The Magic of Crystal Structure-Based Inhibitor Optimization: Development of a Butyrylcholinesterase Inhibitor with Picomolar Affinity and in Vivo Activity

Article abstract of DOI:10.1021/acs.jmedchem.7b01086

The enzymatic activity of butyrylcholinesterase (BChE) in the brain increases with the progression of Alzheimer's disease, thus classifying BChE as a promising drug target in advanced Alzheimer's disease. We used structure-based drug discovery approaches to develop potent, selective, and reversible human BChE inhibitors. The most potent, compound 3, had a picomolar inhibition constant versus BChE due to strong cation-π interactions, as revealed by the solved crystal structure of its complex with human BChE. Additionally, compound 3 inhibits BChE ex vivo and is noncytotoxic. In vitro pharmacokinetic experiments show that compound 3 is highly protein bound, highly permeable, and metabolically stable. Finally, compound 3 crosses the blood-brain barrier, and it improves memory, cognitive functions, and learning abilities of mice in a scopolamine model of dementia. Compound 3 is thus a promising advanced lead compound for the development of drugs for alleviating symptoms of cholinergic hypofunction in patients with advanced Alzheimer's disease.

Full text of DOI:10.1021/acs.jmedchem.7b01086

Subscriber access provided by READING UNIV
Article
The Magic of Crystal Structure-Based Inhibitor Optimization: Development of
a Butyrylcholinesterase Inhibitor with Picomolar Affinity and In Vivo Activity
Urban Košak, Boris Brus, Damijan Knez, Simon Žakelj, Jurij Trontelj, Anja Pišlar, Roman Šink,
Marko Juki#, Marko Živin, Adrian Podkowa, Florian Nachon, Xavier Brazzolotto, Jure Stojan,
Janko Kos, Nicolas Coquelle, Kinga Salat, Jacques-Philippe Colletier, and Stanislav Gobec
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b01086 • Publication Date (Web): 11 Dec 2017
Downloaded from http://pubs.acs.org on December 11, 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.
Journal of Medicinal Chemistry 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 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The Magic of Crystal StructureꢀBased Inhibitor
Optimization: Development of a
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Butyrylcholinesterase Inhibitor with Picomolar
Affinity and In Vivo Activity
‡
‡
‡
‡
‡
‡
Urban Košak, Boris Brus, Damijan Knez, Simon Žakelj, Jurij Trontelj, Anja Pišlar, Roman
‡
‡
▲
∆
ø
ø
Šink, Marko Jukič, Marko Živin, Adrian Podkowa, Florian Nachon, Xavier Brazzolotto,
■
‡
§
∆
§
Jure Stojan, Janko Kos, Nicolas Coquelle, Kinga Sałat, JacquesꢀPhilippe Colletier and
‡
,
Stanislav Gobec *
‡
Faculty of Pharmacy, University of Ljubljana, Aškerčeva 7, 1000 Ljubljana, Slovenia
▲
Institute of Pathological Physiology, Faculty of Medicine, University of Ljubljana, Vrazov trg
2
, 1000 Ljubljana, Slovenia
ø
Institut de Recherche Biomédicale des Armées, 91223 Brétigny sur Orge, France
Faculty of Pharmacy, Jagiellonian University, Medyczna 9 St., 30ꢀ688 Krakow, Poland
∆
■
Institute of Biochemistry, Faculty of Medicine, University of Ljubljana, Vrazov trg 2, 1000
Ljubljana, Slovenia
§
University Grenoble Alpes, and CNRS and CEA, IBS, Fꢀ38044 Grenoble, France
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Abstract
The enzymatic activity of butyrylcholinesterase (BChE) in the brain increases with the
progression of Alzheimer’s disease, thus classifying BChE as a promising drug target in
advanced Alzheimer’s disease. We used structureꢀbased drugꢀdiscovery approaches to develop
potent, selective and reversible human BChE inhibitors. The most potent, compound 3, had a
picomolar inhibition constant versus BChE due to strong cationꢀπ interactions, as revealed by the
solved crystal structure of its complex with human BChE. Additionally, compound 3 inhibits
BChE ex vivo, and is nonꢀcytotoxic. In vitro pharmacokinetic experiments show that compound 3
is highly protein bound, highly permeable, and metabolically stable. Finally, compound 3 crosses
the bloodꢀbrain barrier, and it improves memory, cognitive functions, and learning abilities of
mice in a scopolamine model of dementia. Compound 3 is thus a promising advanced lead
compound for the development of drugs for alleviating symptoms of cholinergic hypofunction in
patients with advanced Alzheimer’s disease.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Page 3 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Introduction
Alzheimer’s disease (AD) is a severe and irreversible progressive neurodegenerative disorder of
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
the central nervous system. Given the aging population in Western countries, the number of
people who suffer from AD is expected to escalate in coming years, with AD prevalence
2
predicted to reach over 100 million by the year 2050. AD is characterized by gradual
deterioration of cognitive abilities, accompanied by behavior that eventually interferes with the
customary activities of daily living. Hallmarks of AD include aggregation and deposition of the
3
4,5
6
protein tau and amyloidꢀβ (Aβ), oxidative stress, and death of cholinergic neurons in the basal
forebrain. This results in reduced levels of the neurotransmitter acetylcholine (ACh) in the
cerebral cortex and other brain areas. Taken together, severe memory and learning deficits are
7–9
associated with AD.
At the neuronal level, cholinergic neurotransmission is terminated by ACh hydrolysis
through two types of cholinesterases (ChEs): acetylcholinesterase (AChE) and
10
butyrylcholinesterase (BChE). The AChE activity is dominant in the healthy brain (80%),
where it catalyzes the hydrolysis of ACh to acetate and choline, while the BChE activity was
1
1
initially considered to be nonꢀessential for full functioning of the central nervous system.
Current treatments of patients with AD focus on drugs that can increase cholinergic
1
2
13
14
neurotransmission by inhibiting the ChEs, which include donepezil, rivastigmine, and the
1
5
alkaloid galanthamine (Figure 1A).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Structures and IC50 values versus the cholinesterases. (A) Donepezil, rivastigmine, and
galantamine. (B) Selective BChE inhibitors, as hit compound 1 and sulfonamide 2.
BChE activity is mainly associated with glial cells, as well as with vascular and neuronal
16–18
structures, with BChEꢀpositive neurons defined for the thalamus, neocortex, and amygdala.
As BChE can efficiently hydrolyze the ACh analog acetylthiocholine and can rescue
cholinesterase function in the AChE nullizygote mouse, BChE might act as a coꢀregulator of the
1
9–23
cholinergic activity in the central nervous system.
Indeed, the levels of BChE in certain parts
of the brain show significant increases in the late stages of AD, with the ratio between BChE and
AChE (BChE/AChE) increasing from 0.2 up to as much as 11.0 in certain parts of the brain; this
ACS Paragon Plus Environment

