Design, synthesis and biological evaluation of benzoylacrylic acid shikonin ester derivatives as irreversible dual inhibitors of tubulin and EGFR

Article abstract of DOI:10.1016/j.bmc.2019.115153

In this study, a series of shikonin derivatives combined with benzoylacrylic had been designed and synthesized, which showed an inhibitory effect on both tubulin and the epidermal growth factor receptor (EGFR). In vitro EGFR and cell growth inhibition assay demonstrated that compound PMMB-317 exhibited the most potent anti-EGFR (IC₅₀ = 22.7 nM) and anti-proliferation activity (IC₅₀ = 4.37 μM) against A549 cell line, which was comparable to that of Afatinib (EGFR, IC₅₀ = 15.4 nM; A549, IC₅₀ = 6.32 μM). Our results on mechanism research suggested that, PMMB-317 could induce the apoptosis of A549 cells in a dose- and time-dependent manner, along with decrease in mitochondrial membrane potential (MMP), production of ROS and alterations in apoptosis-related protein levels. Also, PMMB-317 could arrest cell cycle at G2/M phase to induce cell apoptosis, and inhibit the EGFR activity through blocking the signal transduction downstream of the mitogen-activated protein MAPK pathway and the anti-apoptotic kinase AKT pathway; typically, such results were comparable to those of afatinib. In addition, PMMB-317 could suppress A549 cell migration through the Wnt/β-catenin signaling pathway in a dose-dependent manner. Additionally, molecular docking simulation revealed that, PMMB-317 could simultaneously combine with EGFR protein (5HG8) and tubulin (1SA0) through various forces. Moreover, 3D-QSAR study was also carried out, which could optimize our compound through the structure-activity relationship analysis. Furthermore, the in vitro and in vivo results had collectively confirmed that PMMB-317 might serve as a promising lead compound to further develop the potential therapeutic anticancer agents.

Full text of DOI:10.1016/j.bmc.2019.115153

Journal Pre-proofs
Design, synthesis and biological evaluation of benzoylacrylic acid shikonin ester
derivatives as irreversible dual inhibitors of tubulin and EGFR
Wen-Xue Sun, Hong-Wei Han, Min-Kai Yang, Zhong-Ling Wen, Yin-Song
Wang, Jiang-Yan Fu, Yun-Ting Lu, Ming-Yue Wang, Jia-Xin Bao, Gui-Hua
Lu, Jin-Liang Qi, Xiao-Ming Wang, Hong-Yan Lin, Yong-Hua Yang
PII:
S0968-0896(19)31152-6
https://doi.org/10.1016/j.bmc.2019.115153
BMC 115153
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
9 July 2019
27 September 2019
3 October 2019
Please cite this article as: Sun, W-X., Han, H-W., Yang, M-K., Wen, Z-L., Wang, Y-S., Fu, J-Y., Lu, Y-T., Wang,
M-Y., Bao, J-X., Lu, G-H., Qi, J-L., Wang, X-M., Lin, H-Y., Yang, Y-H., Design, synthesis and biological evaluation
of benzoylacrylic acid shikonin ester derivatives as irreversible dual inhibitors of tubulin and EGFR, Bioorganic &
Medicinal Chemistry (2019), doi: https://doi.org/10.1016/j.bmc.2019.115153
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will
undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing
this version to give early visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
©
2019 Elsevier Ltd. All rights reserved.

Design, synthesis and biological evaluation of benzoylacrylic acid
shikonin ester derivatives as irreversible dual inhibitors of tubulin
and EGFR
1
,2†
1,2†
1,2
1,2
Wen-Xue Sun , Hong-Wei Han , Min-Kai Yang , Zhong-Ling Wen , Yin-Song
1
,2
1,2
1,2
1,2
1,2
Wang , Jiang-Yan Fu , Yun-Ting Lu , Ming-Yue Wang , Jia-Xin Bao , Gui-Hua
1
,2
1,2
1,2
1,2
1,2
Lu , Jin-Liang Qi *, Xiao-Ming Wang *, Hong-Yan Lin *, Yong-Hua Yang *
1State Key Laboratory of Pharmaceutical Biotechnology, Institute of Plant Molecular
Biology, School of Life Sciences, Nanjing University, Nanjing 210023, PR China
2Co-Innovation Center for Sustainable Forestry in Southern China, Nanjing Forestry
University, Nanjing, 210037, PR China
†These authors contributed equally to this work
*
Corresponding author:
Yong-Hua Yang,
State Key Laboratory of Pharmaceutical Biotechnology,
Nanjing University,
Nanjing 210023, PR China.
Tel/Fax: 86-25-89686305,
Email: yangyh@nju.edu.cn.
1

