Computational modeling and target synthesis of monomethoxy-substituted o-diphenylisoxazoles with unexpectedly high antimitotic microtubule destabilizing activity

Article abstract of DOI:10.1016/j.bmcl.2020.127608

The ability of monomethoxy-substituted o-diphenylisoxazoles 2a-d to interact with the colchicine site of tubulin was predicted using computational modeling, docking studies, and calculation of binding affinity. The respective molecules were synthesized in high yields by three steps reaction using easily available benzaldehydes, acetophenones, and arylnitromethanes as starting material. The calculated antitubulin effect was confirmed in vivo in a sea urchin embryo model. Compounds 2a and 2c showed high antimitotic microtubule destabilizing activity compared to that of CA4. Isoxazole 2a also exhibited significant cytotoxicity against human cancer cells in NCI60 screen. For the first time, isoxazole-linked CA4 derivatives 2a and 2c with only one methoxy substituent were identified as potent antimitotic microtubule destabilizing agents. These molecules could be considered as promising structures for further optimization.

Full text of DOI:10.1016/j.bmcl.2020.127608

Bioorganic&MedicinalChemistryLetters30(2020)127608
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Computational modeling and target synthesis of monomethoxy-substituted
o-diphenylisoxazoles with unexpectedly high antimitotic microtubule
destabilizing activity
⁎
Victor S. Stroylova,b, Igor V. Svitankoa,b, Anna S. Maksimenkoa, , Victor P. Kislyia,
Marina N. Semenovac, Victor V. Semenova
a N. D. Zelinsky Institute of Organic Chemistry RAS, 47 Leninsky Prospect, 119991 Moscow, Russian Federation
b National Research University Higher School of Economics (HSE), 20 Myasnitskaya Street, 101000 Moscow, Russian Federation
c N. K. Kol’tsov Institute of Developmental Biology RAS, 26 Vavilov Street, 119334 Moscow, Russian Federation
A R T I C L E I N F O
A B S T R A C T
Keywords:
The ability of monomethoxy-substituted o-diphenylisoxazoles 2a-d to interact with the colchicine site of tubulin
was predicted using computational modeling, docking studies, and calculation of binding affinity. The respective
molecules were synthesized in high yields by three steps reaction using easily available benzaldehydes, acet-
ophenones, and arylnitromethanes as starting material. The calculated antitubulin effect was confirmed in vivo in
a sea urchin embryo model. Compounds 2a and 2c showed high antimitotic microtubule destabilizing activity
compared to that of CA4. Isoxazole 2a also exhibited significant cytotoxicity against human cancer cells in NCI60
screen. For the first time, isoxazole-linked CA4 derivatives 2a and 2c with only one methoxy substituent were
identified as potent antimitotic microtubule destabilizing agents. These molecules could be considered as pro-
mising structures for further optimization.
