Discovery of Potent Human Glutaminyl Cyclase Inhibitors as Anti-Alzheimer’s Agents Based on Rational Design

Article abstract of DOI:10.1021/acs.jmedchem.7b00098

Glutaminyl cyclase (QC) has been implicated in the formation of toxic amyloid plaques by generating the N-terminal pyroglutamate of β-amyloid peptides (pGlu-Aβ) and thus may participate in the pathogenesis of Alzheimer’s disease (AD). We designed a library of glutamyl cyclase (QC) inhibitors based on the proposed binding mode of the preferred substrate, Aβ_3E?42. An in vitro structure-activity relationship study identified several excellent QC inhibitors demonstrating 5- to 40-fold increases in potency compared to a known QC inhibitor. When tested in mouse models of AD, compound 212 significantly reduced the brain concentrations of pyroform Aβ and total Aβ and restored cognitive functions. This potent Aβ-lowering effect was achieved by incorporating an additional binding region into our previously established pharmacophoric model, resulting in strong interactions with the carboxylate group of Glu327 in the QC binding site. Our study offers useful insights in designing novel QC inhibitors as a potential treatment option for AD.

Full text of DOI:10.1021/acs.jmedchem.7b00098

Subscriber access provided by University of Newcastle, Australia
Article
Discovery of Potent Human Glutaminyl Cyclase Inhibitors
as Anti-Alzheimer’s Agents based on Rational Design
Van-Hai Hoang, Phuong-Thao Tran, Minghua Cui, Van T.H. Ngo, Jihyae
Ann, Jongmi Park, Jiyoun Lee, Kwanghyun Choi, Hanyang Cho, Hee Kim,
Heejin Ha, Hyun-Seok Hong, Sun Choi, Young-Ho Kim, and Jeewoo Lee
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b00098 • Publication Date (Web): 24 Feb 2017
Downloaded from http://pubs.acs.org on February 26, 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 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of Potent Human Glutaminyl Cyclase Inhibitors
as AntiꢀAlzheimer’s Agents based on Rational Design
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
†
†
,
⊥
‡
†
†
VanꢀHai Hoang, PhuongꢀThao Tran, Minghua Cui, Van T.H. Ngo, Jihyae Ann, Jongmi
‡
∥
§
§
§
§
Park, Jiyoun Lee, Kwanghyun Choi, Hanyang Cho, Hee Kim, HeeꢀJin Ha, HyunꢀSeok
§
‡
§
,
†
Hong, Sun Choi, YoungꢀHo Kim, Jeewoo Lee*
†
Laboratory of Medicinal Chemistry, Research Institute of Pharmaceutical Sciences, College of
Pharmacy, Seoul National University, Seoul 08826, Republic of Korea
‡
National Leading Research Laboratory of Molecular Modeling & Drug Design, College of
Pharmacy and Graduate School of Pharmaceutical Sciences, Ewha Womans University, Seoul
0
3760, Republic of Korea
§
Medifron DBT, Sandanro 349, DanwonꢀGu, AnsanꢀCity, GyeonggiꢀDo 15426, Republic of
Korea
∥
Department of Global Medical Science, Sungshin University, Seoul 01133, Republic of Korea
⊥
Department of Pharmaceutical Chemistry, Hanoi University of Pharmacy, Hanoi, Vietnam
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 66
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
Glutaminyl cyclase (QC) has been implicated in the formation of toxic amyloid plaques by
generating the Nꢀterminal pyroglutamate of βꢀamyloid peptides (pGluꢀAβ), thus may participate
in the pathogenesis of Alzheimer’s disease. We designed a library of glutamyl cyclase (QC)
inhibitors based on the proposed binding mode of the preferred substrate, Aβ3Eꢀ42. An in vitro
structureꢀactivity relationship study identified several excellent QC inhibitors demonstrating 5ꢀ to
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
40ꢀfold increases in potency compared to a known QC inhibitor. When tested in mouse models of
Alzheimer’s disease (AD), compound 212 significantly reduced the brain concentrations of
pyroform Aβ and total Aβ and restored cognitive functions. This potent Aβꢀlowering effect was
achieved by incorporating an additional binding region into our previously established
pharmacophoric model, resulting in strong interactions with the carboxylate group of Glu327 in
the QC binding site. Our study offers useful insights in designing novel QC inhibitors as a
potential treatment option for AD.
