1,4,5,6,7,8-Hexahydroquinolines and 5,6,7,8-tetrahydronaphthalenes: A new class of antitumor agents targeting the colchicine binding site of tubulin

Article abstract of DOI:10.1016/j.bioorg.2020.103831

New series of 2-amino-1,4,5,6,7,8-hexahydroquinoline-3-carbonitriles 3a,b and 2-amino-5,6,7,8-tetrahydronaphthalene-1,3-dicarbonitriles 4a-h were synthesized and evaluated for their antitumor activity. In vitro antitumor screening of the new members again

Full text of DOI:10.1016/j.bioorg.2020.103831

Journal Pre-proofs
1,4,5,6,7,8-Hexahydroquinolines and 5,6,7,8-tetrahydronaphthalenes: A new
class of antitumor agents targeting the colchicine binding site of tubulin
Mennatallah A. Shaheen, Ali A. El-Emam, Nadia S. El-Gohary
PII:
S0045-2068(20)30510-1
DOI:
https://doi.org/10.1016/j.bioorg.2020.103831
YBIOO 103831
Reference:
To appear in:
Bioorganic Chemistry
Received Date:
Revised Date:
Accepted Date:
3 March 2020
3 April 2020
5 April 2020
Please cite this article as: M.A. Shaheen, A.A. El-Emam, N.S. El-Gohary, 1,4,5,6,7,8-Hexahydroquinolines and
5,6,7,8-tetrahydronaphthalenes: A new class of antitumor agents targeting the colchicine binding site of tubulin,
Bioorganic Chemistry (2020), doi: https://doi.org/10.1016/j.bioorg.2020.103831
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2020 Elsevier Inc. All rights reserved.

1,4,5,6,7,8-Hexahydroquinolines
and
5,6,7,8-
tetrahydronaphthalenes: A new class of antitumor agents targeting
the colchicine binding site of tubulin
Mennatallah A. Shaheena,b, Ali A. El-Emama, Nadia S. El-Goharya,*
aDepartment of Medicinal Chemistry, Faculty of Pharmacy, Mansoura University, Mansoura 35516, Egypt
bDepartment of Pharmaceutical Chemistry, Faculty of Pharmacy, Horus University, New Damietta, Egypt
Corresponding author: dr.nadiaelgohary@yahoo.com, nadiaelgohary2005@gmail.com
ABSTRACT: New series of 2-amino-1,4,5,6,7,8-hexahydroquinoline-3-carbonitriles 3a,b and
2-amino-5,6,7,8-tetrahydronaphthalene-1,3-dicarbonitriles 4a-h were synthesised and evaluated
for their antitumor activity. In vitro antitumor evaluation of the new members against HepG2,
HCT-116 and MCF-7 cancer cells showed that the tetrahydronaphthalene-1,3-dicarbonitrile 4c
has the highest potency against the three tested cancer cells (IC50 = 6.02, 8.45 and 6.28 µM,
respectively). In addition, 4c displayed low cytotoxicity against WI38 and WISH normal cells
(IC50 = 51.78 and 42.36 µM, respectively), and it might be utilized as a potent and selective
antitumor agent. Compound 4c was further studied for its effect on tubulin polymerization,
different phases of cell cycle, apoptosis and caspases 3/9 levels. Results revealed that analog 4c
has tubulin polymerization inhibitory effect with IC50 value of 3.64 μM. Additionally, it
induced significant accumulation of the tested cancer cells in G2/M phase, and induced cell
death primarily via apoptosis. Besides, it showed evident increase in caspase-3 level in HepG2
and HCT-116, and caspase-9 level in MCF-7 cells. Further, docking study proved the exact fit
of 4c into the colchicine binding site of tubulin.
Keywords:
Hexahydroquinolines;
Tetrahydronaphthalenes;
Antitumor;
Tubulin
polymerization inhibition; Cell cycle analysis; Apoptosis; Computational studies
- 1 -

1. Introduction
Cancer is a dreadful disease affecting human in developed as well as developing countries.
It is known that one-quarter of adults’ mortality is due to cancer [1]. The most common
treatment options include surgery [2], chemotherapy [3] and radiotherapy [4]. Tubulin
polymerization inhibition is an attractive therapy in cancer treatment. Tubulin consists of two
subunits: -tubulin and -tubulin which are composed of 450 and 455 amino acids,
respectively [5]. -Tubulin dimers could be grouped to form protofilaments in a head-to-tail
fashion [6]. Thirteen protofilaments interact side to side forming cylinder which is polar and
hollow, and they are called microtubules (MTs). Cell division, cell migration and intracellular
transport were regulated by MTs [7]. Therefore, disruption of the function of MTs will lead to
mitotic arrest and tumor cell death by apoptosis.
1.1. Rational drug design
One of the most fundamental pockets in tubulin is the colchicine binding site that has been
focused on in the design of tubulin polymerization inhibitors. Nguyen and coworkers [8]
selected fifteen colchicine binding site inihbitors (CBSIs) based on their harmonious topology
and binding modes as well as their ability to occupy the same chemical space. They chosed
tubulin-DAMA-colchicine crystal structure as a template for their study, and results revealed
that CBSIs interact with tubulin primarily through different hydrogen bonding interactions.
They concluded that the fifteen CBSIs showed hydrogen bonding interactions with thiol of Cys
239. Out of the fifteen CBSIs, eleven members formed hydrogen bonding interactions with
Val 179 nitrogen atom. Additionally, hydrogen bonds formed with the nitrogen atoms of Asp
249, Ala 248 and Leu 250 were detected in eight members, whereas one hydrogen bonding
interaction was formed with the oxygen atom of Thr 179 in four members. Accordingly, they
constructed a common pharmacophore model based on the binding modes of these compounds.
The consistent structural features and repeated ligand-tubulin interactions were the basis for the
pharmacophoric points of CBSIs. Therefore, the different classes of CBSIs were suggested
based on the seven-point pharmacophore comprised of one hydrogen bond donor (D1), three
hydrogen bond acceptors (A1, A2 and A3), one planar group (R1) and two hydrophobic
- 2 -