o-Diarylisoxazoles
Computational modeling
Docking
Microtubule destabilization
Sea urchin embryo
A study of relationship between molecular structure and biological
activity of a compound (structure–activity relationship, SAR) is an
important methodological approach in drug discovery, allowing for
identification of structural fragments, pharmacophores, which are re-
sponsible for a particular biological effect. Virtual screening using
molecular docking is one of the computational techniques in SAR stu-
dies, providing data about spatial configuration of a molecule required
for the optimal binding to an active site of a target protein.1 Selective
targeting tubulin and microtubules is a well validated approach for
chemotherapy of malignancies.2,3 A large number of small molecule
ligands interact with one of four known specific binding sites in tubulin
molecule: taxane/epothilone, laulimalide, vinca alkaloid, and colchi-
cine sites.4 Several potent natural antimitotics that inhibit tubulin
polymerization by interacting with colchicine binding site, such as
colchicine, combretastatin A-4 (CA4), steganacin, and podophyllotoxin,
feature trimethoxyphenyl functionality, an important pharmacophore
for occupation of the relevant space in tubulin molecule (Fig. 1).4–6
Numerous SAR studies suggest that the presence of trimethoxyphenyl
ring in a compound structure can serve as an evidence of antitubulin
effect. For example, known potent type 1 bone morphogenetic protein
receptor inhibitor K02288 (Fig. 1),7 showed strong antimitotic micro-
tubule destabilizing activity (Table 1).8 Similarly, the 3,4,5-trimethox-
yphenyl moiety of CA4 (Fig. 1), a stilbene originally isolated from the
bark of African willow tree Combretum caffrum,9 is considered to be
fundamental for antimitotic microtubule destabilizing potential.10,11 As
an evidence, monomethoxy-substituted CA4 analogue, 1-(4-methox-
yphenyl)-2-phenylethene (MPE, Fig. 1), was not cytotoxic against B16
murine melanoma cells,12 exhibited five order of magnitude less cyto-
toxicity against K562 human leukemia cell line than the parent CA4,
and did not inhibit in vitro tubulin polymerization.13 The removal of m-
or p-methoxy groups from CA4 ring A markedly reduced cytotoxicity as
well, although the ability to inhibit purified tubulin polymerization was
much less affected.14 CA4 derivatives with the trifluoro-substituted ring
A retained cytotoxicity and inhibition of tubulin polymerization, but
were unable to cause colchicine displacement from the binding site of
tubulin, whereas compounds with the ethoxy- and methyl-substituted A
ring were able to interact with colchicine site of tubulin.13,15
Along with the 3,4,5-trimethoxy-substituted A-ring, a cis-config-
uration of CA4 biaryl scaffold is essential for efficient coupling with the
colchicine binding site of tubulin.11,16–18 To avoid the undesirable
⁎ Corresponding author.
E-mail addresses: victor.stroylov@gmail.com (V.S. Stroylov), svitanko@mail.ru (I.V. Svitanko), chem.annamaks@gmail.com (A.S. Maksimenko),
vkislyi@yandex.ru (V.P. Kislyi), ms@chemical-block.com (M.N. Semenova), vs@zelinsky.ru (V.V. Semenov).
https://doi.org/10.1016/j.bmcl.2020.127608
Received 28 July 2020; Accepted 4 October 2020
Availableonline07October2020
0960-894X/©2020ElsevierLtd.Allrightsreserved.

V.S. Stroylov, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127608
Fig. 1. Structures of antimitotics with trimethoxyphenyl moiety and o-diaryl isoxazoles.
cis–trans transformation, which leads to a significant decrease in anti-
mitotic antitubulin activity, diverse cis-restricted heterocyclic analo-
gues of CA4 have been synthesized. These compounds were reported to
retain effects comparable with those of parent CA4.11 Specifically, o-
diphenylisoxazoles showed both high antimitotic microtubule destabi-
lizing activity and significant cytotoxicity against human cancer cells,
and 3,4,5-trimethoxy/trialkoxy-substituted benzene ring was identified
as essential for these effects.19–21 However, in the sea urchin embryo
assay mono-substituted 3-phenyl-4-methoxyphenylisoxazole 2a (Fig. 1)
was determined unexpectedly as a more potent microtubule destabilizer
evolved from exotic and expensive alchemical calculations to an useful
and affordable tool for predicting binding affinity in complex biological
systems via the most rigorous physical approach. In the present study,
we sought to evaluate the potential of MD and FEP/MD methods to
predict the affinity of new tubulin polymerization inhibitors.
Molecular docking simulations placed all four isomeric molecules
into colchicine binding site consistently with experimental colchicine
binding in a way that mono methoxy-substituted benzene ring was
bound as ring B in corresponding colchicine position.24 Accordingly,
the position of ring A was occupied by an unsubstituted phenyl (Fig. 2).