ACS Paragon Plus Environment

Page 3 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
■
INTRODUCTION
Alzheimer’s disease (AD) and related dementias are the leading cause of disabilities in
old age and have become enormous socioeconomic burdens to many countries with rapidly
growing elderly populations. The currently available treatment options for AD, such as
acetylcholinesterase inhibitors (AChEI) and NꢀmethylꢀDꢀaspartate (NMDA) receptor antagonists,
offer only limited symptomatic relief, rather than modifying the disease course. Among the
various alternatives that have been extensively investigated, βꢀamyloid (Aβ)ꢀtargeted therapies
have been the center of attention because many research findings indicated that elevated levels of
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
1
Aβ peptides in the brain promote the formation of amyloid plaques and neuronal death. What
causes the elevated levels of Aβ peptides is still under debate, but either facilitating the clearance
or suppressing the formation of Aβ is likely to reduce amyloid burdens in the brain, possibly
slowing down the neurodegenerative process. Unfortunately, several Aβꢀlowering drugs based on
this premise, such as bapineuzumab, solanezumab, and semagacestat, have failed to provide
significant therapeutic benefits in clinical trials. One possible explanation for this recent failure is
that the brain Aβ peptides are a heterogeneous mixture of peptides, each form of which displays
2
,3
different structural and functional features. Given that Aβ peptides participate in such diverse
4
5ꢀ7
8,9
physiological processes including neurogenesis, neuronal survival, oxidative stress, and
1
0
innate immunity, targeting specific forms of Aβ peptides that are prone to aggregation and are
neurotoxic may provide a more effective therapy.
Aβ peptides are generated by the proteolytic cleavage of the amyloid precursor protein
(APP), a transmembrane glycoprotein. Depending on the sequential cleavage sites, Aβ peptides
with 30ꢀ51 amino acid residues are produced. Aβ1ꢀ40 and Aβ1ꢀ42 are the most abundant isoforms
in the human brain and are thought to be the major initiators of AD pathogenesis. The brains of
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 66
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
AD patients also contain a significant amount of Nꢀterminally truncated species, such as Aβnꢀ40/42
,
1
1ꢀ13
where n ranges from 2 to 11.
Many studies have found that these Nꢀterminally truncated
species are further cyclized to form pyroglutamate (pGlu)ꢀAβ. Due to their increased
hydrophobicity, pGluꢀAβ peptides are prone to rapid aggregation and are much more resistant to
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
proteolytic degradation. In fact, Aβ3(pE)ꢀ42 constitutes approximately 15ꢀ20% of the total Aβ and
42
1
4ꢀ16
is deposited in the center of senile plaques in the brains of AD patients.
It has been reported
that pGluꢀAβ peptides are more neurotoxic than Aβ1ꢀ40 and Aβ1ꢀ42 and act as a seed for amyloid
1
7ꢀ19
and tau plaques.
The pGluꢀAβ peptides are generated by glutaminyl cyclase (QC). Mammalian QC is
mainly found in the pituitary, hypothalamus, and brain and participates in the maturation of
2
0ꢀ22
neuropeptides and hormones.
Notably, QC is overexpressed in the brains of AD patients and
2
3,24
25
animal models,
and the knockꢀout of QC rescued the cognitive function in AD model mice.
2
6
Small molecule QC inhibitors also reduced brain pGluꢀAβ levels and Aβ plaques, in addition to
24
decreasing gliosis and restoring memory deficits in AD mice. In addition, the results from a
recent phase I trial indicate that QC inhibitors are well tolerated and metabolically safe for both
27
young and elderly patients, suggesting that the inhibition of QC offers an alternative approach
to conventional Aβꢀlowering agents.
In contrast to the emerging importance of QC in AD pathogenesis, relatively few QC
28ꢀ32
inhibitors have been reported so far.