Abstract: In this study, a series of shikonin derivatives combined with benzoylacrylic
had been designed and synthesized, which showed an inhibitory effect on both tubulin
and the epidermal growth factor receptor (EGFR). In vitro EGFR and cell growth
inhibition assay demonstrated that compound PMMB-317 exhibited the most potent
anti-EGFR (IC = 22.7 nM) and anti-proliferation activity (IC = 4.37 μM) against
5
0
50
A549 cell line, which was comparable to that of Afatinib (EGFR, IC = 15.4 nM; A549,
5
0
IC = 6.32 μM). Our results on mechanism research suggested that, PMMB-317 could
5
0
induce the apoptosis of A549 cells in a dose- and time-dependent manner, along with
decrease in mitochondrial membrane potential (MMP), production of ROS and
alterations in apoptosis-related protein levels. Also, PMMB-317 could arrest cell cycle
at G2/M phase to induce cell apoptosis, and inhibit the EGFR activity through blocking
the signal transduction downstream of the mitogen-activated protein MAPK pathway
and the anti-apoptotic kinase AKT pathway; typically, such results were comparable to
those of afatinib. In addition, PMMB-317 could suppress A549 cell migration through
the Wnt/β-catenin signaling pathway in a dose-dependent manner. Additionally,
molecular docking simulation revealed that, PMMB-317 could simultaneously
combine with EGFR protein (5HG8) and tubulin (1SA0) through various forces.
Moreover, 3D-QSAR study was also carried out, which could optimize our compound
through the structure-activity relationship analysis. Furthermore, the in vitro and in vivo
results had collectively confirmed that PMMB-317 might serve as a promising lead
compound to further develop the potential therapeutic anticancer agents.
Chemical compounds studies in this article：
2

Shikonin (PubChem CID: 479503); Maleic anhydride (PubChem CID: 7923);
Colchicine (PubChem CID: 6167); Paclitaxel (PubChem CID: 36314); Afatinib
(PubChem CID: 10184653); Benzene (PubChem CID: 241); Methylbenzene
(PubChem CID: 1140); Fluorobenzene (PubChem CID: 10008); Chlorobenzene
(PubChem CID: 7964); Bromobenzene (PubChem CID: 7961).
Keywords: Shikonin, benzoylacrylic, tubulin, epidermal growth factor receptor,
anticancer
3

1
. Introduction
With the continuous exploration of tumor biology and the application of targeted
drug delivery methods, targeted drug therapy has gradually become a research hotspot
as a result of its outstanding clinical advantages, which can target specific binding sites
in the course of tumor treatment, improve the therapeutic effect and reduce the drugs-
related side effects[1].
As a kind of receptor tyrosine kinase, epidermal growth factor receptor (EGFR)
can induce a complicated signaling cascade reaction, which plays a vital part in the
biological functions to regulate cell support, survival, adhesion, and migration[2].
Specifically, the EGFR protein signaling pathways include the mitogen-activated
protein kinase (MAPK) pathway[3], the anti-apoptotic kinase AKT pathway[4, 5], and
several DNA transcription-related pathways in nucleus apoptosis. Each signaling
pathway is linked with each other, which is involved in the normal physiological
activities of cells and the cancerization process[6]. Over-expression of EGFR may
account for an important cause of tumor formation and proliferation, and specifically
inhibiting the EGFR activity can restrain tumor growth[7].
There are a series of small molecular compounds being reported to inhibit the
EGFR activity, which can compete with the transmembrane domain in EGFR receptor
to block the self-phosphorylation of the receptor. The small molecular EGFR tyrosine
kinase inhibitors can be mainly classified as three categories, including the reversible
single-target EGFR inhibitor, reversible double-target EGFR/HER2 inhibitor and non-
reversible EGFR inhibitor[8]. Typically, gefitinib (Fig. 1A) is a reversible single-target
EGFR inhibitor, which can highly integrate with the EGFR receptor (it can compete
4

with the ATP binding site in EGFR receptor[9]) to inhibit cell proliferation, and
promote cell cycle arrest at the G0/G1 phase, thus ultimately inducing apoptosis.
Icotinib (Fig. 1B), regarded as the "domestic Iressa"[10], is mainly indicated for the
advanced non-small cell lung cancer (NSCLC), and it is found in clinical data to have
higher safety than gefitinib[11]. Erlotinib and Afatinib (Fig. 1C and D) are the efficient
reversible and irreversible EGFR inhibitors, respectively[12], which can block the cell
cycle at G1 phase[13, 14]. Notably, the above EGFR inhibitors have shared a similar
mechanism and post an inevitable drug resistance. Therefore, the existing approach to
avoid drug resistance is to develop the effective multi-target inhibitors.
Fig. 1. Reported natural and synthetic EGFR inhibitors.
Notably, the β-Aroylacrylic acid derivatives have attracted wide attention because
of their antibacterial, antitumor, fungicidal and other physiological activities; besides,
they are also the important synthetic intermediates in the fields of medical science,
agriculture, biology, and perfume[15]. The 4-phenyl-4-oxo-2-butenoic acid
(benzoylacrylic acid) is structurally similar to chalcone and aroylacrylic acids, which
has attracted our interest for further investigation since it can selectively and covalently
combine the diverse pharmacological targets. In addition, results of 2D QSAR study
had revealed the potency of such compounds in the covalent inhibition of EGFR.
Moreover, shikonin and its derivatives have possessed the antibacterial, anti-tumor,
5