Page 5 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
implies a shift from a supportive role to a leading role for BChE in the hydrolysis of the excess
24–27
ACh.
Due to this altered BChE/AChE ratio and the demonstration of promising procognitive
2
8,29
effects of BChE inhibition in vivo using selective BChE inhibitors,
targeting BChE has
2
8–31
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
become a viable alternative for the treatment of patients with mildꢀtoꢀlate stages of AD.
Furthermore, the elevation of depleted ACh levels and the increase in the cognitive functions
promoted by selective BChE inhibitors in vivo is not accompanied by the peripheral
2
9,31
32–
(parasympathomimetic) adverse effects
that are known to limit the use of AChE inhibitors.
34
Selective inhibition of brain BChE thus represents a useful strategy to ameliorate the
cholinergic deficit and improve the cognitive performance of patients in the late stages of AD.
Using a hierarchical virtual screening protocol, we recently defined hit compound 1
(Figure 1B) as a potent, selective and reversible slowꢀtight–binding inhibitor of human BChE
35
(
huBChE). We then used the structure of hit compound 1 to develop sulfonamide 2 (Figure 1B),
29
which is the only in vivo–active noncovalent BChE inhibitor among the large number of
2
8,29,36–41
selective BChE inhibitors to have been reported.
However, the crystal structure of
huBChE in complex with hit compound 1 (PDB ID 4TPK) revealed several additional
opportunities for structural modifications of compound 1 to further improve its huBChE
inhibitory potency.
Here, we report the design, synthesis and biochemical evaluation of a focused library of
amide derivatives of hit compound 1. The most potent of these, compound 3, shows selective
reversible huBChE inhibition in the picomolar concentration range. The crystal structure of its
complex with the target enzyme (i.e., huBChE) reveals that two strong cationꢀπ interactions are
the basis of its greatly increased inhibitory potency. Additionally, compound 3 improves learning
abilities, memory, and cognitive functions of mice in a model of the cholinergic deficit that
characterizes AD, without producing any acute cholinergic adverse effects.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Results and Discussion
Design of the Inhibitors
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Based on the chemical structure of hit compound 1, class I–VII inhibitors were initially
designed (Figure 2), with specific attention given to the production of inhibitors with drugꢀlike
4
2
properties. The class I inhibitors were designed to investigate the importance of the
naphthalene moiety of hit compound 1. This moiety was replaced with other aromatic groups,
including smaller substituted benzene rings, fiveꢀ and sixꢀmembered heterocycles, and 6ꢀ
substituted naphthalenes. Simultaneous modification of the naphthalene moiety and removal of
the ꢀ(CH ) OMe group from the amide nitrogen produced the class II inhibitors. In the class III
2
2
inhibitors, the 2,3ꢀdihydroꢀ1Hꢀinden moiety was replaced with a benzyl group, and the 2ꢀ
naphthalene was either retained or replaced with modified naphthalenes. The class IV inhibitor
was designed to examine the impact of simultaneous replacement of the 2,3ꢀdihydroꢀ1Hꢀinden
moiety with a benzyl group and of the removal of the ꢀ(CH ) OMe group. Benzoxazoleꢀ or
2
2
benzothiazoleꢀcontaining compounds have been shown to bind to Aβ and to inhibit its
4
3
aggregation. Potential multifunctional compounds combining BChE inhibition and antiꢀAβ
aggregation properties were therefore designed here by replacing the 2,3ꢀdihydroꢀ1Hꢀinden
moiety of hit compound 1 with benzoxazole and benzothiazole in the class V inhibitors. To
investigate the effects of the length of the Nꢀalkyl chain attached to the amide nitrogen in
compound 1, compounds with an extended Nꢀalkyl chain were also designed, as the class VI
inhibitors. Finally, the class VII inhibitors are 1,4ꢀdisubstituted piperidines, which were designed
to determine the influence of the piperidine ring disubstitution pattern (Figure 2).
ACS Paragon Plus Environment