In general, these inhibitors share two distinct structural
features: first, a zincꢀbinding moiety, such as imidazole, that can bind to the active site zinc ion
and second, a large hydrophobic residue that can occupy the penultimate position to the Nꢀ
terminal glutamine. Previously, our group designed a series of QC inhibitors based on this
ACS Paragon Plus Environment

Page 5 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
general scaffold, which we divided into three pharmacophoric regions: Aꢀregion representing the
zincꢀbinding group, Bꢀregion containing a hydrogen bond donor, and Cꢀregion mimicking the
3
2
aromatic ring of the phenylalanine side chain. We studied the structureꢀactivity relationship
specifically focused on the Cꢀregion, while having 5ꢀmethylimidazole in the Aꢀregion and a
propylthiourea moiety in the Bꢀregion, based on the structure of the previously reported QC
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
8
inhibitor, compound 1 (Figure 1). Although we found a promising candidate with an IC value
5
0
of 58 nM in our previous study, we wanted to further expand our library of QC inhibitors and to
study their SAR as well as their in vivo activity. To identify an additional pharmacophore, we
turned our attention to the antepenultimate Arg of Aβ3Eꢀ42. While the substrate specificity of this
specific position has not been clearly elucidated, we speculated that the additional side chain
might be able to act as a positional anchor to provide enhanced binding interactions. Therefore,
we designed a novel scaffold that has an additional pharmacophoric region D, as described in
Figure 1. These new series of compounds contain various amineꢀtype functional groups in the Dꢀ
region, which mimic the guanidine moiety of Arg. In this work, we synthesized newly designed
QC inhibitors focused on the Dꢀregion and investigated their SAR based on in vitro inhibition for
human QC. Furthermore, we evaluated the efficacy of the selected inhibitors by measuring their
ability to reduce the formation of Aβ3(pE)ꢀ42 and to prevent memory impairment in AD model
mice. Finally, we analyzed the specific binding interactions between the selected inhibitors and
the QC active site by performing molecular docking studies.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 66
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. Newly designed scaffold for QC inhibitors
■
RESULTS AND DISCUSSION
Chemistry.
To synthesize compounds with the newly designed scaffold that includes the Dꢀregion, we
first synthesized a library of 1ꢀalkyloxyꢀ2ꢀmethoxyꢀ4ꢀnitrobenzene fragments, as shown in
Schemes 1ꢀ3. The syntheses of 1ꢀacyclic and cyclic aminoalkyloxyꢀ2ꢀmethoxyꢀ4ꢀnitrobenzene
derivatives are described in Scheme 1. The Williamson reaction of 4ꢀnitroguaialcol 2 with
dibromoalkanes followed by Nꢀalkylation provided piperidine, morpholine and piperazine
derivatives 6ꢀ17. 4ꢀSubstituted piperidinyl derivatives, 18ꢀ20, were synthesized by the Mitsunobu
3
3, 34
reaction of 2 with corresponding 4ꢀ(hydroxyalkyl)piperidine fragments,
followed by an Nꢀ
ACS Paragon Plus Environment

Page 7 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Boc deprotection reaction to afford 21 and 23. Compounds 21 and 23 were then reductively
methylated to provide 22 and 24. Acyclic aminoalkyl derivatives, 25ꢀ28, were synthesized by
Williamson or Mitsunobu reactions with 2. Alkyl phthalimide intermediates, 29 and 30, were
prepared from 2 and deprotected to give primary amines 31 and 32, which were converted to
compound 33 by reductive amination and to compounds 38ꢀ41 by monoalkylation using 2ꢀ
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
3
5
nitrobenzenesulfonamide as a protecting group, respectively. 1ꢀHeteroarylalkyloxyꢀ2ꢀmethoxyꢀ
ꢀnitrobenzene fragments were prepared as described in Scheme 2. The Sonogashira reaction of
4
3
6
aryl bromides 42ꢀ45, 47 with corresponding terminal alkynes provided the alkyne intermediates
0ꢀ58, which were further reduced and underwent a Mitsunobu reaction to afford 70ꢀ74, 76ꢀ79.
5
Compound 75 was prepared from commercially available 48 by reduction and protection
followed by the Williamson reaction. 1ꢀPyridinylalkoxyꢀ2ꢀmethoxyꢀ4ꢀnitrobenzene fragments
were synthesized as described in Scheme 3. The Cꢀalkylation of picolines and aminopicolines
with bromoalkane derivatives or diethylcarbonate provided chloroethyl, TBSꢀprotected
hydroxyalkyl, and ethoxycarbonyl intermediates (85ꢀ89, 92ꢀ93). Chloroalkyl intermediates (85ꢀ
8
7) were conjugated with 4ꢀnitroguaiacol via the Williamson reaction to produce compounds 96ꢀ
8, while TBSꢀprotected hydroxyalkyl intermediates (88ꢀ89) were first deprotected with TBAF,
9
and subsequently conjugated with 4ꢀnitroguaiacol via the Mitsunobu reaction to yield compounds
9ꢀ100. The ethoxycarbonyl group in compounds 92ꢀ93 was first reduced to an ethylalcohol
9
group (94ꢀ95) by using LAH, and further conjugated with 4ꢀnitroguaiacol via the Mitsunobu
reaction to generate compounds 101ꢀ102.
