Combretastatin A-4 inspired novel 2-aryl-3-arylamino-imidazo-pyridines/ pyrazines as tubulin polymerization inhibitors, antimitotic and anticancer agents

Article abstract of DOI:10.1039/c3md00357d

Based on the pharmacophoric features of the natural product combretastatin A-4 (CA-4) and its synthetic analogues that inhibit tubulin polymerization, a series of novel 2-aryl-3-arylamino-imidazo-pyridines/pyrazines as potential antitubulin anticancer agents were designed. They were synthesized by a one-pot method involving preparation of isocyanides from the anilines via formylation and subsequent dehydration followed by their reactions with heterocyclic-2- amidines and aldehydes. Compounds 1, 2, 14, and 15 were found to exhibit significant tubulin polymerization inhibition and disruption of tubulin-microtubule dynamics similar to that of CA-4. They showed potent anticancer activities in kidney, breast and cervical cancer cell lines, and relatively low toxicity to normal cells, compared to CA-4. The compounds induced DNA and chromosomal damage, and apoptosis via cell cycle arrest in the G2/M phase. The molecular docking and molecular dynamics (MD) simulation studies revealed that disruption of microtubule dynamics might occur by interaction of the compounds at the colchicine binding site of the α,β-tubulin heterodimer interface, similar to that of CA-4. Molecular modelling analysis showed that two of the three methoxy groups at ring A of all four potent compounds (1, 2, 14, and 15) were involved in bifurcated hydrogen bonding with Cysβ241, an important molecular recognition interaction to show tubulin inhibitory activity. In comparison to CA-4, the bridging NH and the imidazo-pyridine/pyrazine moieties in the title compounds provide flexibility for attaining the required dihedral relationship of two aryls and additional pharmacophoric features required for the interaction with the key residues of the colchicine binding site.

Full text of DOI:10.1039/c3md00357d

View Article Online
View Journal
MedChemComm
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: S. K. Guchhait, N.
Sanghai, V. Jain, R. Preet, S. Kandekar, S. Das, N. Trivedi, P. Mohapatra, G. Priyadarshani, M. Kashyap, D.
Das, S. R. Sathapathy, S. Siddharth, C. N. Kundu and Prasad. V. Bharatam, Med. Chem. Commun., 2014,
DOI: 10.1039/C3MD00357D.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/medchemcomm

Page 1 of 44
Medicinal Chemistry Communications
View Article Online
DOI: 10.1039/C3MD00357D
Combretastatin A-4 Inspired Novel 2-Aryl-3-arylamino-imidazo-
pyridines/pyrazines as Tubulin Polymerization Inhibitors, Antimitotic and
Anticancer Agents
Nitesh Sanghai,a Vaibhav Jain,b Ranjan Preet,c Somnath Kandekar,a Sarita Das,c Neha Trivedi,b
Purusottam Mohapatra,c Garima Priyadarshani,a Maneesh Kashyap,a Dipon Das,c Shakti
Ranjan Sathapathy,c Sumit Siddharth,c Sankar K. Guchhait,*a Chanakya N. Kundu,*c and
Prasad V. Bharatama,b
aDepartment of Medicinal Chemistry, National Institute of Pharmaceutical Education and
Research (NIPER), Sector 67, S. A. S. Nagar (Mohali), Punjab–160062, India
bDepartment of Pharmacoinformatics, National Institute of Pharmaceutical Education and
Research (NIPER), Sector 67, S. A. S. Nagar (Mohali), Punjab–160062, India
cSchool of Biotechnology, KIIT University, Campus–11, Patia, Bhubaneswar, Odisha–751024,
India
Corresponding Authors*
For S.K.G.: phone, +91 (0)172 2214683; fax, +91 (0)172 2214692;
E-mail, skguchhait@niper.ac.in.
For C.N.K.: phone, +91 (0)674 2725466; fax, + 91 (0)674 2725732;
E-mail, cnkundu@gmail.com.
Graphical Abstract
1

Medicinal Chemistry Communications
Page 2 of 44
View Article Online
DOI: 10.1039/C3MD00357D
Abstract
Based on the pharmacophoric features of the natural product combretastatin A-4 (CA-4) and its
synthetic analogues that inhibit tubulin polymerization, a series of novel 2-aryl-3-arylamino-
imidazo-pyridines/pyrazines as potential antitubulin anticancer agents were designed. They were
synthesized by a one-pot method involving preparation of isocyanides from the anilines via
formylation and subsequent dehydration followed by their reactions with heterocyclic-2-
amidines and aldehydes. Compounds 1, 2, 14, and 15 were found to exhibit significant tubulin
polymerization inhibition and disruption of tubulin-microtubule dynamics similar to that of CA-
4. They showed potent anticancer activities in kidney, breast and cervical cancer cell lines, and
relatively low toxicity to normal cells, compared to CA-4. The compounds induced DNA and
chromosomal damage, and apoptosis via cell cycle arrest in the G2/M phase. The molecular
docking and molecular dynamics (MD) simulation studies revealed that disruption of
microtubule dynamics might occur by interaction of the compounds at the colchicine binding site
at the interface of α,β-tubulin heterodimer, similar to that of CA-4. Molecular modelling analysis
showed that two of the three methoxy groups at ring A of all the four potent compounds (1, 2,
14, and 15) were involved in bifurcated hydrogen bonding with Cysβ241, an important
molecular recognition interaction to show tubulin inhibitory activity. In comparison to CA-4, the
bridging NH and the imidazo-pyridine/pyrazine moieties in the title compounds provide
flexibility for attaining the required dihedral relationship of two aryls and additional
pharmacophoric features required for the interaction with the key residues of the colchicine
binding site.
Keywords: Combretastatin, Imidazo-pyridine, Imidazo-pyrazine, Antitubulin, Antimitotic,
Anticancer, Molecular Docking, Molecular Dynamics Simulation
2

Page 3 of 44
Medicinal Chemistry Communications
View Article Online
DOI: 10.1039/C3MD00357D
Introduction
Microtubules are cytoskeletal filaments composed of
α
,β-tubulin heterodimer assembly.1 Their
polymerization-depolymerization dynamics are involved in various cellular processes like
mitotic spindle formation during cell division, migration, cell shape organization, and
transportation of vesicles and mitochondria.1,2 Agents that disrupt tubulin-microtubule dynamics
induce mitotic arrest and eventually cell death by apoptosis.3,4 Marketed anticancer drugs that
target such dynamics are Catharanthus (Vinca) alkaloids (vincristine, vinblastine and
vinorelbine), taxanes (paclitaxel and docetaxel), and epothilones.5,6 Therefore, tubulin is
considered as an important target in anticancer drug discovery research.3,7-9
Three binding sites are present in tubulin; the taxane-binding site, the vinca-binding site,
and the colchicine binding site.10,11 Ligands that target the taxane binding pocket act as
microtubule stabilizing agents preventing depolymerization,12,13 whereas agents that interact at
the vinca or the colchicine binding sites inhibit the assembly of the α,β-tubulin towards
polymerization.14 Natural ligands that interact at the colchicine binding site include colchicine,
steganacin, podophyllotoxin, and combretastatins. They possess similar structural scaffolds,
biaryl (for colchicine and steganacin), 1,1-diarylmethane, and 1,2-diarylethene, respectively (Fig.
1). Among combretastatins, CA-4 exhibits pronounced antitubulin cytotoxic effects.15,16 It
specifically interacts at the interface of α,β-tubulin heterodimer and interferes with the flipping in
and out of the β-tubulin loop, which in turn prevents the transition from curved to straight
heterodimer required for the formation of microtubule assembly.6,17 Certain agents that bind to
the colchicine binding site also lead to tumour vascular disruption.18-25 CA-4 suffers from poor
aqueous solubility, low bioavailability, and short biological half life.26 To overcome these
limitations, various derivatization and formulations of CA-4 have been explored in the last
3

