Discovery of Novel Pyrazolopyrimidinone Derivatives as Phosphodiesterase 9A Inhibitors Capable of Inhibiting Butyrylcholinesterase for Treatment of Alzheimer's Disease

Article abstract of DOI:10.1021/acschemneuro.7b00268

Discovery of multitarget-directed ligands (MTDLs), targeting different factors simultaneously to control the complicated pathogenesis of Alzheimer's disease (AD), has become an important research area in recent years. Both phosphodiesterase 9A (PDE9A) and butyrylcholinesterase (BuChE) inhibitors could participate in different processes of AD to attenuate neuronal injuries and improve cognitive impairments. However, research on MTDLs combining the inhibition of PDE9A and BuChE simultaneously has not been reported yet. In this study, a series of novel pyrazolopyrimidinone-rivastigmine hybrids were designed, synthesized, and evaluated in vitro. Most compounds exhibited remarkable inhibitory activities against both PDE9A and BuChE. Compounds 6c and 6f showed the best IC₅₀ values against PDE9A (6c, 14 nM; 6f, 17 nM) together with the considerable inhibition against BuChE (IC₅₀, 6c, 3.3 μM; 6f, 0.97 μM). Their inhibitory potencies against BuChE were even higher than the anti-AD drug rivastigmine. It is worthy mentioning that both showed moderate selectivity for BuChE over acetylcholinesterase (AChE). Molecular docking studies revealed their binding patterns and explained the influence of configuration and substitutions on the inhibition of PDE9A and BuChE. Furthermore, compounds 6c and 6f exhibited negligible toxicity, which made them suitable for the further study of AD in vivo.

Full text of DOI:10.1021/acschemneuro.7b00268

Subscriber access provided by BOSTON UNIV
Article
Discovery of novel pyrazolopyrimidinone derivatives as PDE9A inhibitors
capable of inhibiting BuChE for treatment of Alzheimer’s disease
Yan-Fa Yu, Ya-Dan Huang, Chen Zhang, Xu-Nian Wu, Qian Zhou, Deyan Wu, Yinuo Wu, and Hai-Bin Luo
ACS Chem. Neurosci., Just Accepted Manuscript • DOI: 10.1021/acschemneuro.7b00268 • Publication Date (Web): 07 Aug 2017
Downloaded from http://pubs.acs.org on August 9, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Chemical Neuroscience is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 42
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
Discovery of novel pyrazolopyrimidinone derivatives as PDE9A
inhibitors capable of inhibiting BuChE for treatment of
Alzheimer’s disease
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
YanꢀFa Yu‡,a YaꢀDan Huang‡,a Chen Zhang, a XuꢀNian Wu, a Qian Zhou, a Deyan Wu,a
Yinuo Wu a,* and HaiꢀBin Luoa,b,*
a School of Pharmaceutical Sciences, Sun YatꢀSen University, Guangzhou 510006, PR China
b Collaborative Innovation Center of High Performance Computing, National University of Defense
Technology, Changsha 410073, China
‡These authors contributed equally.
KEYWORDS: Phosphodiesterases 9 (PDE9), Butyrylcholinesterase (BuChE), Multiꢀtargetꢀdirected
ligands, Alzheimer’s disease
1
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 2 of 42
1
2
3
4
5
6
7
8
9
ABSTRACT: Discovery of multiꢀtargetꢀdirected ligands (MTDLs) targeting at different factors
simultaneously to control the complicated pathogenesis of Alzheimer's disease (AD), has become an
important research area in recent years. Both PDE9A and BuChE inhibitors could participate in different
processes of AD to attenuate neuronal injuries and improve cognitive impairments. However, research
on the MTDLs combining the inhibition of PDE9A and BuChE simultaneously has not been reported
yet. In this study, a series of novel pyrazolopyrimidinoneꢀrivastigmine hydrids were designed,
synthesized, and evaluated in vitro. Most compounds exhibited remarkable inhibitory activities against
both PDE9A and BuChE. Compounds 6c and 6f showed the best IC50 values against PDE9A (6c: 14
nM, 6f:17 nM, respectively) together with the considerable inhibition against BuChE (IC50, 6c: 3.3 ꢁM,
6f: 0.97 ꢁM). Their inhibitory potencies against BuChE were even higher than the antiꢀAD drug
rivastigmine. Worthy to mention is that both showed moderate selectivity for BuChE over AChE.
Molecular docking studies revealed their binding patters and explained the influence of configuration
and substitutions on the inhibition of PDE9A and BuChE. Furthermore, compounds 6c and 6f exhibited
negligible toxicity, which made them suitable for the further study of AD in vivo.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
ACS Paragon Plus Environment

Page 3 of 42
ACS Chemical Neuroscience
1
2
3
4
1. Introduction
5
6
7
8
9
Alzheimer's disease (AD) is an ageꢀrelated neurodegenerative disorder, which is characterized by
progressive memory loss, decline in language skills, and cognitive dysfunction.1 In the report by
Alzheimer’s association (USA) in 2016,2 AD has already become the third cause of death after
cardiovascular diseases and cancer. About 46.8 million people worldwide are currently suffering from
AD and about 10 million new cases were diagnosed in 2015. The number will rise to 131 million by
2050. The global cost for medication and care in AD patients was about $818 billion in 2015, which will
rise to $1,000 billion in 2018 and $2,000 billion in 2030, respectively, bringing heavy economic burden
to society and families.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The pathogenesis of AD is complicated and still not fully understood yet.3 Low level of
neurotransmitter acetylcholine (ACh) has been regarded as one of the main reasons for cognitive and
memory impairments of AD patients.4 Acetylcholinesterase inhibitors (AChEIs) could inhibit
acetylcholinesterase (AChE) reversibly or irreversibly, increasing the concentration of ACh in brain to
improve cognitive impairments. Currently, three out of four drugs approved by FDA in AD therapies are
AChEIs (donepezil, rivastigmine, and galantamine).5 However, the concentration of AChE would fall to
10ꢀ15% in advanced AD patients, which made selective AChEIs work not as efficient as that in the early
stage. Butyrylcholinesterase (BuChE) is the other major form of cholinesterases, the inhibition of which
could also improve the levels of ACh. Compared to low level of AChE in AD advanced stage, the
concentration of BuChE remained the same or even increased up to 2ꢀfold.6 Now, dual inhibitors of
AChE and BuChE or selective BuChE inhibitors have been considered as new therapeutic strategies for
AD patients, especially for those in the advanced stage.7
Phosphodiesterases 9 (PDE9), belonging to the phosphodiesterase (PDE) super family of enzymes,
could hydrolyze the intracellular second messenger cyclic guanosine monophosphate (cGMP)
specifically with the highest affinity among all the 11 PDE subfamilies.8 PDE9A is the only isoform of
3
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 4 of 42
1
2
3
4
5
6
7
8
9
PDE9, which is highly expressed in the cortex, basal ganglia, hippocampus, and cerebellum of brain.
The inhibition of PDE9A could decrease the concentration of cGMP in brain to active
NO/cGMP/PKG/CREB signaling, increasing the level of brainꢀderived neurotrophic factor (BDNF),
restoring proteins associated with memory and attenuating neuronal injuries in hippocampal area. Thus,
PDE9 has been regarded as a novel and promising therapeutic target for AD.9 PDE9 inhibitors BAY 73ꢀ
6691 and PFꢀ04447943 could improve memory performance in several rodent memory tasks including
social recognition and object recognition tasks, Morris water maze and etc.10,11 Compared with AChE
inhibitor donepezil, PDE9 inhibitor BAY 73ꢀ6691 could not only prolong both early and late longꢀterm
potentiation (LTP) in hippocampal slices, but also transform early LTP into late LTP.12 In the APP
transgenic Tg2576 mice, BAY 73ꢀ6691 could restore Aβ42 oligomers induced LTP and PFꢀ04447943
could regulate dendritic spine density in hippocampal neurons.13,14 Furthermore, BAY 73ꢀ6691 could
protect Aβ25–35 induced oxidative damage in hippocampus and attenuate damage in hippocampal
neurons.15 Currently, PDE9 inhibitors BIꢀ409306 and PFꢀ04447943 are subjected to AD clinical trials.
Based on the multifactorial nature of AD, multiꢀtargetꢀdirected ligands (MTDLs), single compounds
which target at different enzymatic systems or receptors involved in AD simultaneously, attracted
widespread attention in recent years, and thus numerous MTDLs were developed.16ꢀ17 The
pharmacophores of current antiꢀAD drugs such as donepezil and rivastigmine were often used in the
discovery of MTDLS with cholinesterase inhibitory activities. Several efficient ones such as ladostigil
and memoquin have been proved to be active in preclinical or even clinical trials.18,19
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
On the other hand, an in vivo study demonstrated that coꢀadministration of PDE4 inhibitor roflumilast
and AChE inhibitor donepezil could fully reverse scopolamineꢀinduced memory deficit in the objective
recognition task, while single administration of each drug could not.20 A clinical study showed that both
AChE inhibitor galantamine and PDE3 inhibitor cilostazol monotherapy increased cognitive, affective,
and activities of daily living functions in AD patients and the combination of them gave better effect
4
ACS Paragon Plus Environment

