Twin drug design, synthesis and evaluation of diosgenin derivatives as multitargeted agents for the treatment of vascular dementia

Bioorg. Med. Chem. 37 (2021) 116109

Contents lists available at ScienceDirect

Bioorganic & Medicinal Chemistry

journal homepage: www.elsevier.com/locate/bmc

Twin drug design, synthesis and evaluation of diosgenin derivatives as

multitargeted agents for the treatment of vascular dementia

,

Gui-Xiang Yang^{a 1}, Jia-Min Sun^{a 1}, Lu-Lu Zheng^{a 1}, Li Zhang^a, Jie Li^a, Hai-Xian Gan^a,

Yan Huang^a, Jin Huang^a, Xing-Xing Diao^b, Yun Tang^{a *}, Rui Wang^{a *}, Lei Ma^a

,

,*

^aShanghai Key Laboratory of New Drug Design, School of Pharmacy, East China University of Science and Technology, 130 Meilong Road, Shanghai 200237, China

^bShanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai 201203, China

A R T I C L E I N F O

A B S T R A C T

Keywords:

A novel series of multitargeted molecules were designed and synthesized by combining the pharmacological role

of cholinesterase inhibitor and antioxidant of steroid as potential ligands for the treatment of Vascular Dementia

(VD). The oxygen-glucose deprivation (OGD) model was used to evaluate these molecules, among which the

most potent compound ML5 showed the highest activity. Firstly, ML5 showed appropriate inhibition of cho-

linesterases (ChEs) at orally 15 mg/kg in vivo. The further test revealed that ML5 promoted the nuclear trans-

location of Nrf2. Furthermore, ML5 has signiﬁcant neuroprotective effect in vivo model of bilateral common

carotid artery occlusion (BCCAO), signiﬁcantly increasing the expression of Nrf2 protein in the cerebral cortex. In

the molecular docking research, we predicted the ML5 combined with hAChE and Keap1. Finally, compound

ML5 displayed normal oral absorption and it was nontoxic at 500 mg/kg, po, dose. We can draw the conclusion

that ML5 could be considered as a new potential compound for VD treatment.

Vascular dementia

Diosgenin

Antioxidant

Anti-cholinesterase

Multi-target-directed ligands

1. Introduction

inhibitors used for the treatment of dementia, including donepezil,

rivastigmine, galantamine.⁷Especially for AChE and butyryl cholin-

Vascular Dementia (VD) is a neurodegenerative disease, which is

related to various factors, including age, low levels of education, female

sex, and vascular risk factors.¹Recently, VD has become the main type

of dementia except for Alzheimer’s disease (AD), occupying roughly

one-ﬁfth of dementia.²Based on current research, VD has serious im-

pacts on people’s mental health and quality of life, the mortality and

morbidity of VD are increasing as the population aging. There is no

speciﬁc drug approved for anti-VD by the Food and Drug Administration

(FDA) yet, although it can be delayed through early prevention.³Thus, it

is necessary to develop new anti-VD drugs.

esterase (BuChE) inhibitory activity, rivastigmine has been approved

by the FDA as a transdermal patch for the treatment of mild, moderate,

and severe AD as well as mild-to-moderate Parkinson’s disease de-

mentia (PDD), and one research showed that rivastigmine had similar

beneﬁts to improving cognition and behavioral symptoms in patients

with VD.⁸

VD is caused by a range of cardiovascular or cerebrovascular con-

ditions, it can cause damages to learn and memory functions of the brain

through pathways of inﬂammation and oxidative stress in blood ves-

sels.^9,10Nuclear factors erythroid 2-related factor 2 (Nrf2), a key tran-

scription factor for the expression of the oxidative stress response is

regulated by Kelch-like ECH-associated protein 1 (Keap1). The interac-

tion between Nrf2 and the antioxidant reaction element (ARE) regulates

the expression of antioxidant protein and phase II detoxifying enzy-

me.¹¹Under normal circumstances, Nrf2 binds to Keap1 in the cyto-

plasm and the ubiquitination rapidly degrades.¹²When the level of

oxidation beyond antioxidant capacity, the conformational change of

Keap1 leads to cysteine residues being modiﬁed then the inherently

Acetylcholine (ACh), the ﬁrst neurotransmitter identiﬁed in the

central nervous system plays an important role in cognitive activities

include learning, memory, and spontaneous behavior.^4,5Damaged

Cholinergic neurons in the brain is a characteristic of AD, VD, and

other forms of dementia such as Lewy bodies dementia (LBD) and

frontotemporal dementia (FTD).⁶AChE inhibitors can accumulate ACh

at the synapse, prolong and increase the effects of ACh by inhibiting

acetylcholinesterase (AChE). Currently, there are three AChE

* Corresponding authors.

E-mail addresses: ytang234@ecust.edu.cn (Y. Tang), ruiwang@ecust.edu.cn (R. Wang), malei@ecust.edu.cn (L. Ma).

1

These authors contributed equally to this work.

https://doi.org/10.1016/j.bmc.2021.116109

Received 15 January 2021; Received in revised form 1 March 2021; Accepted 6 March 2021

Available online 19 March 2021

G.-X. Yang et al.

Bioorganic & Medicinal Chemistry 37 (2021) 116109

combined Nrf2 was released to the nucleus.¹³Then, Nrf2 activated

by binding to ARE in the nucleus, together with subsequent enhance-

ment in the expression of its downstream antioxidant genes, including

NADP(H) quinone oxidoreductase (NQO1) and heme oxygenase (HO-

1).^11,14,15Therefore, increasing Nrf2 stability through inhibiting

the interaction between Nrf2 and Keap1 can increase antioxidants

pathways.

2.2. Neuroprotection

Oxygen glucose deprivation (OGD) is one of the oxidative stress

models, which was used to assess the antioxidant effects of compounds

in vitro. Here, in order to evaluate the neuroprotection of compounds

ML1-ML23, the antioxidant activities of target compounds were tested

on SH-SY5Y cells against oxidative injuries stimulated by OGD. The level

of cell viability was quantiﬁed by MTT assay, which was used to char-

acterize the protective effect against OGD.

Currently, the development of multitarget-directed drugs is considered

the most efﬁcient approach for diseases with complex etiology, particu-

larly the neurodegenerative disorders.^16,17In our previous work, we

designed Sarsasapogenin derivatives AA13, AA36, and Diosgenin deriva-

tive M15 as antioxidants in neurodegenerative diseases.^18–20(Fig. 1) It is

also reported that phenyl carbamate as potent cholinesterase inhibitors is

often used in neuroprotective drug design.²¹For these purposes, we have

selected twin drug strategy and steroid and phenyl carbamate to prepare

multifunctional potent drugs for VD. In the present study, we designed 23

new compounds (ML1-ML23), next we summarized structure–activity re-

lationships (SAR) on these compounds, obtained a new potent drug

candidate for anti-VD. Especially compound ML5 fused rivastigmine and

compound 3, which were pharmacophoric features of cholinesterase in-

hibition and antioxidant (Fig. 2).

Compound treatment groups were treated with corresponding com-

pounds while control and model cells were treated with medium for 2 h

and then compound treatment cells and model cells were exposure to

OGD for 4 h and reperfusion for 2 h. Control cells were incubated in the

high-glucose medium and normal oxygen environment. As shown in

Fig. 4, the model group had lower cell viability (OGD, 60.16%)

compared with the control (100%, p < 0.001). In general, compounds

ML2-ML5, ML8-ML10, ML13 displayed high cell viability compared to

the model. Compounds ML2 (10–50

M), ML5 (1–50 M), ML6 (1 M), ML8 (10–50

ML10 (100 M), ML12 (0.1–1 M), ML13 (1 M), ML14 (1–10

ML15 (0.1–1 M) signiﬁcantly increased cell viability compared with

μ

M), ML3 (100

M), ML9 (1–50

M) and

μ

M), ML4 (20–50

μ

M),

μ

Compound ML5 signiﬁcantly increased cell viability against oxida-

tive injuries stimulated by oxygen-glucose deprivation, and it also has a

good ChEs inhibitory activity in vivo. Mechanistic studies revealed that

compound ML5 was a potent Nrf2 activator to promote the nuclear

translocation of Nrf2, which facilitates the expression of a series of

Nrf2-driven antioxidant molecules, including superoxide dismutase

(SOD) and glutathione (GSH). As a new potential drug candidate for

the treatment of VD, ML5 showed strong antioxidant capacity, appro-

priate inhibition of ChEs, and induced neuroprotection following in

vivo ischemia–reperfusion.

model (Fig. 4, Fig. 1S, Supporting Information). In addition, at the

indicated concentrations, some compounds shown in Fig. 4 have a ten-

dency to increase cell viability in SH-SY5Y cells, but there was no sta-

tistical signiﬁcance compared with the model. Compound ML7 did not

exhibit the effect to increase cell viability in the OGD model (Fig. 1S,

Supporting Information). It was remarkable that the linear alkyl and

branched alkyl carbamate derivatives ML2-ML4 were much more

effective in increasing the survival rate of the SH-SY5Y cells. In addition,

ML1 and ML7, bearing ethyl and 3-phenylpropyl, respectively, showed

weaker neuroprotective abilities compared with compounds ML2-ML4.

This demonstrated that introducing phenyl groups, which could

strengthen the structural rigidity of the compound, would enhance the

neuroprotective effect.

2. Results and discussion

2.1. Chemistry

Disubstituted compounds ML5 and ML6 showed signiﬁcant differ-

ences in the activity on OGD model, compound ML5 with N-ethyl-N-

methyl showed higher cell viability than ML1 and ML6 with N-ethyl and

N, N-diethyl. ML8 and ML9 with cyclopentyl and cyclohexyl can afford

neuroprotection, of which no clear SAR (structure–activity relationship)

can be deﬁned. Moreover, pyrrolidinyl substituted compound ML10

showed higher cell viability than piperidyl substituted compound ML11,

as shown in Figure 1S. It was notable that compounds ML12-ML15

exhibited higher cell viability in OGD model of SH-SY5Y cells, which

indicated that introducing substituted or unsubstituted benzyl carba-

mate would be beneﬁcial to improve the activity of cytoprotective.

However, when hetero atoms were introduced to the aryl substituents,

The synthetic routes for compounds ML1-ML23 were shown in

Scheme 1. Compound 1, 4,5-Didehydrotigogenone was obtained using

Oppenauer oxidation methods began with commercially available Dio-

sgenin. In the presence of tertbutyldimethylsilyl chloride (TBSCl),

compound

1 reacted with 2,3-dicyano-5,6-dichlorobenzoquinone

(DDQ) to produce compound 2 in 95% yield, which reacted to com-

pound 3 with Li, diphenyl and diphenyl methane in tetrahydrofuran

(THF). Compound 3, Bis (trichloromethyl) Carbonate (BTC) and N,N-

◦

diisopropylethylamine (DIPEA) were stirred in DCM at 0 C, followed

by addition of different amines, which led to the produce of compounds

ML1-ML23, as shown in Fig. 3.

compounds displayed potent inhibitory action at 10–100

μM. Cell

viability was increased by the addition of compounds ML19-ML23 at

Fig. 1. The structure and protective activities of compounds AA13, AA36 and M15.

2

G.-X. Yang et al.

Bioorganic & Medicinal Chemistry 37 (2021) 116109

Fig. 2. Twin drug strategy design for multifunctional anti-VD agents.

Scheme 1. Synthetic route of compound 1, 2, 3 and ML1-ML23. ^aReagents and conditions: (i) aluminium isopropoxide, cyclohexanone, PhMe, 105 ^◦C; (ii) DDQ,

TBSCl, 1,4-dioxane, rt; (iii) Li, Diphenyl, Diphenyl methane, THF, 70 ^◦C; (iv) BTC, amines, DIPEA, DCM, 0 ^◦C.

low concentration compared with model while being below the model

value at high concentration. Nevertheless, we have not found a clear

SAR. In general, the neuroprotection of compounds was related to

concentration, which displayed neuroprotection activity at low con-

centration and inhibitory effect at high concentration.

in the mice cerebral cortex at 30 min after oral administration. At 15 min

after gavage of 15 mg/kg ML5, the ChEs activity decreased to 76.3%

compared with the normal saline group. The ChEs inhibition activity

was strongest at 30 min after received ML5 (64.6%) and lasted until 45

min (66.2%). The results showed that ML5 signiﬁcantly inhibited ChEs

activity in the brain and provided evidence that ML5 could cross the

blood–brain barrier.

