Identification and optimization of piperlongumine analogues as potential antioxidant and anti-inflammatory agents via activation of Nrf2

Article abstract of DOI:10.1016/j.ejmech.2020.112965

Oxidative stress and inflammation are significant risk factors for neurodegenerative disease. The Keap1-Nrf2-ARE pathway is one of the most promising defensive systems against oxidative stress. Here, dozens of piperlongumine analogues were designed, synth

Full text of DOI:10.1016/j.ejmech.2020.112965

Journal Pre-proof
Identification and optimization of Piperlongumine analogues as potential antioxidant
and anti-inflammatory agents via activation of Nrf2
Limei Ji, Lailiang Qu, Cheng Wang, Wan Peng, Shang Li, Huali Yang, Heng Luo,
Fucheng Yin, Dehua Lu, Xingchen Liu, Lingyi Kong, Xiaobing Wang
PII:
S0223-5234(20)30937-5
DOI:
https://doi.org/10.1016/j.ejmech.2020.112965
EJMECH 112965
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 28 August 2020
Revised Date: 14 October 2020
Accepted Date: 21 October 2020
Please cite this article as: L. Ji, L. Qu, C. Wang, W. Peng, S. Li, H. Yang, H. Luo, F. Yin, D. Lu, X. Liu,
L. Kong, X. Wang, Identification and optimization of Piperlongumine analogues as potential antioxidant
and anti-inflammatory agents via activation of Nrf2, European Journal of Medicinal Chemistry, https://
doi.org/10.1016/j.ejmech.2020.112965.
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.
©
2020 Elsevier Masson SAS. All rights reserved.

Graphical abstract:

1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
Identification and optimization of Piperlongumine analogues as potential
antioxidant and anti-inflammatory agents via activation of Nrf2
Limei Ji, Lailiang Qu, Cheng Wang, Wan Peng, Shang Li, Huali Yang, Heng Luo,
Fucheng Yin, Dehua Lu, Xingchen Liu, Lingyi Kong*, Xiaobing Wang *
1
1
1
1
1
1
1
Jiangsu Key Laboratory of Bioactive Natural Product Research and State Key
Laboratory of Natural Medicines, Department of Natural Medicinal Chemistry, School
of Traditional Chinese Pharmacy, China Pharmaceutical University, Nanjing 210009,
China
1
1
7
8
*Corresponding Authors.
Tel/Fax: +86-25-83271405;
1
2
9
0
E-mail: xbwang@cpu.edu.cn (Xiaobing Wang); cpu_lykong@126.com (Lingyi
Kong)
2
1

2
2
Abstract
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
Oxidative stress and inflammation are significant risk factors for
neurodegenerative disease. The Keap1-Nrf2-ARE pathway is one of the most
promising defensive systems against oxidative stress. Here, dozens of piperlongumine
analogues were designed, synthesized, and tested on PC12 cells to examine
neuroprotective effects against H
2
O2 and 6-OHDA induced damage. Among them, 6d
was found to be able to alleviate the accumulation of ROS, inhibit the production of
NO and downregulate the level of IL-6, which indicated its potential antioxidant and
anti-inflammatory activity. Further studies proved that 6d could activate Nrf2 signaling
pathway, induce the translocation of Nrf2 from cell cytosol to nucleus and upregulate
the related phase II antioxidant enzymes including NQO1, HO-1, GCLC, GCLM and
TrxR1. These results confirmed that 6d exerted antioxidant and anti-inflammatory
activities by activating Nrf2 signaling pathway. Moreover, the parallel artificial
membrane permeability assay indicated that 6d can cross the blood-brain barrier. In
general, 6d is promising for further development as a therapeutic drug against
oxidative stress and inflammation related neurodegenerative disorders.
Keywords: Piperlongumine analogues; Oxidative stress; Neuroprotective effect;
Keap1-Nrf2-ARE pathway

4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
6
6
6
6
6
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
1. Introduction
Neurodegenerative diseases (ND), including Alzheimer's disease (AD),
Parkinson's disease (PD), and multiple Sexual sclerosis (MS), are general term for
diseases caused by chronic progressive degeneration of central nervous tissue [1 - 3].
These diseases have a common pathological feature, such as changes in the tissue
redox balance accompanied by the activation of microglial cells [4]. The pathological
processes of neurodegenerative diseases are associated with generation of reactive
oxygen species (ROS), which cause oxidative stress [5]. Oxidative stress is supposed
to play a key role in the development and progression of various diseases [5, 6]. For
AD, oxidative stress can aggravate the deposition of Aβ and phosphorylation of tau
protein, and promote Aβ and tau-mediated neurotoxicity. Furthermore, the presence of
Aβ and tau is also resulted in the increase of ROS [6, 7]. These consequences, however,
will further aggravate the oxidative stress response, which forms a vicious circle.
Moreover, oxidative stress leads to cellular dysfunction and demise, especially playing
a major role in the degeneration of dopaminergic neurons in the pathogenesis of PD [8].
Consequently, preventing ROS production and reducing oxidative stress may be a
crucial therapeutic target for ND treatment.
The Keap1-Nrf2-ARE system plays a key role in antioxidant and
anti-inflammatory mechanisms, which is one of the main cellular defense mechanisms
against oxidative stress. As a dominant regulator in cell, Nrf2 fights against oxidative
stress by activating anti-oxidative stress proteins, such as NQO1, GCLM, TrxR1, HO-1,
and GCLC, and phase II detoxification enzymes [8 - 11]. Recently, related studies have

