Design, synthesis and biological evaluation of homoerythrina alkaloid derivatives bearing a triazole moiety as PARP-1 inhibitors and as potential antitumor drugs

Article abstract of DOI:10.1016/j.bioorg.2019.103385

A series of homoerythrina alkaloid derivatives containing a 1,2,3-triazole moiety as PARP-1 inhibitors were designed and synthesized. And their anti-proliferative activity was further evaluated. Compound 10n had excellent activity to inhibit proliferation of A549 cells (IC₅₀ = 1.89 μM), which was higher than harringtonine (IC₅₀ = 10.55 μM), pemetrexed (IC₅₀ = 3.39 μM), and rucaparib (IC₅₀ = 4.91 μM). Furthermore, the selectivity index of compound 10n was higher than rucaparib and pemetrexed for lung cancer cells. Flow cytometry analysis showed that compound 10n significantly arrested the cell cycle in the S phase, then induced apoptosis of A549 cells (apoptosis rate is 46%), which effectively inhibited cell proliferation. Simultaneously, western blot analysis revealed that compound 10n could prevent the biosynthesis of PAR. Further analysis results revealed that compound 10n could inhibit the expression of cyclin A, down-regulate the expression of bcl-2/bax, activate caspase-3, and ultimately induce apoptosis of A549 cells. All the results indicated that compound 10n had potential research value as a novel PARP-1 inhibitor in antitumor, and it provided a new reference for further development of PARP-1 inhibitors.

Full text of DOI:10.1016/j.bioorg.2019.103385

Journal Pre-proofs
Design, synthesis and biological evaluation of homoerythrina alkaloid deriva-
tives bearing a triazole moiety as PARP-1 inhibitors and as potential antitu-
mor drugs
Shuai Li, Xin-yang Li, Ting-jian Zhang, Mohamed Olounfeh Kamara, Jing-
wei Liang, Ju Zhu, Fan-hao Meng
PII:
S0045-2068(19)30786-2
DOI:
https://doi.org/10.1016/j.bioorg.2019.103385
YBIOO 103385
Reference:
To appear in:
Bioorganic Chemistry
Received Date:
Revised Date:
Accepted Date:
15 May 2019
8 October 2019
21 October 2019
Please cite this article as: S. Li, X-y. Li, T-j. Zhang, M. Olounfeh Kamara, J-w. Liang, J. Zhu, F-h. Meng,
Design, synthesis and biological evaluation of homoerythrina alkaloid derivatives bearing a triazole moiety as
PARP-1 inhibitors and as potential antitumor drugs, Bioorganic Chemistry (2019), doi: https://doi.org/10.1016/
j.bioorg.2019.103385
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 Published by Elsevier Inc.

Design, synthesis and biological evaluation of homoerythrina alkaloid
derivatives bearing a triazole moiety as PARP-1 inhibitors and as
potential antitumor drugs
Shuai Li, Xin-yang Li, Ting-jian Zhang, Mohamed Olounfeh Kamara,
Jing-wei Liang, Ju Zhu, Fan-hao Meng*
School of Pharmacy, China Medical University, 77 Puhe Road, North New
Area, Shenyang 110122, China
Correspondence: Professor Fan-hao Meng, School of Pharmacy, China
Medical University, Shenyang 110122, P. R. China.
E-mail: fhmeng@cmu.edu.cn
Fax: +86-24-31939448
Abstract
A series of homoerythrina alkaloid derivatives containing a 1,2,3-
triazole moiety as PARP-1 inhibitors were designed and synthesized. And
their anti-proliferative activity was further evaluated. Compound 10n had
excellent activity to inhibit proliferation of A549 cells (IC = 1.89 μM),
5
0
which was higher than harringtonine (IC = 10.55 μM), pemetrexed (IC
5
0
50
=
3.39 μM), and rucaparib (IC = 4.91 μM). Furthermore, the selectivity
5
0
index of compound 10n was higher than rucaparib and pemetrexed for lung
cancer cells. Flow cytometry analysis showed that compound 10n
significantly arrested the cell cycle in the S phase, then induced apoptosis
of A549 cells (apoptosis rate is 46%), which effectively inhibited cell
1

proliferation. Simultaneously, western blot analysis revealed that
compound 10n could prevent the biosynthesis of PAR. Further analysis
results revealed that compound 10n could inhibit the expression of cyclin
A, down-regulate the expression of bcl-2/bax, activate caspase-3, and
ultimately induce apoptosis of A549 cells. All the results indicated that
compound 10n had potential research value as a novel PARP-1 inhibitor in
antitumor, and it provided a new reference for further development of
PARP-1 inhibitors.
Keywords: Homoerythrina alkaloid; PARP-1 inhibitor; Antitumor;
Apoptosis; Drug discovery; 1,2,3-Triazole
1
. Introduction
Poly ADP-ribose polymerase (PARP) plays an essential role in DNA
damage repair and maintenance of genomic stability. PARP inhibitors can
exert a synthetic lethal effect in tumors with specific genetic mutations.
Therefore, PARP is considered an ideal target for cancer chemotherapy [1].
Natural products, especially alkaloids, are one of the most critical
sources of antitumor drugs. Up to now, a lot of natural products and their
derivatives have been widely used in clinical practice, such as paclitaxel,
vinblastine, harringtonine, and homoharringtonine [2-8]. Among them,
harringtonine and homoharringtonine, derived from the genus
Cephalotaxus, have been clinically used for the treatment of cancer for
many years (Figure 1. A) [9-16]. However, some disadvantages, such as
2

lots of adverse effects and low concentration in plants had been reported
17]. Homoerythrina is another alkaloid isolated from Cephalotaxus, which
[
greatly aroused our interest because of its seven-member spiral ring
skeleton, as shown in Figure 1. Studies had reported that homoerythrina-
type alkaloids had weaker anti-proliferative activity than harringtonine-
type alkaloids and homoharringtonine-type alkaloids [18,19]. Therefore,
finding antitumor drugs with homoerythrina as a lead compound is an
interesting and significant study.
3