Page 7 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Design of the class I–VII inhibitors. The modifications introduced during each design step are
indicated in red.
Compared to hit compound 1, all of these class I–VII inhibitors were less potent huBChE
inhibitors (Supporting Table S1). This thus called for rational structureꢀbased drug design, which
yielded the class VIII inhibitors (Figure 3). The crystal structure of the complex of compound 1
with huBChE (PDB ID 4TPK) had indeed revealed that the ꢀ(CH ) OMe group of compound 1 is
2
2
oriented toward the catalytic residues (i.e., Ser198, His438) and Trp82 in the cholineꢀbinding
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
5
pocket, although it did not appear to interact specifically with any of these residue (Figure 3A).
We hypothesized that the simple replacement of the ꢀOMe group in compound 1 by ꢀNMe might
2
improve the potency by contributing an additional cationꢀπ interaction with Trp82. This Trp82
4
4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
indeed stabilizes the natural substrate (and other ligands) through such interactions. To enable
these interactions with Trp82, the ꢀNMe group was used as the simplest basic center. However,
2
compounds with an additional methylene group in the side chain [e.g., ꢀ(CH ) NMe ] were also
2
3
2
designed (Figure 3B), to determine the optimal length between the amide nitrogen and the ꢀNMe2
group for optimal huBChE inhibition. Due to the limited size of the cholineꢀbinding pocket,
further prolongation of the side chain did not appear reasonable.
Synthesis of the Inhibitors
The synthesis of class VIII inhibitors is shown in Figure 3. Nipecotic acid (4) was converted into
4
5
1
1
benzoyl amide 5, which was then reacted with N ,N ꢀdimethylethaneꢀ1,2ꢀdiamine in the
1
1
presence of TBTU to produce amide 6. Compound 5 was also reacted with N ,N ꢀ
dimethylpropaneꢀ1,3ꢀdiamine in the presence of TBTU to produce amide 7. LiAlH was used to
4
reduce both amide groups of compounds 6 and 7 and produce secondary amines 8 and 9, which
was reacted with 2ꢀnaphthoyl chloride to produce amides 10 and 11. Secondary amines 8 and 9
were also converted into orthogonally protected piperidinꢀ3ꢀyl methanamines 12 and 13, which
were debenzylated to produce secondary amines 14 and 15. This secondary amines was reacted
with 1,3ꢀdihydroꢀ2Hꢀindenꢀ2ꢀone in the presence of sodium triacetoxyborohydride
[
NaBH(OAc) ] to produce tertiary amines 16 and 17. TFA was used to remove the tertꢀ
3
butyloxycarbonyl protecting group from compounds 16 and 17 to provide amine tris(2,2,2ꢀ
trifluoroacetates) 18 and 19, which were then reacted with 2ꢀnaphthoyl chloride to produce
ACS Paragon Plus Environment

Page 9 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
amides 3 and 20. Compound 3 was converted into dichloride salt using 2 M solution of HCl in
Et O.
2
Tertꢀbutyl (2ꢀhydroxyethyl)(methyl)carbamate (21) was converted into mesylate 22,
which then was used to alkylate amide 23 to produce compound 24. Concentrated aqueous HCl
in EtOH was used to remove the tertꢀbutyloxycarbonyl protecting group from compound 24 to
provide amide 25 dichloride. Details of the synthesis of class IꢀVII inhibitors are given in
Supporting Figures S1 to S7 and Supporting Discussion.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
O
O
H
N
a
N
b
N
c
N
H
H
OH
OH
N
N
R
R
O
O
O
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
5
6: R = (CH ) NMe
: R = (CH ) NMe
8: R = (CH ) NMe
2 2
9: R = (CH ) NMe
2
2
2
2
2
2
7
2
3
2 3
d
N
R
N
e
O
10: R = (CH ) NMe
2
1: R = (CH ) NMe
2
2
1
2 3
2
3
F3CCOO
H
H
N
h
N
g
N
f
N
R
R
N
R
R
N
N
H
H
N
Boc
Boc
Boc
1
1
2 2
8: R = (CH ) NHMe2
16: R = (CH ) NMe
7: R = (CH ) NMe
14: R = (CH ) NMe
15: R = (CH ) NMe
2 2
2
2
12: R = (CH ) NMe
13: R = (CH ) NMe
2 3
2 2
2
2
2 2
2
2
1
2 3
2
3
9: R = (CH ) NHMe
2
2
3
d
2
Cl
H
N
R
N
Me
Me
N
H
O
i
N
3
2
: R = (CH ) NMe
2 2 2
0: R = (CH ) NMe
2
3
2
N
O
j
Me
OH
Me
O
Me
3
dichloride
N
N
S
O O
Boc
Boc
21
22
2
Cl
H
Me
N
Boc
Me
H
N
N
H
N
N
k
N
l
H
N
N
O
O
O
2
3
24
25 dichloride
ACS Paragon Plus Environment