At last, we deﬁned the neuroprotection activity as the percentage

from model value, deﬁned as 0%, to reach the control value, as 100%

(Table 1). We found that most compounds emerged potent neuro-

protection, especially the new phenyl carbamate ML5, providing

strongest neuroprotective activity of 76.00 ± 4.02% cytoprotection rate

2.4. ML5 prevented loss of cell viability and reduced ROS levels in SH-

SY5Y cells exposed to H₂O₂

at 10

μM (Table 1). In addition, ML5 exhibited excellent bioactivities

from 1 to 50

μ

M especially at low concentrations. Although some other

Oxidative stress produced by reactive radical species has been

recognized as a common pathological feature in VD. Reactive oxygen

species (ROS), together with the loss of function of many antioxidant

defense enzymes which could cause an imbalance between the forma-

tion of cellular oxidants and the antioxidative processes. First, we

evaluated the cytoprotective effects of ML5 at different concentrations

compounds, such as ML9, similarly improve the cytoviabilities at 1 μM,

their cytoprotectivity still not as good as ML5 at each concentration.

Then, we focused on ML5 for its ChEs inhibitory activity.

2.3. Effect of ML5 on ChEs activity in vivo.

(0.01–100 μM) on SH-SY5Y cells exposed to H₂O₂at 200 μM using the 3-

The in vivo ChEs inhibitory activity of ML5 was evaluated and

compared with rivastigmine which was used as the positive controls

(Fig. 5). The results showed that rivastigmine (2 mg/kg, i.g., dose)

inhibited ChEs activity to 49.9% compared with the normal saline group

(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)

assay and ROS assay. As depicted in Figure 6A, the MTT assay indicated

that H₂O₂at 200 μM for 24 h induced a decrease in the viability of SH-

SY5Y cell (77.8%) compared with control (100%, p < 0.05). However,

3

G.-X. Yang et al.

Bioorganic & Medicinal Chemistry 37 (2021) 116109

Fig. 3. Structures of the compounds ML1-ML23.

Fig. 4. Protection against Oxygen ꢀ Glucose Deprivation (OGD) of compounds ML2-ML5, ML8, ML9, ML12, ML13. Data are expressed in percentage with respect to

100% control. Data were expressed as mean ± SEM (n > 4). ***p < 0.001 compared with control group; #p < 0.05, ##p < 0.01, ###p < 0.001 compared with

model group by one-way ANOVA analysis followed by a post hoc (Student-Newman-Keuls) test for multiple comparisons.

the H₂O₂-induced the decrease in cell viability was signiﬁcantly atten-

uated when cells were pretreated with ML5 for 2 h at 1 M (87.7%), 5

M (95.2%), 10 M (96.5%) (p < 0.05). Based on these data, we decided

to utilize the ML5 at 1, 5, 10 M in the next study.

We next studied whether ML5 at 1, 5, 10 M would be effective in

decreasing ROS after 200

H₂O₂stimulation. The 2^′-

7^′dichloroﬂuorescin diacetate (DCFH-DA) is

non-ﬂuorescent

compound. This probe is cell-permeable and is hydrolyzed to the 2^′-

7^′dichlorodihydroﬂuorescein (DCFH) carboxylate anion in the cell,

which is retained in the cell. DCFH is later oxidized by intracellular ROS

into 2^′-7^′dichloroﬂuorescein (DCF) which is a ﬂuorescent compound.

The ﬂuorescence intensity can reﬂect ROS levels. Cells were pretreated

μ

M

with ML5 for 2 h, and then 10

μM DCFH-DA was added, and followed by

a

H₂O₂(200 M), then intracellular ﬂuorescence intensity were analyzed

μ

4

G.-X. Yang et al.

Bioorganic & Medicinal Chemistry 37 (2021) 116109

levels compared to the group treated with H₂O₂alone (5

μM 114.2%, 10

Table 1

Neuroprotection for compounds on SH-SY5Y cells after Oxygen-Glucose Depri-

vation (OGD).

μ

M 111.4%, p < 0.05). To further conﬁrm the results of the ROS, we

detected DCFH-DA green dye which, by ﬂuorescence microscope, cor-

relates a proportional green intensity to the quantity of ROS present in a

cellular system. As shown in Figure 6C, the results were consistent with

the quantitative results measured by the microplate reader. The ﬂuo-

rescence of the H₂O₂group was stronger than that of the control group,

and the ﬂuorescence of the other three groups treated with ML5 were

decreased to some extent compared with H₂O₂group.

Compd.

Conc. (μM)

Neuroprotection (%)

ML2

10

20

50

100

20

50

1

37.43 ± 2.02

67.66 ± 6.10

56.94 ± 4.73

43.36 ± 2.05

33.76 ± 3.46

49.76 ± 5.14

62.07 ± 5.46

76.00 ± 4.02

53.38 ± 4.74

43.16 ± 4.08

33.77 ± 3.32

49.20 ± 3.97

37.07 ± 3.16

38.30 ± 2.12

34.36 ± 2.30

46.42 ± 2.72

41.89 ± 1.61

48.77 ± 1.91

37.77 ± 2.13

29.68 ± 2.65

49.82 ± 3.92

36.41 ± 2.19

25.08 ± 5.95

27.45 ± 8.18

20.43 ± 5.05

25.93 ± 7.22

ML3

ML4

These results and those described above showed that ML5 induced a

signiﬁcant neuroprotective effect on neuronal cells subjected to H₂O₂,

increased cell activity and decreased ROS.

ML5

10

20

50

1

ML6

ML8

2.5. ML5 exerted antioxidant properties in H₂O₂-treated SH-SY5Y cells.

10

20

50

1

ROS production is a consequence of the aerobic metabolism that is

normally controlled by the endogenous antioxidant systems.¹⁹Living

cells contain a large number of antioxidant species and antioxidant en-

zymes, both of which serve to counteracted harmful effects caused by

free radicals.²²GSH is a powerful antioxidant and SOD is one of the most

important antioxidants enzymes.

ML9

10

20

50

100

0.1

1

ML10

ML12

The antioxidative effects of ML5 were studied on SH-SH5Y cells by

evaluating their SOD and GSH levels after pre-protected by ML5 (1, 5,

ML13

ML14

1

10

μ

M) for 2 h and then exposed to 200

μ

M hydrogen peroxide for 24 h.

10

0.1

1

ML15

As shown in Figure 7A and 7B, when treated with 200 μM H₂O₂for 24 h,

the activity of SOD of SH-SY5Y cells in the H₂O₂group was signiﬁcantly

decreased to 50.7% than control (p < 0.05), and the GSH levels were

signiﬁcantly reduced to 74.4% compared with the control group (p <

0.05).

Neuroprotection was deﬁned as the percentage from model value (deﬁned as

0%) until reaching the control value (deﬁned as 100%). Data were expressed as

mean ± SEM (n > 4).

However, the SOD activities in cells that pretreated with 10 μM ML5

for 2 h were signiﬁcantly restored to 101.5% compared with the H₂O₂

group (p < 0.05), meanwhile pretreated with 1 μM and 5 μM ML5 for 2

h, restored SOD activities to 82.7% and 83.3%, respectively, although

the difference did not reach a statistical signiﬁcance. The GSH levels in

the cells pretreated with 10 μM ML5 for 2 h were signiﬁcantly restored

to 93.6% compared with the H₂O₂group (p < 0.05). The cells pretreated

with 1 μM and 5 μM of ML5 for 2 h restored the GSH levels to 81.5% and

91.0% respectively. These results indicated that ML5 is effective to

alleviate oxidative damage in SH-SY5Y cells induced by H₂O₂, which is

mainly related to enhancing the activities of SOD and GSH.

2.6. ML5 activated Nrf2 pathway in SY5Y cells

Under normal physiological conditions, Nrf2 binds to Keap1 in the

cytoplasm and degraded by programmed ubiquitination and proteaso-

mal degradation, thus maintaining a low level. When Nrf2 pathway is

activated, its translocation to the nucleus induces the expression of

several genes implicated in antioxidant defences.²³

We studied whether ML5 would induce translocation of Nrf2 to the

nucleus in SH-SY5Y cells by western blot analysis. As shown in Fig. 8, the

result of western blot showed that 10 µM ML5 signiﬁcantly increased

Nrf2 localization in the nucleus compared with that seen in the un-

treated control after 12 h treatment in SH-SY5Y cells.

Fig. 5. Anti-ChEs activity of ML5. ML5 inhibited ChEs activity signiﬁcantly.

Bar graph shows the cholinesterase activity after gavage administration of

normal saline or rivastigmine 30 min and ML5 15 min, 30 min, 45 min, 3 h, 12

h, respectively. Cholinesterase activity of control brains were considered as

100% (black bar). Data are expressed in percentage with respect to 100%

control. Data were expressed as mean ± SEM (n = 6 mice in each group, male).

*p < 0.05, **p < 0.01, ***p < 0.001 compared with control group by one-way

ANOVA analysis followed by a post hoc (Student-Newman-Keuls) test for

multiple comparisons.

2.7. ML5 ameliorated learning and memory impairments, reduced

neuronal loss in the hippocampus after bilateral common carotid artery

occlusion (BCCAO)

In the above experiments, we found that ML5 has ChEs inhibitory

activity and can protect the oxidative damage of H₂O₂to SH-SY5Y cells.

The hypothesis that chronic cerebral hypoperfusion contributes to the

progression of dementia was proposed long ago, and recent studies have

shown that hypoperfusion resulted in a chain of disruption of homeo-

static interactions including oxidative stress.²⁴Furthermore, substantial

evidence has shown that 2VO can impair learning²⁵. Therefore, we

decided to assay ML5 in animal models of BCCAO.

in a microplate reader and ﬂuorescence microscope. As shown in

Figure 6B, 30 min after H₂O₂stimulation, the level of ROS in the H₂O₂

group increased to 141.8% compared to the control group (100%, p <

0.05). ML5 at 1

μM was not effective in preventing the increased of ROS

induced by H₂O₂, while 5

μ

M and 10 μM ML5 signiﬁcantly reduced ROS

5

G.-X. Yang et al.

Bioorganic & Medicinal Chemistry 37 (2021) 116109

Fig. 6. Effect of ML5 on cell viability of SH-SY5Y exposed to 200 µM H₂O₂. (A) Bar graph shows ML5 pretreated at different concentrations (0.01–100

the viability of SH-SY5Y cells exposed to 200 M H₂O₂for additional 24 h. Data are expressed as percentage with respect to the control which was presented as 100%.

(B) Bar graph shows the effects of 2 h pretreatment of ML5 at 1 M, 5 M, 10 M on the cellular levels of reactive species generation, H₂O₂was utilized at 200 M for

further 30 min (by microplate reader). ROS detected in control cells was considered as 100%. (C) The ﬂuorescence images qualitatively show the intracellular ROS

μM) for 2 h on

μ

level after pretreatment with ML5 at 1 μM, 5 μM, 10 μM for 2 h and 200 μM H₂O₂for further 30 min (Fluorescence microscope shoot). Data were expressed as mean

± SEM (n = 3). #p < 0.05, ##p < 0.01, ###p < 0.001 compared with control; *p < 0.05, **p < 0.01, ***p < 0.001 compared with H₂O₂group by one-way ANOVA

analysis followed by a post hoc (Student-Newman-Keuls) test for multiple comparisons.