In regard to binding free energies predicted by docking, they were ex-
pected to be almost the same and amounted to approximately −7 kcal/
mol (Ki ~ 7.5 μM; hereinafter calculated from dG = -RTlnKi). For
comparison, the same docking procedure estimated CA4 and compound
1 binding affinity as −8.7 and −8.1 kcal/mol (0.4 and 1.1 μM, re-
spectively). No hydrogen bonds were formed in binding poses predicted
by docking, and as a rule of thumb it means poor binding. Another
possibility to be investigated was the formation of water bridges be-
tween protein and ligand.
than
3-(4-methoxyphenyl)-4-(3,4,5-trimethoxyphenyl)isoxazole
1
(Fig. 1),20 which was inconsistent with generally accepted SAR results.
This observation prompted us to analyze the capability of the corre-
sponding three monomethoxy-substituted o-diphenylisoxazole isomers
2b-d (Fig. 1) to affect tubulin. First, molecular docking with subsequent
molecular dynamic simulation of compounds 2a-d was performed in
order to study the interaction of the compounds with colchicine binding
site of tubulin, including the impact of methoxy-substituted benzene
ring topology in relation to atoms O and N in isoxazole heterocycle.
Next, a target synthesis of the respective o-diphenylisoxazoles was
carried out followed by in vivo biological evaluation using a sea urchin
embryo model.
Indeed, molecular dynamics simulations have revealed that com-
pound 2a formed relatively strong hydrogen bond with backbone car-
bonyl oxygen of Asp251 residue through a water molecule, and this
state was populated around 50% of simulation time (Fig. 3). This in-
teraction was also accompanied by another water molecule bound to
Asn249 backbone oxygen atom (around 20% of simulation time). These
interactions provided a good structural basis to explain the unusual
potency of compound 2a.
Computational modeling
Initial computer modeling of active structures with their subsequent
target synthesis can be considered as a promising strategy for the ap-
proach to biologically active compounds. In the present study the mo-
lecular-mechanical simulation for target synthesis of compounds with
the required properties was applied using modern methods, which de-
monstrated the best performance in various tests.22 To accelerate pro-
cess of hit and lead discovery, a method was required that yields a quick
and accurate answer to the question: given a chemical structure, what
affinity would it have to the target? In recent decades, free energy
perturbation (FEP) calculations based on using molecular dynamics
(MD) methodology23 with full-atom models and explicit solvation have
In case of compounds 2b and 2d no such interactions were detected,
which strongly indicated a significantly less potency. However, the
most interesting case was with compound 2c. Lacking any significant
solvation of isoxazole nitrogen, it demonstrated highly stable hydrogen
bond that involved oxygen atom from methoxy substituent, residues
Asn349, Lys352, and a water molecule that formed two hydrogen bonds
with protein and another hydrogen bond with the ligand (Fig. 4). These
hydrogen bonds did not always co-exist: the bond with Asn349 back-
bone oxygen was strong and occupies around 70% of temporal domain,
2

V.S. Stroylov, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127608
Fig. 2. Molecular docking studies of 2a-d with colchicine binding site in tubulin (PDB ID 1SA0). Experimental binding pose for tubulin is shown in red. Binding poses
of compounds 2a-d are shown in licorice and colored according to atom type.
whereas the bond with Lys352 side chain was understandably weaker
(only approximately 25%). These observations suggested that com-
pound 2c might be a potent binder as well.
energy in ligand/protein system, useful results can still be obtained
using their combination. Combined application of docking, MD, and
FEP provide a reasonable approach to directed organic synthesis,
especially to ligand optimization, and deserve wider application. The
results of computational modeling allowed us to assume the presence of
potent structures among four examined diarylisoxazoles. For experi-
mental confirmation of this hypothesis, a synthetic protocol was de-
veloped to afford the respective isomers 2b-d.
To compare, CA4 and compound 1, despite their similar starting
binding pose, behaved slightly different. Both molecules have buried
their trimethoxyphenyl moiety inside the protein without forming any
polar interactions. In case of CA4, an extensive hydrogen bond network
was formed by ring B which involved two water-mediated and one
direct hydrogen bonds with total occupancy of 200%. Compound 1
formed only one water-mediated hydrogen bond of 50% occupancy,
suggesting less activity than that of CA4 (Fig. S1, S2, Supplementary
Data).