Page 5 of 42
ACS Chemical Neuroscience
1
2
3
21
than the monotherapy.
All these results indicated that controlling the PDE inhibition and ChE
4
5
6
7
8
inhibition simultaneously may have better results than single one. However, multiꢀtargetꢀdirected ligand
combining the inhibition of PDEs and ChE simultaneously has not been reported yet.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Aiming at discovery of multiꢀtargetꢀdirected ligands combining dual inhibitory activities of PDE9A
and ChE, a series of novel pyrazolopyrimidinoneꢀrivastigmine hydrids were designed and synthesized in
this study. The inhibition of PDE9A, AChE, and BuChE were evaluated in vitro. The antioxidant
activity and toxicity were also assessed. Molecular docking and dynamics simulations were used to
explore binding modes and underlying mechanisms of lead compounds with enzymes, providing
evidence for further structural modifications.
2. Result and discussion
2.1. Rational design of PDE9A inhibitors capable of inhibiting ChE.
We used to report an efficient PDE9A inhibitor C33, which showed IC50 of 16 nM against PDE9A
and exhibited good pharmacokinetic properties.22 Experiments in vivo demonstrated that C33 has
memory improving effect on learning and memory dysfunction animal model induced by scopolamine
or Aβ1ꢀ42. From the Xꢀray crystal structure of the C33ꢀPDE9A complex, below information was
obtained. C33 formed a hydrogen bond with Gln453 and an aromatic πꢀπ stack against Phe456 via its
pyrazolopyrimidinone scaffold (Figure 1). As Gln453 and Phe456 are two conservative residues in the
binding pocket of PDE9A, the interactions with them are important for the affinity of inhibitors with
PDE9.23ꢀ25 Thus, pyrazolopyrimidinone motif was kept as the scaffold of our designed compounds for
their potency with PDE9A.
Among current AChEI drugs approved for AD, rivastigmine is the only one inhibiting both AChE
and BuChE simultaneously.26 Clinical studies proved that mild to moderately severe AD patients
treating with rivastigmine have apparent ameliorative changes in the brain.27 This result is in concert
5
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 6 of 42
1
2
3
4
5
6
7
8
9
with the hypothesis that inhibition of both AChE and BuChE may have neuroprotective and diseaseꢀ
modifying effects in the advanced AD patients. According to the crystal structures of the AChEꢀ
rivastigmine and BuChEꢀrivastigmine complexes, the carbamate moiety of rivastigmine linked to the
activeꢀsite serine of cholinesterase covalently, which is the pharmacophore for its potency against AChE
and BuChE.28
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Based on these evidences, a series of novel pyrazolopyrimidinoneꢀrivastigmine hydrids were
designed, which comprised of pyrazolopyrimidinone and carbamate moiety in one molecule. The
phenylethylamine motif, existing in the structure of both C33 and rivastigmine, was used as part of the
linker according to drug fusion principle (Figure 1).
Figure 1. Rational design of PDE9 inhibitors capable of inhibiting ChE.
2.2. Chemistry
The synthetic route of compounds 6aꢀ6f was outlined in Scheme 1.22,24 The starting material
trichloropyrimidine (1) could be obtained in high yield by the VilsmeierꢀHaack reaction of barbituric
acid in the presence of phosphorus oxychloride. The pyrazole ring of compound 1 was then closed by
6
ACS Paragon Plus Environment

Page 7 of 42
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
reaction with cyclopentyl hydrazine (2), giving the pyrazolo[3,4ꢀd]pyrimidine 3. Hydrolysis of
compound 3 by KOH afforded the pyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀone 4. Subsequently, compound 4
reacted with corresponding amines to yield the series of compounds 5 in good yields. The
carbamoylation of compounds 5aꢀ5e with the N, Nꢀethyl(methyl)carbamic chloride or N, Nꢀ
dimethylcarbamic chloride in DMF with the assistance of NaH led to the designed compounds 6aꢀ6f.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Cl
O
Cl
O
R
HCl
NH2
HN
N
HN
N
H
R'
NH2
b
a
N
N
+
N
N
Cl
N
Cl
N
Cl
N
Cl
c
1
2
3
4
O
N
O
N
R
HN
R
HN
N
N
d
N
N
N
N
R"
H
R'
H
6
5
5a, R=H, R'=3-OH
6a, R=H, R''=3-OC(O)NEtMe
5b, R=H, R'=3-OH,4-OMe
5c, R=H, R'=3-OMe,4-OH
5d, R=(R)-Me, R'=3-OH
5e, R=(S)-Me, R'=3-OH
6b, R=H, R''=3-OC(O)NEtMe,4-OMe
6c, R=H, R''=3-OMe,4-OC(O)NEtMe
6d, R=(R)-Me, R''=3-OC(O)NEtMe
6e, R=(S)-Me, R''=3-OC(O)NEtMe
6f, R=H, R''=3-OMe,4-OC(O)NMe2
o
Scheme 1. Reagents and conditions: (a) Triethylamine, ꢀ78 C, THF; (b) 2N KOH, 60 oC; (c)Amine, iꢀ
PrOH, Triethylamine, reflux; (d) N, Nꢀethyl(methyl)carbamic chloride or N, Nꢀdimethylcarbamic
chloride, NaH, DMF, r.t.
The synthetic route of 13aꢀ13b is outlined in Scheme 2.29 The condensation of 2ꢀ
(methoxymethylene)malononitrile (7) with cyclopentylhydrazine (8) provided intermediate 9 in the
presence of triethylamine. The intermediate 9 was then hydrolyzed with ammonium hydroxide and
hydrogen peroxide to obtain intermediate 10. And then, intermediate 10 reacted with (R)ꢀethylꢀ2ꢀ
(((benzyloxy)carbonyl)amino)propanoate in the next step to give compound 11 under basic condition.
Compound 11 was then hydrogenated with Pd/C to remove the Cbz group, providing intermediate 12 in
7
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 8 of 42
1
2
3
4
high yield. Compounds 13a and 13b were prepared in good yields by reaction of the corresponding
5
6
aldehydes and sodium borohydride.
7
8
9
O
O
N
O
NC
H
Cl
NH2
O
N
H2N
H2N
HN
OEt
HN
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
NC
CN
N
N
N
b
O
a
N
+
N
N
H2N
c
OMe
NHCbz
7
8
9
11
10
O
N
CHO
O
O
HN
R
N
N
O
HN
MeO
d
N
N
R'
OMe
13a
13b
: R = Me, R' = Me
: R = Me, R' = Et
N
N
NH
e
O
NH2
R
N
O
12
R'
Scheme 2. Reagents and conditions: (a) Triethylamine, 0 oC, EtOH; (b) NH3·H2O, H2O2, EtOH, r.t.; (c)
(R)ꢀethylꢀ2ꢀ(((benzyloxy)carbonyl)amino)propanoate, 60% NaH, dry THF, r.t.; (d)Pd/C, MeOH, H2; (e)
iꢀPrOH, NaBH4, r.t.
2.3. The inhibitory activities of target compounds towards PDE9A in vitro
The inhibitory activities of compounds 6aꢀ6f against PDE9A were determined by bioassay with the
same protocol as our previous reported work.30 As compounds 5aꢀ5e have the pyrazolopyrimidinone
scaffold, the inhibitory activities of compounds 5aꢀ5e against PDE9 were also evaluated. Bay 73ꢀ6691
was used as the reference compound for the bioassay (IC50 = 51 nM).31 The results were summarized in
Table 1.
As expected, most designed compounds were potent PDE9A inhibitors. In the series of compounds
5, with the hydroxyl group at the 3 or 4 position on the phenyl ring, the IC50 values against PDE9A
ranged from 13 to 69 nM. Compound 5b with 3ꢀ hydroxylꢀ4ꢀmethoxy groups and compound 5c with 3ꢀ
methoxyꢀ4ꢀhydroxyl groups substituted on the phenyl ring, gave comparable IC50 values of 39 and 35
nM respectively, while compound 5a with only a hydroxyl group gave a slight lower IC50 of 53 nM. In
our previous study, C10 with no substituted group and C33 with 4ꢀchloro substituted on the phenyl ring
8
ACS Paragon Plus Environment

