Chalcone derivatives targeting mitosis: synthesis, evaluation of antitumor activity and lipophilicity

Article abstract of DOI:10.1016/j.ejmech.2019.111752

This study describes the synthesis of a series of chalcones, including pyrazole and α,β-epoxide derivatives, and evaluation of their cell growth inhibitory activity in three human tumor cell lines, as well as their lipophilicity using liposomes as a biomi

Full text of DOI:10.1016/j.ejmech.2019.111752

Journal Pre-proof
Chalcone derivatives targeting mitosis: Synthesis, evaluation of antitumor activity and
lipophilicity
Patricia Pinto, Carmen Mariana Machado, Joana Moreira, José Diogo P. Almeida,
Patrícia M.A. Silva, Ana C. Henriques, José X. Soares, Jorge A.R. Salvador, Carlos
Afonso, Madalena Pinto, Hassan Bousbaa, Honorina Cidade
PII:
S0223-5234(19)30904-3
DOI:
https://doi.org/10.1016/j.ejmech.2019.111752
EJMECH 111752
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 12 August 2019
Revised Date: 22 September 2019
Accepted Date: 1 October 2019
Please cite this article as: P. Pinto, C.M. Machado, J. Moreira, José.Diogo.P. Almeida, Patrí.M.A.
Silva, A.C. Henriques, José.X. Soares, J.A.R. Salvador, C. Afonso, M. Pinto, H. Bousbaa, H. Cidade,
Chalcone derivatives targeting mitosis: Synthesis, evaluation of antitumor activity and lipophilicity,
European Journal of Medicinal Chemistry (2019), doi: https://doi.org/10.1016/j.ejmech.2019.111752.
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.

1
2
Chalcone derivatives targeting mitosis: synthesis, evaluation of antitumor activity
and lipophilicity
3
4
5
6
Patricia Pintoa,#, Carmen Mariana Machadob,#, Joana Moreirab,c,#, José Diogo P.
Almeidad, Patrícia M. A. Silvad, Ana C. Henriquesd, José X. Soarese, Jorge A. R.
Salvadora,f, Carlos Afonsob,c, Madalena Pintob,c, Hassan Bousbaad,*, Honorina
Cidadeb,c,*
7
8
9
aLaboratório de Química Farmacêutica, Faculdade de Farmácia, Universidade de
Coimbra, Pólo III - Polo das Ciências da Saúde, Azinhaga de Santa Comba, 3000-354
Coimbra, Portugal
10
11
12
bLaboratório de Química Orgânica e Farmacêutica, Departamento de Ciências
Químicas, Faculdade de Farmácia, Universidade do Porto, Rua de Jorge Viterbo
Ferreira 228, 4050-313 Porto, Portugal
13
14
15
cCentro Interdisciplinar de Investigação Marinha e Ambiental (CIIMAR), Universidade
do Porto, Terminal de Cruzeiros do Porto de Leixões, Av. General Norton de Matos s/n
4450-208 Matosinhos, Portugal
16
17
dCESPU, Instituto de Investigação e Formação Avançada em Ciências e Tecnologias da
Saúde (IINFACTS), Rua Central de Gandra 1317, 4585-116 Gandra PRD, Portugal.
18
19
20
eLAQV-REQUIMTE, Laboratório de Química Aplicada, Departamento de Ciências
Químicas, Faculdade de Farmácia, Universidade do Porto, Rua de Jorge Viterbo
Ferreira 228, 4050-313 Porto, Portugal
21
22
23
fCentro de Neurociências e Biologia Celular, Coimbra, Portugal.
Abstract
24
25
26
This study describes the synthesis of a series of chalcones, including pyrazole and α,β-
epoxide derivatives, and evaluation of their cell growth inhibitory activity in three
human tumor cell lines, as well as their lipophilicity using liposomes as a biomimetic

27
28
29
30
31
32
33
34
35
36
37
38
membrane model. Structure-activity and structure-lipophilicity relationships were
established for the synthetized chalcones. From this work, nine chalcones (3, 5, 9, 11,
15-19) showing suitable drug-like lipophilicity with potent growth inhibitory activity
were identified, being the growth inhibitory effect of compounds 15-17 associated with
a pronounced antimitotic effect. Compounds 15-17 affected spindle assembly and, as a
consequence, arrested cells at metaphase in NCI-H460 cells, culminating in cell death.
Amongst the compounds tested, compound 15 exhibited the highest antimitotic activity
as revealed by mitotic index calculation. Moreover, 15 was able to enhance
chemosensitivity of tumor cells to low doses of paclitaxel in NCI-H460 cells. The
results indicate that 15 exerts its antiproliferative activity by affecting microtubules and
causing cell death subsequently to a mitotic arrest, and thus has the potential for
antitumor activity.
39
40
41
Keywords: chalcones; antitumor activity; mitosis; lipophilicity; paclitaxel
chemosensitivity
42
43
44
*Corresponding authors email addresses: hassan.bousbaa@iucs.cespu.pt (H.
Bousbaa) and hcidade@ff.up.pt (H. Cidade)
45
46
#Authors contributed equally to this work

47
1. Introduction
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
Microtubules are long hollow cylindrical protein polymers made of α- and β-tubulin
heterodimers, and, together with actin filaments and intermediate filaments, constitute
the major cytoskeletal components in all eukaryotic cells [1]. They have structural
roles, such as cell shape and polarity, and dynamic roles such as cell motility,
intracellular transport of vesicles and organelles, and chromosome movements in
mitosis [1]. Microtubules form the mitotic spindle in charge of delivering chromosomes
to the two daughter cells and, hence, the success of mitosis depends on proper
microtubule function. Microtubule targeting agents (MTAs), also known as antimitotic
drugs, disrupt microtubule function and activate the spindle assembly checkpoint (SAC)
in response to defective spindle assembly or spindle chromosome attachment. As a
consequence, cells are chronically arrested in mitosis and eventually die by apoptosis,
making microtubules interesting targets for the development of anticancer drugs [2].
MTAs include taxanes and vinca alkaloids which continue to be widely used in the
clinic [3]. Taxanes, such as paclitaxel and docetaxel, are microtubule stabilizing agents
by promoting tubulin polymerization. Vinca alkaloids, such as vinblastine and
vincristine, are microtubule destabilizers by inhibiting microtubule polymerization.
However, these MTAs have limited effectiveness in many malignancies due to
hematopoietic and neurologic toxicities owing to the role of microtubules in non-tumor
cells [3]. Also, their efficacy can be reduced due to drug resistance as a consequence of
upregulation of transmembrane efflux pumps that lowers intracellular drug
concentrations, and tubulin mutations that affect drug binding sites or alter the
expression of different tubulin isotypes [4]. Therefore, the search for new MTAs, that
can overcome the disadvantages associated with the MTAs currently in use, continues
to be a very active area of research [5].
72
73
74
75
76
77
Chalcones represent one of the major classes of flavonoids that have long been known
for their wide range of biological activities, including antitumor activity [6-11].
Moreover, the α,β-unsaturated ketone moiety of chalcones makes these compounds
valuable chemical substrates for the synthesis of bioactive derivatives, namely
heterocyclic ring systems like pyrazoles [12] . Over the last few years, the discovery of
chalcones with antimitotic activity has drawn much attention [13]. Inspired by this, our