Figure 1. A: Subordinate diagram of plant constituents B: Chemical
structures of homoerythrina, homoerythrina-type alkaloid, harringtonine,
homoharringtonine, and rucaparib.
In the PARP family, PARP-1 is one of the most abundant proteins and
is considered to be the major isoform. As a promising antitumor target,
PARP-1 inhibitors are one of the hotspots of current research [20,21].
4

PARP-1 inhibitors achieve the purpose of treating cancer by inhibiting
DNA damage repair and DNA synthesis of tumor cells, then inducing
apoptosis of tumor cells [22-24]. In recent years, several PARP-1 inhibitors
had been approved, such as rucaparib, niraparib, and others. Among them,
rucaparib is a first-in-class PARP-1 inhibitor that had been granted by the
FDA in 2016 for the clinical treatment of cancer. 5H-dibenzo[b,e]azepine-
6
,11-dione derivatives are another kind of PARP-1 inhibitors reported by
our team through the structural optimization of rucaparib [25].
Interestingly, as well as homoerythrina, these compounds all possessed
similar benzo[c]azepine fragment (red parts in Figure 2). This inspired us
to find novel PARP-1 inhibitors with homoerythrina as a starting point.
Besides, 1,2,3-triazole is a kind of essential nitrogen heterocyclic
compound which widely applied in the construction of drug molecules,
especially in the field of anticancer [26,27]. Over the past few years, many
anticancer drugs based on the 1,2,3-triazole had been reported in the
literature [28-30]. 1,2,3-triazole structural motif, the product obtained from
the Cu(I)-catalyzed, alkyne–azide 1,3-dipolar cycloaddition (CuAAC)
reaction, has emerged as attractive and desirable scaffolds [26]. As an
essential class of heterocycle, 1,2,3-triazole and its derivatives have
attracted much attention in the field of medicinal chemistry due to their
broad range of pharmacological properties [31-33]. Therefore, we
introduced a triazole group into the homoerythrina scaffold, expecting to
5

obtain a series of novel homoerythrina alkaloid derivatives with
outstanding antitumor ability.
Figure 2. Design strategy of the target compounds
Therefore, in this study, we linked a (1-phenyl-1H-1,2,3-triazol-4-
yl)methyl piece to the N of homoerythrina to mimic the 5-phenyl-1H-
pyrrole piece of rucaparib and the 5-aryl-1,3,4-oxadiazole fragment of 5H-
dibenzo[b,e]azepine-6,11-dione (blue parts in Figure 2), to obtain a series
of novel homoerythrina alkaloid derivatives as PARP-1 inhibitors. Besides,
to simulate the H-bond interactions between the carbonyl group of
rucaparib and Ser904, Gly863 residues of PARP-1 [34], a carbonyl group
was inserted on the benzo[c]azepine ring of homoerythrina (purple parts in
6

Figure 2). Herein, we describe our preliminary results of 22 compounds
that are synthesized and their biological evaluation of the inhibition of
cancer cell proliferation.
2
. Results and discussion
2
.1 Chemistry
The
3,4-dihydrospiro[benzo[c]azepine-1,1'-cyclohexane]-5(2H)-one
derivatives (10a-10v) were synthesized by the synthesis method illustrated
in Scheme 1.
Among them, the preparation of 1-phenylcyclohexan-1-amine
(compound 4) was done according to the reported method by Theodoros S
[35,36]. Based on the literature, we purified the compound by extracting
impurities. Compound 4 underwent the Michael addition reaction, the
Hydrolysis reaction, the Friedel-Crafts reaction and the Alkylation reaction
[37-39] to form compound 9, and then it was reacted with the
corresponding azidobenzene by the click chemistry to give the target
compounds (10a-10v). The click chemistry is the classic method of
constructing 1,2,3-triazole. The yields and operational simplicity of this
method made it a particularly attractive protocol for laboratory-scale
synthesis. Compound 8 was considered to be the key intermediates to
synthesize the target compounds. In the post-treatment process, we
operated under low-temperature conditions to prevent danger.
In the process of synthesizing the target compounds (10a-10v), we used
7

