A comprehensive assessment of a new series of 5′,6′-difluorobenzotriazole-acrylonitrile derivatives as microtubule targeting agents (MTAs)

Article abstract of DOI:10.1016/j.ejmech.2021.113590

Microtubules (MTs) are the principal target for drugs acting against mitosis. These compounds, called microtubule targeting agents (MTAs), cause a mitotic arrest during G2/M phase, subsequently inducing cell apoptosis. MTAs could be classified in two groups: microtubule stabilising agents (MSAs) and microtubule destabilising agents (MDAs). In this paper we present a new series of (E) (Z)-2-(5,6-difluoro-(1H)2H-benzo[d] [1,2,3]triazol-1(2)-yl)-3-(R)acrylonitrile (9a-j, 10e, 11a,b) and (E)-2-(1H-benzo[d] [1,2,3]triazol-1-yl)-3-(R)acrylonitrile derivatives (13d,j), which were recognised to act as MTAs agents. They were rationally designed, synthesised, characterised and subjected to different biological assessments. Computational docking was carried out in order to investigate the potential binding to the colchicine-binding site on tubulin. From this first prediction, the di-fluoro substitution seemed to be beneficial for the binding affinity with tubulin. The new fluorine derivatives, here presented, showed an improved antiproliferative activity when compared to the previously reported compounds. The biological evaluation included a preliminary antiproliferative screening on NCI60 cancer cells panel (1–10 μM). Compound 9a was selected as lead compound of the new series of derivatives. The in vitro XTT assay, flow cytometry analysis and immunostaining performed on HeLa cells treated with 9a showed a considerable antiproliferative effect, (IC50 = 3.2 μM), an increased number of cells in G2/M-phase, followed by an enhancement in cell division defects. Moreover, β-tubulin staining confirmed 9a as a MDA triggering tubulin disassembly, whereas colchicine-9a competition assay suggested that compound 9a compete with colchicine for the binding site on tubulin. Then, the co-administration of compound 9a and an extrusion pump inhibitor (EPI) was investigated: the association resulted beneficial for the antiproliferative activity and compound 9a showed to be client of extrusion pumps. Finally, structural superimposition of different colchicine binding site inhibitors (CBIs) in clinical trial and our MDA, provided an additional confirmation of the targeting to the predicted binding site. Physicochemical, pharmacokinetic and druglikeness predictions were also conducted and all the newly synthesised derivatives showed to be drug-like molecules.

Full text of DOI:10.1016/j.ejmech.2021.113590

European Journal of Medicinal Chemistry 222 (2021) 113590
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
A comprehensive assessment of a new series of 50,60-
difluorobenzotriazole-acrylonitrile derivatives as microtubule
targeting agents (MTAs)
,
Federico Riu a, Luca Sanna b, Roberta Ibba a *, Sandra Piras a, Valentina Bordoni b,
,
, f, **
M. Andrea Scorciapino c, Michele Lai d e, Simona Sestito a, Luigi Bagella b
,
Antonio Carta a
a Department of Chemistry and Pharmacy, University of Sassari, Via Vienna 2, 07100, Sassari, Italy
b Department of Biomedical Sciences, University of Sassari, Viale San Pietro 43/b, 07100, Sassari, Italy
c Department of Chemical and Geological Sciences, University of Cagliari, S.P. 8 Km 0.700, 09042, Monserrato (CA), Italy
d Retrovirus Centre, Department of Translational Medicine and New Technologies in Medicine and Surgery, University of Pisa, Strada Statale Del Brennero, 2,
Pisa, Italy
e CISUP - Centre for Instrumentation Sharing - University of Pisa, Lungarno Pacinotti 43, Pisa, Italy
f Sbarro Institute for Cancer Research and Molecular Medicine, Center for Biotechnology, College of Science and Technology, Temple University, Philadelphia,
PA, 19122, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Microtubules (MTs) are the principal target for drugs acting against mitosis. These compounds, called
microtubule targeting agents (MTAs), cause a mitotic arrest during G2/M phase, subsequently inducing
cell apoptosis. MTAs could be classified in two groups: microtubule stabilising agents (MSAs) and
microtubule destabilising agents (MDAs). In this paper we present a new series of (E) (Z)-2-(5,6-difluoro-
(1H)2H-benzo[d] [1,2,3]triazol-1(2)-yl)-3-(R)acrylonitrile (9a-j, 10e, 11a,b) and (E)-2-(1H-benzo[d] [1,2,3]
triazol-1-yl)-3-(R)acrylonitrile derivatives (13d,j), which were recognised to act as MTAs agents. They
were rationally designed, synthesised, characterised and subjected to different biological assessments.
Computational docking was carried out in order to investigate the potential binding to the colchicine-
binding site on tubulin. From this first prediction, the di-fluoro substitution seemed to be beneficial
for the binding affinity with tubulin. The new fluorine derivatives, here presented, showed an improved
antiproliferative activity when compared to the previously reported compounds. The biological evalu-
Received 16 December 2020
Received in revised form
21 May 2021
Accepted 27 May 2021
Available online 1 June 2021
Keywords:
Microtubule targeting agents
Benzotriazoles
Tubulin
Extrusion pump inhibition
Cell growth inhibition
Molecular docking
ation included a preliminary antiproliferative screening on NCI60 cancer cells panel (1e10 mM). Com-
pound 9a was selected as lead compound of the new series of derivatives. The in vitro XTT assay, flow
cytometry analysis and immunostaining performed on HeLa cells treated with 9a showed a considerable
antiproliferative effect, (IC50 ¼ 3.2
m
M), an increased number of cells in G2/M-phase, followed by an
-tubulin staining confirmed 9a as a MDA triggering
enhancement in cell division defects. Moreover,
b
tubulin disassembly, whereas colchicine-9a competition assay suggested that compound 9a compete
with colchicine for the binding site on tubulin. Then, the co-administration of compound 9a and an
extrusion pump inhibitor (EPI) was investigated: the association resulted beneficial for the anti-
proliferative activity and compound 9a showed to be client of extrusion pumps. Finally, structural su-
perimposition of different colchicine binding site inhibitors (CBIs) in clinical trial and our MDA, provided
an additional confirmation of the targeting to the predicted binding site. Physicochemical, pharmaco-
kinetic and druglikeness predictions were also conducted and all the newly synthesised derivatives
showed to be drug-like molecules.
© 2021 Elsevier Masson SAS. All rights reserved.
* Corresponding author. Department of Chemistry and Pharmacy, University of Sassari, Via Vienna 2, 07100, Sassari, Italy.
** Corresponding author. Department of Biomedical Sciences, University of Sassari, Viale San Pietro 43/b, 07100, Sassari, Italy.
E-mail addresses: ribba@uniss.it (R. Ibba), lbagella@uniss.it (L. Bagella).
https://doi.org/10.1016/j.ejmech.2021.113590
0223-5234/© 2021 Elsevier Masson SAS. All rights reserved.

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
1. Introduction
oxidation [26]. On this basis, our project aimed to design and
synthesise new optimised benzotriazole-acrylonitrile derivatives
suitably functionalised with two small halogen atoms, such as
fluorine, in position 50 and 60 on the benzotriazole moiety as
depicted in Fig.1. Therefore, in this paper we present a new series of
(E) (Z)-2-(5,6-difluoro-(1H)2H-benzo[d] [1e3]triazol-1(2)-yl)-3-(R)
acrylonitrile (9a-j, 10e, 11a,b) and (E)-2-(1H-benzo[d] [1e3]triazol-
1-yl)-3-(R)acrylonitrile derivatives (13d,j), which were rationally
designed, synthesised, characterised and submitted to different
biological assessments.
Cancer is one of the noncommunicable diseases (NCDs) and it is
expected to become the leading death cause and the principal
obstacle for the increase of life expectancy in the 21st century [1].
Deaths caused by cancer represent approximately 22% of all NCD
casualties [2]. The actual shortage of efficient and safe drugs for
cancer treatment is a driving force for research efforts in this field
[3]. One of the main feature of a tumour cell is its uncontrolled
division. Cell division cycle consists of four successive phases called
G1 (Gap 1), S (DNA synthesis), G2 (Gap 2), and M (mitosis). Cell
cycle phases convey important information regarding the cytotoxic
potential of anticancer drugs [4]. A crucial element involved during
cell replication is the mitotic spindle, mainly composed by micro-
2. Material and methods
2.1. Molecular docking
tubules (MTs), which are filaments consisting in a- and b-tubulin
heterodimers. An attractive approach pursued for the development
of new therapeutic agents for cancer treatment is the interference
with the MT dynamic equilibrium [5e7]. A wide variety of com-
pounds targeting mitotic spindle have been designed and investi-
gated [8]. However, little is known about the mechanism of cell
death triggered by those drugs. Generally, drugs acting against cell
division cycle lead to mitotic arrest during G2/M phase, subse-
quently inducing cell apoptosis. Nowadays, microtubule targeting
agents (MTAs) are some of the most effective drugs used in the
treatment of both solid and haematological tumours [9]. MTAs
could be classified in two groups on the basis of their mechanism of
action: microtubule stabilising agents (MSAs) and microtubule
destabilising agents (MDAs). Both types of agents showed to sup-
press cellular microtubule dynamics [10]. MTAs could also be cat-
egorised into 5 classes based on their interactions with the taxane-,
vinca-, colchicine-, laulimalide/peloruside-, pironetin- or the
maytansine-binding site on tubulin [11]. Colchicine binding site
inhibitors (CBSIs), comprise a variety of small molecules which
include colchicine and combretastatins A-1 (CA-1) and A-4 (CA-4),
AutoDock Vina 1.1.2 [27]. version 1.5.6 was selected for its
known sampling and scoring functions [28]. Tubulin-Colchicine
complex (PDB code: 4O2B; resolution: 2.30 Å) structure was pro-
vided by the Protein Data Bank (rcsb.org) [29]. Protein structures
have been employed removing H2O molecules, MES (buffer, 2-(N-
morpholino)-ethanesulfonic acid), GOL (glycerol), ions (Ca2þ
,
Mg2þ), GTP (Guanosine-50-triphosphate), GDP (Guanosine-50-
diphosphate), but also chains C, D, E, F, in order to analyze the
docking poses of ligands in a free binding pocket. The structures
were visualised and analysed in PyMOL 2.3.4. AutoDockTools (ADT)
[30] was used to obtain not only the PDBQT files of both protein
structures and all ligands, but also to determine the docking Grid
Box (30-30-30 Å for x,y,z coordinates). Docking was performed
using the protein in its rigid form, or by making some amino acids
flexible: Sera178, Thra179, Alaa180, Alaa181. These amino acids are
known to establish important H-bonds with the colchicine-binding
pocket of tubulin. The exhaustiveness value was set to 8 and the
number of poses to 10 for each Vina docking calculation. The
binding energy range was imposed to be ꢀ 2 kcal/mol above that of
the best binding pose for each ligand. 2D-pictures of the binding
interactions between ligands and the protein binding pocket were
made with two molecular-graphics proþgrams: LigPlotþ v.2.2. and
which bind in the colchicine site (CBS) located on
b-tubulin at its
interface with -tubulin [12,13]. Although some CBISs, such as
a
combretastatin A-4 phosphate (CA-4P) and A-1 diphosphate (CA-
1P), 2-methoxyestradiol and verubulin have been evaluated in
phase 1 and 2 clinical trials [14], no CBIS compound has been
approved as anticancer agent so far. Hence, this binding site pro-
vides new challenging opportunities for drug discovery [15e17].
We already published a group of benzotriazolacrylonitrile de-
rivatives which turned out to act as CBSIs [18e23]. All derivatives
can be grouped into two different chemical classes: 1H-benzo-
triazole derivatives which showed good activity, and the parental
2H-benzotriazole derivatives with a weaker antiproliferative ac-
tivity [18]. Among all synthesised derivatives, (E)-2-(1H-benzo[d]
[1e3]triazol-1-yl)-3-(4-methoxyphenyl)acrylonitrile was identi-
fied as lead compound, hereinafter and in literature [19] labelled as
compound 34. Cell cycle analysis on K-562 cells (leukaemia)
revealed that compound 34 inhibits the G2/M phase of cell cycle
€
Maestro (Schrodinger package). LigPlot processed in 2D-dimen-
sion the protein-ligand interaction with a cutoff of 5 Å from the
centre of the ligand. Maestro's Ligand Interaction Diagram is a 2D
binding site representation known for the accuracy in the place-
ment of residues. Structural superimposition of the docked pose of
compound 9a and the crystal structure of tubulin co-crystallised
with colchicine, as well as other CBSIs, were represented with
PyMOL. The crystallised structures of tubulin/ligand complexes
were obtained from the Protein Data Bank: colchicine (PDB code:
4O2B), tivantinib (TIV, PDB code: 5CB4), plinabulin (PLI, PDB code:
5C8Y), crolibulin (CRO, PDB code: 6JCJ).The docking calculations
were conducted on a PC with an Intel® Core™ i7-9705H CPU @
2.60 GHz with 8 GB RAM (operating system: Ubuntu 16.04, 6 CPUs
for each calculation).
[21], with an IC50 value of 0.2 mM on CCRF-CCM leukaemia cells
[19]. Overall, our experimental and computational investigations
supported the hypothesis that these compounds act as anti-
proliferative agents by interacting with tubulin in the colchicine-
binding site. Therefore, compound 34 could be classified as
MDAs, since it inhibits the tubulin assembly [21]. The novelty of the
present work is based on the introduction of fluorine atoms into the
benzotriazole moiety. The replacement of hydrogens with fluorine
atoms is a widely used approach in medicinal chemistry to improve
relevant drug-like properties such as metabolic stability, bioavail-
ability and ligand-protein interactions (by altering the pKa) [24]. As
a matter of fact, aromatic fluorination can also increase the com-
pound lipophilicity [25], and could be used to protect some meta-
bolically weak sites easily affected by Cytochrome P450-dependent
Fig. 1. General structure of previously reported compounds (left) and newly designed
di-fluoro benzotriazole-acrylonitrile derivatives (right).
2