6
6
6
6
7
7
7
7
7
7
7
7
7
7
8
8
8
8
8
8
8
8
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
shown that the Keap1-Nrf2-ARE pathway could be rapidly activated during the
development of neurodegenerative diseases [12], which is related to the production of
ROS. In addition, Nrf2 also regulates the expression of HO-1 to achieve
anti-inflammatory effect [13, 14]. In recent years, crystal structure of Keap1-
sulfhydryl composite has been published, and covalent modification of cysteine-rich
Keap1 protein by electrophilic molecules becomes an important strategy for the
activation of Nrf2 [14]. The modulators are classified into Michael acceptors, polyenes,
isothiocyanates, oxidizable, organosulfur compounds, trivalent arsenicals, diphenols,
heavy metal species and selenium-containing compounds [15]. Dimethyl fumarate
(Tecfidera) is the first Nrf2-inducer approved by FDA for the treatment of
remitting-relapsing multiple sclerosis [16, 17]. Thus, chemicals with an electrophilic
scaffold, especially α, β-unsaturated ketone structure, might be potent activators of
Nrf2 [18, 19]. Many natural products, such as resveratrol [20], butein [21], caffeic acid
[22] and curcumin [23], exhibit the potential to activate Nrf2. These scaffolds could
covalently bind to cysteine residues of Keap1, resulting in dissociation of Nrf2 from
Keap1, and translocation of Nrf2 from cytosol to the nucleus [24].
Piperlongumine (PL) is an alkaloid isolated from long pepper [25]. According to
previously published studies, PL could increase the ROS level in cancer cells [26].
Researchers have found that PL has protective effect against AD, such as lowering
cholinesterase levels, reducing neuroinflammation and inhibiting amyloid plaque
formation [27, 28]. Based on the previous research, several general structure-activity
relationships (SAR) have been identified. It reveals that the presence of 7, 8-olefin is

8
8
9
9
9
9
9
9
9
9
9
9
8
9
0
1
2
3
4
5
6
7
8
9
required. In addition, studies have shown that 2-chloro substitution may increase the
electrophilicity of 2, 3-olefins, thereby enhancing activity [29, 30]. But few
modifications on the position 2 and the lactam ring of PL were reported. In this study,
the lactam ring, the position 2 of PL, and the aromatic ring were changed to study the
SAR of PL. We reported herein the design, synthesis and biological evaluation of a
series of PL analogues.
2. Results and discussion
2.1 Chemistry
PL analogues were synthesized by coupling acyl chlorides and commercially
available lactams (Scheme 1) according to the published procedures with minor
1
13
modifications [10]. All target compounds were characterized by H NMR, C NMR,
and HRMS.
a
1
00
Scheme 1. Synthesis of PL analogs
1
01

O
O
a
b
R
Cl
R
N
O
NH
1
O
-4
O
O
O
O
c
O
R
N
R
OH
R
N
O
HN
6-28
O
O
NH
O
O
N
NH
c
N
N
R3
F
F
F
6
a-6l
O
O
O
O
O
b, c
c
O
OH
N
N
NH
N
N
6
m
5
F
F
1
1
02
03
a
Reagents and conditions: (a) NaH, THF, 0 °C, 30 min; (b) Trimethylacetyl chloride,
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
TEA, DCM, 0 °C, 30 min; (c) NaH, DCM, 0 °C, 30 min.
2.2 Protection of PC12 cells against H O - and 6-OHDA-induced cell damage by PL
2
2
analogues
The cytotoxicity of all compounds toward the PC12 cells were determined by the
MTT assay. As shown in Figure S2, PL, PL-1, 6a, 6b, 6i and 6l displayed significant
toxicity to the cells at high concentration (50 μM), and there were no apparent toxicity
of the other tested compounds toward the PC12 cells at 50 μM. Two classic cellular
models, 6-OHDA and H O induced PC12 cell damage models, were established to
2
2
evaluate the neuroprotective effect of the target compounds [30]. The specific
concentration of H O (200 μM) and 6-OHDA (150 μM) were chosen for the further
2
2
study, where cell viability was approximately 50%. (Figure S1).
The biological evaluation suggested that the 2, 3-position Michael acceptor
moiety in PL could increase the cytotoxicity. When the carbon atom of the lactam ring
of PL was replaced by the nitrogen atom, such as PL-3 which showed better
neuroprotection activity than that of PL-2. It was concluded that removing the α,