Medicinal Chemistry Communications
Page 4 of 44
View Article Online
DOI: 10.1039/C3MD00357D
decade.27-35 The prodrugs, combretastatin A-4 3-O-phosphate (CA-4-P, Zybrestat),
combretastatin A-1 2,3-O-diphosphate (CA-1-P, OXi4503), and amino acid analogue of CA-4
(AVE8062) are currently in clinical trials.19,36,37
Fig. 1 Natural products that target the colchicine binding site and possess structural resemblance:
Combretastatins A-1 and A-4; their (di)phosphates and amino acid analogue prodrugs are in
clinical trials.
The important pharmacological features and structural simplicity of CA-4 have made it
an attractive lead compound for the development of novel antitubulin anticancer agents. From
the structure-activity relationship studies of CA-4, it has been evident that the 3,4,5-trimethoxy
substitutions on the ring A, 4-methoxy group on ring B and a cis configuration between rings A
and
B
are pharmacophoric requirements for the optimal tubulin inhibitory
activity/cytotoxicity.26,30,38-40
The 3-hydroxy group in ring B is amenable for isosteric replacement or further
derivatization for conjugation.41,42 A study of cis and trans-stilbenes related to CA-4 showed that
cis-stilbene configuration provided optimal activity.43 The stilbenoid structure of CA-4 makes it
liable to isomerisation from less stable cis to more stable but inactive trans isomer. This
isomerisation is triggered by light, heat and protic media and thus reduces its therapeutic
4

Page 5 of 44
Medicinal Chemistry Communications
View Article Online
DOI: 10.1039/C3MD00357D
efficacy.15,44 The in vitro study of metabolic stability of CA-4 revealed its susceptibility for Z-E
isomerization of the double bond.45 Isomerization of cis to trans form can be prevented by
locking the cis-type bridge between the two aryl rings. With this perspective, numerous cis-
locked analogues of CA-4 with the incorporation of various heterocyclic, carbocyclic, and
functional bridging groups as double bond replacements have been explored (Fig. 2).30,38,43,46-49
A wide range of heterocyclic bridges have been investigated; they include pyrazoles, thiazoles,
triazoles, tetrazoles, oxazoles, imidazoles, furans, furanones, furazans, dioxolanes, thiophenes,
and indoles.33,48-50 Carbocyclic moieties that replace double bond are cyclopropane,51 and
cyclopentane.48,52 The functional bridging groups involve ethers, ketones, sulfonamides,
sulfonates, amines, and amide.53 Recently, pharmacomodulation strategies for the design of
potent CA-4 analogues are reported by using introduction of β-lactam ring44,54,55 and
isocombretastatin.56,57
Fig. 2 A few representative analogues of CA-4 replacing double bond with A) heterocyclic, B)
carbocyclic, and C) functional bridging moieties.
As a part of our program aimed at natural product based anticancer drug discovery,58 we
were interested in exploration of novel antitubulin anticancer agents and envisaged CA-4
5

Medicinal Chemistry Communications
Page 6 of 44
View Article Online
DOI: 10.1039/C3MD00357D
inspired novel 2-aryl-3-arylamino-imidazo-pyridines/pyrazines as potential tubulin
polymerization inhibitors (Fig. 3).50,59-62
R
MeO
X
OMe
O
MeO
MeO
OMe
Can provide hydrophobic and
hydrogen bond acceptor/donor in
addition to cis lock
X = NH, O, S
R= H, OH
cis-lock
R6
Hydrogen bond
acceptor
Hydrophobic
alkene/Ar/
HtAr
Im-pyd
/pyz
B
R5
:
Im-pyd/pyz
Imidazo[1,2-a]pyridine/pyrazine
MeO
A
R4
NH
OH
OMe
B
R1
R2
A
MeO
OMe
Can enable required dihedral
relationship of two rings
R3
Hydrogen bond
acceptor
Hydrophobic
Novel compounds designed as
potential antitubulin anticancer agents
CA-4 anlogues
Fig. 3 Design of novel 2-aryl-3-arylamino-imidazo-pyridines/pyrazines as potential antitubulin
anticancer agent.
These compounds possess imidazo-pyridine/pyrazine ring as double bond replacement to
lock cis-conformation of rings A and B. Imidazo[1,2-a]-pyridine/pyrazine is a class of scaffold
present in several marketed drugs and many bioactive compounds.63-65 Therefore, its
incorporation could afford, along with cis-lock, favourable physicochemical properties,
improved pharmacokinetic profiles, and additional hydrogen bond donor/acceptor points as well
as hydrophobic interactions with α,β-tubulin heterodimer. It has been established that for
effective interaction at colchicine binding site in tubulin, CA-4 class of compounds require two
phenyl rings to be tilted to each other at a particular range of dihedral angles (around 45º -
55º).66,67 Natural products colchicine, podophyllotoxin, steganacin, and combretastatins also
possess similar dihedral relationships between rings
A
and B. 2‑Aryl-3-aroyl-
6

Page 7 of 44
Medicinal Chemistry Communications
View Article Online
DOI: 10.1039/C3MD00357D
indoles/benzofurans/benzothiophens have been reported as potent inhibitors of tubulin
polymerization.50,60,62 Considering these, we therefore speculated that in our designed
compounds also the NH unit could provide flexibility for spatial orientation of two aromatic
rings A and B in the colchicine binding site of α,β-tubulin heterodimer.
In this article, we present the synthesis, bio-evaluation, and molecular modelling studies
of CA-4 inspired novel 2-aryl-3-arylamino-imidazo-pyridines/pyrazines (1-20). The synthesized
compounds were tested in vitro and ex vivo for their tubulin assembly inhibitory activities,
anticancer activities in kidney, breast and cervical cancer cells, and cytotoxicity towards normal
cells. The activities were compared with CA-4 and a clinically used anticancer drug 5-
fluorouracil. Various biochemical and immunological studies like DAPI (4̕, 6–diamidino–2–
phenylindole) nuclear staining, expression of representative apoptotic protein markers (caspase–
3, PARP cleavage, and p53), caspase staining (immunocytochemistry), comet assay, and
cytokinesis-block micronucleus (CBMN) assay were carried out to establish the potential of
investigated compounds in apoptosis, and damage in DNA as well as chromosomal level. Flow
cytometry was also carried out to identify the phase of cell cycle arrest for apoptosis.
Furthermore, with an intention of revealing the mode of binding, atomistic level interactions and
binding affinity of the title compounds with the α,β-tubulin heterodimer, molecular docking
analysis was carried out on all the investigated compounds. Molecular docking results for the
potent antitubulin, anticancer compounds were compared with that of the lead compounds
colchicine and CA-4. In addition, molecular dynamics (MD) simulations were also performed on
these potent compounds to assess the stability of intermolecular interactions in the dynamical
conditions and to estimate the binding free energy of ligands.
Results and Discussion
7