Chemistry. Target synthesis
Regioselective preparation of 5-unsubstituted diarylisoxazoles is
challenging due to formation of a mixture of isomers. Recently a
modified simple synthetic pathway for the single isomer synthesis was
published,20 which included three main stages (Scheme 1). Initially,
condensation of arylnitromethanes 3a,b with Schiff bases 4a,b under
slightly acidic conditions yielded nitrostilbenes 5a,b. Importantly, the
use of Schiff bases instead of aldehydes duplicated the yields of the
intermediate nitrostilbenes 5. Subsequent reaction with ethox-
ycarbonylmethylpyridinium bromide afforded isoxazoline-N-oxides 6,
which were transformed to the target 5-unsubstituted isoxazoles 2a and
2b through recyclization and decarboxylation steps.20
To assess binding in a quantitative way we have resorted to FEP
calculations25 to estimate what free binding energy difference we might
expect from compound 2a-d (Table 1). Calculated and experimentally
determined values for compound 1 and CA4 as reference were in fairly
good agreement. Compounds 2b and 2d were of 2–3 kcal/mol less fa-
vorable than 2a, which was consistent with observations made from
MD trajectories (Fig. S3-S6, Supplementary Data). In case of compound
2c its potency underestimated by 1.7 kcal/mol, however, this was most
likely the result of different time frames employed in pure MD and FEP/
MD workflows. In pure MD, simulation time of 100 ns was sufficient for
any conformational changes to happen in ligand/protein system,
whereas in FEP/MD protocol only 2 ns timeframe was used, and protein
was additionally restrained to improve convergence.
More available diarylisoxazoles 2c and 2d were obtained via three-
stage synthetic route using simple reagents and reactants (Scheme 2).
Condensation of acetophenones 7 with benzaldehydes 8 yielded
chalkones 9, the subsequent rearrangement of which under addition of
diacetoxyiodobenzene (DIB) in acidic media with the 1,2-shift of aryl-
ring afforded acetals 10.28 Reaction of acetals 10c,d with hydro-
xylamine led to isoxazoles 2c,d with high yields.
To conclude, the calculations have resulted in the following com-
pound activity order: 2a > 2c > 2b > 2d. The data obtained during
current computational investigation suggest that while no single
method could provide substantial accuracy in predicting binding free
3

V.S. Stroylov, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127608
Fig. 3. Snapshot from molecular dynamics simulation of compound 2a bound to tubulin. A system of hydrogen bonds was formed that involved the ligand, two water
molecules, and protein residues Asn249 and Asp251.
Biological evaluation
timeframe was used, and protein was additionally restrained to improve
convergence. Interchange of position of atoms O and N in isoxazole ring
(2a versus 2c) only slightly affected the activity. On the contrary,
comparison of effects for diarylisoxazole pairs featuring interchanged
rings A and B (2a versus 2b and 2c versus 2d) revealed significant ac-
tivity decrease in molecules with 4-methoxyphenyl moiety position
close to the heteroatom of the isoxazole ring. Noteworthy, in the sea
urchin embryo assay monomethoxy-substituted isoxazoles 2a and 2c
showed antimitotic activity higher than activity of trimethoxyphenyl-
containing natural products steganacin, podophyllotoxin, and colchi-
cine, and was comparable to that of CA4. In addition, compounds 2a
and 2c were more potent than their close analogue, trimethoxy-sub-
stituted compound 1. The higher activity of monosubstituted isoxazole
2a with the same position of phenyl rings as in compound 1 in relation
to isoxazole heteroatoms may be suggestive of the irrelevance of 3,5-
methoxy groups in the ring A and 4-methoxy group in the ring B,
particularly for 3,4-o-diphenylisoxazoles. The observed strong anti-
tubulin effect of monomethoxy-substituted o-diphenylisoxazoles to-
gether with an impact of a proper isoxazole ring orientation could be
explained as follows. The appropriate position of O and N atoms in
isoxazole linker in relation to methoxy-substituted phenyl ring fa-
cilitated the formation of an additional strong hydrogen bond in the
colchicine binding site, thereby compensating for the energy loss
caused by the lack of the 3,4,5-trimethoxyphenyl moiety.