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
2.2. Chemistry
registered in solutions in deuterated DMSO and recorded with a
Bruker Avance III 400 NanoBay (400 MHz). 1H NMR chemical shifts
are reported in parts per million (ppm) downfield from tetrame-
thylsilane (TMS) used as internal standard. Chemical shift values
2.2.1. Materials
3,4-difluoroaniline (1) and the aldehydes (8aej) were
commercially available.
are reported in ppm (d) and coupling constants (J) are reported in
Hertz (Hz). The assignment of exchangeable protons (OH and NH)
was confirmed by the addition of D2O. Signal multiplicities are
represented as s (singlet), d (doublet), dd (doublet of doublets), ddd
(doublet of doublets of doublets), t (triplet), m (multiplet) and wm
(wide multiplet). 13C NMR chemical shifts are reported in downfield
from tetramethylsilane (TMS) used as internal standard, jmod (J-
modulated spin-echo for X-nuclei coupled to H-1 to determine
number of attached protons) was selected as most suitable method.
For the E/Z-isomers characterisation (Paragraph 1.1.), 1H NMR
spectra were acquired on an Agilent INOVA-500 spectrometer with
2.2.2. Synthetic methods
The synthesis, as depicted in Scheme 1, starts with the
commercially available 3,4-difluoroaniline (1), which was in the
first place acetylated with acetic anhydride, gaining compound 2.
The following nitration of the amidic compound 2 with a solution of
KNO3 in H2SO4conc (concentrated H2SO4), resulted in the formation
of nitro-derivative 3. Then, 4,5-difluoro-2-nitroaniline 4 was ob-
tained by hydrolytic deprotection with H2SO4conc. The reduction of
nitro group brought to 4,5-difluorobenzene-1,2-diamine (5). The
dianiline derivative 5 is cyclized with NaNO2 in HCl, obtaining 50,60-
difluorobenzotriazole (6). Reaction of 6 with chloroacetonitrile
(ClCH2CN) was performed, in the presence of KOH in acetonitrile, to
obtain two geometric isomers bearing an acetonitrile chain on the
triazole ring 7a,b. The final step corresponds to a Knoevenagel
condensation between each acetonitrile isomer (7a,b), separately,
and the appropriate, commercially purchased, aldehydes (8a-j),
bringing to final (E) (Z)-2-(5,6-difluoro-(1H)2H-benzo[d] [1e3]tri-
azol-1(2)-yl)-3-(R)acrylonitrile derivatives 9a-j, 10e, and 11a,b. The
final step generally brings to the sole formation of the E-isomer,
only in one case a Z-isomer compound was identified and purified
from the reaction mixture. A Knoevenagel condensation between
2-(1H-benzo[d] [1e3]triazol-1-yl)acetonitrile 12a and aldehydes
8d,j gave final compounds 13d,j. Compounds were purified by flash
chromatography or by crystallisation from ethanol. For flash chro-
matography, Merck silica gel 60 was used with a particle size of
0.040e0.063 mm (230e400 mesh ASTM), and a proper eluent
mixture of petroleum spirit (PS), diethyl ether (DE), ethyl acetate
(EA), chloroform (CHCl3) and methanol (CH3OH).
an operative Larmor frequency of 500.3 MHz, by using a 6.1 ms pulse
(90ꢁ), 1 s delay time, 1.5 s acquisition time, 8 transients and a
spectral width of 9 kHz. 1He1H correlation TOCSY experiments
were recorded over the same spectral window using 2048 complex
points and sampling each of the 256 increments with 8 scans and
by applying 80 ms spin-lock with the MLEV-17 mixing scheme. The
same acquisition parameters were applied for the acquisition of the
NOESY experiments with 200 ms mixing time. In addition, a series
of selective DPFGSE (Double Pulse Field Gradient Spin-Echo) one-
dimensional NOESY [31] were performed with the same parame-
ters as for standard 1H acquisition with 128 transients and either
250, 500 and 750 ms mixing time. Resonance assignments were
obtained on the basis of relative intensity, chemical shift and fine
structure, together with results from TOCSY and NOESY 2D spectra
as well as the database at www.nmrdb.org [32].
2.3. Biology
2.3.1. NCI60 testing screen: in vitro antiproliferative assay
The first in vitro anti-cancer screening was performed through
the NCI60 test, provided by the Developmental Therapeutics Pro-
gram of the National Cancer Institute (NCI, Bethesda, USA). At the
beginning, a single high dose of 10 mM of the compound is tested in
the full NCI60 cell panel. If the inhibition results agree with the
selection criteria, the same compound can be tested again in
2.2.3. Chemical characterisation
Evaporation was performed in vacuo (rotating evaporator). So-
dium sulfate was always used as the drying agent. Celite® 545 was
used as filter agent. Commercially available chemicals were pur-
chased from Sigma Aldrich and Carlo Erba Reagents. Retention
factors (Rf) were obtained using a mixture of petroleum spirit (PS)
and ethyl acetate (EA) as eluents (8/2 respectively), to develop Thin
Layer Chromatographies (TLCs), Merck F-254 plates were used.
5 ꢃ 10-fold dilutions, from 100
mM up to 0.01 mM. More details
about the NCI60 experiments are reported in the Supplementary
Material (Table S1 and Figures S34e53) and in the NCI website
(https://dtp.cancer.gov).
€
Melting points (m.p.) of the compounds were measured in a Kofler
hot stage in open capillaries or Digital Electrothermal melting point
apparatus and are uncorrected. The compounds were dissolved in
CH3CN for HPLC (concentration of 1.0e2.0 ppm) for ESI-MS char-
acterisation. Mass spectra (full mass) were obtained on a Q Exactive
Plus Hybrid Quadrupole-Orbitrap mass spectrometer of Thermo
Fisher Scientific, in the positive-ion and negative-ion mode. The
2.3.2. HeLa cells culture
HeLa cells (human cervix epithelioid carcinoma) (ATCC, Rock-
ville, MD) have been cultured with Dulbecco's Modified Eagle
Medium (DMEM) (Gibco). Medium has been supplemented with
10% of Fetal Bovine Serum (FBS), 100
mg/mL streptomycin, 100
solutions were infused at a flow rate of 5.00
m
L/min into the ESI
units/mL penicillin (Gibco) and 1% of L-glutamine. Cells were
chamber. The spectra were recorded in the m/z range 150e800 at a
resolution of 140 000 and accumulated for at least 2 min in order to
increase the signal-to-noise ratio. Measurements conditions for
positive-ion were as follows: spray voltage 2300 V, capillary tem-
perature 250 ꢁC, sheath gas 10 (arbitrary units), auxiliary gas 3
(arbitrary units), sweep gas 0 (arbitrary units) and probe heater
temperature 50 ꢁC. Measurements conditions for negative-ion were
as follows: spray voltage ꢂ1900 V, capillary temperature 250 ꢁC,
sheath gas 20 (arbitrary units), auxiliary gas 5 (arbitrary units),
sweep gas 0 (arbitrary units), probe heater temperature 50 ꢁC. ESI-
MS spectra were analysed by using Thermo Xcalibur 3.0.63 soft-
ware (Thermo Fisher Scientific), and the Xtract tool (integrated into
the software) was used to extract the average deconvoluted mon-
oisotopic masses. Nuclear Magnetic Resonance (NMR) spectra were
incubated at 37 ꢁC in a humidified atmosphere containing 5% of
CO2.
2.3.3. Proliferation assay on HeLa cells
HeLa cells were seeded at a density of 1500 cells/wells in a 96
wells-plate. After 24 h from seeding, cells were treated with
different concentrations of 9a in a range between 0.1
mM and 5 mM
for 24 and 48 h in a final volume of 100 L. Before using, 9a has been
m
resuspended in DMSO which has been used as control. After 24 and
48 h of incubation with 9a, XTT assay was performed using Cell
Proliferation Kit II (Roche, Basel, Switzerland) and following the
manufacture's protocol. Specifically, a volume of 100
74.5 L of medium, 25 L of labelling reagent and 0.5
electron coupling was added to each well and then incubated at
m
L with
m
m
mL of XTT
3

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
Scheme 1. (A) Synthesis of (E) (Z)-2-(5,6-difluoro-(1H)2H-benzo[d] [1e3]triazol-1(2)-yl)-3-(R)acrylonitrile (9a-j, 10e, 11a,b) and (B) (E)-2-(1H-benzo[d] [1e3]triazol-1-yl)-3-(R)acry-
lonitrile derivatives (13d,j). Reaction conditions: (a) (CH3COO)2O, 10 min, 0 ꢁC; (b) KNO3, H2SO4, 2h, r.t.; (c) H2SO4 conc, 2h, reflux; (d) H2, Pd/C, 2 h, r.t.; (e) NaNO2, HCl, 20 h, r.t.; (f)
ClCH2CN, KOH, DMF, o/n, reflux; (g): (1) TEA, toluene, 110 ꢁC, or (2) DIMCARB, CH3CN, 60 ꢁC or (3) piperidine, CH3CN, 60 ꢁC; properly chosen to get the full conversion.
37 ꢁC for 4 h. After incubation, the absorbance was quantified at
490 nm using a spectrophotometric plate reader (SPECTRAMax 384
PLUS). Later on, data obtained were used for the calculation of IC50
value using Graphpad Prism software (San Diego, CA, USA). More
than three independent experiments were carried out for each
treatment and performed in triplicate.
adherent and floating cells (dead cells) were initially centrifuged
for 5 min at 3000 rpm, then washed in PBS and fixed with ice-cold
ethanol at 70% in agitation on a vortex to dissolve cluster and
permeabilise the membranes. After incubation overnight
at ꢂ20 ꢁC, fixed HeLa cells were washed twice with cold PBS and
finally resuspended in 200 mL of a mix containing PBS and 20 mL/
test of 7-AAD (Bioscience, San Diego, CA) and incubated at room
temperature for 20 min. Flow cytometry was performed for cell
cycle analysis using BD FACS CANTO II. For each sample 20 000
events were collected and then analysed through BD FACS DIVA
software.
2.3.4. Cell cycle analysis
Based on their phenotype and population doubling, HeLa cells
were seeded at 50% of confluence on 6 cm dishes. After 24 h, cells
were treated with a concentration of 5 or 10 mM of 9a. Later on,
4

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
2.3.5. Immunostaining
of compounds in complete medium were added to each sample to
obtain indicated treatment concentrations. Plating of cells, prepa-
ration of serial dilutions and addition of compounds to cells were
performed using an automated liquid handling platform (Gilson
Pipetmax). Immediately before compound addition to cells and
after 24, 48 and 72 h from treatment, images for far-red fluores-
cence (Cy5 filter), green fluorescence (GFP filter) and phase contrast
were taken with objective 4x using automated digital widefield
microscopy system BioTek Cytation 5. Cy5 filter: led cube 625 nm,
filter cube excitation 650 30, emission 800 90. GFP filter: led
cube 465 nm, filter cube excitation 469 25, emission 525 25.
During each reading, cells were maintained at 37 ꢁC, 5% CO2. For
each well, four images were taken and merged to cover the entire
well. Number of nuclei stained by SiR-DNA in far-red fluorescence,
and number of dead cells stained by CellTox Green Dye in green
fluorescence, were automatically counted using BioTek Gen5 soft-
ware. Alive cells were calculated by subtracting number of dead
cells from total count of nuclei.
HeLa cells (105/well) were seeded 24 h before treatment with 9a
(0.1 and 0.3 mM). After 48 h cells were fixed with PFA 4% (para-
formaldehyde). Cells were permeabilised via three washes in PBS
containing 0.3% Triton X-100 (PBST) and then blocked in PBST
containing 1% BSA. Primary antibody
b-tubulin (ab179512) was
added in blocking solution for 2 h. Then, 488 anti-rabbit secondary
antibody (Alexa Fluor, Thermo Scientific) was incubated for 1 h.
Images were taken using 20x objectives (Evos FL2 Auto, Thermo
Scientific). Statistical analysis was performed on 3 biological and
technical replicates, using the Kruskal-Wallis test with multiple
comparisons.
2.3.6. High-content imaging of HeLa cells and EC50 calculation
Imaging experiments were performed using an Operetta CLS
high-content imaging device (PerkinElmer, Hamburg, Germany),
and analysed with Harmony 4.6 software (PerkinElmer). To assess
the tubulin integrity, we analysed HeLa cells using 63x magnifica-
tion, taking 25 fields per sample in biological and technical tripli-
cates. Data were analysed using the following building blocks: 1-
Find Nuclei, 2- Find Cytoplasm (Tubulinþ). EC50 was obtained by
counting the number of nuclei in each field. Cell dimensions were
assessed using the following building blocks: 1- Find Nuclei, 2- Find
Cytoplasm (Tubulinþ), 3- Calculate Morphology properties (Area),
as previously described [33].
2.4. Physicochemical, pharmacokinetic and druglikeness
predictions
SwissADME (http://www.swissadme.ch) predictions were per-
formed for each final compound [34]. The prediction does not
discriminate between E- and Z-isomer. The physicochemical prop-
erties comprise typical molecular characteristics, including the TPSA
(Topological Polar Surface Area). Log Po/w is calculated through
differentmethods, andan average value isreported asconsensus Log
Po/w. iLOGP, XLOGP3, WLOGP, MLOGP, SILICOS-IT methods contrib-
uted to the averaged Log P [35]. The solubility and the Log S of each
compound are also determined through several methods. Pharma-
cokinetics properties are represented by the prediction of the GI
(gastrointestinal), BBB (blood-brain barrier) and skin absorption,
but also the affinity (substrate or inhibitor) for some important
proteins (e.g. P-gp) and metabolic enzymes. Druglikeness is
expressed through different methods, e.g. the Lipinski ‘rule of five’
[36], and the bioavailability score. Other druglikeness predictions
were calculated with the Ghose, Veber, Egan and Muegge filters.
Final considerations are the synthetic accessibility and the lead-
likeness. The latter parameter outline the similarity of structure and
physico-chemical properties with a definable “lead” compound.
2.3.7. Colchicine-9a competition assay
Tubulin polymerization assay was carried out using the kit
supplied by Cytoskeleton (BK006P, Citoskeleton, Denver, CO). It is
based on the principle that light is scattered by microtubules to an
extent that fluorescence is proportional to the concentration of the
microtubule polymer. The absorbance was recorded at 340 nm at
37 ꢁC for 30 min with fixed acquisitions every 2 s. Colchicine-9a
competition assay was performed incubating equimolar concen-
trations of tubulin with
colchicine, 9a or colchicineþ9a. After 30 min fluorescence
emitted by tubulin polymerization was recorded by reading
absorbance for 30 min.
2.3.8. Cytotoxicity evaluation of compound 9a in presence or
absence of extrusion pump inhibitor (EPI)
Cells were plated in a 384 multiwell plate and treated with 9a (5
concentrations) in absence or presence (co-treatment) of EPI SS26.
As negative control, cells were treated with vehicle solution (DMSO
0.3%). All treatments were performed at 0.3% DMSO concentration.
Three technical replicates were performed. The number of alive and
dead cells was quantified in each well of the multi-well plate
immediately before compound addition and at 24, 48 and 72 h of
treatment by Kinetics of Cytotoxicity and Proliferation assay. Per-
centages of growth, percentages of dead cells, GI50, TGI, and LC50
were calculated at each time point to evaluate reduction of prolif-
eration and induction of cell death, data reported in Table S2.
3. Results and discussion
3.1. Molecular docking
In order to evaluate the feasibility and the benefit of the fluorine
substitution on the benzotriazole moiety, molecular docking sim-
ulations were performed ahead of the synthetic efforts. Aiming to
compare their binding affinity with the tubulin active site, colchi-
cine as a known ligand, compound 34 and the corresponding newly
designed difluoro-derived compound (labelled derivative 9b) were
docked in the colchicine-binding site on tubulin. The X-ray struc-
ture of tubulin co-crystallised with colchicine, PDB ID: 4O2B, was
selected. Colchicine was removed from its binding pocket in tubulin
and re-docked in the same binding pocket with the same co-
ordinates, showing affinity energy of ꢂ9.8 kcal/mol. The same
protein structure, without colchicine in place, was employed for
docking studies on compounds 34 and 9b. The best-docked pose for
derivative 34 presented a binding energy value of ꢂ6.7 kcal/mol,
meanwhile the best-docked pose of compound 9b ranked affinity
energy of ꢂ8.6 kcal/mol. All dockings were analysed with PyMOL
[37], Maestro [38] and LigPlot [39]. As depicted in Fig. 2, 9b has two
F,,,N electrostatic interactions, better-called fluorine bonds [40],
involving the electron-acceptor fluorine in position 5 on the
2.3.8.1. Cell culture. A498, A-704, A549, and HT1197 cells were
cultured in DMEM high glucose supplemented with 10% FBS, 2 mM
L
-Glutamine, Non-essential amino acids, 1 mM Sodium Pyruvate,
and Antibiotic Antimycotic Solution. Cells were cultured at 37 ꢁC,
5% CO2 humidified air.
2.3.8.2. Kinetics of cell proliferation and cytotoxicity. 24 h before
compound addition, 500 cells suspended in 20
mL phenol-red free
complete medium containing SiR-DNA 0.5 M (Tebu-bio SC007)
m
and CellTox Green Dye 0.8x (Promega G8731) were plated in each
well of a 384 Well Flat Clear Bottom Black Polystyrene TC-Treated
Microplates (Corning 3764). The day after, 10 mL of serial dilutions
5