1
1
1
1
1
1
19
20
21
22
23
24
β-unsaturated in δ-valerolactam ring of PL could improve the neuroprotection. As
expected, PL-4 displayed lower cytotoxicity (Figure S2A) and better protection
(Figure 1) in the initial screening. Few studies report the antioxidant activity of PL by
replacing its lactam ring with 2-imidazolidone structures. We paid our attention on the
modifications of the PL lactam with substituted imidazolidones, and PL-4 was selected
as a leading compound for the follow-up studies.
1
1
25
26
Figure 1 Initial screening of PL analogues against 6-OHDA- or H O -induced PC12
2
2
1
1
1
27
28
29
cell damage. Data are the mean ± SD of three independent experiments. (**p) ≤ 0.01
^
compared with the control group; ( p) ≤ 0.05 compared with H O -treated or
2
2
6-OHDA-treated group.
1
30

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
Figure 2. The docking models of icompond 1 (A) and 6 (B) with the BTB domain of
Keap1 (PDB: 4CXT).
Glide program in Schrodinger was utilized to predict the potential binding model
between 1 and Keap1 BTB domain (Figure 2A). The covalent docking results showed
that 1 could form a covalently bond with residue CYS151 of Keap1, the benzene ring
from 1 formed π-π stacking with residue HIE154, and the carbonyl group from the α,
β-unsaturated ketone and the lactam carbonyl group formed hydrogen bonds with
residues GLY148, LYS131 and GLY145. The docking results displayed that the residue
LYS150 as a hydrogen bond donor around the interface pocket is very close to the
ligand. With this in mind, compound 6 having a hydrogen bond acceptor was designed,
and the docking result showed that the added pivaloyl fragment did form hydrogen
bond with residue LYS150 (Figure 2B). In this point, compounds 1, 2, 3 and 4 without
pivaloyl and compounds 6, 15, 20 and 21 with pivaloyl were designed and synthesized.
The experimental results showed that the added pivaloyl play crucial roles in their
neuroprotective activities (6, 15, 20 and 21 vs. 1, 4, 3 and 2, Tables 1 and 2). To
further verify the influence of added pivaloyl to the expression of Nrf2. PC12 cells
were treated with compounds 1 and 6 for 24 h, and the expression of Nrf2 was detected
by western blot. As showed in Figure 3, the expression of Nrf2 was significantly
upregulated, and the upregulation of Nrf2 by 6 was more obvious than that of
compound 1. The results suggested that the pivaloyl group is vital to activate Nrf2.
To validate the SAR, a variety of additional aryl analogues (1 - 28) with different
benzyl groups varying from para and ortho to meta substituents were synthesized

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
(Tables 1 and 2). All the target compounds showed negligible cytotoxicity (Figure
S2A). Different substituted groups were introduced to phenyl. The 4'-F substituted
compound 6 exhibited potent neuroprotection with cell viability (73.58%) better than
that of 14 and 13 (4'-F > 2'-F > 3'-F). The 4'-Cl substituted 8 exhibited potent
neuroprotection better than that of 17 and 18 (4'-Cl > 2'-Cl > 3'-Cl). It is well known
that the majority nature of heterocycles is served as hydrogen bond donors and
receptors. In this study, heterocycles were utilized to replace the aromatic group in the
scaffold of PL. Several compounds substituted by furan (20), thiophene (21), pyridine
(22), naphthalene (24), 1, 3-benzodioxole (25) were synthesized. Unfortunately, the
neuroprotection was not improved compared to the compounds containing phenyl.
Among the tested compounds, the compounds with electron-withdrawing groups
showed better neuroprotective activity than that with the electron-donating groups. The
monosubstituted compounds 6 (73.58%), 13 (70.97%) and 14 (72.43%) show better
neuroprotective activity than the polysubstituted compounds 16 (63.55%) and 26
(59.15%). It is remarkable that 6 showed best protection against 6-OHDA and H O
2
2
induced PC12 cell damage (Tables 1 and 2).
Table 1 The neuroprotection of PL analogues substituted with different aryl group
Cell viability (%)
Compd.
Structure
H O
6-OHDA
2
2
Model
46.00 ± 4.12
68.57 ± 2.88
45.23 ± 4.37
O
N
O
1
69.13 ± 3.4
NH
F