78
79
80
81
82
83
84
85
research group has reported several chalcones with notable growth inhibitory activity in
human tumor cell lines [14-18]. The study of the mechanisms underlying this inhibitory
effect resulted in the identification of two chalcone derivatives, 1 and PC2 (2), with
antimitotic effect (Figure 1) [19, 20]. Both chalcones caused abnormal spindle
apparatus assembly and a prolonged mitotic arrest followed by cell death [19, 20].
These results support the idea that chalcones 1 and 2 represent a good starting basis for
the design of new potential antitumor agents targeting mitosis with improved antitumor
properties.
86
87
88
Figure 1. Chemical structure of chalcones 1 and PC2 (2).
89
90
91
92
93
94
95
96
97
The action of drug relies on the correct balance between its pharmacokinetic (PK) and
pharmacodynamic (PD) behavior. The concept of drug-likeness emerged from the seek
for this PK/PD balance [21], being defined by a set of molecular descriptors and/or
physicochemical properties which are used to identify the suitable features for a
compound to became a drug [22, 23]. Inspired by the pivotal work of Lipinski and his
“rule of five”, several authors have proposed different set of criteria to evaluate drug-
likeness [24-30]. As the majority of molecular descriptors and physicochemical
properties can be calculated or predicted by in silico methods, its values can be used to
roughly evaluate the compound drug-likeness even before its synthesis.
98
99
Among the several drug-like features, lipophilicity is one of the most important
parameters [31] , namely in the case of taxanes, the most prominent class of MTA
agents. In fact, taxanes are very lipophilic agents, exhibiting poor oral bioavailability,
and requiring the use of co-solvents which are associated with hypersensitivity reactions
100
101

102
103
104
105
106
107
108
109
110
and peripheral neuropathy [32, 33]. Usually, lipophilicity is assessed by measuring the
partition coefficient (Log P) drug between water and n-octanol [21, 31, 34] . However,
the n-octanol:water system fails to encode all interactions that occur during drug
permeation across biological membranes [35]. In this context, biomimetic models have
been gaining increasing relevance as alternative to the classical log P determination
because they mimic the duality hydrophobic/hydrophilic environment present in
biological membranes [35-37]. Among the different biomimetic models, 1,2-
dimyristoyl-sn-glycero-3-phosphorylcholine (DMPC) liposomes are suitable models to
assess drug lipophilicity [38].
111
112
113
114
115
116
117
118
119
120
Hence, in continuation of our efforts to obtain new chalcone derivatives with improved
antitumor and antimitotic activity, a small library of chalcones structurally related with
1 and 2, has been envisioned and preliminary assessed for their drug-likeness properties.
These compounds were designed to evaluate the effect of the substitution pattern of A
and B phenyl rings in the antitumor and antimitotic activities. The analogues were
synthesized and evaluated for their in vitro growth inhibitory activity against three
human tumor cell lines, A375-C5 (melanoma), MCF-7 (breast adenocarcinoma) and
NCI-H460 (non-small cell lung cancer). The antimitotic effect of the chalcones showing
potent growth inhibitory effect was characterized. Furthermore, the compounds
lipophilicity was determined using liposomes as biomimetic model.
121
122
2. Results and Discussion
2.1. Drug-likeness of the chalcones library
123
124
125
126
127
128
129
130
131
Inspired by 1 and PC2 (2), a small library of chalcones was designed and in silico
studies were performed to predict drug-likeness. The drug-likeness was evaluated
accordingly to the following chemical features: i) size, inferred by molecular weight
(MW); ii) flexibility, inferred by the number of rotatable bonds (RB); iii) unsaturation,
inferred by fraction of carbon sp3 (Fsp3); iv) solubility, inferred by log S; v) polarity,
inferred by polar surface area; and vi) lipophilicity, inferred by the log P. For each
designed chalcone, the abovementioned molecular descriptors/physic-chemical
properties were calculated (Table S1 – Table S4 – Supporting materials). For each
chemical feature, the obtained mean values were plotted against the range of values

132
133
134
135
136
preconized by the drug-likeness guidelines (Figure 2a). Solubility, polarity,
lipophilicity, size, and flexibility were within the drug-like chemical space, while
unsaturation is the only feature that was not within the preconized limits. However, the
high unsaturation degree is a characteristic of the chalcone scaffold and per si cannot
invalidate the design of this library.
137
138
139
140
141
Figure 2. (a) Polar plot of the chalcones chemical space. For each feature, mean values
were plotted in polar coordinates. Green colored zone represent the range of values
preconized for the drug-like chemical space. (b) color map of the compliance with the
rules of Medicinal Chemistry.
142
143
144
Molecular descriptors and physicochemical properties of the designed chalcones were
framed accordingly to the “rules” of medicinal chemistry (Figure 2b). The designed
compounds fulfill the requirements preconized by different medicinal chemistry “rules”,

145
146
147
148
with the exceptions of the MDDR-like rule. The designed derivatives that violated one
criteria of MDDR-like have number of rings < 3, and the ones that violated two criteria
have also a number of rotatable bonds < 6. Once again, these features were mostly
imposed by the chalcone scaffold.
149
2.2. Synthesis and tumor cell growth inhibitory activity evaluation
150
151
152
153
154
155
156
Chalcones 3-23 were synthesized according to the strategy illustrated on Scheme 1a.
The synthesis of chalcones 3-23 was accomplished through base-catalyzed aldol
condensation reactions of appropriately substituted acetophenones and benzaldehydes
1
by microwave (MW) irradiation (Scheme 1a). According to the H NMR data all
chalcones showed trans configuration (JHα-Hβ =15.4-15.7 Hz). 13C NMR assignments
were determined by 2D heteronuclear single quantum correlation (HSQC) and
heteronuclear multiple bond correlation (HMBC) experiments.
157

158
159
Scheme 1. Synthesis of chalcones. a) 40% NaOH/CH3OH, MW b) H2O2/5% NaOH,
CH3COCH3/CH3OH (3:2), r.t. or H2O2/5% K2CO3/CH3OH, r.t.; c) Hydrazine hydrate,
p-toluenesulfonic acid, Xylenes and dichloromethane, 100 ºC, 3-5 h.
160
161
162
163
164
165
166
The effect of chalcones on the in vitro growth of three human tumor cell lines, A375-C5
(melanoma) MCF-7 (breast adenocarcinoma) and NCI-H460 (non-small cell lung
cancer) was evaluated according to the procedure adopted by the National Cancer
Institute (NCI, USA) which uses the protein-binding dye sulforhodamine B (SRB) to
assess cell growth as described in the materials and methods section. A dose response

167
168
curve was obtained for each cell line, and the concentration of each tested compound
that caused cell growth inhibition of 50% (GI50) was determined.
169
Table 1. Growth inhibitory effect of compounds 3-27 on human tumor cell lines.
GI50 (µM)
A375-C5
3.63 ± 0.58
11.12 ± 0.96
4.15 ± 0.85
17.77 ± 5.08
5.38 ± 1.47
7.25 ± 2.97
3.21 ± 0.45
6.96 ± 0.65
3.33 ± 1.18
11.27 ± 1.30
7.14 ± 1.87
12.14 ± 1.87
5.70 ± 1.45
6.90 ± 1.10
8.57 ± 1.06
2.89 ± 0.19
7.10 ± 0.62
3.45 ± 0.54
3.81 ± 0.765
4.51 ± 1.30
4.14 ± 0.70
38.50 ± 4.26
6.63 ± 3.37
> 37.5
MCF-7
NCI-H460
5.06 ± 0.20
13.62 ± 2.61
7.12 ± 0.20
17.76 ± 2.97
8.34 ± 2.02
8.44 ± 2.13
3.02 ± 0.01
7.48 ± 0.41
4.44 ± 0.87
15.28 ± 2.85
11.85 ± 3.46
15.50 ± 5.66
6.28 ± 0.31
6.61 ± 0.63
8.35 ± 0.31
5.60 ± 1.20
8.74 ± 1.03
10.84 ± 1.92
8.95 ± 1.01
9.61 ± 2.54
27.68 ± 1.91
61.78 ± 2.04
16.88 ± 3.48
> 37.5
5.95 ± 0.88
12.60 ± 2.68
7.70 ± 2.32
23.92 ± 7.18
11.65 ± 4.57
12.12 ± 2.33
3.26 ± 0.11
10.06 ± 3.70
4.28 ± 2.17
10.78 ± 4.44
12.17 ± 2.79
22.54 ± 1.84
5.56 ± 1.51
6.89 ± 0.41
9.75 ± 1.24
3.97 ± 0.82
8.52 ± 1.03
6.49 ± 0.30
13.15 ± 0.44
8.41 ± 3.63
15.10 ± 0.39
59.92 ± 12.70
14.01 ± 1.73
> 37.5
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
16.08 ± 2.94
16.34 ± 1.40
16.15 ± 0.56
Results are expressed as GI50 (µM) values, tested using A375-C5, MCF-7 and NCI-