Page 9 of 42
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
gave the IC50 values of 39 nM and 16 nM, respectively.22 Combined with these data, we concluded that
the properties of substitutions at phenyl nucleus would affect the inhibitory activities against PDE9A but
not significantly. Compared with compound 5a, compound 5d with a methyl group at the R position
gave the best IC50 (13 nM) among all the compounds in series 5, indicating that the steric hinderance has
important effect on the affinity of compounds with PDE9A. However, compound 5e, the stereoisomer of
5d with different configuration, gave the IC50 of 69 nM against PDE9A. This difference in inhibitory
activities caused by configuration also occurred on the two stereoisomers of C33 but with totally
contrary results. The Sꢀisomer of C33 gave better IC50 value (11 nM) than Rꢀisomer (58 nM). The
crystal structure of the C33ꢀPDE9 complex revealed that the chlorophenyl tail of (S)ꢀC33 interacted
with a small hydrophobic pocket while that of (R)ꢀC33 pointed out the molecular surface. The slightly
better binding affinity of (S)ꢀC33 may come from more van der Waals interactions in the hydrophobic
pocket. As the major structural difference among C33, 5d, and 5e is the hydroxyl group on the phenyl
ring, we concluded that the hydroxyl group of 5d and 5e might form additional interactions with
residues in the binding pocket of PDE9A, leading to the different inhibitory activities of 5d and 5e.
In the series of compounds 6, with carbamate attaching on the hydroxyl group of compounds 5, the
results were quite different from structureꢀactivity relationship (SAR) of 5. Compound 6c, with 3ꢀ
methoxyl and 4ꢀcarbamate groups on the phenyl ring, gave the best IC50 of 14 nM. Compound 6a with
3ꢀcarbamate group gave a comparable IC50 against PDE9A (IC50: 19 nM). The two stereoisomers 6d and
6e with different configurations, gave the IC50 of 37 nM and 30 nM, respectively. It seemed that the
difference in inhibitory activities of the two stereoisomers 5d and 5e disappeared after the carbamate
groups attaching on the hydroxyl group of 5. This result further supported our speculation that the
hydroxyl group of compounds 5d and 5e may form additional interactions with PDE9A. Furthermore, as
the volume of N, Nꢀethylmethylcarbamate group is relative large, we speculated that the attachment of
this group on compounds 6aꢀ6f might affect spatial orientation of phenyl group, resulting in the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 10 of 42
1
2
3
4
5
6
7
8
9
difference in affinity with PDE9A. Compound 6f, with a N, Nꢀdimethylcarbamate group was also
synthesized and evaluated for the comparison with compound 6c. The IC50 of compound 6f against
PDE9A was 17 nM, which was almost the same with compound 6c. Compounds 13a and 13b with
different linkers were also designed and synthesized. However, both of them showed low inhibition
against PDE9A at 100 nM. We concluded that the size of linker used in compounds 13a and 13b might
be too large to fit well in the binding pocket of PDE9A for high affinity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.4. The inhibitory activities of target compounds towards AChE and BuChE in vitro
The inhibitory activities against AChE (from electric eel) and BuChE (from Equine serum) were
measured by the Ellman’s method.32 Rivastigmine and donepezil were used as the reference
compounds.33,34 The results were summarized in Table 2.
As our expected, compounds 5aꢀ5e without the carbamate group have almost no inhibition of AChE
and BuChE at a concentration of 20 µM. Compounds 6aꢀ6f with the introduction of carbamates on the
hydroxyl group of 5aꢀ5e, have an increased inhibitory capacity of both AChE and BuChE. To our
surprise, most compounds displayed good inhibitory activities against BuChE, with IC50 values ranging
from 0.96 to 18.8 ꢁM, while the inhibition against AChE at 20 µM were only between 17% to 28%.
These results indicated that compounds 6aꢀ6f were selective BuChE inhibitors. As is mentioned in the
introduction, the level of butyrylcholinesterase remained the same or even increased up to 2ꢀfold while
the level of acetylcholinesterase decreased in the advanced AD stage. Dual inhibitors of both AChE and
BuChE or selective inhibitors of BuChE may be more suitable for therapy advanced AD patients.
Both the substituted groups on the phenyl ring and the configuration at the R position have effects
on the inhibitory activities. Among compounds 6aꢀ6e with N, Nꢀmethylethyl carbamate group,
compound 6c with the 3ꢀmethoxyl and 4ꢀcarbamate group gave the best IC50 of BuChE (3.3 ꢁM). The
two stereisomers 6d and 6e, showed the IC50 of BuChE 3.7 and 17.9 ꢁM, respectively, implying that the
10
ACS Paragon Plus Environment

Page 11 of 42
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
difference in configurations plays an important role in the inhibition of BuChE. Compound 6f with N, Nꢀ
dimethyl carbamate groups at 4ꢀposition was also synthesized and evaluated, showing a better inhibitory
activity (0.96 ꢁM) than compound 6c (3.3 ꢁM). Both are more potent than current drugs rivastigmine
(2.8 ꢁM) and donepezil (8.1 ꢁM). The selectivities of compounds 6c and 6f (BuChE/AChE) are above
7ꢀfold and 20ꢀfold, respectively. It is worthy to mention that compounds 6c and 6f also exhibited
remarkable inhibitory affinity towards PDE9A with the IC50 of 14 and 17 nM, respectively. These are in
accordance with our aim at designing dual inhibitors of PDE9A and cholinesterase.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The inhibition of AChE and BuChE by compounds 13a and 13b were also tested. Among all the
designed compounds with carbamate groups, these two compounds showed the highest BuChE
inhibitory activities, with the IC50 of 0.27 and 0.57 ꢁM respectively, which was 10ꢀfold and 6ꢀfold than
that of rivastigmine. The selectivity of compound 13a (BuChE/AChE) could be up to 100. Although the
low inhibition against PDE9A limited them to be dual inhibitors of PDE9A and cholinesterase,
structural evidence provided by these two compounds may be useful for further modification.
2.5. Antioxidant activity by the ORAC method.
Oxidative stress has been considered as an important factor in the AD onset and progression. The
oxidative damage caused by ROS and RNS may accelerate the aggregation of amyloid plaques in AD
patients.35, 36 Thus, antioxidants have been widely applied in the MTDLS for AD treatment. Hydroxylꢀ
phenyl motif existed in many antioxidants such as resveratrol, ferulic acid and has been regarded as the
pharmacophore for their antioxidant activities.37,38 Oxygen radical absorbance capacity assay (ORACꢀ
FL) is a widely applied method to test the antioxidant activities of hydrophilic antioxidants including
phenolic derivatives. As compounds 5aꢀ5e with good inhibition against PDE9A contained a hydroxylꢀ
phenyl motif, antioxidant activities of these compounds were evaluated by ORACꢀFL method. The
results were summarized in Table 2. As expected, all these compounds have good antioxidant activities,
11
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 12 of 42
1
2
3
4
5
6
7
8
9
with ORACꢀFL values ranging from 0.9ꢀ1.8 Trolox equivalents. Compound 5d, with the best IC50 value
of 13 nM against PDE9A, gave an ORAC value of 1.0 Trolox. Compound 5c, with a relatively lower
IC50 value of 35 nM against PDE9A, gave the best ORAC value of 1.8 Trolox, which is even more
active than the reference compound ferulic acid (1.1). These results indicated that compounds 5d and 5c
were potent PDE9A inhibitors with antioxidant activities.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.6. Inhibition of Aβ1ꢀ42 selfꢀaggregation by ThT assay.
One important hallmarks in the AD patients is amyloid beta (Aβ) aggregation.39 Polyphenols, such as
caffeic acid, feurlic acid have been reported that could inhibit Aβ selfꢀaggregation.40 Thus, a ThT
fluorescence assay was performed on the designed compounds to evaluate their effects on Aβ1ꢀ42 selfꢀ
aggregation. The results were shown in Table 2. Compounds 5a, 5b, 6a, and 6b were able to inhibit Aβ
selfꢀaggregation to some extent at the concentration of 50 ꢁM. Compound 5a containing a hydroxyl
group, gave a comparable activity (80%) to that of curcumin, a positive compound widely used for the
evaluation of Aβ selfꢀaggregation.
2.7. The selectivity of compounds 5d and 6c over PDEs.
The PDE enzyme superfamily are encoded with 21 genes and divided into 11 families according to
the difference in structure and distribution. However, the binding pockets of some PDEs subfamilies are
quite similar. Thus, PDE inhibitor may target at more than one PDE protein simultaneously, resulting in
low selectivity. As each PDE subfamily is distributed differently and involved in different biological
process, low selectivity over other PDEs may cause severe drug adverse effects and limit clinical usage.
Thus, compounds 5d and 6c, with high IC50 against PDE9A (5d: 13 nM, 6c: 14 nM) were choose to
explore the selectivities over PDE1A, PDE4A, PDE5A and PDE8A. The results were shown in table 3.
For the PDE9A inhibitors, selectivities over PDE1A and PDE8A are usually difficult to be obtained.23ꢀ25
12
ACS Paragon Plus Environment