centers (H1 and H2) (Fig. 1). Based on these formerly reported facts, we can conclude that
compounds exhibiting these seven pharmacophoric features will be considered as promising
tubulin polymerization inhibitors. Taking colchicine as a lead for synthesis of the new
members, its structure consists of three rings: ring A, ring B (linker) and ring C. Structure–
activity relationship study revealed that rings A and C of colchicine are prerequisites for its
high binding affinity to colchicine binding site of tubulin. The changes of the linker region
influence the cytotoxic activity of the most reported CBSIs [8]. Fig. 2 illustrates the
pharmacophoric points of colchicine as a lead for synthesis of new CBSIs.
Lots of CBSIs of natural and synthetic origins have been developed, and they are
structurally in common [9–11]. Literature survey proved that naphthalene derivatives A-D have
promising antitumor activity through tubulin polymerization inhibition (Fig. 3) [12-17]. Also,
quinoline derivatives E-G were proved to have excellent antitumor activity through tubulin
polymerization inhibition (Fig. 3) [18-23]. Further, the 4,6-diarylpyridine-3-carbonitrile H [24],
and 7-aryl-5-indolylimidazo[1,2-a]pyridine-8-carbonitriles I [25] and J [26] were reported as
antitumors and CBSIs (Fig. 4). Recently, 4-aryl-4H-chromene derivatives K [27] and L, M
[28] were identified as anticancer agents that exert their potency via tubulin polymerization
inhibition (Fig. 4). In addition, the 4-aryl-4H-chromenes were reported as apoptosis inducers
through elevation of caspases levels [29]. Crolibulin (Figs. 4 and 5) is a 4-aryl-4H-chromene
derivative, and it is a promising antitumor agent that inhibits tubulin polymerization into MTs
via binding to the colchicine binding site of tubulin, leading to cell cycle arrest in G2/M phase,
apoptosis and suppression of tumor cell proliferation [27]. Structure-activity relationship study
of 4-aryl-4H-chromenes revealed that presence of cyano group at 3-position played a
significant role on their antitumor activity, and introduction of a phenyl moiety at 4-position
considerably improved the potency [27-29]. In the current research, 4H-chromene ring in
crolibulin was replaced by hexahydroquinoline or tetrahydronaphthalene, and the cyano group
at 3-position as well as the aromatic ring at 4-position were retained in order to prepare a new
series of 4-arylhexahydroquinolines 3a,b (Fig. 5); in addition, an attempt to improve the
binding interaction with the target protein was achieved by the introduction of an extra cyano
group at 1-position to prepare a new series of 4-aryltetrahydronaphthalene-1,3-dicarbonitriles
4a-h with expected antitumor and tubulin polymerization inhibitory activities (Fig. 5).
- 3 -

2. Results and discussion
2.1. Chemistry
As described in Scheme 1, the arylidenemalononitriles 1a,b and 2a-h [30-36] were
prepared via condensation of the appropriate aromatic aldehydes with malononitrile in ethanol
[36]. The arylidenemalononitriles 1a,b were allowed to react with cyclohexanone and excess
ammonium acetate via heating in ethanol to yield the expected 2-amino-4-aryl-1,4,5,6,7,8-
hexahydroquinoline-3-carbonitriles 3a and 3b in 85 and 72% yields, respectively, this reaction
was previously described by El-Ashmawy et al. [37]. The structures of the hexahydroquinoline-
1
3-carbonitrile derivatives 3a,b were confirmed based on IR, H NMR, 13C NMR and
1
Electrospray Ionization mass spectral (ESI-MS) data. H NMR spectra of both compounds
13
showed characteristic singlets at  5.10 and 5.04 ppm corresponding to C4-H, and C NMR
spectra showed signals at  33.2 and 32.7 ppm corresponding to C-4. Additionally, ESI-MS
spectra of compounds 3a,b were in agreement with their exact mass values, for example, mass
spectrum of 3a showed (M++H) peak at 320.0 corresponding to its exact mass 319.06. Unlike
the reaction of arylidenemalononitriles 1a,b with cyclohexanone and ammonium acetate, the
reaction of arylidenemalononitriles 2a-h with cyclohexanone and ammonium acetate under the
same reaction conditions did not yield the expected hexahydroquinoline-3-carbonitrile analogs,
and instead they yielded the unexpected 2-amino-4-aryl-5,6,7,8-tetrahydronaphthalene-1,3-
dicarbonitrile derivatives 4a-h in 66-90% yields. A similar observation was previously reported
by Al-Youbi et al. [38] and Asiri et al. [39,40]. The possible mechanism of the formation of
5,6,7,8-tetrahydronaphthalene-1,3-dicarbonitriles 4a-h is described in Scheme 2 following the
previously reported mechanism [41].
1
13
The structural assignments of compounds 4a-h were relied on their H NMR, C NMR
1
and ESI-MS spectral data. H NMR spectra revealed the absence of the singlet characteristic
for C4-H in all derivatives. In addition, ESI-MS spectra of compounds 4a-h were in agreement
with their exact mass values, for example, mass spectrum of 4d showed (M++H) peak at 310.2
corresponding to its exact mass 309.11. The structures of the tetrahydronaphthalene-1,3-
dicarbonitrile derivatives 4a-h were further confirmed by single crystal X-ray crystallography
of compounds 4b (Fig. 6A) and 4d (Fig. 6B).
- 4 -