H460 cell lines. Doxorubicin (positive control): A375, GI50 = 0.014 ± 0.002 µM; MCF-
7, GI50 = 0.009 ± 0.001 µM; and NCI-H460, GI50 = 0.009 ± 0.002 µM. Data represent
mean ± SD of at least three independent experiments performed in duplicate.
170
171
172
173
174
As shown in Table 1, compounds 3, 5, 9, 11, 15-19, and 22 demonstrated the best
results of GI50 (< 10 µM) in the three human tumor cell lines studied and compounds 7,
8, 10, 13, 20, 21, and 23 also showed promising antiproliferative activity, with a GI50
value lower than 10 µM for at least one tumor cell line.
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
Although the number of compounds is limited, an attempt was made to draw some
considerations concerning structure-activity relationship (SAR). Considering the
substitution pattern of B ring, some interesting results were observed. Regarding the
methoxylated chalcones, when comparing GI50 values of 3,4,5-trimethoxychalcones 3, 9
and 18 with the corresponding 2,4,5-trimethoxychalcones 4, 10 and 19, the results
revealed that the presence of three methoxy groups on positions 3,4,5 is more favorable
for their growth inhibitory activity than the presence on positions 2,4,5. Moreover, the
comparison of the GI50 values of 3,4-dimethoxychalcones 6, 12 and 21 with 3,5-
dimethoxychalcones 5, 11 and 20 revealed that the presence of two methoxy groups on
positions 3,5 is more favorable than the presence on positions 3,4. These results suggest
the importance of 3,4,5-trimethoxyphenyl and 3,5-dimetoxyphenyl B rings for this
activity. Regarding the growth inhibitory activity of the chlorinated chalcones, the
results showed that the presence of 3,5-dichlorophenyl ring (7, 13 and 22) is more
favorable than a 3,4-dichlorophenyl ring (8, 14 and 23) for the studied effect, mainly in
A375-C5 and MCF-7 cell lines.
190
191
192
193
194
195
196
It has been proposed that the presence of conjugated enones in the chalcone scaffold
may allow Michael addition of intracellular thiol compounds, such as glutathione, to the
olephinic double bond contributing to the enhancement of their antiproliferative activity
in human tumor cell lines [39] . Nevertheless, some studies have shown that molecular
modification of this enone moiety may led to derivatives with enhanced antiproliferative
effect. In fact, for some chalcones the replacement of this moiety by a pyrazole ring has
led to chalcone derivatives with improved activity [40] . Therefore, to explore the

197
198
199
200
201
202
203
204
205
206
207
208
impact of modification of the enone moiety on the antiproliferative effect of the two
most potent chalcones, 3 and 9, the pyrazole derivatives 26 and 27 were obtained by
molecular modification of 3 and 9, respectively, according to scheme 1b. Firstly,
chalcone α,β-epoxide derivatives 24 and 25 were synthesized with 46% and 41% yield,
respectively, by the reaction of chalcones 3 and 9 with hydrogen peroxide in the
presence of potassium carbonate and methanol. To improve the yields, the epoxidation
reaction was then performed using the same oxidant with a stronger basic solution (5%
NaOH) in a mixture of acetone:methanol (3:2). Using these conditions both derivatives
24 and 25 were obtained with higher yields (61% and 58%, respectively). After the
epoxidation step, pyrazole derivatives 27 and 28 were synthesized by the reaction with
hydrazine hydrate in the presence of p-toluenesulfonic acid in xylenes and
dichloromethane, in very low yields (4-8%).
209
210
211
212
213
214
The antiproliferative activity of these chalcone derivatives on A375-C5, MCF-7, and
NCI-H460 cell lines was also evaluated. The comparison of GI50 values for the epoxide
(24 and 25) and pyrazole (26 and 27) derivatives with those of corresponding chalcones
3 and 9, suggests that the substitution of the enone moiety by a three carbon bridge with
an epoxide or a pyrazole ring is associated with a decrease in the growth inhibitory
effect, particularly for compound 3 (Table 1).
215
216
2.3. Characterization of the growth inhibitory activity
2.3.1. NCI-H460 cells arrest in mitosis in response to 15-17 treatment
217
218
219
220
221
222
223
224
225
226
Chalcones 3, 5, 9, 11, 15-19, and 22 with best results of GI50 (< 10 µM) in the three
human tumor cell lines were tested for their potential antimitotic effect in NCI-H460
cells. For that, NCI-H460 cells were treated with 3xGI50 of each compound so as to
enhance the potential effect, making antimitotic activity, if any, identifiable within 15 h
incubation with the compound. As positive control, cells were treated with 1 µM of
nocodazole, a microtubule depolymerizing agent which induces a well-known arrest in
mitosis [41]. Phase contrast microscopy examination of treated cells showed an
accumulation of bright round-shaped cells, characteristic of mitotic cells, that mirrored
that of nocodazole-treated cells, suggesting an antimitotic activity for all compounds
tested (Figure 3a). DAPI staining of treated cells, to visualize the DNA, confirmed the

227
228
229
230
presence of an increasing number of cells arrested in mitosis, through observation of
condensed chromosomes (Figure 3a). In contrast, and as expected, untreated and
DMSO-treated cells showed only few cells in mitosis, with the majority of cells at
interphase.
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
We noted that compounds 15-17 exhibited the highest antimitotic activity compared
with the other selected compounds tested. This observation was confirmed when we
determined the mitotic index (MI) for all the selected compounds, after phase contrast
microscopy quantification of mitotic and interphase cells (Figure 3b). As expected,
untreated cells presented a mitotic index of 11.3%, while nocodazole-treated cells had
an increase to 73.1%. All compounds tested induced a significant increase in mitotic
index, compared to control treatments. It is noteworthy that cells treated with
compounds 15-17 exhibited the strongest antimitotic activity, with a mitotic index
between 25.80% and 49.37% (Figure 3b). Collectively, these results suggest that
compounds 3, 5, 9, 11, 15-19, and 22 exert their antiproliferative effect by blocking
mitotic progression, being 15-17 the compounds with the strongest antimitotic activity.
Interestingly, similarly to the previously identified antimitotic chalcones 1 and PC2 (2)
[19, 20], and to several MTAs, such as podophylotoxin, combretastatin A4 and
colchicine, chalcones 15-17 with the highest mitotic index also possess a 3,4,5-
trimethoxyphenyl ring, already described as important for the interaction with tubulin
[20]. These results reinforce the importance of a 3,4,5-trimethoxyphenyl ring to obtain
compounds with antimitotic effect.
248

249
250
251
252
253
254
255
Figure 3. Treatment with 3, 5, 9, 11, 15-19, and 22 arrest NCI-H460 cells in mitosis,
being this effect associated with alteration in mitotic spindle morphology. (a)
Representative microscopy fields of mitotic cells (rounded) accumulated in cell culture
after 15 h treatment with the indicated compounds, as shown by phase contrast
microscopy (upper panel, Bar = 20 µm) and confirmed by DAPI staining of DNA
(lower panel, Bar = 5 µm). Similar results were observed with the other compounds.