The Morris water maze is one of the most commonly used tests to

measure the hippocampus-related spatial learning capacity in chronic

cerebral hypoperfusion in rats. Therefore, we examined the learning and

memory functions by analyzing escape latency and frequency in the

platform quadrant using the Morris Water Maze test (MWM). SD rats

were divided into sham operation, BCCAO, ML5 low-dose (3 mg/kg, i.

g.), ML5 high-dose (15 mg/kg, i.g.), and donepezil (1 mg/kg, i.g.)

groups, respectively. As shown in Figure 8D, the mean escape latency of

all groups over the course of the 5 learning days decreased at 4 weeks

after BCCAO.

learning and memory deﬁcits in our BCCAO rats. Importantly, the

escape latency, either of ML5 groups or the donepezil group, decreased

faster than the BCCAO group (Figure 9A). In the probe trial, the group

received ML5 with a dose of 3, 15 mg/kg, and the group received

donepezil 1 mg/kg stayed longer in the target quadrant than BCCAO

group (p < 0.05, Figure 9B). At the same time, no statistical difference in

swimming speed between any of the groups (Figure 9C). The speciﬁc

movement trajectory of each group of rats was shown in Figure 9D.

Figure 8E showed the overview of the hippocampal and enlarged views

of CA1 regions by HE staining. The number of neurons in the sham group

was higher than in other groups. Whether ML5 treated groups or

donepezil treated groups, the number of neurons was higher than the

BCCAO groups, indicated that ML5 treatment signiﬁcantly reduced the

loss of neurons after BCCAO. These results suggested that ML5 would

The escape latency of the BCCAO group decreased slower than the

sham group’s during the training sessions, and during the probe trial

phase, the BCCAO group stayed shorter in the target quadrant than the

that in the sham group (p < 0.05, Figure 9B). Both of above-validated

6

G.-X. Yang et al.

Bioorganic & Medicinal Chemistry 37 (2021) 116109

Fig. 7. The effects of ML5 on the levels of SOD and GSH in SH-SY5Y cells treated with 200

μ

M H₂O₂. (A) Bar graph shows the activity of SOD in SH-SY5Y cells after

ML5 pretreatment at 1, 5, 10 μM for 2 h and exposed to 200 μM H₂O₂for additional 24 h. (B) Bar graph shows the level of GSH in SH-SY5Y cells after ML5

pretreatment at 1, 5, 10 μM for 2 h and exposed to 200 μM H₂O₂for additional 24 h. Data are expressed in percentage with respect to 100% control. Data were

expressed as mean ± SEM (n = 3–5). #p < 0.05, ##p < 0.01, ###p < 0.001 compared with control; *p < 0.05, **p < 0.01, ***p < 0.001 compared with H₂O₂group

by one-way ANOVA analysis followed by a post hoc (Student-Newman-Keuls) test for multiple comparisons.

Fig. 8. Compound ML5 activates Nrf2 pathway. (A) Western blot analyses of cytoplasm Nrf2 protein content from SH-SY5Y cells treated for 12 h with 10

μ

M ML5 (n

M ML5

= 5) and relative expression levels of cytoplasm Nrf2. (B) Western blot analyses of nucleus Nrf2 protein content from SH-SY5Y cells treated for 12 h with 10

μ

(n = 5) and relative expression levels of nucleus Nrf2. Data were expressed as mean ± SEM (n = 5). #p < 0.05, ##p < 0.01, ###p < 0.001 compared with control;

*p < 0.05, **p < 0.01, ***p < 0.001 compared with control group by one-way ANOVA analysis followed by a post hoc (Student-Newman-Keuls) test for multiple

comparisons.

mitigate the learning and memory deﬁcits caused by BCCAO.

stress and neuroinﬂammation. Therefore, the Nrf2 pathway is being

increasingly considered a therapeutic target for Neurodegenerative

disease.

2.8. ML5 exerted an antioxidant and neuroprotective effect by activating

In our above research, we found that ML5 could reverse the increase

of ROS and the decrease of SOD and GSH caused by H₂O₂damage in SH-

SY5Y cells. Therefore, we decided to explore whether the neuro-

protective effect of ML5 on BCCAO rats is related to its antioxidant ca-

pacity and whether it is achieved by activating the Nrf2 pathway. We

then tested the SOD activity and GSH content in the cerebral cortex of

each group of rats to observe whether ML5 alleviated oxidative damage

in BCCAO rats.

Nrf2 pathway.

Transcription factor Nrf2 is the main protein that regulates oxidative

stress in the body and it is a transcription factor that induces the

expression of a great number of cytoprotective and detoxiﬁcation

genes.²⁶There are pieces of evidence that Keapl-Nrf2-ARE signal

pathway is the most important signal pathway in the body to resist

excessive ROS and maintain redox homeostasis, as it reduces oxidative

7

G.-X. Yang et al.

Bioorganic & Medicinal Chemistry 37 (2021) 116109

Fig. 9. Neuroprotective effect and cognitive recovery of ML5 on BCCAO rats. (A) Timeline of ML5 treatment and behavioral assessment. Mice were treated with ML5

(4 mg/kg or 20 mg/kg) two week prior to the beginning of behavioral testing and treatment continued throughout the experiment. After testing, animals were

sacriﬁced and tissue was harvested. ML5 treatment lasted a total of 46 days. (B) Bar graph shows the quantitative analysis of the mean speed in 1 min at day 5 of

MWM test between the different groups. (C) Bar graph shows the time spent in the target quadrant in the probe trial. (D) Line chart shows the escape latency changes

in the different groups from days 1 to 5 of MWM test. (E) Representative photographs of probe test in the WMW test. (F) Representative photographs of HE staining

(magniﬁcation: ×40) in the hippocampal and the enlarged CA1 region (magniﬁcation: ×100). Data are presented as mean ± SEM; n = 8–13. #p < 0.05, ##p < 0.01,

###p < 0.001 compared with sham; *p < 0.05, **p < 0.01, ***p < 0.001 compared with BCCAO group by one-way ANOVA analysis followed by a post hoc (Student-

Newman-Keuls) test for multiple comparisons.

8

G.-X. Yang et al.

Bioorganic & Medicinal Chemistry 37 (2021) 116109

As shown in Figure 10A-B, the SOD activity and GSH levels of BCCAO

formed by benzene ring of ML5 and TRY 341 further stabilizes the

group were decreased to 59.8% and 60.6% severally compared with

control (p < 0.05). ML5 suppressed the BCCAO-induced reduction in the

activity of the SOD and the levels of GSH. The SOD activity of the 3 mg/

kg treatment group (85.9%) and the 15 mg/kg (98.4%) treatment group

was signiﬁcantly higher than that of the BCCAO surgery group (both p <

0.05). In addition, the GSH level in the 15 mg/kg (81.4%) treatment

group was also signiﬁcantly higher than the BCCAO surgery group, but

the GSH level in the 3 mg/kg treatment group (68%) was not signiﬁ-

cantly recovered due to the larger standard error.

binding of ML5 to hAChE. Due to the traction of cation-π interaction

formed by TRY 286 and protonated nitrogen, the pose of rivastigmine is

rotated by about 90^◦compared to ML5 (Figure 11A and C). Neverthe-

less, the interactive features of the phenyl carbamate portion of riva-

stigmine are extremely similar to that of ML5. TYR 124 forms a

hydrogen bond interaction with the oxygen atom of the phenyl ester

group. The π-π stacking interaction is also conserved by interacting with

PHE 338, which could reinforce the interaction between rivastigmine

and hAChE. In addition, ML5 and rivastigmine can act on AS region TRP

86 via hydrophobic interaction. The MM-GBSA ΔGbinding values for

ML5 and rivastigmine with hAChE are ꢀ 63.166 kcal/mol and ꢀ 56.262

kcal/mol, respectively. The binding afﬁnity of ML5 is slightly superior

to that of rivastigmine, probably because ML5 could be stabilized by a

hydrophobic hole composed of residues located at the entrance of an

active site gorge, such as HIS 287, TRP 286, and GLU 292. It is evi-

denced that PAS (TYR 72, ASP 74, TYR 124, TRP 286) and AS (TRP 86)

are promising binding sites for the treatment of AD and TYR 124 is a

signiﬁcant anchoring point for the binding of small molecular in-

hibitors.^27,28The strong interactions between ML5 with PAS and AS,

especially with PAS, combined with the superior binding afﬁnity be-

tween ML5 and hAChE would guarantee ML5 as a bioactive molecule

with effective neuroprotection.

Aiming to investigate the mechanism underlying the antioxidant

stress and neuroprotective effects of ML5, we examined the protein

expression level of Nrf2 in the cerebral cortex. The Western blot results

revealed that ML5 treatment signiﬁcantly increased the expression of

Nrf2 protein in the cerebral cortex compared with the BCCAO group (p

< 0.05). Therefore, the antioxidant and neuroprotective effects of

compound ML5 are closely related to the activation of Nrf2.

2.9. Insights into bioactive mechanisms

Given that ML5 possesses potent inhibitory activities targeting

hAChE (PDB ID: 6F25) compared with its counterpart rivastigmine,

molecular docking studies were performed to compare their potential

bioactive mechanisms. There are ﬁve substrate-binding regions in the

active site gorge of hAChE, including catalytic triad residues, oxyanion

hole, anionic site (AS), acyl pocket, and peripheral anionic site (PAS).²⁷

As shown in Figure 11A, both ML5 and rivastigmine could properly

occupy the active site gorge of hAChE. As for ML5 (Figure 11B), its

phenyl carbamate moiety goes deeply into the PAS region of hAChE.

The carbonyl group of phenyl carbamate moiety forms a stable

The Keap1 (PDB ID: 3VNH) plays a key role in the negative regula-

tion of Nrf2-mediated cytoprotective response. C-terminal Kelch domain

of Keap1 is a recognition module for Nrf2 and inhibition of Kelch-Nrf2

interaction is recognized as an effective strategy for the therapy of

neurodegenerative diseases.^29,30We docked ML5 to the Kelch domain of

Keap1 to investigate its potential bioactive mechanism. As illustrated in

Fig. 12, ML5 is properly anchored into the Kelch domain of Keap1 with

MM-GBSA ΔGbinding = ꢀ 49.683 kcal/mol. The oxygen atom of phenyl

hydrogen bond interaction with TYR 124. The π-π stacking interaction

Fig. 10. The effects ML5 on the levels of antioxidants and the expression of Nrf2 in BCCAO rats. (A) Bar graph shows the activity of SOD in all groups at 4 weeks after

BCCAO. (B) Bar graph shows the levels of GSH in all groups at 4 weeks after BCCAO. (C) Quantitative analysis of the western blotting shows the Nrf2 expression levels

in the cortex of all groups. Data are presented as mean ± SEM; n = (8–13). #p < 0.05, ##p < 0.01, ###p < 0.001 compared with sham; *p < 0.05, **p < 0.01, ***p

< 0.001 compared with BCCAO group by one-way ANOVA analysis followed by a post hoc (Student-Newman-Keuls) test for multiple comparisons.

9

G.-X. Yang et al.

Bioorganic & Medicinal Chemistry 37 (2021) 116109

Fig. 11. The alignment of docking poses of ML5 and rivastigmine (A), binding models of ML5 (B) and rivastigmine (C) interacting with hAChE (PDB ID: 6F25).

Proteins are shown as green cartoons. Small molecules and key residues are drawn as sticks. Carbons in ML5, rivastigmine and key residues are colored in pink,

orange and green, respectively. All oxygen, nitrogen and hydrogen (polar hydrogen only) atoms are colored in red, blue and white, respectively. Aromatic ring

centers and positron centers are presented as yellow and blue spheres. Hydrogen bonds,

and cyan dotted lines, respectively.

