The insight of in vitro and in silico studies on cholinesterase inhibitors from the roots of Cimicifuga dahurica (Turcz.) Maxim.

Abstract

Cholinesterases (ChEs) are enzymes that break down neurotransmitters associated with cognitive function and memory. We isolated cinnamic acids (1 and 2), indolinones (3 and 4), and cycloartane triterpenoid derivatives (5–19) from the roots of Cimicifuga dahurica (Turcz.) Maxim. by chromatography. These compounds were evaluated for their inhibitory activity toward ChEs. Compound 1 was determined to have an IC50 value of 16.7 ± 1.9 μM, and to act as a competitive inhibitor of acetylcholinesterase (AChE). Compounds 3, 4 and 14 were found to be noncompetitive with IC50 values of 13.8 ± 1.5 and 6.5 ± 2.5 μM, and competitive with an IC50 value of 22.6 ± 0.4 μM, respectively, against butyrylcholinesterase (BuChE). Our molecular simulation suggested each key amino acid, Tyr337 of AChE and Asn228 of BuChE, which were corresponded with potential inhibitors 1, and 3 and 4, respectively. Compounds 1 and 4 were revealed to be promising compounds for inhibition of AChEs and BuChEs, respectively.

Full text

RESEARCH PAPER
The insight of in vitro and in silico studies on cholinesterase inhibitors from the
roots of Cimicifuga dahurica (Turcz.) Maxim.
Jang Hoon Kim
a
, Nguyen Phuong Thao
b
, Yoo Kyong Han
c
, Young Suk Lee
c
, Bui Thi Thuy Luyen
d
,
Ha Van Oanh
d
, Young Ho Kim
c
and Seo Young Yang
c
a
Radiation Breeding Research Center, Advanced Radiation Technology Institute, Korea Atomic Energy Research Institute, Jeoungeup, Jeollabuk-
do, Republic of Korea;
b
Institute of Marine Biochemistry (IMBC), Vietnam Academy of Science and Technology (VAST), Hanoi, Vietnam;
c
College
of Pharmacy, Chungnam National University, Daejeon, Republic of Korea;
d
Department of Pharmaceutical Industry, Hanoi University of
Pharmacy, Hanoi, Vietnam
ABSTRACT
Cholinesterases (ChEs) are enzymes that break down neurotransmitters associated with cognitive function
and memory. We isolated cinnamic acids (1and 2), indolinones (3and 4), and cycloartane triterpenoid
derivatives (5–19) from the roots of Cimicifuga dahurica (Turcz.) Maxim. by chromatography. These com-
pounds were evaluated for their inhibitory activity toward ChEs. Compound 1was determined to have an
IC
50
value of 16.7 ± 1.9 lM, and to act as a competitive inhibitor of acetylcholinesterase (AChE).
Compounds 3,4and 14 were found to be noncompetitive with IC
50
values of 13.8 ± 1.5 and 6.5 ± 2.5 lM,
and competitive with an IC
50
value of 22.6 ± 0.4 lM, respectively, against butyrylcholinesterase (BuChE).
Our molecular simulation suggested each key amino acid, Tyr337 of AChE and Asn228 of BuChE, which
were corresponded with potential inhibitors 1, and 3and 4, respectively. Compounds 1and 4were
revealed to be promising compounds for inhibition of AChEs and BuChEs, respectively.
ARTICLE HISTORY
Received 10 April 2018
Revised 16 May 2018
Accepted 19 June 2018
KEYWORDS
Cimicifuga dahurica;
Ranunculaceae; cholinester-
eases inhibitor;
molecular simulation
Introduction
Cholinesterases (ChEs), which are enzymes that hydrolyse choline
esters, are classified as acetylcholinesterase (EC 3.1.1.7, AChE) and
butyrylcholinesterase (EC 3.1.1.8, BuChE)
1
. AChE is responsible for
the conversion of acetylcholine (ACh) into choline and acetic acid
in cholinergic synapses. AChE is formed as a tetramer of 70-kDa
monomeric subunits
1,2
. Its 3D structure was revealed by examin-
ing the enzyme of electric eels
3
. AChE has an active site with
a-helix and b-sheet structures and a catalytic triad of serine, histi-
dine, and glutamic acid
1,3
. BuChE, an enzyme that breaks down
artificial butyrylcholine, is known to hydrolyse ACh and other ester
derivatives in the body
4,5
. BuChE, which is a tetrameric serine
esterase consisting of monomers of 90-kDa molecular mass,
showed over 65% structural similarity to AChE
4,6
. ACh is a neuro-
transmitter that is produced from the acetylation reaction of cho-
line and acetyl-CoA by choline acetyltransferase, and is distributed
in the central and peripheral nervous systems
7
. ACh plays a key
role in nerve-nerve communication by binding to ACh receptors
8
.