The targeted diarylisoxazoles 2a–d were evaluated for antimitotic
microtubule destabilizing activity in a phenotypic sea urchin embryo
assay.21,26 First, the antimitotic activity could be easily determined by
cleavage alteration/arrest that was observed after the exposure of fer-
tilized eggs to a tested molecule. Second, after treatment of hatched
blastulas, a specific change of motility pattern, namely, embryo spin-
ning at the bottom of the culture vessel instead of forward swimming
near the surface of the seawater, clearly indicated the microtubule
destabilizing mode of action. The results are presented in Table 1 with
steganacin, podophyllotoxin, colchicine, and CA4 as a positive control.
All monomethoxy-substituted o-diphenylisoxazoles 2a-d exhibited
antimitotic activity. Embryo spinning after the exposure to compounds
2a-c served as an evidence of targeting tubulin and microtubule de-
stabilization. Isoxazole 2d could be considered as microtubule desta-
bilizer as well, owing to formation of tuberculate eggs characteristic for
microtubule destabilizing agents.21,26 The antimitotic antitubulin effect
of compounds decreased in the following order: 2c > 2a > 2b >
2d, which was correlated in general with calculated binding affinity of
these molecules to colchicine site of tubulin (Table 1). Calculated
higher potency of 2a in comparison with 2c might be a consequence of
the methodological error. Specifically, underestimated potency of 2c by
1.7 kcal/mol was most likely the result of different time frames em-
ployed in pure MD and FEP/MD workflows. In pure MD, simulation
time of 100 ns was sufficient for any conformational changes to happen
in ligand/protein system, whereas in FEP/MD protocol only 2 ns
o-Diarylisoxazole 2a with high antimitotic activity in the sea urchin
embryo model exhibited considerable cytotoxicity in NCI60 screening
with mean GI50 value of 0.457 μM for 60 human cancer cell lines
4

V.S. Stroylov, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127608
Fig. 4. Snapshot from molecular dynamics simulation of compound 2c bound to tubulin. A system of hydrogen bonds was formed that involves the ligand, protein
residues Asn349 and Lys352 and a water molecule.
(Supplementary data). MDA-MB-435 melanoma cell line was the most
sensitive to 2a (GI50 = 0.026 μM). However, in contrast to sea urchin
embryo assay data, in NCI60 cytotoxicity screen the compound was less
potent than 3,4,5-trimethoxyphenyl-containing analogue 1 (mean GI50
value of 0.17 μM20). Comparison of cytotoxicity of CA4 and its deri-
vatives against K562 leukemia cells generated noteworthy results.
Monomethoxy-substituted stilbene MPE (Fig. 1) exhibited five order of
magnitude less cytotoxicity (IC50 = 5 μM) than the parent CA4
Table 1
Tubulin binding parameters of o-diarylisoxazoles and their effects on sea urchin embryos.
a
Compd
dG (kcal-mol−1
)
Calculated binding affinity to tubulin (μM)
Sea urchin embryo effects, EC (μM)b
Calculated
Experimental
Cleavage alteration
Cleavage arrest
Embryo spinning
K02288c
NDd
NDd
NDd
0.1
1
1
Steganacinc
NDd
NDd
NDd
NDd
0.02
0.05
0.002
0.01f
0.005
0.2
0.3e
0.05
0.1
NDd
0.5
0.05
0.05
1f
Podophyllotoxinc
NDd
NDd
NDd
Colchicinec
NDd
NDd
NDd
CA4c
1
−11.35
−10.38
−10.96
−8.91
−9.78
−8.08
−11.88
−10.92
−11.33
−9.15
−11.88
−9.15
0.005
0.025
0.01
0.01
0.1f
0.05
2
2a
2b
2c
0.1
2
0.298
0.069
1.210
0.002
0.2
0.02
2 (TE)g
0.1
> 5
2d
a dG: Gibbs free energy of binding.