a microwave reactor to accelerate the reaction rate and obtained superior
results. And, the procedure was straightforward. The experimental and
post-treatment methods for most of the compounds were different from
those reported in the literature and had described explicitly in the general
methods.
Br
ⅰ
ⅱ
ⅲ
ⅳ
OH
NH2
N3
2
3
4
1
ⅴ
ⅵ
ⅶ
HN
O
O
HN
O
HN
O
7
Cl
6
OH
5
N
N
N
NH
ⅷ
ⅸ
N
R
N
click chemistry
O
O
O
8
9
10a-10v
b, R=2-Br e, R=2-CH3 h, R=2-Cl k, R=2-OCH3
n, R=2-F
o, R=3-F
q, R=2-OH t, R=2-CN
a, R=H c, R=3-Br f, R=3-CH3 i, R=3-Cl
d, R=4-Br g, R=4-CH3 j, R=4-Cl
l, R=3-OCH3
r, R=3-OH
u, R=3-CN
v, R=4-CN
m, R=4-OCH3 p, R=4-F s, R=4-OH
Scheme 1. Reagents and conditions: (ⅰ) Mg, cyclohexanone,
ethoxyethane, 35 ℃, 3 h, 95.5%; (ⅱ) NaN , 20 ℃, 12 h; (ⅲ) LiAlH , 40 ℃,
3
4
1
2 h, 89.43%; (ⅳ) Methyl-acrylate, no solvent required, 40 ℃, 12 h, 79.5%;
(
ⅴ) NaOH, water, 50 ℃, 2 h, 97.15%; (ⅵ) SOCl , 35 ℃, 2 h; (ⅶ) AlCl ,
2
3
2
5 ℃, 24 h, 35.2%; (ⅷ) 3-bromoprop-1-yne, 70 ℃, 12 h, 89.96%; (ⅸ)
Corresponding azidobenzene, 8 minutes, 35 ℃, 75%-85%.
.2 Biological evaluation
2
8

2
.2.1 MTT assay and SAR analysis.
Taking harringtonine, pemetrexed, and rucaparib as reference drugs, the
target compounds (10a-10v) were evaluated for the anti-proliferative
activity against four cancer cell lines: A549, OVCAR-3, HCT-116, and
MCF-7 by 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyltetrazolium bromide
(
MTT) assay. The IC values were used to express the results and were
5
0
summarized in Table 1.
Table 1
The IC values of compounds, harringtonine, pemetrexed, and rucaparib
5
0
against A549, OVCAR-3, HCT-116, and MCF-7 cell lines.
Compounds
IC50 [a](µM)
A549
OVCAR-3
HCT-116
> 50
MCF-7
Compound 8
39.78 ± 0.72
5.41 ± 0.17
4.51 ± 0.71
7.67 ± 0.96
9.86 ± 0.22
9.02 ± 0.24
9.94 ± 0.26
15.88 ± 0.06
6.65 ± 0.42
9.94 ± 0.26
36.57 ± 1.01
4.49 ± 0.46
7.08 ± 0.73
11.31 ± 0.77
1.89 ± 0.23
5.58 ± 0.70
14.9 ± 2.59
6.16 ± 0.84
7.53 ± 0.73
> 50
> 50
> 50
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0a
0b
0c
0d
0e
0f
2.83 ± 0.38
7.11 ± 0.75
10.44 ± 0.53
37.44 ± 1.09
13.81 ± 1.47
14.32 ± 3.47
16.71 ± 0.88
5.08 ± 0.43
11.08 ± 0.83
22.91 ± 1.68
9.63 ± 0.84
20.15 ± 1.21
23.69 ± 0.48
4.29 ± 0.11
4.96 ± 0.42
18.74 ± 0.95
31.76 ± 0.99
32.61 ± 2.90
> 50
3.35 ± 0.47
5.87 ± 0.79
8.36 ± 0.42
8.89 ± 0.52
6.31 ± 0.72
7.12 ± 0.66
7.20 ± 1.94
5.14 ± 0.48
7.03 ± 1.23
23.20 ± 2.90
4.20 ± 0.16
13.68 ± 0.86
31.72 ± 3.08
3.97 ± 0.70
4.33 ± 0.39
4.93 ± 0.59
> 50
3.99 ± 0.51
6.96 ± 0.77
13.64 ± 0.60
17.02 ± 0.20
8.70 ± 0.78
13.58 ± 0.24
43.81 ± 1.65
4.14 ± 0.80
7.27 ± 0.43
18.55 ± 1.42
9.81 ± 0.87
12.42 ± 0.61
39.97 ± 1.27
4.12 ± 0.62
8.86 ± 0.99
9.01 ± 4.68
> 50
0g
0h
0i
0j
0k
0l
0m
0n
0o
0p
0q
0r
0s
7.84 ± 0.42
> 50
20.79 ± 0.65
> 50
0t
> 50
> 50
> 50
> 50
0u
> 50
> 50
> 50
> 50
9

1
0v
> 50
> 50
> 50
> 50
Harringtonine
Pemetrexed
Rucaparib
10.55 ± 0.26
3.39 ± 0.15
4.91 ± 0.11
10.87 ± 0.08
7.45 ± 0.60
2.41 ± 0.13
11.13 ± 0.11
4.74 ± 0.22
13.51 ± 0.17
11.72 ± 0.15
4.18 ± 0.04
12.98 ± 0.28
[
a] IC values were used to express inhibition effect (IC = Mean ± SD).
5
0
50
From MTT assay after 24 h of treatment; the values were average from at
least 3 independent experiments.
From the results of MTT assay, we found that when R was replaced by
2
-F in all compounds, the compound 10n had the best anti-proliferative
activity (IC = 1.89 ± 0.23 μM) than the other compounds, which was
5
0
excellent for A549 cells. For the other target compounds, most compounds
of the electron-withdrawing group (-F, -Cl, -Br) were superior to the
electron-donating group (-CH , -OCH , -OH) in terms of anti-proliferative
3
3
activity. Although -CN (10t-10v) was an electron-withdrawing group, its
activity was the worst among all compounds. We believed that this
phenomenon was caused by the massive steric hindrance of the -CN, which
is linear, and its spatial conformation was more affected than the other
electron-donating groups, such as -OCH , so its activity was the worst
3
among all compounds.
Simultaneously, for the electron-withdrawing group, the F group (10n-
1
0p) was superior to the Cl group (10h-10j) and the Br group (10b-10d).
For the electron-donating group, when R was replaced by -OCH (10k-
3
1
0m), the anti-proliferative activity of most compounds was better than
replaced by -CH (10e-10g) and -OH (10q-10s). For all target compounds,
3
1
0