This molecule is associated with maintenance of cognitive func-
tion and memory
5,8
. In particular, Alzheimer’s disease (AD)
patients are characterised by a decline in ACh levels
8
. Two ChEs
have been regarded as target enzymes for treatment of AD
1,5,8
.
Cimicifuga dahurica (Turcz.) Maxim., in the family
Ranunculaceae, is commonly called “shengma”and is distributed
throughout Korea, Japan, China and Russia
9
. In China, the rhi-
zomes of C. dahurica have been used as a traditional medicine to
treat headaches and toothaches
10
. Phytochemical studies of this
plant indicated the presence of cycloartane triterpenoids and cin-
namic acid derivatives
9
. These compounds exhibit neuroprotective
activity and enhance cell viability by eliminating H
2
O
2
in PC12
cells
9,11
. Cycloartane triterpenoids have anti-tumour activities
including induction of apoptosis and G
2
/M cell cycle arrest in solid
tumours, blood tumours and drug-resistant tumours
10
.
These findings led us to search for products that block the
catalytic reaction of ChE. We isolated compounds 1–19 from the
roots of C. dahurica using open column chromatography (CC).
These compounds were tested for interactions with both AChE
and BuChE in vitro. Through molecular simulation, the inhibitor-
ChE complex structure was predicted visually using the Autodock
4.2 programme. The complex that was constructed considered the
interaction between the inhibitor and ChE in terms of molecular
dynamics (MDs).
Materials and methods
General experimental procedures
Optical rotations were measured using a JASCO P-2000 polarim-
eter (JASCO, Oklahoma, OK, USA). IR spectra were obtained on a
Bruker TENSOR 37 FT-IR spectrometer (Bruker, Billerica, MA, USA).
NMR spectra were recorded on JEOL JNM-AL 400 MHz and JEOL
ECA 600 MHz spectrometer (JEOL, Peabody, MA, USA), chemical
CONTACT Seo Young Yang syyang@cnu.ac.kr; Young Ho Kim yhk@cnu.ac.kr College of Pharmacy, Chungnam National University, Daejeon 34134,
Republic of Korea
Supplemental data for this article can be accessed here.
These authors contributed equally to this work.
ß2018 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly cited.
JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY
2018, VOL. 33, NO. 1, 1174–1180
https://doi.org/10.1080/14756366.2018.1491847

shift (d) are expressed in ppm with reference to the TMS signals.
Gas chromatography spectra were recorded on a Shimadzu-2010
spectrometer (Shimadzu, Kyoto, Japan), SPB-1 capillary (30
m0.25 mm and 30 m 0.32 mm); Mightysil RP-18 GP, Kanto
Chemical, 10 250 mm. The electrospray ionisation and the high-
resolution electrospray ionisation mass spectrometer were oper-
ated in the positive-ion mode, with sodium iodide being used for
mass calibration from an Agilent 6530 Accurate-Mass Q-TOF LC/
MS system (Micromass, Wythenshawe, UK). CC was conducted
using on 65–250 or 230–400 mesh silica gel (Sorbent
Technologies, Atlanta, GA, USA), porous polymer gel (Diaion
V
R
HP-
20, 20–60 mesh, Mitsubishi Chemical, Tokyo, Japan), Sephadex
TM
LH-20 (Supelco, Bellefonte, PA, USA), octadecyl silica (ODS, 50 lm,
Cosmosil 140 C
18
-OPN, Nacalai Tesque), and YMC RP-C
18
resins
(30–50 lm, Fuji Silysia Chemical). Analytical thin layer chromatog-
raphy (TLC) systems were performed on precoated silica gel 60
F
254
(1.05554.0001, Merck) and RP-18 F
254S
plates (1.15685.0001,
Merck) and compounds were visualised by spraying with 10%
H
2
SO
4
in water and then heating for 1.5–2 min. All procedures
were carried out with solvents purchased from commercial sour-
ces that were used without further purification.
Chemicals and reagents
AChE (C3389), acetylthiocholine iodide (A5751), BuChE (C1057),
butyrylthiocholine iodide (B3253) and 5,5-dithiobis(2-nitrobenzoic
acid) (DTNB) were purchased from Sigma-Aldrich (St Louis,
MO, USA).