Page 11 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 3. Synthesis of class VIII inhibitors. Conditions and reagents: (a) (i) PhCOCl, K CO , THF–
2
3
H O, 0 °C to rt, 22 h; (ii) 6 M HCl (aq), 0 °C (95%); (b) H Nꢀ(CH ) ꢀNMe2, TBTU, Et N, DCM, rt, 24 h
2
2
2 n
3
(
93% crude); (c) LiAlH , anhydrous THF, rt to reflux, under argon, 3.5 h (67%–78% crude); (d) 2ꢀ
4
naphthoyl chloride, Et N, DCM, 0 °C to rt, 24 h (39%–85%); (e) Boc O, Et N, DCM, 0 °C to rt, 24 h
3
2
3
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
(
41%–55%); (f) cyclohexene, Pd(OH) /C cat., MeOH, rt to reflux, under argon, 24 h (98%–99% crude);
2
g) 1,3ꢀdihydroꢀ2Hꢀindenꢀ2ꢀone, NaBH(OAc) , AcOH, 1,2ꢀdichloroethane, under argon, 24 h (47%–
3
6
9%); (h) TFA, DCM, rt, 24 h (95%–98%); (i) 2 M HCl solution in Et O, MeOH, under argon, 0 °C (15
2
min) to rt (90 min) (98%); (j) MeSO Cl, Et N, DCM, 0 °C to rt, 2 h (97% crude); (k) (i) NaH (60%
2
3
dispersion), dry DMF, 0 °C to rt, 2 h; (ii) 88, rt to 50 °C, 72 h (51 %); (l) conc. aqueous HCl, EtOH, rt to
0 °C, 16 h (70%).
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Structureꢀbased drug design of the class VIII inhibitors. (A) Schematic representation of the
binding mode of hit compound 1 (aquamarine sticks) with huBChE (PDB ID 4TPK). The residues of the
acylꢀbinding pocket are shown as gray sticks. The Tyr332 that can form the crucial cationꢀπ interactions is
shown as green sticks. Possible interactions (hydrogen bonds and/or cationꢀπ interactions) with catalytic
residues (purple sticks) and Trp82 (orange sticks) are shown as yellow dashes. (B) The modifications to
compound 1 that were introduced during this design step are indicated in red.
In vitro Cholinesterase Inhibitory Activities and Structure–Activity Relationships
The synthesized compounds were first evaluated for their relative inhibition of recombinant
huBChE hydrolase activity, using the BChEꢀspecific substrate butyrylthiocholine iodide and the
4
6
method of Ellman. The inhibitory potencies against murine AChE (mAChE) were determined
for all of the synthesized compounds using the specific AChE substrate acetylthiocholine iodide.
mAChE was chosen as the surrogate for human AChE since the distribution of the aminoꢀacid
ACS Paragon Plus Environment

Page 13 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4
7
residues along the active site gorge of these two enzymes is strictly conserved. The IC values
5
0
were determined for all of the compounds that showed at least 50% relative inhibition at 10 µM,
whereas the other compounds were classified as inactive (IC50 >10 µM). Table 1 lists the
structures, inhibitory potencies (i.e., IC ) and selectivities towards huBChE [i.e., IC (mAChE)/
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
50
IC (huBChE)] of the most potent of these class I–VIII inhibitors. Supporting Table S1 provides
5
0
the data that were obtained with all 49 of the amides produced in the present study. The
inhibitory potencies and selectivities toward BChE of initial hit compound 1 and sulfonamide 2
are indicated in Table 1 for comparison.
Altogether, these data highlight the following concepts that are needed to preserve the
huBChE inhibitory activity of these compounds: (i) no modification or replacement of the
naphthaleneꢀ2ꢀyl moiety, or replacement of the 2,3ꢀdihydroꢀ1Hꢀinden moiety with a benzyl
group; (ii) a basic (protonatable) piperidine nitrogen is essential; (iii) a 1,3ꢀdisubstituted
piperidine is superior to a 1,4ꢀdisubsituted analog; and (iv) the ꢀ(CH ) OMe chain can only be
2 2
replaced by a ꢀ(CH ) NMe group, which greatly improves the inhibitory potency. Further details
2
2
2
regarding the structure–activity relationship of these compounds can be found in the Supporting
Discussion.
Compound 3 was the most potent inhibitor of this series, and therefore the only one that
was further evaluated for biological activity. In all of these biological studies, compound 3 was
used in the form of its hydrochloride salt (3 dichloride) (Supporting Figure S8), with the
exception of the enzyme kinetics studies, in which compound 3 was used as its free base.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 1. Structures and inhibitory potencies of hit compound 1, sulfonamide 2, and the most potent
of the type I–VIII inhibitors.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
N°
Inhibitor
class
Piperidine
disubstitution
pattern
Subsitution
X
IC ±SEM (µM)
Selectivity
50
1
2
b
R
R
R
huBChE
mAChE
ratio
1
Initial
hit
1,3
(CH
(CH
2
)
)
2
OMe CO
0.021
102.1
±2.5
4862
±0.002
2
3
/
1,3
1,3
2
2
OMe SO
2
0.0049
>10
>10
>2041
>9709
±
0.0003
VIII
(CH
)
NMe
CO
0.00103
2
2
2
c
±
0.00004
2
2
2
6
7
8
I
1,3
1,3
1,3
(CH
2
)
2
OMe CO
0.410
>10
>10
>10
>24
>149
>43
±
0.053
II
H
CO
0.067
±0.001
III
(CH ) OMe CO
2 2
0.230
±
0.008
ACS Paragon Plus Environment

