotoxic activity in the CEM
cells comparable
to that of the parental line, indicating that these derivatives are not a substrate for P-gp.
Table 3. Cytotoxicity of 3p and 4c in multidrug-resistant cells
a
Compound
GI (nM)
50
wt
Vbl100
CEM
CEM
1.9±0.2
0.52±0.1
3
p
0.40±0.1
4
c
0.35±0.05
a
Compound concentration required to reduce cell growth by 50%.
Values are the mean ± SEM for three separate experiments.
3
.4. In vitro inhibition of tubulin polymerization and colchicine binding.
To investigate whether the antiproliferative activities of the most potent compounds of
the series derived from an interaction with tubulin, derivatives 3a-d, 3g, 3i-m, 3p-r and
4
a-c were evaluated for their inhibition of tubulin polymerization and for effects on the
3
binding of [ H]colchicine to tubulin (Table 4). For comparison, CA-4 and 2c were
examined in contemporaneous experiments.
All tested compounds strongly inhibited tubulin assembly, with compound 3a found to
be the most active (IC , 0.30 µM), and it was almost twice as potent as CA-4 (IC ,
5
0
50
0
.54 µM). Several compounds (3b, 3i, 3p, 3r, 4b and 4c) were 1.5-fold more potent
than CA-4, while all remaining compounds showed activity comparable to that of CA-4.
1
9

Table 4. Inhibition of tubulin polymerization and colchicine binding by compounds 2c,
3a-d, 3g, 3i-m, 3p-r, 4a-c and CA-4 (1a)
a
b
Tubulin assembly
Colchicine binding
% ±S.D
IC ±S.D (µM)
5
0
Compound
5
µM drug
0.5 µM drug
3
°
0.30±0.01
0.37±0.04
0.48±0.06
0.61±0.05
0.52±0.03
0.37±0.06
0.57±0.02
0.42±0.05
0.52±0.09
0.44±0.09
0.36±0.01
0.48±0.02
0.39±0.05
0.57±0.01
0.34±0.05
0.35±0.01
0.41±0.04
0.54±0.06
93±2
88±0.8
91±2
69±2
n.d
3
b
3
c
49±2
n.d.
3
d
85±5
3
g
85±1
n.d.
3
i
86±0.6
85±4
n.d.
3
j
n.d.
3
k
93±0.8
85±3
55±2
n.d.
3
l
3
m
81±0.1
95±0.9
85±1
n.d.
3
p
q
69±5
n.d.
3
3
4
r
90±0.3
86±5
62±2
n.d.
°
4
b
93±0.8
96±0.7
91±4
79±2
90±1
68±0.2
82±2
4
2
c
c
CA-4 (1a)
97±0.8
a
Inhibition of tubulin polymerisation. Tubulin was at 10 µM.
b
3
Inhibition of [ H] colchicine binding. Tubulin and colchicine were at 0.5 and 5 µM concentrations,
respectively.
n.d.=not determined
When comparing inhibition of tubulin polymerization with the growth inhibitory effects,
we found a good correlation for most, but not all, of the active compounds. While 3b
was generally less potent than 3a as an antiproliferative agent, the two compounds were
similar as inhibitors of tubulin assembly. Although several compounds, such as 3c, 3d,
2
0

3
g, 3j, 3m, 3r and 4a showed lower antiproliferative activity on HeLa, HL-60 and SEM
cancer cells when compared with CA-4, they were comparable to CA-4 as inhibitors of
tubulin assembly.
In the colchicine binding studies, all compounds when tested at the higher concentration
(5 µM) had quantitatively similar effects, varying within a narrow range (81-96%
inhibition), and they showed potency comparable to that CA-4, which in these
experiments inhibited colchicine binding by 97%. For the most active compounds (90%
or higher inhibition when present in the reaction mixture at the same concentration as
colchicine), even when tested at a ten-fold reduced concentration (0.5 µM), compounds
4
b and 4c were as potent as CA-4, which in these latter experiments inhibited colchicine
binding by 82%, while derivatives 2c, 3a, 3c, 3k, 3p and 3r were less potent (49-69%
range inhibition).
The potent inhibition observed with these compounds indicates that they bind to tubulin
at a site overlapping the colchicine site. It is thus significant that one compound (4c) in
the present series had activities superior to that of CA-4 as inhibitors of tubulin
assembly and, less frequently observed, comparable to that CA-4 as inhibitors of
colchicine binding to tubulin at the two concentrations (0.5 and 5 µM) tested.
3
.5. Molecular modeling studies.
Molecular docking studies using Glide SP [26] were conducted on selected compounds
to elucidate their potential interaction with the colchicine site of tubulin. The crystal
structure of tubulin co-crystallized with the microtubule-destabilizing CA-4 (PDB ID:
5
LYJ) was used [27]. The docked molecules occupied the active site similarly to the co-
crystallized CA-4, with the trimethoxyphenyl ring deeply buried into the hydrophobic
2
1