256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
Nocodazole was used as a positive control. (b) Mitotic Index graph showing
accumulation of mitotic cells after 15 h of treatment with all selected compounds.
Statistical significance of samples treated with the compounds when compared with
control, *P<0.05; **P<0.01; ***P<0,001; ****P<0.0001. (c) Immunofluorescence
staining with anti-α-tubulin antibody. Control metaphase cells exhibited aligned
chromosomes (blue) attached to normal microtubule fibers (green) that are assembled
into a well-organized bipolar spindle, while 15-treated cells presented multipolar
spindles with misaligned chromosomes. DNA was stained with DAPI (blue). Bar = 5
µm. Similar results were observed with compounds 16 and 17. (d) Mitotic spindles were
quantified in control- and 15-treated cells, the results are expressed as a percentage of
multipolar spindles over the total of spindles. Statistical significance of samples treated
with 15 when compared with control (*P<0.05). Data represent mean±SD of three
independent experiments. (e) NCI-H460 cell death was induced upon prolonged
exposure to 15. Extended treatment with compound (24 h) promoted micronucleation
and cell death (arrow); the inset shows a fragmented cellular nucleus reminiscent of cell
dead by apoptosis. DNA stained with DAPI. Bar = 5µm.
272
2.3.2. Mitotic spindle morphology is affected by 15-17
273
274
275
276
277
278
279
280
281
282
283
284
285
286
Chalcones 15-17 with the strongest antimitotic activity were selected for subsequent
analysis. Mitotic arrest is often linked to spindle microtubules perturbation, resulting in
persistent activation of SAC signaling activation [41] . Accordingly, and since 15-17
treated cells were arrested in mitosis, we then treated NCI-H460 cells with these
compounds as above, and performed an immunostaining using an anti-α-tubulin
antibody, to visualize the spindle microtubules. DNA was also stained with DAPI.
Through fluorescence microscopy observation, we noticed that untreated control cells
exhibited microtubule fibers assembled into a well-organized bipolar spindle, while
compounds-treated cells showed an abnormal mitotic spindle morphology, mainly
consisting of a multipolar spindle (Figure 3c). Consistent with microtubule defects,
many chromosomes failed to align at the metaphase plate. As the three compounds
exhibited similar results, we pursued further analysis with the compound 15 only,
because it showed the best antimitotic activity. The morphology of spindles was
quantified revealing an increase in multipolar spindles, from 2% of total mitotic

287
288
289
spindles in untreated-cells to 28% in 15-treated cells (Figure 3d). These results suggest
that the compounds arrest NCI-H460 cells in mitosis, by interfering with spindle
microtubule assembly.
290
2.3.3. Assessment of cell fate after prolonged treatment with 15
291
292
293
294
295
296
297
In order to characterize the fate of NCI-H460 cells in response to prolonged treatment
with 15, NCI-H460 cells were exposed to 15 treatment for 24 h and stained with DAPI,
for DNA evaluation. Through fluorescence microscopy observation, it was verified that
cells presented abnormal nuclear morphology, with micronucleation, suggesting cell
death (Figure 3e). This observed morphology is reminiscent of cells dying by
apoptosis, probably as a consequence of prolonged mitotic cell arrest or cell division
failure. Further analyses are needed to confirm if the 15-induced cell death is apoptosis.
298
2.3.4. Treatment with 15 sensitizes tumor cells to low nanomolar doses of paclitaxel
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
Given its antimitotic activity, we wondered if 15 could enhance chemosensitivity and
efficacy of tumor cells to paclitaxel, an antimitotic agent currently used in cancer
therapy. The rational is that the two agents should target different aspects of the mitotic
spindle, and consequently should mutually increase their efficacy in terms of antimitotic
activity and cytotoxicity. Therefore, we performed an MTT viability assay using a
combination of 15 compound (at 1x and 0.5xGI50 concentrations) and paclitaxel at
doses ranging from 2 to 25 nM, for 48 h. As shown in Figure 4, concentrations of
paclitaxel lower than 25 nM had no significant effect on cell viability. Of note, we
found that at 1xGI50 15 was able to sensitize cancer cells to doses as low as 1 nM of
paclitaxel, as judged by the resulting decrease in cell viability (46.8%) comparatively to
cell viability in the 15 (67.9%) or paclitaxel (92.3%) individual treatments (Figure 4).
Interestingly, while cell viability was not significantly affected (96.9%) with 0.5xGI50
concentration of 15, its combination with paclitaxel from 5 nM doses greatly
compromised cell viability (51.8%). Collectively, these observations suggest that the
compound 15 increases the cytotoxic effect of paclitaxel, and that the doses of both
antimitotic agents could be lowered in a combination therapy, thereby providing an
opportunity to minimize resistance or side effects issues.

316
317
318
319
320
321
Figure 4. The chalcone 15 enhances sensitivity of cancer cells to paclitaxel. Relative
cell viability of NCI-H460 cells treated with 15 (at 0.5xGI50 or 1xGI50) in combination
with paclitaxel at the indicated concentrations for 48 h. Statistical significance of treated
samples when compared with control, *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001.
Results are expressed as mean ± SD, from at least three independent experiments.
322
2.4. Lipophilicity Evaluation
323
324
325
326
327
328
329
330
331
332
Lipophilicity was evaluated using the gold standard octanol:water system (log P) and
the biomimetic system composed by liposomes and buffer at physiological conditions
(pH 7.4 and 37 °C). For each compound, log P values were predicted by different in
silico methods (Table S4 – Support materials). As depicted in Figure 5a, the log P
values varied significantly, with standard deviation up to 1 logarithm unit, depending on
the in silico method used for the prediction. It should be highlighted that compound
were identified in pair-wise positional isomers, with exceptions of compounds 13, 14,
and 15. Positional isomers had equal predicted log P values (for instance, compound 3
and 4). This is attributed to the fact that the contributions of each atom/fragment for log
P calculation is independent of the atom/fragment position.
333
334
Partition coefficients in biomimetic models (log Kp) were determined using 1,2-
dimyristoyl-sn-glycero-3-phosphorylcholine (DMPC) liposomes as it is a suitable

335
336
337
338
339
340
341
model for this class of compounds [38] . Log Kp were determined by recording its
second-derivative absorption spectra in the presence of increasing lipid concentrations
at physiological conditions (pH 7.4 and 37 °C) [36] . The experimental data, second
derivative vs. lipid concentration, at the absorption maximum wavelength is then fitted
into Equation 1 to calculate Kp value [42-44] . As an example, Figure 5b shows the
plot of the second derivative vs. lipid concentration at 334 nm for chalcone 17 and its
corresponding best fitted curve.
342

343
344
345
346
Figure 5. (a) Mean, median and standard deviation of the Log P predicted by different
in silico methods. (b) Chalcone 17 best fitting curve to the experimental second-
derivative data at a wavelength of 334 nm. (c) Experimentally obtained Log Kp values,
and mean and median of predicted log P for the studied chalcones.
347
348
349
350
351
352
353
354
355
356
357
358
359
360
The experimentally determined log Kp values are in Table S5 – Support materials, and
summarized in Figure 5c. The majority of chalcones showed log Kp values which were
slightly higher than the predicted log P values. All compounds were not ionized at pH
7.4, the difference between log Kp and log P relied on the non-aqueous phase. As
liposomes are able to take into account more interactions, namely the electrostatic ones,
the partition coefficient values obtained with this system were slightly higher.
Exceptions to this general trend were the benzothiophene based chalcone 16, with a log
KP smaller than the log P; the benzofuran based chalcone 17, with log KP substantial
higher than the predicted log P; and the di-ortho chloro-substituted analogues 8, 14, and
23, with log KP substantial lower than the predicted log P. The behavior verified for
chalcone 17 was in accordance with previous reports for very lipophilic compounds (log
P around 5) and compounds bearing phenol groups [38] . Regarding the
dichlorobenzene based analogues, which had similar predicted log P values, different
log Kp values were obtained depending on the position of chloro groups.
361
362
363
364
365
366
367
368
369
370
371
372
373
From the obtained results, a structure-property relationship (SPR) can be proposed
(Figure 6a). Concerning the A ring, 1,2-dimetoxybenzene based chalcones had slight
lower log Kp values when compared with thiophene based chalcones. For the thiophene
based chalcones, the substitution with a methyl group, independently of position,
promotes a very slight increase of the log KP value. Moreover, the substitution of the
thiophene moiety for a benzothiophene also promotes a small increase of the log KP,
whereas the substitution for a 7-hydroxybenzofuran promotes a huge increase of the log
KP. When the chalcone enone moiety is substituted by the α,β-epoxide moiety,
lipophilicity is decreased, whereas the substitution by a pyrazole moiety lead to an
lipophilicity increment. Regarding the B ring, while di-ortho chloro-substituted
analogues showed log KP values lower than di-meta substituted analogues, di-ortho
methoxy-substituted analogues showed log KP values slightly higher than di-meta
substituted analogues. This feature is more prominent with di-ortho chloro-substituted

