Combretastatin A-4 and Derivatives: Potential Fungicides Targeting Fungal Tubulin

Article abstract of DOI:10.1021/acs.jafc.5b05119

Combretastatin A-4, first isolated from the African willow tree Combretum caffrum, is a tubulin polymerization inhibitor in medicine. It was first postulated as a potential fungicide targeting fungal tubulin for plant disease control in this study. Combretastatin A-4 and its derivatives were synthesized and tested against Rhizoctonia solani and Pyricularia oryzae. Several compounds have EC₅₀ values similar to or better than that of isoprothiolane, which is widely used for rice disease control. Structure-activity relationship study indicated the the cis configuration and hydroxyl group in combretastatin A-4 are crucial to the antifungal effect. Molecular modeling indicated the binding sites of combretastatin A-4 and carbendazim on fungal tubulin are totally different. The bioactivity of combretastatin A-4 and its derivatives against carbendazim-resistant strains was demonstrated in this study. The results provide a new approach for fungicide discovery and fungicide resistance management.

Full text of DOI:10.1021/acs.jafc.5b05119

Article
pubs.acs.org/JAFC
Combretastatin A‑4 and Derivatives: Potential Fungicides Targeting
Fungal Tubulin
Zhong-lin Ma,† Xiao-jing Yan,§ Lei Zhao,† Jiu-jiu Zhou,† Wan Pang,† Zhen-peng Kai,*,†
and Fan-hong Wu*,†,#
†School of Chemical and Environmental Engineering, Shanghai Institute of Technology, Shanghai 201418, People’s Republic of China
§Institute of Plant Protection, China Academy of Agricultural Sciences, Beijing 100193, People’s Republic of China
#Shanghai Engineering Research Center of Pharmaceutical Progress, Shanghai 201203, People’s Republic of China
ABSTRACT: Combretastatin A-4, first isolated from the African willow tree Combretum caffrum, is a tubulin polymerization
inhibitor in medicine. It was first postulated as a potential fungicide targeting fungal tubulin for plant disease control in this study.
Combretastatin A-4 and its derivatives were synthesized and tested against Rhizoctonia solani and Pyricularia oryzae. Several
compounds have EC50 values similar to or better than that of isoprothiolane, which is widely used for rice disease control.
Structure−activity relationship study indicated the the cis configuration and hydroxyl group in combretastatin A-4 are crucial to
the antifungal effect. Molecular modeling indicated the binding sites of combretastatin A-4 and carbendazim on fungal tubulin are
totally different. The bioactivity of combretastatin A-4 and its derivatives against carbendazim-resistant strains was demonstrated
in this study. The results provide a new approach for fungicide discovery and fungicide resistance management.
KEYWORDS: combretastatin, fungicide, resistance, tubulin, molecular modeling
tubulin. Herein we report the syntheses of combretastatin A-4
and its derivatives via the Wittig reaction and the evaluation of
their in vitro antifungal bioactivities. Molecular modeling was
performed to determine the binding site of combretastatin A-4.
The potential use of combretastatin A-4 and its derivatives for
fungal resistance management was also evaluated in this study.
INTRODUCTION
■
Fungi can cause serious crop diseases, resulting in critical losses
of yield and quality. Use of resistant varieties, biological control,
and chemical fungicides are the common strategies for managing
crop diseases. At present, chemical treatment with fungicides is
the most common method to control those diseases. However,
fungicide resistance has been detected in many fungal species
and causes unsatisfactory control efficacy. It is necessary to
develop new fungicides that have new structures and new
antifungal mechanisms to address this problem.
Combretastatin A-4, which was first isolated from the African
willow tree Combretum caffrum,1 was identified as an attractive
lead compound for the development of novel anticancer agents
that utilize starvation tactics to attack widespread necrosis of
solid tumors, including multidrug-resistant ones. It exerts a
potent inhibition of tubulin polymerization by binding to the
colchicine site and, as a consequence, demonstrates strong
suppressive activity on tumor blood flow (TBF).2−4
Microtubules are biopolymers that are composed of self-
associating α- and β-tubulin heterodimers. Because they play
an important role in a number of cellular processes, including
maintaining the structure of the cell, cell movement, intra-
cellular transportation, and cell division, microtubules are
attractive targets for drug and pesticide design.5 Carbendazim
and other benzimidazole fungicides, well-known β-tubulin inter-
ferers, have been widely used to control various plant diseases
for decades in China and in many other parts of the world.