Medicinal Chemistry Communications
Page 8 of 44
View Article Online
DOI: 10.1039/C3MD00357D
Chemistry
The 3-aminoimidazo[1,2-a]pyridine/pyrazine class of compounds are, in general, prepared by an
Ugi–type multicomponent reaction (also referred as Groebke–Blackburn–Bienayme reaction)68-70
of a heterocyclic-2-amidine with aldehyde and isocyanide. Versatile aldehydes are known to
undergo this reaction and can generate substitution diversity at C2 of the product. However, it
suffers from the generation of substitution diversity at C3-amine in the scaffold because
isocyanides are relatively expensive, unavailable commercially and difficult to prepare.64,71-73 In
our study, the CA-4 inspired novel imidazo-pyridine/pyrazine derivatives required the
incorporation of relevant aromatic diversity at C3-amine. A new approach of one-pot preparation
of isocyanide directly from amine and its utilization in the Ugi-type reaction was recently
developed by our research group.74 It employs three reactions, formylation of amine, dehydration
of formamide to isocyanide, and its reaction with aldehyde and heterocyclic amidine (Scheme 1).
The nature of the dehydrating agent and base were found to play a key role in the one-pot
process. The method has enabled the incorporation of differently substituted ring A at C3-amine
and ring B at C2 in the products with high yields. The process being one-pot does not require any
tedious work-up and purification of any of the intermediates, and minimizes waste products.
Using this methodology, CA-4 inspired imidazo[1,2-a]-pyridines and pyrazines, imidazo[2,1-
b]thiazole, and imidazo[2,1-b]benzothiazole analogs (compounds 1-20, Fig. 4) with relevant
variation at 2-aryl and 3-arylamine were prepared.
Scheme 1. One-pot method for preparation of investigated compounds (1-20)
8

Page 9 of 44
Medicinal Chemistry Communications
View Article Online
DOI: 10.1039/C3MD00357D
Fig. 4 Synthesized 2-aryl and 3-arylamino substituted imidazo[1,2-a]-pyridines and pyrazines,
imidazo[2,1-b]thiazole, and imidazo[2,1-b]benzothiazole analogs (1-20).
9

Medicinal Chemistry Communications
Page 10 of 44
View Article Online
DOI: 10.1039/C3MD00357D
Biological studies
CA-4 inspired imidazo-pyridines/pyrazines inhibit tubulin polymerization and disrupt
microtubule dynamics
To study the tubulin polymerization inhibitory activity of novel 2-aryl-3-arylamino-imidazo-
pyridines/pyrazines, an ex vivo assay in kidney cancer cells (HEK 293T) was performed
according to the protocol described under experimental section. All the synthesized compounds
(1-20) were tested with CA-4 as positive control. Interestingly, they were found to possess
moderate to good activity of tubulin assembly inhibition (Table 1). The most potent activity was
found for compounds 1, 2, 14, and 15 with the IC50 (concentration required to inhibit 50% of
tubulin assembly) value of 12, 10, 11, and 14 µM, respectively, which were significant in
comparison to that of CA-4 (IC50: 9 µM). We were then curious to further confirm the activity of
these potent compounds (1, 2, 14, and 15) for disruption of tubulin assembly by immunostaining
method. HEK 293T cells were treated with these compounds and CA-4 at their respective IC50
concentrations and then immunostained with tubulin antibody (Fig. 5). The upper panel
represents the tubulin protein assembly of the treated as well as untreated HEK 293T cells
labeled with FITC. The middle panel represents the propidium iodide stained nucleus of the cells
and the lower panel is the merge of the above two panels. The results imply that the CA-4 and
investigated compounds 1, 2, 14, and 15 induced significant distortions in the tubulin assembly
of the treated HEK 293T cells leading to the formation of abnormal mitotic spindles in
comparison with control. Moreover, investigated compounds showed also significant disruption
of tubulin assembly in a similar manner to that of CA-4.
10

Page 11 of 44
Medicinal Chemistry Communications
View Article Online
DOI: 10.1039/C3MD00357D
Table 1 IC50 of tubulin assembly inhibition using cellular lysate of HEK 293T for investigated
compounds and CA-4. Data is the mean SD of three different experiments.
Compound IC50 (µM)
Compound IC50 (µM)
Compound IC50 (µM)
1
8
15
12 ± 2.0
10 ± 1.8
38 ± 2.3
36 ± 2.1
20 ± 2.5
40 ± 2.0
39 ± 1.2
32 ± 3.0
36 ± 1.8
34 ± 1.7
44 ± 1.6
46 ± 2.4
38 ± 1.8
11 ± 2.0
14 ± 2.2
34 ± 1.5
48 ± 1.9
42 ± 3.0
54 ± 1.8
38 ± 2.0
9 ± 1.2
2
9
16
3
10
17
4
11
18
5
12
19
6
13
20
7
14
CA-4
Fig. 5 Immunostaining of tubulin assembly in HEK 293T cells. The upper, middle and lower
panels represent the tubulin assembly stained with FITC, the propidium iodide and the merge of
above panels, respectively. Images were taken under confocal microscope (Leica TCS SP5).
Micrographs presented here is the best of three individual experiments.
Cytotoxic activities of 2-aryl-3-arylamino-imidazo-pyridines/pyrazines
MTT assay was carried out to investigate the cytotoxic activity of all synthesized 2-aryl-3-
arylamino-imidazo-pyridines/pyrazines (1-20) in both kidney cancer cells (HEK 293T) and its
11