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
benzotriazole scaffold (depicted in Fig. 2B as green dashes). These
dipole-dipole interactions involve intrinsic nitrogen atoms of
Then, colchicine was re-docked in its binding pocket. The
eptacyclic (tropolone) ring has a hydrophobic interaction with
Cys
b
241 and Leu
b
242 (~4 Å). Leu
b
242 contributes also with hy-
Ala
wards the hydrophobic portion of the tubulin-binding pocket.
Met 259, Asn 258, Leu 255, Lys 254, Asp 251, Ala 250, Leu 248,
Leu 242 and Cys 241 are the representative aas of this portion of
a180. Trimethoxyphenyl moiety of colchicine is oriented to-
drophobic interactions between the carbons of the isobutyl side
chain of the amino acid (aa) and the benzotriazole carbons. The
benzene moiety of the benzotriazole provides hydrophobic in-
b
b
b
b
b
b
b
b
b
teractions with Ala
chain benzene moiety establishes van der Waals interactions with
aas located in the opposite portion of the binding pocket: Leu 248,
Glu 183, Lys 254 and Asn 258. The amidic nitrogen atom of
Asn 101 interacts with the acceptor oxygen of the methoxyl group
b
250, Asp
b
251 and Leu
b
255, while the side-
the pocket. Most of these aas also establish hydrophobic in-
teractions with compound 9b, especially with the heterocyclic
scaffold of the molecule, but also with the acrylonitrile linkage, as
shown in Fig. 2B. Fig. 3 shows polar and hydrophobic interactions
b
a
a
b
b
among compound 9b and a- and b-tubulin subunits in the
of compound 34, forming a hydrogen bond (4.18 Å, Fig. 2A). The
best binding pose of derivative 9b showed the methoxyl group in a
different conformation if compared with 34; accordingly, this de-
terminates only a weak interaction with the same asparagine res-
idue. Moreover, the carbon atom of the methoxyl group contributes
colchicine-binding pocket. Methoxyl group is shown to be the most
solvent-exposed portion of the molecule, pointing toward the outer
portion of tubulin active site.
In the binding pocket, four aas (Ser
Ala 181) were recognised to be crucial for the binding of colchicine.
Using AutoDockTools [30], these aas residues were set as flexible
a178, Thra179, Alaa180,
a
to the binding with a hydrophobic interaction with Lysb254.
Fig. 2. Schematic comparison between the binding pose of compounds 34 (A) and 9b (B) inside the tubulin-binding pocket. H-bonds are depicted as green dashed lines and hydrophobic
contacts as red hash marks. Atom colours: carbon, black; nitrogen, blue; fluorine, green; oxygen, red. Amino acids are labelled with the residue name, the chain and the correspondent
number. The hashes in the ligand show atoms having hydrophobic interactions with aas. Distances are given in Å. Figure made by using LigPlotþ. [39].
6

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
Fig. 3. Ligand Interaction Diagram and LID legend of the different interactions between compound 9b and different amino acids in tubulin binding pocket. Residue markers are coloured
according to residue type (green ¼ hydrophobic, cyan ¼ polar, red ¼ negative, purple ¼ positive). Green and cyan lines represent the interactions between 9b and the different portions of
the binding pocket. The gap in the line shows the opening of the pocket. Grey atom markers on the background of ligand structure represent the solvent-accessible surface area (SASA) of that
atom, and the marker size represents the amount of exposure. Each protein residue is depicted as a “guitar pick”: it points away from the ligand when the aa backbone faces towards the
ligand, or it points to the ligand when the aa side chain faces towards the ligand. Cut-off default of 4 Å was used. Figure made by using Maestro. [38].
during the docking calculations. Compound 34 best-predicted pose
showed an affinity energy value of ꢂ6.7 kcal/mol, meanwhile,
compound 9b presented a ranked affinity energy of ꢂ7.4 kcal/mol.
This preliminary docking prediction suggests that compound 9b
possesses a good binding profile and that the difluoro substitution
on the main benzotriazole scaffold, could be considered beneficial
for the affinity to the colchicine-binding site of tubulin, pushing us
forward to continue our work. Therefore, the new series of (E) (Z)-
2-(5,6-difluoro-(1H)2H-benzo[d] [1e3]triazol-1(2)-yl)-3-(R)acry-
lonitrile derivatives was designed, synthesised and tested in vitro
on different tumour cancer cells, as further reported in the present
work.
3.2. Chemistry
The synthetic route followed to synthesise final compounds 9a-
j, 10e, 11a,b and 13d-j is described in Scheme 1A, B. The synthesis,
as depicted in Scheme 1A, started with the commercially available
3,4-difluoroaniline (1), which was in first place acetylated, afford-
ing compound 2. The following nitration, hydrolysis and reduction
led to 4,5-difluorobenzene-1,2-diamine (5), which was cyclized
with NaNO2 in HCl, obtaining the 50,60-difluorobenzotriazole 6.
Reaction of 6 with chloroacetonitrile (ClCH2CN), carried out in the
presence of KOH, yielded the geometric isomers 7a,b bearing an
acetonitrile chain on the triazole ring (Scheme 1A).
7

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
As a final step (Scheme 1A), the Knoevenagel condensation
between each acetonitrile isomer (7a,b) and the appropriate
commercial aldehydes (8a-j), afforded the final (E) (Z)-2-(5,6-
difluoro-(1H)2H-benzo[d] [1e3]triazol-1(2)-yl)-3-(R)acrylonitrile
derivatives 9a-j, 10e, and 11a,b. Knoevenagel condensation gener-
ally provided the sole E-isomer. In some cases, the reaction
generated also the Z-isomer in very low quantity, In some cases, the
reaction generated also the Z-isomer in very low quantity, not
sufficient to be fully characterised (as explained in Paragraph 1.1 for
compound 9i). Z-isomer isolation from the reaction mixture was
possible only for the compound (10e). The reaction was performed
by one out of three different synthetic conditions: (1) TEA (trie-
thylamine) in toluene (as previously described by us [18e23]); (2)
DIMCARB (dimethylammonium dimethylcarbammate) used as
base and reaction catalyst, in acetonitrile and (3) piperidine in
acetonitrile (Scheme 1A). The presence of DIMCARB (in CH3CN) also
allowed to obtain the p-NO2-derived Z-isomer 10e, while using TEA
(in toluene) as base provided the E-isomer 9e as the major product.
Compounds 13d,j were obtained by Knoevenagel condensation
between the 2-(1H-benzo[d] [1e3]triazol-1-yl)acetonitrile isomer,
synthesised as previously reported [23], and the appropriate al-
dehydes 8d,j (Scheme 1B). The substituents on the acrylonitrile
moiety were selected based on the antimitotic activity properties of
the series of compounds previously reported by us [18e23] or by
colleagues [41]. As two new substituents were selected from the
literature (4-hydroxyphenyl and isoquinolin-5-yl), we also
designed and synthesised new parental compounds of the lead 34,
thus two new (E)-2-(1H-benzo[d] [1e3]triazol-1-yl)-3-(R)acrylo-
nitrile derivatives (13d,j), as new parental compounds of the lead
34.
Regarding 1H-benzotriazole derivatives (13d,j), compound 13d
demonstrated a low GI rate, but, contrariwise, isoquinolin-5-yl
derivative 13j showed an interesting antiproliferative activity and
was selected for the 5-doses dilution assay. At 10 mM, 13j was able
to inhibit the proliferation of the majority of tumour cell lines with
percentage values ranging between 70 and 100% (some of them are
reported in Fig. 5). At 1 mM, Compound 13j showed antiproliferative
activity against six cell lines belonging to four different tumours
type (Fig. 5), exhibiting growth inhibition percentages of 50e60%
against most of the reported cell lines. However, 13j showed the
best activity against CCRF-CEM leukaemia line, inhibiting
completely the tumour cell growth.
In light of the reported results for all the newly synthesised
compounds (Figs. 4 and 5 and Figures S34e53), 9a showed an
overall better antiproliferative potency and a wide range of activity.
Hence, derivative 9a was selected as lead compound for further
biological assessments.
3.3.2. IC50 and mechanism of action of compound 9a against HeLa
cells
In order to understand if 9a could be defined as a microtubule
destabilising agent (MDA), as well as its previously reported for
analogue compounds [21,22], its antiproliferative activity was
tested against HeLa cells through XTT assay. XTT is a colorimetric
assay useful to analyze the number of viable cells by the ability their
metabolic activity: tetrazolium salts added to the culture medium
are actively absorbed into cells and reduced by dehydrogenase
enzymes of metabolically active cells to yield a chromogenic for-
mazan product. Cells were treated at different concentrations of 9a
ranging between 0.1 and 5 mM, DMSO was used as control. XTT
assay was carried out after 24 and 48 h of treatment, the absor-
bance data were obtained and processed to calculate the percent-
age of cell viability [42]. A reduction of 50% of alive cells was
detected after 48 h of treatment with 9a with an IC50 value of
3.3. Biology
3.3.1. NCI60 in vitro screening
In order to evaluate the potential anticancer profile of our
compounds, we firstly focused on their antiproliferative ability in
cancer cells. The first in vitro anti-cancer screening was performed
through the NCI60 test (NCI, Bethesda, USA). In the beginning, a
3.2
m
M.
To prove that 9a was able to affect the HeLa cell cycle, flow
cytometry was used to evaluate the changes in the DNA content
after treatment. As shown in Fig. 6, the administration of 9a at 5
mM
single high dose of 10
m
M of the compound was tested on the full
causes an increase of cells number in G2/M-phase (54%) compared
to the control (30%) after 24 h of treatment. This effect was even
more evident at a higher concentration of 10 mM with a percentage
of G2/M-phase cells of 66% after 24 h. The increase of cells in G2/M-
phase causes a contemporary decrease of G1-phase cells both at
NCI60 cell panel. The panel comprises a set of different cancer lines
including solid (non-small cell lung, colon, central nervous system -
CNS, ovarian, renal, prostate, breast cancers, and melanoma) and
haematological (leukaemia) tumours. Results showed that, at
10
m
M concentration, 50,60-difluoro-substituted 1H-benzotriazole
5
m
M (39%) and 10
mM (20%) compared to the control (65%). Results
derivatives (9a-j, 10e) possess higher percentages of growth inhi-
bition (GI) in comparison to 2H-benzotriazole derivatives (11a,b).
All data from NCI60 are reported in Supplementary Material
(Figures S34e53). Among all the new synthesised derivatives,
compounds 9a, 9e, 9g and 9h satisfied the pre-determined
threshold inhibition criteria in a minimum number of cell lines
after 48 h showed slight recovery of cells in G1-phase at 5
but a dramatic increase of cells in G2/M-phase (87%) proportional
to the decrease of G1-phase rate (10%) after 48 h of treatment at
m
M (41%),
10 mM.
3.3.3. Compound 9a slows the cytokinesis in HeLa cells
and progressed to the second step of the 5-dose assay, from 100
m
M
Aiming to confirm the cell cycle impairment observed in Fig. 6,
we performed a fluorescence microscopy screening on HeLa cells
treated with 9a for 48 h. Based on the previously described cell
accumulation in G2/M phase we decided to count the number of
incomplete cytokinesis events per field. As shown in Fig. 7AeB, the
to 10 nM. Among them, derivative 9a exhibits GI values in the range
60e100% against all the sixty solid and haematological cell lines,
reaching GI percentages around 100% against all leukaemia cell
lines. More in depth, compound 9a turned out cytotoxic against
seventeen solid and haematological tumour cell lines with lethal
amount of incomplete cell divisions increases starting from 0.3 mM
percentages up to 90% (U251, CNS cancer cell line). At 1
m
M, com-
compared to controls. We also observed an overall increase in cell
dimension compared to controls (mock, untreated).
pound 9a presents GI percentages ranging from 40 to 60% against
eight tumour cell lines and 70e100% against further seven cell lines
(93% against NCIeH522 cell line of non-small cell lung cancer).
3.3.4. High-content imaging on HeLa cells
General GI50 values are up to 0.1
come up to 1 M.
In conclusion, compound 9a showed the best scores and was
selected as lead of this series. Antiproliferative activities of 9a on
NCI60 cancer cells are reported in Fig. 4.
m
M, while LC50 and TGI values
In order to gain insights in the potential targeting of microtu-
bules elicited by 9a, we analysed tubulin behaviour in a cellular
setting in the presence of our lead compound. Briefly, HeLa cells
were treated for 24 h with 9a at various concentrations (10, 25, 50,
m
100 and 500 nM). Cells were then stained for
b-tubulin and
8