Plant material
The roots (3.5-years old) of C. dahurica were purchased from a
herbal company, Naemome Dah, Ulsan, Korea, in February 2016.
This sample was identified by Prof. Y.H. Kim. A voucher specimen
(CNU-16003) representing this collection has been deposited at
the Herbarium of the College of Pharmacy, Chungnam National
University, Daejeon, Korea.
Extraction and isolation
The roots of C. dahurica (2.5 kg) were extracted three times with
5.0 L of 95% ethanol at 40 C. Concentrated ethanol extract
(65.3 g) was suspended in distilled water and progressively frac-
tioned with n-hexane (9.6 g), dichloromethane (15.2 g) and water
(40.5 g) fractions.
Dichloromethane fraction was subjected to silica gel CC by
using gradient solvent system of n-hexane and EtOAc (from 95:5
to 2.5:5) to achieve seven fractions (D1–D7). D3 fraction was chro-
matographed by Sephadex LH-20 CC with MeOH-H
2
O solvent
(95:5) to isolate compounds 7(10.2 mg) and 10 (8.9 mg).
Compounds 5(13.5 mg), 6(4.0 mg), 9(3.8 mg) and 13 (3.1 mg)
were purified by silica gel CC with an isocratic solvent system of
n-hexane and EtOAc (2:1) from D4 fraction. D5 fraction was chro-
matographed by RP-C
18
CC using mixture solvent system of acet-
one and H
2
O (1:4) to afford compounds 1(3 mg), 8(5.6 mg) and
12 (3.0 mg). D7 fraction was purified by over Sephadex LH-20 CC
with MeOH solvent to give three fractions (D7.1–D7.3). D7.1 frac-
tion was subjected to RP-C
18
CC by using solvent system of MeOH
and H
2
O (5.5:1) to obtain compounds 3(20 mg) and 4(5 mg).
Compounds 2(32.6 mg) and 11 (5 mg) were isolated by over silica
gel CC with solvent system of CH
2
Cl
2
and acetone (4:1) from D7
fraction. H
2
O fraction was subjected to a Diaion HP-20 CC by using
gradient solvent system of MeOH and H
2
O (from 25:75 to 100:0)
to give four fractions (H1–H4). H2 fraction was isolated as com-
pounds 14 (11.7 mg), 15 (18.2 mg), 18 (15.2 mg) and 19 (10.1 mg)
by RP-C
18
CC with gradient solvent system of MeOH and H
2
O
(from 35:65 to 100:0). H4 fraction was purified by over RP-C
18
CC
with solvent system of MeOH and H
2
O (35:65) to achieve com-
pounds 16 (10.3 mg) and 17 (3.5 mg).
ChE assay
AChE and BuChE inhibition assays were performed as described
by Othman et al.
12
with some modifications. Briefly, each 130 lL
of AChE (0.05 U/mL) and BuChE (0.05 U/mL) in 50 mM phos-
phate buffer (pH 7.4) was added to 96-well plates containing
20 lL of MeOH or sample dissolved in MeOH. 25 lL acetylthiocho-
line iodide (5 mM) or butyrylthiocholine iodide (5 mM), and 25 lL
DTNB (1 mM) were added into the mixture in order. After initiating
ChE reaction at 37 C, the products were scanned at 475 nm UV-
Vis photometer for 20 min. The inhibition activity was calculated
using the following equation:
Inhibitory activity ð%Þ¼½ðDcontrolDsampleÞ=Dcontrol100:
Where control and sample were the intensity of control and
inhibitor after 20 min, respectively.
The ChE inhibitory activity of each sample was expressed as
IC
50
(mM required to inhibit the hydrolysis of the ChE substrates
by 50%) determined from the log-dose inhibition curve.
Molecular docking of inhibitor with ChE
Molecular docking was performed as previously described using
the Autodock 4.2 programme (La Jolla, CA, USA)
12
. Single bond of
ligand was flexibly assigned by using torsion tree of
Autodocktools. Each pdb files of AChE (pdb ID: 1C2B) and BuChE
(pdb ID: 4BDS) were downloaded from RCSB protein data bank.
Achieved protein was added in hydrogens, and then this was
assigned with compute gasteiger charges. For the docking, the
grid containing activity site or all protein was set. Ligand was
docked into that with default values of genetic algorithm parame-
ters (number of GA runs: 50, maximun number of evals:
25,000,000). The result was presented with Ligplot (Cambridge,
UK) and Chimaera (San Francisco, CA, USA).