As described in Scheme 4, all of the synthesized 1ꢀalkyloxyꢀ2ꢀmethoxyꢀ4ꢀnitrobenzene
fragments were reduced to corresponding amines, 103ꢀ145, which were subsequently converted
to isothiocyanate and coupled with 3ꢀ(5ꢀmethylꢀ1Hꢀimidazolꢀ1ꢀyl)propanꢀ1ꢀamine to afford the
final thiourea compounds in situ. Compounds 146ꢀ166 were further subjected to NꢀBoc
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 66
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
deprotection, while compounds 167 and 168 were reacted with hydrazine to deprotect a
phthalimide group, yielding primary amine final compounds (172ꢀ174, 191ꢀ193, 197ꢀ198, 203ꢀ
205, 210ꢀ211, 214ꢀ216, 219ꢀ220) as well as amine intermediates (169ꢀ171, 199, 212). The final
modification of the primary amine group of these intermediates generated acetylated (178, 213),
guanidinylated (179), and methylated (202) analogues. The 2ꢀchloropyrimidine derivatives (180ꢀ
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
184) were synthesized via the nucleophilic substitution of the amine intermediates to afford the
corresponding final compounds.
a
Scheme 1. Synthesis of alkyl Dꢀregion
a
Reagents and conditions: (a) Bromoalkyl derivatives, K CO , DMF, 100°C, 1 h for 25, 26, and 28; (b)
2
3
DMF, base (piperidine, morpholine, 1ꢀmethylpiperazine or NꢀBoc piperazine), heat; (c) 4ꢀ
(hydroxyalkyl)piperidine, DEAD, PPh , DCM, r.t, 2 h; (d) TFA, DCM, r.t, overnight; (e) (HCHO) , ZnCl ,
3
n
2
ACS Paragon Plus Environment

Page 9 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
DCM, r.t, 1 h; then NaBH , reflux, overnight; (f) N H .H O, EtOH, r.t, overnight; (g) 2ꢀ
4
2
4
2
nitrobenzenesulfonyl chloride, TEA, DCM, r.t, 5 h; (h) CH I, NaH, THF, r.t, overnight; (i) thiophenol,
3
K CO , MeCN, r.t, overnight; (j) Boc O, Et N, DCM, r.t, 2 h; (k) HO(CH) N(CH )Boc, DEAD, PPh ,
2
3
2
3
2
3
3
DCM, r.t, 2 h for 27
.
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
a
Scheme 2. Synthesis of aryl Dꢀregion
a
Reagents and conditions: (a) CuI, Pd(PPh ) , Et N, DCM, 50°C, overnight or alkynꢀ1ꢀol, DMF, Et N,
(Ph P) PdCl , CuI, r.t, 15 h; (b) Pd/C, MeOH, H , r.t, overnight; (c) 4ꢀnitroguaiacol, DEAD, PPh , DCM,
r.t, 2 h; (d) HCOONH , Zn, MeOH, r.t, 15 min; (e) 4ꢀnitroguaiacol, Cs CO , DMF, 100°C; (f) Boc O,
3
4
3
3
3
2
2
2
3
4
2
3
2
Et N, DCM, r.t, overnight.
3
a
Scheme 3. Synthesis of aryl Dꢀregion
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 66
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
a
Reagents and conditions: (a) bromoalkane derivatives or diethylcarbonate, nꢀBuLi, THF, ꢀ78 °C; (b) 4ꢀ
nitroguaiacol, K CO , DMF, 100°C, 1 h or 4ꢀnitroguaiacol, DEAD, Ph P, DCM, r.t; (c) Boc O, tꢀBuOH,
r.t, overnight; (d) TBAF, THF, r.t, 2 h; (e) LAH, THF, 0°C, 1 h.
2
3
3
2
a
Scheme 4. Synthesis of final QC 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
Reagents and conditions: (a) Pd/C, H , MeOH, r.t, overnight; (b) 3ꢀ(5ꢀmethylꢀ1Hꢀimidazolꢀ1ꢀyl)propanꢀ
2
1
ꢀamine, TCDI, Et N, DCM, r.tꢀ40°C, overnight; (c) TFA, DCM, r.t, 2 h; (d) 2ꢀchloropyrimidine
3
derivatives, EtOH, Et N, reflux, 2 days; (e) (HCHO) , ZnCl , DCM, r.t, 1 h; then NaBH , reflux, overnight;
(
HCl, iPr EtN, MeOH, r.tꢀ50°C, 2 h.