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
Fig. 4. Graphical illustration of antiproliferative activity for compound 9a tested against NCI60 cell panel. Percentage of Growth Inhibition (%) is reported for each cell line tested,
grouped for type of cancer. For each group, a range of GI50, TGI and LC50 are reported; values are expressed as molar concentration (M).
processed using high-content confocal microscopy screening. As
shown in Fig. 8, we observed -tubulin disassembly starting from
colchicine-competition assay was planned in order to verify that 9a
specifically binds the colchicine site in tubulin, as previously pre-
dicted. Fig. 9 shows how colchicine and 9a affect the tubulin
polymerization process, and the competion of compound 9a and
colchicine for the same binding site, when co-administered. Results
confirmed that 9a slows tubulin polymerization down, as shown by
the blue-marked curve (Fig. 9). Moreover, lead compound 9a was
observed to compete with colchicine for the binding site as proved
by the decrease of colchicine effectiveness on tubulin assembly
(red-marked curve in Fig. 9).
b
50 nM. At the same concentration, the high-content screening
revealed that HeLa cells increased dimension by three times
compared to controls (Fig. 8B). This phenomenon indirectly con-
firms the cell-cycle blockade observed in Fig. 6. The increased
cytoplasmic area is compatible with senescent-like phenotype.
3.3.5. Colchicine-9a competition assay
Once 9a was proved to interfere with tubulin assembly, a
9

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
Fig. 5. Graphical illustration of representative antiproliferative activity results for compound 13j tested against NCI60 cell panel. Seven cell lines were selected and reported
coloured by type of cancer. The bar chart reports the percentage of Growth Inhibition for each cancer cell line when treated with 13j at 1 or 10 mM concentration. The table shows a
range of GI50, TGI and LC50 values (M) for each reported type of cancer.
3.3.6. Cytotoxicity evaluation of compound 9a in presence or
absence of extrusion pump inhibitor (EPI)
highlight that the association with EPI SS26 resulted beneficial for
the antiproliferative activity of 9a against all the 4 tested tumour
cell lines. The most striking results were obtained after 72 h of
treatment for all the tested cell lines. The cell growth inhibition was
highly increased when 9a þ SS26 were tested against A-704 cell
line at both 48 and 72 h. The most representative data concerned
Extrusion pumps (EPs) represent one of the main issues in
anticancer therapy since they are often overexpressed in drug-
resistant cancer cells. In order to investigate whether this chemi-
cal class of molecules could be client of EPs, we tested our lead
compound 9a in cell-based association assay. Four cancer cell lines
with EPs overexpression were selected [43]. Two kidney cancer
(A498, A-704), one non-small cell lung cancer (A549) and one
bladder cancer (HT1197) cell lines were selected for this experi-
ment. Two of them, A498 and A549 cell lines were already used in
the NCI screening and the results from the NCI60 cell line screening
were used to select a concentration range of compound 9a to be
used in the association assay. The four cancer cell lines were treated
with the sole derivative 9a or in association with an extrusion
pump inhibitor (EPI) from our group library, compound here
labelled as SS26. Compound SS26 (structure not showed for Intel-
lectual Property rights) EP inhibition activity was previously vali-
dated by association with Doxorubicin (a know EP client) in HCT-15
colon cancer cell line, as shown in Figure S54. The association assay
results are reported in Fig. 10 and Figure S55 and S56. Fig. 10 shows
the Percentage of Growth (%) for each tumour cell line, at each time
point at 5 different concentration of compound 9a (1, 2, 4, 8 and
the two concentrations of 4 and 8
mM of compound 9a since at
16 M concentration the cells were saturated and the EPI effect was
m
not detectable any longer. Figure S55 presents the raw cell growth
values, while Figure S56 shows in bar charts the concentration-
dependent efficacy.
We obtained two main pieces of information from this associ-
ation assay: i) we proved that these small molecules are client of
extrusion pumps with a negative effect on the antiproliferative
activity; ii) we demonstrated that, by increasing the concentration
of derivative 9a in the cytosol, the antiproliferative activity is
potentiated.
3.3.7. Compound 9a docks in the colchicine-binding site on tubulin
As a result of the biological assessments previously reported,
derivative 9a was selected as the lead compound of this series of
molecules. Molecular docking studies were conducted on 9a in the
colchicine-binding site on tubulin, using the same docking pa-
rameters and procedures already followed for colchicine, com-
pounds 34 and 9b. Docking calculations evidenced the role of the
fluorine bonds: the fluorine atom in position 6 established two
16
m
M), alone and in association with our EPI at a fixed concen-
M). The data are reported as Loess model with 95%
tration (1
m
confidence interval. The data collected at 24 h time-point are pre-
sented for completeness but were not considered for the cytotoxic
effect analysis since the cell growth is not yet stable at 24 h after
seeding. The results of the experiment at 48 and 72 h clearly
different bonds with the amino groups of Ala
Leu 255 (4.96 Å) of -tubulin. Moreover, the oxygen of Val
-tubulin engages in H-bond with the N3 in the benzotriazole
b
250 (4.75 Å) and
b
b
b238 of
b
10

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
Fig. 6. Cell cycle phase distribution in HeLa cells after 9a treatment. A. Representative cell cycle profiles of HeLa cells treated with 5
mM or 10 mM of 9a for 24 or 48 h. Treatment with
5
m
M of 9a causes an accumulation of cells in G2/M phase after 24 h with a completed block at 10 M after 48 h post-treatment. B. Percentages of cells in the different phases are
m
shown.
nucleus of 9a (4.28 Å), while the carbonyl oxygen of Ala317 of
b-
Then, the conformation state of the top-ranked pose of com-
tubulin interacts with the cyan group of the acrylonitrile linker of
pound 9a was superimposed with the crystal structure of colchicine
9a (4.29 Å). Compound 9a also establishes different nonpolar in-
in its binding site, at the interface of a- and b-tubulin. As shown in
teractions in the colchicine binding site, especially with the
subunit of tubulin. In particular, the benzotriazole scaffold mainly
interacts with the aa residues Asp 251, Ile 318 and Ala 354, while
the tolyl moiety establishes hydrophobic interactions with
Leu 248, Asn 258, Met 259 and Lys 352. Fig. 11 illustrates the
docking conformation of compound 9a.
b
-
Fig. 12, the benzotriazole scaffold occupies the trimethoxyphenyl
portion of the colchicine-binding site, while the CH2eCN chain
almost overlaps with one of the methoxy groups located in the
same aromatic portion. Additionally, the tolyl moiety of 9a overlaps
with the phenoxy ring of colchicine (Fig. 12). Overall, our results
suggest that the favourable binding mode of lead compound 9a
b
b
b
b
b
b
b
11

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
Fig. 7. Fluorescence microscopy screening of cytokinesis. A. Images, taken 48 h treatment with 9a and controls stained with b-tubulin, show an increase of incomplete cell divisions starting
from 0.3
mM. B. Kruskal-Wallis test with multiple comparisons performed on the number of incomplete cell division per field. ***P(<0,0001); ns ¼ not significant.
Fig. 8. High content confocal screening of HeLa cells. A. Increasing doses of 9a were used to treat HeLa cells for 24 h. Then, cells were stained for b-tubulin (green) and nuclei (DAPI-
blue). Starting from 50 nM we observed the disruption of microtubules. B. One way ANOVA analysis performed on cytoplasm area using high-content screening. Data are expressed
as mean SD. Experiments were performed with three biological and three technical replicates.
docks similarly to the crystal structure of colchicine in its binding
pocket of the -tubulin dimer.
PDB code: 5C8Y) [44] and crolibulin (CRO, PDB code: 6JCJ) [45].
Tivantinib (TIV) is a well-known CBSI in phase I/II/III clinical trials
for the treatment of different tumor form [46e48]. The direct
binding of TIV to the colchicine-binding site was recently described
[49] and the analysis of the tubulineTIV crystal structure evidenced
how TIV binding site and mode overlap with that of colchicine. In
a,b
To gain further insights, we decided to compare the mode of
binding of the new proposed CBSI 9a with some of the most
important and well-known CBSIs actually in clinical trials (phase I,
II or III). In Fig. 13, we underlined the difference in binding mode
and affinity to tubulin of the different studied molecular structures:
compound 9a, tivantinib (TIV, PDB code: 5CB4) [44], plinabulin (PLI,
particular, TIV establishes several direct H-bonds with Asn
b
256
179
Ala 315 of etubulin, but also interacts with Leu 246 and Thr
b
b
b
a
12

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
profile. As we noticed from our previously developed compounds
[18e22], the results confirmed that 1H-benzotriazole derivatives
(9a-j, 10e) possess a better profile activity than 2H-benzotriazole
ones (11a,b). Overall, in vitro results suggested the p-CH3 derived
compound 9a as the most active compound: at 1 mM concentration,
it showed GI percentages between 40 and 60% against eight tumour
cell lines and 70e100% against further seven cell lines. On this basis,
derivative 9a was selected as lead compound of this library of
molecules to be subjected to further investigations. To prove its
activity as MDA, effects of 9a on cell cycle were analysed. Our lead
candidate demonstrated its antiproliferative activity against HeLa
cells with an IC50 ¼ 3.2
mM, blocking most of the cells in G2/M phase
and causing an increased number of cell division defects. Fluores-
cence microscopy analysis showed a slowed cytokinesis in the same
cell line after treatment with 9a, while high-content confocal mi-
Fig. 9. Colchicine-competition assay. Tubulin was pre-incubated with equimolar con-
centration of colchicine, 9a and 9a þ colchicine. Then, tubulin polymerization was
assessed by measuring absorbance every 2 s for a total of 30 min. Statistical analysis
were performed using one-sample T test and expressed as mean SD, *p ꢀ 0,01. Side
scale bars compare tubulin/9a (upper bar) and 9a þ colchicine/colchicine (lower bar)
curves.
croscopy on Hela cells stained for
b-tubulin showed tubulin
disassembly, allowing us to speculate microtubules as potential
target of our lead compound. The binding site of lead compound 9a
was conclusively proved through colchicine-9a competition assay.
Structural superimposition of CBSIs in clinical trial and 9a further
suggested the common binding to the colchicine binding site,
via water molecules [44]. Plinabulin is a vascular disrupting agent
with depolymerising activity, currently in phase I/II for the treat-
ment of several types of cancer [50,51]. The tubulineplinabulin co-
crystallised structure shows that the small molecule binds a deeper
enlightening
teractions. Compound 9a was also combined with an extrusion
pump inhibitor (SS26, 1 M): the co-administration of a microtu-
a new pattern of electrostatic/hydrophobic in-
m
portion of
b-tubulin, where it engages in H-bonds with Glu
b198
bule destabilising agent and an EPI determined a more potent
antiproliferative activity against four drug-resistant cancer cell
lines. All designed and synthesised compounds showed good
properties of predicted pharmacokinetics and druglikeness (as re-
ported in SM). In conclusion, we developed a series of small mol-
ecules as new MDAs; among them, candidate 9a showed the best
overall activity. Future investigations will be focused on exploring
more in depth its pharmacokinetic and pharmacodynamic features,
paying particular attention to the intimate interaction between
tubulin and 9a.
and Val 236, and interacts with Gly
b
b
235 and with Thr 179 via
a
water molecules. Also the crystal structure of plinabulin could be
overlapped with the one of colchicine [44]. Finally, crolibulin is a
chromene-derived inhibitor of tubulin polymerization, able to
affect the cell cycle in G2/M phase [52]. Although it has completed
phase I and phase II trial for anaplastic thyroid in co-administration
with cisplatin [53,54], its therapeutic application includes prostate
adenocarcinoma [55] and gliomas [56]. Unfortunately, its clinical
use is limited due to cardiovascular and neurotoxicity [57,58]. In the
tubulin-crolibulin crystal complex, crolibulin occupies the well-
known colchicine-binding site, forming three H-bonds between
the 2-amino, 7-amino and 3-cyano groups of the chromene moiety
5. Experimental section
of crolibulin and the amino acids Thr
As aforementioned, the lead 9a docks mainly in the
the interface between -dimer and engages four main electro-
static interactions with two alanine, one leucine and one valine
residues. Although some of the crystal structures of the selected
CBSIs overlap with colchicine (Fig. 13), the most favourable
conformational state of 9a binds the colchicine-binding pocket in a
slightly different way (Fig. 12), giving a new pattern of electrostatic/
hydrophobic interactions, and indicating 9a as an interesting
compound to be further investigated.
a
179, Ala
b
248, respectively.
5.1. N-(3,4-difluorophenyl)acetamide (2)
b
-subunit at
a,b
Acetic anhydride (1 ml) was added dropwise to 3,4-
difluoroaniline (1) (1 g, 0.8 ml, 7.8 mmol) in an ice-bath (10 min).
The white solid was filtrated, washed to reach neutral pH, then
dried. C8H7F2NO, MW: 171.14; 80% yield (1.1 g, 6.4 mmol), m.p.
107.3e109.4 ꢁC; Rf 0.10.1H NMR (DMSO‑d6):
d 10.16 (1H, s, NH), 7.77
(1H, ddd, 1JH-F ¼ 13.2 Hz, 2Jmeta ¼ 7.6 Hz, 3Jmeta ¼ 2.4 Hz, H-2), 7.36
2
(1H, dd, 1JH-F ¼ 19.3 Hz, Jmeta ¼ 9.2 Hz, H-5), 7.25 (1H, wm, H-6),
2.04 (3H, s, CH3). 13C NMR (DMSO‑d6):
d 168.51 (C]O), 148.84 (C,
dd, 1JC-F ¼ 242 Hz, 2JC-F ¼ 13 Hz, CeF), 144.99 (C, dd, 1JC-F ¼ 240 Hz,
4. Conclusions
2JC-F ¼ 13 Hz, CeF), 136.31 (CeN), 117.32 (CH), 115.11 (C, d, JC-
¼ 8 Hz, CH-CF), 107.85 (C, d, JC-F ¼ 22 Hz, CH-CF), 23.86 (CH3). ESI-
F
In summary, in this work we developed a series of 50,60-
difluorobenzotriazolacrylonitrile derivatives and we reported
computational and experimental data suggesting them to act as
MDAs. More in depth, we demonstrated, through molecular dock-
ing studies, the different polar (fluorine bonds with cysteine and
leucine residues, H-bonds, e.g. with different leucine residues) and
nonpolar (with the hydrophobic portion of the pocket) interactions
between the representative compound 9b and the tubulin binding
site. The binding mode of 9b was also compared to the best-docked
poses of colchicine and compound 34. E/Z-designed derivatives
(9a-j, 10e, and 11a,b, 13d,j) were obtained through Knoevenagel
condensation as a final step of the synthetic route and properly
characterised through NMR spectroscopy analysis. Then, all com-
pounds were subjected to a preliminary in vitro screening against
the NCI60 cancer cell lines to evaluate their potential anticancer
MS (m/z): calcd for C8H7F2NO 172.05685, found 172.05676 [M þ
H]þ.
5.2. N-(4,5-difluoro-2-nitrophenyl)acetamide (3)
Compound 2 (1 g, 6.0 mmol) was dissolved in concentrated
H2SO4 (H2SO4 conc) in an ice-bath under magnetic stirring. A solu-
tion of KNO3 (1.22 g, 12.0 mmol) in H2SO4 conc (3.26 ml) was added
dropwise. The solution was brought to r.t. and stirred for 2 h. Next,
the reaction was quenched with ice. The obtained precipitate was
filtered off and washed with water to reach neutral pH. Pale-yellow
solid; C8H6F2N2O3, MW: 216.14; 79% yield (1 g, 4.6 mmol); m.p.
101.4e102.5 ꢁC; Rf 0.60.1H NMR (DMSO‑d6):
d 10.33 (1H, s, NH), 8.22
(1H, t, 1JH-F ¼ 8.8 Hz, H-5), 7.83 (1H, dd, 1JH-F ¼ 11.2 Hz, 2JH-F ¼ 7.6 Hz,
H-2), 2.09 (3H, s, CH3). 13C NMR (DMSO‑d6):
d 168.82 (C]O), 151.90
13