b The sea urchin embryo assay was conducted as described previously.21,26 Fertilized eggs and hatched blastulae were exposed to 2-fold decreasing concentrations of
compounds. Duplicate measurements showed no differences in effective threshold concentration (EC) values.
c Structures of K02288, steganacin, podophyllotoxin, cholchicine, and CA4 are presented in Fig. 1.
d ND: not determined.
e Data from27
.
f Data from20
.
gTE: tuberculate eggs typical of microtubule destabilizing agents.
5

V.S. Stroylov, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127608
Scheme 1. Synthesis of 3,4-diarylisoxazoles 2a,b.
Scheme 2. Synthesis of 4,5-diarylisoxazoles 2c,d.
(IC50 = 0.001 μM),13 which confirmed well-established value of the
3,4,5-trimethoxyphenyl pharmacophore. However, isoxazole derivative
of CA4, compound 1, was only a little more active than its mono-
methoxy-substituted analogue 2a (GI50 value of 0.035 μM20 and
0.047 μM (Supplementary data), respectively). Hence, the replacement
of ethene linker in MPE by isoxazole heterocycle (2a) enhanced cyto-
toxicity against K562 cells by ~ 100 times. It could be assumed that
monomethoxy-substituted isoxazole derivative of CA4 acquired struc-
tural features favorable for interaction with tubulin even in the absence
of the 3,4,5-trimethoxyphenyl moiety due to formation of additional
water-mediated hydrogen bonds between isoxazole heterocycle and
amino acid residues in binding site (Fig. S3, Supplementary Data).
comparable to that of the parent CA4. Effects of compound 2a, observed
in the sea urchin embryo model, were confirmed by its considerable
cytotoxicity in NCI60 screening against 60 human cancer cell lines.
Therefore, for the first time it was shown that the 3,4,5-trimethox-
yphenyl motif was not necessary for interaction with colchicine site of
tubulin. It was suggested that a certain relative position of the isoxazole
heterocycle and the p-methoxyphenyl ring provided structural hall-
marks for the formation of the additional strong hydrogen bond in the
colchicine binding site of tubulin, providing high binding affinity in the
absence of the 3,4,5-trimethoxyphenyl pharmacophore.
Thus, simple isomers of o-diphenylisoxazoles, 2a and 2c, synthe-
sized by a facile procedure could be considered as promising molecules
for further research and optimization.
Conclusions
Declaration of Competing Interest
Isoxazole linked CA4 analogues 2a-d without 3,4,5-trimethox-
yphenyl pharmacophore were identified as potent antimitotic micro-
tubule destabilizing agents. Their antitubulin effects were predicted
using computational modeling by MD and FEP/MD methods, docking
studies and calculated affinity with colchicine site of tubulin. The cor-
responding isomers were selectively synthesized with high yields using
easily available benzaldehydes, acetophenones, and arylnitromethanes.
The calculated antitubulin effects have been validated in vivo using the
phenotypic sea urchin embryo assay. Compounds 2a and 2c with the
unsubstituted phenyl ring position close to heteroatoms N or O of the
isoxazole linker showed antimitotic microtubule destabilizing activity
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
Acknowledgments
The authors gratefully acknowledge support from the Russian
Science Foundation Grant 18-13-00044. The work of MNS was con-
ducted under the IDB RAS Government basic research program in 2020
6

V.S. Stroylov, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127608
No. 0088-2019-0005.
trimethoxy unit? Org Biomol Chem. 2003;1(17):3033–3037. https://doi.org/10.
1039/B306878A.
We thank the National Cancer Institute (NCI) (Bethesda, MD, USA)
for screening compound 2a by the Developmental Therapeutics
Program at NCI (Anticancer Screening Program; http://dtp.cancer.gov).