3
n
2
4
f) Ac O, pyridine, MC, r.t, 4ꢀ5 h; (g) N H .H O, EtOH, r.t, overnight; (h) 1Hꢀpyrazoleꢀ1ꢀcarboxamidineꢀ
2 2 4 2
2
In vitro QC Inhibition.
To evaluate the biological activity of the compound library, we determined the QC
inhibition of each compound by using a fluorogenic substrate, GlnꢀAMC (Lꢀglutamine 7ꢀamidoꢀ
37
4
ꢀmethylcoumarin), and pyroglutamyl peptidase (pGAP) as an auxiliary enzyme. In this assay,
QC first converts GlnꢀAMC into pGluꢀAMC, which is then hydrolyzed by pGAP to generate
AMC. AMC can be detected by measuring the fluorescence at the excitation and emission
wavelengths of 380 and 460 nm. These results are summarized in Table 1ꢀ3, along with the IC50
value of compound 1 (IC = 29.2 nM) for comparison.
5
0
ACS Paragon Plus Environment

Page 11 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
First, we analyzed the structureꢀactivity relationship of aliphatic amines in the Dꢀregion,
as shown in Table 1. The group I (169ꢀ184) contain various terminal amine, amide, guanidine,
and pyrimidylamine functional groups, while having an aliphatic linker that was varied in length.
In general, compounds with a 2ꢀcarbon linker (169, 172, 175, 180) showed a similar activity to 1,
whereas compounds with a 3ꢀcarbon (170, 173, 176, 181, 183) or 4ꢀcarbon (171, 174, 177, 182,
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
184) linker demonstrated more potent inhibition. In particular, 170, 177 and 182 were found to be
6
ꢀ8ꢀfold more potent than 1, having IC values of 5.3, 3.7, and 5.7 nM, respectively. Among the
5
0
various terminal functional groups, amines and pyrimidylamines showed greater inhibition than
amide or guanidine groups, suggesting that the electron density in this region might control the
binding interactions rather than the size or number of hydrogens. Furthermore, compound 182
showed 3ꢀfold more potent inhibition than compound 184, which contained an extra fluorine
atom in the pyrimidylamine ring, supporting our hypothesis.
Table 1. IC values for inhibition of hQC by benzocyclic compounds (Group I)
5
0
a
Compound
R
n
IC50 (nM)
1
29.2 (±4.0)
30.2 (±4.3)
5.3 (±1.0)
13.8 (±3.1)
38.7 (±7.1)
6.9 (±1.4)
16.2 (±2.1)
38.6 (±4.3)
169
2
3
4
2
3
4
2
1
1
1
70
71
72
173
1
1
74
75
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 66
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
1
76
77
3
4
13.0 (±1.7)
1
3.7 (±1.1)
1
78
79
3
3
40.8 (±2.4)
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
1
41.2 (±8.2)
1
80
81
2
3
4
3
4
39.4 (±10.4)
17.5 (±2.3)
5.7 (±2.8)
1
182
1
83
84
18.5 (±3.8)
17.5 (±4.1)
1
a
The values indicate the mean of at least three experiments.
Next, we investigated the SAR of the cyclic amineꢀcontaining compounds, group II, as
shown in Table 2. Interestingly, all the compounds in this group proved to be highly potent
inhibitors, with IC values ranging from 0.7 to 24.7 nM, regardless of the type of cyclic amine
5
0
and the length of the linker. In particular, 1ꢀpiperazinyl analogs (191ꢀ193) appeared to be the
most potent in this group, with compound 191 demonstrating an IC value of 0.7 nM, is was 42ꢀ
5
0
fold more potent than 1. Both nitrogen atoms in the piperazine appeared to be important in
binding interaction, because removing any one of either nitrogen reduced inhibitory activity as
observed in the 1ꢀpiperidinyl (185ꢀ187), 4ꢀmorpholinyl (188ꢀ190), 4ꢀpiperidinyl (197, 199)
except 198 (n=3), and 4ꢀ(1ꢀmethylpiperadinyl) (200ꢀ202) analogues. In addition, the 4ꢀNH
hydrogen might play a crucial role for favorable binding interactions, because the 1ꢀpiperidinyl
(
185ꢀ187) analogues were much less potent than the 4ꢀpiperidinyl analogues (197ꢀ199) as well as
the 1ꢀpiperazinyl analogues (191ꢀ193). This apparent importance of the NH at the 4ꢀposition is
not observed with n =3 derivatives (186, 189, 192, 195, 198, and 201), suggesting that the
ACS Paragon Plus Environment

Page 13 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
distance between the Rꢀgroup and Cꢀregion also affects the hydrogen bond interaction of the 4ꢀ
NH.