we found that most compounds with ortho-substituent had the superior
anti-proliferative activity over meta-substituent and para-substituent
compounds.
In conclusion, the results showed that when R was replaced by 2-F, the
compound 10n had the best anti-proliferative activity (IC = 1.89 ± 0.23
5
0
μM) against A549 cells in all target compounds, and the electron-
withdrawing group (-F, -Cl, -Br) was superior to the electron-donating
group (-CH , -OCH , -OH). Most compounds with ortho-substituent
3
3
showed better activity. The compound 10n was expected to be a novel lead
compound in anti-tumor.
The MTT assay results indicated that compound 10n had a better
inhibitory effect on lung cancer cells than the reference drug. MTT assay
was performed on healthy cells corresponding to these four cancer cells,
and the safety of the compound 10n was evaluated by the selectivity index
(SI). As shown in Table 2, the compound 10n had a higher SI than the
other three cells for HPAEpiC, and its SI even higher than pemetrexed.
This indicated that compound 10n had the characteristics of potent
inhibition and high safety, and it had the value of continuing research on
lung cancer. Therefore, this article focuses on the effect of compound 10n
on A549 cells.
Table 2
Selectivity index of compound 10n, rucaparib, and pemetrexed against
1
1

HPAEpiC, IOSE80, HIEC, and HTB-125 cell lines.
Compound
SI (Selectivity index) [b]
HPAEpiC [c]
15.98
IOSE80 [d]
HIEC [e]
HTB-125 [f]
7.84
1
0n
11.61
6.94
Rucaparib
8.32
11.86
3.98
4.32
Pemetrexed
6.50
2.92
4.83
5.95
From MTT assay after 24 h of treatment; the values were average from at
least 3 independent experiments.
[
[
[
[
[
b] SI = CC / IC ;
5
0
50
c] Human normal alveolar epithelial cells;
d] Human normal ovarian epithelial cells;
e] Human normal intestinal epithelial cells;
f] Human normal breast cell.
2
.2.2 Compound 10n could arrest cell cycle in the S phase.
PARP-1 is a crucial enzyme in DNA replication and repair. Extensive
research showed that PARP-1 inhibitors could inhibit cell cycle
progression, then achieve the purpose of inducing apoptosis [23]. This also
makes the study of PARP-1 inhibitors essential. MTT assay results showed
that most of the compounds had excellent activity against cancer cells.
Among them, compound 10n as a hitting compound had a very eye-
catching inhibitory potential.
To further investigate whether the inhibitory effect of the compound 10n
on cell viability was associated with cell cycle inhibition, we used flow
cytometry to analyze the cell cycle. A549 cells were treated with different
concentrations of the compound 10n (0, 1.0, 2.0, 4.0 μM) for 24 h, stained
1
2

with propidium iodide (PI), and then analyzed for cell cycle distribution by
flow cytometry. The results of flow cytometry showed that with the
increase of concentration, the percentage of the cells of the S phase
increased significantly, and the G0/G1 phase and G2/M phase decreased
significantly (Figure 3. B). A large number of cells could not enter the
G2/M phase through the S phase, which in turn causes cell cycle arrest in
the S phase.
These results implied that compound 10n could cause cancer cell cycle
arrest in the S phase, thereby inhibiting the proliferation of A549 cells. The
result is significantly different (P < 0.001).
1
3

Figure 3. Compound 10n could cause cell cycle arrest in A549 cells. (A):
Effect of the compound 10n on cell cycle progression of A549 cells was
evidenced by PI staining and FACS analysis. (B): Percentage of cells in
1
4

each period after treatment of A549 cells with different concentrations of
the compound 10n. Three independent experiments were performed, each
column represented the mean ± SD of triplicate determinations (*P < 0.05;
*
*P < 0.01 and ***P < 0.001. Compound 10n-treated group VS control).
.2.3 Compound 10n could induce apoptosis to further inhibit cell
2
proliferation.
Apoptosis is an essential pathway for cell death and is critical for
maintaining cell homeostasis. Induction of apoptosis is an essential
strategy for tumor suppression. If DNA repair is impaired, the cell will die
when the cell cycle is inhibited. PARP-1 is a necessary enzyme in the
process of DNA repair. When its function is inhibited, the arrested cell
cycle cannot be repaired and then proceeds to apoptosis [40]. To verify that
whether compound 10n eventually caused apoptosis when causing cycle
arrest, flow cytometry analysis was performed.
The concentration and treatment conditions as the above experiment,
A549 cells were stained with Annexin V/ PI. Flow cytometry results
showed that the apoptotic rate of A549 cells was 3.1%, 8.3%, 23.1%, and
4
6.0% for each concentration, respectively (Figure 4. A). With the
increase in concentration, early apoptotic rate and late apoptotic rate
significantly increased (Figure 4. B).
It was suggested that compound 10n could induce apoptosis to inhibit
the proliferation of A549 cells further. The result is significantly different
1
5