Medicinal Chemistry Communications
Page 12 of 44
View Article Online
DOI: 10.1039/C3MD00357D
corresponding normal cells (Vero). CA-4 was used as a positive control in the assay. Cells were
treated with investigated compounds and CA-4 of increasing concentrations and the viability was
measured in concentration dependent manner according to the protocol described under
experimental section. It was noted that compounds 1, 2, 14, and 15 resulted in 50% cell death
(LC50) in HEK 293T cells at relatively lower concentrations (10, 8, 12, and 10 µM, respectively)
than other compounds and 50% cell death in Vero cells at higher concentrations viz. 50, 48, 70,
and 65 µM, respectively (Table 2 and Fig. S1). Thus, these results demonstrated that the
compounds 1, 2, 14, and 15 are more potent cytotoxic agents than other tested compounds and
posses significant cytotoxicity in HEK 293T, but less cytotoxicity for Vero cells. To further
extend the cytotoxic potential of CA-4 and the investigated compounds, we have measured
cytotoxicity in a panel of cell lines with different time points (24 h and 48 h) of treatment (Table
3). The data showed that SiHa and HeLa (cervical cancer cell lines) were more susceptible to
CA-4 in comparison to MCF-7 and HEK 293T cells. To measure the anti-colony forming
potential of the compounds against a single cell towards formation of a colony or clone in long
term, a widely used clonogenic cell survival assay was performed. It was noticed that compounds
1, 2, 14 and 15 showed significant reduction in colony formation in HEK 293T cells (Fig. S2).
The investigated compounds (1, 2, 14 and 15) showed also significant anchorage dependent anti-
cell migration potential in HEK 293T cells (Fig. S3). Interestingly, these cytotoxicity results
were in accordance with the tubulin polymerization inhibitory activities of the investigated
compounds and CA-4. This indicates the disruption of tubulin-microtubule dynamics as a major
pathway for cytotoxic properties of the investigated 2-aryl-3-arylamino-imidazo-
pyridines/pyrazines class of compounds.
12

Page 13 of 44
Medicinal Chemistry Communications
View Article Online
DOI: 10.1039/C3MD00357D
Table 2 LC50 values (ꢀM) of CA-4 and investigated compounds (1-20) for kidney cancer cells
(HEK 293T) and its corresponding normal cells (Vero).
LC50 (Vero)
LC50 (Vero)
Compound
LC50
Compound
LC50
(HEK 293T)
10 ± 1.0
8 ± 0.5
(HEK 293T)
34 ± 1.8
38 ± 2.4
12 ± 0.6
10 ± 0.3
34 ± 2.0
48 ± 1.9
47 ± 2.6
54 ± 1.8
63 ± 2.0
8 ± 0.6
1
2
12
13
50 ± 2.0
48 ± 3.0
50 ± 2.3
62 ± 2.5
26 ± 2.0
75 ± 2.4
63 ± 1.6
32 ± 2.7
48 ± 1.9
62 ± 1.5
72 ± 2.0
45 ± 2.0
62 ± 2.5
70 ± 1.6
65 ± 2.0
54 ± 1.6
58 ± 2.4
74 ± 3.0
56 ± 1.6
74 ± 2.2
11 ± 0.5
3
14
38 ± 2.0
36 ± 3.0
22 ± 2.1
54 ± 2.2
42 ± 2.3
25 ± 2.2
36 ± 1.7
34 ± 2.0
56 ± 2.5
4
15
5
16
6
17
7
18
8
19
9
20
10
11
CA-4
Table 3 LC50 values (ꢀM) of CA-4 and investigated compounds (1, 2, 14, and15) against various
cancer cells with different time of treatment.
Compound
SiHa (µM)
48 h
0.2 0.01
0.1
1.5 0.1
MCF-7 (µM)
24 h 48 h
0.8
HEK-293T (µM)
HeLa (µM)
24 h 48 h
24 h
24 h
48 h
0.6
0.02
7
1
2
8
1
0.2
6
1
0.8
0.2
10
CA-4
1
0.01
5
1
2
20
2.3
15
12
1
1
10
1
5
1
15 2.1
1
4
7
1
1
2
10
6
8
2
3
1.4
1.2
15
17
2
2
7
9
1
2
3
2
0.9
0.9
18
1.5
1
12
2
2
3
1
1
14
5
1.2
8
1
5
1
10
5
1
9
1
5
15
Determination of phase of the cell cycle arrest in kidney cancer cells
To investigate the regulation of cell cycle profile in HEK 293T cells, fluorescence activated cell
sorter (FACS) analysis was carried out. The cells were treated with CA-4 and the compounds 1,
2, 14, and 15 for 24 h, stained with propidium iodide (PI), and then FACS analysis was
performed. The DNA content was measured at each phase of the cell cycle and the data is
represented in Fig. S4. It was observed that all the four compounds induced cell cycle arrest in
13

Medicinal Chemistry Communications
Page 14 of 44
View Article Online
DOI: 10.1039/C3MD00357D
G2/M phase, which is in accordance to that of CA-4 and tubulin polymerization inhibition
property.
Apoptotic effect of the compounds 1, 2, 14, and 15 in kidney cancer cells
A nuclear staining experiment using HEK 293T cells was performed to investigate the apoptotic
effect of 2-aryl-3-arylamino-imidazo-pyridines/pyrazines 1, 2, 14, and 15. Chromatin
condensation and nuclear fragmentation of the cancer cells was analyzed by fluorescence
microscopy using the DNA-binding fluorescent dye DAPI after treating the cells with the
investigated compounds and CA-4 at their respective LC50 concentrations for 24 h using a
protocol as described in experimental section. It was observed that the chromatin condensation
and nuclear fragmentation increased significantly in the treated cells in comparison to untreated
cells (control) that suggested the presence of apoptotic nuclei (Fig. 6A). The apoptotic nuclei
resulted in the cancer cells treated with the investigated compounds increased substantially in
comparison with the untreated cells and was comparable to that of CA-4 (Fig. 6B). To further
corroborate the apoptotic effect of the compounds in HEK 293T cells, the expression level of
representative apoptotic marker proteins such as p53, PARP and its cleaved products was studied
(Fig. 6C). In the experiment, GAPDH was used as loading control. In comparison with the
untreated cells, treated cells showed a significant increase in the expression of p53 and the level
of cleaved products (86 kDa) of PARP, and reduced level of full length (116 kDa) of PARP.
These results indicate that the investigated compounds possess apoptotic effect in HEK 293T
cancer cells. To examine whether the induction of apoptosis by the investigated compounds is
associated with the activation of caspase-3, an immunocytochemical assay was performed. The
results indicate that the investigated compounds activate caspase-3 in HEK 293T cells (Fig. S5).
14

Page 15 of 44
Medicinal Chemistry Communications
View Article Online
DOI: 10.1039/C3MD00357D
Fig. 6 CA-4 and the investigated compounds 1, 2, 14, and 15 induce apoptosis in HEK 293T
cells. (A) DAPI nuclear staining experiment: Images were taken using fluorescent microscope
(Nikon, Japan) at 40X magnification. Images represented here are one of the replicates of three
different experiments. (B) A graphical representation of apoptotic nuclei of (A) (*P<0.05). The
red and green bars represent the non-apoptotic and apoptotic nuclei, respectively. (C) Western
blot analysis showing the expression level of p53, and PARP and its cleaved products in HEK
293T cells. GAPDH was used as loading control.
Compounds induce DNA and chromosomal damage
To determine whether the investigated compounds contribute to DNA damage, a well known
comet assay was performed. The HEK 293T cells were treated with CA-4 and the compounds 1,
2, 14, and 15 at their respective LC50 concentrations for 24 h and then the comet assay was
carried out according to the protocol described in experimental section. In treated cells, there
were lesions in DNA and the comet lengths were increased drastically, while no comet formation
was seen in untreated cells (Fig. S6A). The bar diagram shown in Fig. S6B represents the comet
lengths in pixels. From these data, it appears that the investigated compounds cause DNA
damage in kidney cancer cells. Further, to examine whether CA-4 and the investigated
15