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
Table 2. IC values for inhibition of hQC by benzocyclic compounds (Group II)
5
0
a
Compound
185
R
n
2
3
4
2
3
4
2
3
4
2
3
4
2
3
4
2
3
4
IC50 (nM)
24.7 (±3.5)
13.4 (±1.1)
12.5 (±1.7)
12.5 (±2.9)
14.1 (±0.9)
9.2 (±2.1)
0.7 (±0.4)
11.4 (±2.3)
4.8 (±1.0)
7.3 (±1.9)
11.7 (±1.0)
8.2 (±0.4)
13.8 (±0.7)
7.2 (±1.4)
7.8 (±1.8)
4.6 (±1.4)
20.2 (±3.8)
12.3 (±2.5)
1
1
1
1
86
87
88
89
190
1
1
1
1
91
92
93
94
195
1
1
1
96
97
98
199
200
2
2
01
02
a
The values indicate the mean of at least three experiments.
We examined the QC inhibition of compounds with aryl amines and nitrogenꢀcontaining
heterocycles, group III, as demonstrated in Table 3. All three aniline analogs (203ꢀ205) appeared
to be potent inhibitors, regardless of the length of the linker, with IC values ranging from 7.7 to
50
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 66
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
9
.0 nM. Within this group, the length of the alkyl linker does not significantly affect the
inhibitory effects; rather, the position of nitrogen in the heterocycle seems to be crucial for
favorable binding interactions, as was also observed for the compounds in group B. For example,
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
3ꢀpyridyl analogue (209, IC = 19.8 nM) was less potent than the 4ꢀpyridyl surrogate (208, IC
5
0
5
0
=
9.1 nM). Among the group C compounds, 212, a 4ꢀ(2ꢀaminopyridyl) analogue with a 4ꢀcarbon
linker, was the most potent inhibitor, having an IC value of 4.5 nM. In contrast, compound 216,
5
0
a positional isomer of 212, exhibited much less potent inhibition (IC50 = 19.3 nM), again
supporting that the binding direction or position of nitrogen atoms in the heterocycle is
particularly important. In addition, to confirm the selectivity of compound 212, we also
3
8,
determined the IC value for isoQC, an isozyme of QC with a differential substrate specificity.
5
0
3
9
The IC value of 212 for isoQC was 502 nM, which was more than 100ꢀfold greater than the
5
0
IC value for QC, indicating that compound 212 is highly specific for QC. 5ꢀPyrimidinyl (217,
5
0
2
18) and 5ꢀ(2ꢀaminopyrimidinyl) (219, 220) analogues were found to be slightly less potent than
pyridine analogs.
Table 3. IC values for inhibition of hQC by benzocyclic compounds (Group III)
5
0
a
Compound
R
n
2
3
4
2
3
4
IC50 (nM)
2
2
03
04
7.7 (±1.4)
9.0 (±1.1)
8.3 (±0.8)
23.5 (±5.2)
14.4 (±4.4)
9.1 (±2.8)
205
206
207
208
2
09
4
2
19.8 (±4.5)
5.5 (±1.9)
210
ACS Paragon Plus Environment

Page 15 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
11
12
3
4
10.7 (±2.5)
2
4.5 (±1.4)
HN
N
2
13
O
4
19.0 (±4.6)
∗
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
2
14
15
3
4
16.6 (±1.2)
15.3 (±3.1)
2
16
4
19.3 (±4.7)
217
3
4
3
4
35.5 (±6.6)
21.3 (±4.0)
20.5 (±1.9)
18.4 (±1.4)
218
219
220
a
The values indicate the mean of at least three experiments.
In vivo activity.