However, these fungicides have generally lost their efficacy
because of the resistance after being used for 2 or 3 years.6
Resistance to benzimidazole fungicides usually resulted from
certain point mutations in the target β-tubulin gene.7
MATERIALS AND METHODS
■
General Experimental Procedures. All of the chemicals and
reagents were commercially available and required no further purifi-
cations. Solvents were dried and freshly distilled before use according
to literature procedures. Isoprothiolane (98% purity) was purchased
from Lianyungang Liben Agro-chemical Co., Ltd. (Liangyungang,
China). Carbendazim (98% purity) was purchased from Zhejiang
Yifan Chemical Industry Co., Ltd. (Wenzhou, China). Melting points
were determined using a WRS-2A melting point apparatus (Shanghai
ShenGuang Instrument Co., Ltd., Shanghai, China). TLC analyses
were performed on precoated silica gel polyester plates (Yantai Jiang
You Silicone Development Co., Ltd., Yantai, China) with a fluorescent
indicator UV 254. Chromatographic separations were performed on
silica gel flash columns. 1H NMR and 13C NMR spectra were recorded
on an AVANCE III at 500 MHz or on an Avance DMX500 spectro-
meter (Bruker, Fallanden, Switzerland) at 101 Hz in chloroform-d
̈
using TMS (δ 0.0 ppm) as an internal standard. IR was recorded on a
6700 FT-IR (Nicolet, Madison, WI, USA). HRMS were recorded on
a solariX 70 FT-MS spectrometer (Bruker) using methanol/water
(1:1, v/v) as solvent. LC-mass spectra were recorded on an LCMS-
2020 spectrometer from Shimadzu Corp. (Kyoto, Japan) with aceto-
nitrile and water as the mobile phase, and the gradient was from 5%
acetonitrile at 0 min to 100% acetonitrile at 10 min. The column used
Received: October 25, 2015
Revised: December 28, 2015
Accepted: December 29, 2015
Published: December 29, 2015
As combretastatin A-4 is a tubulin polymerization inhibitor
in medicine, we postulated it also has an effect on the fungal
© 2015 American Chemical Society
746
DOI: 10.1021/acs.jafc.5b05119
J. Agric. Food Chem. 2016, 64, 746−751

Journal of Agricultural and Food Chemistry
Article
Figure 1. Chemical structures of combretastatin A-4 and derivatives.
Figure 2. Synthesis of combretastatin A-4 derivatives.
was a 150 mm × 4.6 mm i.d., 5 μm, Gemini RP-18 (Phenomenex,
Torrance, CA, USA).
Molecular Modeling. The sequence of R. solani β-tubulin was
retrieved from GenBank using a combination of BLAST and keyword
searches. A crystal structure of cow brain β-tubulin (PDB ID 1Z2Bb)
was used as the 3D coordinate template for the homology modeling.
The homology model for R. solani β-tubulin was generated using
FUGUE10 and ORCHESTRAR module in Sybyl.11 The initial model
was optimized energetically using the minimize program with steepest
descent algorithm, AMBER7 FF99 as the force field and Gasteiger−
Huckel as the atomic point charges. The minimization was terminated
when the RMS gradient convergence criterion of 0.05 kcal/(mol·Å)
was reached.