Medicinal Chemistry Communications
Page 16 of 44
View Article Online
DOI: 10.1039/C3MD00357D
compounds have potential to cause damage at chromosomal level, a cytokinesis-block
micronucleus (CBMN) assay was performed. Micronuclei originate from the chromosome
fragments or whole chromosomes that lag behind at anaphase during nuclear division. CBMN
assay has evolved into a comprehensive method for measuring chromosome breakage, DNA
miss-repair, chromosome loss, necrosis, apoptosis, and cytostasis.75 The HEK 293T cells were
treated with CA-4 and the investigated compounds at their LC50 concentrations and then CBMN
assay was performed. It was observed that micronuclei formed in treated cells, whereas no
micronuclei formation was seen in untreated cells, indicating the chromosomal damage (Fig.
S6C). Therefore, these results confirm that the CA-4 and the investigated compounds not only
induce DNA damage but also the chromosomal damage simultaneously.
Molecular Docking analysis
Molecular docking has emerged as one of the very robust and efficient approaches for
understanding the interaction between macromolecules and small molecules, which further leads
to the rationalization of lead optimization efforts. In the past, several molecular modeling
studies76 like ligand-based pharmacophore mapping,66,77 molecular docking,78-80 and virtual
screening81,82 have been performed by various research groups on antitubulins acting on
colchicine binding site. These studies revealed an insight into the atomistic level details of
ligands interacting with α,β-tubulin heterodimer, their binding affinity and important
pharmacophoric features necessary for inhibition of microtubule assembly. Based on the
important pharmacophoric features considered in the design of the investigated compounds, it is
obvious to speculate that the compounds can interact at the colchicine binding site of tubulin
similar to that of CA-4. Accordingly, molecular docking studies were performed to investigate
the binding pose, intermolecular interactions and binding affinity of ligands with the α,β-tubulin
16

Page 17 of 44
Medicinal Chemistry Communications
View Article Online
DOI: 10.1039/C3MD00357D
heterodimer. The docking results of four potent compounds (1, 2, 14, and 15) are discussed
herein; the results of all investigated compounds (1-20) have been provided in Table S1
(supporting information). Moreover, the interaction profile and docking score of the potent
compounds were also compared with that of colchicine and CA-4 (for molecular docking
analysis of colchicine and CA-4, see supporting information, Fig. S7, Fig. S8).
Docking of potent 2-aryl-3-arylamino-imidazo-pyridines/pyrazines
The result of the ligands preparation in Ligprep module of Schrödinger molecular modeling
package83 showed that 2-aryl-3-arylamino-imidazo-pyridines/pyrazines at pH 7 2 can exist in
two different charged states (predicted by Epik program), protonated (charge +1, protonation
occurs at N1 nitrogen of imidazole ring) and neutral (charge 0). Further, to identify the most
plausible charged state of these species in terms of their stability, potent compounds (1, 2, 14,
and 15) were subjected to quantum chemical calculations using density functional theory (DFT)
method at B3LYP/6-31+G(d,p) level by GAUSSIAN03 programme.84 This analysis suggested
that the calculated protonation energy for these compounds falls in the range of 237-245
kcal/mol (Table S2, supporting information), which is quite favourable and thus they prefer to
exist in protonated form. Further, the pka value (~7.1) of imidazole nitrogen predicted by
ChemAxon MarvinView software85 corroborated that the compounds are basic in nature.
Therefore, the docking results for the protonated state of these compounds are presented in this
work (Table 4). In order to see the differences in the docking scores and intermolecular
interactions of neutral species (1-20), they were also docked into the colchicine binding site of
tubulin and the results are shown in Table S3 (supporting information). Minor variation in the
docking score and interactions in comparison to protonated state were observed for the neutral
state compounds.
17

Medicinal Chemistry Communications
Page 18 of 44
View Article Online
DOI: 10.1039/C3MD00357D
Docking of the potent investigated compounds (1, 2, 14, and 15) into the colchicine
binding site of α,β-tubulin heterodimer (Fig. S9) revealed that the major portion of these
compounds was buried in the β-subunit (showing maximum interactions) and a fewer
interactions with residues of α-subunit, colchicine and CA-4 (Fig. 7 and Fig. S9). The binding
pose and interactions for CA-4 and investigated compounds 1, 2, and 14 revealed that imidazo-
pyridine/pyrazine in the compounds can mimic the 3-hydroxy-4-methoxyphenyl moiety of CA-4
(see Fig. S10 for superimposition). This is an important observation in the aspect that the 3-
hydroxy-4-methoxyphenyl moiety in the CA-4 or its cis-locked analogues can be replaced by a
suitable heterocycle moiety such as imidazo-pyridine/pyrazine for retaining antitubulin activity.
The Glide docking scores for the potent compounds 1, 2, 14, and 15 (-9.33, -10.02, -8.82 and -
8.59, respectively) were comparable to that of known tubulin inhibitors colchicine (-10.19) and
CA-4 (-8.38) (Table 4). In addition, intermolecular interactions (hydrogen bonding, hydrophobic,
C-H· · ·π and N-H· · ·π) of the docked potent compounds were found to be consistent with the
binding model of colchicine and CA-4. These results are in agreement with the tubulin
polymerization inhibitory assay which showed that the potent compounds 1, 2, 14, and 15 (IC50:
12, 10, 11 and 14 µM, respectively) have comparable activity to that of CA-4 (IC50: 9 µM). The
methoxy group of all the potent compounds showed bifurcated hydrogen bonding with Cysβ241,
an essential interaction to show tubulin inhibition activity.76 Moreover, protonated nitrogen (N1)
of imidazo-pyridine/pyrazine ring of compounds 1, 2, and 14 formed hydrogen bonding with the
main chain carbonyl of Thrα179, whereas in compound 15 bridging NH showed relatively
weaker hydrogen bonding with the same residue (distance 3.74 Å). This is due to the lack of
flexibility between ring A and imidazo[1,2-a]-pyridine ring of compound 15, which flips the
orientation of ring B inside the binding site in comparison to other three compounds. The
18