Based on the in vitro activity, we selected the eight most potent compounds, which
demonstrated IC values less than ca. 5 nM, along with compound 1 as a control for further
5
0
animal studies. We first examined the cytotoxicity of each compound in an immortalized
hippocampal neuronal cell line (HTꢀ22) and found that all compounds were nonꢀtoxic at a 10 ꢀM
concentration. For the acute model studies, we injected human Aβ3ꢀ40 (5 ꢀg) and each compound
(25 mg/kg) into deep cortical/hippocampus of ICR mice (male, six weeks old). On the next day,
the inhibitory activities were determined by measuring the levels of human AβN3pEꢀ40 in the brain
extracts of the mice. As described in Table 4, compounds 170, 191, and 212 suppressed more
than 60% of the formation of AβN3pEꢀ40 compared to the vehicle control, while compound 1
showed 41.6% inhibition. Specifically, compound 191, which showed the most potent activity in
vitro, also demonstrated the most potent AβN3pEꢀ40 lowering effects (Table 4, Figure 2a).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 66
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
a
Table 4. QC inhibition in acute model studies in vivo.
In vitro
IC50
Cytotoxicity
at 10 ꢀM
(% of control)
% inhibition of
human AβN3pEꢀ40 formation human AβN3pEꢀ40 (ꢀlogPe)
(deep cortical/hippocampus
injected)
% inhibition of PAMPA
(nM)
formation
(i.p. injected)
NE
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
29.2
21.5)
5.3
~100
41.6
6.3
6.2
b
(
1
1
1
70
77
82
~100
~100
~100
~100
~100
~100
~100
~100
64.6
35.7
27.3
77.2
29.2
18.7
38.0
66.1
NE
3.7
5.7
0.7
4.8
4.6
5.5
4.5
191
3.6
6.1
1
2
93
00
210
12
2
54.7
4.5
b
(12.6)
a
5
ꢀL of human Aβ3−40 in PBS (1 ꢀg/ꢀL) was injected into the deep cortical/hippocampus to 5 weeks old ICR mice
25 g, n = 4, male) using a stereotaxic frame to induce acute Aβ toxicity. Test compounds were administrated via icv
or ip injection. Sandwich ELISA was performed for the quantification of the brain AβN3pEꢀ40
measured additionally by using mouse QC.
(
b
;
IC50 values were
Next, we performed another acute model study by intraperitoneally (i.p.) injecting the
three most potent compounds (170, 191, and 212) to assess their in vivo efficacy, which can be
translated to the ability to penetrate the bloodꢀbrain barrier (BBB). Again, human Aβ3ꢀ40 was
injected deep cortical/hippocampus prior to the i.p. administration of each inhibitor into ICR
mice. The inhibitory activities were estimated by measuring the levels of human AβN3pEꢀ40 in the
brain extracts of the mice. As shown in Figure 2b, among all the injected compounds, 212
exhibited significant AβN3pEꢀ40 lowering effects (54.7% inhibition), whereas compounds 1, 170
ACS Paragon Plus Environment

Page 17 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
and 191 were found to be inactive via i.p. injection. A parallel artificial membrane permeability
4
1
assay (PAMPA) of these four compounds also correlated with the i.p. in vivo results, showing
that 212 was the most permeable compound, while the other three compounds showed low
permeabilities with similar –logPe values (Table 4). Additionally, we determined mouse liver
microsomal stability of compound 212, and found that percent remaining after 30 minutes was 50
percent, indicating that 212 is moderately stable for metabolism in vivo. These observations
supported that a significant amount of 212 penetrated the BBB and successfully inhibited the
brain QC.
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
(a)
(b)
1
50
150
%
of
00
1
% o 1f 0V 0e hicle
Vehicl
e
5
0
50
0
**
*
*
*
*
0
1
170 177 182 191 193 200 210 212
QCI Conc. (10 mM)
Vehicle
170 191 212
Vehicle
QCI conc. (25 mg/kg)
Figure 2. Inhibition of QC in acute AD model mice showing percent inhibition of the formation
of human AβN3pEꢀ40 levels in brain after (a) deep cortical/hippocampus injection (10 mM); (b) i.p.
injection of each inhibitor (25 mg/kg). Data were analyzed by Oneꢀway ANOVA with Bonferroni
postꢀhoc test in SPSS. (*: p < 0.05, **: p < 0.01)
To validate the desired therapeutic effects of compound 212, we next performed longꢀ
term studies in transgenic model mice of AD. 212 was administered to the APP/PS1 mice via
deep cortical/hippocampus brain infusion for 21 days. After the brain infusion, the brain
concentrations of AβN3pEꢀ42 and Aβ1ꢀ42 were measured. As described in Figure 3, compound 212
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 66
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
significantly reduced the brain concentrations of AβN3pEꢀ42 (Figure 3a), which corresponds to the
observed percent inhibition in acute AD mice. Compound 212 also significantly reduced the total
Aβ levels (Figure 3b) without significant toxicity, indicating that the inhibition of QC not only
reduced the amount of pGluꢀAβ but also contributed to the overall reduction of amyloid
formation.