(Z)-2-Methoxy-5-(3,4,5-trimethoxystyryl)]phenol (combretastatin
A-4) and (Z)-2-methoxy-5-(3,4,5-trimethoxystyryl)]phenyl disodium
phosphate (combretastatin A-4P) were synthesized following the
methods of Shen et al.8 and Pettit et al.1 Combretastatin A-4 derivatives
were synthesized with the methods previously reported.9
Bioassays. Antifungal effects of compounds against Rhizoctonia
solani Kuhn (rice sheath blight disease), Pyricularia oryzae Cav (rice
̈
blast disease), and carbendazim-resistant strains Fusarium oxysporum
Schl. f. sp. vasinfectum (Atk.) Snyd. & Hans (cotton Fusarium wilt)
and Sclerotinia sclerotiorum (Lib.) de Bary (rape Sclerotinia rot) were
evaluated using mycelium growth rate test. All fungal strains were
provided by the Institute of Plant Protection, Chinese Academy of
Agricultural Sciences. Carbendazim was dissolved in 0.1 M HCl at
10 mg/mL as a stock solution and was added to autoclaved potato
sucrose agar (PSA) that had been cooled to 45−50 °C. The pH
value was adjusted with HCl to 6.8 in all media. Isoprothiolane,
combretastatin A-4, and derivatives were dissolved in dimethyl sulfoxide
(DMSO) and then mixed with the autoclaved PSA. A mycelial plug
(7 mm in diameter) taken from the edge of a 4-day-old colony of
strains was grown in the dark for 3 days at 25 °C in a 9 cm diameter
Petri plate containing PSA amended with different concentrations of
fungicides. The radial growth from the edge of the plug to the edge of
the colony of each strain was measured after 3 days at 25 °C. For each
plate, the average colony radial growth, measured in two perpendicular
directions, was used for calculation of the inhibition. Data presented
as percentages were log-transformed before statistical analyses. Data
were analyzed using a one-way analysis of variance (ANOVA) with a
Dunnett’s multiple-comparison test as the post hoc determination of
significance using GraphPad Prism version 5.0 (GraphPad Software,
San Diego, CA, USA). Dose−response curves were also prepared
using the computer program GraphPad Prism. Values are expressed as
mean standard errors (SE) with n indicating the number of samples
measured (n = 3−5).
Combretastatin A-4 and carbendazim were constructed using the
2D sketcher module in Sybyl. Minimum energy conformations of
all structures were calculated using the Minimize module of Sybyl.
The force field was MMFF94 with an 8 Å cutoff for nonbonded
interactions, and the atomic point charges were also calculated with
MMFF94.12 Minimizations were achieved using the steepest descent
method for the first 100 steps, followed by the Broyden−Fletcher−
Goldfarb−Shanno (BFGS)13 method until the root-mean-square
(RMS) of the gradient became <0.005 kcal/(mol·Å).14
The Surflex-Dock15 module implemented in the Sybyl program
was used for the docking studies. Each inhibitor was docked into the
corresponding protein’s binding site by an empirical scoring function
and a patented search engine in Surflex-Dock, applied with the automatic
docking. Other parameters were established by default in the software.
RESULTS AND DISCUSSION
■
Chemistry. (Z)-2-Methoxy-5-(3,4,5-trimethoxystyryl)]-
phenol (combretastatin A-4) and (Z)-2-methoxy-5-(3,4,5-
trimethoxystyryl)]phenyl disodium phosphate (combretastatin
A-4P) were synthesized following previously reported methods.1,7
The double bond in part A of combretastatin A-4 was
hydrogenated, the hydroxyl group in part B was replaced by an
amino or nitro or disodium phosphate group, and the methoxy
747
DOI: 10.1021/acs.jafc.5b05119
J. Agric. Food Chem. 2016, 64, 746−751

Journal of Agricultural and Food Chemistry
Article
748
DOI: 10.1021/acs.jafc.5b05119
J. Agric. Food Chem. 2016, 64, 746−751

Journal of Agricultural and Food Chemistry
Article
group in part C was replaced by an ethoxy group to afford
12 combretastatin A-4 derivatives, respectively (Figure 1).
The synthesis of the derivatives is outlined in Figure 2 with the
methods previously reported.9 All of the structure elucida-
tion data are presented in Table 1. E-Diphenylethene and
Z-diphenylethene were characterized by the different 1H NMR
coupling constants. The J values of 8 (E-diphenylethene) and
7 (Z-diphenylethene) are 16.0 and 12.0 Hz, respectively.