π-π stacking interactions and cation-π bonds are depicted in purple, yellow

carbamate moiety directly interacts with SER 602 and forms a stable

hydrogen bond with its hydroxy group. Moreover, the tail tetrahy-

dropyran group is toward the edge of the active site and interacts with

ASN 382 by hydrogen bond interaction. The benzene ring of ML5 forms

most promising drug candidate ML5 was selected to study its acute

toxicity by oral administration at a dose of 500 mg/kg (n = 7). All the

animals were observed for 14 days after administration. After observa-

tion, all the animals were alive and performed healthily from the color of

their fur, weight changes, behavioral symptoms, water, and food con-

sumption. All mice were sacriﬁced on the 15th day to observe the

changes in the heart, liver, and kidneys. All organs observed were

normal. This result displayed that ML5 had well-tolerated at a dose of

500 mg/kg with no acute toxicity.

a π-π interaction with TYR 334, which would effectively stabilize the

binding of ML5 into Keap1. In addition, hydrophobic interaction stem-

med from ARG 380 and ARG 415 contributes to the molecular binding to

some extent. Shih-Ching Lo et al.3 found that residues ARG 380, ASN

382, and ARG 415 of Keap1 participate in hydrogen bond interactions

with the side chain atoms of Nrf2, and residues in Keap1 including TYR

334, ASN 382, and SER 602 participate in hydrogen bonding in-

teractions with the backbone of Nrf2. The mutagenesis analysis of Keap1

showed that mutation of individual residues for ARG 380, TYR 334, ASN

382 and ARG 415 signiﬁcantly disrupts the capability of Keap1 binding

to Nrf2, and these residues are required for repression of Nrf2-dependent

transcription. Therefore, ML5 could ﬁrmly occupy the Kelch domain of

Keap1, which would interfere with the recognition between Keap1 and

Nrf2, thus exert promising neuroprotective activity.

3. Conclusions

We designed, synthesized and biologically evaluated 23 new candi-

date compounds for VD treatment, which fused steroid and phenyl

carbamate together according to our previous work. Some of the test

compounds (ML2, ML3, ML4, ML5, ML6, ML8, ML9, ML10, ML12,

ML13, ML14 and ML15) signiﬁcantly increased cell viability with the

OGD model. Among the series, compound ML5 showed the highest

neuroprotection activity in the OGD model and appropriate inhibition of

2.10. Acute toxicity study

ChEs at oral 15 μM in vivo, which indicated ROS scavenging and cell

viability increasing against SH-SY5Y cells damage induced by H₂O₂.

Further pharmacological experiments showed that ML5 could promote

Nrf2 nuclear translocation and hence promote the levels of SOD and

GSH, affording protection of SH-SY5Y cells from oxidative insults. In the

These studies indicated that compound ML5 had obviously anti-VD

effect in vivo and in vitro. Acute toxicity study is considered to be an

essential basis for the safety evaluation of new drugs. Therefore, the

10

G.-X. Yang et al.

Bioorganic & Medicinal Chemistry 37 (2021) 116109

Fig. 12. Predicted binding model of ML5 interacting with Keap1 (PDB ID: 3VNH). Painting rules are the same as mentioned in Fig. 11.

BCCAO-induced VD model, ML5 mitigated the learning and memory

deﬁcits, which signiﬁcantly reversed the BCCAO-induced loss of neurons

and improved the levels of SOD and GSH in the hippocampus. In the

molecular studies, ML5 exhibited good binding afﬁnity in the active site

of AChE and ﬁrmly occupied the Kelch domain of Keap1, which would

interfere with the recognition between Keap1 and Nrf2, thus exert

promising neuroprotective activities. In addition, ML5 was found to be

well-tolerated and nontoxic up to 500 mg/kg, po, dose. Put all together,

the most promising multitargeted drug ML5 could be deemed as a po-

tential drug in anti-VD for further study.

4.2. General method for synthesis of compound 1

In a 500 mL round bottomed ﬂask, the diosgenin (82.46 g, 0.2 mol),

aluminium isopropoxide (1.5 equiv), and cyclohexanone (1.5 equiv)

were suspended in 300 mL toluene. The mixture was stirred for 4 h at

90 ^◦C until the reaction was completed according to TLC detection. The

toluene was removed under reduced pressure. The residue was dissolved

in dichloromethane, ﬁltered with Buchner funnel. The ﬁltrate was dried

with Na₂SO₄and concentrated under vacuum. The mixture was puriﬁed

on a silica gel column using mixtures of ethyl acetate/petroleum ether =

1:20 as eluent, obtaining the corresponding compound 1 as a white

amorphous powder.

4. Experimental section

4.1. General

4.3. General method for synthesis of compound 2

Reagents and solvents were purchased from common commercial

suppliers, which were not treated further, unless speciﬁed. Reactions

were detected by TLC (Yantai Huanghai, China) and tested with UV

light (λ = 254 nm) or 1% vanillin sulfuric acid solution. Products were

puriﬁed by chromatographic separations (200–300 mesh), which were

produced from Yantai Huanghai, China. ¹H spectra were recorded by a

Bruker AV-400 instrument in CDCl₃with tetramethylsilane (TMS,

Me₄Si) as an internal standard. ¹H spectra were expressed by δ (ppm),

J, spin multiplicities. ¹³C spectra were recorded by Ascend-600 in-

strument in CDCl₃which were characterized by δ (ppm). High-

resolution mass (HRMS, ESI) spectral data were acquired by

ABSCIEX 4600 TOF electrospray mass spectrometer in positive ion

mode.

Chlorotrimethylsilane (543.2 mg, 0.05 equiv) was added slowly to a

solution of compound 1 (41.03 g, 0.1 mol) in tetrahydrofuran and 1,4-

dioxane (200 mL, V_THF/V_DO= 1/1) at 0 ^◦C. Then, 2,3-dicyano-5,6-

dichlorobenzoquinone (1.5 equiv) was added. After 12 h of reaction,

the mixture was diluted with water, extracted with DCM, dried over

anhydrous sodium sulfate, ﬁltered and the solvent evaporated. The

resultant crude mixture was isolated by column chromatography using

mixed eluent (ethyl acetate/petroleum ether = 1:15), the compound 2

was acquired as a white amorphous powder.

4.4. General method for synthesis of compound 3

Compounds ML1-23 were analyzed by HPLC to determine their

purity. The analyses were performed on Agilent 1260 series HPLC sys-

In a 250 mL double-neck ﬂask, lithium metallic block (9.92 g, 1.43

mol), biphenyl (58.6 g, 380 mmol) and diphenyl methane (28.5 mL)

were suspended in 300 mL dry THF under N₂atmosphere. The mixture

tem (C-18 column, Shimadzu, 5 μm, 4.6 mm × 150 mm) at room tem-

perature. The HPLC chromatograms of compounds ML1-23 are

included in the Supporting Information. All the tested compounds were

dissolved in acetonitrile. Acetonitrile and water (98:2) were used as

mobile phase, and the ﬂow rate was set at 0.6 mL/min. The maximal

absorbance at the range of 203 nm was used as the detection wave-

length. The purity of compounds ML1-ML23 was determined by the

HPLC analysis (>95%), which meets the purity requirement by the

Journal.

◦

was reacted at 70 C for 30 min until mixed system had turned blue.

Then, the solution of compound 2 (95 mmol) was added. After the

reaction was complete (detected by TLC), the mixture was cooled to

room temperature before adjust the pH by adding 10% aq. HCl and

extracted with DCM. The organic layer was dried with Na₂SO₄and

concentrated under vacuum. The compound 3 was obtained by col-

umn chromatography using ethyl acetate/petroleum ether = 1:20 as

eluent.

11

G.-X. Yang et al.

Bioorganic & Medicinal Chemistry 37 (2021) 116109

4.5. General method for synthesis of compounds ML1-ML23

ML7: White amorphous powder, yield 84%. ¹H NMR (400 MHz,

CDCl₃) δ ¹H NMR (400 MHz, CDCl₃) δ 7.27 (5H, m), 7.13 (1H, d, J = 8.7

Hz), 6.62 (1H, dd, J = 8.4 Hz, 2.5 Hz), 6.56 (1H, s), 4.70 (1H, s), 4.45

(1H, m), 3.90 (1H, t, J = 7.8 Hz), 3.60 (2H, m), 3.49 (1H, d, J = 5.9 Hz),

3.39 (1H, t, J = 10.8 Hz), 2.83 (2H, m), 1.00 (3H, d, J = 6.8 Hz), 0.82

(3H, s), 0.80 (3H, d, J = 6.3 Hz); ¹³C NMR (151 MHz, CDCl₃) δ 153.64,

145.03, 138.10, 132.46, 128.41(2C), 127.81(2C), 126.29, 126.09,

115.26, 112.64, 109.46, 80.94, 66.89, 62.84, 62.22, 55.30, 51.34,

43.73, 41.65, 40.80, 39.91, 38.34, 35.46, 32.63, 31.49, 31.36, 30.27,

29.56, 28.74, 27.86, 26.44, 24.24, 17.13, 16.42, 14.50. HRMS (ESI):

Calcd for C₃₆H₄₈NO₄[M+H]⁺558.3583; Found: 558.3585.

Compound 3 (184.6 mg, 0.5 mmol) and bis(trichloromethyl)car-

bonate (59.4 mg, 0.2 mmol) were dissolved in dry DCM under N₂at-

◦

mosphere at 0 C. After 30 min of reaction, the different amines (1.5

equiv) were added. TLC was used to determine the reaction end point.

Then, diluted the mixture with water, washed with brine, dried over

Na₂SO₄, concentrated under vacuum and isolated by colum-

n chromatography. At last, we obtained products ML1-ML23.

ML1: White amorphous powder, yield 93%. ¹H NMR (400 MHz,

CDCl₃) δ 7.24 (1H, d, J = 8.5 Hz), 6.87 (1H, dd, J = 8.2 Hz, 1.8 Hz), 6.83

(1H, s), 4.94 (1H, t, J = 5.2 Hz), 4.46 (1H, m), 3.49 (1H, dd, J = 10.1 Hz,

4.5 Hz), 3.39 (1H, t, J = 10.9 Hz), 3.31 (2H, m), 2.85 (2H, m), 1.21 (3H,

t, J = 7.3 Hz), 1.00 (3H, d, J = 6.7 Hz), 0.80 (6H, m); ¹³C NMR (151

MHz, CDCl₃) δ 154.80, 148.76, 137.96, 137.50, 126.07, 121.61, 118.66,

109.30, 80.84, 66.89, 62.30, 55.35, 43.93, 41.64, 40.76, 39.91, 38.05,

36.10, 31.56, 31.41, 30.31, 29.50, 28.80, 27.73, 26.31, 17.17, 16.40,

15.15, 14.52. HRMS (ESI): Calcd for C₂₉H₄₂NO₄[M+H]⁺468.3114;

Found: 468.3096.

ML8: White amorphous powder, yield 87%. ¹H NMR (400 MHz,

CDCl₃) δ 7.24 (1H, d, J = 8.3 Hz), 6.87 (1H, d, J = 8.4 Hz), 6.84 (1H, s),

4.91 (1H, d, J = 7.5 Hz), 4.46 (1H, m), 4.05 (1H, dd, J = 14.1 Hz, 7.0

Hz), 3.49 (1H, dd, J = 8.3 Hz, 3.9 Hz), 3.42 (1H, t, J = 6.7 Hz), 2.87 (2H,

m), 1.00 (3H, d, J = 6.7 Hz), 0.80 (6H, m). ¹³C NMR (151 MHz, CDCl₃) δ

154.33, 148.77, 137.90, 137.41, 126.03, 121.61, 118.67, 109.28, 80.83,

66.87, 62.30, 55.35, 52.95, 43.93, 41.64, 40.76, 39.91, 38.05, 33.17

(2C), 31.56, 31.41, 30.31, 29.49, 28.80, 27.74, 26.31, 23.55 (2C), 17.15,

16.39, 14.52. HRMS (ESI): Calcd for C₃₂H₄₆NO₄[M+H]⁺508.3427;

Found: 508.3412.