Page 15 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
29
30
IV
V
1,3
1,3
H
H
CO
CO
1.437
>10
>10
>7
±0.409
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0.632
>16
±
0.016
31
VI
1,3
1,4
(CH
(CH
2
)
)
3
OMe CO
0.03824
>10
>10
>262
±0.00162
3
2
VII
2
2
OMe CO
0.702
>14
8.6
±
0.033
35
Tacrine
/
/
/
/
0.012
0.003
0.106
±0.01
±
a
The values are expressed as mean ± SE of at least two independent experiments.
b
Selectivity ratio: IC (mAChE)/IC (huBChE)
5
0
50
c
IC50 value does not reflect the true affinity due to the concentration of the huBChE surpassing the
concentration of inhibitor in the in vitro assays. The IC value for the tight binder, compound 3, is thus
50
converging to the total enzyme concentration (1 nM).
Chiral HPLC Resolution and Kinetic Studies of Compound 3
Semiꢀpreparative reversedꢀphase chiral HPLC (Supporting Figure S8A; Experimental Section)
was used to resolve compound 3 into its pure enantiomers. The optical purities of the enantiomers
were determined by calculation of the enantiomeric excess (e.e.) using analytical reversedꢀphased
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
chiral HPLC (Supporting Figure S8B, C; Experimental Section) and by the measurement of the
optical rotation (Experimental Section). Both enantiomers were obtained with e.e. of 99%.
The affinity of compound 3 for its target enzyme huBChE was particularly high, such that
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
the value of K was lower than the total concentration of enzyme used in the assay system (i.e., 1
i
nM). This situation is referred to as tightꢀbinding inhibition, and it presents some unique
challenges for the appropriate quantitative assessment of the potency of compound 3. To
determine the actual affinity of the pure enantiomers of compound 3, the enzyme kinetic
experiments were carried out on a stoppedꢀflow apparatus according to Copeland’s guidelines for
4
8
evaluation of tightꢀbinding inhibitors. Briefly, the full progress curves of the formation of the
enzymatic products were followed in the absence and presence of each enantiomer of compound
3
. The curves were analyzed simultaneously with the ENZO application, using a model that also
accounted for the depletion of the free inhibitor concentration due to the formation of the
4
9
enzymeꢀinhibitor complex. Accordingly, a singleꢀstep competitive mechanism with a slow k
on
(Figure 5A) allowed full reproduction of these progress curves. Of note, this mechanism is
48
typical of most tightꢀbinding inhibitors. Curvature in the initial phase of the (+)ꢀ3 inhibited
enzyme reaction was observed, which is indicative of slowꢀonset of inhibition (Figure 5B).
ACS Paragon Plus Environment

Page 17 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. Determination of the inhibition mechanism and K of compound (+)ꢀ3. (A) Competitive
i
singleꢀstep inhibition mechanism with slow association (low k ) that fully reproduced the progress curves
on
obtained. S, substrate; P, product; E, enzyme; EA, acylated enzyme; I, (+)ꢀ3; K, (–)ꢀ3; k0–k5, kinetic
constants. (B) Progress curves for hydrolysis of 50.4 ꢁM butyrylthiocholine iodide by 1.07 nM huBChE in
the absence (no (+)ꢀ3; red curves) and presence of the pure enantiomer of compound (+)ꢀ3 at 0.4 (orange
curves), 0.8 (green curves) and 2 nM (black curves). Data were obtained using a stoppedꢀflow apparatus.
Each concentration was performed in duplicate. Fitted curves for all of the conditions are shown as
smooth lines.
A more than 50ꢀfold difference in the huBChE inhibitory potencies was seen between the pure
enantiomers, with the eutomer of compound 3, compound (+)ꢀ3, showing a particularly low
inhibition constant (11.1 pM). The determined rates and the calculated inhibition constants for
each of the enantiomers, along with the K and k values for butyrylthiocholine iodide, are
m
cat
8
9
listed in Table 2. The k values of substrates and small inhibitors are in the range of 10 to 10
on
–
1 –1
M s , respectively and do not differ substantially. On the other hand, the k value reflects all
off
interactions between the ligand and the protein, and depends on the availability of the compound,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
in this case the two enantiomers, to realize as many interactions (binding energy) as sterically
possible.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 2. Characteristic kinetic constants for the hydrolysis of butyrylthiocholine iodide (BTCI)
by huBChE, and its inhibition by compounds 1, 3 and bisꢀtacrine at pH 8. The values are
expressed as mean ± SE of at least two independent experiments.
Compound
K (µM) kcat (s )
Kinetic constant
–
1
–1 –1
–1
K (pM)
i
k (M s )
k (s )
m
on
off
BTCI
12
ꢀꢀ
859
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
8
8
–2
–3
(–)ꢀ3
603 ± 35.7
10.8 ± 0.46
1.02 ± 0.02 × 10
1.49 ± 0.02 × 10
6.17 ± 0.22× 10
(
+)ꢀ3
+)ꢀ1
ꢀꢀ
ꢀꢀ
1.60 ± 0.05 × 10
–2
2
.11 ±
5.72 ± 0.03 × 10
3
5
3
(
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
2710 ± 31.4
27050 ± 1604
163 ± 6
7
35
35
0
.01 × 10
1.28 ±
3.45 ±
5
(–)ꢀ1
7 35
–1 35
0
.04 × 10
0.11 × 10
Bis(7)ꢀ
8
ꢀꢀ
1.85 ± 0.04 × 10
0.030 ± 0.0006
5
0,51
tacrine
Crystal Structure of HuBChE in Complex with Compound 3
The coꢀcrystal structure of compound 3 with huBChE was solved at 2.1 Å resolution. The
crystals of huBChE in complex with compound 3 were obtained by coꢀcrystallization methods.
This involved setting hanging drops after preꢀincubation of 500 µM compound 3 dichloride in
1
% dimethylsulfoxide aqueous solution with the enzyme solution (90 µM) for 3 h. Although
ACS Paragon Plus Environment