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
Fig. 10. Association assay. Percentage of Growth in cells treated with compound 9a alone and in association with EPI SS26. Loess model, 95% Confidence interval.
1
2
1
(C, dd, JC-F ¼ 253 Hz, JC-F ¼ 13 Hz, CeF), 145.04 (C, dd, JC-
d
7.99 (1H, dd, 1JH-F ¼ 11.0 Hz, 2JH-F ¼ 8.4 Hz, H-5), 7.56 (2H, s, NH2),
¼ 246 Hz, 2JC-F ¼ 14 Hz, CeF), 137.45 (CeN), 129.55 (CH), 114.89 (C,
6.96 (1H, dd, JH-F ¼ 12.8 Hz, JH-F ¼ 7.2 Hz, H-2). 13C NMR
1
2
F
1
2
d, JC-F ¼ 10 Hz, CH-CF),113.43 (C, d, JC-F ¼ 22 Hz, CH-CF), 23.39 (CH3).
ESI-MS m/z calcd for C8H6F2N2O3 217.10478, found 217.04193 [M þ
H]þ.
(DMSO‑d6):
d
154.67 (C, dd, JC-F ¼ 126 Hz, JC-F ¼ 15 Hz, CeF),
144.63 (C, d, JC-F ¼ 12 Hz, C), 139.89 (CeF), 125.06 (CeNH2), 110.93
(CH-CF), 105.41 (C, d, JC-F ¼ 21 Hz, CH-CF). ESI-MS m/z calcd for
C6H4F2N2O2 175.03136, found 174.95352 [M þ H]þ.
5.3. 4,5-Difluoro-2-nitroaniline (4)
5.4. 4,5-Difluorobenzene-1,2-diamine (5)
Intermediate 3 (1 g, 4.6 mmol) was dissolved in H2SO4
conc
(10 ml) under reflux (2 h). Next, the reaction was quenched with
ice. A precipitate was obtained, then filtered off and washed with
water to reach pH. Yellow solid; C6H4F2N2O2, MW: 174.11; 61% yield
(1.0 g, 5.7 mmol); m.p. 95.6e105.5 ꢁC; Rf 0.53.1H NMR (DMSO‑d6):
Compound 4 (2 g, 11.5 mol) was dissolved in ethanol (200 ml).
Pd/C (0.20 g, 10% w/w) was added. The hydrogenation was per-
formed with the shaking hydrogenation reactor (open tank,
1.5 h). The Pd/C was filtered off and the solution was
14

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
Fig. 11. 3D structure of the docking pose of 9a in the colchicine-binding site in tubulin. 2D representation of lead compound 9a showing its polar/nonpolar interactions with the
colchicine-binding pocket on tubulin.
Fig. 12. Superimposition of the binding pose of compound 9a (light grey) and the crystal structure of colchicine (yellow, PDB ID: 4O2B) in the colchicine-binding pocket at the interface of
and -tubulin. Tubulin -chains are distinguished by different colours: the -monomer is salmon-coloured, the -one is in brick red tonality.
a-
b
a,
b
b
a
1
concentrated in vacuo to obtain a brown oil, which was purified
via flash chromatography with DE; Brown solid; C6H6F2N2, MW:
144.12; 98% yield (1.62 g, 11.2 mmol); m.p. 111.6e117.2 ꢁC; Rf
(4H, s, 2NH2). 13C NMR (DMSO‑d6):
d
141.14 (2C, dd, JC-
2
¼ 231 Hz, JC-F ¼ 15Hz, CeF), 131.46 (2CeNH2), 102.34-101.71
F
(2C, m, CHeF). ESI-MS m/z calcd for C6H6F2N2 145.05718, found
145.05730 [M þ H]þ.
0.10.1H NMR (DMSO‑d6):
d
7.04 (2H, t, 1JH-F ¼ 10.4 Hz, H-2,5), 5.33
15

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
Fig. 13. Comparison of binding conformation of the best docking pose of compound 9a (light grey, A, with 4O2B apo-form) and the crystal structures of tivantinib (blue marine, B, 5CB4),
plinabulin (dark green, C, 5C8Y) and crolibulin (violet, D, 6JCJ). The a-tubulin portion is coloured in brick red, while b-tubulin is salmon-coloured.
5.5. 5,6-Difluoro-1H-benzo[d] [1e3]triazole (6)
(1C, d, 1JC-F ¼ 11.0 Hz, C]N), 128.79 (1C, d, 1JC-F ¼ 12.0 Hz, C]N),
114.76 (C^N), 106.65 (1C, dd, 1JC-F ¼ 20.4 Hz, CH), 98.53 (1C, d, 1JC-
Compound 5 (1.65 g, 11.0 mmol) was dissolved in HCl 2 N
(114 ml) in an ice-bath. An aqueous solution of NaNO2 (1.63 g,
0.3 mol) was added dropwise. The solution was brought to r.t., then
extracted with DE. Yellow solid; C6H3F2N3, MW: 155.11; 43% yield
(0.8 g, 5.0 mmol); m.p. 179e181 ꢁC; Rf 0.16. Spectra correspondent
¼ 25.0 Hz, CH), 36.01 (CH2). ESI-MS m/z calcd for C6H4F2N4
F
193.03203, found 193.03242 [M ꢂ H]-.
5.7. General procedure for final compounds 9a-j, 10e, 11a,b and
13d,j
to lit [59].
d ESI-MS m/z calcd for C6H4F2N4 154.02113, found
154.02127 [M ꢂ H]-.
Compounds 9a-j and 10e were obtained from 7b and 8a-j (ratio
1:1), 11a,b from 7a and 8a,b (ratio 1:1) and 13d,j from 12a and 8d,j
(ratio 1:1). Triethylamine (TEA, ratio 1:1 þ 20%) was used as a base
for products 9a-c,e-j, 10e and 11a,b, in toluene as a solvent at
110 ꢁC. DIMCARB (ratio 1:1 þ 20%) was used as a catalyst for
products 9d and 13d, in acetonitrile at 60 ꢁC. Piperidine was used as
a base for product 13j, in acetonitrile at 60 ꢁC. Some of the final
compounds were (1) filtrated and crystallised from EtOH, (2) some
were worked-up through liquid chromatography.
5.6. Synthesis and characterisation of the intermediates 2-(5,6-
difluoro-1H-benzo[d] [1e3]triazol-1-yl)acetonitrile (7a) and 2-(5,6-
difluoro-1H-benzo[d] [1e3]triazol-1-yl)acetonitrile (7b)
Compound (6) (2.4 g, 15.3 mmol), KOH (0.95 g, 16.9 mmol) and
chloroacetonitrile (0.9 ml, 1.0 g, 13.8 mmol) were dissolved in
acetonitrile (20 ml), at 80 ꢁC under reflux overnight. The solution
was concentrated in vacuo and purified via flash chromatography
(PS/EA 8/2). These reaction conditions have a good influence on the
selective formation of the 1-isomer compared to the 2-isomer
(ratio 3:1). (7a) Orange solid; C8H4F2N4, MW: 194.14; 9% yield
5.8. (E)-2-(5,6-difluoro-1H-benzo[d] [1e3]triazol-1-yl)-3-(p-tolyl)
acrylonitrile (9a)
(0.26 g, 1.3 mmol); m.p. 97.3e99.1 ꢁC; Rf 0.20.1H NMR (DMSO‑d6):
Work-up procedure (2): PS/EA 9.5/0.5. White solid; C16H10F2N4,
1
d
8.17 (2H, t, JH-F ¼ 8.8 Hz, H-4,7), 6.28 (2H, s, CH2). 13C NMR
MW: 296,28; 45% yield (0.24 g, 0.8 mmol); m.p. 121.4e122.5 ꢁC; Rf
2
1
2
(DMSO‑d6):
d
150.85 (2C, d, 1JC-F ¼ 249.2 Hz, JC-F ¼ 19.3 Hz, CeF),
0.72.1H NMR (DMSO‑d6):
d
8.50 (1H, dd, JH-F ¼ 10.0 Hz, JH-
140.13 (2C, t, 1JC-F ¼ 6.0 Hz, C]N), 114.12 (C^N), 104.30 (2C, m, CH-
CF), 44.0 (CH2). ESI-MS m/z calcd for C8H4F2N4 193.03203, found
193.03246 [M ꢂ H]-. (7b) Yellow solid; C8H4F2N4, MW: 194.14; 33%
yield (0.98 g, 5.0 mmol); m.p. 129.2e131.3 ꢁC; Rf 0.18 (PS/EA 7/3). 1H
¼ 7.2 Hz, H-40), 8.38 (1H, dd, 1JH-F ¼ 9.8 Hz, JH-F ¼ 7.2 Hz, H-70),
2
F
8.28 (1H, s, ¼CH), 7.98 (2H, d, 1JH-H ¼ 8.0 Hz, H-200,600), 7.51 (2H, d,
1JH-H ¼ 8.4 Hz, H-300,500), 2.48 (3H, s, CH3). 13C NMR (DMSO‑d6):
d
151.34 (1C, dd, 1JH-F ¼ 241.2 Hz, 2JH-F ¼ 16.6 Hz, CeF), 148.63 (1C,
2
NMR (DMSO‑d6):
d
8.29 (1H, dd, 1JH-F ¼ 10.0 Hz, 2JH-F ¼ 7.2 Hz, H-4),
dd, 1JH-F ¼ 244.5 Hz, JH-F ¼ 16.2 Hz, CeF), 142.61 (CeCH3), 142.43
(¼CH), 140.6 (1C, d, JH-F ¼ 10.4 Hz, C]N), 129.80 (4C, m, CH), 128.08
(1C, d, JH-F ¼ 12.0 Hz, C]N), 127.68 (C), 114.15 (C^N), 107.04 (1C, d,
1JC-F ¼ 20.8 Hz, CH-CF), 104.71 (C]CH), 99.65 (1C, d, 1JC-F ¼ 24.6 Hz,
8.21 (1H, dd, 1JH-F ¼ 9.6 Hz, 2JH-F ¼ 6.8 Hz, H-7), 6.14 (2H, s, CH2). 13
C
NMR (DMSO‑d6):
d
150.87 (1C, dd, 1JC-F ¼ 240.6 Hz, 2JC-F ¼ 16.8 Hz,
CeF), 148.41 (1C, dd, 1JC-F ¼ 233.5 Hz, 2JC-F ¼ 22.0 Hz, CeF), 140.32
16