ML2: White amorphous powder, yield 91%. ¹H NMR (400 MHz,

CDCl₃) δ 7.23 (1H, d, J = 8.8 Hz), 6.85 (1H, d, J = 8.7 Hz), 6.82 (1H, s),

4.46 (1H, dd, J = 14.8 Hz, 5.6 Hz), 3.58 (2H, m), 3.49 (3H, m), 3.39 (1H,

t, J = 11.2 Hz), 2.85 (2H, m), 1.00 (3H, d, J = 6.7 Hz), 0.80 (6H, m); ¹³C

NMR (151 MHz, CDCl₃) δ 154.92, 148.79, 137.94, 137.47, 126.06,

121.59, 118.64, 109.29, 80.84, 66.88, 62.30, 55.35, 43.94, 41.64,

40.95, 40.77, 39.91, 38.05, 31.92, 31.57, 31.41, 30.31, 29.50, 28.81,

27.74, 26.31, 19.91, 17.14, 16.40, 14.52, 13.74. HRMS (ESI): Calcd for

ML9: White amorphous powder, yield 93%. ¹H NMR (400 MHz,

CDCl₃) δ ¹H NMR (400 MHz, CDCl₃) δ 7.23 (1H, d, J = 8.3 Hz), 6.87 (1H,

d, J = 8.5 Hz), 6.83 (1H, s), 4.87 (1H, d, J = 7.6 Hz), 4.45 (1H, m), 3.56

(1H, dd, J = 10.9 Hz, 6.9 Hz), 3.49 (1H, dd, J = 10.3 Hz, 3.2 Hz), 3.39

(1H, t, J = 10.9 Hz), 2.86 (2H, m), 1.00 (3H, d, J = 6.7 Hz), 0.80 (6H, d,

J = 8.5 Hz); ¹³C NMR (151 MHz, CDCl₃) δ 153.98, 148.78, 137.87,

137.35, 126.00, 121.59, 118.65, 109.27, 80.83, 66.87, 62.30, 55.34,

50.04, 43.93, 41.64, 40.75, 39.91, 38.05, 33.26 (2C), 31.56, 31.40,

30.30, 29.49, 28.80, 27.74, 26.30, 25.47, 24.76 (2C), 17.14, 16.39,

14.52. HRMS (ESI): Calcd for C₃₃H₄₈NO₄[M+H]⁺522.3583; Found:

522.3576.

C

₃₁H₄₆NO₄[M+H]⁺496.3427; Found: 496.3419.

ML3: White amorphous powder, yield 90%. ¹H NMR (400 MHz,

CDCl₃) δ 7.24 (1H, d, J = 8.5 Hz), 6.87 (1H, d, J = 8.4 Hz), 6.84 (1H, s),

4.95 (1H, t, J = 5.8 Hz), 4.46 (1H, dd, J = 13.5 Hz, 7.4 Hz), 3.49 (1H, dd,

J = 11.6 Hz, 2.8 Hz), 3.39 (1H, t, J = 10.9 Hz), 3.25 (2H, dd, J = 13.4 Hz,

6.8 Hz), 2.87 (2H, m), 1.00 (3H, d, J = 6.7 Hz), 0.92 (3H, t, J = 6.7 Hz),

0.80 (6H, m); ¹³C NMR (151 MHz, CDCl₃) δ 154.91, 148.79, 137.94,

137.46, 126.05, 121.59, 118.64, 109.29, 80.84, 66.88, 62.30, 55.35,

43.94, 41.64, 41.24, 40.77, 39.91, 38.05, 31.57, 31.41, 30.31, 29.53,

29.50, 28.90, 28.81, 27.74, 26.31, 22.34, 17.15, 16.40, 14.52, 14.00.

HRMS (ESI): Calcd for C₃₂H₄₈NO₄[M+H]⁺510.3583; Found: 510.3577.

ML4: White amorphous powder, yield 89%. ¹H NMR (400 MHz,

CDCl₃) δ 7.24 (1H, d, J = 8.5 Hz), 6.87 (1H, d, J = 8.4 Hz), 6.84 (1H, s),

4.80 (1H, d, J = 7.5 Hz), 4.46 (1H, m), 3.89 (1H, dd, J = 13.5 Hz, 6.8

Hz), 3.49 (1H, dd, J = 10.1 Hz, 3.3 Hz), 3.39 (1H, t, J = 10.9 Hz), 2.86

(2H, m), 1.23 (6H, d, J = 6.5 Hz), 1.00 (3H, d, J = 6.8 Hz), 0.80 (6H, m);

¹³C NMR (151 MHz, CDCl₃) δ 154.04, 148.76, 137.94, 137.45, 126.06,

121.64, 118.70, 109.30, 80.85, 66.90, 62.32, 55.36, 43.95, 43.39,

41.66, 40.78, 39.93, 38.07, 31.58, 31.42, 30.32, 29.51, 28.82, 27.75,

26.32, 22.95, 17.16, 16.41, 14.54. HRMS (ESI): Calcd for C₃₀H₄₄NO₄

[M+H]⁺482.3270; Found: 482.3252.

ML10: White amorphous powder, yield 84%. ¹H NMR (400 MHz,

CDCl₃) δ 7.24 (1H, d, J = 8.5 Hz), 6.88 (1H, dd, J = 8.4 Hz, 2.3 Hz), 6.85

(1H, s), 4.45 (1H, m), 3.54 (2H, t, J = 6.5 Hz), 3.48 (3H, m), 3.39 (1H, t,

J = 10.9 Hz), 2.87 (2H, m), 1.00 (3H, d, J = 6.7 Hz), 0.81 (H, s), 0.80

(3H, d, J = 6.5 Hz); ¹³C NMR (151 MHz, CDCl₃) δ 153.51, 149.15,

137.84, 137.25, 125.99, 121.71, 118.76, 109.28, 80.85, 66.88, 62.31,

55.36, 46.40, 46.31, 43.94, 41.64, 40.77, 39.93, 38.07, 31.57, 31.41,

30.32, 29.51, 28.81, 27.77, 26.32, 25.82, 24.99, 17.15, 16.41, 14.53.

HRMS (ESI): Calcd for C₃₁H₄₄NO₄[M+H]⁺494.3270; Found: 494.3248.

ML11: White amorphous powder, yield 92%. ¹H NMR (400 MHz,

CDCl₃) δ 7.23 (1H, d, J = 8.4 Hz), 6.86 (1H, d, J = 8.4 Hz), 6.83 (1H, s),

4.95 (1H, t, J = 5.6 Hz), 4.45 (1H, m), 3.48 (1H, m), 3.39 (1H, t, J = 10.9

Hz), 3.26 (2H, dd, J = 13.2 Hz, 6.7 Hz), 2.85 (2H, m), 1.00 (3H, d, J =

6.7 Hz), 0.95 (4H, t, J = 7.3 Hz), 0.80 (6H, m); ¹³C NMR (151 MHz,

CDCl₃) δ 154.09, 149.28, 137.85, 137.27, 126.00, 121.72, 118.77,

109.26, 80.84, 66.87, 62.31, 55.36, 45.46, 45.03, 43.93, 41.64, 40.76,

39.92, 38.06, 31.57, 31.41, 30.31, 29.49, 28.80, 27.76, 26.31, 25.92,

25.52, 24.34, 17.15, 16.40, 14.53. HRMS (ESI): Calcd for C₃₂H₄₆NO₄

[M+H]⁺508.3427; Found: 508.3423.

ML5: White amorphous powder, yield 90%. ¹H NMR (400 MHz,

CDCl₃) δ 7.24 (1H, d, J = 8.5 Hz), 6.86 (1H, d, J = 8.4 Hz), 6.82 (1H, s),

4.45 (1H, m), 3.49 (1H, m), 3.43 (1H, m), 3.39 (1H, t, J = 8.4 Hz), 3.01

(3H, s), 2.86 (2H, m), 1.00 (3H, d, J = 6.8 Hz), 0.81 (3H, s), 0.80 (3H, d,

J = 6.5 Hz); ¹³C NMR (151 MHz, CDCl₃) δ 154.92, 149.25, 137.87,

137.31, 126.01, 121.68, 118.77, 109.26, 80.83, 66.87, 62.31, 55.35,

44.01, 43.93, 41.64, 40.76, 39.92, 38.07, 34.21, 31.57, 31.41, 30.31,

29.49, 28.80, 27.75, 26.32, 17.14, 16.40, 14.52, 13.19. HRMS (ESI):

Calcd for C₃₀H₄₄NO₄[M+H]⁺482.3270; Found: 482.3257.

ML12: White amorphous powder, yield 94%. ¹H NMR (400 MHz,

CDCl₃) δ 7.33 (5H, m), 7.24 (1H, d, J = 8.6 Hz), 6.89 (1H, d, J = 8.4 Hz),

6.86 (1H, s), 5.29 (1H, t, J = 5.7 Hz), 4.45 (3H, d, J = 5.9 Hz), 3.49 (1H,

dd, J = 10.6 Hz, 3.9 Hz), 3.39 (1H, t, J = 10.9 Hz), 2.86 (2H, m), 1.00

(3H, d, J = 6.7 Hz), 0.81 (3H, s), 0.80 (3H, d, J = 6.5 Hz); ¹³C NMR (151

MHz, CDCl₃) δ 154.99, 148.73, 138.10, 138.02, 137.65, 128.77 (2C),

127.73 (2C), 127.68, 126.11, 121.56, 118.61, 109.30, 80.84, 66.89,

62.30, 55.35, 45.34, 43.94, 41.65, 40.77, 39.91, 38.04, 31.57, 31.41,

30.32, 29.51, 28.81, 27.73, 26.31, 17.15, 16.40, 14.53. HRMS (ESI):

Calcd for C₃₄H₄₄NO₄[M+H]⁺530.3270; Found: 530.3252.

ML6: White amorphous powder, yield 87%. ¹H NMR (400 MHz,

CDCl₃) δ 7.24 (1H, d, J = 8.5 Hz), 6.86 (1H, d, J = 8.5 Hz), 6.83 (1H, s),

4.45 (1H, m), 3.49 (1H, d, J = 6.9 Hz), 3.39 (5H, m), 2.87 (2H, m), 1.00

(3H, d, J = 6.7 Hz), 0.80 (6H, m); ¹³C NMR (151 MHz, CDCl₃) δ 154.53,

149.24, 137.84, 137.23, 125.99, 121.68, 118.76, 109.27, 80.84, 66.88,

62.31, 55.36, 43.93, 42.17, 41.82, 41.64, 40.76, 39.92, 38.09, 31.57,

31.41, 30.31, 29.49, 28.80, 27.76, 26.33, 17.15, 16.40, 14.53, 14.23,

13.42. HRMS (ESI): Calcd for C₃₁H₄₆NO₄[M+H]⁺496.3427; Found:

496.3405.

ML13: White amorphous powder, yield 89%. ¹H NMR (400 MHz,

CDCl₃) δ 7.32 (2H, dd, J = 8.2 Hz, 5.5 Hz), 7.24 (1H, d, J = 8.6 Hz), 7.04

(2H, t, J = 8.5 Hz), 6.88 (1H, d, J = 8.6 Hz), 6.85 (1H, s), 5.29 (1H, t, J =

5.6 Hz), 4.46 (1H, m), 4.41 (2H, d, J = 5.9 Hz), 3.49 (1H, dd, J = 11.0

Hz, 3.1 Hz), 3.39 (1H, t, J = 10.9 Hz), 2.87 (2H, m), 1.00 (3H, d, J = 6.7

12

G.-X. Yang et al.

Bioorganic & Medicinal Chemistry 37 (2021) 116109

Hz), 0.80 (6H, m); ¹³C NMR (151 MHz, CDCl₃) δ 162.28 (d, J = 245.7

Hz), 154.98, 148.66, 138.05, 137.72, 133.95 (d, J = 3.6 Hz), 129.44 (d,

J = 8.1 Hz, 2C), 126.12, 121.53, 118.58, 115.58 (d, J = 21.8 Hz, 2C),

109.29, 80.82, 66.88, 62.29, 55.34, 44.60, 43.93, 41.64, 40.75, 39.90,

38.03, 31.56, 31.40, 30.30, 29.50, 28.80, 27.71, 26.30, 17.14, 16.39,

14.52. HRMS (ESI): Calcd for C₃₄H₄₃FNO₄[M+H]⁺548.3176; Found:

548.3170.