Molecular simulation of inhibitor with ChE
MDs were performed to simulate the complex of ligand with pro-
tein by the Gromacs version 4.6.5 package. Itp and gro files of lig-
and were built at Prodrg server. Gro and topology files of ChE
were generated by pdbgmx utility. These were edited to add lig-
and files. The ligand with protein was dissolved in water mole-
cules of a cubic box with a size of 12 12 12 containing six
sodium ions (1.0 Å distance). Moreover, then this complex was
minimised until it reach the maximal force of 10 kJ/mol. Each NVT
and NPT was simulated at 300 K temperature and 1 bar pressure
for 100 ps in the order, respectively. Lastly, equilibrated complex
was subjected to MD simulation for 10,000 ps.
Statistical analysis
Statistical significance was determined using a one-way analysis of
variance and Students t-test (Systat Inc., Evanston, IL, USA). A p
JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY 1175

value <0.01 was considered significant. All results are presented
as the mean ± SEM.
Results and discussion
Isolation and identification
An ethanol extract of the roots of C. dahurica was progressively par-
titioned into n-hexane, dichloromethane, and water fractions. The
dichloromethane and water fractions were subjected to various CC
methods to obtain compounds 1–13 and 14–19, respectively. These
compounds were investigated based on spectroscopic data and
comparison with previous reports. The nineteen extracted com-
pounds were identified as cimiciphenone (1)
13
, ferulic acid methyl
ester (2)
13
,(E)-3-(30-methyl-20-butenylidene)-1-methyl-2-indolinone
(3)
13
,(E)-3-(30-methyl-20-butenylidene)-2-indolinone (4)
13
, 7,8-dide-
hydrocimigenol (5), 24-epi-24-O-acetyl-7,8-didehydroshengmanol
(6)
14
, 25-triepoxy-12b-acetoxy-3b,26-dihydroxy-9,19-cyclolanost-7-
ene (7)
15
, 25-O-acetyl-7,8-didehydrocimigenol (8)
16
, 25-anhydro-7,8-
didehydrocimigenol (9)
15
, 24-epi-7,8-didehydrocimigenol (10)
16
,
25-O-acetylcimigenol (11)
15
, 24-epi-24-O-acetyl-7,8-didehydrosheng-
manol (12)
14
, 25-anhydrocimigenol (13)
15
, 25-O-acetyl-7,8-didehy-
drocimigenol 3-O-b-D-xylopyranoside (14)
14
, 25-anhydrocimigenol-
3-O-b-D-xylopyranoside (15)
17
,24-epi-7,8-didehydrocimigenol
3-O-b-D-xylopyranoside (16)
14
,3-O-b-D-xylopyranosyl-24S,25-dihy-
droxy-15-oxo-acta-(16R,23R)-16,23-monoxoside (17)
18
, cimiricaside
A(18)
19
, and 7,8-didehydro-25-anhydrocimigenol-3-O-b-D-xylopyra-
noside (19)
17
(Figure 1).
ChE assay
To screen for the ability of the isolated compounds 1–19 to block
catalytic reaction of ChE, they were analysed in vitro at 100 lM con-
centration using a UV-spectrophotometer. As shown in Figure 2(A)
and Table 1, compounds 1–4and 6–8, and compounds 2–6,9and
14–18 exhibited over 50% inhibitory activity against AChE and
BuChE, respectively. To calculate their IC
50
values, these compounds
were subjected to an enzyme assay at a variety of concentrations.
They caused decreases in activities of the two ChEs, with gradual or
sharp slopes in activity curves in a dose-dependent manner (Figure
2(B,C),Table 1). These results showed that compounds 1–4and 6–
8had IC
50
values ranging from 16.7 ± 1.9 to 95.8 ± 5.1 lM against
AChE. In particular, compound 1had an IC
50
value of 16.7 ± 1.9 lM.