2.2. Biology
2.2.1. Cytotoxicity testing
2.2.1.1. In vitro antitumor testing against cancer cells
MTT assay was followed to evaluate the antitumor activity of the new derivatives 3a,b and
4a-h against hepatocellular carcinoma (HepG2), colorectal carcinoma (HCT-116) and breast
carcinoma (MCF-7) [42,43]. Concentrations of compounds that induce 50% inhibition of cell
viability (IC50, µM) were determined and compared to doxorubicin and colchicine as reference
agents (Table 1).
Results showed that 4c is the most potent member against all chosen cancer cells (IC50 =
6.02-8.45 µM). On the other hand, 4b exhibited strong activity against MCF-7 cancer cells
(IC50 = 10.15 µM), and moderate activity against HepG2 and HCT-116 cells (IC50 = 26.80 and
12.40 µM, respectively). Further, 4a and 4g were proved to have moderate activity against all
tested cell lines (IC50 = 11.39-30.93 µM). On the opposite side, 3b was proved to be
moderately active against HepG2 and HCT-116 cell lines (IC50 = 32.59 and 36.56 M,
respectively). The remaining members displayed low or no activity against the three chosen
cancer cells.
Structure-activity relationship
Concerning the hexahydroquinolines 3a,b, presence of 2,4-difluorophenyl moiety at 4-
position of hexahydroquinoline ring resulted in moderate activity against HepG2 and HCT-116
cells, and weak activity against MCF-7 cells (compound 3b). Replacement of 2,4-
difluorophenyl moiety in 3b with 2,3-dichlorophenyl counterpart resulted in weak activity
against the three tested cancer cells (compound 3a).
With respect to the activity of tetrahydronaphthalenes 4a-h, the 4-(2-chloro-6-
fluorophenyl)tetrahydronaphthalene derivative 4c showed the highest activity against the three
chosen cancer cells. Replacement of 2-chloro-6-fluorophenyl moiety in 4c with 2,6-
dichlorophenyl counterpart gave compound 4a with moderate activity against the three cell
lines, whereas its replacement with 3,4-dichlorophenyl moiety gave compound 4b with strong
- 5 -

activity against MCF-7 cells, and moderate activity against HepG2 and HCT-116 cells. In
addition, replacement of 2-chloro-6-fluorophenyl moiety in 4c with 2,6-difluorophenyl
counterpart resulted in weak or abolished activity against the selected cancer cells (derivative
4d), and this might be attributed to the additional chlorine-binding interaction of 4c with target
protein (chlorine is a good halogen-bond donor [44]), whereas fluorine in 4d is generally considered to
be a poor halogen-bond donor [44], and hence 4d is expected to show no halogen-binding interaction
with the target protein. Likewise, replacing 2-chloro-6-fluorophenyl moiety in 4c with biphenyl
resulted in weak activity against the selected cancer cells (derivative 4e). Further, presence of
naphthalen-2-yl moiety at 4-position of tetrahydronaphthalene ring resulted in moderate
activity against the three tested cancer cells (compound 4g). Replacing naphthalen-2-yl in 4g
with anthracen-9-yl led to decreased activity against the three tested cancer cells (compound
4h), whereas its replacement with [1,3]benzodioxol-4-yl resulted in weak activity against
HepG2, and abolished activity against the selected colorectal and breast cancer cells
(compound 4f versus 4g). Comparing the activity of 3a,b and 4a-h, it is obvious that
compounds 4a-c and 4g,h are more potent antitumor agents than 3a,b, and this might be
attributed to the presence of an extra cyano group at 1-position of tetrahydronaphthalene
nucleus in 4a-c and 4g,h that was proved to be involved in additional hydrogen bonding
interaction with the colchicine binding site of tubulin.
2.2.1.2. In vitro cytotoxicity testing of 4c against normal cells
Cytotoxicity of 4c against WI38 lung fibroblast and WISH amnion epithelial normal cells
was evaluated following MTT assay [42,43]. IC50 values (µM) were determined, and results are
compared to doxorubicin. Compound 4c showed lower cytotoxicity toward WI38 and WISH
normal cells (IC50 = 51.78 and 42.36 µM, respectively) than doxorubicin (IC50 = 6.72 and 3.80
µM, respectively), and it was evidenced to be safe against the tested normal cells at their
cytotoxic concentrations against the tested cancer cells.
- 6 -

2.2.2. Mechanistic studies
2.2.2.1. Screening of tubulin polymerization inhibitory activity
Crolibulin was proved to exert its antitumor activity through inhibition of tubulin
polymerization [27]. Due to structural similarity between the most active antitumor member in
the current research (compound 4c) and crolibulin (Fig. 5), and as an attempt to explore the
possible mode of action of 4c, its in vitro tubulin polymerization inhibitory activity was
assessed [45,46]. Results indicated that 4c is a good tubulin polymerization inhibitor (IC50 =
3.64 μM) in comparison to colchicine (IC50 = 1.28 μM) (Fig. 7). In addition, 4c induced %
tubulin polymerization inhibitory activity of 54.11% at concentration of 10 μM in comparison
to colchicine with % inhibition value of 61.25% at same concentration.
2.2.2.2. Flow cytometry analysis of cell cycle
The structural similarities between the most active antitumor compound 4c and crolibulin,
as well as the promising inhibition of tubulin dynamics led us to hypothesize that it exerts its
antiproliferative properties through inducing G2/M arrest in cancer cells similarly to tubulin
polymerization inhibitors that were reported to induce G2/M arrest in cancer cells [46,47].
Therefore, compound 4c was studied for its ability to arrest cell cycle adopting the propidium
iodide (PI) flow cytometry assay kit [48-52].
Compound 4c was incubated with HepG2, HCT-116 and MCF-7 cancer cells for 24 h, then
changes within the cell cycle were observed. The cell cycle histograms of the three cancer cells
treated with compound 4c (at IC50 of the corresponding cell line) are illustrated in Fig. 8. The
observations of cell cycle arrest induced by 4c in the three cancer cell lines are shown in Fig. 8
and compared to that of the untreated cells. Regarding the results of cell cycle arrest in HepG2,
compound 4c increased the percentage of cells in pre-G1 phase (from 1.29% to 24.03%) and
G2/M phase (from 5.62% to 35.72%), and decreased the percentage of cells in S phase (from
36.12% to 24.82%) and Go/G1 phase (from 58.26% to 39.46%) (Fig. 8). Referring to the
results of cell cycle arrest in HCT-116, compound 4c increased the percentage of cells in pre-
G1 phase from (2.19% to 26.59%) and G2/M phase (from 21.20% to 45.91%), and decreased
the percentage of cells in S phase (from 31.52% to 25.66%) and Go/G1 phase (from 47.28% to
- 7 -