Page 13 of 42
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
Compounds 5d and 6c showed the selectivities of 8 and 9 over PDE1A, respectively. These results were
almost the same as C33. In our previous report, we found that the interaction with Tyr424 in the binding
pocket of PDE9A was important for high selectivity over other PDEs. The crystal of the C33ꢀPDE9
complex revealed that C33 didn’t form the interaction with Tyr424 but entered into a small selective
pocket, resulting in its moderate selectivity over PDE1A. Thus, we speculated that both compounds 5d
and 6c didn't form the hydrogen bond with Tyr424, either. Furthermore, PDE8A protein contains a
tyrosine residue at the corresponding Tyr424 position of PDE9A, the interaction with tyrosine residue
might result in the low selectivity over PDE8A. In our tests, the selectivities of compounds 5d and 6c
over PDE8A could get up to 700. This results further supported our speculation that compounds 5d and
6c didn't form the hydrogen bond with Tyr424. The inhibition ratio of 5d at 1ꢁM and 10ꢁM is 29% and
84%, respectively, while that of 6c is 27% and 81%, respectively. Thus, the selectivites of compounds
5d and 6c over PDE4D were above 77 folds. The selectivities of compounds 5d and 6c over PDE5A
were also evaluated, the selectivities of compound 5d over PDE5A is 8 folds and that of 6c is 3 folds.
Combined with these results, we concluded that both 5d and 6c were selective PDE9A inhibitors.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.6. Effects of test compounds on the cell viability
The toxicity of compounds 5c, 5d, 6c and 6f were evaluated by using cell viability assays on the SHꢀ
SY5Y cell line (Figure 2). For compound 5c with moderate IC50 against PDE9A and best antioxidant
activity (1.8), the cell viability were 37% and 85% at 100 µM and 40 µM, respectively. For compound
5d with the best IC50 against PDE9A and antioxidant activity(1.0), the cell viability were 70% and 97%
at 100 µM and 40 µM, respectively. For compound 6c, the dual inhibitor of PDE9A and BuChE, the cell
viability were 53% and 83% at 100 µM and 40 µM, respectively. For compound 6f, the other dual
inhibitor of PDE9A and BuChE, the cell viability is 58% and 79% at 100 µM and 40 µM, respectively.
13
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 14 of 42
1
2
3
4
5
6
These results indicated that all these inhibitors we obtained were almost nonꢀtoxicity and suitable for
further exploration.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Cell viability in SHꢀSY5Y cells of compounds 5c, 5d, 6c and 6f
2.7. Molecular modeling studies on the PDE9Aꢀinhibitor complexes.
For compounds 5d and 6c, which showed the best IC50 value against PDE9A of 13 and 14 nM,
respectively, molecular docking studies were performed to explore their binding modes with PDE9A.
Furthermore, in order to explain the influence on affinity towards PDE9 caused by chiral conformation
of inhibitors, binding modes of compounds 5e, 6d, and 6e were also investigated.
In the binding pocket of PDE9A, Gln453 and Phe456 are two key residues for the affinity of
inhibitor with PDE9A. As shown in Figure 3, compound 6c formed an additional hydrogen bond with
the key residue Gln453 besides two common hydrogen bonds with Gln453 and πꢀπ stack with Phe456,
resulting in its high affinity of PDE9A. The phenyl tail of two stereoisomers 5d and 5e entered into a
small pocket of PDE9 protein like C33 (Figure 4A, C, and E), but stretched to a different position
14
ACS Paragon Plus Environment

Page 15 of 42
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
because of chiral configuration. The hydroxyl group on the compound 5d with R configuration, formed
an additional hydrogen bond with Ala452 beside the two hydrogen bonds with Gln453 whereas
compound 5e not. This might be the main reason that 5d with R configuration gave better affinity of
PDE9 than 5e with S configuration.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3: The binding mode of compound 6c with PDE9A
After the carbamate groups were attached on the hydroxyl groups of 5d and 5e, forming the
corresponding 6d and 6e, the influence on affinities caused by chiral configuration disappeared.
Observed from the binding modes, compounds 6d and 6e adopted a similar binding mode (Figure 4B, D,
and F), forming two hydrogen with Gln453 and an additional hydrogen bond with Ala452 via the
nitrogen atom on the linker. However, dislike compounds 5d and 5e, the tail of compounds 6d and 6e
switched to another position due to the steric hindrance of carbamate groups, forming a face to edge πꢀπ
interaction with Phe456. Thus, compounds 6d and 6e gave a comparable IC50 value against PDE9A.
For the compound 13a and 13b, which showed low inhibitory activities toward PDE9A, molecular
docking method was adopted in the same way but could not obtain an appropriate configuration of
complexes between 13a or 13b and PDE9A. The volume of side chains attached on the
15
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 16 of 42
1
2
3
4
pyrazolopyrimidinone ring of 13a and 13b were too large to enter the binding pocket of PDE9A,
5
6
resulting in their low inhibitory activities.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4: The binding mode of compounds 5d, 5e, 6d, and 6e with PDE9A protein. A) 5dꢀPDE9A; B)
6dꢀPDE9A; C) 5eꢀPDE9A; D) 6eꢀPDE9A; E) 5d+5eꢀPDE9A; F) 6d+6e ꢀPDE9A
16
ACS Paragon Plus Environment

Page 17 of 42
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5: The binding patterns of compounds 6f and 13a with the AChE and BuChE.
2.8. Molecular modeling studies on the ChEꢀinhibitor complexes.
The binding modes of compounds 6f and 13a with AChE and BuChE were investigated by molecular
docking, in order to explore the underlying mechanism for their high selectivity of BuChE over AChE.
As shown in Figure 5, compounds 6f and 13a adopted similar binding modes. The carbamate group of
17
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 18 of 42
1
2
3
4
5
6
7
8
9
them inserted the catalytic domain of AChE and BuChE, forming van der Waals interactions with key
amino residues Ser203, Glu334, and His447 in AChE and Ser198, Glu325, and His438 in BuChE,
respectively. The pyrazolopyrimidinone motif formed πꢀπ stacking interactions with Trp286 in AChE
and Tyr332 in BuChE, respectively. However, two relative large amino residue Phe295 and Phe297
existed in the binding pocket of AChE, which made the midꢀgorge site of AChE relatively small. The
tail of compounds 6f and 13a could not fit well in this area. This might be the possible reason for the
high selectivities of 6f and 13a over AChE.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The binding free energies of compound 13a, which showed the best IC50 of 0.27 µM against BuChE
and selectivity above100 over AChE, was calculated by the MMꢀPBSA method (Table 4).41 The
calculation results were in accordance with our experimental results. The absolute binding free energy of
13aꢀBuChE was higher than that of 13aꢀAChE. Electrostatic components (ꢂGele) played a key role in
the binding of 13a with AChE and so did van der Waals (ꢂGvdw) in the binding of 13a with BuChE. In
addition, it was clearly visible that nonꢀpolar component to solvation (ꢂGnonpol, solv) values was an
unfavorable factor for the binding of 13a with both AChE and BuChE.
3. Conclusion
In conclusion, a series of novel pyrazolopyrimidinoneꢀrivastigmine hydrids were designed,
synthesized, and evaluated in vitro. Most designed compounds exhibited dual inhibitory activities
towards both PDE9A and BuChE. Compounds 6c and 6f, sharing a similar structure, were identified to
be the most efficient ones. Both of them showed excellent inhibitory activities against PDE9A (6c: 14
nM and 6f: 17 nM) and also were highly selective inhibitors towards BuChE (6c: 3.3 ꢁM; 6f: 0.97 ꢁM).
As a result, the inhibitory potency against BuChE and selectivity of BuChE/AChE of 6f are much higher
than current drug rivastigmine. Molecular docking studies were performed to explore structureꢀactivity
relationship of these compounds, which revealed the detailed binding modes and influence on inhibitory
18
ACS Paragon Plus Environment