Page 19 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
racemic compound 3 was used for the experiments, the observed electronic density suggests that
the R isomer was bound in the active site. This assumption is based on the local environment,
which provides the R isomer with an extra hydrogen bond between the piperidine nitrogen of
compound 3 and a surrounding water molecule (127). Examination of the binding mode explains
the exceptional binding affinity of the inhibitor compound 3 towards huBChE (Figure 6). Similar
to the parent hit of compound 1, the naphthalene ring is Tꢀstacked with Trp231 in the acylꢀ
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
binding pocket of huBChE. The fact that this pocket is comparatively smaller in AChE likely
explains why compound 3 displays high selectivity towards huBChE over mAChE (Table 1).
Furthermore, the piperidine nitrogen forms a cationꢀπ interaction with the indole ring of Tyr332.
The ꢀ(CH ) NMe side chain replaces the ꢀ(CH ) OMe chain of hit compound 1, and points
2
2
2
2 2
toward the cholineꢀbinding pocket. The tertiary amine forms a cationꢀπ interaction with Trp82,
with a distance of 3.7 Å to the closest carbon of the indole ring. This additional interaction of
compound 3 can be clearly correlated with its picomolar inhibitory potency. This thus validates
the rational drugꢀdesign approach undertaken on the basis of the previous structural data, which
resulted in this increased inhibitory potency.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. Crystal structure of compound 3 (R isomer) in complex with huBChE (PDB ID 5NN0).
Compound 3 (magenta) and the key residues of huBChE involved in the binding are shown as sticks, as
are the catalytic residues (i.e., His438, Ser198). Note that Ser198 was observed in the crystal in two
alternate conformers. Trp231, Phe398, and Leu286 form the acylꢀbinding pocket, into which the
naphthalene ring of compound 3 fits. Water molecules are presented as red spheres, hydrogen bonds are
black, and cationꢀπ interactions are green dashes between the positively charged nitrogen atom and the
center of the aromatic rings. The piperidine nitrogen forms a cationꢀπ interaction with Tyr 332 (4.6 Å),
while the tertiary amine (ꢀ(CH ) NMe ) of compound 3 is 4.2 Å and 4.4 Å from the benzene and pyrrole
2 2
2
rings of Trp82, respectively.
Thermodynamic and Computational Analysis of Binding
ACS Paragon Plus Environment

Page 21 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Thermodynamics of ligand binding for compounds (+)ꢀ1 and (+)ꢀ3 was investigated by enzyme
kinetic experiments at three different temperatures (Supporting Figure S9), and thermodynamic
parameters were calculated from Arrenhius plots (Supporting Figure S10). From this
thermodynamic profiling the Gibbs free energies of binding (∆G), which partition into enthalpic
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(∆H) and enthropic (T∆S) binding contributions, were obtained. Compound (+)ꢀ1 (∆G = –49.88
kJ/mol) was found to be predominantly enthalpic binder with ∆H contribution of –46.36 kJ/mol
and small favorable entropic portion of –3.52 kJ/mol. Also for compound (+)ꢀ3 (∆G= –60.65
kJ/mol), very strong enthalpic interactions of –97.23 kJ/mol were determined, however they were
partially compensated by entropic penalty of +36.58 kJ/mol (Figure 7).
Figure 7. Thermodynamic signatures for compounds (+)ꢀ1 and (+)ꢀ3. Histograms of thermodynamic
signatures of hBChEꢀcompound complexes, with Gibbs free energy of binding (ꢂG; blue), enthalpy (ꢂH;
green) and entropy (–TꢂS; red).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
These results are consistent with in silico analysis of binding modes and interaction energies
calculations. Using Schrödinger Glide XP docking protocol, compounds (+)ꢀ1 and (+)ꢀ3 were
scored within their respective crystal complexes, and their docking scores were –8.687 and –
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
5.197 kcal/mol, respectively. A key difference in interaction profiles between both compounds
was observed with compound (+)ꢀ3 demonstrating favorable electrostatic/πꢀcation interactions
Electro and πꢀcat terms –1.26 and –0.91, respectively), which were not present in compound (+)ꢀ
(
1 (–0.72 and 0). Furthermore, while only one waterꢀbridged Hꢀbond was observed between
methoxy group and carbonyl oxygen of compound (+)ꢀ1 with crystal water 781 (Supporting
Figure S11), a network of water molecules in a vicinity of compound (+)ꢀ3 (crystal waters: 54,
60, 81, 97, 103, 119, 127, 148, 186, 191, 221 and 383) could be observed (Supporting Figure
S12). Crystal waters 54 and 127 enable further waterꢀbridged Hꢀbonds (with Ser198 and His485,
and Pro285, respectively), while others form a water lattice around the compound (+)ꢀ3. We can
postulate that the observed water molecules contribute favorably to the enthalpy of binding, while
simultaneously represent a high entropic penalty.
Rat Brain Slices Experiments
To confirm that compound 3 can efficiently inhibit BChE in an environment that is more similar
to in vivo conditions, inhibition of BChE was performed in rat brain slices. The modified Koelleꢀ
Karnovsky histochemical method was used on the brain sections where the highest BChE activity
2
9, 52–54
was detected.
The formation of the dark brown/ black granular BChE reaction product was
observed in both vascular and neuronal structures of the laterodorsal thalamic nucleus. Complete
BChE inhibition was observed when these sections were incubated with 10 ꢁM compound 3
(Figure 8). Of note, 300 ꢁM of sulfonamide 2 was needed to achieve the same effect, highlighting
2
9
a correlation between the inhibition constants determined in vitro and the results obtained here.
ACS Paragon Plus Environment