2
3
4
5
65.15 ± 3.86
64.08 ± 3.55
67.66 ± 4.15
53.56 ± 1.29
64.2 ± 1.74
60.00 ± 2.72
65.32 ± 4.72
56.32 ± 2
a
1
1
1
1
70
71
72
73
Cell viability (%) of PC12 cells were detected by the MTT assay after 24 h of
incubation with compounds at the concentration of 20 μM with H O (200 μM) or
2
2
6-OHDA (150 μM). Data are the mean ± SD of three independent experiments.
Table 2 The neuroprotection of PL analogues substituted with different aryl group
1
74
a
Cell viability (%)
Cell viability (%)
Compd.
R
Compd.
R
H O
6-OHDA
H O
2 2
6-OHDA
2
2
Model
46.00 ± 4.12
73.58 ± 3.55
70.15 ± 3.86
72.05 ± 2.15
72.22 ± 2.11
45.23 ± 4.37
72.23 ± 2.72
69.21 ± 0.74
71.12 ± 3.23
70.79 ± 2.82
17
18
19
20
21
22
23
70.01 ± 4.55 69.01 ± 2.39
69.21 ± 1.87 68.32 ± 4.06
60.36 ± 2.42 56.73 ± 1.30
69.66 ± 4.15 68.31 ± 4.72
69.23 ± 3.86 68.21 ± 4.55
70.91 ± 5.32 69.58 ± 2.51
6
7
8
9
1
0
1
72.1 8 ± 0.76 72.32 ± 2.25
N
1
71.14 ± 1.81
70.32 ± 3.09
71.41 ± 4.5
71.12 ± 4.02

1
1
1
2
3
4
61.52 ± 3.22
70.97 ± 2.54
72.43 ± 2.54
60.08 ± 5.27
69.32 ± 4.19
72.33 ± 1.20
24
25
26
72.33 ± 3.54 71.79 ± 5.49
63.33 ± 1.05 62.31 ± 2.94
59.15 ± 4.34 56.38 ± 5.36
1
1
5
6
72.5 ± 2.88
71.12 ± 3.40
61.77 ± 2.54
27
28
73.22 ± 2.73 72.13 ± 6.40
72.65 ± 4.48 71.21 ± 2.62
63.55 ± 1.64
a
1
1
1
75
76
77
Cell viability (%) of PC12 cells were detected by the MTT assay after 24 h of
incubation with compounds at the concentration of 20 μM with H O (200 μM) or
2
2
6-OHDA (150 μM). Data are the mean ± SD of three independent experiments.
1
1
78
79
Figure 3 The expression of Nrf2 in PC12 cells after treatment with compounds 1 and 6.
(**p) ≤ 0.05, Data are presented by mean ± SD (n = 3).
1
1
1
1
1
1
80
81
82
83
84
85
As exhibited in Table 3, compounds 6a, 6b, 6i, and 6l displayed obvious
cytotoxicity at low concentrations (Figure S2B). In addition, their neuroprotective
effect were eliminated or weakened, which may be related to their toxicity. When the
position 2 of N-terminus was connected to the aromatic structure, such as 6k, 6l and 6j,
the neuroprotective effects of them were reduced. When carbonyl was connected to

1
1
1
1
1
1
1
1
1
1
1
1
86
87
88
89
90
91
92
93
94
95
96
97
piperidine (6e), N, N-dimethyl (6c) and N-methyl, N-ethyl (6d), the neuroprotective
effects were significantly enhanced. It is remarkable that 6d showed excellent
neuroprotective effect (75.01%). To verify whether α, β-unsaturated ketone is an
essential group for neuroprotection, compounds PL-6, 5 and 6m were synthesized and
evaluated their neuroprotective effects. As expected, compounds PL-4 (68.80%), 1
(68.57%) and 6d (75.01%) exhibited potency neuroprotective effect with cell viability
better than that of PL-6 (55.78%), 5 (53.56%) and 6m (59.50%), respectively. In
general, we synthesized a series of PL analogues following the strategy shown in
Figure 4 and evaluated their neuroprotective activities. Among them, 6d showed lower
cytotoxicity (Figure S2B), best neuroprotective effect (Table 3), and was chosen for
the follow-up studies.
Table 3 The neuroprotection of PL analogues substituted with different groups
a
Cell viability (%)
Compd.
Model
Structure
(
H O -induced)
(6-OHDA-induced)
2
2
46.96 ± 4.12
38.05 ± 3.20
43.96 ± 5.43
73.82 ± 1.39
45.65 ± 4.37
6
6
a
30.23 ± 4.35
43.76 ± 5.01
71.32 ± 2.87
b
6
c
6
d
75.01 ± 2.02
72.8 ± 1.19
75.44 ± 2.29
72.14 ± 2.60
6
e