28.43%) (Fig. 8). Regarding the results of cell cycle arrest in MCF-7, compound 4c increased
the percentage of cells in pre-G1 phase (from 1.85% to 22.42%), G2/M phase (from 16.69% to
35.77%) and S phase (from 29.49% to 31.42%), and decreased the percentage of cells in Go/G1
phase (from 53.82% to 32.81%) (Fig. 8). From the above results, we can conclude that 4c
demonstrated significant increase in the cell population of the three tested cancer cell lines in
G2/M phase, and induced cellular apoptosis in pre-G1 phase (Fig. 8), and accordingly stopped
the mitotic cycle. This action came in agreement with the well known fact that tubulin
polymerization inhibitors prevent the cancer cells from beginning mitosis [46,47].
2.2.2.3. Cell apoptosis
Cancer is one of the diseases where too little apoptosis occurs, resulting in development
of malignant cells that will not die. Apoptosis is well-known as a popular target in cancer
treatment [53]. The mechanism of apoptosis is complex and involves many pathways [53].
Several studies evidenced that tubulin polymerization inhibitors are capable of inducing
cellular apoptosis [27,46,47]. Therefore, compound 4c was studied for its capability to induce
apoptosis in the three cancer cell lines applying the annexin V/PI double staining flow
cytometry assay [48-52]. The dot plot flow cytometry assay data of the cells pigmented with PI
and annexin fluorescein isothiocyanate (V-FITC) are shown in Fig. 9. The annexin V/PI double
staining disclosed that after 24 h of exposure, compound 4c (at IC50 of the corresponding cell
line) induced early and late apoptosis in the three cancer cell lines in comparison to the
untreated cells. The proportion of total apoptosis in HCT-116 cells treated with compound 4c
was more than that of MCF-7 and HepG2 cells (Fig. 9). On contrary, compound 4c induced
weak necrosis in the three treated cancer cells (Fig. 9) confirming that cell death induced by
compound 4c occurs primarily through apoptosis.
2.2.2.4. Caspases 3/9 activation assay
Caspases represent a family of cysteine proteases that are present in the cytosol, and they
are correlated with cell apoptosis through the mitochondrial pathway [54]. The most common
members in the caspases family are caspases 3/9, the activation of which triggers cell apoptosis
[55]. Therefore, drugs that elevate the levels of caspases are considered to be potent apoptosis
- 8 -

inducers [56]. Caspase-3 is expressed in HepG2 and HCT-116 cell lines [57,58], whereas
MCF-7 cancer cells do not express this type of caspase [59,60], and instead they express
caspase-9 [61,62]. Consequently, compound 4c (at concentration of 10 μM) was assessed for its
ability to elevate the levels of caspase-3 in HepG2 and HCT-116 [63], and caspase-9 in MCF-7
[63]. Results (Fig. 10A) revealed that 4c increased the level of caspase-3 in HepG2 and HCT-
116 by about 10 and 5 folds, respectively compared to the corresponding untreated cells. On
the other hand, 4c increased the level of caspase-9 in MCF-7 cancer cells by about 8 folds
compared to the corresponding untreated cells (Fig. 10B).
3. Computational aided studies
3.3. Analysis of 4c-tubulin interaction
Docking study was applied to identify the possible binding interactions between the active
candidate 4c and tubulin, and hence to explore the key structural features affecting its tubulin
polymerization inhibitory activity. Docking study will provide some insights into the additional
structural alterations and development of new more potent and selective tubulin polymerization
inhibitors. In the present research, a molecular modeling study relied on tubulin-(DAMA-
colchicine) complex crystal structure (PDB code: 4O2B) [64] was executed applying
“molecular operating environment (MOE) version 2019.01” Chemical Computing Group Inc.
software [65] in order to attain an idea about the binding mode of the efficacious member 4c to
the colchicine binding site of tubulin which is present at the interface between the  and 
chains of tubulin.
Primarily, the molecular docking setup was validated by implementing the re-docking of
colchicine close to the colchicine binding site. The re-docking validation step reproduced the
experimental binding mode of the co-crystallized ligand quite efficiently indicating the
suitability of the used setup for the intended docking study, and this is confirmed by the small
root-mean-square deviation (RMSD) of 0.66 Å (<2 Å) between the docked pose and the co-
crystallized ligand, and by the capability of the docking poses to reproduce the key interactions
achieved by the co-crystallized ligand with the hot spot amino acid (Val181) in the active site
(Fig. 11).
- 9 -

Analysis of docking results (Fig. 12) revealed that compound 4c showed a consistent
binding mode (docking energy score (ΔG) = -6.05 kcal/mol) like that of the co-crystallized ligand
(DAMA-colchicine) (docking energy score (ΔG) = -8.09 kcal/mol), including two hydrogen
bonding interactions of the cyano group in compound 4c with Val181 and Ala180 residues of
tubulin, and one hydrogen bonding interaction of amino group in 4c with Met259 of
tubulin. Furthermore, 4c displayed an extra binding interaction with tubulin through the
electrophilic chlorine atom in 4c with nucleophilic oxygen atom in Lys352 (Fig. 12).
Also, superimposition of the co-crystallized colchicine ligand and the docking pose of
compound 4c at the colchicine binding site showed the overlap of the aryl rings in 4c with rings
A and C of colchicine (Fig. 13). To further elucidate the structural similarity between compound
4c and crolibulin, flexible alignment was carried out and they showed high structural similarity
(Fig. 14). On the other hand, docking study revealed that the weak antitumor activity of 4d
(carrying 2,6-difluorophenyl substituent at 4-position of tetrahydronaphthalene nucleus)
compared to the potent analog 4c (carrying 2-chloro-6-fluorophenyl substituent at 4-position of
tetrahydronaphthalene nucleus) might be due to the absence of halogen-bonding interaction in 4d
(Fig. 15), since fluorine atom in 4d is considered to be a poor halogen-bond donor [44], whereas chlorine
atom in 4c is considered to be a good halogen-bond donor [44]. The obtained results proved that 4c
can bind to colchicine binding site to a good extent, and it occupied the active site similarly to the
co-crystallized colchicine (Fig. 13), and hence it might exert its antitumor activity via tubulin
ploymerization inhibition.
4. Conclusion
Results of antitumor assay assured that compound 4c is the most active member against
HepG2, HCT-116 and MCF-7 cancer cells (IC50 = 6.02, 8.45 and 6.28 µM, respectively).
Furthermore, it displayed low cytotoxicity against WI38 and WISH normal cells (IC50 = 51.78
and 42.36 µM, respectively), and it might be utilized as a potent and selective antitumor agent.
Results of tubulin polymerization inhibition assay proved that 4c has good tubulin
polymerization inhibitory activity. Cell cycle analysis demonstrated that 4c induces cell cycle
arrest in G2/M phase in the three cancer cell lines and this result comes in agreement with the
fact that tubulin polymerization inhibitors strongly arrest cell cycle in G2/M phase. In addition,
- 10 -