Page 19 of 44
Medicinal Chemistry Communications
View Article Online
DOI: 10.1039/C3MD00357D
dihedral angle between ring A and imidazo-pyridine/pyrazine ring of compounds 1, 2 and 14 was
found to be 46.70°, 49.93° and 51.92°, respectively, while for compound 15 the dihedral angle
between rings A and B was 48.90°. Therefore, this endorses further the reported studies 66,67 that
optimum range of dihedral angle is necessary between the aryl rings for the compounds to bind
at colchicine site of the tubulin. This also signifies the importance of bridging NH in providing
the flexibility to the title compounds for showing maximum interactions with the α,β-tubulin
heterodimer. The imidazo-pyridine/pyrazine ring of all the four potent compounds were also
found to be involved in N-H· · ·π (except compound 15) and hydrophobic interactions with the
residues of the colchicine binding site, implying the significance of the heterocyclic scaffold
considered in the title compounds.
The binding model of the best docked compound 2 (Glide score -10.19) showing
interactions with the tubulin heterodimer is depicted in Fig. 7. In detail, methoxy groups of
compound 2 formed key interaction with the side chain -SH of Cysβ241 (hydrogen bond distance
1.84 Å and 2.42 Å). In addition, the 3-hydroxyl group of compound showed hydrogen bond with
the side chain carboxyl group of Gluα183 (2.01 Å) and protonated nitrogen of imidazole ring
formed hydrogen bond with the main chain carbonyl of Thrα179 (2.41 Å) (not present in the
unprotonated form of compound), which is further providing stability to the protein-ligand
complex. Other residues like Valα181, Alaα180, Leuβ242, Alaβ250, Alaβ316, Ileβ318 and
Ileβ378 were found to be involved in hydrophobic interactions, while some other weaker
contacts like CH· · ·π interaction with Leuβ255 and NH· · ·π interaction with Asnβ258 were also
observed for compound 2. Molecular docking of other investigated compounds 1, 14, and 15
showed almost similar intermolecular interactions as those of compound 2, CA-4, and
colchicine. This molecular docking study indicates that the title compounds have strong potential
19

Medicinal Chemistry Communications
Page 20 of 44
View Article Online
DOI: 10.1039/C3MD00357D
to bind at the interface of α,β-tubulin heterodimer, which in turn destabilize/inhibit the tubulin
polymerization as observed in the experimental study.
Table 4 Comparison of molecular docking results (Glide score and non-bonding interactions) of
known tubulin inhibitors colchicine and CA-4 with the potent investigated compounds (sorted
according to Glide score). Value 2 in parenthesis of column 4 indicates the bifurcated hydrogen
bond.
Compound
Glide
Score
Hydrogen Bonding
interaction
Hydrophobic
interaction
C-H· · ·π
interaction
N-H· · ·π
interaction
Backbone
Valα181
Side chain
-10.19
-10.02
-9.33
-8.82
-8.59
-8.38
Cysβ241(2)
Alaα180,Leuβ242,
Alaβ250 Alaβ316,
Ileβ318, Ileβ378
Alaα180,Valα181,
Alaβ250, Alaβ316,
Ileβ318, Ileβ378
Alaα180,Valα181,
Alaβ250, Alaβ316,
Ileβ318, Ileβ378
Alaα180,Valα181,
Alaβ250, Alaβ316,
Ileβ318, Ileβ378
Alaα180,Valα181,
Alaβ250, Alaβ316,
Ileβ318, Ileβ378
Alaα180,Valα181,
Alaα180, Leuβ242,
Alaβ250, Alaβ316,
Ileβ318, Ileβ378
Leuβ255
Leuβ255
Leuβ255
Leuβ255
Leuβ255
Leuβ255
Asnβ258
Asnβ258
Asnβ258
Asnβ258
Asnβ258
Asnβ258
Colchicine
(Charge 0)
Thrα179 (2.41
Å)
Gluα183
Cysβ241(2)
2
(Charge +1)
Thrα179 (2.63
Å)
Gluα183
Cysβ241(2)
1
(Charge +1)
Thrα179 (2.85
Å), Asnβ249
Cysβ241(2)
Cysβ241(2)
Cysβ241(2)
14
(Charge +1)
Thrα179 (3.74
Å)
15
(Charge +1)
Valα181
CA-4
(Charge 0)
20

Page 21 of 44
Medicinal Chemistry Communications
View Article Online
DOI: 10.1039/C3MD00357D
Fig. 7 Binding model of ligands at the interface of α,β-tubulin heterodimer. Compound 2 is
shown in stick models (magenta color). For clarity non-polar hydrogen atoms have been deleted
from the ligands. Hydrogen bonding interactions are shown as black dashes, and residues
involved are represented also in stick models, while residues involved in other intermolecular
interactions are represented as lines. Green and cyan color represents α- and β-tubulin,
respectively.
Molecular Dynamics (MD) Simulation Analysis
In the previously reported studies, MD simulations were implemented to refine the tubulin-
inhibitor complex,80 calculation of potential energy of tubulin inhibitors in order to identify the
correct binding pose,79 and stability of interactions during the dynamical process to distinguish
active from inactive ligand.86 In another study, constrained molecular dynamics simulation and
hydropathy analysis were performed to investigate the reason behind the lower and higher
activity of the ligands.66 In comparison to these MD studies, which run simulation for a few
hundreds of picoseconds, the present simulation was performed for long run (10 ns) to get more
detailed insight into the stability of α,β-tubulin-ligand complex and their intermolecular
interactions (special emphasis on Cysβ241) in dynamical conditions, and to estimate the binding
free energy of ligands. All the full atomistic MD simulations were performed using AMBER 11
package.87 For the purpose of MD simulations, best docking pose of the four potent investigated
21

Medicinal Chemistry Communications
Page 22 of 44
View Article Online
DOI: 10.1039/C3MD00357D
compounds (1, 2, 14, and 15 in their protonated states) and known tubulin inhibitor CA-4 with
α,β-tubulin heterodimer were selected. Root-mean-square deviation (RMSD) analysis of Cα
backbone atoms of protein and all the atoms of ligand throughout the simulation run are provided
in the supporting information (Fig. S11).
Hydrogen bond analysis
Hydrogen bonding analysis between protein and ligand was carried out to investigate the
stability of these intermolecular interactions under dynamical conditions and to identify the
predominant protein residues involved in such type of contacts. Hydrogen bond formation ability
of potent compounds (1, 2, 14, and 15) and CA-4 analyzed during the last 2 ns of total simulation
run is presented in Fig. S12. There is a clear distinction in the number of hydrogen bonds formed
by CA-4 and the potent compounds with the residues of α,β-tubulin heterodimer in each frame.
CA-4 formed only one persistent hydrogen bond throughout the trajectory, while potent
investigated compounds showed 1-3 hydrogen bonds. This exemplify that the incorporation of
additional pharmacophores (bridging NH and imidazo-pyridine/pyrazine moiety) into CA-4
resulted in enhanced intermolecular interactions.
The contribution of each amino acid residues involved in the hydrogen bonding during
the last 2 ns trajectory (1000 frames) was also assessed in terms of their occupancy (Fig. 10).
The maximum hydrogen bond occupancy in protein complexed with CA-4 was shown by
hydrophobic residue Valβ315 (9.90%), this residue interacts with the –OH group at ring B of the
ligand. In case of protein complexed with potent compounds 1, 2, 14, and 15 the highest
occupancy was exhibited by residues Serα178 (74.20%), Thrα179 (39.60%), Leuβ248 (37.60%)
and Asnβ249 (26.40%), respectively. It was observed that the protonated nitrogen at imidazole
ring of the compounds 1, 2 and 15 formed hydrogen bonding interactions with the respective
22