6
f
70.42 ± 4.51
71.58 ± 4.55
70.15 ± 3.86
43.25 ± 4.15
72.27 ± 2.21
70.23 ± 3.72
69.26 ± 1.74
37.12 ± 3.23
6
6
g
h
O
O
O
6
i
N
N
Cl
F
6
j
62.25 ± 2.11
63.28 ± 0.76
60.79 ± 2.82
60.32 ± 5.25
6
k
6
l
55.14 ± 4.81
59.5 ± 4.20
52.32 ± 2.09
62.45± 1.20
6
m
a
1
1
2
98
99
00
Cell viability (%) of PC12 cells were detected by the MTT assay after 24 h of
incubation with compounds at the concentration of 20 μM with H O (200 μM) or
2
2
6-OHDA (150 μM). Data are the mean ± SD of three independent experiments.
2
01
02
2
Figure 4 The design of the modifications of PL
2
03
2.3. Antioxidant effect of 6d
2
2
04
05
2.3.1. Defense of PC12 Cells against H O or 6-OHDA induced cell damage in a dose
2
2
dependent manner by 6d
2
2
06
07
Initially, we investigated the cytotoxicity of all compounds on the PC12 cells (a
rat pheochromocytoma cell line), BV2 cells (mouse microglia), and L02 cells (an

2
2
2
2
2
2
2
2
2
2
2
2
08
09
10
11
12
13
14
15
16
17
18
19
immortal hepatic). As shown in Figure S3, no apparent toxicity was observed at 50
μM. In Figures 5A and 5B, compared with control group, cells pretreated with 6d (5,
10 and 20 μM) for 24 h followed by treatment with H O - and 6-OHDA exhibited
2
2
stronger viability in dose-dependent manners. Lactate dehydrogenase (LDH) is an
important indicator of membrane integrity [31]. To confirm the cytoprotection of 6d,
the content of LDH leakage after H O - or 6-OHDA-treated was determined. As shown
2
2
in Figures 5C and 5D, the content of LDH was up-regulated by H O or 6-OHDA, and
2
2
the release amount of LDH increased 2.8-fold and 1.8-fold that of the control
respectively, which was the same as the expected result. After pretreatment with 6d,
the leakage of LDH was significantly reduced. The above results indicated that the
non-toxic concentration of 6d could prevent the neurotoxicity induced by H O and
2
2
6-OHDA in PC12 cells.

2
2
20
21
Figure 5. Protection of 6d against H O - (A) and 6-OHDA-induced (B) PC12 cell
2
2
2
2
2
2
2
2
2
2
2
2
22
23
24
25
26
27
28
29
30
31
damage, determined by MTT assay. Protection of 6d against H O -induced (C) and
2 2
6-OHDA-induced (D) PC12 cell damage was measured by the LDH release assay.
Data are the mean ± SD of three independent experiments. (**p) ≤ 0.01 and (***p) ≤
^
^^
^^^
0.001 compared with the control group; ( p) ≤ 0.05, ( p) ≤ 0.01 and ( p) ≤ 0.001
compared with H O -treated or 6-OHDA-treated group.
2
2
2.3.2 Prevention of ROS accumulation in PC12 cells
Oxidative stress response to ROS is a key initiator of oxidative damage induced
by H O and 6-OHDA. In order to study the protection of 6d to the oxidative damage
2
2
in PC12 cells, the level of ROS was quantified by dichlorofluorescein diacetate
(DCFH-DA) using flow cytometry [32, 33]. When PC12 cells were exposed to H O or
2
2

2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
6-OHDA, the intracellular ROS fluorescence intensity increased significantly.
However, when pretreated with an increasing concentration of 6d (5, 10 and 20 μM),
the ROS levels decreased in a dose-dependent manner (Figures 6A and 6B). The result
suggested that 6d effectively reduced H O - and 6-OHDA-induced ROS accumulation.
2
2
2.3.3 Alleviation of H O - and 6-OHDA-induced intracellular mitochondrial
2
2
dysfunction by 6d
Mitochondria are important mediators of cell metabolism and main producers of
ROS. The reduction of mitochondrial membrane potential (MMP) is a sign of early
apoptosis. The reduction of cell membrane potential can be detected by the red
fluorescence of JC-1 converted to green fluorescence [34]. Figures 6C and 6D showed
that H O and 6-OHDA reduced the MMP of PC12 cells, indicating that H O - and
2
2
2
2
6-OHDA caused the mitochondrial dysfunction. In contrast, pretreatment of PC12 cells
with 6d before exposure to H O or 6-OHDA could significantly increase the MMP.
2
2
These results showed that 6d had potent protection against H O - or 6-OHDA-induced
2
2
mitochondrial dysfunction in PC12 cells.