the death of cancer cells induced by 4c was proved to occur primarily through apoptosis.
Referring to results of caspases 3/9 assay, compound 4c elevated caspase-3 (in HepG2 and HCT-
116 cells) and caspase-9 levels (in MCF-7 cells) confirming that apoptosis might have been
occured via increasing caspases 3/9 levels. Further, docking study proved the exact fit of 4c into
the colchicine binding site of tubulin. The previously presented results proved that rational
design of the new hexahydroquinolines and tetrahydronaphthalenes with expected antitumor and
tubulin polymerization inhibitory activities was adequate, and the most active member in the
current study (compound 4c) will be subjected to extra structural modifications in order to obtain
new members with higher potency.
5. Experimental
5.1. Chemistry
Melting points (°C) were determined by Stuart melting point (SMP30) apparatus. IR spectral
1
analysis were carried out in KBr disc on Unicam SP 1000 IR spectrometer (ν in cm1). H and
13C NMR spectral analysis were carried out on Bruker Avance III HD FT-high resolution (400
MHz), (500 MHz) and (700 MHz). Chemical shifts are expressed in  ppm with reference to
tetramethylsilane. Electrospray Ionization mass spectra (ESI-MS) were recorded on Agilent 6410
Triple Quad tandem mass spectrometer at 4.0 and 3.5 kV for positive and negative ions,
respectively. Elemental analyses (% C, H, N) were determined, and they were found to be in
agreement with the expected structures within ±0.4% of the calculated values. Reaction times
were controlled by TLC plates (Silica gel 60 F254, E. Merck), and spots were visualized by UV.
Hexane/ethyl acetate (9:1) was used for elution. The arylidenemalononitriles 1a,b and 2a-h were
synthesized following the reported procedure [36].
5.1.1. Synthesis of 2-amino-4-(substituted phenyl)-1,4,5,6,7,8-hexahydroquinoline-3-
carbonitriles 3a,b and 2-amino-4-aryl-5,6,7,8-tetrahydronaphthalene-1,3-dicarbonitriles
4a-h
A mixture of the appropriate arylidenemalononitrile 1a,b or 2a-h (0.005 mol),
cyclohexanone (0.49 g, 0.005 mol) and ammonium acetate (0.77 g, 0.01 mol, excess) was
- 11 -

refluxed in absolute ethanol for 6-12 h. The precipitated solid was filtered and crystallized from
ethanol to produce compounds 3a,b and 4a-h, respectively.
5.1.1.1. 2-Amino-4-(2,3-dichlorophenyl)-1,4,5,6,7,8-hexahydroquinoline-3-carbonitrile (3a)
1
Yield 85%, m.p. 285-287 °C. IR: 3443, 3300 (NH2); 3195 (NH); 2213 (C≡N). H NMR (500
MHz, CDCl3): δ 1.68–1.74 (m, 2H, CH2), 1.85 (t, J = 6.0 Hz, 2H, CH2), 2.24-2.30 (m, 2H, CH2),
2.86-2.91 (m, 2H, CH2), 5.10 (s, 1H, C4-H), 5.49 (s, 1H, NH, D2O-exchang.), 7.03-7.10 (dd, J =
13
2.0, 1.5 Hz, 1H, Ar-H), 7.35 (t, J = 8 Hz, 1H, Ar-H), 7.57-7.59 (m, 1H, Ar-H). C NMR: δ
21.7, 22.2, 25.6, 29.5, 33.3, 89.5, 114.8, 127.8, 128.0, 129.0, 130.7, 131.1, 134.0, 147.4, 161.0,
162.0. ESI-MS, m/z: 320.0 [M+H]+. Anal. Calcd. (Found) for C16H15Cl2N3 (319.06): C, 60.01
(60.37); H, 4.72 (4.43); N, 13.12 (13.46).
5.1.1.2. 2-Amino-4-(2,4-difluorophenyl)-1,4,5,6,7,8-hexahydroquinoline-3-carbonitrile (3b)
1
Yield 72%, m.p. 148-149 °C. IR: 3421, 3354 (NH2); 3307 (NH); 2216 (C≡N). H NMR (500
MHz, CDCl3): δ 1.65-1.75 (m, 2H, CH2), 1.80-1.86 (m, 2H, CH2), 2.20-2.41 (m, 2H, CH2), 2.80-
2.98 (m, 2H, CH2), 5.04 (s, 1H, C4-H), 5.29 (s, 1H, NH, D2O-exchang.), 6.94-7.03 (m, 2H, Ar-
13
H), 7.17-7.23 (m, 1H, Ar-H). C NMR: δ 22.2, 22.4, 25.6, 29.5, 32.7, 90.9, 104.5-104.9 (m),
112.1 (d, J = 21.5 Hz), 115.1 (d, J = 16.3 Hz), 115.7, 119.6-119.8 (dd, J = 3.7, 3.6 Hz), 127.2,
131.2-131.6 (m), 158.3 (t, J = 12.2 Hz), 160.3 (t, J = 11.9 Hz), 162.8 (t, J = 13.3 Hz). ESI-MS,
m/z: 288.1 [M+H]+. Anal. Calcd. (Found) for C16H15F2N3 (287.12): C, 66.89 (66.52); H, 5.26
(5.59); N, 14.63 (14.31).
5.1.1.3. 2-Amino-4-(2,6-dichlorophenyl)-5,6,7,8-tetrahydronaphthalene-1,3-dicarbonitrile (4a)
Yield 83%, m.p. 220-222 °C. IR: 3419, 3304 (NH2); 2212 (2C≡N). 1H NMR (400 MHz, CDCl3):
δ 1.71-1.74 (m, 2H, CH2), 1.84-1.88 (m, 2H, CH2), 2.28-2.38 (m, 2H, CH2), 2.84 (t, J = 8.1 Hz,
2H, CH2), 5.10 (s, 2H, NH2, D2O-exchang.), 7.12-7.14 (dd, J = 2.0, 2.0 Hz, 1H, Ar-H), 7.38 (d, J
13
= 2.0 Hz, 1H, Ar-H), 7.58 (d, J = 8.2 Hz, 1H, Ar-H). C NMR: δ 22.4, 22.7, 26.4, 33.3, 89.5,
92.1, 116.2, 120.6, 127.6, 130.1, 130.9, 133.4, 135.9, 151.7, 157.0, 162.1. ESI-MS, m/z: 342.1
[M+H]+. Anal. Calcd. (Found) for C18H13Cl2N3 (341.05): C, 63.17 (63.51); H, 3.83 (3.49); N,
12.28 (12.51).
5.1.1.4. 2-Amino-4-(3,4-dichlorophenyl)-5,6,7,8-tetrahydronaphthalene-1,3-dicarbonitrile (4b)
Yield 90%, m.p. 233-235 °C. IR: 3420, 3304 (NH2); 2212 (2C≡N). 1H NMR (400 MHz, CDCl3):
δ 1.67-1.75 (m, 2H, CH2), 1.83-1.89 (m, 2H, CH2), 2.28-2.39 (m, 2H, CH2), 2.82-2.85 (m, 2H,
- 12 -