Compounds 2–6,9and 14–18 were revealed to have IC
50
values
ranging from 6.5 ± 2.5 to 90.9 ± 6.0 lM against BuChE. Compounds
3,4and 14 exhibited greater inhibitory activity against BuChE than
OO
O
H
H
OH
H
H
OH
O
OAc
HO
HO
18
O
HO
H
H
7
OAc
O
O
OH
OO
HO
H
H
OH
H
H
OH
5
OO
O
H
H
OH
H
H
OAc
O
OH
HO
HO
14
OO
HO
H
OH
H
H
9
H
OO
HO
H
H
OH
H
H
OAc
8
O
HO
H
OH
H
H
OH
6
OAc
OH
O
O
O
HO
HO
OH
OMe
1
N O
CH3
3
N O
H
4
MeO
HO
O
OMe
2
OO
O
H
H
OH
H
H
O
OH
HO
HO
15
OO
O
H
H
OH
H
H
OH
O
OH
HO
HO
16
O
O
H
OH
O
H
H
OH
O
OH
HO
HO
17
OO
HO
H
H
OH
H
H
OH
10
O
HO
H
OAc
OH
H
H
OH
OH
12
OO
HO
H
OH
H
H
13
H
OO
O
H
H
OH
H
H
O
OH
HO
HO
19
OH
O
O
HH
H
H
HO
OAc
H
11
Figure 1. Structures of isolated compounds 1–19 from C. dahurica.
1176 J. H. KIM ET AL.

the other compounds tested, with IC
50
values of 13.8 ± 1.2, 6.5 ± 2.5
and 22.6 ± 0.4 lM.
Enzyme kinetics on AChE and BuChE
As indicated in Figure 2(D), compound 1was competitive inhibitor
which observed to have same V
max
value, and different K
m
values
at 6.2–50 lM concentration on AChE. Compounds 3and 4were
confirmed as noncompetitive mode due to various V
max
values
and a K
m
value according to respective concentrations on BuChE
(Figure 2(E,F)). Whereas, compound 14 was revealed to take the
binding into activity site by competing with substrate (Figure
2(G)). Additionally, these results were calculated with K
i
values of
the potential inhibitors using secondary replot. Compound 1was
calculated to be 16.2 ± 0.9 lM on AChE. Compounds 3,4and 14
(D)
(G) (H)
(E) (F)
(A) (B) (C)
Figure 2. Inhibitory activity of compounds 1–19 at 100 lM on AChE and BuChE (A). IC
50
values of them on AChE (B) and BuChE (C). Lineweaver-Burk plots of com-
pound 1on AChE (D) and of compounds 3,4,and14 on BuChE (E-G). Secondary plot of compounds 1,3,4and 14 (H).
Table 1. Inhibitory activity of compounds 1–19 on two ChEs.
Compounds
AChE BuChE
100 lM (%) IC
50
(lM) 100lM (%) IC
50
(lM)
1 65.4 ± 0.9 16.7 ± 1.9 13.6 ± 0.7 N.T
2 58.0 ± 0.6 52.4 ± 3.7 62.3 ± 0.3 37.1 ± 1.5
3 53.8 ± 0.1 95.8 ± 5.1 88.7 ± 1.3 13.8 ± 1.2
4 61.0 ± 4.6 48.0 ± 6.8 93.1 ± 0.6 6.5 ± 2.5
5 43.1 ± 5.6 N.T 53.6 ± 0.9 60.9 ± 1.2
6 53.7 ± 1.6 94.9 ± 3.5 57.3 ± 0.3 90.9 ± 6.0
7 52.0 ± 0.1 92.1 ± 2.2 41.0 ± 0.1 N.T
8 55.4 ± 2.6 69.5 ± 1.2 18.0 ± 0.4 N.T
9 48.6 ± 0.6 N.T 51.0 ± 1.9 35.1 ± 1.2
10 31.8 ± 0.5 N.T 41.1 ± 2.2 N.T
11 15.8 ± 4.3 N.T 14.0 ± 2.6 N.T
12 28.0 ± 4.0 N.T 18.6 ± 0.3 N.T
13 42.8 ± 0.8 N.T 44.9 ± 1.1 N.T
14 24.9 ± 6.2 N.T 55.0 ± 0.7 22.6 ± 0.4
15 37.1 ± 0.5 N.T 60.7 ± 0.6 33.3 ± 5.0
16 19.0 ± 2.5 N.T 62.0 ± 0.8 31.0 ± 5.5
17 47.1 ± 1.6 N.T 55.4 ± 0.8 84.1 ± 5.7
18 23.1 ± 6.8 N.T 54.4 ± 1.3 51.0 ± 0.5
19 21.5 ± 1.7 N.T 38.5 ± 0.1 N.T
Tacrine
b
0.123 ± 1.5 0.011 ± 0.4
N.T: not test.
a
Compounds were tested three times.
b
Positive control.
Table 2. Enzyme kinetics of compounds 1,3,4and 14 against two ChEs.