antiviral, anti-inflammatory, anti-allergic and other pharmacological effects[16-18].
Our recent studies have mainly focused on obtaining the shikonin derivatives with high
efficiency and low toxicity, and it is found that various types of modified shikonin
derivatives have outstanding anti-tubulin activity[19]. Inspired by these previous
research results, this study was mainly carried out aiming to design a kind of shikonin
derivative with improved tumor targeting capacity. Typically, we would like to
introduce a group of benzoylacrylic acid pharmacophore with EGFR-targeting
inhibitory activity to the shikonin structure, which had combined a new type of reactive
molecule containing both shikonin and benzoylacrylic acid. Results of activity
validation had proved the function of this new molecular structure consisting of double
enzyme or protein targets, together with various protein change-related regulatory
pathways in the cell system. In addition, molecular docking simulation, along with other
molecular activity detection, had rendered more reasonable molecular drug design.
6

2
. Results
2
.1 Chemistry
All the synthesized compounds for the benzoyl acrylic acid shikonin ester
derivatives PMMB311-325 followed the general pathway outlined in Scheme 1. These
compounds were obtained by two steps, which are elucidated in experimental section.
1
13
All of them were firstly reported (Table 1) and characterized by H NMR, C NMR,
melting test and mass spectroscopy in full accordance with depicted structures (SI
Appendix). All docking runs were carried out by using CDOCKER Dock protocol of
Discovery Studio 3.5.
Scheme 1. Synthesis of compounds PMM-311~PMM-325. Reagents and conditions:
(
a) AlCl , CH Cl , reflux, 12 h; (b) DCC, DMAP, CH Cl , 0 °C, 4 h.
3 2 2 2 2
Table 1
Chemical structures and binding energy of PMM-311~PMM-325, Shikonin and
Afatinib with 1SA0 and 5HG8.
7

R2
R5
R1
R3
R4
O
OH
OH
O
O
O
O
1
SA0
5HG8
CC50
(μM)
L02
△
Gb(kca △Gb(kca
Compound
PMMB-311
PMMB-312
PMMB-313
PMMB-314
PMMB-315
PMMB-316
PMMB-317
PMMB-318
PMMB-319
PMMB-320
PMMB-321
PMMB-322
PMMB-323
PMMB-324
PMMB-325
Shikonin
R1
CH3
H
R2
H
R3
R4
CH3
H
R5
l/mol)
l/mol)
7
6
8
8.1
4.2
8.4
100
2.4
100
9.2
100
100
100
3.3
100
100
3.5
100
2.2
5.3
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
-60.25
-52.13
-56.89
-58.96
-59.28
-60.46
-62.52
-54.63
-56.73
-52.54
-49.37
-53.66
-60.42
-64.35
-54.69
-65.53
-54.85
-42.57
H
CH3
CH3
CH3
-53.46
-49.34
-52.11
-56.36
-48.21
-58.76
-45.86
-46.34
-48.54
-49.03
-55.54
-50.46
-48.52
-51.64
-54.13
-57.37
H
CH3
H
H
>
Cl
H
H
9
H
CH CH
H
2
3
>
H
H
Cl
F
H
8
H
H
H
>
H
F
CH3
CH3
CH3
H
H
>
>
CH3
H
H
H
Br
H
H
8
CH3
H
Cl
H
>
>
H
H
F
H
CH3
H
H
9
CH3
H
H
F
>
H
Br
H
1
2
Afatinib
2
2
.2 Bioactivity
.2.1. PMMB-317 selectively inhibits cancer cells proliferation
8

MTT assay were performed for antiproliferative activities by the new compound
PMMB311-325 against five human cancer cell lines, including the human breast cancer
cells (MCF-7), human cervical cancer cells (HeLa), human non-small cell lung cancer
cell lines (NCI H1975), human lung adenocarcinoma cell line (A549), human lung
carcinoma cells (NCI H460) and human normal liver cell line (L02). Results showed in
Table 1 indicated that all compounds had lower cytotoxic against the non-tumorigenic
L02 cells than shikonin itself. As shown in Table 2, almost all the compounds showed
the anti-proliferative activities, Specifically, PMMB-317 (IC50 = 4.37 ± 0.46 μM)
performed better anti-proliferation activities than afatinib (IC = 6.32 ± 1.57 μM) and
5
0
shikonin (IC = 6.13 ± 1.68 μM) against A549 cells. Based on these results, we
5
0
speculated that intervention of benzoyl acrylic acid derivatives could increase the anti-
proliferative activities against cancer cells but significantly reduce the cytotoxicity of
shikonin against non-cancer cells. To further investigate the anti-cancer mechanisms of
these drugs, we chose compound PMMB-317 against A549 cells for downstream
application.
Table 2
In vitro cell proliferation of compounds PMMB-311~ PMMB-325, Shikonin and
Afatinib against five cancer cell lines.
Compound
IC ± SD (μM)
50
Hela
MCF-7
5.28 ± 0.17
9.14 ± 0.41
14.32 ± 1.82
A549
H460
H1975
PMMB-311
PMMB-312
PMMB-313
10.15 ± 0.24
8.96 ± 0.25
10.13 ± 1.36
13.13 ± 0.75
9.72 ± 1.94
9.12 ± 1.82
13.21 ± 1.67
11.43 ± 2.44
11.33 ± 1.42
10.19 ± 0.96
10.17 ± 0.66
15.68 ± 0.58
9