(2H, m), 1.00 (3H, d, J = 6.7 Hz), 0.82 (3H, s), 0.80 (3H, d, J = 6.3 Hz);

¹³C NMR (151 MHz, CDCl₃) δ 151.91, 148.26, 138.18, 138.03, 137.51,

129.13 (2C), 128.63, 126.22, 123.79, 121.61, 118.63 (2C), 109.32,

80.84, 66.89, 62.29, 55.34, 43.94, 41.65, 40.76, 39.90, 38.02, 31.56,

31.41, 30.31, 29.51, 28.80, 27.70, 26.31, 17.15, 16.40, 14.53. HRMS

(ESI): Calcd for C₃₃H₄₂NO₄[M+H]⁺516.3114; Found: 516.3091.

ML20: White amorphous powder, yield 87%. ¹H NMR (400 MHz,

CDCl₃) δ 7.40 (2H, dd, J = 8.1 Hz, 4.3 Hz), 7.28 (1H, d, J = 8.6 Hz), 7.03

(2H, t, J = 8.6 Hz), 6.93 (1H, dd, J = 8.5 Hz, 2.4 Hz), 6.89 (1H, s), 4.46

(1H, dd, J = 13.7 Hz, 7.7 Hz), 3.49 (1H, m), 3.40 (1H, t, J = 10.9 Hz),

2.88 (2H, m), 1.00 (3H, d, J = 6.7 Hz), 0.82 (3H, s), 0.80 (3H, d, J = 6.3

Hz). ¹³C NMR (151 MHz, CDCl₃) δ 152.07, 148.20, 138.24, 138.13,

133.48, 126.25, 121.55, 120.39, 118.57, 115.87 (2C), 115.72 (2C),

109.32, 80.83, 66.89, 62.29, 55.34, 43.94, 41.65, 40.76, 39.89, 38.02,

31.56, 31.41, 30.31, 29.52, 28.80, 27.70, 26.31, 17.14, 16.40, 14.53.

HRMS (ESI): Calcd for C₃₃H₄₁FNO₄[M+H]⁺534.3020; Found:

534.2998.

ML14: White amorphous powder, yield 88%. ¹H NMR (400 MHz,

CDCl₃) δ 7.26 (3H, m), 6.89 (3H, m), 6.85 (1H, s), 5.21 (1H, s), 4.45 (1H,

m), 4.38 (2H, d, J = 5.8 Hz), 3.81 (3H, s), 3.49 (1H, dd, J = 10.0 Hz, 3.2

Hz), 3.39 (1H, t, J = 10.8 Hz), 2.87 (2H, m), 1.00 (3H, d, J = 6.7 Hz),

0.80 (6H, m); ¹³C NMR (151 MHz, CDCl₃) δ 159.15, 154.92, 148.74,

137.99, 137.59, 130.23, 129.14 (2C), 126.08, 121.57, 118.62, 114.11

(2C), 109.29, 80.83, 66.88, 62.30, 55.34, 55.31, 44.80, 43.93, 41.64,

40.76, 39.90, 38.04, 31.56, 31.41, 30.31, 29.50, 28.80, 27.72, 26.31,

17.14, 16.39, 14.52. HRMS (ESI): Calcd for C₃₅H₄₆NO₅[M+H]⁺

560.3376; Found: 560.3361.

ML15: White amorphous powder, yield 90%. ¹H NMR (400 MHz,

CDCl₃) δ 7.62 (2H, d, J = 7.9), 7.47 (2H, d, J = 8.0), 7.24 (1H, d, J = 8.6

Hz), 6.89 (1H, d, J = 8.4 Hz), 6.86 (1H, s), 5.41 (1H, t, J = 6.1 Hz), 4.47

(3H, m), 3.49 (1H, dd, J = 10.1 Hz, 3.2 Hz), 3.39 (1H, t, J = 10.9 Hz),

2.87 (2H, m), 1.00 (3H, d, J = 6.7 Hz), 0.80 (6H, m); ¹³C NMR (151 MHz,

CDCl₃) δ 155.11, 148.60, 142.25, 138.11, 137.83, 129.92(q, J = 32.9

Hz), 127.85 (2C), 126.16, 125.70 (2C), 125.67, 121.49, 118.54, 109.30,

80.83, 66.89, 62.30, 55.34, 44.77, 43.93, 41.65, 40.76, 39.89, 38.03,

31.56, 31.41, 30.31, 29.50, 28.81, 27.70, 26.31, 17.14, 16.39, 14.52.

HRMS (ESI): Calcd for C₃₅H₄₃F₃NO₄[M+H]⁺598.3144; Found:

598.3144.

ML21: White amorphous powder, yield 83%. ¹H NMR (400 MHz,

CDCl₃) δ 7.32 (2H, d, J = 7.9 Hz), 7.29 (1H, d, J = 8.5 Hz), 7.13 (2H, d, J

= 8.3 Hz), 6.93 (1H, dd, J = 8.4 Hz, 2.4 Hz), 6.90 (1H, d, J = 2.3 Hz),

6.83 (1H, s), 4.46 (1H, m), 3.49 (1H, dd, J = 10.7 Hz, 5.4 Hz), 3.40 (1H,

t, J = 10.9 Hz), 2.88 (2H, m), 2.32 (3H, s), 1.00 (3H, d, J = 6.7 Hz), 0.82

(3H, s), 0.80 (3H, d, J = 6.4 Hz). ¹³C NMR (151 MHz, CDCl₃) δ 151.95,

148.32, 138.15, 137.96, 134.90, 133.36, 129.63 (2C), 126.20, 121.62,

118.72 (2C), 118.65, 109.31, 80.84, 66.89, 62.30, 55.35, 43.95, 41.65,

40.77, 39.90, 38.03, 31.57, 31.41, 30.31, 29.52, 28.81, 27.71, 26.32,

20.78, 17.15, 16.40, 14.53. HRMS (ESI): Calcd for C₃₄H₄₄NO₄[M+H]⁺

530.3270; Found: 530.3267.

ML16: White amorphous powder, yield 84%. ¹H NMR (400 MHz,

CDCl₃) δ 8.57 (2H, m), 7.72 (1H, d, J = 7.8 Hz), 7.30 (1H, dd, J = 7.7 Hz,

5.0 Hz), 7.24 (1H, d, J = 8.6 Hz), 6.88 (1H, d, J = 8.5 Hz), 6.85 (1H, s),

5.44 (1H, t, J = 5.9 Hz), 4.47 (3H,m), 3.49 (1H, dd, J = 10.9 Hz, 4.0 Hz),

3.39 (1H, t, J = 10.9 Hz), 2.86 (2H, m), 1.00 (3H, d, J = 6.7 Hz), 0.80

(6H, m); ¹³C NMR (151 MHz, CDCl₃) δ 155.17, 149.07, 148.97, 148.59,

138.07, 137.79, 135.66, 133.91, 126.14, 123.63, 121.50, 118.55,

109.29, 80.81, 66.87, 62.28, 55.32, 43.91, 42.76, 41.63, 40.74, 39.88,

38.01, 31.54, 31.39, 30.29, 29.48, 28.79, 27.69, 26.29, 17.14, 16.38,

14.52. HRMS (ESI): Calcd for C₃₃H₄₃N₂O₄[M+H]⁺531.3223; Found:

531.3208.

ML22: White amorphous powder, yield 81%. H NMR (400 MHz,

1

CDCl₃) δ 7.86 (1H, m), 7.25 (4H, m), 7.11 (1H, t, J = 7.5 Hz), 6.95 (1H,

m), 6.91 (1H, s), 6.74 (1H, s), 4.46 (1H, m), 3.49 (1H, dd, J = 10.5 Hz,

4.2 Hz), 3.40 (1H, t, J = 10.9 Hz), 2.87 (2H, m), 2.67 (2H, q, J = 7.5 Hz),

1.00 (3H, d, J = 6.7 Hz), 0.82 (3H, s), 0.80 (3H, d, J = 6.3 Hz). ¹³C NMR

(151 MHz, CDCl₃) δ 152.37, 148.38, 138.16, 137.97, 134.79, 133.24,

128.55, 126.88, 126.21, 124.71, 121.58, 121.29, 118.60, 109.30,

80.84, 66.89, 62.31, 55.36, 43.95, 41.65, 40.77, 39.91, 38.03, 31.58,

31.42, 30.32, 29.52, 28.81, 27.72, 26.31, 24.21, 17.15, 16.41, 14.53,

13.96. HRMS (ESI): Calcd for C₃₅H₄₆NO₄[M+H]⁺544.3427; Found:

544.3416.

ML17: White amorphous powder, yield 87%. ¹H NMR (400 MHz,

CDCl₃) δ 8.59 (2H, d, J = 5.1 Hz), 7.28 (3H, m), 7.25 (1H, d, J = 8.6 Hz),

6.90 (1H, dd, J = 8.3 Hz, 2.0 Hz), 6.86 (1H, s), 5.48 (1H, t, J = 6.1 Hz),

4.46 (3H, m), 3.49 (1H, dd, J = 10.0 Hz, 3.1 Hz), 3.39 (1H, t, J = 10.9

Hz), 2.88 (2H, m), 1.00 (3H, d, J = 6.7 Hz), 0.81 (3H, s), 0.80 (3H, d, J =

6.5 Hz); ¹³C NMR (151 MHz, CDCl₃) δ 155.21, 149.92 (2C), 148.56,

147.44, 138.13, 137.89, 126.18, 122.23 (2C), 121.45, 118.50, 109.30,

80.82, 66.88, 62.29, 55.33, 44.07, 43.92, 41.64, 40.75, 39.88, 38.02,

31.55, 31.40, 30.30, 29.50, 28.80, 27.70, 26.30, 17.14, 16.39, 14.52.

ML23: White amorphous powder, yield 85%. ¹H NMR (400 MHz,

CDCl₃) δ 7.36 (2H, d, J = 8.7 Hz), 7.28 (1H, s), 6.93 (1H, dd, J = 8.4 Hz,

2.4 Hz), 6.88 (3H, m), 6.78 (1H, s), 4.46 (1H, m), 3.49 (1H, dd, J = 11.0

Hz, 5.6 Hz), 3.39 (1H, t, J = 10.9 Hz), 2.89 (2H, m), 1.00 (3H, d, J = 6.7

Hz), 0.82 (3H, s), 0.80 (3H, d, J = 6.3 Hz). ¹³C NMR (151 MHz, CDCl₃) δ

156.15, 152.19, 148.37, 138.14, 137.93, 130.57, 126.19, 121.61,

120.54, 118.64, 114.33, 109.31, 80.84, 66.89, 62.30, 55.50, 55.35,

43.95, 41.65, 40.77, 39.90, 38.03, 31.57, 31.41, 30.31, 29.71, 29.52,

28.81, 27.71, 26.31, 17.15, 16.40, 14.53. HRMS (ESI): Calcd for

HRMS (ESI): Calcd for

531.3208.

C

₃₃H₄₃N₂O₄[M+H]⁺531.3223; Found:

C

₃₄H₄₄NO₅[M+H]⁺546.3219; Found: 546.3217.