(P < 0.001).
Figure 4. Compound 10n could induce apoptosis in A549 cells. (A): The
1
6

effect of compound 10n on A549 cell death was evidenced by Annexin V-
FITC/PI double staining and FACS analysis. Apoptotic cells were Annexin
V [+] and PI [-], late apoptotic cells were Annexin V [+] and PI [+],
necrotic cells were Annexin V [-] and PI [+] and living cells were Annexin
V [-] and PI [-]. (B): Percentage of apoptotic cells after treatment of A549
cells with different concentrations of the compound 10n. Three
independent experiments were performed, each column represented the
mean ± SD of triplicate determinations (*P < 0.05; **P < 0.01 and ***P <
0
.001. Compound 10n-treated group VS control).
2
.2.4 Verification of the target of compound 10n at the protein level.
Studies had shown that inhibition of PARP-1 could reduce DNA repair
function and enhance the therapeutic effect of radiotherapy and
chemotherapy on tumors. PAR (Poly (ADP-ribose)) is the active product
of PARP-1. When PARP-1 is inhibited, the amount of biosynthesis of PAR
will decrease, thereby inhibiting cell cycle progression and further
inducing apoptosis [41].
We directly tested the expression of PAR and PARP-1 using western
blot. And the results showed that with the increase of the concentration of
compound 10n, the expression of PARP-1 was unchanged, but the
biosynthesis of PAR decreased in a concentration-dependent manner
(Figure 5. A(i)). Simultaneously, at the same 4.0 μM concentration,
compound 10n was more potent than rucaparib in inhibiting PARP-1
1
7

enzyme activity and had greater anti-tumor proliferative potential (Figure
5
. A(ⅱ)).
Combined with the above results, the compound 10n can inhibit the
enzymatic activity of PARP-1 more effectively, thereby reducing the
biosynthesis of PAR, and achieving the purpose of inhibiting the
proliferation of A549 cells. This also makes the compound 10n have the
value of continuing research. The result is significantly different (P <
0
.001).
1
8

Figure 5. (A)(i): Expression of PAR and PARP-1 protein at different
concentrations of compound 10n in A549; (ii): Effect of rucaparib and
compound 10n on the protein expression of PAR and PARP-1 at 4.0 μM
in A549. (B): Effect of compound 10n on the cyclin A and the apoptotic
1
9

protein expression of caspase-3, bcl-2/bax and cleaved caspase-3 in A549.
Each column represented the mean ± SD of triplicate determinations (*P <
0
.05; **P < 0.01 and ***P < 0.001. Compound 10n -treated group VS
control).
.2.5 Effect of compound 10n on the cyclin A protein and the apoptotic
2
protein expression of caspase-3, bcl-2/bax and cleaved caspase-3 in A549.
Studies had shown that when PARP-1 is inhibited, DNA synthesis is
arrested, thereby inhibiting the cycle of cancer cells, triggering the process
of apoptosis. Therefore, to further verify the specific mechanism by which
compound 10n inhibits cancer cell proliferation, we explored the
expression of cycle-associated proteins and apoptosis-related proteins.
From the results of flow cytometry analysis, the cell cycle progression
of A549 cannot be adequately performed due to incomplete DNA
replication, and most cells could not enter the division phase (G2/M), and
then got arrested in the S phase. Numerous studies had shown that cyclin
A is a critical protein in the S phase checkpoint, so we performed western
blot analysis on cyclin A of A549 cells [42]. And, the results of Annexin
V/PI staining showed that the compound 10n could induce apoptosis.
Therefore, we also further performed western blot analysis on the
expression of apoptotic proteins.
Western blot results showed that in Figure 5, the relative protein
expression of cyclin A decreased in a concentration-dependent manner
2
0

(Figure 5. B). Moreover, the ratio of bcl-2/bax was significantly reduced
and activated protein expression of caspase-3, but there was no significant
difference at the concentration of 1.0 μM.
The results showed that compound 10n inhibited the expression of
cyclin A, down-regulated the expression of bcl-2/bax, activated caspase-3,
and ultimately caused inhibition of cell cycle progression and activation of
the apoptosis pathway. The result is significantly different (P < 0.001).
3
. Conclusions
We designed and synthesized a series of homoerythrina alkaloid
derivatives bearing a triazole moiety as PARP-1 inhibitors. And we
evaluated their anticancer activity against four human cancer cell lines
(A549, OVCAR-3, HCT-116, and MCF-7). These compounds all had the
novel seven-membered spiro skeleton as homoerythrina. Most of the target
compounds showed better activity, and the compound 10n had excellent
inhibitory effect on A549 cells (IC = 1.89 μM) than harringtonine (IC =
5
0
50
1
0.55 μM), pemetrexed (IC = 3.39 μM), and rucaparib (IC = 4.91 μM).
5
0
50
The SI of the compound 10n was higher than pemetrexed. Flow cytometry
analysis indicated that compound 10n significantly arrested the cell cycle
in the S phase, then induced apoptosis of A549 cells (apoptosis rate is 46%).
Western blot analysis manifested that compound 10n could inhibit the
enzymatic activity of PARP-1, thereby preventing the biosynthesis of PAR.
Further analysis results revealed that compound 10n could inhibit the
2
1

expression of cyclin A, down-regulate the expression of bcl-2/bax, activate
caspase-3, and ultimately achieve the purpose of inducing apoptosis.
In conclusion, we successfully modified the structure of homoerythrina
and found that compound 10n, as a novel PARP-1 inhibitor, had a better
antitumor ability by activating the apoptosis signaling pathway. The
compound 10n provided a better choice for drug research and the
development of novel PARP-1 inhibitors.
4
. Experimental procedures
4
.1 General methods.
Unless otherwise indicated, reagents and solvents were purchased from
commercial sources and used without further purification. All reactions
were monitored by TLC using silica gel aluminum cards (0.2 mm thickness)
with a fluorescent indicator at 254 nm. The column chromatography was
performed using silica gel (200-300 mesh) from Qingdao Ocean Chemicals
(Qingdao, Shandong, China). Melting points were recorded on the RD-1
1
13
melting apparatus and were uncorrected. The spectra of H-NMR and C-
1
NMR were recorded on the Varian NMR spectrometer operating at H at
1
3
6
00 MHz and C at 150 MHz. DMSO-d was used as the solvent to
6
measure all chemical shifts. High-resolution mass spectra (HRMS) were
measured in an ESI mode of an Agilent Accurate-Mass Q-TOF 6540
(Agilent, Santa Clara, CA, USA).
4
.1.1 The synthesis of 1-phenylcyclohexan-1-ol.
2
2