CH2), 5.20 (s, 2H, NH2, D2O-exchang.), 7.12-7.14 (dd, J = 2.8, 2.8 Hz, 1H, Ar-H), 7.38 (d, J =
2.8 Hz, 1H, Ar-H), 7.59 (d, J = 8.0 Hz, 1H, Ar-H). 13C NMR: δ 22.4, 22.7, 26.4, 33.3, 89.5, 91.5,
116.2, 117.3, 120.6, 127.6, 130.1, 130.9, 133.1, 133.4, 135.9, 151.7, 157.0, 162.1. ESI-MS, m/z:
342.1 [M+H]+. Anal. Calcd. (Found) for C18H13Cl2N3 (341.05): C, 63.17 (63.42); H, 3.83 (3.51);
N, 12.28 (12.59).
5.1.1.5. 2-Amino-4-(2-chloro-6-fluorophenyl)-5,6,7,8-tetrahydronaphthalene-1,3-dicarbonitrile
(4c)
1
Yield 66%, m.p. 120-122 °C. IR: 3448, 3378 (NH2); 2215, 2169 (2C≡N). H NMR (400 MHz,
CDCl3): δ 1.70-1.76 (m, 2H, CH2), 1.86-1.90 (m, 2H, CH2), 2.25-2.29 (m, 2H, CH2), 2.88 (t, J =
6.4 Hz, 2H, CH2), 5.33 (s, 2H, NH2, D2O-exchang.), 7.14-7.18 (m, 1H, Ar-H), 7.36-7.45 (m, 2H,
13
Ar-H). C NMR: δ 21.8, 22.1, 25.1, 31.7, 92.1, 98.0, 114.5, 114.7, 114.9, 122.7, 125.9 (d, J =
3.5 Hz), 131.7 (d, J = 9.2 Hz), 133.3 (d, J = 4.1 Hz), 148.6, 155.8, 157.7, 159.3, 160.2. ESI-MS,
m/z: 324.5 [M-H]+. Anal. Calcd. (Found) for C18H13ClFN3 (325.08): C, 66.36 (66.64); H, 4.02
(4.31); N, 12.90 (12.63).
5.1.1.6. 2-Amino-4-(2,6-difluorophenyl)-5,6,7,8-tetrahydronaphthalene-1,3-dicarbonitrile (4d)
Yield 80%, m.p. 215-217 °C. 1H NMR (700 MHz, CDCl3): δ 1.61-1.63 (m, 2H, CH2), 1.72-1.74
(m, 2H, CH2), 2.15-2.17 (m, 2H, CH2), 2.87-2.89 (m, 2H, CH2), 6.64 (s, 2H, NH2, D2O-
exchang.), 7.32-7.34 (m, 2H, Ar-H), 7.62-7.65 (m, 1H, Ar-H). 13C NMR: δ 21.8, 22.1, 26.4,
29.4, 96.2, 97.9, 112.1-112.3 (dd, J = 6.3, 6.4 Hz), 113.8 (t, J = 29.0 Hz), 115.3 (d, J = 8.4 Hz),
125.6, 131.7 (t, J = 14.0 Hz), 137.9, 148.3, 151.2, 158.5 (d, J = 9.2 Hz), 160.5 (d, J = 9.2 Hz).
ESI-MS, m/z: 310.2 [M+H]+. Anal. Calcd. (Found) for C18H13F2N3 (309.11): C, 69.89 (70.11);
H, 4.24 (4.57); N, 13.58 (13.23).
5.1.1.7. 2-Amino-4-((1,1'-biphenyl)-4-yl)-5,6,7,8-tetrahydronaphthalene-1,3-dicarbonitrile (4e)
Yield 90%, m.p. 251-253 °C. IR: 3416, 3307 (NH2); 2210 (2C≡N). 1H NMR (400 MHz, DMSO-
d6): δ 1.59-1.61 (m, 2H, CH2), 1.73-1.76 (m, 2H, CH2), 2.27 (t, J = 6.0 Hz, 2H, CH2), 2.72 (t, J =
6.4 Hz, 2H, CH2), 6.62 (s, 2H, NH2, D2O-exchang.), 7.36-7.43 (m, 3H, Ar-H), 7.51 (t, J = 7.6
Hz, 2H, Ar-H), 7.76 (d, J = 7.2 Hz, 2H, Ar-H),7.81 (d, J = 8.4 Hz, 2H, Ar-H). 13C NMR: δ 22.6,
22.9, 26.6, 33.4, 89.9, 92.1, 116.7, 120.9, 127.2, 127.3, 127.4, 127.6, 128.7, 128.8, 134.9, 140.3,
141.7, 154.2, 157.1, 161.5. ESI-MS, m/z: 348.2 [M-H]+. Anal. Calcd. (Found) for C24H19N3
(349.16): C, 82.49 (82.13); H, 5.48 (5.76); N, 12.03 (12.35).
- 13 -