ML18: White amorphous powder, yield 86%. ¹H NMR (400 MHz,

CDCl₃) δ 7.25 (1H, d, J = 7.6 Hz), 7.03 (1H, d, J = 2.7 Hz), 6.98 (1H,

dd, J = 5.0 Hz, 3.5 Hz), 6.89 (1H, dd, J = 8.4 Hz, 2.2 Hz), 6.85 (1H, s),

5.33 (1H, t, J = 5.7 Hz), 4.62 (2H, d, J = 5.8 Hz), 4.45 (1H, m), 3.49

(1H, dd, J = 10.1 Hz, 3.2 Hz), 3.39 (1H, t, J = 10.9 Hz), 2.88 (2H, m),

4.6. Biochemical studies

4.6.1. Oxygen-glucose deprivation-reperfusion (OGD/R)

The cells were pretreated with ML5 for 2 h and then exposed to OGD

medium (glucose-free Dulbecco’s Modiﬁed Eagle medium), then placed

in an anaerobic chamber containing 92% N₂, 3% O₂and 5% CO₂for 4 h

1.00 (3H, d, J = 6.8 Hz), 0.81 (3H, s), 0.80 (3H, d, J = 6.5 Hz); ¹³

C

NMR (151 MHz, CDCl₃) δ 154.65, 148.65, 140.71, 138.05, 137.72,

126.98, 126.14, 126.12, 125.36, 121.54, 118.59, 109.30, 80.84,

66.89, 62.30, 55.35, 43.94, 41.65, 40.76, 40.05, 39.90, 38.04,

31.57, 31.41, 30.31, 29.50, 28.81, 27.72, 26.31, 17.15, 16.40, 14.53.

HRMS (ESI): Calcd for C₃₂H₄₂NO₄S [M+H]⁺536.2835; Found:

536.2833.

◦

at 37 C. OGD was terminated by removing glucose-free medium and

exposed to high-glucose medium and cell cultures were reintroduced to

the regular CO₂incubator at 37 ^◦C for 2 h. Control cells were incubated

in the high-glucose medium at 37 ^◦C in a regular 74% N₂, 21% O₂and

5% CO₂incubator.

ML19: White amorphous powder, yield 91%. ¹H NMR (400 MHz,

CDCl₃) δ 7.44 (2H, d, J = 7.7 Hz), 7.34 (2H, t, J = 7.9 Hz), 7.28 (1H, d, J

= 8.5 Hz), 6.94 (1H, dd, J = 8.4 Hz, 2.4 Hz), 6.90 (1H, d, J = 2.2 Hz),

4.46 (1H, m), 3.49 (1H, d, J = 8.0 Hz), 3.40 (1H, t, J = 10.9 Hz), 2.89

4.6.2. Cholinesterases (ChEs) activity assay

Male ICR mice (5–6 weeks old, about 18–22 g) were purchased from

Shanghai Sippr-bk Laboratory Animal co, Ltd (Shanghai, China). In the

13

G.-X. Yang et al.

Bioorganic & Medicinal Chemistry 37 (2021) 116109

ChEs activity inhibition test, the animals are grouped as follows:

cat#P0010). All the experiments were repeated three times

independently.

Group

Drugs and Doses

Number of mice

Control group

Normal saline

6

4.6.5. ML5 activated Nrf2 pathway in SY5Y cells

Rivastigmine group

ML5-15 min group

ML5-30 min group

ML5-45 min group

ML5-3 h group

Rivastigmine, 2 mg/kg

ML5, 15 mg/kg

Cytoplasmic and nuclear protein was extracted using the Nuclear and

Cytoplasmic Protein Extraction Kit (Beyotime, Jiangsu, China;

cat#P0028), respectively. The relative expression of cytoplasmic and

nuclear Nrf2 protein were detected by western blot analysis.

ML5-12 h group

4.6.6. Animal experiment

Male SD rats (6–8 weeks old, about 200–220 g) were purchased from

Shanghai Sippr-bk Laboratory Animal Co, Ltd (Shanghai, China). They

were housed in environmentally controlled conditions (temperature of

25 ± 2 ^◦C, relative humidity of 60 ± 5%, and 12/12 h light/dark cycle).

In the BCCAO experiment, the rats were divided into control group

(sham group), model group (BCCAO group), low-dose ML5 group (3

mg/kg), high-dose ML5 group (15 mg/kg), and donepezil group (1 mg/

kg) (n = 10, 13, 8, 12, 10 respectively).

The mice cortex was homogenized in cold 75 mM sodium phosphate

buffer (pH 7.4). The assay solution consisted of 50 L of 0.1 mM phos-

phate buffer (pH 7.4), 100 L of deionized water, 20 L of mice cortex

homogenate, and 30 L of 2 mM acetylthiocholine iodide (Sigma;

cat#01480) as the substrate of the ChEs enzymatic reaction, respec-

tively. After reaction at 37 ^◦C for 8 min, 50

L 3% SDS was added to stop

the reaction. Finally, after adding 50

L of 0.2% 5,5^′-dithiobis (2-

μ

nitrobenzoic acid) (DTNB, Sigma; cat#D8130), the ChEs assay was

measured with a microplate reader (Biotek Instruments, Inc) at 440 nm

and at least three independent runs were performed for each sample.

The compound was pre-administered for 2 weeks. Before surgery, the

animals fasted for 12 h. An intraperitoneal injection of 10% chloral

hydrate (0.1 mL/100 g body weight) was used as anesthesia. Then, a

midline incision of the neck was followed by blunt muscle dissection and

the common carotid arteries were exposed. Then the para-arterial vagus

nerve was separated with a glass minute needle. The same procedure

was performed in the sham operation group rats without carotid artery

occlusion.

4.6.3. ML5 prevented the loss of cell viability and reduced ROS levels in

SH-SY5Y cells exposed to H₂O₂

Human dopaminergic neuroblastoma SH-SY5Y cells (American Type

Culture Collection, Virginia, USA) were cultured in Dulbecco’s modiﬁed

Eagle’s medium (DMEM)/F-12 HAM nutrient medium (1:1 mixture)

supplemented with 10% fetal bovine serum (FBS), penicillin (1000

units/mL), cultured in a 5% CO₂three-gas incubator (Thermo Scientiﬁc)

at 37 ^◦C.

The Morris water maze (MWM) test was performed at 28 days after

the surgery. The maze was made of a circular tub (height 50 cm,

diameter 180 cm) ﬁlled to a depth of 30 cm with water maintained at 25

± 1 ^◦C. Day 1: visual platform trial; Day 2–5: during the hidden platform

trial, the rats were trained twice a day for 3 days. The training period

was ﬁxed every day. If the rats did not ﬁnd the platform within 1 min,

they were guided to ﬁnd the platform and stayed on the platform for

10–15 s. Day 6: during the probe test, remove the platform and record

the movement track of the rats and the escape latency. In the course of

the experiment, the video acquisition system of AniLab Software Tech-

nology Co, Ltd. was used to record the mouse motion trajectory, and the

AVIAS (Animal Video Iracking Analysis Systcm) software was used for

trajectory analysis.

The viability of SH-SY5Y cells was analyzed by using the 3-(4,5-

dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT, Sigma;

cat#M2128) assay. SH-SY5Y cells growing to the fusion state were laid

on the 96-well plate at a concentration of 5*10⁴/mL. In order to test the

ability of ML5 in preventing the oxidative damage triggered by H₂O₂, a

pretreatment with ML5 (dissolved in DMSO) at 0.01–100

μM for 2 h and

200 M H₂O₂for another 24 h was performed in SH-SY5Y cells. Then 10

μ

L MTT (5 mg/ml) was added to the cells. The cells incubated for

another 4 h at 37 ^◦C. Subsequently, the medium was replaced with

DMSO in each well. Finally, the absorbance was measured at 490 nm by

a microplate reader (BioTek Instruments, Inc). The experiment was

repeated three-ﬁve times. Graph Pad Prism and SPSS software was used

for statistical analysis and all the results were expressed as mean ±

standard error (SEM).

Part of rat brain samples were collected after perfusion with ice-cold

ﬁxative solution (0.1 M phosphate buffer (PB), pH 7.3 and 4% para-

formaldehyde) for HE staining. First, after deparafﬁnization and rehy-

dration, 4 μm parafﬁn sections were rinsed in distilled water and then

stained with hematoxylin solution (Beyotime Biotechnology;

cat#C0105S) for 8 min. Next, the sections were stained with eosin so-

lution (Beyotime Biotechnology; cat#C0105S) for 3 min, followed by

dehydration with graded alcohol and clearing in xylene. Finally, the

sections were mounted using a synthetic resin.

The intracellular production of ROS was assessed by the nonpolar

compound 2^′-7^′-dichlorodihy droﬂuorescein diacetate (DCFH-DA,

Sigma; cat#D6883). The DCFH-DA was diluted to 10 μM with medium.

The cells were pretreatment with ML5 for 2 h and loaded with 10 μM

DCFH-DA. After this, washed cells three times with PBS to remove excess

DCFH-DA and changed to fresh medium. Then the cells were stimulated

4.6.7. Western blot

with 200

μm H₂O₂for 30 min. The conversion of DCFH-DA to the

The animals were euthanized after BCCAO surgery, and part of brain

samples were collected as fresh tissue for western blot. Similarly, after

treatment, the SH-SY5Y cells were lysed with a lysis buffer and centri-

fuged, then supernatant was collected as total extracts. To prepare the

cytoplasmic and nuclear fraction, cells were lysed using nuclear and

cytoplasmic protein extraction kit (Beyotime; cat#P0028) according to

the manufacturer’s instructions. The brain homogenates and cell lysates

were quantiﬁed with Bicinchoninic Acid protein assay kit (Beyotime;

ﬂuorescent product DCF was quantiﬁed using microplate reader (BioTek

Instruments, Inc) (Em/Ex = 535/485 nm) and qualitatively observed

using a ﬂuorescence microscope (Nikon, Tokyo, Japan). All the experi-

ments were repeated three times independently.

4.6.4. GSH and SOD assays

Cells or brain cortex were lysed. The total GSH was measured by a

commercially available Total Glutathione Assay kit (Nanjing Jiancheng

Bioengineering Institute, Nanjing, China; cat#A006-2-1). The SOD ac-

tivity was detected by using SOD Assay Kit (WST-8 method) obtained

from Beyotime (Beyotime Institute of Biotechnology, Nantong, China;

cat#S0101M). All protocols completely complied to the manufacturer’s

instructions. The absorbance was assessed at 450 nm using a microplate

reader. The absorbance was assessed at 405 nm (GSH) and 450 nm

(SOD) using a microplate reader. The protein concentration was esti-

mated by the Bicinchoninic Acid protein assay kit (Beyotime;

cat#P0010). The homogenates and lysates (30 μg protein) were frac-

tionated by 10% SDS-PAGE and transferred to polyvinylidene ﬂuoride

membrane (PVDF, Millipore; cat#IPVH00010). After transfer, mem-

branes were blocked in 5% skim milk and incubated overnight at 4 ^◦C

with primary antibody, and incubated with anti-rabbit or anti-mouse

HRP antibody (1 : 2000 dilution) for 2 h at room temperature. A rab-

bit polyclonal antibody (1: 500 in Tris-buffered saline with Tween

(TBST)) for Nrf2 (Proteintech; cat#16396-1-AP), a rabbit monoclonal

14

G.-X. Yang et al.

Bioorganic & Medicinal Chemistry 37 (2021) 116109

antibody for β-tubulin (Servicebio; cat#GB11017), a mouse monoclonal

antibody for Pcna (abcam; cat#ab29) was used to detect the corre-

sponding proteins. After peroxidase-conjugated goat anti-rabbit/anti-

mouse secondary antibody (Santa Cruze; cat#sc-2004, cat#sc-2005)

was used, ECL (Millipore) detection reagent was used for visualization

according to the manufacturer’s instructions. Densitometry analysis of

the bands was performed using Image J software.

Appendix A. Supplementary material

Supplementary data to this article can be found online at https://doi.

org/10.1016/j.bmc.2021.116109.

References

1 Elahi FM, Miller BL. A clinicopathological approach to the diagnosis of dementia.

Nature Rev Neurol. 2017;13:457–476. https://doi.org/10.1038/nrneurol.2017.96.

2 O’Brien JT, Thomas A. Vascular dementia. Lancet . 2015;386:1698–1706. https://doi.

org/10.1016/S0140-6736(15)00463-8.

4.6.8. Statistical analysis

All the results were expressed as mean ± standard error (SEM).