Mg (8.13 g, 0.33 mol), catalyst amount of I , and compound 1 (10.00 g,
2
0
.064 mol) were added to the diethyl ether at room temperature, and the
mixture was heated to 35 ℃. Then, the solution of compound 1 (40.00 g,
.25 mol) in 50 mL of diethyl ether was dropped into the above mixture
0
over 0.5 h, and then cyclohexanone (34.38 g, 0.35 mol) was added to the
reaction system. The mixture was stirred at room temperature for 3 h. This
series of experimental operations needed to be carried out under anhydrous
and anaerobic conditions. After the reaction was completed, the mixture
was quenched with saturated aqueous ammonium chloride (100 mL). Next,
the mixture was extracted with Et2O (3 × 100 mL), and the combined
extracts were dried (MgSO4). Evaporation of the solvent to give crude
compound 2 as yellow oil. The crude product was purified by column
chromatography. Yield: 95.50%.
4
.1.2 The synthesis of 1-phenylcyclohexan-1-amine.
Sodium azide (8.11 g, 0.12 mol) and trifluoroacetic acid (25.88 g, 0.23
mol) were added to the solution of compound 2 (20.00 g, 0.11 mol) in DCM,
and this step was carried out at -5 ℃. The reaction was stirred at room
temperature for 12 h. Evaporation the solvent and dissolved it with
tetrahydrofuran. Note that a large amount of saturated anhydrous sodium
carbonate solution should be added before the solvent was evaporated to
destroy the formed hydrazoic acid and perform the liquid separation. The
mixture was then added to lithium aluminum hydride (4.77 g, 0.13 mol)
2
3

dissolved in tetrahydrofuran at -5 ℃. The reaction was stirred at 40 ℃ for
2 h. After the reaction was completed, water was added dropwise to the
1
reaction system at -5 ℃ to achieve the purpose of quenching lithium
aluminum hydride. The precipitate was then suction filtered, and the
solvent was evaporated in vacuo. Next, the pH of the mixture was adjusted
to 2-3. And then the mixture was extracted with Et2O (3 × 100 mL) to
remove impurities. After that, the pH of the aqueous layer should be
adjusted to 9-10. The mixture was extracted with DCM (3 × 100 mL), and
the combined extracts were dried (MgSO4). Evaporation of the solvent to
give compound 4 as white oil. Yield: 89.43%.
4
.1.3 The synthesis of methyl 3-((1-phenylcyclohexyl) amino) propanoate.
The compound 4 (20.00 g, 0.11 mol), DBU (8.69 g, 0.057 mol) and
methyl acrylate (10.81 g, 0.13 mol) were added to a 250 mL reaction flask
at 40 ℃ for 12 h to give crude compound 5 as white oil. The crude product
was purified by column chromatography. Yield: 79.5%. Note that no
solvent was needed for this experimental procedure.
4
.1.4 The synthesis of 3-((1-phenylcyclohexyl)amino)propanoic acid.
To a solution of compound 5 (15.00 g, 0.057 mol) in water, sodium
hydroxide (4.59 g, 0.115 mol) was added, and the reaction was stirred at
0 ℃ for 2 h. As the reaction progresses, compound 5 gradually dissolved
5
in the form of sodium salt in water. Then we adjusted the pH of the solution
to 6 and evaporated the solvent to give compound 6 as white oil, which
2
4

was used directly in the next step. Yield: 97.15%.
.1.5 The synthesis of 3,4-dihydrospiro[benzo[c]azepine-1,1'-
4
cyclohexan]-5(2H)-one.
To a solution of compound 6 (20.00 g, 0.081 mol) in DCM, thionyl
chloride (19.24 g, 0.162 mol) was added, and the reaction was stirred at 35
℃
for 2 h. Evaporation of the solvent and dissolved it with DCM. Then we
added anhydrous aluminum trichloride (12.94 g, 0.097 mol) to the reaction
system. The reaction was stirred at 25 ℃ for 24 h to give crude compound
8
as white oil. During the post-treatment, the reaction solution was slowly
poured into ice water and continuously stirred, and then liquid separation
was carried out. Finally, the organic layer was alkalized for extraction. The
crude product was purified by column chromatography. Yield: 35.2%.
4
.1.5.1
compound 8)
A white powder, yield: 35.2%. Mp: 101.6-103.4 ℃. H NMR (600 MHz,
DMSO-d ) δ 7.50 (s, 1H), 7.41 (s, 1H), 7.22 (d, J = 8 Hz, 2H), 3.02 (s, 2H),
3,4-dihydrospiro[benzo[c]azepine-1,1'-cyclohexan]-5(2H)-one
(
1
6
1
3
2
1
2
2
4
.71 (s, 2H), 1.60 (m, 10H); C NMR (150 MHz, DMSO-d ) δ 206.34,
6
50.37, 140.53, 130.48, 127.94, 127.35, 126.54, 60.53, 46.22, 37.17, 36.85,
+
5.43, 21.80. ESI-HRMS calcd. for C H NO [M + H] 230.1539, found:
1
5
20
30.1550.
.1.6
The
synthesis
of
2-(prop-2-yn-1-yl)-3,4-
dihydrospiro[benzo[c]azepine-1,1'-cyclohexan]-5(2H)-one.
2
5