pocket formed by different residues of the β-tubulin chain, including βVal238,
βCys241, βLeu242, βLeu248, βAla250, βLeu255, βAla316, βIle318 and βIle378
(Figure 3). Interestingly, all the new derivatives formed an H-bond between their 4-
methoxy group and βCys241, an interaction believed to be fundamental for colchicine
site agents. The central pyrazole ring overlapped with the cis-double bond of CA-4,
confirming its function as a suitable mimic for the cis-olefin bridge. The different
substituted phenyl rings and the 2’-benzo[b]thienyl ring of 3a were placed in a
hydrophobic area formed by the α and β-tubulin interface. Hydrophobic interactions
with the surrounding amino acids, including βAsn258, βMet259, βLys352, αThr179,
αAla180 and αVal181, further stabilized the molecules in the active site. However,
from the docking results, it appeared that this hydrophobic area can accommodate the
different substituted phenyl ring, including the n-propyl (3h) and the tri-substituted (3n)
derivatives, compounds found not active in the antiproliferative assay. Also, derivatives
3
e, 3f and 3o presented a binding mode in line with the other docked molecules,
indicating that the loss in antiproliferative activity observed could be associated with
some other factors rather than their inability to bind the colchicine site. It should also be
noted that the regioisomeric heterocyclic compounds (3p vs. 4c) bound in the active site
in a similar manner, in line with their ability to inhibit tubulin polymerization.
In order to further elucidate the suggested binding mode and possibly to find a potential
structural discriminant between active and inactive compounds, a series of 50 ns
molecular dynamic (MD) simulations on selected compounds (3a, 3e, 3h, 3m, 3n, 3o,
3
p, 4c) were performed using the Desmond software package [28]. The compounds
relative binding free energies (∆Gbinding) were then calculated using the Prime/MM-
GBSA based calculation method [29, 30]. All the protein-ligand systems reached
2
2

stability after an initial 25 ns of equilibration, as shown by the C-alpha RMSD variation,
and therefore only the second 25 ns of the simulation was considered in our analysis. In
general, the position of the trimethoxyphenyl ring and the interaction between the 4-
methoxy group and βCys241 were maintained by all the derivatives during the entire
MD, potentially contributing to the protein-ligand stability. The orientation of the
different substituted phenyl rings and 2’-benzo[b]thienyl group (3a) in the nearby
hydrophobic area was constant during the entire simulation, with the hydrophobic
interactions with the surrounding residues maintained. Unfortunately, the ∆Gbinding
calculated values for all the ligand-protein complexes were very similar and not
significantly different to justify a potential correlation between the biological activity
and the calculated values. In conclusion, molecular docking and MD studies have given
insight into the potential binding mode for this family of compounds, but both
molecular modeling techniques were not able to find a significant correlation between
the calculated binding energy values and the experimental antiproliferative data. Other
factors, such as physical-chemical properties of the compounds rather than their ability
to interact with tubulin could be the cause of the loss of biological activity seen in the
cell-based assays.
2
3