374
375
analogues. For instance, log KP of the di-meta chlorochalcone 13 was 4.43, and log KP
of the di-ortho chlorochalcone 14 was 2.80.
376
377
378
379
380
Figure 6. (a) Structure-lipophilicity relationship of chalcone derivatives. (b)
Comparison between the GI50 of the chalcone derivatives on NCI-H460 cell line and the
log KP. The insert displays the comparison of the most potent compounds (GI50 < 8 µM)
and their molecular weight.
381
382
383
384
385
386
GI50 and log KP of the compounds were examined in order to investigate relationship
between the compounds lipophilicity and the antiproliferative activity. Figure 6b
depicts the relationship between lipophilicity and the antiproliferative activity on NCI-
H460 cell line. Similar relationships were found on the other cell lines (Figure S1 –
Support materials). None of the compounds showing a log KP below 3 exhibited GI50
values lower than 10 µM. The most potent compounds (GI50 < 8 µM) have log Kp

387
388
389
390
391
392
393
values between 3.30-3.68 (dashed box, Figure 6b). Interestingly, these compounds, that
share similar lipophilicity behavior, do not share the same chemical features, as it can
been inferred by broad distribution of their molecular weight (as illustrated in the insert
of Figure 6b). In fact, these compounds have the broadest distribution of size as 274.3
and 358.4 corresponds to the lowest and highest molecular weight of all synthetized
compounds, respectively. Moreover, increasing the lipophilicity above 4 did not
improve significantly the activity.
394
395
3. Material and Methods
3.1. General information
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
Microwave reactions were performed using a glassware setup for atmospheric-pressure
reactions or a 100 mL Teflon reactor (internal reaction temperature measurements with
a fiber-optic probe sensor), using an Ethos MicroSYNTH 1600 Microwave Labstation
from Milestone. All reactions were monitored by thin-layer chromatography (TLC).
The purification of compounds by flash chromatography and preparative TLC was
performed using Macherey-Nagel silica gel 60 (0.04-0.063 mm) and Macherey-Nagel
silica gel 60 (GF254) plates, respectively. Melting Points were obtained in a Köfler
microscope and are uncorrected. Optical rotation measurements were carried out on a
1
Polartronic Universal polarimeter (ADP 410 polarimeter). H and 13C NMR spectra
were taken in CDCl3 at room temperature, on Bruker Avance 300 and 500 instruments
1
13
(300.13 MHz or 500 MHz for H and 75.47 or 120 MHz for C). Chemical shifts are
expressed in δ (ppm) values relative to tetrametylsilane (TMS) used as an internal
reference; 13C NMR assignments were made by 2D (HSQC and HMBC) NMR
experiments (long-range 13C-1H coupling constants were optimized to 7 Hz). The
spectral treatment was executed using the MestReNova v6.0.2-5475 software. HRMS
mass spectra of compounds 3-23 were performed on an APEXQe FT-ICR MS (Bruker
Daltonics, Billerica, MA), equipped with a 7T actively shielded magnet, at C.A.C.T.I.-
University of Vigo, Spain. Ions were generated using a Combi MALDI-electrospray
ionization (ESI) source. HRMS mass spectra of compounds 24-27 were performed on
an LTQ OrbitrapTM XL hybrid mass spectrometer (Thermo Fischer Scientific, Bremen,
Germany) controlled by LTQ Tune Plus 2.5.5 and Xcalibur 2.1.0. at CEMUP-

417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
University of Porto, Portugal. 1,2-Dimyristoyl-sn-glycero-3-phosphorylcholine
(DMPC) was purchased from Avanti Polar Lipids, Inc. (Alabaster, AL, USA). The
remaining commercially available reagents were purchased from Sigma Aldrich (St.
Louis, MO, USA) and Alfa Aesar (Ward Hill, MA, USA). HEPES buffer (HEPES: 10
mmol L–1, pH 7.4) was prepared using double deionized water from Arium® water
purification system (resistivity > 18 MΩ cm, Sartorius, Goettingen, Germany), and the
ionic strength was adjusted with NaCl (I=0.1 M). Absorption spectra for log KP
determination were recorded with a microplate reader (Synergy HT; Bio-Tek
Instruments, Winooski, VT, USA) and the double beam spectrophotometer (Cary 1E,
Varian Inc., Palo Alto, CA, USA) was a JASCO V660. Log S was predicted by 6
programs/online servers, including SwissADME [45], PreADMET [46] , pkCSM [47] ,
AquaSol [48] , and ALOGPs [49] . PSA was calculated by ChemDraw (v. 18.1),
SwissADME [45], and Molinspiration [50]. Log P calculations were given by 6
programs/online servers, including ChemDraw (v. 18.1), SwissADME [45],
PreADMET [46], pkCSM [47] , Molinspiration [50], and ALOGPs [49] .
432
3.2. Synthesis
433
434
435
436
437
438
439
440
441
442
443
444
445
Synthesis of chalcones 3-23. To a solution of appropriately substituted ketone (1.01 -
7.93 mmol, 0.19 - 1.05 g) in methanol (25 mL) was added an aqueous solution of 40%
sodium hydroxide (methanol/water) until pH 13 - 14, under stirring and on ice. Then, a
solution of appropriately substituted benzaldehyde (2.03 - 15.86 mmol, 0.28 - 3.11 g) in
methanol was slowly added to the reaction mixture. The reaction was submitted to
successive 15 min periods of microwave irradiation of 180 W of power, with the final
temperature of 69 °C. Total irradiation time 2 h - 3 h, and was monitored by TLC. Upon
completion, the reaction mixture was poured into ice and the pH was adjusted to
approximately 7 with diluted hydrochloride acid. For all chalcones, except 4, 10, 12, 19
and 20 the resulting residue was filtered, washed with water, and purified as described
below. For chalcones 4, 11, 12, 19 and 20 the resulting residue was taken in chloroform
and further rinsed with water, dried over anhydrous sodium sulfate, evaporated under
reduced pressure and the obtained residue was purified as described below.