5.1.1.8.
2-Amino-4-([1,3]benzodioxol-4-yl)-5,6,7,8-tetrahydronaphthalene-1,3-dicarbonitrile
(4f)
Yield 79%, m.p. 255-257 °C. IR: 3430, 3301 (NH2); 2212 (2C≡N). 1H NMR (400 MHz, DMSO-
d6): δ 1.60-1.64 (m, 2H, CH2), 1.72-1.75 (m, 2H, CH2), 2.35-2.19 (m, 2H, CH2), 2.71 (t, J = 6.0
Hz, 2H, CH2), 6.02 (s, 1H, Dioxole-H), 6.06 (s, 1H, Dioxole-H), 6.63 (s, 2H, NH2, D2O-
exchang.), 6.75-6.77 (dd, J = 1.2, 1.2 Hz, 1H, Ar-H), 6.96 (t, J = 8.0 Hz, 1H, Ar-H), 7.02-7.04
13
(dd, J = 1.2, 1.2 Hz, 1H, Ar-H). C NMR: δ 22.6, 22.8, 25.6, 33.2, 88.5, 101.6, 109.5, 116.9,
117.9, 119.4, 122.3, 122.5, 127.2, 127.7, 144.5, 147.7, 148.7, 158.4, 161.7. ESI-MS, m/z: 318.1
[M+H]+, 316.2 [M-H]+. Anal. Calcd. (Found) for C19H15N3O2 (317.12): C, 71.91 (71.56); H, 4.76
(4.52); N, 13.24 (13.55).
5.1.1.9. 2-Amino-4-(naphthalen-2-yl)-5,6,7,8-tetrahydronaphthalene-1,3-dicarbonitrile (4g)
Yield 87%, m.p. 220-222 °C. IR: 3408, 3306 (NH2); 2210 (2C≡N). 1H NMR (400 MHz, CDCl3):
δ 1.67-1.69 (m, 2H, CH2), 1.84-1.90 (m, 2H, CH2), 2.38-2.39 (m, 2H, CH2), 2.88 (t, J = 6.4 Hz,
2H, CH2), 5.32 (s, 2H, NH2, D2O-exchang.), 7.36-7.38 (dd, J = 1.6, 1.6 Hz, 1H, Ar-H), 7.56-7.58
(m, 2H, Ar-H), 7.77 (s, 1H, Ar-H), 7.89-7.93 (m, 2H, Ar-H), 7.98 (d, J = 8.4 Hz, 1H, Ar-H). 13C
NMR: δ 22.5, 22.8, 26.5, 33.3, 90.1, 90.2, 116.0, 120.7, 125.5, 126.6, 126.8, 127.6, 127.8, 128.2,
128.5, 133.5, 133.6, 154.4, 157.1, 161.5. ESI-MS, m/z: 324.1 [M+H]+. Anal. Calcd. (Found) for
C22H17N3 (323.14): C, 81.71 (81.43); H, 5.30 (5.67); N, 12.99 (12.67).
5.1.1.10. 2-Amino-4-(anthracen-9-yl)-5,6,7,8-tetrahydronaphthalene-1,3-dicarbonitrile (4h)
Yield 85%, m.p. 296-299 °C. 1H NMR (500 MHz, CDCl3): δ 1.49-1.57 (m, 2H, CH2), 1.82-1.86
(m, 2H, CH2), 1.95 (t, J = 6.5 Hz, 2H, CH2), 2.94 (t, J = 6.5 Hz, 2H, CH2), 5.19 (s, 2H, NH2,
D2O-exchang.), 7.47-7.55 (m, 5H, Ar-H), 8.05-8.10 (m, 4H, Ar-H). 13C NMR: δ 22.5, 25.3, 28.6,
33.4, 91.9, 106.7, 116.0, 121.1, 124.6, 125.4, 125.8, 125.9, 126.8, 127.4, 128.9, 131.2, 140.9,
152.3, 157.3, 157.4. ESI-MS, m/z: 374.2 [M+H]+. Anal. Calcd. (Found) for C26H19N3 (373.16):
C, 83.62 (83.93); H, 5.53 (5.37); N, 11.25 (11.58).
- 14 -

5.2. Biology
Detailed methods of biological assays are described in the Supplementary file.
5.2.1. In vitro antitumor assay against cancer cells and cytotoxicity against normal cells
MTT assay was adopted for evaluation of antitumor activity against cancer cells and
cytotoxicity against normal cells [42,43].
5.2.2. Mechanistic study
5.2.2.1. Tubulin polymerization inhibition assay
Inhibition of tubulin polymerization was evaluated adopting the reported method [45,46].
5.2.2.2. Flow cytometry assay
Flow cytometry assay was employed for assessment of cell cycle arrest in HepG2, HCT-116
and MCF-7 cancer cells according to the reported procedure [48-52].
5.2.2.3. Caspases 3/9 assay
Levels of caspases 3 and 9 were assessed following the reported procedure [63].
5.3. Computational aided studies
5.3.1. Docking studies
The molecular docking calculations were done utilizing “molecular operating environment
(MOE) version 2019.01” Chemical Computing Group Inc. software [65]. The crystal structure
of tubulin (PDB code: 4O2B) complex was attained from the RCSB Protein Data Bank [64].
Energy minimization of compound 4c was executed utilizing MMFF94x force field until a root-
mean-square (RMS) gradient of 0.01 kcal/mol was reached. Additionally, all hydrogens were
initially added and the forcefield partial charges were computed.
- 15 -