PMMB-314
PMMB-315
PMMB-316
PMMB-317
PMMB-318
PMMB-319
PMMB-320
PMMB-321
PMMB-322
PMMB-323
PMMB-324
PMMB-325
Shikonin
10.13 ± 1.45
15.23 ± 2.58
11.42 ± 2.52
7.93 ± 0.49
8.27 ± 1.16
18.55 ± 2.37
17.15 ± 2.28
9.34 ± 0.25
17.56 ± 0.65
15.1 ± 1.84
14.21 ± 2.68
8.02 ± 0.19
7.84 ± 0.67
6.39 ± 0.58
6.63 ± 0.53
9.37 ± 1.33
9.62 ± 0.69
8.90 ± 0.13
7.68 ± 1.23
7.89 ± 1.2
6.23 ± 1.1
14.54 ± 0.63
8.52 ± 0.61
4.37 ± 0.46
9.39 ± 1.81
10.18 ± 0.98
15.37 ± 1.31
13.22 ± 1.74
7.51 ± 1.28
12.63 ± 0.57
13.42 ± 0.59
9.19 ± 0.53
6.13 ± 1.68
6.32 ± 1.57
8.16 ± 1.21
16.35 ± 1.41
11.71 ± 1.23
5.61 ± 1.03
10.12 ± 2.53
9.17 ± 1.23
4.21 ± 0.45
10.32 ± 0.71
4.59 ± 0.03
13.63 ± 2.5
8.35 ± 1.32
6.51 ± 0.98
5.04 ± 1.18
6.14 ± 0.67
9.56 ± 1.29
12.39 ± 0.22
11.85 ± 0.29
9.43 ± 1.26
7.25 ± 0.92
16.55 ± 0.82
22.54 ± 1.66
18.28 ± 1.03
10.26 ± 1.09
9.56 ± 0.99
7.26 ± 0.42
13.28 ± 0.94
4.28 ± 1.08
5.55 ± 1.26
14.28 ± 0.49
18.60 ± 0.44
19.43 ± 2.18
6.64 ± 0.36
10.08 ± 1.45
9.31 ± 0.43
9.55 ± 1.43
7.62 ± 2.35
Afatinib
IC , the mean value of three separate determinations.
5
0
2
.2.2. In vitro EGFR inhibitory activities
The EGFR inhibitory activities of all the synthesized shikonin derivatives were
further evaluated in vitro. As shown in Table 3, all the tested compounds exhibited
EGFR inhibitory activity at varying degree and some of them are even comparable to
that of afatinib. Especially, compound PMMB317 (IC = 22.7 nM) and PMMB324
5
0
(
IC = 23.4 nM) inhibited the EGFR activity remarkably, which was at least 6-fold
50
stronger than shikonin (IC =139.2 nM) and almost comparable to afatinib (IC = 15.4
5
0
50
nM). Generally, the substituent groups on the aromatic ring affected the activity
significantly. According to the structure-activity relationship, we clearly found that
compounds with substitution on the phenyl ring enhanced inhibitory activities.
Moreover, electron-withdrawing group (F, Cl) improved the enzymatic inhibitory
1
0

activity more significantly than the electron donating group (Et, Me).
Table 3
The EGFR inhibitory activity of PMMB-311~PMMB-325, Shikonin and Afatinib.
EGFR kinase
Compound
PMMB-311
PMMB-312
PMMB-313
PMMB-314
PMMB-315
PMMB-316
PMMB-317
PMMB-318
PMMB-319
PMMB-320
PMMB-321
PMMB-322
PMMB-323
PMMB-324
PMMB-325
Shikonin
% inhibition at 150 nM
IC (nM)
50
40.6
67.1
55.8
59.1
65.3
76.1
80.3
59.5
40.9
33.1
50.1
19.4
56.3
83.3
54.6
50.9
85.6
>150
45.2
118.6
105.3
50.8
32.5
22.7
107.2
>150
>150
146.2
>150
120.2
23.4
119.7
139.2
15.4
Afatinib
2
.2.3. PMMB-317 caused cell apoptosis in a dose- and time- dependent manner
To evaluate the cell inhibitory activity of the compound PMMB-317, we conducted
the cellular apoptosis assay. A549 cells were treated with PMMB-317 at various
concentrations (0, 2, 4, 8 and 12 μM) for 24 h and 8 μM for 0, 12, 24 and 36 h at 37 ℃.
As shown in Fig. 2A to D, with increasing concentrations or time of PMMB-317, the
1
1

percentage of apoptotic cells significantly increased, which suggested that PMMB-317
induced apoptosis in a time- and dose-dependent fashion. Additionally, we observed
that PMMB-317 could significantly decrease Bcl-2, caspase-3 and caspase-9, but
increased Bax protein expression and the cleavage of PARP, caspase-3 and caspase-9
in A549 cells, which was in full accordance with cell apoptosis results by western
blotting. (Fig. 2E to H). Treatment with PMMB-317 at 4 μM and 8 μM for 24 h induced
2
8.20% and 66.10% cells apoptosis respectively. Notably, 18.92% and 59.2% cells
induced apoptosis when treated with PMMB-317 at 8 μM for 12 h and 24 h respectively.
Hence, it was concluded that PMMB-317 had a better ability to induce cell apoptosis
especially at lower concentration and shorter time.
1
2