Page 23 of 44
Medicinal Chemistry Communications
View Article Online
DOI: 10.1039/C3MD00357D
residue showing highest occupancy, whereas in compound 14, indole ring –NH interacted with
the residue exhibiting maximum occupancy. Interestingly, in all the protein-ligand complexes,
hydrogen bonding was observed between the side chain -SH of Cysβ241 (critical residue for
antitubulin activity) of a protein and oxygen of methoxy groups at ring A of the ligands. The
hydrogen bond occupancy of Cysβ241 for protein complexed with ligands CA-4, 1, 2, 14, and 15
was found to be 1.70%, 4.60%, 4.60%, 6.40% and 6.70%, respectively. Although, Cysβ241
represent low occupancy (due to stringent criterion of hydrogen bonding) in all the cases but it
was always in close proximity to the two of the methoxy oxygens. Being a critical residue in the
interaction of tubulin and inhibitors, Cysβ241was analysed for the proximity to methoxy groups
of ligands (see supporting information for detailed analysis, Fig. S13). The results indicate that
side chain of Cysβ241 is in a distance range close enough to form hydrogen bonds with two of
the three ring A methoxy groups.
Binding free energy calculations
The length of a simulation run plays a significant role to obtain the reliable binding free energy
estimate of the protein-ligand complex.88 Therefore, to observe the fluctuations in the binding
free energy during the whole simulation run, calculations were initially performed on 100
snapshots of each 1 ns trajectory until 10 ns, using Molecular Mechanics-Generalized Born
Surface Area (MM-GBSA) method.89 The binding free energy with respect to simulation time
for the known tubulin inhibitor CA-4 and potent investigated compounds (1, 2, 14, and 15) with
α,β-tubulin heterodimer are depicted in Fig. S14. It is clearly reflected from the graph that
binding free energy gets equilibrated during the last 2 ns of the simulation run i.e. from 9-10 ns.
This suggested that simulation time was long enough to study the thermodynamics and energetic
contribution of various intermolecular forces within a system.
23

Medicinal Chemistry Communications
Page 24 of 44
View Article Online
DOI: 10.1039/C3MD00357D
Based on these results, average binding free energy of ligands for the last 2 ns trajectory
is reported in this article (Table 5) along with the contribution of different energy components
(van der Waals, electrostatic, polar solvation and nonpolar solvation interaction energies). The
binding free energy for the ligands CA-4, 1, 2, 14, and 15 via MM-GBSA was predicted to be -
37.57, -40.51, -34.24, -30.95 and -36.87 kcal/mol, respectively. Therefore, in accordance with
the ex vivo tubulin polymerization inhibitory activities (similar IC50) and molecular docking
results (similar Glide score), the binding free energies of the potent compounds are comparable
to that of CA-4. Total relative binding free energy (ꢀGbind) equals to the sum of gas phase energy
(ꢀGgas) and solvation free energy (ꢀGsolv). In case of CA-4, gas phase and solvation free energies
were found to be -62.78 and 25.21 kcal/mol, respectively, signifying that the former has
favourable and later has unfavourable energetic contribution to the total binding free energy.
Likewise, in all the other potent investigated compounds, gas phase energy was favourable (-546
to -596 kcal/mol) while solvation free energy was unfavourable (510 to 562 kcal/mol).
It can also be observed from Table 5 that van der Waals energy of potent compounds (1,
2, 14, and 15) is more (about 10-15 kcal/mol) as compared to CA-4, this could be due to the
presence of additional hydrophobic imidazo-pyridine/pyrazine ring in the compounds. In
addition, electrostatic energy component of compounds (-488 to -532 kcal/mol) was much higher
in comparison to that of CA-4 (-14.22 kcal/mol). This difference is attributed to the presence of
protonated nitrogen in imidazole ring of compounds, which makes them positively charged and
hence can interact with the negatively charged residues of the colchicine binding site. It should
also be noted that the formal charge of the α,β-tubulin heterodimer along with GTP is -38.00,
implying that the protein is majorly composed of negatively charged residues. Another reason
behind the more favourable electrostatic contribution in case of potent compounds than CA-4
24

Page 25 of 44
Medicinal Chemistry Communications
View Article Online
DOI: 10.1039/C3MD00357D
may be because of higher hydrogen bond formation ability of former with the residues of protein
as compared to later (Fig. S12). Moreover, in all the potent compounds electrostatic energy
component is predominant than van der Waals contribution to the overall binding energy, while
in CA-4 van der Waals energy is the main driving force behind the stabilization of protein-ligand
complex.
Table 5 Average binding free energy results (last 2 ns) for α,β-tubulin-ligand complexes along
with its different energy components.a
Compound
VDW
-48.56
-57.10
-64.60
-59.98
-57.81
EEL
EGB ESURF
ꢀGgas
-62.78
ꢀGsolv
25.21
ꢀGbind
-14.22
30.77
517.60
569.52
520.14
515.81
-5.57
-6.34
-7.22
-6.90
-6.23
-37.57 2.62
-40.51 4.21
-34.24 4.20
-30.95 3.33
-36.87 3.48
CA-4
1
-494.67
-531.94
-484.21
-488.65
-551.77
-596.54
-544.19
-546.46
511.26
562.30
513.24
509.58
2
14
15
a The meaning of the different terms used in this table is as follows: VDW = van der Waals energy as calculated by
the MM force field. EEL = electrostatic energy as calculated by the MM force field. EGB = the electrostatic
contribution to the solvation free energy calculated by GB. ESURF = nonpolar contribution to the solvation free
energy calculated by an empirical model. ꢀGgas = total gas phase energy i.e. sum of van der Waals and electrostatic
energy from MM. ꢀGsolv = total solvation free energy i.e. sum of electrostatic and nonpolar contributions from
solvation. ꢀGbind = final estimated binding free energy calculated from the terms above (kcal/mol).
25