We also wish to thank Dr. Yu. A. Strelenko from N. D. Zelinsky Institute
of Organic Chemistry RAS for the development of NMR spectra pre-
sentation software.
14. Cushman M, Nagarathnam D, Gopal D, He HM, Lin CM, Hamel E. Synthesis and
evaluation of analogs of (Z)-1-(4-methoxyphenyl)-2-(3,4,5-trimethoxyphenyl)ethene
as potential cytotoxic and antimitotic agents. J Med Chem. 1992;35(12):2293–2306.
https://doi.org/10.1021/jm00090a021.
15. Hall JJ, Sriram M, Strecker TE, et al. Design, synthesis, biochemical, and biological
evaluation of nitrogen-containing trifluoro structural modifications of com-
bretastatin A-4. Bioorg Med Chem Lett. 2008;18(18):5146–5149. https://doi.org/10.
1016/j.bmcl.2008.07.070.
16. Cushman M, Nagarathnam D, Gopal D, Chakraborti AK, Lin CM, Hamel E. Synthesis
and evaluation of stilbene and dihydrostilbene derivatives as potential anticancer
agents that inhibit tubulin polymerization. J Med Chem. 1991;34(8):2579–2588.
https://doi.org/10.1021/jm00112a036.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bmcl.2020.127608.
17. Pettit GR, Rhodes MR, Herald DL, Hamel E, Schmidt JM, Pettit RK. Antineoplastic
agents. 445. Synthesis and evaluation of structural modifications of (Z)- and (E)-
combretastatin A-4. J Med Chem. 2005;48(12):4087–4099. doi:10.1021/
jm0205797.
References
18. Gaspari R, Prota AE, Bargsten K, Cavalli A, Steinmetz MO. Structural basis of cis- and
trans-combretastatin binding to Tubulin. Chemistry. 2017;2(1):102–113. https://doi.
org/10.1016/j.chempr.2016.12.005.
1. Sliwoski G, Kothiwale S, Meiler J, Lowe EW. Computational Methods in Drug
Discovery, Barker EL, ed. Pharmacol Rev. 2014;66(1):334–395. https://doi.org/10.
1124/pr.112.007336.
19. Sun C-M, Lin L-G, Yu H-J, et al. Synthesis and cytotoxic activities of 4,5-diarylisox-
azoles. Bioorg Med Chem Lett. 2007;17(4):1078–1081. https://doi.org/10.1016/j.
bmcl.2006.11.023.
2. Dumontet C, Jordan MA. Microtubule-binding agents: a dynamic field of cancer
therapeutics. Nat Rev Drug Discov. 2010;9(10):790–803. https://doi.org/10.1038/
nrd3253.
20. Chernysheva NB, Maksimenko AS, Andreyanov FA, et al. Regioselective synthesis of
3,4-diaryl-5-unsubstituted isoxazoles, analogues of natural cytostatic combretastatin
A4. Eur J Med Chem. 2018;146:511–518. https://doi.org/10.1016/j.ejmech.2018.01.
070.
3. Duan Y, Liu W, Tian L, Mao Y, Song C. Targeting tubulin-colchicine site for cancer
therapy: inhibitors, antibody- drug conjugates and degradation agents. Curr Top Med
Chem. 2019;19(15):1289–1304. https://doi.org/10.2174/
21. Semenova MN, Demchuk DV, Tsyganov DV, et al. Sea urchin embryo model as a
reliable in vivo phenotypic screen to characterize selective antimitotic molecules.
Comparative evaluation of combretapyrazoles, -isoxazoles, -1,2,3-triazoles, and
-pyrroles as Tubulin-Binding Agents. ACS Comb Sci. 2018;20(12):700–721. https://
doi.org/10.1021/acscombsci.8b00113.
1568026619666190618130008.
4. Li L, Jiang S, Li X, Liu Y, Su J, Chen J. Recent advances in trimethoxyphenyl (TMP)
based tubulin inhibitors targeting the colchicine binding site. Eur J Med Chem.