Fig. 2. PMMB-317 induced cell apoptosis in A549 cells. (A) PMMB-317 induced cell
apoptosis in A549 cells in a dose-dependent manner. (B) PMMB-317 induced cell
apoptosis in A549 cells in a time-dependent manner. (C) The percentage of apoptotic
cells in A549 cells treated with various concentrations of PMMB-317 for 24 h. (D) The
percentage of apoptotic cells in A549 cells treated with 8 μM PMMB-317 for different
time. (E) Western blot analyses of apoptosis-related proteins (Bax, Bcl2) separated by
SDS-PAGE. (F) Relative expression level was analyzed by Image J software. GAPDH
served as a loading control. Images are representative of three independent experiments.
Data are mean ± S.E.M. from three independent experiments (*P < 0.05; **P < 0.01).
(G) Western blot analyses of apoptosis-related proteins (PARP, cleaved PARP, Caspase
3
, cleaved caspase 3, Caspase 9 and cleaved caspase 9) separated by SDS-PAGE. (H)
and (I) Relative band intensity was determined by Image J software. GAPDH served as
a loading control. Images are representative of three independent experiments. Data are
mean ± S.E.M. from three independent experiments (*P < 0.05; **P < 0.01).
2
.2.4. PMMB-317 could induce the early apoptosis in a dose-dependent manner
The early apoptosis of cells is accompanied by production of ROS and the decrease
of MMP[20]. After treatment with 4, 8 or 12 μM of PMMB-317, microscopy revealed
that in untreated A549 cells, the pattern of well-polarized mitochondria marked by red
fluorescent staining was replaced by green fluorescence (Fig. 3). In addition, flow
cytometry showed that the percentage of cells with high membrane potential decreased
from 90.6% to 67.8% after exposure to 4 μM PMMB-317 for 12 h and even decreased
1
3

to 38.6% in 12 μM PMMB-317 treated group. To eliminate the impact of late apoptosis
on MMP depolarization, cells were treated with PMMB-317 for only 12 h, and through
the PI staining results, there was scarely any late apoptotic cells.
Fig. 3. Mitochondrial transmembrane potential was analysed in PMMB-317-treated
A549 cells by JC-1 staining. (A) Cytofluorimetric analysis of mitochondrial membrane
potential (△Ψm) by JC-1 staining. (B) Flow cytometry analysis of A549 cells with high
△
Ψm (%) by JC-1 staining. (C) Flow cytometry analysis of A549 cells present PI
positive. Each point represents the mean ± S.E.M. from three replicates (**P < 0.01).
Moreover, ROS generation represent a result of intracellular mitochondrial
depolarization according the previous study[21]. Therefore, it is essential for us to
investigate whether the PMMB-317 could induce ROS production in A549 cells. As
shown in Fig. S1, ROS inducing ability were significantly increased with the treatment
of PMMB-317 in a dose dependent manner. Taken together, all of these can further
1
4

confirm that PMMB-317 could induce the early apoptosis of A549 cells in a dose-
dependent manner.
2
.2.5. Effect of PMMB-317 on cell cycle arrest
Then to elucidate whether the anti-proliferative effect induced by the derivatives
was due to cell cycle arrest, A549 cells were treated with the compound PMMB-317.
According to the data illustrated in Fig. 4A to D, treatment of A549 cells with
compound PMMB-317 lead to G2/M arrest in a dose and time mechanism. When the
concentration of PMMB-317 increased to 8 μM, 33.01% of cells showed G2/M phase
arrest compared with the control groups. Consistently, 46.55% of cells showed G2/M
arrest, when the treatment time was 36 h. In keeping with the result, immunoblot
analysis of cell cycle related proteins, Cyclin A, CDK1 and CDK2 were performed. As
shown in Fig. 4E and F, the accumulation levels of CDK1 markedly increased
compared to the control group, which indicated that the G2/M arrest was induced by
PMMB-317. while the level of cyclin A and CDK1 were moderately down regulated.
Therefore, we predict that PMMB-317 was not very efficient to induce cell apoptosis
via cell cycle arrest.
1
5

Fig. 4. The effects of different concentrations of PMMB-317 on cell cycle distribution
in A549 cells. (A) Cells were treated with 0, 2, 4, 8 μM PMMB-317 for 24 h. (B) Cells
were treated with 8 μM PMMB-317 for different time (G1 phase, green; S phase, yellow
and G2/M phase, blue). (C) Statistic analysis of dose-dependent assay. (D) Statistic
analysis of time-dependent assay. (E) Representative image of immunoblot analysis of
cell cycle related proteins in A549 cells treated with PMMB-317. (F) Relative band
intensity was determined by Image J software. GAPDH served as a loading control.
Images are representative of three independent experiments. Data are mean ± S.E.M.
from three independent experiments (*P < 0.05; **P < 0.01).
1
6