Binding mode K
i
(lM)
1 Competitive type (AChE) 16.2 ± 0.9
3 Non-competitive type (BuChE) 4.9 ± 2.1
4 Non-competitive type (BuChE) 3.5 ± 1.5
14 Competitive type (BuChE) 10.7 ± 1.3
JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY 1177

were solved to be 4.9 ± 2.1, 3.5 ± 1.5 and 10.7 ± 1.3 lM on BuChE
(Figure 2(H),Table 2).
Molecular docking of inhibitors with AChE and BuChE
These findings suggest that compounds 1,3,4and 14 may bind
with either AChE or BuChE. An inhibitor (1) was found to dock
into the active site of AChE, thus acting as a competitive inhibitor
of this enzyme. This inhibitor was fitted into the binding site in a
stable position with an Autodock score of –9.42 kcal/mol.
Compound 1formed five hydrogen bonds (Ser203: 2.95 Å;
Phe295: 2.86 Å; Phe338: 2.73 Å; His447: 2.59 Å and 2.67 Å) with
four amino acids in AChE and had a hydrophobic interaction with
amino acids surrounding the active site (Figure 3(A,B)).
Compounds 3,4and 14 exhibited molecular docking with BuChE.
The noncompetitive inhibitors (3and 4) were simulated in a blind
docking test to search for possible binding with BuChE. The com-
petitive inhibitor (14) docked into BuChE in the method described
earlier. As shown in Figure 3(C–F), compounds 3and 4were con-
firmed to have hydrophobic interactions with amino acids, but
not to form hydrogen bonds. The predicted binding site was pro-
posed as the location where the inhibitor clustered with a low
Autodock score. In particular, compound 4was stably placed in
the active site, with the top five positions scoring from –7.25 to
7.26 kcal/mol (Supplementary Figure S1). Enzyme kinetic results
showed that compound 4preferentially bound to the allosteric
site. Therefore, the catalytic site was excluded as a binding site for
this compound (4). Our results predicted the binding site as that
with the lowest Autodock score for the next cluster, similar to the
results for compound 3. In addition, compound 14 exhibited
hydrophobic interactions with seventeen residues and formed six
hydrogen bonds (Asp70: 2.97Å; Glu119: 2.71 Å; Glu276: 2.71 Å;
Asn289: 3.12 Å; Trp430: 2.74 Å; Tyr440: 2.60 Å) with six amino
acids in the active site of BuChE (Figure 3(G,H)).
MDs of inhibitors with AChE and BuChE
We performed MD simulations to study the stability of the inhibi-
tor-ChE complex in solution at 300 K under 1 bar of pressure. The
complexes of AChE with compound 1, and BuChE with com-
pounds 3and 4were simulated stably with potential energies of
about –2.35 10
6
kJ/mol (Figure 4(A)). As indicated in Figure
4(B,C), each enzyme exhibited root mean square derivation
(RMSD) values below 3.5 Å distance and root mean square fluctua-
tions (RMSF) below 5.0 Å distance. Compound 1formed about
1–3 hydrogen bonds with the active site of AChE for 10 ns (Figure
4(D)). Compounds 3and 4created 0–1 hydrogen bonds in the
allosteric site of BuChE (Figure 4(E,F)). Furthermore, these results
indicated the key amino acids interacting with inhibitors during
MD simulation. Analysis of complex formation during a 1-ns inter-
val of MD simulation showed that compound 1 participated in
hydrogen bonding with Tyr337 in the active site of AChE (Figure
4(G,J)). Compounds 3and 4were located within 3.5 Å of Asn228
in the predicted allosteric site of BuChE (Figure 4(H,I,K,L)).
Compound 1maintained a 3.5 Å distance with Tyr337 for 10 ns.
Compounds 3and 4occasionally approached Asn228 at a 3.5
Å distance.
Figure 3. Hydrogen bonds (A) and hydrophobic interaction (B) between compound 1and AChE. Hydrogen bonds and hydrophobic interaction between compounds 3
(C and D), 4(E and F) and 14 (G and H) with BuChE.
1178 J. H. KIM ET AL.

Conclusions
AD is a neurodegenerative disease caused by destruction of neu-
rons in the central nervous system
20
. Cycloartane triterpenoids
and cinnamic acid derivatives from the roots of C. dahurica have
been reported to have a neuroprotective effect on PC12 cells
9,11
.
The cholinergic hypothesis of AD is supported by increased mem-
ory and cognition function after binding of ACh to ACh receptors
in the brain
8,21
. Therefore, AChE and BuChE are considered prom-
ising target enzymes for treating AD disease due to their effect of
decreasing ACh levels
4,8
.