Page 19 of 42
ACS Chemical Neuroscience
1
2
3
4
5
activities caused by configurations. Cell viability assay in the SHꢀSY5Y cell line showed that both 6c
and 6f have negligible toxicity, being worthy to be further explored in vivo study of AD.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4. Experimental section
4.1. Chemistry
Unless otherwise mentioned, all reagents were purchased from commercial suppliers and used without
purification. Thin layer chromatography was performed on precoated silica gel 60 F254 plates. Silica gel
1
(230ꢀ400 mesh) was used for column chromatography. H NMR spectra were recorded on a 400 MHz
spectrometer. Spectra were referenced internally to the residual proton resonance in CDCl3 (δ 7.26 ppm)
as the internal standard. Chemical shifts (δ) were reported as part per million (ppm) in δ scale downfield
from TMS. 13C NMR spectra were recorded on a 100 MHz spectrometer and the spectra were referenced
to CDCl3 (δ 77.0 ppm, the middle peak). Coupling constants (J) were reported in Hertz (Hz). Highꢀ
resolution mass spectra (HRMS) were obtained on a LTꢀTOF.
4.1.1. General for the preparation of compound 5aꢀ5e.
To the solution of intermediate 4 in iꢀpropanol (2.0 mL), the corresponding amine (0.4 mmol) and
o
trimethylamine (4 mg, 0.4 mmol) were added. The mixture was stirred at 80 C for 1 hour. After the
intermediate 4 was disappeared on the TLC, the solvent was removed by evaporation. The residue was
purified on column chromatography (CH2CI2: MeOH=50:1) to give the compounds 5aꢀ5e in good yield.
4.1.2. 1ꢀcyclopentylꢀ6ꢀ((3ꢀhydroxybenzyl)amino)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4(5H)ꢀone (5a)
It was synthesized with 3ꢀ(aminomethyl)phenol as the starting material. Elution with methanol and
dichloromethane (1:50) afforded compound 5a (0.20 g, 61%) as a white solid.1H NMR (400 MHz,
DMSO) δ 10.48 (s, 1H), 7.75 (s, 1H), 7.13 (t, J = 7.8 Hz, 1H), 6.98 (s, 1H), 6.78 (d, J = 7.4 Hz, 2H),
6.65 (dd, J = 7.0, 1.4 Hz, 1H), 4.97 ꢀ 4.88 (m, 1H), 4.45 (d, J = 5.7 Hz, 2H), 2.02 ꢀ 1.80 (m, 6H), 1.62
13
(dd, J = 9.9, 4.8 Hz, 2H). C NMR (101 MHz, MeOD) δ 157.55, 133.83, 129.10, 118.06, 114.06,
19
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 20 of 42
1
2
3
4
5
6
113.84, 99.75, 57.61, 44.28, 31.27, 24.24. HRMS (ESIꢀTOF) m/z [M + H]+calcd for C17 H19 N5 O2
326.1612 , found 326.1605 .
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4.1.3. 1ꢀcyclopentylꢀ6ꢀ((3ꢀhydroxyꢀ4ꢀmethoxybenzyl)amino)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidin ꢀ4(5H) ꢀ
one (5b)
It was synthesized with 5ꢀ(aminomethyl)ꢀ2ꢀmethoxyphenol as the starting material. Elution with
1
methanol and dichloromethane (1:50) afforded compound 5b (0.19 g, 53%) as a white solid. H NMR
(400 MHz, DMSO) δ 8.93 (s, 1H), 7.74 (d, J = 1.7 Hz, 1H), 6.87 (d, J = 8.2 Hz, 1H), 6.86 ꢀ 6.79 (m,
2H), 6.76 (dd, J = 8.2, 2.1 Hz, 1H), 4.99ꢀ 4.91 (m, 1H), 4.38 (d, J = 5.6 Hz, 2H), 3.74 (s, 3H), 2.08ꢀ 1.79
(m, 8H), 1.70 ꢀ 1.59 (m, 2H). 13C NMR (101 MHz, DMSO) δ 158.25, 154.02, 153.56, 147.32, 146.93,
134.31, 131.84, 118.89, 115.57, 112.62, 100.20, 57.14, 56.16, 44.04, 32.14, 24.78. HRMS (ESIꢀTOF)
m/z [M + H]+calcd for C18 H21 N5 O3 356.1717, found 356.1708.
4.1.4. 1ꢀcyclopentylꢀ6ꢀ((4ꢀhydroxyꢀ3ꢀmethoxybenzyl)amino)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidin ꢀ4(5H)ꢀ
one (5c)
It was synthesized with 4ꢀ(aminomethyl)ꢀ2ꢀmethoxyphenol as the starting material. Elution with
methanol and dichloromethane (1:50) afforded compound 5c (0.17 g, 49%) as a white solid.1H NMR
(400 MHz, CDCl3) δ 11.01 (s, 1H), 7.52 (s, 1H), 6.88 (d, J = 8.7 Hz, 3H), 6.44 (t, J = 5.3 Hz, 1H), 5.97
(s, 1H), 5.07 ꢀ 5.01 (m, 1H), 4.56 (d, J = 5.3 Hz, 2H), 3.80 (s, 3H), 2.07 (dd, J = 7.2, 5.0 Hz, 4H), 1.98 ꢀ
1.92 (m, 2H). 13C NMR (101 MHz, Acetone) δ 205.31, 205.11, 133.65, 120.66, 114.83, 111.73, 57.19,
55.42, 44.54, 31.74, 29.46, 29.26, 29.07, 28.88, 28.69, 28.50, 28.30, 24.47. HRMS (ESIꢀTOF) m/z [M +
H]+calcd for C18 H21 N5 O3 356.1717, found 356.1717.
4.1.5. (R)ꢀ1ꢀcyclopentylꢀ6ꢀ((1ꢀ(3ꢀhydroxyphenyl)ethyl)amino)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidin ꢀ4(5H)ꢀ
one (5d)
20
ACS Paragon Plus Environment