2018;151:482–494. https://doi.org/10.1016/j.ejmech.2018.04.011.
5. Jordan A, Hadfield JA, Lawrence NJ, McGown AT. Tubulin as a target for anticancer
drugs: Agents which interact with the mitotic spindle. Med Res Rev.
1998;18(4):259–296 doi:10.1002/(SICI)1098-1128(199807)18:4 < 259::AID-
MED3 > 3.0.CO;2-U.
22. Stroganov OV, Novikov FN, Medvedev MG, et al. The role of human in the loop:
lessons from D3R challenge 4. J Comput Aided Mol Des. 2020;34(2):121–130. https://
doi.org/10.1007/s10822-020-00291-4.
23. Williams-Noonan BJ, Yuriev E, Chalmers DK. Free energy methods in drug design:
prospects of “alchemical perturbation” in medicinal chemistry. J Med Chem.
2018;61(3):638–649. https://doi.org/10.1021/acs.jmedchem.7b00681.
24. Ravelli RBG, Gigant B, Curmi PA, et al. Insight into tubulin regulation from a com-
plex with colchicine and a stathmin-like domain. Nature. 2004;428(6979):198–202.
https://doi.org/10.1038/nature02393.
6. Lu Y, Chen J, Xiao M, Li W, Miller DD. An overview of tubulin inhibitors that interact
with the colchicine binding site. Pharm Res. 2012;29(11):2943–2971. https://doi.
org/10.1007/s11095-012-0828-z.
7. https://www.medchemexpress.com/products.html.
8. Semenova MN. Personal Communications.
9. Pettit GR, Singh SB, Hamel E, Lin CM, Alberts DS, Garcia-Kendal D. Isolation and
structure of the strong cell growth and tubulin inhibitor combretastatin A-4.
Experientia. 1989;45(2):209–211. https://doi.org/10.1007/BF01954881.
10. Tron GC, Pirali T, Sorba G, Pagliai F, Busacca S, Genazzani AA. Medicinal chemistry
of combretastatin A4: present and future directions. J Med Chem.
2006;49(11):3033–3044. https://doi.org/10.1021/jm0512903.
11. Mikstacka R, Stefański T, Różański J. Tubulin-interactive stilbene derivatives as
anticancer agents. Cell Mol Biol Lett. 2013;18(3):368–397. https://doi.org/10.2478/
s11658-013-0094-z.
25. Zeifman AA, Stroylov VV, Novikov FN, Stroganov OV, Kulkov V, Chilov GG.
Alchemical FEP calculations of ligand conformer focusing in explicit solvent. J Chem
Theory Comput. 2013;9(2):1093–1102. https://doi.org/10.1021/ct300796g.
26. Semenova MN, Kiselyov A, Semenov VV. Sea urchin embryo as a model organism for
the rapid functional screening of tubulin modulators. Biotechniques.
2006;40(6):765–774. https://doi.org/10.2144/000112193.
27. Wang R-W-J, Rebhun LI, Kupchan MS. Antimitotic and antitubulin activity of the
tumor inhibitor steganacin. Cancer Res. 1977;37:3071–3079.
28. Tsyganov DV, Semenova MN, Konyushkin LD, Ushkarov VI, Raihstat MM, Semenov
VV. A convenient synthesis of cis-restricted combretastatin analogues with pyrazole
and isoxazole cores. Mendeleev Commun. 2019;29(2):163–165. https://doi.org/10.
1016/j.mencom.2019.03.015.
12. Pati HN, Wicks M, Holt HLJ, et al. Synthesis and biological evaluation of cis-com-
bretastatin analogs and their novel 1,2,3-triazole derivatives. Heterocycl Commun.
2005;11(2):117–120. https://doi.org/10.1515/HC.2005.11.2.117.
13. Gaukroger K, Hadfield JA, Lawrence NJ, Nolan S, McGown AT. Structural require-
ments for the interaction of combretastatins with tubulin: how important is the
7