Medicinal Chemistry Communications
Page 26 of 44
View Article Online
DOI: 10.1039/C3MD00357D
Conclusions
In conclusion, a series of combretastatin A-4 inspired novel 2-aryl-3-arylamino-imidazo-
pyridines/pyrazines were designed. They were synthesized by a one-pot approach involving the
preparation of isocyanides directly from amines via formylation and subsequent dehydration
followed by their reaction with heterocyclic-2-amidines and aldehydes. Ex vivo tubulin
polymerisation assay and in vitro immunostaining experiment revealed that among the
investigated compounds (1-20), compounds 1, 2, 14, and 15 were found to possess significant
activities for inhibition of tubulin polymerization and disruption of tubulin-microtubule
dynamics, comparable to that of CA-4. MTT and clonogenic assays showed that these
compounds exhibited potent anticancer activities in various cancer cell lines (kidney, breast, and
cervical), and relatively low toxicity to normal cells, compared to CA-4. Various biophysical and
immunological studies including DAPI nuclear staining, expression of representative apoptotic
protein markers, immunocytochemistry, comet assay, and cytokinesis-block micronucleus assay
implied their good efficacies in apoptosis, and DNA as well as chromosomal damage in HEK
293T cells. Furthermore, flow cytometry indicated that the compounds induced cell-arrest in
G2/M phase of cell cycle. Molecular docking and MD simulation studies suggest that these
compounds occupied the colchicine binding site at the interface of α,β-tubulin heterodimer,
similar to that of CA-4, and stabilized the complex by forming multiple interactions with the key
residues (including the important molecular recognition residue Cysβ241) of the protein.
Moreover, the compounds showed docking score and binding free energy values comparable to
that of CA-4. Interestingly, molecular modeling studies revealed that in comparison to CA-4, the
bridging NH and the imidazo-pyridine/pyrazine ring in the title compounds provided additional
pharmacophoric features for the inhibition of tubulin polymerization. The bridging NH group
26

Page 27 of 44
Medicinal Chemistry Communications
View Article Online
DOI: 10.1039/C3MD00357D
afforded the flexibility in attaining the required dihedral angel between two aryl moieties for
maximum binding with tubulin. The protonated nitrogen (N1) of imidazo-pyridine/pyrazine ring
showed hydrogen bonding, and also the ring was found to be involved in NH· · ·π and
hydrophobic interactions.
In overall, the results of tubulin polymerization inhibition, disruption of microtubule
networks, cytotoxic activities, relative potencies, and cell cycle arrest in G2/M phase were found
to be in accordance, and supported by in silico results of mode of interaction with tubulin. These
suggested the inhibition of tubulin polymerization via interacting on colchicine binding site as
mode of action for exhibiting cytotoxic and anti-proliferative properties of the investigated
compounds. This work will provide impetus in rational structural modulation by considering the
important pharmacophoric features of the lead compounds for the identification of novel
anticancer agents that can target at colchicine binding site of tubulin. The relevant studies of the
investigated potent compounds for their further development are underway.
27

Medicinal Chemistry Communications
Page 28 of 44
View Article Online
DOI: 10.1039/C3MD00357D
Experimental Section
Chemistry
General considerations: All reactants and reagents were obtained from commercial source and
used without further purification except solvent. The NMR spectra were recorded on a Bruker
Avance DPX 400 MHz spectrometer in CDCl3/DMSO–d6/CD3OD solvents using TMS as an
internal standard. J values are given in Hz. Mass spectra were recorded with LCMS–MAT–LCQ
[for ESI and APCI], Bruker–Maxis [for HRMS] mass spectrometers. IR spectra were recorded
on a Nicolet FT–IR Impact 410 instrument as thin film (neat). The reactions were monitored by
TLC (Merck®, Silica gel 60 F254). The purity of synthesized compounds was analyzed by HPLC
(Shimadzu LC–6AD system), Phenomenex RP–C18 column (250 x 4.60 mm), particle size 5
µm, flow rate 1mL/min, using water-acetonitrile.
Synthesis of 2-Aryl-3-arylamino-imidazo-pyridine/pyrazine derivatives (1-20)
Representative synthesis of 2-(3-Hydroxy-4-methoxyphenyl)-3-(3,4,5-trimethoxyphenyl-
amino)imidazo[1,2-a]pyridine (compound 1)
A mixture of trimethoxyaniline (1 mmol) and formic acid (1.5 mmol) was stirred at 80 °C. After
completion of reaction (2.5 h), the resultant mixture was cooled to 0 °C and the assembly was
flushed with nitrogen. Anhyd. acetonitrile (1.5 mL), DABCO (2.5 mmol, 280 mg) and p-TsCl
(2.5 mmol, 475 mg) were subsequently added to the mixture at 0 °C under nitrogen. The reaction
temperature was allowed to gradually elevate to RT (25–28 oC) by removing the ice bath and the
reaction was continued at RT. To realize the progress of conversion of N-formamide to
isocyanide, a TLC-monitoring process that can circumvent the pungent smell of isocyanide was
performed. A little portion of the reaction mixture was quenched with saturated aqueous solution
of CuSO4. The aqueous solution was mini-extracted with EtOAc in a vial and the EtOAc solution
28

Page 29 of 44
Medicinal Chemistry Communications
View Article Online
DOI: 10.1039/C3MD00357D
was checked for TLC. After complete conversion of N-formamide to isocyanide as indicated by
TLC (4 h), 2-aminopyridine (1 mmol) and 3-hydroxy-4-methoxybenzaldehyde (1 mmol) were
added. The reaction mixture was heated at 50 °C. The resultant mixture after completion of the
reaction as indicated by TLC (16 h) was extracted with EtOAc (2x25 mL). The combined
organic solution was washed with water (2x5 mL) and brine (1x5 mL), dried with anhyd.
Na2SO4, and concentrated under reduced pressure. The column chromatographic purification of
crude mass on silica gel eluting with EtOAc-hexane provided 2-(3-Hydroxy-4-methoxyphenyl)-
3-(3,4,5-trimethoxyphenylamino)imidazo[1,2-a]pyridine in 48% isolated yield.
(202 mg, 48%); Green solid; mp 170-172 °C; IR (KBr) max/cm-1 3310, 2933, 1607, 1504, 1438,
1231, 1010; δH (400 MHz; DMSO-d6; Me4Si) 3.63 (6 H, s), 3.67 (3 H, s), 3.87 (3 H, s), 5.87 (2
H, s), 6.92-6.97 (2 H, m), 7.36 (1 H, d, J 7.3), 7.46 (1 H, dd, J 1.8, J 8.4 ), 7.51 (1 H, d, J 1.8 ),
7.57 (1 H, d, J 9.0), 7.98 (1 H, d, J 6.8); δC(100 MHz, DMSO-d6; Me4Si) 56.3, 56.4,
61.3(OCH3), 91.7, 112.6, 113.7, 115.2, 117.2, 119.8, 120.2, 124.3, 127.1, 127.3, 131.8, 139.7,
143.7, 143.8, 147.5, 149.2, 155.5; MS (APCI) m/z 422 (M+H+); HRMS (ESI) Calcd for
C23H23N3O5Na [M + Na]+ 444.1537, found m/z 444.1534.
All other compounds (2-20) were prepared following this reaction procedure. The structures of
compounds were characterized by 1H and 13C NMR, IR and Mass (ESI or APCI) spectroscopies
and confirmed by HRMS (ESI). The purity of the investigated compounds in HPLC were found
to be >95%.
Biology
Cell line and culture conditions
The human embryonic kidney cancer cells (HEK 293T, cat. # CRL-11268), normal monkey
kidney cells (Vero, cat. # CCL-81), breast cancer cells (MCF 7, cat. #HTB-22) and cervical
29