Figure 3. Proposed binding modes for compounds 3a (A), 3p (B), 4c (C), 3h (D), 3n
E), 3o (F) in comparison with CA-4 in the tubulin colchicine site (PDB ID: 5LYJ). Co-
(
crystallized CA-4 is shown in green, compound 3a in pink, compound 3p in turquoise,
compound 4c in gold, compound 3h in orange, compound 3n in salmon and compound
3
o in purple. The carbon atoms of residues from α-tubulin chain are shown in dark
purple, whereas carbon atoms of residues from β-tubulin are colored in green. The
hydrophobic area at the dimer interface is represented as a light purple surface.
3
.6. Compounds 3p and 4c induced cell cycle arrest in G2/M along with alteration of
cell cycle checkpoint proteins.
The effects of two of the most active compounds (3p and 4c) on cell cycle progression
was examined by flow cytometry in MDA-MB-231 cells (Figure 4). After a 24 h
2
4

treatment, both compounds induced a G2/M arrest in these cells, but 4c (Figure 4, Panel
B) was more potent than 3p since a maximum effect (60% G2/M cells) occurred with 10
nM 4c. A comparable effect with 3p required 100 nM compound. With both
compounds, the increase in G2/M cells was accompanied by an equivalent reduction of
cells in the S and G1 phases.
A
B
.
Figure 4. Cell cycle analysis of MDA-MB-231 cells treated with 3p (Panel A) or 4c
(Panel B) for 24 h, at the indicated concentrations. Cells were fixed and labeled with PI
and analyzed by flow cytometry as described in the Experimental Section. Data are
represented as mean of two independent experiments ± SEM.
To further investigate the effects of the two compounds on the cell cycle, we evaluated
the expression of proteins involved in regulation of the cell cycle and in spindle
2
5

assembly in MDA-MB-231 cells by western blot analysis. As shown in Figure 5, both
compounds induced a substantial increase in the expression of p21, a well known cyclin
dependent kinase inhibitor. Although it is known that a p21 increase contributes to
arrest cells in G1, a p21 increase can also inhibit cdc2-cyclin B complex formation and
thus contribute to mitotic arrest [31].
Moreover, cdc25c expression was strongly reduced, in particular with 4c (Figure 5
Panel B) both after 24 and 48 h treatments even at the lower concentration (50 nM)
used. Similarly, there was also a decrease in the dephosphorylation of cdc2. Note that
dephosphorylation of this protein is needed to activate the cdc2/cyclin B complex, and
this effect is stimulated by cdc25c [32, 33]. On the other hand, neither compound caused
a significant change in cyclin B expression after either a 24 or 48 h treatment. This
suggests that the induced G2/M arrest is not due to defects in G2/M regulatory proteins
but, rather, is closely linked with acceleration of entry into mitosis. These results are in
agreement with findings with other antimitotic derivatives synthesized by our group
[
34-36] and suggest that cdc2/cyclin B1 complexes failed to be activated, preventing
cells from exiting mitosis, which would eventually lead to apoptotic cell death.
2
6

A
B
Figure 5. Effect of compounds 3p (Panel A) and 4c (Panel B) on cell cycle checkpoint
proteins. MDA-MB-231 cells were treated for 24 or 48 h with the indicated
concentrations of compounds. The cells were harvested and lysed for detection of the
expression of the indicated protein by western blot analysis. To confirm equal protein
loading, each membrane was stripped and reprobed with anti-GAPDH antibody.
3
.7. Compounds 3p and 4c induced apoptosis.
To evaluate the mode of cell death induced by 3p and 4c, we performed a bi-parametric
cytofluorimetric analysis using propidium iodide (PI) and annexin-V-FITC, which stain
DNA and phosphatidylserine (PS) residues, respectively. We analysed the effects on
apoptosis induction by both these molecules in MDA-MB-231 cells after treatments for
2
4 or 48 h.
2
7

Both compounds induced apoptosis in a time and concentration dependent manner
Figure 6). The apoptotic effects were evident with both compounds at the lowest
(
concentration examined (50 nM).
A
B
Figure 6. Flow cytometric analysis of apoptotic cells after treatment of MDA-MB-231
cells with compounds 4c or 3p at the indicated concentrations after incubation for 24
(Panel A) or 48 h (Panel B). The cells were harvested and labeled with annexin-V-FITC
and PI and analyzed by flow cytometry. Dual staining for annexin-V and with PI permits
-
-
+
-
discrimination between live cells (annexin-V /PI ), early apoptotic cells (annexin-V /PI
)
+
+
-
+
, late apoptotic cells (annexin-V /PI ) and necrotic cells (annexin-V /PI ). Data are
represented as mean ± SEM of three independent experiments.
2
8