Image J program was used to analyze the density of western blot and

immunoﬂuorescence images. Graph Pad Prism 8.0 (GraphPad Software,

La Jolla CA, USA) was used for data processing and IBM SPSS Statistics

25 (SPSS Inc., Chicago, IL) was used for statistical analysis. The two-way

ANOVA was used to determine the statistical signiﬁcance (p < 0.05) of

the MWM data. A one-way ANOVA followed by Student’s t-test (non-

parametric Mann-Whitney and Wilcoxon tests) were used to determine

the statistical signiﬁcance (p < 0.05) of the other obtained data.

3 Wu Y, Zhou Q, Zhang T, et al. Discovery of potent, selective, and orally bioavailable

inhibitors against phosphodiesterase-9, a novel target for the treatment of vascular

dementia. J Med Chem. 2019;62:4218–4224. https://doi.org/10.1021/acs.

jmedchem.8b01041.

4 Bartus RT, Dean RL, Beer B, Lippa AS. The cholinergic hypothesis of geriatric

memory dysfunction. Science. 1982;217:408–414. https://doi.org/10.1126/

science.7046051.

5 Perry EK, Walker MP, Grace J, Perry R. Acetylcholine in mind: a neurotransmitter

correlate of consciousness? Trends Neurosci. 1999;22:273–280. https://doi.org/

10.1016/S0166-2236(98)01361-7.

6 Happich M, Kirson NY, Desai U, et al. Excess costs associated with possible

misdiagnosis of alzheimer’s disease among patients with vascular dementia in a Uk

Cprd population. Alzheimer’s Dementia. 2014;10:911–912. https://doi.org/10.1016/

j.jalz.2014.07.110.

4.6.9. Molecular docking studies

The crystal structures of hAChE (PDB ID: 6F25) and Keap1 (PDB ID:

3VNH) were downloaded from the Protein Data Bank (https://www.

7 Dembitsky VM, Dzhemileva L, Gloriozova T, D’yakonov V. Natural and synthetic

drugs used for the treatment of the dementia. Biochem Biophys Res Commun. 2020;

524:772–783. https://doi.org/10.1016/j.bbrc.2020.01.123.

¨

rcsb.org/) as receptors. Maestro 10.5 embedded in Schrodinger soft-

8 Emre M, Aarsland D, Albanese A, Byrne EJ, Deuschl G, De Deyn PP, et al.

Rivastigmine for dementia associated with Parkinson’s disease. N Engl J Med. 2004;

24:2509–2518. https://doi.org/10.1056/NEJMoa041470.

ware package (2016-1) was employed to perform all molecular

modeling studies. Protein structures of hAChE and Keap1 were reﬁned

by Protein Preparation Wizard. Protein energies were minimized with

Root Mean Square Deviation (RMSD) value of 0.3 Å using OPLS-2005

force ﬁeld. LigPred was used to prepare 3D conformations of ligands

(ML5 and rivastigmine). Epik generated the protonation states of ML5

and rivastigmine at pH values from 5.0 to 9.0. Docking sites were

centered by original ligands complexed with hAChE and Keap1 The grid

boxes of receptors were set to 20 Å to accommodate all ligand atoms.

Then the Glide extra precision (XP) model was utilized to dock all

reﬁned ligands to the active sites of receptors. Binding afﬁnities between

the most superior conformers of ligands and the receptors were calcu-

lated by Prime MM-GBSA under VSGB solution model using OPLS-2005

force ﬁeld. Interaction model ﬁgures were generated by PyMOL.

9 de la Torre JC. Vascular risk factor detection and control may prevent Alzheimer’s

disease. Ageing Res Rev. 2010;9:218–225. https://doi.org/10.1016/j.

arr.2010.04.002.

10 Lin MS, Chiu MJ, Wu YW, et al. Neurocognitive improvement after carotid artery

stenting in patients with chronic internal carotid artery occlusion and cerebral

ischemia. Stroke. 2011;42:2850–2854. https://doi.org/10.1161/

STROKEAHA.111.613133.

11 Peng S, Zhang B, Meng X, Yao J, Fang J. Synthesis of piperlongumine analogues and

discovery of nuclear factor erythroid 2-related factor 2 (Nrf2) activators as potential

neuroprotective agents. J Med Chem. 2015;58:5242–5255. https://doi.org/10.1021/

acs.jmedchem.5b00410.

12 Fishbein JC, Holland R. Chemistry of the cysteine sensors in Keap1. Antioxid Redox

Signal. 2010;13:1749–1761. https://doi.org/10.1089/ars.2010.3273.

13 Uruno A, Furusawa Y, Yagishita Y, et al. The Keap1-Nrf2 system prevents onset of

diabetes mellitus. Mol Cell Biol. 2013;33:2996–3010. https://doi.org/10.1128/

MCB.00225-13.

14 Benchekroun M, Romero A, Egea J, et al. The antioxidant additive approach for

Alzheimer’s disease therapy: new ferulic (lipoic) acid plus melatonin modiﬁed

tacrines as cholinesterases inhibitors, direct antioxidants, and nuclear factor

(Erythroid-Derived 2)-Like 2 activators. J Med Chem. 2016;59:9967–9973. https://

doi.org/10.1021/acs.jmedchem.6b01178.

4.6.10. Acute toxicity study.

All animal experiments were carried out in accordance with the

National Institutes of Health Guide for the Care and Use of Laboratory

Animals of China for animal experimentation, and were approved by the

Animal Ethics Committee of the East China University of Science and

Technology. A total of 14 healthy male mice (20–30 g) were randomly

divided into 2 groups for acute toxicity study of the compound ML5.

These animals were left without food for 12 h before the experiment.

Compound ML5 was suspended in normal saline and given orally to

divided into 2 groups, including 0 and 500 mg/kg, n = 7 per group. After

the administration of the test compound, the mice were observed closely

for the ﬁrst 4 h for the behavior and mortality.

15 Zhang M, An C, Gao Y, Leak RK, Chen J, Zhang F. Emerging roles of Nrf2 and phase II

antioxidant enzymes in neuroprotection. Prog Neurobiol. 2013;100:30–47. https://

doi.org/10.1016/j.pneurobio.2012.09.003.

16 Bolognesi ML, Cavalli A, Valgimigli L, et al. Multi-target-directed drug design

strategy: from a dual binding site acetylcholinesterase inhibitor to a trifunctional

compound against Alzheimer’s disease. J Med Chem. 2007;50:6446–6449. https://

doi.org/10.1021/jm701225u.

17 Cavalli A, Bolognesi ML, Minarini A, et al. Multi-target-directed ligands to combat

neurodegenerative diseases. J Med Chem. 2008;51:347–372. https://doi.org/

10.1021/jm7009364.

18 Yang GX, Huang Y, Zheng LL, et al. Design, synthesis and evaluation of diosgenin

carbamate derivatives as multitarget anti-Alzheimer’s disease agents. Eur J Med

Chem. 2019;187, 111913. https://doi.org/10.1016/j.ejmech.2019.111913.

19 Pan H, Van Khang P, Dong D, Wang R, Ma L. Synthesis and SAR study of novel

sarsasapogenin derivatives as potent neuroprotective agents and NO production

inhibitors. Bioorg Med Chem Lett. 2017;27:662–665. https://doi.org/10.1016/j.

bmcl.2016.11.070.

Then, each animal was weighed every day, and recorded the mor-

tality. After 2 weeks, the animals were sacriﬁced to observe the changes

in the heart, liver, and kidneys.

20 Yang GX, Ge SL, Wu Y, et al. Design, synthesis and biological evaluation of 3-

piperazinecarboxylate sarsasapogenin derivatives as potential multifunctional anti-

Alzheimer agents. Eur J Med Chem. 2018;156:206–215. https://doi.org/10.1016/j.

ejmech.2018.04.054.

Declaration of Competing Interest

The authors declare that they have no known competing ﬁnancial

interests or personal relationships that could have appeared to inﬂuence

the work reported in this paper.

21 Kinarivala N, Patel R, Boustany RM, Al-Ahmad A, Trippier PC. Discovery of aromatic

carbamates that confer neuroprotective activity by enhancing autophagy and

inducing the anti-apoptotic Protein B-cell lymphoma 2 (Bcl-2). J Med Chem. 2017;60:

9739–9756. https://doi.org/10.1021/acs.jmedchem.7b01199.

Acknowledgments

22 Poprac P, Jomova K, Simunkova M, Kollar V, Rhodes CJ, Valko M. Targeting free

radicals in oxidative stress-related human diseases. Trends Pharmacol Sci. 2017;38:

592–607. https://doi.org/10.1016/j.tips.2017.04.005.

This work was supported by the National Natural Science Foundation

of China (Grants 81973196 and 81673318), the 111 Project (Grant

BP0719034) and the Natural Science Foundation of Shanghai (Grant

19ZR1413900).

23 Kobayashi M, Yamamoto M. Nrf2–Keap1 regulation of cellular defense mechanisms

against electrophiles and reactive oxygen species. Adv Enzyme Regul. 2006;46:

113–140. https://doi.org/10.1016/j.advenzreg.2006.01.007.

15

G.-X. Yang et al.

Bioorganic & Medicinal Chemistry 37 (2021) 116109

24 Du SQ, Wang XR, Xiao LY, et al. Molecular mechanisms of vascular dementia: what

can be learned from animal models of chronic cerebral hypoperfusion? Mol

Neurobiol. 2017;54:3670–3682. https://doi.org/10.1007/s12035-016-9915-1.

25 Leon-Moreno LC, Castaneda-Arellano R, Rivas-Carrillo JD, Duenas-Jimenez SH.

Challenges and improvements of developing an ischemia mouse model through

bilateral common carotid artery occlusion. J Stroke Cerebrovasc Dis. 2020;29,

104773. https://doi.org/10.1016/j.jstrokecerebrovasdis.2020.104773.

by docking calculations and molecular dynamics simulations. Eur J Med Chem. 2014;

81:158–175. https://doi.org/10.1016/j.ejmech.2014.05.008.

28 Zueva I, Dias J, Lushchekina S, et al. New evidence for dual binding site inhibitors of

acetylcholinesterase as improved drugs for treatment of Alzheimer’s disease.

Neuropharmacology. 2019;155:131–141. https://doi.org/10.1016/j.

neuropharm.2019.05.025.

29 Lo SC, Li X, Henzl M, Beamer LJ, Hannink M. Structure of the Keap1:Nrf2 interface

provides mechanistic insight into Nrf2 signaling. EMBO J. 2006;25:3605–3617.

https://doi.org/10.1038/sj.emboj.7601243.

26 Kensler TW, Wakabayashi N, Biswal S. Cell survival responses to environmental

stresses via the Keap1-Nrf2-ARE pathway. Annu Rev Pharmacol Toxicol. 2007;47:

89–116. https://doi.org/10.1146/annurev.pharmtox.46.120604.141046.

27 Vitorovic-Todorovic MD, Koukoulitsa C, Juranic IO, Mandic LM, Drakulic BJ.

Structural modiﬁcations of 4-aryl-4-oxo-2-aminylbutanamides and their acetyl- and

butyrylcholinesterase inhibitory activity. Investigation of AChE-ligand interactions

30 Heightman TD, Callahan JF, Chiarparin E, et al. Structure-activity and structure-

conformation relationships of aryl propionic acid inhibitors of the Kelch-like ECH-

Associated Protein 1/Nuclear Factor Erythroid 2-Related Factor 2 (KEAP1/NRF2)

protein-protein interaction. J Med Chem. 2019;62:4683–4702. https://doi.org/

10.1021/acs.jmedchem.9b00279.

16

Article Doi

Twin drug design, synthesis and evaluation of diosgenin derivatives as multitargeted agents for the treatment of vascular dementia

DOI: 10.1016/j.bmc.2021.116109

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.bmc.2021.116109

Full text of DOI:10.1016/j.bmc.2021.116109

Products guided by the article

R&D Labs maybe for 2137-22-6

Relevant to this article

Hot Product