446
447
448
449
The structure elucidation of compounds 3-6, 8, 11, 12, and 18-23 was established by 1H
13
and C NMR techniques and data were in accordance to the previously reported [51-
62]. The structure elucidation of new compounds 7, 9-10, 13-17, and 24-27 was
established on basis of HRMS and NMR techniques.
450
451
452
453
454
455
456
457
458
(E)-3-(3,5-dichlorophenyl)-1-(thiophen-2-yl)prop-2-en-1-one
(7).
Purified
by
1
crystallization (methanol: chloroform). Yield: 46%; mp: 101-103 ºC; H NMR (CDCl3
300.13 MHz): δ 7.89 (1H, dd, J=3.8, 1.1 Hz, H-5’), 7.73 (1H, dd, J=4.5, 1.1 Hz, H-3’),
7.70 (1H, d, J=15.6 Hz, H-β), 7.50 (1H, d, J=1.8 Hz, H-6), 7.41 (1H, d, J=15.6 Hz, H-
α), 7.40 (1H, t, J=1.8 Hz, H-4), 7.21 (1H, dd, J=4.9, 3.9 Hz, H-4’), 6.50 (1H, d, J=1.8
Hz, H-2,-6); 13C NMR (CDCl3, 75.47 MHz): δ 181.3 (C=O), 145.1 (C-2’), 140.9 (C-β),
137.7 (C-3,-5), 135.7 (C-1), 134.6 (C-3’), 132.3 (C-5’), 130.1 (C-4), 128.4 (C-4’), 126.5
(C-2,-6), 124.0 (C-α); ESI-HRMS (+) m/z: Anal. Calc. for C13H9Cl2OS (M+H)+:
282.97457; found: 282.97482.
459
460
461
462
463
464
465
466
467
(E)-1-(4-methylthiophen-2-yl)-3-(3,4,5-trimethoxyphenyl)prop-2-en-1-one (9). Purified
1
by crystallization (methanol). Yield: 52%; mp 140-142 °C; H NMR (CDCl3, 300.13
MHz): δ 7.76 (1H, d, J=15.5 Hz, H-β), 7.69 (1H, d, J=1.2 Hz, H-3’), 7.29 (1H, dd,
J=1.3, 1.1 Hz, H-5’), 7.27 (1H, d, J =15.5 Hz, H-α), 6.86 (2H, s, H-2,-6), 3.93 (6H, s, 3-
,5-OCH3), 3.91 (3H, s, 4-OCH3), 2.34 (3H, s, 4’-CH3); 13C NMR (CDCl3 75.47 MHz): δ
181.9 (C=O), 153.5 (C- 3,-5), 145.0 (C-2’), 144.0 (C-β), 140.5 (C-4), 139.0 (C-4’),
133.7 (C-3’), 130.7 (C-1), 129.8 (C-5’), 120.9 (C-α), 105.7 (C-2,-6), 61.0 (4-OCH3),
56.3 (3-,5-OCH3), 15.7 (4’-CH3); ESI-HRMS (+) m/z: Anal. Calc. for C17H19O4S
(M+H)+: 319.09986; found: 319.09920.
468
469
470
471
472
473
474
475
(E)-1-(4-methylthiophen-2-yl)-3-(2,4,5-trimethoxyphenyl)prop-2-en-1-one (10). Purified
by flash column chromatography (n-hexane: acetone, 7:3) followed by crystallization
1
(methanol: chloroform). Yield: 23 %; mp: 107-109 ºC; H NMR (CDCl3 300.13 MHz):
δ 8.11 (1H, d, J=15.6 Hz, H-β), 7.65 (1H, d, J=1.2 Hz, H-3’), 7.35 (1H, d, J= 15.6 Hz,
H-α), 7.24 (1H, dd, J= 1.3, 1.1 Hz, H-5’), 7.12 (1H, s, H-6), 6.53 (1H, s, H-3), 3.95 (3H,
s, 2-OCH3), 3.92 (3H, s, 4-OCH3); 3.91 (3H, s, 5-OCH3), 2.33 (3H, s, 4’-CH3); 13C
NMR (CDCl3, 75.47 MHz): δ 182.5 (C=O), 154.8 (C-2), 152.5 (C-4), 145.5 (C-2’),
143.2 (C-5), 139.2 (C-β), 138.9 (C-4’), 133.3 (C-3’), 129.0 (C-5’), 119.9 (C-α), 115.4

476
477
478
(C-1), 111.7 (C-6), 96.8 (C-3), 56.6 (4-OCH3), 56.4 (5-OCH3), 56.1 (2-OCH3), 15.7 (4’-
CH3); ESI-HRMS (+) m/z: Anal. Calc. for C17H19O4S (M+H)+: 319.09986; found:
319.09957.
479
480
481
482
483
484
485
486
(E)-3-(3,5-dichlorophenyl)-1-(4-methylthiophen-2-yl)prop-2-en-1-one (13). Purified by
1
crystallization (methanol: chloroform). Yield: 65%; mp: 112-115 ºC; H NMR (CDCl3
300.13 MHz): δ 7.69 (1H, d, J=1.2 Hz, H-3’), 7.67 (1H, d, J=15.7 Hz, H-β), 7.49 (2H, d,
J=1.9 Hz, H-2,-6), 7.39 (1H, t, J=1.9 Hz, H-4), 7.37 (1H, d, J=15.7 Hz, H-α), 7.23 (1H,
dd, J=1.3, 1.1 Hz, H-5’), 2.34 (3H, s, 4’-CH3); 13C NMR (CDCl3, 75.47 MHz): δ 181.6
(C=O), 145.0 (C-2’), 141.0 (C-β), 139.7 (C-4’), 138.2 (C-1), 136.1 (C-3,-5), 134.7 (C-
3’), 131.0 (C-5’), 130.4 (C-4), 126.9 (C-2,-6), 124.5 (C-α), 16.1 (4’-CH3); ESI-HRMS
(+) m/z: Anal. Calc. for C14H11Cl2OS (M+H)+: 296.99022; found: 296.99025.
487
488
489
490
491
492
493
494
495
(E)-3-(3,4-dichlorophenyl)-1-(4-methylthiophen-2-yl)prop-2-en-1-one (14). Purified by
1
crystallization (methanol: chloroform). Yield: 67%; mp: 120-122 ºC; H NMR (CDCl3
300.13 MHz): δ 7.72 (1H,d, J=1.9 Hz, H-2), 7.70 (1H, d, J=15.6 Hz, H-β), 7.69 (1H, d,
J=1.2 Hz, H-3’), 7.50 (1H, d, J=8.4 Hz, H-5), 7.45 (1H, dd, J=8.4, 1.9 Hz, H-6), 7.35
13
(1H, d, J=15.6 Hz, H-α), 7.31 (1H, dd, J=1.3, 1.1 Hz, H-5’), 2.33 (3H, s, 4’-CH3); C
NMR (CDCl3, 75.47 MHz): δ 181.4 (C=O), 144.6 (C-2’), 141.0 (C-β), 139.2 (C-4’),
134.8 (C-1), 134.4 (C-4), 134.2 (C-3’), 133.3 (C-3), 131.0 (C-5’), 130.4 (C-5), 129.7
(C-2), 127.6 (C-6), 123.2 (C-α), 15.7 (4’-CH3); ESI-HRMS (+) m/z: Anal. Calc. for
C14H11Cl2OS (M+H)+: 296.99022; found: 296.99019.
496
497
498
499
500
501
502
503
504
505
(E)-1-(5-methylthiophen-2-yl)-3-(3,4,5-trimethoxyphenyl)prop-2-en-1-one (15). Purified
by flash column chromatography (n-hexane: ethyl acetate, 7:3) followed by
crystallization (methanol: chloroform). Yield: 16 %; mp: 126-128 ºC; 1H NMR (CDCl3
300.13 MHz): δ 7.74 (1H, d, J=15.4 Hz, H-β), 7.70 (1H, d, J=3.5 Hz, H-3’), 7.26 (1H, d,
J= 15.5 Hz, H-α), 6.87 (1H, d, J= 3.5 Hz, H-4’), 6.85 (2H, s, H-2,-6), 3.93 (6H, s, 3-, 5-
13
OCH3), 3.90 (3H, s, 4-OCH3), 2.57 (3H, s, 5’-CH3); C NMR (CDCl3, 75.47 MHz): δ
181.5 (C=O), 153.5 (C-3, -5), 150.1 (C-5’), 143.6 (C-2’), 143.4 (C-β), 140.3 (C-4),
132.4 (C-3’), 130.4 (C-1), 126.9 (C-4’), 120.8 (C-α), 105.6 (C-2,-6), 61.0 (4- OCH3),
56.2 (3-,5- OCH3), 16.2 (5’-CH3); ESI-HRMS (+) m/z: Anal. Calc. for C17H19O4S
(M+H)+: 319.09986; found: 319.09904.