Page 23 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 8. Representative coronal section from a rat brain in the region containing the laterodorsal
thalamic nucleus. The sections were processed for BChE histochemical staining using 4 mM
butyrylthiocholine iodide (BTCI), in the absence (left) and presence (right) of 10 ꢁM compound 3.
Selective AChE inhibitor 1,5ꢀbis(4ꢀallyldimethylammoniumphenyl)pentanꢀ3ꢀone dibromide was added to
completely block AChE activity. Magnification, 10×. The relative optical density scores for the staining
intensity for BChE activity are shown on the right.
Cytotoxicity of Compound 3
The cytotoxicity profile in human neuroblastoma SHꢀSY5Y cells (Figure 9) demonstrated a LC50
of 6.63 µM for compound 3, which is more than 6,000ꢀfold greater than the concentration needed
for 50% inhibition of huBChE. Cytotoxicity profile was also demonstrated in the HepG2 cell
line, where compound 3 showed even lower impairment of the cell metabolic activity (LC =
5
0
33.87 µM) (Figure 9).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 9. Cytotoxicity of compound 3. Concentrationꢀdependent cytotoxicity of compound 3 in human
neuroblastoma SHꢀSY5Y cells and in human hepatocellular HepG2 cells. SHꢀSY5Y cells and HepG2
cells were incubated in the presence of the indicated concentrations of compound 3. Cell viability was
evaluated after 48 h (for SHꢀSY5Y cells) and 24 h (for HepG2 cells) using the MTS assay. The control
group (with DMSO) was considered as 100% cell viability. Data are means ±SD of three independent
experiments, each carried out in triplicate. The X axis is in log scale.
In Vitro Pharmacokinetics and In Vivo Blood Plasma–Brain Distribution of Compound 3
The in vitro pharmacokinetics and in vivo blood plasma–brain distribution of compound 3 were
29
determined as previously reported for sulfonamide 2. The permeability values (P ) of
app
compound 3 in heterogeneous human epithelial colorectal adenocarcinoma (Cacoꢀ2) cells were
determined in both the absorptive [apicalꢀtoꢀbasolateral; (A–B)] and eliminatory [basolateralꢀtoꢀ
apical; (B–A)] directions, allowing the calculation of the efflux ratio as Papp(B–A)/Papp(A–B). The
–6
Papp(A–B) (absorptive) of compound 3 was 28.0 ±0.9 ×10 cm/s, which is classified as “high”
ACS Paragon Plus Environment

Page 25 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5
5
according to the biopharmaceutical classification system of drug permeability. This would thus
not be a limiting factor in the bioavailability of compound 3. Papp(B–A) (eliminatory) of compound
–6
3
was 46.1 ±5.2 ×10 cm/s. These data yielded an “efflux ratio” (Papp(B–A)/Papp(A–B)) of 1.65,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
which is classified as “low” and indicates very limited active transport of compound 3 in the
56
eliminatory direction.
Further analysis showed that 93% of compound 3 was bound to plasma proteins
5
7
(Supporting Table S2), which classifies this compound as highly protein bound. Compound 3
was relatively stable in human plasma, with 76% still present and intact after 120 min (i.e., t1/2
>
120 min) (Supporting Table S3). Likewise, the hepatic metabolism of compound 3 in
cryopreserved hepatocytes was moderate (t1/2 >120 min) (Supporting Table S4). Compound 3
showed a considerably longer halfꢀlife compared to other high extraction ratio reference
compounds cleared by the cytochrome P450 enzymes used as positive controls in the metabolic
stability assays (i.e., flurazepam, naloxone, propranolol; Supporting Table S5). The moderate
hepatocyte clearance (<8.2 µL/min/million cells) (Supporting Table S4) of compound 3 is in
agreement with the long halfꢀlife and single metabolite (monomethyl derivative 11; Supporting
Table S6) identified using liquid chromatography–high resolution mass spectrometry. As
metabolite 11 is a potent and selective huBChE inhibitor (IC50 = 0.018 µM), the parent
compound 3 might have residual pharmacological activity.
As AD is a brain disorder, any drug used for its treatment needs to cross the blood–brain
barrier. In vivo measurements of the distribution between the blood plasma and the central
5
8
nervous system is the most reliable evaluation of this crossing of the blood–brain barrier. One
hour after rats had an intraperitoneal injection of 10 mg/kg compound 3, the mean (±SEM)
concentration of compound 3 in the brain tissue was 248 ±37 µg/kg, while that in the blood
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
plasma was 169 ±13 µg/L. The brainꢀtoꢀblood plasma ratio of compound 3 was 1.47, which is
29
more than threeꢀfold higher than the brainꢀtoꢀblood plasma ratio of sulfonamide 2.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In Vivo Activity of Compound 3
Encouraged by the selective and potent huBChE inhibitory activity, inhibition of BChE in rat
brain slices, low cytotoxicity, favorable in vitro pharmacokinetics, and proven penetration into
the brain, we set out to test the effects of compound 3 on behavioral activities. The Morris water
maze task was performed to test the potential procognitve properties of compound 3.
Scopolamine was used to induce cognitive deficits in test animals (1 mg/kg; intaperitoneal
injection prior to the acquisition trials in each test). Scopolamine is a nonꢀselective muscarinic
ACh receptor antagonist, and it produces cognitive deficits in test subjects that are comparable to
59–62
those caused by the cholinergic deficit characteristic of AD.
ChE inhibitors alleviate
scopolamineꢀinduced cognitive deficits by increasing the levels of synaptic ACh (notably in the
6
0,63
brain), which results in displacement of scopolamine from muscarinic ACh receptors.
Scopolamineꢀinduced amnesia is thus frequently used in AD research for preclinical evaluation
5
9,64–67
of proꢀcognitive properties of ChE inhibitors.
The selective AChE inhibitor donepezil was
used in the behavioral tasks as the positive control. C57BL/6J and CDꢀ1 Krf mice were used for
the experiments. These mice have an unaltered brain BChE/AChE enzymatic activity ratio, and
they were used because test animals with an increased BChE/AChE enzymatic activity ratio are
currently not available.
The Morris water maze task is a hippocampusꢀdependent memory and spatial learning
task used to investigate (i) effects on longꢀterm memory, through decreased time to reach an
escape platform (i.e., escape time latency); (ii) effects on cognition through the measure of the
overall distance travelled to reach the escape platform; and (iii) induction of sensorimotor deficits
ACS Paragon Plus Environment