Our study led to isolation of cinnamic acids (1and 2), indoli-
nones (3and 4), and triterpenoid derivatives (5–19) from the roots
of C. dahurica. We analysed these compounds to evaluate their
inhibitory activity on both AChE and BuChE. Compound 1has an
IC
50
value of 16.7 ± 1.9lM against AChE, while compounds 3and 4
were determined to have IC
50
values of 1 3.8 ± 1.2 and 6.5 ± 2.5 lM
against BuChE, respectively. According to BuChE assay results, triter-
penoid glycosides showed more potent inhibitory activities than
those of their aglycones except for compound 19. Above all, indoli-
none derivatives (3and 4) were highly potential inhibitors
(D) (E) (F)
(G) (H) (I)
(J) (K) (L)
(A) (B) (C)
Figure 4. The potential energy (A), RMSD (B), RMSF (C), and hydrogen bonds (D–F) of respective compounds 1,3and 4with receptor for 10,000 ps. The distance
(G–I) of respective compounds 1,3and 4with key amino acids (J–L).
JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY 1179

compared to the others. In reported studies, alkaloid derivatives,
such as atherosperminine, (þ)-N-methylisococlaurine, berberine, 9-
amino-1,2,3,4-tetrahydro acridine, and rivastigmine, were found to
be famous ChEs inhibitor
12,20,21
. Moreover, blood brain barrier (BBB)
plays a role to keep neuronal cells from neurotoxic substances of
outside. However, BBB transporters of glucose, phenylalanine, aragi-
nine and lactate are responsible for transporting small molecules,
such as deoxyglucose, galactose, lysine, pyruvate and guanosine,
into brain
22,23
. Especially, compound 4having 199 Da alkaloid may
overcome the block of BBB and potentially invade into brain.
Based on the enzyme kinetic study, compound 1was shown to
block catalytic reaction by interacting with the active site of AChE.
Compounds 3and 4were revealed to have affinity for the allosteric
site of BuChE. Their binding positions were predicted for the active
or allosteric sites using the Autodock 4.2 package. Moreover, MD ana-
lysis led us to propose the key amino acid involved in ligand-receptor
interactions. As a result, the ketone form of the ester in cimiciphe-
none (1) exhibited hydrogen bonding with the aromatic hydroxyl
group of Tyr337 in the active site of AChE during simulation. To
develop a new cinnamic acid moiety of AChE, chemists should con-
sider compounds that are capable of interaction with Tyr337. (E)-
3–(30-methyl-20-butenylidene)-1-methyl-2-indolinone (3)and(E)-3–(30-
methyl-20-butenylid-ene)-2-indolinone (4) participated in hydrogen
bonding with Asn228 located at its predicted binding site on BuChE.
It is necessary to develop a prenyl indolinone derivative as a non-
competitive inhibitor to enhance hydrogen bonding with this polar
amino acid (Asn228). In our research, we identified the key amino
acids, which could not be found through molecular docking, using
MD analysis. Among cycloartane triterpenoids and cinnamic acid
derivatives that show neuroprotective activity
9,11
, cimiciphenone (1),
(E)-3–(30-methyl-20-butenylidene)-1-methyl-2-indolinone (3)and(E)-
3–(30-methyl-20-butenylidene)-2-indolinone (4)showedpromiseas
potential inhibitors of AChE and BuChE, respectively. Compound 1
was determined to be the optimal compound for development as a
competitive inhibitor of AChE, while compounds 3and 4may pro-
vide a new skeleton for noncompetitive inhibitors of BuChE.
Disclosure statement
The authors have declared no conflict of interest.
Funding
This study was supported by the Priority Research Centre
Programme through the National Research Foundation of Korea
(NRF) funded by the Ministry of Education, Science and
Technology [2009–0093815], Republic of Korea.
References
1. Orhan IE, Senol FS, Shekfeh S, et al. Pteryxin-A promising
butyrylcholinesterase-inhibiting coumarin derivative from
Mutellina purpurea. Food Chem Toxicol 2017;109:970–4.
2. Zhang X-J, Greenberg D. Acetylcholinesterase involvement
in apoptosis. Front Mol Neurosci 2012;5:1–6.
3. Dvir H, Silman I, Harel M, et al. Acetylcholinesterase: From
3D structure to function. Chem Biol Interact 2010;187:10–22.
4. Ali R, Sheikh IA, Jabir NR, Kamal MA. Comparative review of
decade’s research on cholinesterase inhibition. Am J
Neuroprot Neuroregen 2012;4:136–8.