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
CH-CF), 21.15 (CH3). ESI-MS m/z calcd for C16H10F2N4 297.09463,
298.09798, found 297.09467, 298.09808 [M þ H]þ.
¼ 10.0 Hz, C]N), 137.61 (¼CH), 137.03 (C), 130.79 (2CH), 127.71
F
(1C, d, JH-F ¼ 12.0 Hz, C]N), 124.13 (2CH), 113.15 (C^N), 109.56 (C]
1
1
CH), 107.23 (1C, d, JC-F ¼ 20.0 Hz, CH-CF), 100.17 (1C, d, JC-
5.9. (E)-2-(5,6-difluoro-1H-benzo[d] [1e3]triazol-1-yl)-3-(4-
methoxyphenyl)acrylonitrile (9b)
¼ 24.0 Hz, CH-CF). ESI-MS m/z calcd for C15H7F2N5O2 328.06406,
F
found 328.06406 [M þ H]þ.
Work-up procedure (1). Brown solid; C16H10F2N4O, MW: 312,28;
5.13. (E)-2-(5,6-difluoro-1H-benzo[d] [1e3]triazol-1-yl)-3-(4-
fluorophenyl)acrylonitrile (9f)
37% yield (0.18 g; 0.6 mmol); m.p. 102.8e103.4 ꢁC; Rf 0.48.1H NMR
(DMSO‑d6):
d
8.43 (1H, dd, 1JH-F ¼ 9.8 Hz, 2JH-F ¼ 7.2 Hz, H-40), 8.29
(1H, dd, 1JH-F ¼ 9.4 Hz, 2JH-F ¼ 7.2 Hz, H-70), 8.18 (1H, s, ¼CH), 8.02
Work-up procedure (1). Light brown solid; C15H7F3N4, MW:
(2H, d, JH-H ¼ 8.8 Hz, H-200,600), 7.21 (2H, d, JH-H ¼ 8.8 Hz, H-300,500),
300,24; 19% yield (0.06 g, 0.2 mmol); m.p. 135.3e136.4 ꢁC; Rf
3.89 (3H, s, OCH3). 13C NMR (DMSO‑d6):
d
162.42 (CeOCH3), 151.30
0.67.1H NMR (DMSO‑d6):
d
8.46 (1H, dd, 1JH-F ¼ 9.8 Hz, 2JH-F ¼ 7.2 Hz,
(1C, dd, 1JH-F ¼ 250.0 Hz, JH-F ¼ 16.7 Hz, CeF), 148.62 (1C, dd, 1JH-
H-40), 8.35 (1H, dd, 1JH-F ¼ 10.0 Hz, JH-F ¼ 6.8 Hz, H-70), 8.28 (1H,
2
2
¼ 244.0 Hz, 2JH-F ¼ 16.3 Hz, CeF), 142.79 (¼CH), 140.52 (1C, d, JH-
s, ¼CH), 8.10 (2H, dd, JH-H ¼ 8.4 Hz e JH-H ¼ 5.2 Hz, H-200,600), 7.51
F
¼ 10.4 Hz, C]N), 131.97 (2CH), 128.20 (1C, d, JH-F ¼ 12.3 Hz, C]N),
(2H, t, H-300,500). 13C NMR (DMSO‑d6):
d
165.11 (C), 162,40 (C), 151.39
F
122.71 (C),114.85 (2CH),114.57 (C^N),106.99 (1C, d, 1JC-F ¼ 20.9 Hz,
CH-CF), 102.57 (C]CH), 99.52 (1C, d, 1JC-F ¼ 24.6 Hz, CH-CF), 55.58
(CH3). ESI-MS m/z calcd for C16H10F2N4O 313.08954, 314.09290,
found 313.08945, 314.09280 [M þ H]þ.
(1C, dd, 1JH-F ¼ 251.3 Hz, JH-F ¼ 20.0 Hz, CeF), 148.65 (1C, dd, 1JH-
2
¼ 245.1 Hz, 2JH-F ¼ 16.1 Hz, CeF), 140.80 (¼CH), 140.64 (1C, d, JH-
F
¼ 10.3 Hz, C]N),132.38 (2C, d, 1JC-F ¼ 9.1 Hz, CH-CF),128.01 (1C, d,
F
JH-F ¼ 12.4 Hz, C]N), 127.16 (C), 116.47 (2C, d, 1JC-F ¼ 22.1 Hz, CH-
1
CF), 113.88 (C^N), 107.09 (1C, d, JC-F ¼ 20.7 Hz, CH-CF), 99.77
1
5.10. (E)-2-(5,6-difluoro-1H-benzo[d] [1e3]triazol-1-yl)-3-(2,3,4-
trimethoxyphenyl)acrylonitrile (9c)
(1C, d, JC-F ¼ 24.9 Hz, CH-CF). ESI-MS m/z calcd for C15H7F3N4
301.06956, 302.07291, found 301.06961, 302.07288 [M þ H]þ.
Work-up procedure (2): PS/EA 9/1. Yellow solid; C18H14F2N4O3,
5.14. (E)-3-(4-chlorophenyl)-2-(5,6-difluoro-1H-benzo[d] [1e3]
triazol-1-yl)acrylonitrile (9g)
MW: 344,32; 21% yield (0.1 g, 0.3 mmol); m.p. 119.3e120.6 ꢁC; Rf
1
2
0.43.1H NMR (DMSO‑d6):
d
8.44 (1H, dd, JH-F ¼ 9.6 Hz, JH-
2
¼ 7.2 Hz, H-40), 8.21 (1H, dd, 1JH-F ¼ 9.6 Hz, JH-F ¼ 6.8 Hz, H-70),
Work-up procedure (2): PS/EA 9/1. Light brown solid;
C15H7ClF2N4, MW: 316.70; 34% yield (0.27 g, 0.9 mmol); m.p.
F
8.11 (1H, s, C]CH), 7.94 (1H, d, JH-H ¼ 8.8 Hz, H-600), 7.13 (1H, d, JH-
1
¼ 8.8 Hz, H-500), 3.94 (3H, s, OCH3), 3.90 (3H, s, OCH3), 3.81 (3H, s,
147.3e149.7 ꢁC; Rf 0.50.1H NMR (DMSO‑d6):
d
8.46 (1H, dd, JH-
H
OCH3). 13C NMR (DMSO‑d6):
d
156.99 (CeOCH3), 153.14 (2CeOCH3),
¼ 9.6 Hz, JH-F ¼ 7.2 Hz, H-40), 8.36 (1H, dd, JH-F ¼ 9.6 Hz, JH-
2
1
2
F
141.52 (C), 140.32 (CF), 138.04 (CH), 128.44 (CF), 123.56 (CH), 116.76
¼ 6.8 Hz, H-70), 8.28 (1H, s, ¼CH), 8.02 (2H, d, JH-H ¼ 8.8 Hz, H-
F
(2C), 114.30 (C^N), 108.47 (CH), 107.06 (1C, d, 1JC-F ¼ 21.6 Hz, CH-
200,600), 7.73 (2H, d, JH-H ¼ 8.8 Hz, H-300,500). 13C NMR (DMSO‑d6):
1
CF), 104.09 (C]CH), 99.20 (1C, d, JC-F ¼ 24.7 Hz, CH-CF), 61.82
d
151.39 (1C, dd, 1JH-F ¼ 249.0 Hz, 2JH-F ¼ 16.0 Hz, CeF), 148.69 (1C,
1
2
(OCH3), 60.43 (OCH3), 56.19 (OCH3). ESI-MS m/z calcd for
dd, JH-F ¼ 245.0 Hz, JH-F ¼ 16.0 Hz, CeF), 140.69 (1C, d, JH-
C
[M
18H14F2N4O3 373.11067, 374.11403, found 373.11124, 374.11432
¼ 11.0 Hz, C]N), 140.13 (¼CH), 136.49 (C), 131.38 (2CH), 129.46
F
þ
H]þ; calcd 395.092962, 396.09597, found 395.09296,
(CeCl), 129.33 (2CH), 127.91 (1C, d, JH-F ¼ 12.0 Hz, C]N), 113.69
396.09625 [M þ Na]þ; calcd 411.06656, found 411.06683 [M þ K]þ.
(C^N), 107.11 (1C, d, 1JC-F ¼ 20.0 Hz, CH-CF), 106.63 (C]CH), 99.85
1
(1C, d, JC-F ¼ 24.0 Hz, CH-CF). ESI-MS m/z calcd for C15H7ClF2N4
5.11. (E)-2-(5,6-difluoro-1H-benzo[d] [1e3]triazol-1-yl)-3-(4-
hydroxyphenyl)acrylonitrile (9d)
317.04001, found 317.04178 [M þ H]þ.
5.15. (E)-3-(4-bromophenyl)-2-(5,6-difluoro-1H-benzo[d] [1e3]
triazol-1-yl)acrylonitrile (9h)
Work-up procedure (1). Yellow solid; C15H8F2N5O, MW: 298.25;
29% yield (0.12 g, 0.40 mmol); m.p. 178.4e180,6 ꢁC; Rf 0.18.1H NMR
1
2
(DMSO‑d6):
d
10.59 (1H, s, OH), 8.41 (1H, dd, JH-F ¼ 9.6 Hz, JH-
Work-up procedure (2): PS/DE 8/2. White powder; C15H7BrF2N4,
¼ 7.2 Hz, H-40), 8.25 (1H, dd, 1JH-F ¼ 9.0 Hz, JH-F ¼ 6.8 Hz, H-70),
MW: 361.15; 47% yield (0.35 g, 1.0 mmol); m.p. 150.0e151.6 ꢁC; Rf
2
F
8.09 (1H, s, ¼CH), 7.92 (2H, d, JH-H ¼ 8.0 Hz, H-200,600), 7.0 (2H, d, JH-
0.3.1H NMR (DMSO‑d6):
d
8.46 (1H, dd, 1JH-F ¼ 9.8 Hz, 2JH-F ¼ 7.6 Hz,
1
2
¼ 8.0 Hz, H-300,500). 13C NMR (DMSO‑d6):
d
161.54 (CeOH), 151.28
H-40), 8.37 (1H, dd, JH-F ¼ 9.6 Hz, JH-F ¼ 6.8 Hz, H-70), 8.26 (1H,
H
(1C, dd, 1JH-F ¼ 249.0 Hz, JH-F ¼ 17.0 Hz, CeF), 148.59 (1C, dd, 1JH-
s, ¼CH), 7.95 (2H, d, JH-H ¼ 8.8 Hz, H-200,600), 7.87 (2H, d, JH-
2
¼ 244.5 Hz, 2JH-F ¼ 17.0 Hz, CeF), 143.62 (CH]C), 140.46 (1C, d, JH-
¼ 8.4 Hz, H-300,500). 13C NMR (DMSO‑d6):
d
151.41 (1C, dd, JH-
1
F
H
¼ 10.0 Hz, C]N), 132.31 (2CH), 128.30 (1C, d, JH-F ¼ 12.0 Hz, C]N),
¼ 250.0 Hz, 2JH-F ¼ 16.4 Hz, CeF), 148.67 (1C, dd, 1JH-F ¼ 245.0 Hz,
F
F
121.11 (C), 116.23 (2CH), 114.81 (C^N), 106.94 (1C, d, 1JC-F ¼ 21.0 Hz,
2JH-F ¼ 16.2 Hz, CeF), 140.69 (1C, d, JH-F ¼ 10.2 Hz, C]N), 140.13
(¼CH), 132.20 (2CH), 131.43 (2CH), 129.75 (C), 127.85 (1C, d, JH-
CH-CF), 101.17 (C]C), 99.39 (1C, d, 1JC-F ¼ 25.0 Hz, CH-CF). ESI-MS
m/z calcd for
C
15H8F2N5O 297.05824, 298.06160, 299.06495,
¼ 12.4 Hz, C]N), 125.43 (CeBr), 113.59 (C^N), 107.01 (1C, d, 1JC-
F
found 297.05939, 298.06268, 299.06622 [M ꢂ H]-.
¼ 19.4 Hz, CH-CF), 106.67 (C]C), 99.72 (1C, d, 1JC-F ¼ 24.8 Hz, CH-
F
CF). ESI-MS m/z [M þ H]þ.
5.12. (E)-2-(5,6-difluoro-1H-benzo[d] [1e3]triazol-1-yl)-3-(4-
nitrophenyl)acrylonitrile (9e)
5.16. (E)-3-(benzo[d] [1,3]dioxol-4-yl)-2-(5,6-difluoro-1H-benzo[d]
[1e3]triazol-1-yl)acrylonitrile (9i)
Work-up procedure (2): PS/EA 8/2. Yellow powder;
C
15H7F2N5O2, MW: 327.25; 24% yield (0.11 g, 0.3 mmol); m.p.
Work-up procedure (1). Yellow solid; C6H8F2N4O2, MW: 326,26;
151.1e152.6 ꢁC; Rf 0.46.1H NMR (DMSO‑d6):
d
8.51-8.42 (2H, m, H-
28% yield (0.13 g, 0.4 mmol); m.p. 164.6e165.2 ꢁC; Rf 0,41.1H NMR
40,70), 8.48 (2H, d, JH-H ¼ 8.8 Hz, H-300,500), 8.43 (1H, s, ¼CH), 8.23
(DMSO‑d6):
d
8.44 (1H, dd, 1JH-F ¼ 9.6 Hz, 2JH-F ¼ 7.2 Hz, H-40), 8.28
(2H, d, JH-H ¼ 8.4 Hz, H-200,600). 13C NMR (DMSO‑d6):
d
151.47 (1C, dd,
(1H, dd, 1JH-F ¼ 9.6 Hz, 2JH-F ¼ 6.8 Hz, H-70), 8.15 (1H, s, ¼CH), 7.62
(1H, d, JH-H ¼ 1.6 Hz, H-200), 7.56 (1H, dd, JH-H ¼ 8.4 Hz e J ¼ 1.6 Hz, H-
600), 7.20 (1H, d, JH-H ¼ 8.4 Hz, H-500), 6.21 (2H, s, CH2). 13C NMR
1JH-F ¼ 249.5 Hz, JH-F ¼ 17.0 Hz, CeF), 148.70 (1C, dd, JH-
2
1
¼ 245.0 Hz, 2JH-F ¼ 16.0 Hz, CeF),148.48 (CeNO2),140.84 (1C, d, JH-
F
17