2
.2.6. Effect of PMMB-317 on the cytoskeleton architecture of A549 cells
Confocal laser experiments can be used to observe the morphology of cell
microtubules under the action of drugs, previous cell cycle experiment showed that the
compound PMMB-317 could block the A549 cell cycle in G2/M phase, thus affecting
cell growth and proliferation. In view of previous reports that the polymerization and
depolymerization of microtubules could influence the cell cycle. To confirm that, the
depolymerization or aggregation of microtubules were observed under confocal laser
scanning microscope. As shown in Fig. 5, the compound PMMB-317 could lead to
tubulin depolymerization distinctly just like colchicine treated group. The initial result
prompt us to further investigate, microtubule assembly assay was conducted in vitro.
PMMB-317 and colchicine caused inhibition of microtubule assembly at the
concentration of 4, 8 μM and 1 μM for 24 h respectively, while paclitaxel promoted the
polymerization of tubulin significantly at the concentration of 1 μM (Fig. 5B and C).
1
7

Fig. 5. Effect of PMMB-317 on the interphase microtubules of A549 cells using
colchicine and paclitaxel as references. (A) Microtubules tagged with rhodamine (red)
and nuclei tagged with DAPI (blue) were observed under a confocal microscope. (B)
PMMB-317 affected microtubule assembly in vitro. After 24 h treatment with PMMB-
3
17 (4, 8 μM), colchicine (1 μM), and paclitaxel (1 μM), polymerized tubulin on the
cell membrane was immunoblotted with b-tubulin antibody and detected by flow
cytometry. (C) PMMB-317 affected microtubule assembly in vitro. After treating A549
cells with PMMB-317 (4, 8 μM), colchicine (1 μM), and paclitaxel (1 μM) for 24 h,
cytosolic (S, soluble) and cytoskeletal (P, polymerized tubulin) tubulin fractions were
separated and immunoblotted with antibody against β-tubulin.
1
8

2
.2.7. Effect of PMMB-317 on cell migration of A549 cells
Tumor cell migration undoubtedly increases the difficulty of curing tumors,
especially once the distal metastasis occured[21]. To investigate whether PMMB-317
could inhibit the migratory capacity of A549 cells, the conventional in vitro wound
healing assay was conducted. As shown in Fig. 6A and B, the control group cells
gradually occupied the cell-free space within the channel over time and the cells almost
filled the channel after 48 h. Whereas, cells migration capacity was greatly inhibited in
a dose-dependent manner exposure to PMMB-317. The Wnt/β-catenin signaling
pathway is vital for many cellular processes including cell proliferation and
migration[21]. Based on the above results, we subsequently investigated the influence
of PMMB-317 on Wnt/β-catenin signaling pathway in A549 cells. Western blot
analysis results shown in Fig. 6C indicated that PMMB-317-treated A549 cells resulted
in a decreased expression of β-catenin in cytoplasm and an increased protein expression
of Axin-2 in a dose-dependent manner. These results together suggest that PMMB-317
might inhibit migration of A549 cells via Wnt/β-catenin signaling pathway.
1
9

Fig. 6. The effects of PMMB-317 on cell migration in A549 cells. (A) Cell migration
was detected by the wound scrape assay. Representative images of cell migration in the
wound scrape model at 0, 24 and 48 h are shown. (B) Statistical analysis of the wound
scrape assay. (C) Western blot analyses of Wnt/β-catenin signaling pathway proteins
(β-catenin and Axin-2) separated by SDS-PAGE. (D) Relative band intensity was
determined by Image J software. GAPDH served as a loading control. Images are
representative of three independent experiments. Data are mean ± S.E.M. from three
independent experiments (*P < 0.05; **P < 0.01).
2
.2.8. Effect of PMMB-317 on EGFR signaling pathway of A549 cells
EGFR activation has been connected to multiple biological processes in cancer
progression, such as cell migration, proliferation and metastasis through many signaling
pathways[22]. To investigate the effect of PMMB-317 on EGFR, we conducted
2
0

immunoblot analysis of EGFR and its downstream signaling pathways proteins. As
shown in Fig. 7, A549 cells that expressed high EGFR was treated with increasing
concentrations of PMMB-317 and demonstrated a dose-dependent decrease in
phosphorylation of EGFR, ERK and AKT.
Fig. 7. The effects of PMMB-317 on EGFR signaling pathway in A549 cells. (A)
Western blot analyses of apoptosis-related proteins (p-EGFR, Ras, p-c-Raf, ERK1/2,
p-ERK1/2) separated by SDS-PAGE. Relative band intensity was determined by Image
J software. GAPDH served as a loading control. Images are representative of three
independent experiments. Data are mean ± S.E.M. from three independent experiments
(*P < 0.05; **P < 0.01). (B) Western blot analyses of apoptosis-related proteins (p-
2
1