2
2
47
48
Figure 6. The effects of 6d on H O -induced (A) and 6-OHDA-induced (B)
2
2
2
2
49
50
intracellular ROS production in PC12 cells. The cells were stained with DCFH-DA and
immediately determined by flow cytometry. The effects of 6d on H O -induced (C)
2
2

2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
and 6-OHDA-induced (D) MMP reduction in PC12 cells. MMP were detected by flow
cytometry after JC-1 staining. Data are presented by mean ± SD (n = 3). (***p) ≤
^
^^
^^^
0.001 compared with the control group; ( p) ≤ 0.05, ( p) ≤ 0.01 and ( p) ≤ 0.001
compared with H O -treated or 6-OHDA-treated group.
2
2
2.3.4 Alleviation of H O - and 6-OHDA-induced PC12 cell apoptosis
2
2
To investigate the anti-apoptosis ability of 6d against H O - and
2
2
6-OHDA-induced cell damage, quantified Annexin V-FITC and propidium iodide (PI)
were used to evaluate apoptotic cells by flow cytometry [35]. As shown in Figure 7A,
compared with the control group, the apoptosis rate of PC12 cells increased from
5.71% (control group) to 25.38% (H O treated group) after treatment with H O for 24
2
2
2
2
h. In contrast, the rates of apoptotic cells were significantly reduced when H O -treated
2
2
PC12 cells were co-incubated with 6d (5, 10 and 20 μM), the cell apoptosis rate was
reduced to 16.45%, 7.62%, and 6.52%, respectively. Similarly, in Figure 7B, the
apoptosis rate of PC12 cells increased from 3.23% (control group) to 27.5% (6-OHDA
treated group) after treatment with 6-OHDA for 24 h, and the cell apoptosis rate was
reduced to 19.17%, 15.07%, and 10.83% after the cells were pretreated with 6d (5, 10
and 20 μM).
Hoechst 33342 is a blue fluorescent dye that can penetrate cell membranes and is
commonly used to detect apoptosis. After staining, images were captured using a
high-content imaging system. As shown in Figures 7C and 7D, both H O and
2
2
6-OHDA could cause PC12 cell apoptosis. The apoptotic nuclei were characterized by
highly fluorescent aggregates, while no obvious apoptotic nuclei were observed in the

2
2
2
73
74
75
control group. Pretreatment of cells with 6d significantly reduced the number of
apoptotic nuclei, which indicated that 6d showed significant neuroprotective effect
against H O - and 6-OHDA-induced apoptosis.
2
2
2
2
76
77
Figure 7. Prevention of PC12 cells from H O -induced (A) and 6-OHDA-induced (B)
2
2

2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
apoptosis by 6d. Apoptotic cells were detected by flow cytometry after AnnexinV and
PI double staining. (C, D) Images showed the apoptotic nuclei by Hoechst 33342
staining. The top panel is phase contrast pictures, and the bottom panel is fluorescent
pictures. Scale bars: 100 μm. Data are presented by mean ± SD (n = 3). (***p) ≤ 0.001
^
^^
compared with the control group; ( p) ≤ 0.05 and ( p) ≤ 0.01 compared with H O - or
2
2
6-OHDA-treated group.
2.3.5 The reduction of H O -induced and 6-OHDA-induced Ca overload on PC12
2
+
2
2
cells by 6d
Increased ROS will react with cellular proteins, nucleic acids, et al., to cause
2
+
2+
cellular barriers to affect cellular Ca influx. Ca has been shown to mediate
cytotoxicity of oxidative stress. Fluo-3 AM fluorescent probe can be used to detect the
2
+
2+
concentration of Ca in the cell, and the relative intensity of Ca is reflected by the
fluorescence intensity [36, 37]. After being stimulated with H O or 6-OHDA for 24 h,
2
2
2
+
the intracellular Ca was almost 3.3 times and 1.8 times, respectively (Figures 8A, B).
2
+
Compared with the control group, the intracellular free Ca fluorescence value
increased from 5537.87 (control group) to 17791.36 (H O treated group) after
2
2
treatment with H O for 24 h. In contrast, pretreatment of PC12 cells with 6d (5, 10
2
2
2
+
and 20 μM) before exposure to H O sharply dropped the intracellular free Ca
2
2
fluorescence value to 8721.73, 7793.41, and 6730.32, respectively. Similarly, the
2
+
intracellular free Ca fluorescence value increased from 3998.83 (control group) to
6708.18 (6-OHDA treated group) after treatment with 6-OHDA for 24 h. Pretreatment
of PC12 cells with 6d (5, 10 and 20 μM) before exposure to 6-OHDA significantly