506
507
508
509
510
511
512
513
514
515
(E)-1-(benzo[b]thiophen-2-yl)-3-(3,4,5-trimethoxyphenyl)prop-2-en-1-one
(16).
Purified by flash column chromatography (n-hexane: ethyl acetate, 7:3) followed by
1
crystallization (methanol: chloroform). Yield: 38 %; mp> 350ºC; H NMR (CDCl3,
300.13 MHz): δ 8.13 (1H, s, H-3’), 7.95-7.90 (2H, m, H-4,’-7’), 7.83 (1H, d, J=15.5
Hz, H-β), 7.58 (1H, d, J=15.5 Hz, H-α), 7.52-7.41 (2H, m, H-5’,-6’), 6.91 (2H, s, H-2,-
13
6), 3.96 (6H, s, 3-,5-OCH3), 3.92 (3H, s, 4-OCH3); C NMR (CDCl3 75.47 MHz): δ
171.2 (C=O), 153.5 (C-3,-5), 145.2 (C-8’), 144.6 (C-β), 143.0 (C-2’), 140.7 (C-4),
139.3 (C-9’), 130.1 (C-1), 128.8 (C-3’), 127.4 (C-7’), 125.9 (C-4’), 125.1 (C-6’), 123.1
(C-5’), 120.4 (C-α), 105.8 (C-2,-6), 61.1 (4-OCH3), 56.3 (3-,5-OCH3); ESI-HRMS (+)
m/z: Anal. Calc. for C20H19O4S (M+H)+: 355.09986; found: 355.09935.
516
517
518
519
520
521
522
523
524
525
526
(E)-1-(7-hydroxybenzofuran-2-yl)-3-(3,4,5-trimethoxyphenyl)prop-2-en-1-one
(17).
Purified by flash column chromatography (n-hexane: ethyl acetate, 5:5) followed by
flash column chromatography (n-hexane: ethyl acetate, 5:5). Yield: 12 %; mp: 190-
1
192ºC; H NMR (CDCl3, 300.13 MHz): δ 7.89 (1H, d, J = 15.6 Hz, H-β), 7.68 (1H, s,
H-3’), 7.43 (1H, d, J=15.6 Hz, H-α), 7.29 (1H, dd, J=7.8, 1.1 Hz, H-4’), 7.21 (3H, t,
J=7.8 Hz, H-5’), 7.06 (1H, dd, J=7.7, 1.1 Hz, H-6’), 6.91 (2H, s, H-2,-6), 3.94 (6H, s, 3-
,5-OCH3), 3.92 (3H, s, 4-CH3); 13C NMR (CDCl3 75.47 MHz): δ 179.5 (C=O), 153.6
(C-2’), 153.5 (C-3,-5), 145.3 (C-β), 144.6 (C-8’), 141.8 (C-7’), 140.9 (C-4), 130.0 (C-
1), 128.8 (C-9’), 125.9 (C-6’), 120.1 (C-α), 115.0 (C-4’), 114.2 (C-3’,-5’), 106.1 (C-2,-
6), 61.1 (4-OCH3), 56.3 (3-,5-OCH3); ESI-HRMS (+) m/z: Anal. Calc. for C20H19O6
(M+H)+: 355.11761; found: 355.11710.
527
528
529
530
531
532
533
534
535
Synthesis of epoxide derivatives 24 and 25. To a solution of chalcones 3 (1.56 mmol, 0.5
g) or 9 (1.50 mmol, 0.5 g) in methanol or methanol:acetone (3:2, 30 mL) was added
powdered K2CO3 (0.622 g, 4.5 mmol) or a solution of 5% NaOH (3.84 mL),
respectively, followed by the dropwise addition of 30% hydrogen peroxide (0.96 mL,
8.5 mmol) at 0-4 ºC. The reaction mixture was stirred at room temperature for 2-3 h,
and after was left in the dark overnight at -20 ºC. The obtained solid was filtered, and
washed with water. The solid obtained was purified by crystallization (dichloromethane:
methanol).

536
537
538
539
540
541
542
543
544
Thiophen-2-yl(3-(3,4,5-trimethoxyphenyl)oxiran-2-yl)methanone (24). Yield: 61%; mp
(dichloromethane : methanol): 111-113 ºC; 1H NMR (CDCl3, 300.13 MHz): δ 8.01 (1H,
dd, J=3.9, 1.1 Hz, H-5’), 7.76 (1H, dd, J=4.9, 1.1 Hz, H-3’), 7.19 (1H, dd, J= 4.9, 3.9
Hz, H-4’), 6.57 (2H, s, H-2,-6), 4.13 (1H, d, J= 1.8 Hz, H-1’’), 4.03 (1H, d, J= 1.8 Hz,
H-2’’), 3.87 (6H, s, 3, 5-OCH3), 3.86 (3H, s, 4-OCH3); 13C NMR (CDCl3, 75.47 MHz):
δ 186.3 (C=O), 153.7 (C-3, -5), 140.9 (C-2’), 138.5 (C-4), 135.3 (C-3’), 133.7 (C-5’),
130.9 (C-1), 128.5 (C-4’), 102.5 (C-2, -6), 62.0 (C-2’’), 60.9 (4-OCH3), 59.7 (C-1’’),
56.2 (3-,5-OCH3); ESI-HRMS (+) m/z: Anal. Calc. for C16H16NaO5S (M+Na)+:
343.06107; found: 343.06327.
545
546
547
548
549
550
551
552
(4-methylthiophen-2-yl)(3-(3,4,5-trimethoxyphenyl)oxiran-2-yl)methanone (25). Yield:
1
58%; mp (dichloromethane: methanol): 104-108 ºC; H NMR (CDCl3, 300.13 MHz): δ
7.79 (1H, d, J=1.2 Hz, H-3’), 7.35 (1H, t, J= 1.2 Hz, H-5’), 6.56 (2H, s, H-2,-6), 4.12
(1H, d, J= 1.8, H-1’’), 4.00 (1H, d, J= 1.8, H-2’’), 3.87 (6H, s, 3-,5-OCH3), 3.86 (3H, s,
4-OCH3), 2.31 (1H,s, 4’ -CH3). 13C NMR (CDCl3, 75.47 MHz): δ= 186.2 (C=O), 153.7
(C-3,-5), 140.6 (C-2’), 139.4 (C-4, -4’), 135.4 (C-3’), 131.3 (C-1), 131.0 (C-5’), 102.4
(C-2,-6), 61.8 (C-2’’), 60.9 (4-OCH3), 59.6 (C-1’’), 56.2 (3-,5-OCH3), 15.5 (4’-CH3);
ESI-HRMS (+) m/z: Anal. Calc. for C17H1905S (M+Na)+: 357.07672; found: 357.07976.
553
554
555
556
557
558
559
560
Synthesis of pyrazole derivatives 26 and 27. To a solution of epoxide derivatives 24
(0.9 mmol, 317 mg) or 25 (1.2 mmol, 387 mg) in xylenes (4-7 mL) and
dichloromethane (2-4 mL) was added hydrazine hydrate (2.7-3.6 mmol) and p-
toluenesulfonic acid monohydrate (0.11-0.15 mmol, 24.5-30.5 mg). The reaction
mixture was stirred during 3-5 h at 100 ºC. After, the xylenes were removed under
nitrogen stream and the solid obtained was purified as indicated below for the referred
pyrazole derivatives.
561
562
563
564
565
3-(thiophen-2-yl)-5-(3,4,5-trimethoxyphenyl)-1H-pyrazole (26). Purified by flash
column chromatography (n-hexane: ethyl acetate, 6:4) followed by preparative TLC (n-
1
hexane: ethyl acetate, 6:4). Yield: 4%; mp 70-72ºC; H NMR (CDCl3, 300.13 MHz): δ
7.36 (1H, d, J= 3.8 Hz, H-5’), 7.27 (1H, dd, J= 5.0, 3.6 Hz, H-3’), 7.03 (1H, d, J=5.0,
3.8 Hz, H-4’), 6.96 (2H, s, H-2,-6), 6.69 (1H, s, H-2’’), 3.90 (6H, s, 3-,5-OCH3), 3.88