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
(DMSO‑d6):
d
151.76 (1C, dd, 1JH-F ¼ 249.6 Hz, 2JH-F ¼ 16.6 Hz, CeF),
151.27 (2C, dd, 1JC-F ¼ 250.5 Hz, 2JC-F ¼ 20.0 Hz, CeF), 140.37 (2C, t,
1JC-F ¼ 6.0 Hz, C]N), 138.07 (C]CH), 132.23 (2CH), 122.32 (C),
114.97 (2CH), 113.43 (C^N), 108.68 (C]CH), 104.24 (2CH, m, CH-
CF), 55.64 (CH3). ESI-MS m/z calcd for C16H10F2N4O 313.08954,
found 313.08975 [M þ H]þ.
151.29 (CeO), 149.10 (1C, dd, 1JH-F ¼ 245.0 Hz, 2JH-F ¼ 16.2 Hz, CeF),
148.42 (CeO), 146.92 (¼CH), 142.92 (CH), 141.06 (1C, d, JH-
¼ 10.4 Hz, C]N), 128.59 (1C, d, JH-F ¼ 12.4 Hz, C]N), 127.52 (CH),
F
124.70 (C), 114.86 (C^N), 109.49 (CH), 108.44 (CH), 107.46 (2CH, d,
1JC-F ¼ 28.0 Hz, CH-CF), 103.68 (C]CH), 102.76 (OeCH2eO), 99.96
(2CH, d, 1JC-F ¼ 24.8 Hz, CH-CF). ESI-MS m/z calcd for C16H8F2N4O2
327.06881, 328.07216, found 327.06906, 328.07233 [M þ H]þ.
5.21. (E)-2-(1H-benzo[d] [1e3]triazol-1-yl)-3-(4-hydroxyphenyl)
acrylonitrile (13d)
5.17. (E)-2-(5,6-difluoro-1H-benzo[d] [1e3]triazol-1-yl)-3-
(isoquinolin-5-yl)acrylonitrile (9j)
Work-up procedure (2): CHCl3/CH3OH 9.9/0.1. Yellow solid;
15H10N4O, MW: 262.27; 28% yield (0.2 g, 0.8 mmol); m.p.
C
1
115.4e116.5 ꢁC; Rf 0.38.1H NMR (DMSO‑d6):
d 8.22 (2H, d, JH-
Work-up procedure (1). Light brown solid; C18H9F2N5, MW:
¼ 8.4 Hz, H-4’), 8.12 (1H, s, ¼CH), 7.98 (1H, d, JH-H ¼ 8.4 Hz, H-70),
H
333.30; 33% yield (0.2 g, 0.6 mmol); m.p. 109.4e111.5 ꢁC; Rf 0.10.1H
7.93 (2H, d, JH-H ¼ 8.4 Hz, H-200,600), 7.73 (1H, d, JH-H ¼ 7.8 Hz, H-60),
NMR (DMSO‑d6):
d
9.47 (1H, s, H-100), 8.91 (1H, s, ¼CH), 8.64 (1H, d,
7.00 (2H, d, JH-H ¼ 8.4 Hz, H-300,500). 13C NMR (DMSO‑d6):
d 161.86
JH-H ¼ 6 Hz, H-300), 8.49 (1H, d, JH-H ¼ 7.2 Hz, H-800), 8.44 (2H, m, H-
(CeOH), 145.71 (C]N), 143.31 (CH), 132.61 (2CH), 132.29 (C]N),
129.63 (CH), 125.68 (CH), 121.75 (CeCH), 120.35 (CH), 116.73 (2CH),
115.39 (C^N), 111.30 (CH), 102.06 (C). ESI-MS m/z calcd for
40,70), 8.39 (1H, d, JH-H ¼ 8.4 Hz, H-600), 8.12 (1H, d, JH-H ¼ 6.0 Hz, H-
400), 7.93 (1H, t, JH-H ¼ 7.8 Hz, H-700). 13C NMR (DMSO‑d6):
d 153.05
(CH), 151.43 (1C, dd, 1JH-F ¼ 249.0 Hz, JH-F ¼ 16.0 Hz, CeF), 148.68
(1C, dd, 1JH-F ¼ 245.0 Hz, 2JH-F ¼ 16.0 Hz, CeF), 144.03 (CH), 140.73
(1C, d, Jc-F ¼ 11.0 Hz, C]N), 138.08 (CH), 133.64 (C]CH), 131.45
(CH), 131.23 (CH), 128.18 (1C, d, Jc-F ¼ 13.0 Hz, C]N), 128.06 (C),
127.50 (C), 127.17 (¼CH), 117.25 (CH), 113.42 (C^N), 109.82 (C),
C
15H10N4O 263.09274, found 263.53491 [M ꢂ H]-.
2
5.22. (E)-2-(1H-benzo[d] [1e3]triazol-1-yl)-3-(isoquinolin-5-yl)
acrylonitrile (13j)
1
1
107.07 (2CH, d, JC-F ¼ 21.0 Hz, CH-CF), 100.16 (2CH, d, JC-
Work-up procedure (1). Brick red-brown powder. C18H11N5,
¼ 25.0 Hz, CH-CF). ESI-MS m/z calcd for C18H9F2N5 334.08988,
MW: 297.32; 35% yield (0.23 g, 0.8 mmol); m.p. 131.4e133.3 ꢁC; Rf
F
335.09323, 336.09659, found 334.09045, 335.09366, 336.09689
0.04.1H NMR (DMSO‑d6):
d
9.47 (1H, s, H-100), 8.92 (1H, s, ¼CH), 8.63
[M þ H]þ.
(1H, d, JH-H ¼ 5.6 Hz, H-300), 8.50 (1H, d, JH-H ¼ 7.2 Hz, H-40), 8.39
(1H, d, JH-H ¼ 8.0 Hz, H-400), 8.28 (1H, d, JH-H ¼ 8.4 Hz, H-800), 8.18
(1H, d, JH-H ¼ 8.4 Hz, H-70), 8.14 (1H, d, JH-H ¼ 6 Hz, H-600), 7.93 (1H, t,
JH-H ¼ 8.0 Hz, H-60), 7.79 (1H, t, JH-H ¼ 8.0 Hz, H-700), 7.62 (1H, t, JH-
5.18. (Z)-2-(5,6-difluoro-1H-benzo[d] [1e3]triazol-1-yl)-3-(4-
nitrophenyl)acrylonitrile (10e)
¼ 8.0 Hz, H-50).13C NMR (DMSO‑d6):
d 153.05 (CH]N isoq), 145.49
H
Work-up procedure (1). Red solid; C15H7F2N5O2, MW: 327.25;
(C]N), 144.06 (CH]N isoq), 137.02 (CH), 133.65 (C]CH), 131.62
(C]N), 131.41 (CH), 131.09 (CH), 129.44 (CH), 128.09 (C), 127.60 (C),
127.20 (CH), 125.53 (CH), 119.99 (¼CH), 117.20 (CH), 113.53 (CN),
111.51 (CH), 110.34 (C). ESI-MS m/z calcd for C15H10N4O, found
263.06778, 264.07114 [M þ H]þ.
26% yield (0.12 g, 0.4 mmol); m.p. 179.3e181.2 ꢁC; Rf 0,68.1H NMR
(DMSO‑d6):
d
8.53 (1H, dd, 1JH-F ¼ 10.0 Hz, 2JH-F ¼ 7.2 Hz, H-40), 8.32
1
2
(2H, d, JH-H ¼ 8.8 Hz, H-300,500), 7.74 (1H, dd, JH-F ¼ 9.6 Hz, JH-
¼ 6.8 Hz, H-70), 7.69 (2H, d, JH-H ¼ 8.8 Hz, H-200,600), 7.27 (1H,
F
s, ¼CH). 13C NMR (DMSO‑d6):
d
151.34 (1C, dd, 1JC-F ¼ 250.5 Hz, 2JC-
¼ 16.0 Hz, CeF), 149.45 (CeNO2), 148.70 (1C, dd, 1JC-F ¼ 227.5 Hz,
Fundings
F
2JC-F ¼ 16.0 Hz, CeF), 147.63 (C), 140.59 (1C, d, JH-F ¼ 10.0 Hz, C]N),
137.20 (C]CH), 129.19 (2CH), 124.18 (1C, d, JH-F ¼ 17.0 Hz, C]N),
124.26 (2CH), 114.83 (C^N), 107.43 (2CH, d, 1JC-F ¼ 20.0 Hz, CH-CF),
99.77 (1C, d, 1JC-F ¼ 25.0 Hz, CH-CF), 99.64 (¼CH). ESI-MS m/z calcd
for C18H9F2N5 311.06467, found 311.25558 [M þ H]þ.
This work was supported by “Regione Autonoma della Sarde-
gna” (Sardinia, Italy), by “Legge Regionale 7 agosto 2007:CRP1_574,
22/41 del 2017” [grant number RASSR01499].
Author contributions
5.19. (E)-2-(5,6-difluoro-2H-benzo[d] [1e3]triazol-2-yl)-3-(p-tolyl)
acrylonitrile (11a)
Conceptualisation, F.R., R.I., S.P. and A.C.; formal analysis, F.R.,
R.I., S.S., V.B., L.S., M.A.S. and M.L.; investigation, F.R., R.I., V.B., L.S.,
M.A.S. and M.L.; resources, A.C.; data curation, F.R., R.I., S.S. V.B., L.S.,
M.A.S., M.L., L.B. and A.C.; writing/original draft preparation, F.R.,
R.I.; writing/review and editing, all authors; visualization, F.R., R.I.,
S.S. and A.C.; supervision, L.B., A.C., funding acquisition, L.B., A.C. All
authors have read and agreed to the published version of the
manuscript.
Work-up procedure (1). Yellow solid; C16H10F2N4, MW: 296,28;
42% yield (0.16 g, 0.5 mmol); m.p. 158.1e160.5 ꢁC; Rf 0,83.1H NMR
(DMSO‑d6):
d
8.66 (1H, s, ¼CH), 8.23 (2H, t, 1JH-H ¼ 8.8 Hz, H-40,70),
7.96 (2H, d, JH-H ¼ 8.0 Hz, H-200,600), 7.44 (2H, d, JH-H ¼ 8.0 Hz, H-
300,500), 2.42 (3H, s, CH3). 13C NMR (DMSO‑d6):
d
151.39 (2C, dd, 1JC-
¼ 250.5 Hz, 2JC-F ¼ 19.0 Hz, CeF),142.87 (2C, t, 1JC-F ¼ 7.0 Hz, C]N),
F
138.24 (C]CH), 130.00 (2CH), 129.96 (2CH), 127.27 (C), 113.08
(C^N), 110.45 (C]CH),104.33 (2CH, m, CH-CF), 21.23 (CH3). ESI-MS
m/z calcd for C16H10F2N4 297.09463, found 297.09482 [M þ H]þ.
Declaration of competing interest
The authors declare that they have no known competing
financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
5.20. (E)-2-(5,6-difluoro-2H-benzo[d] [1e3]triazol-2-yl)-3-(4-
methoxyphenyl)acrylonitrile (11b)
Work-up procedure (1). Brown solid; C16H10F2N4O, MW: 312,28;
Acknowledgments
39% yield (0.17 g, mmol); m.p. 148.6e149.3 ꢁC; Rf 0,43.1H NMR
(DMSO‑d6):
d
8.62 (1H, s, ¼CH), 8.21 (2H, t, 1JH-H ¼ 8.8 Hz, H-40,70),
We thank the National Cancer Institute, Developmental Thera-
peutics Program, NCI (Bethesda, MD, USA; https://dtp.cancer.gov)
as the source of the data from in vitro anti-cancer screening. We
8.06 (2H, d, JH-H ¼ 8.8 Hz, H-200,600), 7.18 (2H, d, JH-H ¼ 8.8 Hz, H-
300,500), 3.88 (3H, s, OCH3). 13C NMR (DMSO‑d6):
d 162.52 (CeOCH3),
18