Biological Evaluations and SAR Assay. The antifungal
effects of combretastatin A-4, combretastatin A-4P, and deriva-
tives were determined from concentration−response curves
with rice sheath blight disease (R. solani) and rice blast disease
(P. oryzae) as the tested strains. The EC50 value of each com-
pound is shown in Table 2. As isoprothiolane is a commercial
fungicide widely used for rice disease control, it was selected as
a positive control in our experiment. For the rice sheath blight
disease control, combretastatin A-4 and 2 showed significant
effects, with EC50 values of 46.5 and 51.6 μM, respectively,
compared with isoprothiolane (EC50 value = 64.5 μM).
Combretastatin A-4, 1, 2, 7, and 11 were more active than
isoprothiolane in the antifungal assay against P. oryzae.
Table 2. Antifungal Activities of Combretastatin A-4 and
Derivatives
a
EC50 value (μM)
inhibition at 50 μg/mL (%)
compound
R. solani
P. oryzae S. sclerotio F. oxysporum
The double bond in part A of combretastatin A-4 was hydro-
genated to afford the final product. The antifungal effects of 1
and combretastatin A-4 are similar, which suggested that the
single bond retains the antifungal activity. It is obvious that the
effect is completely lost when the natural cis configuration is
transformed to the trans configuration. Structural modifications
to the hydroxyl group in part B indicate that the amino group
maintained the activity; however, the nitro and disodium
phosphate groups completely eliminated the antifungal activity.
The bioassay results for compounds 2 and 4 indicated that the
introduction of an ethoxy group replacing the methoxy group
in part C enhanced bioactivity against the rice diseases.
Molecular Modeling. Molecular modeling studies were
undertaken to determine the interaction between combretas-
tatin A-4 and fungal tubulin. The studies were performed
with homology modeling and the Surflex-Dock module in the
Sybyl program. The homology model of β-tubulin of R. solani
combretastatin A-4
46.5
104
27.7
−0.6
−1.5
24.7
0.3
53.3
17.2
15.1
31.9
16.7
21.9
29.2
27.4
36.6
16.4
15.1
12.0
24.0
14.4
21.7
−
combretastatin A-4P no effect no effect
1
75.6
51.6
355
238
271
99.6
383
369
2
3
4
4.9
5
no effect no effect
no effect no effect
−1.2
−0.6
11.6
−1.5
0.9
6
7
94.0
no effect no effect
261 624
no effect no effect
188 291
no effect no effect
159
8
9
10
−3.7
4.9
11
12
−0.9
5.8
b
carbendazim
isoprothiolane
−
−
324
64.5
−
a
b
Inhibition against carbendazim resistance strains. −, not tested.
Figure 3. Binding sites of combretastatin A-4 and carbendazim in R. solani β-tubulin. Combretastatin A-4 is shown with green spacefill and
carbendazim with red spacefill. Hydrogen bonds are shown with dotted yellow lines. The conserved amino acid residues are highlighted in a red box.
749
DOI: 10.1021/acs.jafc.5b05119
J. Agric. Food Chem. 2016, 64, 746−751

Journal of Agricultural and Food Chemistry
Article
(GenBank CUA76508.1) was generated with the crystal
structure of cow’s brain β-tubulin (PDB ID 1Z2Bb) as the
template. A sequence alignment showed their sequence identity
value was 86%. Combretastatin A-4 and carbendazim were used
to identify the receptor binding pockets and to analyze the
binding characteristics with a blind docking calculation in this
study. The colchicine binding site of tubulin has been validated
as the binding pocket of combretastatin A-4 (Figure 3) between
two adjacent subunits of tubulin.16 Thus, the colchicine binding
site was used for the docking simulation of combretastatin
A-4 and derivatives. Hydrogen-bonding interactions show that
combretastatin A-4 is bound to Val236 and Asn256 of subunit
A of the β-tubulin model. This suggested a hydrogen-bonding
interaction between the hydroxyl group of combretastatin A-4
and Val236 is critical for full biological activity, which explains
why the biological activity was lost if the hydroxyl group was
replaced by a nitro or disodium phosphate group. Figure 2
shows that the carbendazim binding site is different from that
of combretastatin A-4. Carbendazim has interactions only
with one subunit of β-tubulin. Qiu et al.17 demonstrated the
possible binding pocket for carbendazim on the β2-tubulin in
Gibberella zeae was formed with residues Tyr50, Phe167,
Glu198, Phe200, and Arg241. Sequence alignment showed
those above residues are conserved (Figure 3). The carbendazim
binding pocket of R. solani β-tubulin was also formed with those
residues. In addition, His6 and Trp21 formed two hydrogen
bonds with carbendazim in our model, explaining why muta-
tions at the codons 6 gene were responsible for resistance to
carbendazim.7
It is obvious that the binding sites of combretastatin A-4
and carbendazim on fungal tubulin are totally different, which
suggested that combretastatin A-4 and derivatives can be the
useful for controlling carbendazim-resistant plant diseases.