PI3K, AKT, p-AKT, mTOR) separated by SDS-PAGE. Relative band intensity was
determined by Image J software. GAPDH and β-actin served as a loading control.
Images are representative of three independent experiments. Data are mean ± S.E.M.
from three independent experiments (*P < 0.05; **P < 0.01).
2
.2.9. PMMB-317 effectively supress tumor growth of A549 xenografts in nude
mice model
To evaluate the antitumor effects of PMMB-317 in vivo, we subcutaneously
inoculated A549 cells into the left flank of female nude mice. Two weeks later, all the
mice got visible tumors. The mice were then randomized into 5 groups. Changes in
tumor volumes and body weight were measured and recorded after every two days. As
shown in Fig. 8B and C, PMMB-317 inhibited the tumor growth compared with control
in a dose-dependent manner. After the last treatment on day 32, average tumor size of
PMMB-317 (4 mg/kg) was 6-fold smaller than vehicle control group (Fig. S2). The
effect of PMMB-317 was similar to that of Afatinib at 4 mg/kg dose and better than
Shikonin (2 mg/kg) treated group. However, PMMB-317 had no significant effect on
body weight of mice while shikonin caused a significant decrease in body weight
compared with the control group. Therefore, we speculated that the inhibitory effect of
PMMB-317 on tumor growth was increased and its cytotoxicity was reduced
significantly compared with shikonin. Meanwhile, overall survival of mice in PMMB-
3
17-treated groups were greatly increased (Fig. 8E).
2
2

Fig. 8. PMMB-317 suppresses tumor growth of A549 xenografts in nude mice. (A)
Four representative tumors were shown. (B) Body weight of the mice. (C) Tumor
volumes of the mice. (D) The percentage of tumor weight were calculated. (E) Survival
curve of mice. Data are the mean ± S.E.M. of 8 mice per group. (*P < 0.05; **P < 0.01
versus vehicle (DMSO)).
2
.3. Molecular docking
Docking study has confirmed that PMMB-317 conjugated in the colchicine site of
tubulin (PDB: 1SA0) perfectly. As shown in Fig. 9A and B, compound PMMB-317
binds to tubulin (PDB: 1SA0) with CDOCK interaction energy of -62.52 kcal/mol
(Table 1) via three hydrogen bonds with ASN 101 and 258, LYS254. Other weak
interactions, such as van der Waals and carbon-hydrogen bonds, also contribute to the
binding affinity of compound PMMB-317 with tubulin. In addition, as shown in Fig.
2
3

9
C and D, PMMB-317 also binds to EGFR (PDB: 5HG8) with CDOCK interaction
energy of -58.76 kcal/mol (Table 1) via two hydrogen bonds with ARG 841 and CYS
97.
7
Fig. 9. Molecular docking model of compound PMMB-317 with the tubulin (PDB code
SA0) and EGFR (PDB code 5HG8) binding sites. (A) Interactions of PMMB-317 with
1
the amino acid residues at the binding pocket of tubulin (carbon atom, gray; oxygen
atom, red; fluorine atom, light green). (B) Binding pose of PMMB-317 in the protein
surface of tubulin (carbon atom, blue; oxygen atom, red; hydrogen atom, white; fluorine
atom, bright green). Hydrogen bonds were displayed as green dashed lines. (C)
2
4

Interactions of PMMB-317 with the amino acid residues at the binding pocket of EGFR
carbon atom, gray; oxygen atom, red; fluorine atom, light green). (D) Binding pose of
(
PMMB-317 in the protein surface of EGFR (carbon atom, blue; oxygen atom, red;
hydrogen atom, white; fluorine atom, bright green). Hydrogen bond was displayed as
green dashed lines.
2
.4. 3D-QSAR analysis
In order to get a better interaction of small molecules with target protein, 3D-QSAR
was used to optimize the small molecules and the calculations were done by built-in
QSAR software of DS 3.5 (Discovery Studio 3.5, Accelrys, Co. Ltd). Eighteen
compounds were utilized to 3D-QSAR study and pIC was transformed from the
5
0
available IC (μM) values of anti-proliferation assay. Compounds PMMB259[23] and
5
0
8
b[24], have been found in the previous research of our laboratory which had similar
structure to benzoylacrylic acid shikonin ester derivatives that were involved in the 3D-
QSAR study to satisfy the minimum requiments of the computer simulation. All
compounds were divided into the training and test parts, which are shown in Table S1.
2
The correlation co-efficient R of this model was 0.9987, which manifests this model
has an ability to predict. The distribution between predicted pIC and experimental
5
0
pIC value for both test sets and training sets were shown in Fig. 10.
5
0
2
5