2
+
3
3
3
3
3
00
01
02
03
04
dropped the intracellular free Ca fluorescence value to 4910.91, 4204.33, and
4161.82, respectively. The experimental results showed that 6d in PC12 cells could
2
+
prevent 6-OHDA- or H O -induced Ca overload. It was in agreement with the results
2
2
of ROS, MMP and apoptosis above, which suggested that 6d could be used as a
potential neuroprotective agent for the treatment of ND.
3
3
05
06
2
+
Figure 8. Effects of 6d on H O -induced (A) and 6-OHDA-induced (B) Ca overload
2
2
3
3
3
07
08
09
in PC12 cells. Data are presented by mean ± SD (n = 3). (*p) ≤ 0.05 and (***p) ≤
^
^^
0.001 compared with the control group; ( p) ≤ 0.05 and ( p) ≤ 0.01 compared with
H O -treated or 6-OHDA-treated group.
2
2
3
10
11
2.3.6 The Activation of the Keap1-Nrf2-ARE Pathway and the Induction of the
3
Downstream Antioxidant Proteins Expression in PC12 Cells by 6d
3
12
As confirmed above, 6d showed potent protection on PC12 cells against

313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
6-OHDA- or H O -induced cell damage, alleviated ROS accumulation, mitochondrial
2 2
2
+
dysfunction, Ca influx, and cell apoptosis. We hypothesized that the antioxidant
activitie of 6d is related to the activation of Nrf2. To verify whether 6d could activate
the Nrf2-Keap1-ARE pathway, the expression of Nrf2 and its downstream antioxidant
proteins were detected by Western blot. The protein level for Nrf2 was maximal after a
3 h treatment of with the highest concentration 6d (20 μM), the protein levels for HO-1,
GCLM reached maximum at 12 h, and the protein levels for TrxR1, NQO1 and GCLC
reached maximum at 24 h (Figure 9A). Western blot analysis demonstrated that
treatment with 6d for 24 h resulted in a remarkable increase of the Nrf2-regulated
proteins (NQO1, HO-1, TrxR1, GCLM, and GCLC) in a dose-dependent manner
(Figure 9B) [38].
DPPH (diphenyl-1-picrylhydrazyl) and ABTS radical scavenging method was
used to assess the antioxidant activities of these compounds in vitro，where Trolox was
used as a positive [39, 40]. 6d is incapable of intercepting either DPPH or ABTS free
radicals (Figure S4), which indicated that 6d exerted neuroprotective effect as an
activator rather than direct radical scavenger.
The translocation of Nrf2 from cytosol to nucleus is prerequisite for the
expression of Nrf2-dependent proteins. To confirm that whether 6d could transfer Nrf2
from the cytoplasm to the nucleus. First, we checked the nuclear and cytoplasmic Nrf2
expression. Nuclear Nrf2 accumulation increased maximally at 6 h. Cytosol Nrf2
accumulation increased maximally at 3 h and declined after 6 h. These results indicated
that 6d promoted the transfer of Nrf2 to the nucleus, which facilitates the binding of

3
3
3
3
3
3
3
3
3
35
36
37
38
39
40
41
42
43
Nrf2 to ARE for the transcription process. (Figure 9C) [41]. Besides, to confirm
whether 6d exerted the anti-inflammatory or antioxidant activity by activating Nrf2
signaling pathway, antioxidant assay was performed with the existence of Nrf2
inhibitor brusatol or the HO-1 inhibitor zinc protoporphyrin IX (ZnPP) existed [30]. As
shown in Figures 9D and 9E, 6d alleviated the H O or 6-OHDA induced cell death at
2
2
20 μM, while this neuroprotective effect was almost abolished in the group pretreated
with brusatol or ZNPP. This result demonstrated that Nrf2 was essential for the
neuroprotective effect of 6d in PC12 cells.

3
3
44
45
Figure 9. 6d could increase the expression of nuclear and cytosolic Nrf2, TrxR1,
NQO-1, HO-1, GCLM and GCLC in a time (A) and dose (B) dependent manner.
Promotion of Nrf2 nuclear accumulation by 6d (C). Brusatol (D), ZnPP (E) affected
the protection of 6d. PC12 cells were incubated for 30 min in the presence of 6d (20
μM) together with brusatol (10 nM) or ZNPP (10 nM) prior to stimulation with H O
3
3
3
3
3
3
46
47
48
49
50
51
2
2
(150 μM) for 24 h, determined by MTT assay. Data are presented by mean ± SD (n =
3).