Page 21 of 42
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
It was synthesized with (R)ꢀ3ꢀ(1ꢀaminoethyl)phenol as the starting material. Elution with methanol
and dichloromethane (1:50) afforded compound 5d (0.27 g, 68%) as a white solid.1H NMR (400 MHz,
CDCl3) δ 7.76 (s, 1H), 7.14 (d, J = 7.8 Hz, 1H), 7.05 (s, 1H), 6.94 (s, 1H), 6.87 (d, J = 7.5 Hz, 1H), 6.75
(d, J = 7.9 Hz, 1H), 5.20 ꢀ 5.11 (m, 1H), 4.90 (s, 1H), 2.06 ꢀ 2.00 (m, 3H), 1.95 (dd, J = 13.1, 6.3 Hz,
3H), 1.65 (d, J = 6.3 Hz, 2H), 1.52 (d, J = 6.8 Hz, 3H). 13C NMR (101 MHz, MeOD) δ 161.19, 158.72,
155.59, 153.63, 147.00, 135.23, 130.58, 118.11, 115.02, 114.05, 101.04, 59.39, 52.06, 32.66, 32.44,
25.63, 25.59, 22.85. HRMS (ESIꢀTOF) m/z [M + H]+ calcd for C18 H21 N5 O2 340.1768, found
340.1763.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4.1.6. (S)ꢀ1ꢀcyclopentylꢀ6ꢀ((1ꢀ(3ꢀhydroxyphenyl)ethyl)amino)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidin ꢀ4(5H)ꢀ
one (5e)
It was synthesized with (S)ꢀ3ꢀ(1ꢀaminoethyl)phenol as the starting material. Elution with methanol and
dichloromethane (1:50) afforded compound 5e (0.17 g, 49%) as a white solid.1H NMR (400 MHz,
CDCl3) δ 7.72 (s, 1H), 7.13 (d, J = 7.5 Hz, 1H), 7.04 (dd, J = 17.8, 9.9 Hz, 2H), 6.82 (d, J = 7.6 Hz, 1H),
6.73 (dd, J = 8.0, 1.7 Hz, 1H), 5.08 (s, 1H), 4.90 ꢀ 4.83 (m, 1H), 2.00 ꢀ 1.90 (m, 4H), 1.87 ꢀ 1.80 (m,
2H), 1.60 (dd, J = 6.5, 5.2 Hz, 2H), 1.47 (d, J = 6.9 Hz, 3H). 13C NMR (101 MHz, CDCl3) δ 161.06,
156.94, 154.17, 151.69, 144.99, 133.98, 129.89, 117.90, 114.89, 112.26, 99.49, 58.00, 49.81, 31.69,
31.46, 29.32, 24.67, 24.60, 21.77. HRMS (ESIꢀTOF) m/z [M + H]+calcd for C18 H21 N5 O2 340.1768,
found 340.1761.
4.1.7. General for the preparation of compound 6aꢀ6f.
To the solution of compound 5 (1.0 mmol) in DMF (15 mL), sodium hydride (4.8 mg, 2.0 mmol) and
N,Nꢀethyl(methyl)carbamic chloride (0.25 g, 2.0 mmol) were added subsequently. The mixture was
stirred at room temperature for 1h. The mixture was then extracted with dichloromethane three times.
The combined organic layer was washed with water three times, dried over sodium sulfate and
21
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 22 of 42
1
2
3
4
evaporated to afford the crude product. The crude product was purified by column chromatography to
5
6
afford the compound 6 as a colorless solid.
7
8
9
4.1.8. 3ꢀ(((1ꢀcyclopentylꢀ4ꢀoxoꢀ4, 5ꢀdihydroꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ6ꢀyl)amino)methyl)
phenyl ethyl(methyl)carbamate (6a)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
It was synthesized with compound 5a as the starting material. Elution with methanol and
dichloromethane (1:100) afforded compound 6a (0.30 g, 73%) as a white solid.1H NMR (400 MHz,
CDCl3) δ 10.47 (d, J = 14.6 Hz, 1H), 7.82 ꢀ 7.72 (m, 1H), 7.32 (t, J = 7.8 Hz, 1H), 7.19 ꢀ 7.09 (m, 2H),
7.01 (d, J = 5.1 Hz, 1H), 6.43 (d, J = 46.7 Hz, 1H), 4.98 (p, J = 7.6 Hz, 1H), 4.56 (d, J = 5.5 Hz, 2H),
3.44 (dq, J = 26.6, 7.1 Hz, 2H), 3.00 (t, J = 20.7 Hz, 3H), 2.09 ꢀ 2.01 (m, 4H), 1.94 ꢀ 1.88 (m, 2H), 1.79
(dd, J = 10.8, 5.6 Hz, 2H), 1.68 (dt, J = 8.9, 3.9 Hz, 2H), 1.27 ꢀ 1.17 (m, 3H). 13C NMR (101 MHz,
CDCl3) δ 160.07, 152.68, 151.63, 140.12, 134.39, 129.47, 121.01, 114.29, 100.10, 57.47, 44.51, 31.92,
24.75, 13.27, 12.57. HRMS (ESIꢀTOF) m/z [M + H]+calcd for C21 H26 N6 O3 411.2139, found
411.2146.
4.1.9. 5ꢀ(((1ꢀcyclopentylꢀ4ꢀoxoꢀ4, 5ꢀdihydroꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ6ꢀyl)amino)methyl) ꢀ2ꢀ
methoxyphenyl ethyl(methyl)carbamate (6b)
It was synthesized with compound 5b as the starting material. Elution with methanol and
1
dichloromethane (1:100) afforded compound 6b (0.29 g, 66%) as a white solid. H NMR (400 MHz,
CDCl3) δ 10.20 (s, 1H), 8.01 (s, 1H), 7.81 (s, 1H), 7.11 (d, J = 7.3 Hz, 2H), 6.88 (d, J = 8.5 Hz, 1H),
6.41 (s, 1H), 5.00 (p, J = 7.4 Hz, 1H), 4.47 (d, J = 5.3 Hz, 2H), 3.80 (s, 3H), 3.48 (dd, J = 14.0, 7.0 Hz,
1H), 3.39 (dd, J = 14.0, 6.9 Hz, 1H), 2.96 (s, 3H), 2.88 (s, 3H), 2.04 (s, 6H), 1.98 ꢀ 1.87 (m, 3H), 1.73 ꢀ
1.64 (m, 2H), 1.25 (t, J = 6.9 Hz, 3H), 1.17 (dd, J = 15.5, 8.5 Hz, 2H). 13C NMR (101 MHz, CDCl3) δ
159.83, 154.84, 154.66, 154.17, 152.76, 151.00, 140.50, 140.42, 134.39, 130.88, 125.59, 125.53, 122.76,
122.70, 112.40, 112.34, 100.07, 100.01, 99.99, 57.39, 56.04, 44.32, 44.14, 34.43, 33.96, 31.95, 24.84,
22
ACS Paragon Plus Environment