Fig. 10. Plot of experimental versus predicted anti-proliferative activities of training set
2
and test set. The model was set up with correlation coefficient R of 0.9962, which
indicated that this model possessed pretty good predicting capability.
Moreover, the molecules aligned with the iso-surfaces of the 3D-QSAR model co-
efficient on van der Waals grids and electrostatic potential grids were also listed in Fig.
1
1. Electrostatic map indicates red contours around regions where high electron density
(negative charge) is expected to increase anti-proliferation activity, and blue contours
represent areas where low electron density (partial positive charge) is expected to
increase anti-proliferation activity. Similarly, steric map indicates green contours
around areas where large steric bulk is predicted to increase activity and yellow
contours around areas where small steric bulk is expected to decrease activity. The map
predicted that the compounds that contained higher negative charge and bulky group
such as F-substituted group in benzene ring would promote the anti-proliferation
activity. This model was based on the actual experiments to guide and optimize the
2
6

tubulin and EGFR inhibitors for the future study.
Fig. 11. (a) 3D-QSAR model coefficients on electrostatic potential grids. Blue
represents positive coefficients; red represents negative coefficients. (b) 3D-QSAR
model coefficients on van der Waals grids. Green represents positive coefficients;
yellow represents negative coefficients.
2
7

3
. Experiments Section
3
.1. Materials and measurements
All chemicals reagents and solvents for the synthesis of the compounds were of
analytical grade, purchased from Nanjing Chemical CO. Ltd (China) and used without
1
further purification, unless otherwise specified. All the H NMR spectra were recorded
1
3
on a Bruker DPX 300 model spectrometer in CDCl . C NMR spectra were recorded
3
1
13
on a Bruker DPX 600 spectrometer in CDCl . Chemical shifts (δ) for H NMR and C
3
NMR spectra were reported in ppm(δ), using TMS as an internal standard. ESI-MS
spectra were obtained on a Mariner Biospectrometry Workstation (ESI-TOF) mass
spectrometer. Melting points (uncorrected) were detected on a XT4 MP Apparatus
(
Taike Corp., Beijing, China), Thin layer chromatography (TLC) was carried out on
glass-backed silica gel plates (Silica Gel 60 Å GF254) and visualized in UV light (λ
54 nm). Column chromatography was performed using silica gel (200-300 mesh) and
2
eluting with ethyl acetate, dichloromethane and petroleum ether (bp 30-60 ℃).
MTT (methyl thiazolyl tetrazolium), PMSF (phenylmethanesulfonyl fluoride),
DAPI (4',6-diamidino-2-phenylindole), RIPA lysis buffers (#P0013B), Anti-tubulin
(#AT819), and Cy3-labeled goat anti-mouse IgG (H+L) (#A0521) were purchased from
Beyotime Institute of Biotechnology (Haimen, China). BCA protein assay kit (#23227)
was purchased from Pierce (Rockford, IL, USA). PVDF membranes were purchased
from Thermo Scientific. JC-1 assay kit (#C2006) was purchased from Beyotime
Institute of Biotechnology (Haimen, China). Anti-Axin-2 (20540-1-AP) was purchased
from Proteintech (Wuhan, China); anti-cleaved PARP-1 (#sc-56196) was purchased
2
8

from Santa Cruz Biotechnology Inc. (Dallas, Texas, USA). anti-GAPDH (#5174) was
purchased from Cell Signaling Technology (Beverly, MA). The secondary antibody
(anti-mouse or anti-rabbit IgG) (#E1WP319 or #E1WP318) were purchased from
Enogene (Nanjing, China). Anti-EGFR (WL03432), Anti-p-EGFR (WL03432), anti-
Raf1 (WL00553), anti-Ras (WL0257), anti-PARP (WL01578), anti-Cleaved PARP
(
WL01932), anti-ERK1/2 (WL01864), anti-CDK1 (WL01718), anti-CDK2
WL02028), anti-p-ERK1/2 (WLP1512), anti-Akt (WL0003b), anti-p-Akt (WLP001a),
(
anti-CyclinA (WL01753), anti-Caspase-9 (WL03229), anti-PI3 Kinase p85
WL02240), anti-Cleaved caspase-9 (WL01838), anti-Caspase-3 (WL01992a), anti-
(
cleaved caspase-3 (WL02348), anti-β-actin (WL01372), anti-mTOR (WL02477), and
anti-β-catenin (WL0962a) were purchased from Wanlei Biotech Co. Ltd (Shenyang,
China), anti-Phospho-c-Raf (Ser338) (56A6) Rabbit mAb (#9427) and anti-Phospho-
PI3 Kinase p85 (Tyr458)/p55 (Tyr199) were purchased from Cell Signaling
Technology (Beverly, MA). The cell cycle and Annexin V-FITC cell apoptosis assay
kits were purchased from KeyGEN Biotech Co. Ltd (Nanjing, China). ECL Kit
(#34077) was purchased from Thermo Scientific (USA). Reactive Oxygen Species
Assay Kit were purchased from YEASEN (Shanghai, China).
3
.2. General procedure for the synthesis of compounds 2a-2o
All the synthesized compounds along with synthetic route were listed in Scheme
. The catalyst 0.02 mol of anhydrous aluminum chloride and 0.01 mol maleic
1
anhydride were dissolved in 40 mL flask, stirring at room temperature 10 min, the
aluminum chloride anhydrous and maleic anhydride were fully dissolved. Then 0.01
2
9