566
567
568
569
(3H, s, 4-OCH3); 13C NMR (CDCl3, 75.47 MHz) δ= 153.5 (C-3,-5), 147.5 (C-1’’),
144.0 (C-2’), 138.8 (C-4), 134.0 (C-3’’), 127.9 (C-3’), 126.0 (C-5’), 125.8 (C-1), 125.2
(C-4’), 103.3 (C-2, -6), 100.1 (C-2’’), 61.0 (4 -OCH3), 56.3 (3-, 5- OCH3); ESI-HRMS
(+) m/z: Anal. Calc. for C16H17N2O3S (M+H)+: 317.09544; found: 317.09495.
570
571
572
573
574
575
576
577
578
3-(4-methylthiophen-2-yl)-5-(3,4,5-trimethoxyphenyl)-1H-pyrazole (27). Purified by
flash column chromatography (n-hexane: ethyl acetate, 6:4) followed by preparative
1
TLC (n-hexane: ethyl acetate, 6:4). Yield: 1%; mp 84-86 ºC; H NMR (CDCl3, 500
MHz): δ 7.64 (1H, brs, H-5’), 7.13 (1H, brs, H-3’), 6.87 (1H, s, H-2’’), 6.63 (2H, s, H-
13
2,-6), 3.89 (6H, s, 3-,5-OCH3), 3.80 (3H, s, 4-OCH3), 2.22 (3H, s, 4’-CH3); C NMR
(CDCl3, 120 MHz) δ= 153.2 (C-3,-5), 143.1 (C-1’’), 137.9 (C-2’), 137.2 (C-4), 133.7
(C-3’’), 128.5 (C-4’), 128.0 (C-3’), 125.1 (C-5’), 121.2 (C-1), 109.3 (C-2’’), 106.1 (C-2,
-6), 61.0 (4 -OCH3), 56.1 (3-,5- OCH3), 15.7 (4’-CH3); ESI-HRMS (+) m/z: Anal. Calc.
for C17H19N2O3S (M+H)+: 331.11109; found: 331.11060.
579
580
3.3. Growth Inhibitory effect in Human Tumor Cell Lines
581
582
583
584
585
Cell Culture. The human tumor cell lines melanoma A375-C5, breast adenocarcinoma
MCF-7, and non-small cell lung cancer NCI-H460 (European Collection of Cell
Culture, UK) were grown in RPMI-1640 culture medium (Lonza, Basel, Switzerland),
supplemented with 5% fetal bovine serum (FBS), and cells were maintained at 37 °C in
a 5% CO2 humidified atmosphere.
586
587
588
589
590
591
Compound preparation. All compounds were dissolved in sterile DMSO at 60 mM
stock, with the exception 26 which was dissolved at 10 mM. Nocodazole and paclitaxel
(Sigma-Aldrich Co., Saint Louis, MO, USA) were dissolved in DMSO as 0.5 mM
stock. To avoid repeated freeze-thaw cycles, the 60 or 10 mM of stock solution were
stored in different aliquots, at -20 ºC. Appropriate dilutions of the compounds were
freshly diluted in RPMI medium culture.
592
593
594
Viability assays. Cells were plated in 96-well plates at 0.05 × 106 cells/well in complete
medium at 37 °C for twenty-four h. Then, two-fold serial dilutions of the test
compounds were added to culture medium, at concentrations ranging from 0 to 150 µM,

595
596
597
598
599
600
601
602
603
604
605
for 48 h. Control groups were treated with equivalent amount, up to 0.25%
concentration, of the solvent dimethyl sulfoxide (DMSO, Sigma-Aldrich). Then, cells
were fixed with 50% (m/v) trichloroacetic acid (Merck Millipore, Darmstadt,
Germany), washed with distilled water and stained with sulforhodamine B (SRB;
Sigma-Aldrich) for 30 min at room temperature. After a 5 times wash with 1% (v/v)
acetic acid (Merck Millipore), plates were left to dry at room temperature before SRB
complexes solubilization with 10 mM Tris buffer (Sigma-Aldrich) for 30 min. Cell
survival was measured through determination of the absorbance at 515 nm in a
microplate reader (Biotek Synergy 2, BioTek Instruments, Inc., Winooski, VT, USA).
The 50% cell growth inhibition concentration (GI50) was calculated using the dose–
response curve established for each test compound.
606
607
608
609
610
611
612
613
614
615
616
Alternatively, for combination experiments with paclitaxel, cell viability was assessed
by MTT assay. Cells were plated in 96-well plates (0.05×106 cells/well) in complete
culture medium and incubated at 37 °C for 24 h. Cells were then incubated for 48 h with
the test compound, paclitaxel, or a combination of both at 37 °C and 5% CO2. Then,
cells were placed in fresh serum-free medium, and 3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide (MTT), previously dissolved in PBS, was added to each
well (0.5 mg/ml) and incubated for 4 h at 37 °C and 5% CO2. Formazan crystals were
then solubilized by adding 100 µL solubilization solution (89% Isopropanol, 10% of
Triton-100, 0.37% HCl) for 2 h. Absorbance was measured at 570 nm in a microplate
reader (Biotek Synergy 2), and retrieved through the Gene5 software (Biotek). Viability
was calculated relative to untreated cells.
617
618
619
620
621
622
623
Determination of antimitotic activity. NCI-H460 cells (2 x 105) were grown in six-well
dishes and treated for 24 h with the test compounds. Treatment with 1 µM nocodazole
served as a positive control. Controls included untreated cells and cells treated with the
highest concentration of DMSO used to dissolve the compounds. Mitotic index,
percentage of mitotic cells over total cell population, was determined by cell-rounding
under phase-contrast microscopy. For each condition, a total of 2000 mitotic and
interphasic cells were counted from more than ten random microscope fields.

624
625
626
627
628
629
630
631
632
633
634
635
636
637
Indirect Immunofluorescence. A total of 0.15x106 cells / well NCI-H460 cells were
grown on poly-L-lysine coated coverslips, in six-well dishes, and were treated with the
test compounds during 15 h. The respective controls were included as described for
mitotic index determination. Cells were fixed with 2% paraformaldehyde (Sigma-
Aldrich) in PBS for 12 min, and washed 3 times with PBS for 5 min. Cells were then
permeabilized in 0.5% Triton X-100 in PBS for 7 min, and washed 3 times with PBS,
for 5 min each. To avoid unspecific binding of the primary antibody, cells were treated
with a blocking solution, consisting of 10% FBS in PBST (PBS plus 0.5% Tween 20),
for 30 min, followed by an 1 h incubation with the primary antibody, mouse anti-α-
Tubulin (Sigma-Aldrich) diluted at 1:2500 in PBST containing 5% FBS. After rinsing
in PBST, cells were incubated for an extra 1 h with the Alexa Fluor 488 (Molecular
Probes) conjugated secondary antibody diluted at 1:1500 in PBST with 5% FBS. DNA
was stained with 2 µg/mL 4′, 6-diamidino-98 2-phenylindole (DAPI, Sigma-Aldrich)
diluted in Vectashield mounting medium (Vector, H-1000).
638
639
640
641
642
643
644
645
646
647
Image acquisition and processing. Phase-contrast microscopy images were recorded
with a 10x objective, on a Nikon TE 2000-U microscope (Amsterdam, Netherlands),
using a DXM1200F digital camera (Amsterdam, Netherlands) and with Nikon ACT-1
software (version 2.62, Melville, NY, USA). Fluorescence images were acquired with
Plan Apochromatic 63x/NA 1.4 objective or with 40x objective on an Axio Observer
Z.1 SD microscope (Carl Zeiss, Germany), coupled to an AxioCam MR3. Z-stacks were
acquired with 0.4 µm intervals and, after image deconvolution with AxioVision Release
SPC software (version 4.8.2, Carl Zeiss, Germany), they were processed using ImageJ
(version 1.44, Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda,
MD, USA).
648
649
650
651
652
Statistical analysis. Statistical analysis was performed using an Unpaired Student t-test,
or two-way ANOVA, in GraphPad Prism version 6 (GraphPad software Inc., CA,
USA). Alpha value was 0.05 and the confidence interval 95%. Data are presented as the
means ± standard deviation (SD) of at least three independent experiments, and was
considered statistically significant when p<0.05.
653
3.4. Lipophilicity evaluation