3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
2.4 Anti-inflammatory effect of 6d
The excessive activation of microglia plays an important part in neuronal damage
and death caused by AD and PD neuroinflammation [42]. NO release is a vital feature
of microglia activation. Excessive NO production can cause inflammation [43].
Besides, many researchers believe that HO-1 regulates the inflammatory process and is
related to the Nrf2 / ARE pathway [44 - 46]. Collectively, activating Nrf2 is an
excellent method to reduce the inflammatory process [46]. The anti-neuroinflammation
property was measured using the Griess method to examine the effects of 6d on the
production of NO. The results showed that after the 18 h incubation with LPS
(1μg/mL), the level of NO increased sharply compared to the control group. However,
the levels of NO dose-dependently decreased when BV2 microglia cells were
pretreated with 6d before exposed to LPS (Figure 10A). It is well known that the
release of inflammatory factors contributes to the development of inflammation. IL-6
is an important indicator to evaluate the activation degree of macrophages and the
progress of inflammatory response [47, 48]. 6d significantly attenuated LPS-induced
production of IL-6 in BV2 microglia cells (Figure 10B). The potential
anti-inflammatory effect may be related to the activation of the Nrf2 signaling pathway.
We determined the expression of Nrf2 and its downstream anti-oxidant proteins using
Western blot. As shown in Figure 10C, treatment of BV2 microglia cells with 6d for
24h could significantly upregulate the expression of these proteins with dose
dependent manner, which was similar with that in PC12 cells. Then, to investigate
whether activation of Nrf2 is responsible for the anti-inflammatory effect of 6d, Nrf2

3
3
3
3
3
3
3
3
3
3
74
75
76
77
78
79
80
81
82
83
was knockdown by siRNA. Then the control siRNA- and Nrf2 siRNA-transfected BV2
microglia cells were pre-treated with 6d (20 µM) and stimulated with LPS for another
18 h. The level of Nrf2 decreased sharply in Nrf2 siRNA-transfected BV2 microglia
cells (Figure 10D). Furthermore, 6d (20 µM) could significantly reduce LPS-induced
NO production in control siRNA-transfected BV2 microglia cells (Figure 10E) [49 -
51]. However, the suppressive effects of 6d (µM) was obviously suppressed in Nrf2
siRNA-transfected BV2 microglia cells. Similar results were observed in experiments
to determine the effects of Nrf2 knockout on the inhibition of IL-6 production (Figure
10F). These results demonstrated that 6d exerted anti-inflammatory effect with a
Nrf2-dependent manner in BV2 microglia cells.

3
3
84
85
Figure 10. Contribution of Nrf2 to the anti-inflammatory effect of 6d. (A) 6d reduced
the production of LPS-stimulated inflammatory mediators NO in BV2 microglia cells.
(B) 6d reduced the production of LPS-induced IL-6 in BV2 microglia cells. The levels
3
3
86
87

3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
88
89
90
91
92
93
94
95
96
97
98
99
00
01
02
03
04
05
06
07
08
09
of IL-6 were measured by ELISA kites. (C) 6d dose dependently induced expression of
Nrf2 and its downstream antioxidant proteins in BV2 microglia cells. (D) Nrf2
expression in Control siRNA- and Nrf2 siRNA-transfected BV2 microglia cells. (***p)
^^
^^^
≤ 0.001 compared with the control group; ( p) ≤ 0.01 and ( p) ≤ 0.001 compared
with LPS-treated group. Transfection of BV2 microglia cells with Nrf2 siRNA
reversed suppressive effects of 6d (20 µM) on NO (E), IL-6 (F) production following
LPS stimulation. **p≤0.01 and *p≤0.05 in comparison with control siRNA-transfected
cells. Data are presented by mean ± SD (n = 3).
2.6 Molecular docking study between 6d and Keap1
Keap1 has strict regulation on the activation of Nrf2. The oxidative or
electrophilic agents covalently bind to the cysteine-rich Keap1 protein, thereby
changing conformational of Keap1 to prevent the ubiquitination of Nrf2, thus activate
Keap1-Nrf2-ARE pathway. According to previous reports, other Michael receptors
may activate the Nrf2 pathway by reacting with CYS151 or other reactive cysteine
residues in Keap1 [4, 52]. Considering that α, β-unsaturated ketone of 6d is a Michael
receptor moiety, which is possible to react with cysteine residues of keap1. So the
covalently docking study was performed used Glide based on the X-ray crystal
structure of BTB domain of KEAP1 in complex with CDDO (PDB: 4CXT) (Figure
11). The result of covalent docking showed that 6d could covalently bonding with
CYS151 of Keap1, the benzene ring formed π-π stacking with residue HIE154, and the
carbonyl group of α, β-unsaturated ketone and carbonyl lactam formed hydrogen bonds
with residues GLY148, LYS131 and ARG135. The N-ethyl, N-methylformamide