Antifungal Effects on Resistant Strains. Two carbenda-
zim-resistant strains (Fusarium oxysporum Schl. f. sp. vasinfectum
(Atk.) Snyd. & Hans and Sclerotinia sclerotiorum (Lib.) de Bary)
were used to test the antifungal effects of combretastatin A-4
and derivatives on the resistant fungi. The results demonstrated
that combretastatin A-4 and 2 have appreciable inhibitory ability
to the resistant fungi, compared with carbendazim (Table 2).
On testing with F. oxysporum, the inhibition by combretastatin
A-4 was 2.5-fold greater than that of carbendazim. Combretas-
tatin A-4 and 2 have nearly 5- and 4-fold more antifungal effects
than carbendazim against S. sclerotiorum. This result showed
that the binding sites of combretastatin A-4 and derivatives
are different from that of carbendazim, which suggested that
combretastatin A-4 is a good lead compound for new fungicide
discovery that can be used for the control of carbendazim-
resistant fungi.
AUTHOR INFORMATION
■
Corresponding Authors
*(Z.-P.K.) Phone: +86 136 71951027. Fax: +86 21 60877220.
E-mail: kaizp@sit.edu.cn.
*(F.-H.W.) Phone: +86 21 60877220. Fax: +86 21 60877220.
Email: wfh@sit.edu.cn.
Funding
This work was supported by grants from the National Natural
Science Foundation of China (No. 21402122 and 21472126).
It was also supported by the Natural Science Foundation of
Shanghai (No. 14ZR1440600) and the Innovation Program of
Shanghai Municipal Education Commission (No. 14YZ148).
Notes
The authors declare no competing financial interest.
REFERENCES
■
(1) Pettit, G. R.; Singh, S. B.; Hamel, E.; Lin, C. M.; Alberts, D. S.;
Kendall, D. G. Isolation and structure of the strong cell growth and
tubulin inhibitor combretastatin A-4. Experientia 1989, 45, 209−211.
(2) Salehi, M.; Ostad, S. N.; Riazi, G. H.; Assadieskandar, A.; Shavi,
T. C.; Shafiee, A.; Amini, M. Synthsis, cytotoxic evaluation, and
molecular docking study of 4,5-diaryl-thiazole-2-thione analogs of
combretastatin A-4 as microtubule-binding agents. Med. Chem. Res.
2014, 23, 1465−1473.
(3) Hamze, A.; Rasolofonjatovo, E.; Provot, O.; Mousset, C.; Veau,
D.; Rodrigo, J.; Bignon, J.; Liu, J. M.; Bakala, J. W.; Thoret, S.; Dubois,
J.; Brion, J. D.; Alami, M. B-Ring-Modified isoCombretastatin A-4
analogues endowed with interesting anticancer activities. ChemMed-
Chem 2011, 6, 2179−2191.
(4) Qin, Y. J.; Li, Y. J.; Jiang, A. Q.; Yang, M. R.; Zhu, Q. Z.; Dong,
H.; Zhu, H. L. Design, synthesis and biological evaluation of novel
pyrazoline-containing derivatives as potential tubulin assembling
inhibitiors. Eur. J. Med. Chem. 2015, 94, 447−457.