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
[17] D.M. Patterson, M. Zweifel, M.R. Middleton, P.M. Price, L.K. Folkes,
M.R.L. Stratford, P. Ross, S. Halford, J. Peters, J. Balkissoon, D.J. Chaplin,
A.R. Padhani, G.J.S. Rustin, Phase I clinical and pharmacokinetic evaluation of
the vascular-disrupting agent OXi4503 in patients with advanced solid tu-
mors, Clin. Canc. Res. 18 (2012) 1415e1425, https://doi.org/10.1158/1078-
0432.CCR-11-2414.
[18] P. Sanna, A. Carta, L. Gherardini, Synthesis and antimycobacterial activity of 3-
aryl- , 3-cyclohexyl- and 3, heteroaryl- 57 (2002) 79e87.
[19] A. Carta, P. Sanna, M. Palomba, L. Vargiu, M. La Colla, R. Loddo, Synthesis and
antiproliferative activity of 3-aryl-2-(1H-benzotriazol-1-yl)acrylonitriles. Part
III, Eur. J. Med. Chem. 37 (2003) 891e900, https://doi.org/10.1016/S0223-
5234(02)01411-3.
acknowledge CISUPdCentre for Instrumentation Sharingd Oper-
etta CLS Facility -University of Pisa. We thank Kitos Biotech for
€
in vitro 9a-EPI combination assay. We acknowledge Schrodinger®
for the short term free license of Maestro package. F.R. acknowl-
ꢀ
edges MIUR (Ministero dell’Istruzione, dell’Universita e della
Ricerca) for the PhD grant, and the Erasmusþ program. F.R. thanks
€
Prof. Goran Widmalm and Alessandro Ruda, for the support during
his PhD Erasmusþ at the Stockholm University, Sweden.
[20] A. Carta, M. Palomba, G. Boatto, B. Busonera, M. Murreddu, R. Loddo, Synthesis
and antiproliferative activity of 3-aryl-2-[1H(2H)-benzotriazol- 1(2)-yl]acry-
lonitriles variously substituted: Part 4, Farmaco 59 (2004) 637e644, https://
doi.org/10.1016/j.farmac.2004.03.004.
[21] A. Carta, I. Briguglio, S. Piras, G. Boatto, P. La Colla, R. Loddo, M. Tolomeo,
S. Grimaudo, A. Di Cristina, R.M. Pipitone, E. Laurini, M.S. Paneni, P. Posocco,
M. Fermeglia, S. Pricl, 3-Aryl-2-[1H-benzotriazol-1-yl]acrylonitriles: a novel
class of potent tubulin inhibitors, Eur. J. Med. Chem. 46 (2011) 4151e4167,
https://doi.org/10.1016/j.ejmech.2011.06.018.
Appendix A. Supplementary data
Supplementary data related to this article can be found at
https://doi.org/10.1016/j.ejmech.2021.113590.
Appendix A
[22] I. Briguglio, E. Laurini, M.A. Pirisi, S. Piras, P. Corona, M. Fermeglia, S. Pricl,
A. Carta, Triazolopyridinyl-acrylonitrile derivatives as antimicrotubule agents:
synthesis, in vitro and in silico characterization of antiproliferative activity,
inhibition of tubulin polymerization and binding thermodynamics, Eur. J.
Supplementary Material reports deeper NMR studies, NCI60
tables, EPI SS26 activity validation, further association assay results
and ADME prediction tables.
Med.
Chem.
141
(2017)
460e472,
https://doi.org/10.1016/
j.ejmech.2017.09.065.
References
[23] P. Sanna, A. Carta, M.E.R. Nikookar, Synthesis and antitubercular activity of 3-
aryl substituted-2-(1H(2H) benzotriazol-1(2)-yl)acrylonitriles, Eur. J. Med.
Chem. 35 (2000) 535e543, https://doi.org/10.1016/S0223-5234(00)00144-6.
[24] H. van deW, D.K.W.D.A. Smith, Methods and principles in medicinal chemis-
try, in: Pharmacokinetics and Metabolism in Drug Design, vol. 31, Wiley-VCH,
Weinheim, 2006.
[1] F. Bray, J. Ferlay, I. Soerjomataram, R.L. Siegel, L.A. Torre, A. Jemal, Global
cancer statistics 2018: GLOBOCAN estimates of incidence and mortality
worldwide for 36 cancers in 185 countries, CA. Cancer J. Clin. 68 (2018)
394e424, https://doi.org/10.3322/caac.21492.
[25] B.E. Smart, Fluorine substituent effects (on bioactivity), J. Fluor. Chem. 109
(2001) 3e11, https://doi.org/10.1016/S0022-1139(01)00375-X.
[26] S. Purser, P.R. Moore, S. Swallow, V. Gouverneur, Fluorine in medicinal
chemistry, Chem. Soc. Rev. 37 (2008) 320e330, https://doi.org/10.1039/
b610213c.
[2] A. Damasceno, Noncommunicable Disease, 2016, https://doi.org/10.1002/
9781119097136.part5.
[3] G. Berube, How to utilize academic research efforts in cancer drug discovery,
Expert Opin, Drug Discov 14 (2019) 331e334, https://doi.org/10.1080/
17460441.2019.1582637.
[4] Y. Imoto, Y. Yoshida, F. Yagisawa, H. Kuroiwa, T. Kuroiwa, The cell cycle,
including the mitotic cycle and organelle division cycles, as revealed by
cytological observations, J. Electron. Microsc. 60 (2011) 117e136, https://
doi.org/10.1093/jmicro/dfr034.
ꢁ
ꢁ
[27] O.A.J. Trott, O, AutoDock Vina: improving the speed and accuracy of docking
with
a new scoring function, efficient optimization, and multithreading,
J. Comput. Chem. 31 (2010) 455e461, https://doi.org/10.1002/jcc.21334.
[28] Z. Wang, H. Sun, X. Yao, D. Li, L. Xu, Y. Li, S. Tian, T. Hou, Comprehensive
evaluation of ten docking programs on a diverse set of protein-ligand com-
plexes: the prediction accuracy of sampling power and scoring power, Phys.
Chem. Chem. Phys. 18 (2016) 12964e12975, https://doi.org/10.1039/
c6cp01555g.
[29] H.M. Berman, T. Battistuz, T.N. Bhat, W.F. Bluhm, P.E. Bourne, K. Burkhardt,
Z. Feng, G.L. Gilliland, L. Iype, S. Jain, P. Fagan, J. Marvin, D. Padilla,
V. Ravichandran, B. Schneider, N. Thanki, H. Weissig, J.D. Westbrook,
C. Zardecki, The protein data bank, Acta Crystallogr. Sect. D Biol. Crystallogr.
58 (2002) 899e907, https://doi.org/10.1107/S0907444902003451.
[30] A.J. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell,
AutoDock4 Olson, AutoDockTools4, Automated docking with selective re-
ceptor flexibility, J. Comput. Chem. 16 (2009) 2785e2791.
[5] S. Florian, T.J. Mitchison, Anti-microtubule drugs, Methods Mol. Biol. 1413
(2016) 403e421, https://doi.org/10.1007/978-1-4939-3542-0_25.
ꢁ
[6] G. Chandrasekaran, P. Tatrai, F. Gergely, Hitting the brakes: targeting micro-
tubule motors in cancer, Br. J. Canc. 113 (2015) 693e698, https://doi.org/
10.1038/bjc.2015.264.
[7] E. Mukhtar, V.M. Adhami, H. Mukhtar, Targeting microtubules by natural
agents for cancer therapy, Mol. Canc. Therapeut. 13 (2014) 275e284, https://
doi.org/10.1158/1535-7163.MCT-13-0791.
[8] A. Jordan, J.A. Hadfield, N.J. Lawrence, A.T. McGown, Tubulin as a target for
anticancer drugs: agents which interact with the mitotic spindle, Med. Res.
Rev. (1998), https://doi.org/10.1002/(SICI)1098-1128(199807)18:4<259::AID-
MED3>3.0.CO;2-U.
[31] K. Stott, J. Keeler, Q.N. Van, A.J. Shaka, One-dimensional NOE experiments
using pulsed field gradients, J. Magn. Reson. 125 (1997) 302e324, https://
doi.org/10.1006/jmre.1997.1110.
[9] R. Kaul, A.L. Risinger, S.L. Mooberry, Microtubule-targeting drugs: more than
antimitotics,
J.
Nat.
Prod.
(2019),
https://doi.org/10.1021/
acs.jnatprod.9b00105.
[32] Y. Binev, M.M.B. Marques, J. Aires-de-Sousa, Prediction of1H NMR coupling
constants with associative neural networks trained for chemical shifts,
J. Chem. Inf. Model. 47 (2007) 2089e2097, https://doi.org/10.1021/ci700172n.
[33] M. Lai, V. La Rocca, R. Amato, G. Freer, M. Costa, P.G. Spezia, P. Quaranta,
G. Lombardo, D. Piomelli, M. Pistello, Ablation of acid ceramidase impairs
autophagy and mitochondria activity in melanoma cells, Int. J. Mol. Sci. 22
(2021) 3247, https://doi.org/10.3390/ijms22063247.
[10] Y.N. Cao, L.L. Zheng, D. Wang, X.X. Liang, F. Gao, X.L. Zhou, Recent advances in
microtubule-stabilizing agents, Eur. J. Med. Chem. (2018), https://doi.org/
10.1016/j.ejmech.2017.11.062.
[11] S.K. Coulup, G.I. Georg, Revisiting microtubule targeting agents: a-Tubulin and
the pironetin binding site as unexplored targets for cancer therapeutics,
Bioorg. Med. Chem. Lett 29 (2019) 1865e1873, https://doi.org/10.1016/
j.bmcl.2019.05.042.
[34] A. Daina, O. Michielin, V. Zoete, SwissADME: a free web tool to evaluate
pharmacokinetics, drug-likeness and medicinal chemistry friendliness of
small molecules, Sci. Rep. 7 (2017) 42717, https://doi.org/10.1038/srep42717.
[12] A. Massarotti, A. Coluccia, R. Silvestri, G. Sorba, A. Brancale, The tubulin
colchicine domain:
a molecular modeling perspective, ChemMedChem 7
(2012) 33e42, https://doi.org/10.1002/cmdc.201100361.
[35] A. Daina, O. Michielin, V. Zoete, iLOGP:
a simple, robust, and efficient
[13] Y. Lu, J. Chen, M. Xiao, W. Li, D.D. Miller, An overview of tubulin inhibitors that
interact with the colchicine binding site, Pharm. Res. (N. Y.) 29 (2012)
2943e2971, https://doi.org/10.1007/s11095-012-0828-z.
[14] A.M. Tsimberidou, W. Akerley, M.C. Schabel, D.S. Hong, C. Uehara, A. Chhabra,
T. Warren, G.G. Mather, B.A. Evans, D.P. Woodland, E.A. Swabb, R. Kurzrock,
Phase I clinical trial of MPC-6827 (Azixa), a microtubule destabilizing agent, in
patients with advanced cancer, Mol. Canc. Therapeut. 9 (2010) 3410e3419,
https://doi.org/10.1158/1535-7163.MCT-10-0516.
description of n -Octanol/Water partition coefficient for drug design using the
GB/SA approach, J. Chem. Inf. Model. 54 (2014) 3284e3301, https://doi.org/
10.1021/ci500467k.
[36] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Experimental and
computational approaches to estimate solubility and permeability in drug
discovery and development settings, Adv. Drug Deliv. Rev. (2012), https://
doi.org/10.1016/j.addr.2012.09.019.
€
[37] The PyMOL Molecular Graphics System, Version 2.0 Schrodinger, LLC., (n.d.).
[15] J.Y. Bruce, J. Eickhoff, R. Pili, T. Logan, M. Carducci, J. Arnott, A. Treston,
€
[38] Maestro, Version 10.2, Schrodinger, LLC, New York, NY, 2015 n.d.
G. Wilding, G. Liu,
A phase II study of 2-methoxyestradiol nanocrystal
[39] L. Ra, S. Mb, LigPlotþ: multiple ligand-protein interaction diagrams for drug
colloidal dispersion alone and in combination with sunitinib malate in pa-
tients with metastatic renal cell carcinoma progressing on sunitinib malate,
Invest. N. Drugs 30 (2012) 794e802, https://doi.org/10.1007/s10637-010-
9618-9.
discovery, J. Chem. Inf. Model. 51 (2011) 2778e2786.
[40] K. Eskandari, M. Lesani, Does fluorine participate in halogen bonding? Chem.
Eur J. 21 (2015) 4739e4746, https://doi.org/10.1002/chem.201405054.
[41] C.S. See, M. Kitagawa, P.J. Liao, K.H. Lee, J. Wong, S.H. Lee, B.W. Dymock,
Discovery of the cancer cell selective dual acting anti-cancer agent (Z)-2-(1H-
indol-3-yl)-3-(isoquinolin-5-yl)acrylonitrile (A131), Eur. J. Med. Chem. 156
(2018) 344e367, https://doi.org/10.1016/j.ejmech.2018.07.011.
[16] P. Liu, Y. Qin, L. Wu, S. Yang, N. Li, H. Wang, H. Xu, K. Sun, S. Zhang, X. Han,
Y. Sun, Y. Shi, A phase i clinical trial assessing the safety and tolerability of
combretastatin A4 phosphate injections, Anti Canc. Drugs 25 (2014) 462e471,
https://doi.org/10.1097/CAD.0000000000000070.
19

F. Riu, L. Sanna, R. Ibba et al.
European Journal of Medicinal Chemistry 222 (2021) 113590
[42] L. Sanna, R. Piredda, I. Marchesi, V. Bordoni, S.V. Forcales, D.F. Calvisi,
L. Bagella, Verteporfin exhibits anti-proliferative activity in embryonal and
alveolar rhabdomyosarcoma cell lines, Chem. Biol. Interact. 312 (2019)
108813, https://doi.org/10.1016/j.cbi.2019.108813.
[50] M.M. Mita, M.A. Spear, L.K. Yee, A.C. Mita, E.I. Heath, K.P. Papadopoulos,
K.C. Federico, S.D. Reich, O. Romero, L. Malburg, M.J. Pilat, G.K. Lloyd,
S.T.C. Neuteboom, G. Cropp, E. Ashton, P.M. LoRusso, Phase 1 first-in-human
trial of the vascular disrupting agent plinabulin (NPI-2358) in patients with
solid tumors or lymphomas, Clin. Canc. Res. (2010), https://doi.org/10.1158/
1078-0432.CCR-10-1096.
[51] M. Millward, P. Mainwaring, A. Mita, K. Federico, G.K. Lloyd, N. Reddinger,
S. Nawrocki, M. Mita, M.A. Spear, Phase 1 study of the novel vascular dis-
rupting agent plinabulin (NPI-2358) and docetaxel, Invest. N. Drugs (2012),
https://doi.org/10.1007/s10637-011-9642-4.
[52] R.O. Carlson, New tubulin targeting agents currently in clinical development,
Expet Opin. Invest. Drugs (2008), https://doi.org/10.1517/13543784.17.5.707.
[53] J.J. Field, A. Kanakkanthara, J.H. Miller, Microtubule-targeting agents are
clinically successful due to both mitotic and interphase impairment of
microtubule function, Bioorg. Med. Chem. (2015), https://doi.org/10.1016/
j.bmc.2014.02.035.
[43] J. Barretina, G. Caponigro, N. Stransky, K. Venkatesan, A.A. Margolin, S. Kim,
ꢁ
C.J. Wilson, J. Lehar, G.V. Kryukov, D. Sonkin, A. Reddy, M. Liu, L. Murray,
ꢁ
M.F. Berger, J.E. Monahan, P. Morais, J. Meltzer, A. Korejwa, J. Jane-Valbuena,
F.A. Mapa, J. Thibault, E. Bric-Furlong, P. Raman, A. Shipway, I.H. Engels,
J. Cheng, G.K. Yu, J. Yu, P. Aspesi, M. De Silva, K. Jagtap, M.D. Jones, L. Wang,
C. Hatton, E. Palescandolo, S. Gupta, S. Mahan, C. Sougnez, R.C. Onofrio,
T. Liefeld, L. MacConaill, W. Winckler, M. Reich, N. Li, J.P. Mesirov, S.B. Gabriel,
G. Getz, K. Ardlie, V. Chan, V.E. Myer, B.L. Weber, J. Porter, M. Warmuth,
P. Finan, J.L. Harris, M. Meyerson, T.R. Golub, M.P. Morrissey, W.R. Sellers,
R. Schlegel, L.A. Garraway, The Cancer Cell Line Encyclopedia enables pre-
dictive modelling of anticancer drug sensitivity, Nature 483 (2012) 603e607,
https://doi.org/10.1038/nature11003.
[44] Y. Wang, H. Zhang, B. Gigant, Y. Yu, Y. Wu, X. Chen, Q. Lai, Z. Yang, Q. Chen,
J. Yang, Structures of a diverse set of colchicine binding site inhibitors in
complex with tubulin provide a rationale for drug discovery, FEBS J. (2016),
https://doi.org/10.1111/febs.13555.
[45] Z. Zhang, C. Wang, L. Ma, X. Jiang, C. Wu, Y. Wang, Y. Jiang, W. Zheng, Y. Yang,
Y. Ma, J. Yang, Molecular mechanism of crolibulin in complex with tubulin
provides a rationale for drug design, Biochem. Biophys. Res. Commun. (2019),
https://doi.org/10.1016/j.bbrc.2019.02.064.
[54] T.F. Greene, S. Wang, L.M. Greene, S.M. Nathwani, J.K. Pollock, A.M. Malebari,
T. McCabe, B. Twamley, N.M. OBoyle, D.M. Zisterer, M.J. Meegan, Synthesis and
biochemical evaluation of 3-Phenoxy-1,4-diarylazetidin-2-ones as tubulin-
targeting antitumor agents, J. Med. Chem. (2016), https://doi.org/10.1021/
acs.jmedchem.5b01086.
[55] J. Kalmuk, M. Folaron, J. Buchinger, R. Pili, M. Seshadri, Multimodal Imaging
Guided Preclinical Trials of Vascular Targeting in Prostate Cancer, Oncotarget,
2015, https://doi.org/10.18632/oncotarget.4463.
[46] E.R. Caremoli, R. Labianca, Tivantinib: critical review with a focus on hepa-
tocellular carcinoma, Expet Opin. Invest. Drugs 23 (2014) 1563e1574, https://
doi.org/10.1517/13543784.2014.949339.
[56] M. Folaron, M. Seshadri, Bioluminescence and MR imaging of the safety and
efficacy of vascular disruption in gliomas, Mol. Imag. Biol. (2016), https://
doi.org/10.1007/s11307-016-0963-8.
[47] Y.K. Kang, K. Muro, M.H. Ryu, H. Yasui, T. Nishina, B.Y. Ryoo, Y. Kamiya,
S. Akinaga, N. Boku, A phase II trial of a selective c-Met inhibitor tivantinib
(ARQ 197) monotherapy as a second- or third-line therapy in the patients
with metastatic gastric cancer, Invest. N. Drugs (2014), https://doi.org/
10.1007/s10637-013-0057-2.
[57] W.L. Read, P. Rosen, P. Lee, S. Anthony, R. Korn, N. Raghunand, B. Tseng,
J. Whisnant, D. Von Hoff, R. Tibes, Pharmacokinetic and pharmacodynamic
results of a 4-hr IV administration phase I study with EPC2407, a novel
vascular disrupting agent, J. Clin. Oncol. (2009), https://doi.org/10.1200/
jco.2009.27.15_suppl.3569.
[48] J.W. Goldman, I. Laux, F. Chai, R.E. Savage, D. Ferrari, E.G. Garmey, R.G. Just,
L.S. Rosen, Phase 1 Dose-Escalation Trial Evaluating the Combination of the
Selective MET (Mesenchymal-epithelial Transition Factor) Inhibitor Tivantinib
(ARQ 197) Plus Erlotinib, Cancer, 2012, https://doi.org/10.1002/cncr.27575.
[49] A. Aoyama, R. Katayama, T. Oh-hara, S. Sato, Y. Okuno, N. Fujita, Tivantinib
(ARQ 197) exhibits antitumor activity by directly interacting with tubulin and
overcomes ABC transporter-mediated drug resistance, Mol. Canc. Therapeut.
(2014), https://doi.org/10.1158/1535-7163.MCT-14-0462.
[58] I.M. Subbiah, D.J. Lenihan, A.M. Tsimberidou, Cardiovascular toxicity profiles
of vascular-disrupting agents, Oncol. (2011), https://doi.org/10.1634/theon-
cologist.2010-0432.
[59] R.J. Faggyas, N.L. Sloan, N. Buijs, A. Sutherland, Synthesis of structurally
diverse benzotriazoles via rapid diazotization and intramolecular cyclization
of 1,2-aryldiamines, Eur. J. Org Chem. 2019 (2019) 5344e5353, https://
doi.org/10.1002/ejoc.201900463.
20