To a solution of compound 8 (1.00 g, 0.004 mol) in acetonitrile, 3-
bromoprop-1-yne (0.57 g, 0.005 mol) was added, and the reaction was
stirred at 70 ℃ for 12 h to give crude compound 9 as yellow powder. This
step required the addition of an equal amount of K CO as an acid-binding
2
3
agent. The crude product was purified by column chromatography. Yield:
8
9.96%.
4
.1.7 General procedure for the synthesis of compounds 10a-10v.
Hydrogen chloride (2.68 g, 0.028 mol), sodium nitrite (1.63 g, 0.024 mol)
and sodium azide (1.54 g, 0.024 mol) were sequentially added to the
corresponding aqueous solution of aniline (2.00 g, 0.021 mol) every 0.5 h
and the reaction was stirred at 0 ℃ for 5 h to give the corresponding
azidobenzene. Then, to a solution of compound 9 (0.50 g, 0.004 mol) in
DMF, corresponding azidobenzene, anhydrous copper sulfate (0.20 g), and
Vitamin C (0.20 g) were added. The reaction was placed in a microwave
reactor and kept stirring at 35 ℃ for 8 minutes to give crude compounds
1
0a-10v. The crude product was purified by column chromatography.
.1.7.1 2-((1-phenyl-1H-1,2,3-triazol-4-yl)methyl)-3,4-
4
dihydrospiro[benzo[c]azepine-1,1'-cyclohexan]-5(2H)-one (10a)
1
A white solid, yield: 83.6%. Mp: 153.5-154.2 ℃. H NMR (600 MHz,
DMSO-d ) δ 8.67 (s, 1H), 7.89 (d, J = 8.0 Hz, 2H), 7.58 (t, J = 7.9 Hz, 2H),
6
7
3
.49–7.42 (m, 3H), 7.34–7.30 (m, 1H), 7.26–7.22 (m, 1H), 3.68 (s, 2H),
.30 (s, 2H), 2.93 (s, 2H), 2.25 (d, J = 12.2 Hz, 2H), 1.79–1.69 (m, 2H),
2
6

1
3
1
.59 (m, 3H), 1.49–1.40 (m, 2H), 1.27–1.14 (m, 1H); C NMR (150 MHz,
DMSO-d ) δ 208.05, 147.38, 143.78, 142.40, 137.20, 130.67, 130.30,
6
1
28.91, 128.42, 127.30, 126.50, 121.96, 120.37, 61.75, 41.92, 41.08, 40.65,
+
3
6.82, 25.84,22.05. ESI-HRMS calcd. for C H N O [M + H] 387.2179,
2
4
27
4
found: 387.2179.
4
.1.7.2
2-((1-(2-bromophenyl)-1H-1,2,3-triazol-4-yl)methyl)-3,4-
dihydrospiro[benzo[c]azepine-1,1'-cyclohexan]-5(2H)-one (10b)
1
A gray solid, yield: 80.9%. Mp: 142.6-145.1 ℃. H NMR (600 MHz,
DMSO-d ) δ 7.64–7.59 (m, 3H), 7.57–7.52 (m, 2H), 7.50–7.42 (m, 2H),
6
7
2
.36–7.30 (m, 1H), 7.25 (m, 1H), 3.71 (s, 2H), 3.32 (s, 2H), 2.30–2.22 (m,
H), 1.76 (d, J = 11.9 Hz, 2H), 1.67–1.55 (m, 4H), 1.48–1.41 (m, 2H), 1.17
1
3
(
t, J = 7.1 Hz, 2H); C NMR (150 MHz, DMSO-d ) δ 208.06, 146.29,
6
1
1
1
4
4
43.86, 142.41, 134.10, 134.06, 132.24, 130.67, 129.37, 129.11, 128.43,
27.28, 126.50, 126.02, 119.27, 61.77, 60.23, 41.95, 25.85, 22.05, 21.22,
+
4.55. ESI-HRMS calcd. for C H BrN O [M + H] 465.1290 and
2
4
26
4
67.1290, found: 465.1271 and 465.1271.
.1.7.3 2-((1-(3-bromophenyl)-1H-1,2,3-triazol-4-yl)methyl)-3,4-
dihydrospiro[benzo[c]azepine-1,1'-cyclohexan]-5(2H)-one (10c)
1
A gray solid, yield: 80.0%. Mp: 141.4-143.7 ℃. H NMR (600 MHz,
DMSO-d ) δ 8.18 (m, 2H), 7.97 (m, 2H), 7.75–7.64 (m, 2H), 7.55 (m, 2H),
6
7
2
.46 (d, J = 5.3 Hz, 1H), 3.69 (s, 2H), 3.32 (s, 2H), 2.26 (d, J = 13.7 Hz,
1
3
H), 1.81–1.41 (m, 5H), 1.20 (m, 5H); C NMR (150 MHz, DMSO-d ) δ
6
2
7