5. Nguyen VTN, Zhao BT, Seong SH, et al. Inhibitory effects of
serratene-type triterpenoids from Lycopodium complanatum
on cholinesterases and b-secretase 1. Chem Biol Interact
2017;274:150–7.
6. Huang Y-J, Huang Y, Baldassarre H, et al. Recombinant
human butyrylcholinesterase from milk of transgenic ani-
mals to protect against organophosphate poisoing. Pro Natl
Acad Sci USA 2007;104:13603–8.
7. de Souza LG, Renn~
a MN, Figueroa-Villar JD. Coumarins as cholin-
esterase inhibitors: A review. Chem Biol Interact 2016;254:11–23.
8. Islam MM, Rohman MA, Gurung AB, et al. Correlation of cho-
linergic drug induced quenching of acetylcholinesterase
bound thioflavon-T fluorescence with their inhibition activ-
ity. Spectrochim Acta A 2018;189:250–7.
9. Lv C, Yang F, Qin R, et al. Bioactivity-guided isolation of
chemical constituents against H
2
O
2
-induced neurotoxicity
on PC12 from Cimicifuga dahurica (Turcz.) Maxim. Bioorg
Med Chem Lett 2017;27:3305–9.
10. Tian Z, Si J, Chang Q, et al. Antitumor activity and mechanisms
of action of total glycosides from aerial part of Cimicifuga
dahurica targeted against hepatoma. BMC Cancer 2007;7:7–10.
11. Qin R, Zhao Y, Zhao Y, et al. Polyphenolic compounds with
antioxidant potential and neuro-protective effect from
Cimicifuga dahurica (Turcz.) Maxim. Fitoterapia 2016;115:52–6.
12. Othman WNNW, Liew SY, Khaw KY, et al. Cholinesterase inhibi-
tory activity of isoquinoline alkaloids from three Cryptocarya
species (Lauraceae). Bioorg Med Chem 2016;24:4464–9.
13. Thao NP, Luyen BTT, Lee JS, et al. Soluble epoxide hydrolase
inhibitors of indolinone alkaloids and phenolic derivatives
from Cimicifuga dahurica (Turcz.) Maxim. Bioorg Med Chem
Lett 2017;27:1874–9.
14. Kusano A, Takahira M, Shibano M, et al. Studies on the
Constituents of Cimicifuga Species. XXVI. Twelve New
Cyclolanostanol Glycosides from the Underground Parts of
Cimicifuga simplex WORMSK. Chem Pharm Bull 1999;47:511–6.
15. Thao NP, Kim JH, Luyen BTT, etal. In silico investigation of cyclo-
artane triterpene derivatives from Cimicifuga dahurica (Turcz.)
Maxim roots for the development of potent soluble epoxide
hydrolase inhibitors. Inter J Biol Macromole 2017;98:526–34.
16. Li JX, Kadota S, Hattori M, et al. Constituents of Cimicifugae
Rhizoma. I. Isolation and Characterization of Ten New
Cycloartenol Triterpenes from Cimicifuga heracleifolia
KOMAROV. Chem Pharm Bull 1993;41:832–41.
17. Thao NP, Luyen BT, Lee JS, et al. Inhibition potential of cycloar-
tane-type glycosides from the roots of Cimicifuga dahurica
against soluble epoxide hydrolase. J Nat Prod 2017;80:1867–75.
18. Imai A, Lankin DC, Nikoli
c D, et al. Cycloartane triterpenes from
the aerial parts of Actaea racemosa. J Nat Prod 2016;79:541–54.
19. Yoshimitsu H, Nishida M, Sakaguchi M, Nohara I. Two new 15-
deoxycimigenol-type and three new 24-epi-cimigenol-type glyco-
sides from Cimicifuga rhizome. Chem Pharm Bull 2006;54:1322–5.
20. Groner E, Ashani Y, Schorer-Apelbaum D, et al. The kinetics of
Inhibition of Human acetylcholinesterase and butyrylcholinesterase
by two series of novel carbamates. Mol Pharmacol 2007;71:1610–7.
21. Jiang Y, Gao H, Turdu G. Traditional Chinese medicinal herbs
as potential AChE inhibitors for anti-Alzheimer’s disease:
A review. Bioorg Chem 2017;75:50–61.
22. Pardridge WM. Drug transport across the blood-brain bar-
rier. J Cereb Blood Flow Metab 2012;32:1959–72.
23. Grabrucker AM, Ruozi B, Belletti D, et al. Nanoparticle trans-
port across the blood brain barrier. Tissue Barriers 2016;
4:e1153568–19.
1180 J. H. KIM ET AL.