Page 27 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
6
8–70
through the measure of the mean speed of swimming.
During the consecutive training days in
all of experimental groups, the escape latency times progressively decreased, as evident from the
characteristic learning curves (Figure 10A). Scopolamineꢀtreated mice showed significantly
attenuated learning abilities and cognitive functions, as compared to vehicleꢀtreated mice, as they
had significantly longer escape latency times (Figure 10A) and travelled longer distances to reach
the escape platform (Figure 10B). This cognitive dysfunction caused by scopolamine was
attenuated by compound 3 (30 mg/kg, intraperitoneally) as demonstrated by both shorter escape
latency times (Figure 10A) and shorter distances travelled to reach the escape platform (Figure
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
0B). This thus illustrates the memoryꢀimproving and procognitive properties of compound 3,
29
which were comparable to sulfonamide 2. Donepezil (10 mg/kg) was also effective in this
behavioral test (Figure 10A, B). This was expected, as donepezil is a selective AChE inhibitor
and compound 3 is a selective BChE inhibitor, and the mice used had an unaltered brain
BChE/AChE enzymatic activity ratio, with AChE accounting for the majority of their brain ChE
enzymatic activity.
Analysis of the mean speed of swimming revealed that scopolamine induced significant
hyperlocomotion in mice, as it increased the mean speed of swimming (vehicle, 0.13 m/s;
scopolamine 0.20 m/s). Both compound 3 and donepezil decreased the hyperlocomotion caused
by scopolamine, as they reduced the mean speed of swimming compared to scopolamineꢀtreated
mice (scopolamine + compound 3, 0.18 m/s; scopolamine + donepezil, 0.15 m/s).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 10. Effects of compound 3 and donepezil on scopolamineꢀinduced memory impairment in the
Morris water maze task. (A) Learning curves of the escape latencies for vehicleꢀtreated mice (green),
scopolamineꢀtreated control mice (black), and scopolamineꢀinduced memoryꢀimpaired mice treated with
3
0 mg/kg compound 3 (brown) or 10 mg/kg donepezil (purple) during the acquisition phase. (B) Distances
traveled to reach the target platform during the acquisition phase in vehicleꢀtreated mice (green bars),
scopolamineꢀtreated control mice (black bars), and scopolamineꢀinduced memoryꢀimpaired mice treated
with 30 mg/kg compound 3 (brown bars) or 10 mg/kg donepezil (purple bars). Data are means ±SEM.
ACS Paragon Plus Environment

Page 29 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Statistical analysis: repeatedꢀmeasures ANOVA, followed by Bonferroni postꢀhoc comparisons. * p <
0
.05; ** p < 0.01; *** p < 0.001 (vs. scopolamine + vehicleꢀtreated mice).
In Vivo Safety of Compound 3
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Peripheral adverse effects of approved drugs for the treatment of patients with AD are the result
of an excess of ACh produced by inhibition of ChEs in the peripheral nervous system, the
3
1
parasympathetic autonomic system, and the basal ganglia. To interpret the data from the Morris
water maze task, the in vivo safety profile of compound 3 and its impact on the general activity
and motor skills of the mice were determined, using the locomotor activity and rotarod tests.
Compound 3 was tested at the concentration effective in the behavioral tests (30 mg/kg
intraperitoneal). Compound 3 did not induce motor deficits in these mice, as shown in the rotarod
tests, at 6 rpm, 18 rpm, or 24 rpm. Although compound 3 reduced the locomotor activity between
2
4 min and 30 min (Supporting Figure S13) of the locomotor activity test, this had no impact on
the results obtained in the behavioral test. Additionally, administration of compound 3 to mice
did not produce any acute adverse cholinergic effects. This was expected, as selective BChE
29,31
inhibition in vivo does not produce peripheral (parasympathomimetic) adverse effects.
Conclusions
A comprehensive series of 49 selective BChE inhibitors was designed and synthesized, and their
structure–activity relationships were determined. Crystalꢀstructureꢀbased design led to the most
potent inhibitor of the series, compound 3, which was resolved into its pure enantiomers using
semiꢀpreparative chiral HPLC. Detailed kinetic studies showed slow, tightꢀbinding inhibition of
compound 3 versus huBChE, where the eutomer demonstrated a picomolar in vitro huBChE
inhibitory potency (K = 11.1 pM). The solved crystal structure of compound 3 in complex with
i
ACS Paragon Plus Environment