2
1
3
4
4
08.01, 143.26, 138.05, 132.31, 132.02, 128.42, 128.35, 127.35, 127.30,
23.82, 123.20, 122.92, 122.16, 119.62, 119.29, 64.39, 60.54, 60.22, 56.06,
+
5.96, 21.21, 14.53. ESI-HRMS calcd. for C H BrN O [M + H]
2
4
26
4
65.1290 and 467.1290, found: 465.1276 and 467.1276.
.1.7.4 2-((1-(4-bromophenyl)-1H-1,2,3-triazol-4-yl)methyl)-3,4-
dihydrospiro[benzo[c]azepine-1,1'-cyclohexan]-5(2H)-one (10d)
1
A gray solid, yield: 81.6%. Mp: 152.4-154.5 ℃. H NMR (600 MHz,
DMSO-d ) δ 7.88 (m, 3H), 7.84–7.75 (m, 2H), 7.48–7.42 (m, 2H), 7.35–
6
7
2
.28 (m, 1H), 7.24 (d, J = 7.0 Hz, 1H), 3.67 (s, 2H), 3.31 (s, 2H), 2.21–
.19 (m, 2H), 1.79–1.66 (m, 2H), 1.60 (m, 3H), 1.44 (m, 2H), 1.18 (m, 3H);
1
3
C NMR (150 MHz, DMSO-d ) δ 208.04, 154.73, 143.27, 136.14, 133.27,
6
1
6
33.18, 128.42, 127.31, 126.50, 123.69, 122.59, 122.27, 121.98, 64.39,
0.55, 60.23, 41.07, 22.05, 21.23, 14.54. ESI-HRMS calcd. for
+
C H BrN O [M + H] 465.1290 and 467.1290, found: 465.1266 and
2
4
26
4
4
67.1266.
4
.1.7.5
2-((1-(o-tolyl)-1H-1,2,3-triazol-4-yl)methyl)-3,4-
dihydrospiro[benzo[c]azepine-1,1'-cyclohexan]-5(2H)-one (10e)
1
An off-white solid, yield: 77.2%. Mp: 136.9-138.6 ℃. H NMR (600
MHz, DMSO-d ) δ 8.32 (s, 1H), 7.49–7.44 (m, 5H), 7.42–7.39 (m, 2H),
6
7
2
2
.26–7.23 (m, 1H), 3.70 (s, 2H), 2.92 (s, 2H), 2.26 (d, J = 12.1 Hz, 2H),
.13 (s, 3H), 1.98 (s, 1H), 1.78–1.69 (m, 2H), 1.64–1.55 (m, 3H), 1.45 (m,
1
3
H), 1.25–1.19 (m, 2H);
C NMR (150 MHz, DMSO-d ) δ 208.10,
6
2
8

1
46.37, 143.92, 142.41, 136.89, 133.42, 131.78, 130.67, 130.10, 128.42,
27.40, 127.28, 126.50, 126.43, 125.43, 61.78, 42.11, 41.15, 40.82, 25.84,
1
+
2
2.04, 21.23, 17.91. ESI-HRMS calcd. for C H N O [M + H] 401.2336,
2
5
29
4
found: 401.2342.
4
.1.7.6
2-((1-(m-tolyl)-1H-1,2,3-triazol-4-yl)methyl)-3,4-
dihydrospiro[benzo[c]azepine-1,1'-cyclohexan]-5(2H)-one (10f)
1
An off-white solid, yield: 74.3%. Mp: 135.3-137.2 ℃. H NMR (600
MHz, DMSO-d ) δ 8.64 (s, 1H), 7.73 (s, 1H), 7.68 (d, J = 7.9 Hz, 1H), 7.44
6
(m, 3H), 7.34–7.30 (m, 1H), 7.28–7.23 (m, 2H), 3.67 (s, 2H), 2.91 (s, 2H),
2
.40 (s, 3H), 2.24 (d, J = 12.2 Hz, 2H), 1.97 (d, J = 5.5 Hz, 1H), 1.78–1.69
1
3
(
m, 2H), 1.63–1.55 (m, 3H), 1.47–1.40 (m, 2H), 1.25–1.14 (m, 2H); C
NMR (150 MHz, DMSO-d ) δ 208.03, 147.28, 143.77, 142.39, 140.03,
6
1
1
37.17, 130.66, 130.07, 129.47, 128.41, 127.29, 126.49, 121.88, 120.75,
17.46, 61.74, 41.90, 41.07, 40.64, 25.84, 22.06, 21.37, 14.54. ESI-HRMS
+
calcd. for C H N O [M + H] 401.2336, found: 401.2339.
2
5
29
4
4
.1.7.7
2-((1-(p-tolyl)-1H-1,2,3-triazol-4-yl)methyl)-3,4-
dihydrospiro[benzo[c]azepine-1,1'-cyclohexan]-5(2H)-one (10g)
1
An off-white solid, yield: 79.6%. Mp: 141.2-143.5 ℃. H NMR (600
MHz, DMSO-d ) δ 8.62 (s, 1H), 7.78 (t, J = 6.6 Hz, 2H), 7.49–7.43 (m,
6
2
2
1
H), 7.39 (m, 2H), 7.35–7.31 (m, 1H), 7.25 (d, J = 7.3 Hz, 1H), 3.67 (s,
H), 2.91 (s, 2H), 2.38 (s, 3H), 2.26 (d, J = 12.7 Hz, 2H), 2.00–1.95 (m,
H), 1.80–1.69 (m, 2H), 1.60 (m, 3H), 1.46 (m, 2H), 1.27–1.15 (m, 2H);
2
9