5.3.2. Molinspiration calculations
Molinspiration software [66] was utilized for analysis of Lipinski’s rule parameters [67],
TPSA and Nrotb of the new compounds [68]. Results (Supplementary Table S1) are described
in the Supplementary file.
5.3.3. Prediction of toxicity risks in human and carcinogenicity in mice and rats
The new derivatives were analyzed for the prediction of various toxicity risks in human
(tumorigenicity, irritancy and reproductive effects) using Osiris software [69], as well as
carcinogenicity in mice and rats using PreADMET software [70]. Results (Supplementary Table
S2) are described in the Supplementary file.
Acknowledgments
Gratitude to Holding Company for Biological Products and Vaccines (VACSERA), Egypt
for carrying out the cytotoxicity, tubulin polymerization inhibition, cell cycle analysis and
caspases 3/9 assays.
- 16 -

Fig. 1. Pharmacophoric features of CBSIs. The pharmacophoric points and distances between them are given in
◦
A . Hydrogen bond acceptors (A1, A2, A3) are colored blue, Hydrogen bond donor (D1) is colored red,
Hydrophobic centers (H1, H2) are colored green, and Planar moiety (R1) is colored grey.
- 17 -

H3C
A2
O
O
H2
A
CH3
CH3
R1
B
O
O
C
N
H3C
H
CH3
H1
O
O
A1
Colchicine
Fig. 2. The pharmacophoric features of colchicine.
- 18 -

OCH3
OCH3
H3CO
OCH3
OCH3
R
H3CO
O
OCH3
OCH3
HO
OCH3
OCH3
OCH3
O
R = 4-(NC2H5)2, 3-NH2-4-OCH3
A
B
D
C
OCH3
OCH3
HO
H3CO
OH
OCH3
N
H2C
H2C
O
NH
N
CH3
N
CH3
N
CH3
E
F
G
Fig. 3. Napthalenes A-D and quinolines E-G with reported antitumor and tubulin polymerization inhibitory activities.
- 19 -

F
F
OCH3
OCH3
H3CO
F
Cl
CN
CN
CN
N
N
N
N
N
N
NH
H
NH
N
F
NH
H3C
J
I
H
R
R
OCH3
Br
H3CO
CN
CN
N
CN
CH3
HO
O
NH2
HO
O
N
H2N
O
NH2
CH3
NH2
R = Br, Cl
L
R = Br, Cl
M
K (Crolibulin)
Fig. 4. 4,6-Diarylpyridine-3-carbonitrile H, 7-aryl-5-indolylimidazo[1,2-a]pyridine-8-carbonitriles I, J and 4-aryl-
4H-chromenes K-M with reported antitumor and tubulin polymerization inhibitory activities.
- 20 -

CN
OCH3
Br
H3CO
N
H
NH2
3a,b
CN
H2N
O
NH2
NH2
Crolibulin
CN
NH2
CN
4a-h
Different aromatic moieties
=
Fig. 5. The designed new analogs 3a,b and 4a-h as potential antitumor agents and tubulin polymerization inhibitors.
- 21 -

Fig. 6. The ORTEP plots of compounds 4b (A) and 4d (B).
- 22 -

Fig. 7. Tubulin polymerization inhibition of 4c.
- 23 -

IC5
0µ
M
IC50(µM)
Fig. 8. Flow cytometric analysis of cell cycle phase distribution in HepG2, HCT-116 and MCF-7 cells after
treatment with 4c (at IC50 of the corresponding cell line) for 24 h.
- 24 -

PI
Annexin V-FITC
Fig. 9. Annexin V-FITC/PI double staining for analysis of apoptosis in HepG2, HCT-116 and MCF-7 cells after
treatment with 4c (at IC50 of the corresponding cell line) for 24 h. Q1 quadrant points to dead (necrotic) cells; Q2
quadrant points to late apoptosis; Q3 quadrant points to live cells; Q4 quadrant points to early apoptosis. Total
apoptosis is the summation of both early and late apoptosis.
- 25 -

Fig. 10. (A) Results of caspase-3 induction in HepG2 and HCT-116 cells. (B) Results of caspase-9 induction in MCF-7
cells.
- 26 -

Fig. 11. (A) 2D Diagram of the interaction between compound
4c and colchicine binding site. (B) 3D Diagram of the
interaction between compound 4c and colchicine binding site
(PDB code: 6JCJ) with bond distances shown. For clarity, only
interacting residues are displayed. Compound 4c is colored
cyan, -tubulin is colored violet, -tubulin was removed for
clarity.
Fig. 11. (A) 2D Diagram of re-docked colchicine binding at the colchicine binding site of tubulin with bond
◦
distances (A ) shown. (B) 3D Diagram of the interaction between re-docked colchicine at the colchicine binding
site. (C) Superimposition of the co-crystallized colchicine ligand (blue) and the docking pose of colchicine
(yellow) at the colchicine binding site, -tubulin is colored green and -tubulin is colored red. For clarity, some
non-interacting residues are deleted. (PDB code: 4O2B).
- 27 -

◦
Fig. 12. (A) 2D Diagram of the interaction between 4c and the colchicine binding site with bond distances (A )
shown. (B) 3D Diagram of the interaction between 4c and the colchicine binding site. Compound 4c is colored
cyan, -tubulin is colored green and -tubulin is colored red. For clarity, some non-interacting residues are
deleted. (PDB code: 4O2B).
- 28 -

Fig. 13. (A) 2D Diagram of tubulin-4c binding mode superposed on tubulin-colchicine complex. Colchicine is
colored red and 4c is colored green. (B) 3D Diagram of tubulin-4c binding mode superposed on tubulin-
colchicine complex. Colchicine is colored yellow, 4c is colored cyan, -tubulin is colored green and -tubulin is
colored red. For clarity, some non-interacting residues are deleted. (PDB code: 4O2B).
- 29 -