(5) Verhey, K. J.; Gaertig, J. The tubulin code. Cell Cycle 2007, 6,
2152−2160.
(6) Bollen, G. J.; Scholten, G. Acquired resistance to benomyl and
some other systemic fungicides in a strain of Bostryis cinerea in
cyclamen. Neth. J. Plant Pathol. 1971, 77, 83−90.
(7) Qiu, J. B.; Huang, T. T.; Xu, J. Q.; Bi, C. W.; Chen, C. J.; Zhou,
M. G. β-Tubulins in Gibberella zeae: their characterization and
contribution to carbendazim resistance. Pest Manage. Sci. 2012, 68,
1191−1198.
(8) Shen, W. P.; Diao, Y. J.; Jin, H. M.; Wang, J. P.; Wang, J. G.
Synthesis of combretastatin A-4 supported by polymer. Chinese Patent
CN 1704393A, 2005.
(9) Zhao, L.; Zhou, J. J.; Huang, X. Y.; Cheng, L. P.; Pang, W.; Kai, Z.
P.; Wu, F. H. Design, synthesis and anti-proliferative effects in tumor
cells of new combretastatin A-4 analogs. Chin. Chem. Lett. 2015, 26,
993−999.
(10) Shi, J.; Blundell, T. L.; Mizuguchi, K. FUGUE: sequence-
structure homology recognition using environment-specific substitu-
tion tables and structure-dependent gap penalties. J. Mol. Biol. 2001,
310, 243−257.
(11) Sutcliffe, M. J.; Haneef, I.; Carney, D.; Blundell, T. L.
Knowledge based modelling of homologous proteins, part I: three-
dimensional frameworks derived from the simultaneous superposition
of multiple structures. Protein Eng., Des. Sel. 1987, 1, 377−384.
(12) Halgren, T. A. MMFF VI. MMFF94s option for energy
minimization stydies. J. Comput. Chem. 1999, 20, 720−729.
(13) Head, J. D.; Zerner, M. C. A Broyden-Fletcher-Goldfarb-Shanno
optimization procedure for molecular geometries. Chem. Phys. Lett.
1985, 122, 264−270.
(14) Kai, Z. P.; Ling, Y.; Liu, W. J.; Zhao, F.; Yang, X. L. The study of
solution conformation of allatostatins by 2-D NMR and molecular
modeling. Biochim. Biophys. Acta, Proteins Proteomics 2006, 1764, 70−
75.
In conclusion, combretastatin A-4 was first isolated from
the African willow tree, and the double bond in part A of
combretastatin A-4 was hydrogenated; the hydroxyl group in
part B was replaced by amino, nitro, or disodium phosphate
group, and the methoxy group in part C was replaced by an
ethoxy group to afford 12 combretastatin A-4 derivatives,
respectively. Combretastatin A-4 and its derivatives were
identified as new potential fungicides targeting fungal tubulin.
Because their binding site is entirely different from that of
carbendazim, they are possibly useful for fungal resistance
management. The results of antifungal assays on resistant
strains confirmed our hypothesis.
750
DOI: 10.1021/acs.jafc.5b05119
J. Agric. Food Chem. 2016, 64, 746−751

Journal of Agricultural and Food Chemistry
Article
(15) Spitzer, R.; Jain, A. N. Surflex-Dock: docking benchmarks and
real-world application. J. Comput.-Aided Mol. Des. 2012, 26, 687−699.
(16) Chen, J.; Liu, T.; Dong, X.; Hu, Y. Z. Recent development and
SAR analysis of colchicine binding site inhibitors. Mini-Rev. Med.
Chem. 2009, 9, 1174−1190.
(17) Qiu, J. B.; Xu, J. Q.; Yu, J. J.; Bi, C. W.; Chen, C. J.; Zhou, M. G.
Localisation of the benzimidazole fungicide binding site of Gibberella
zeae β2-tubulin studied by site-directed mutagenesis. Pest Manage. Sci.
2011, 67, 191−198.
751
DOI: 10.1021/acs.jafc.5b05119
J. Agric. Food Chem. 2016, 64, 746−751