Page 23 of 42
ACS Chemical Neuroscience
1
2
3
4
5
24.77, 12.94, 12.37. HRMS (ESIꢀTOF) m/z [M + H]+calcd for C22 H28 N6 O4 441.2245, found
441.2241.
6
7
8
9
4.1.10. 4ꢀ(((1ꢀcyclopentylꢀ4ꢀoxoꢀ4, 5ꢀdihydroꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ6ꢀyl) amino) methyl) ꢀ2ꢀ
methoxyphenyl ethyl(methyl) carbamate (6c)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
It was synthesized with compound 5c as the starting material. Elution with methanol and
dichloromethane (1:100) afforded compound 6c (0.24 g, 54%) as a white solid.1H NMR (400 MHz,
CDCl3) δ 10.36 (s, 1H), 7.81 (s, 1H), 7.02 (dd, J = 8.0, 4.5 Hz, 1H), 6.84 (s, 1H), 6.80 (d, J = 8.0 Hz,
1H), 6.15 (s, 1H), 5.00 ꢀ 4.95 (m, 1H), 4.50 (d, J = 5.6 Hz, 2H), 3.78 (s, 3H), 3.49 (dt, J = 17.5, 7.1 Hz,
2H), 3.11 (s, 2H), 3.03 (s, 2H), 2.07 ꢀ 2.02 (m, 4H), 1.92 (dd, J = 13.4, 9.3 Hz, 2H), 1.29 ꢀ 1.25 (m, 3H),
1.21 (t, J = 7.1 Hz, 2H).13C NMR (101 MHz, CDCl3) δ 159.95, 154.15, 152.79, 151.67, 139.68, 136.86,
134.26, 123.14, 119.43, 111.60, 100.09, 99.98, 57.52, 55.95, 44.59, 44.31, 34.43, 31.93, 24.74, 12.90,
12.34. HRMS (ESIꢀTOF) m/z [M + H]+calcd for C22 H28 N6 O4 441.2245, found 441.2246.
4.1.11. (R)ꢀ3ꢀ(1ꢀ((1ꢀcyclopentylꢀ4ꢀoxoꢀ4, 5ꢀdihydroꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ6ꢀyl) amino)
ethyl) phenyl ethyl(methyl)carbamate (6d)
It was synthesized with compound 5d as the starting material. Elution with methanol and
dichloromethane (1:100) afforded compound 6d (0.24 g, 56%) as a white solid.1H NMR (400 MHz,
CDCl3) δ 10.40 (s, 1H), 7.77 ꢀ 7.74 (m, 1H), 7.30 (t, J = 7.8 Hz, 1H), 7.19 (dd, J = 11.0, 5.7 Hz, 2H),
7.00 (s, 1H), 6.72 (d, J = 8.7 Hz, 1H), 5.19 (p, J = 6.6 Hz, 1H), 4.95 (p, J = 7.6 Hz, 1H), 3.41 (dq, J =
25.0, 7.1 Hz, 2H), 3.00 (d, J = 32.2 Hz, 3H), 2.10 ꢀ 1.98 (m, 5H), 1.98 ꢀ 1.88 (m, 3H), 1.73 ꢀ 1.63 (m,
2H), 1.59 (d, J = 6.9 Hz, 3H), 1.19 (dt, J = 16.9, 7.1 Hz, 3H). 13C NMR (101 MHz, CDCl3) δ 160.14,
154.24, 151.91, 151.61, 134.20, 129.40, 123.03, 120.70, 119.75, 100.01, 57.69, 50.23, 44.13, 34.28,
33.84, 31.90, 31.76, 24.77, 24.75, 22.25, 13.21, 12.47. HRMS (ESIꢀTOF) m/z [M + H]+calcd for C22
H28 N6 O3 425.2296, found 425.2298.
23
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 24 of 42
1
2
3
4
5
6
4.1.12.
(S)ꢀ3ꢀ(1ꢀ((1ꢀcyclopentylꢀ4ꢀoxoꢀ4,
5ꢀdihydroꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ6ꢀyl)amino)
7
8
ethyl)phenyl ethyl(methyl) carbamate(6e)
9
It was synthesized with compound 5e as the starting material. Elution with methanol and
dichloromethane (1:100) afforded compound 6e (0.262 g, 62%) as a white solid.1H NMR (400 MHz,
CDCl3) δ 7.81 (s, 1H), 7.34 (d, J = 7.9 Hz, 1H), 7.25 ꢀ 7.19 (m, 2H), 7.03 (s, 1H), 6.56 (d, J = 6.7 Hz,
1H), 5.25 ꢀ 5.18 (m, 1H), 4.98 (dd, J = 15.2, 7.6 Hz, 1H), 3.50 ꢀ 3.38 (m, 2H), 3.07 (s, 1H), 2.98 (s, 2H),
2.12 ꢀ 2.02 (m, 4H), 1.97 ꢀ 1.90 (m, 2H), 1.69 (td, J = 6.6, 3.2 Hz, 2H), 1.61 (d, J = 6.9 Hz, 3H), 1.27 ꢀ
1.18 (m, 3H). 13C NMR (101 MHz, CDCl3) δ 151.84, 134.26, 77.33, 77.01, 76.69, 57.70, 50.28, 31.89,
31.77, 24.74, 22.14. HRMS (ESIꢀTOF) m/z [M + H]+calcd for C22 H28 N6 O3 425.2296, found 425.2299.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4.1.13. 4ꢀ(((1ꢀcyclopentylꢀ4ꢀoxoꢀ4,5ꢀdihydroꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ6ꢀyl)amino) methyl) ꢀ 2ꢀ
methoxyphenyl dimethylcarbamate (6f)
It was synthesized with compound 5c as the starting material. Elution with methanol and
dichloromethane (1:100) afforded compound 6f (0.242 g, 57%) as a white solid.1H NMR (400 MHz,
CDCl3) δ 10.38 (s, 1H), 7.75 (s, 1H), 7.01 (d, J = 8.0 Hz, 1H), 6.87 (d, J = 1.7 Hz, 1H), 6.82 (dd, J = 8.1,
1.8 Hz, 1H), 6.41 (t, J = 5.6 Hz, 1H), 4.98 (p, J = 7.6 Hz, 1H), 4.51 (d, J = 5.6 Hz, 2H), 3.77 (s, 3H),
3.13 (s, 3H), 3.02 (s, 3H), 2.05 (dd, J = 13.0, 5.9 Hz, 4H), 1.96 ꢀ 1.86 (m, 2H), 1.70 ꢀ 1.61 (m, 2H). 13C
NMR (101 MHz, CDCl3) δ 159.94, 155.23, 154.12, 152.78, 151.63, 139.53, 136.95, 134.34, 123.16,
119.33, 111.40, 100.10, 57.42, 55.95, 44.53, 36.94, 36.63, 31.98, 24.78. HRMS (ESIꢀTOF) m/z [M +
H]+calcd for C21 H26 N6 O4 427.2088, found 427.2093.
4.1.14. General for the preparation of compound 13aꢀ13b.
24
ACS Paragon Plus Environment

Page 25 of 42
ACS Chemical Neuroscience
1
2
3
To a solution of intermediated 12 (0.12 g, 0.5 mmol) in DMF(5.0 mL), sodium borohydride (34 mg,
4
5
6
o
0.8 mmol) and aldehyde (0.8 mmol) were added at 0 C. The mixture was stirred for 30 min at room
7
8
9
temperature. Water (15 mL) was added. The mixture was extracted with ethyl acetate three times. The
combined organic layer was washed with brine, dried over anhydrous sodium sulfate, and evaporated to
give crude product. The crude product was purified by column chromatography to afford the compound
13 as a colorless solid.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4.1.15.
(R)ꢀ5ꢀ(((1ꢀ(1ꢀcyclopentylꢀ4ꢀoxoꢀ4,5ꢀdihydroꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ6ꢀyl)ethyl)
amino)methyl)ꢀ2ꢀmethoxyphenyl dimethylcarbamate (13a)
It was synthesized with 4ꢀformylꢀ2ꢀmethoxyphenyl dimethylcarbamate as the starting material.
Elution with methanol and dichloromethane (1:100) afforded compound 13a (0.22 g, 48%) as a white
solid. 1H NMR (400 MHz, CDCl3) δ 10.17 (s, 1H), 8.08 (s, 1H), 7.06 (d, J = 7.9 Hz, 1H), 6.93 ꢀ 6.85 (m,
2H), 5.25 ꢀ 5.13 (m, 1H), 3.87 (s, 3H), 3.81 ꢀ 3.68 (m, 2H), 3.14 (s, 3H), 3.03 (s, 3H), 2.12 (dd, J = 12.0,
6.2 Hz, 4H), 2.00 (dd, J = 13.3, 9.7 Hz, 2H), 1.77 ꢀ 1.72 (m, 2H), 1.48 (d, J = 6.8 Hz, 3H). 13C NMR
(101 MHz, CDCl3) δ 161.67, 158.35, 154.73, 152.40, 151.80, 139.88, 137.06, 134.57, 123.35, 120.14,
112.25, 104.98, 57.79, 56.27, 56.04, 52.15, 36.80, 36.56, 32.47, 32.37, 24.75, 21.54. HRMS (ESIꢀTOF)
m/z [M + H]+calcd for C23 H30 N6 O 4455.2401, found 455.2402.
4.1.16.
(R)ꢀ5ꢀ(((1ꢀ(1ꢀcyclopentylꢀ4ꢀoxoꢀ4,5ꢀdihydroꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ6ꢀyl)ethyl)
amino)methyl)ꢀ2ꢀmethoxyphenyl ethyl(methyl)carbamate(13b)
It was synthesized with 4ꢀformylꢀ2ꢀmethoxyphenyl ethyl(methyl)carbamate as the starting material.
Elution with methanol and dichloromethane (1:100) afforded compound 13b (0.18 g, 39%) as a white
solid. 1H NMR (400 MHz, CDCl3) δ 10.29 (s, 1H), 8.08 (s, 1H), 7.05 (d, J = 7.9 Hz, 1H), 6.87 (dd, J =
8.9, 7.4 Hz, 2H), 5.18 (p, J = 7.5 Hz, 1H), 3.85 (s, 3H), 3.81 ꢀ 3.66 (m, 2H), 3.46 (dd, J = 32.5, 6.3 Hz,
2H), 3.05 (d, J = 38.6 Hz, 3H), 2.18 ꢀ 2.09 (m, 4H), 2.00 (dd, J = 12.6, 9.7 Hz, 2H), 1.78 ꢀ 1.72 (m, 2H),
25
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 26 of 42
1
2
1.47 (d, J = 6.8 Hz, 3H), 1.24 (d, J = 27.2 Hz, 3H). 13C NMR (101 MHz, CDCl3) δ 161.84, 158.62,
154.49, 154.31, 152.38, 151.76, 139.89, 139.80, 137.13, 134.54, 123.30, 120.13, 112.29, 112.21, 104.91,
57.76, 56.24, 55.94, 52.00, 44.22, 44.13, 34.33, 33.89, 32.45, 32.36, 26.89, 24.80, 24.74, 24.69, 21.48,
12.96, 12.40. HRMS (ESIꢀTOF) m/z [M + H]+calcd forC24 H32 N6 O4 469.2558, found 469.2560.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4.2. In vitro bioassay test for the inhibition of PDE9
The PDE9A2 protein was purified by the protocols according to our previous reports.25 3HꢀcGMP as
the substrate for the biological test against PDE9A2. 3HꢀcGMP was diluted with the assay buffer which
contained 20ꢀ50 mM TrisꢀHCl (pH 8.0),10 mM MgCl2 and 1 mM DTT to 20 000ꢀ30 000 cpm per assay.
o
And then, the reaction was performed at 25 C for 15 min, which was terminated by the addition of 0.2
3
M ZnSO4. After the addition of 0.2 N Ba(OH)2, a precipitate was formed and the unreacted HꢀcGMP
left in the supernatant. The radioactivity in the supernatant was measured in 2.5 mL of Ultima Gold
liquid scintillation cocktails (PerkinElmer) using a PerkinElmer 2910 liquid scintillation counter. For the
measurement of IC50 of inhibitors, eight different concentrations were used and each measurement was
repeated three times. The IC50 values against PDE9 were calculated by nonlinear regression. The mean
values of the measurements were considered as the final IC50 values with the SD values of the
measurements.
4.3. AChE/BuChE inhibition activity
The inhibition activity of designed compounds on eeAChE and eqBuChE were tested using the the
Ellman’s method.32 Compounds were diluted using the enzyme buffer (0.1 M KH2PO4/K2HPO4 buffer,
pH 8.0). The final concentration of 5, 5’ ꢀdithiobis (2ꢀnitrobenzoic acid) was 0.25 M. The mixture was
o
incubated with 0.25 unit of AChE at 37 C for 15 min. After mixing with ATC (0.25 M), the mixture
was scanned immediately at 412 nm using Thermo Scientific VarioskanFlash. The average absorbance
variation for 5 consecutive minutes was calculated. The inhibitory rate for each compound was
26
ACS Paragon Plus Environment

Page 27 of 42
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
calculated by comparing the average absorbance of samples containing compounds with the absence of
test compounds. The doseꢀresponse curve and concentration for inhibit 50% enzyme activity (IC50) was
calculated in GraphPad version 5 by fitting replicates with the lg(inhibitor) vs normalized response
variable slope equation. The BuChE test was performed in the same procedure with BTC as the
substrate for BuChE
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4.4. The ORAC test for the antioxidant activities
The antioxidant activity was tested using the modified oxygen radical absorbance capacity fluorescein
(ORACꢀFL) method 42,43. Each assays was diluted with phosphate buffer (75 mM, pH 7.4) to make the
final reaction mixture 200 ꢁL. Test compound (20 ꢁL) and fluorescein (120 ꢁL, 150 nM final
concentration) were placed in the wells of a black 96ꢀwell plate. The mixture was then incubated at 37
oC for 15 min. AAPH solution (60 ꢁL, 12 mM) was added rapidly. The fluorescence of this plate was
recorded every minute for 4h with excitation at 485 nm and emission at 535 nm in a Spectrafluor Plus
plate reader (Tecan, Crailsheim, Germany). Trolox was used as standard. All the reaction mixture was
prepared at least three independent assays were performed. The ORACꢀFL values were expressed as
Trolox equivalents.
4.4. In vitro cell viability assay
SHꢀSY5Y cells was used for the cytotoxicity test by using the MTT (3ꢀ(4, 5ꢀdimethylꢀ2ꢀ thiazolyl)ꢀ
2,5ꢀdiphenyl ꢀ2ꢀHꢀtetrazolium bromide) method.44 The SHꢀSY5Y cells (5×103) were inoculated into a
96ꢀwell plate. After 24 h, compounds at the concentration of 10 ꢁM, 20 ꢁM, 40 ꢁM, 60 ꢁM, 80 ꢁM and
100 ꢁM, were added. After incubation for 48 h, MTT solution was added to each cell well and
o
inoculated for 4 h at 37 C. After the purple formazan dye was dissolved, 100 ꢁL DMSO was added to
the cells. The absorbance of each cell well was measured at 570 nm using Flex Station 3 (Molecular
Devices).
4.5. Molecular docking and MD simulations for PDE9A
27
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 28 of 42
1
2
The Xꢀray crystal structure of 28ꢀPDE9A complex (PDB ID: 4GH6) reported in our previous studies25
was selected for molecular modeling. Surflexꢀdock45 embedded in the software Tripos Sybyl 2.0 was
used. Two metal ions crucial for the PDE’s catalytic activity in the catalytic domain and water
molecules coordinate these two metal ions were retained. Hydrogen atoms were added, and the ionizable
residues were protonated at the neutral pH. The protomol, was generated using the parameters by
default. The parameters of proto_thresh and proto_bloat were assigned 0.5 and 0, respectively. After the
protomol was the prepared, molecular docking was performed for test molecules.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
After molecular docking was completed, MD simulations were used to equilibrate the whole system. 8
ns MD simulations were performed after minimization, heating and equilibration steps of each system.
The 100 snapshots were isolated from the final 1.0 ns period of the MD simulation trajectories, and then
used for bindingꢀfreeꢀenergy calculations by the MMꢀPBSA approach46ꢀ48
.
The PBSA program in the Amber 1649, which could numerically solves the PoissoneBoltzmann
equations, was used to calculate the electrostatic contribution to the solvationꢀfree energy. MMꢀPBSA
binding free energy (ꢂGbind) calculations were performed by the extraction of 100 snapshots of the last 1
ns trajectories by equation 1 using default parameters. Gcomplex, Grec and Glig represent the free energies
of complex, receptor and ligand, respectively. Each free energy was evaluated by the sum of the MM
energy EMM, the solvation free energy Gsolv, and the entropy contribution S, respectively, leading to
equation 2.
ꢂGbind = Gcomplex ꢀ Grecꢀ Glig
ꢂGbind = ꢂEMM + ꢂGsolvꢀ TꢂS
(1)
(2)
ꢂEMM is the gas phase interaction energy and can be decomposed into EMM, comp, EMM, rec and EMM, lig
.
Solvation free energy is evaluated by the sum of the electrostatic solvation free energy and nonpolar
solvation free energy. The electrostatic solvation free energy, ꢂGPB, was solved by the Poisson
28
ACS Paragon Plus Environment

Page 29 of 42
ACS Chemical Neuroscience
1
2
3
4
5
Boltzmann (PB) equation, while the nonpolar solvation free energy varies proportionally with the
solvation accessible surface area (SASA), leading to by equation 3 and 4
6
7
8
9
ꢂGsolv = ꢂGPB + ꢂGnp
ꢂGnp = γ SASA + b
(3)
(4)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Although entropy contribution term can be evaluated by the normal mode analysis, the process is
extremely timeꢀconsuming for large proteinꢀligand systems. In order to get the compromise between
efficiency and accuracy, entropy contribution term was omitted for ꢂGbind in equation (2). The default
parameters were adopted, with γ = 0.0072 kcal/(Å2) and b = 0 kcal/mol.The charges of Mg2+ and Zn2+
were assigned 2.0 for PB calculations, and their bond raddi were used for SA calculations.
4.6 Molecular docking and MD simulations for AChE and BuChE
The crystal structures of human AChE and BuChE with bound inhibitors (PDB ID: 4EY7 and
4TPK) were used for molecular docking, respectively. The residues of each receptor was protonated in
pH of 7.0 and atoms’ partial charges were assigned by the CHARMm force field. The docking site was
defined as spheres with a 8 Å radius covering the binding site. CDOCKER protocol embedded in
Discovery Studio 2.5 was utilized for docking of 6f and 13a to AChE and BuChE. The other docking
parameters were set by default. Twenty topꢀranked docking poses were retained and carefully analyzed.
The most appropriate pose was retained and was subjected to MD simulations. 8 ns MD simulations
were performed and 100 snapshots were isolated from the final 2 ns of the trajectory for the following
bindingꢀfreeꢀenergy calculations using the MMꢀPBSA method in the Amber 16 with the same procedure
in 4.5.
29
ACS Paragon Plus Environment