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)
(b)
5
0
Aβ1ꢀ
N3pEꢀAβ42 Conc. [pg/mg protein]
00
4
40
4
2
300
*
Con
c.
3
0
2
1
00
00
0
*
*
20
[ng/
1
0
0
mg
2
12
212
Vehicle
Vehicle
Figure 3. Inhibition of the formation of human AβN3pEꢀ42 and Aβ1ꢀ42 in APP/PS1 mice. (a) Brain
concentrations of AβN3pEꢀ42; (b) Brain concentrations of Aβ1ꢀ42. Data were analyzed by unpaired tꢀ
test or oneꢀway ANOVA. (ns: p > 0.05, *: p < 0.05, **: p < 0.01)
Next, we tested compound 212 in 5XFAD transgenic mice, a wellꢀknown AD mouse
4
0
model suitable to study Aβ ꢀinduced neurodegeneration and amyloid plaque formation. We
4
2
specifically choose 5XFAD mice because this particular AD model generates severe AD related
pathologies that were resulted from the massive increase of Aβ and the pyroform of Aβ in the
2
5
brains. Aβ accumulates in the 5XFAD brain starting at 2 months of age, and develops other
42
typical AD pathologies such as reduced synaptic markers, neuronal loss and memory
impairments. We injected 212 into 5XFAD mice from 19 weeks to 34 weeks of age and
performed the Morris Water Maze test to assess changes in the spatial memory impairment. The
ACS Paragon Plus Environment

Page 19 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
efficacy of 212 in 5XFAD mice was evaluated based on the subsequent improvement in cognitive
functions, as shown in Figure 4. The escape latency of 212ꢀtreated mice became significantly
shorter over repeated sessions compared with vehicleꢀtreated mice (Figure 4a), indicating that
the treatment of 212 improved cognitive function. Additionally, the duration of time spent in the
target quadrant was significantly longer with 212 treated mice (Figure 4b), also supporting that
the QC inhibitor 212 effectively recovered learning and memory function without fatal side
effects.
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
(
a)
(b)
6
0
Tim
e
spe 40
nt
in
2
0
0
tar
get
2
12
Vehicle
Figure 4. Effect of compound 212 on cognitive function of 5xFAD mice. (a) Escape latency of
each group in the hiddenꢀplatform training of the Morris water maze. Data were analyzed by twoꢀ
way repeatedꢀmeasures ANOVA; (b) Percentage of time spent in the target quadrant. Data were
analyzed by independent samples tꢀtest in SPSS. (ns: p > 0.05, *: p < 0.05)
Molecular Modeling Studies.
Among the compounds we designed based on the Nꢀterminal tripeptide (GluꢀPheꢀArg) of
Aβ3Eꢀ42, compounds 191 and 212 exhibited excellent in vitro activity. Specifically, 191
demonstrated more than 40ꢀfold higher potency compared with the reported dipeptide mimic
compound 1. Considering that the addition of the Dꢀregion pharmacophore improved the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 66
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
inhibitory effects to such a great extent, we decided to study the mode of binding of these
compounds in the hQC active site. For our docking studies, we utilized the Xꢀray crystal structure
4
2
of hQC complexed with a known QC inhibitor, PBD150 (PDB id: 3PBB). Since the ligand
protonation states are important for an accurate docking study, we first performed the protonation
state prediction of our ligands. At the physiological pH of 7.4, the Dꢀregion in 191 was
monoprotonated in its major form and that of 212 was predicted to have neutral and protonated
forms with the ratio of 2:1. Glide SP (Standard Precision) docking was also carried out, and both
191 and 212 docked very well into the active site of hQC. The 5ꢀmethyl imidazole in the Aꢀ
region chelated with zinc and formed an Hꢀbonding interaction with the indole NH of Trp329.
Additionally, it displayed hydrophobic interactions with Leu249, Trp207, and Ile321. The
thiourea group in the Bꢀregion acted as a linker, and the phenyl ring in the Cꢀregion exhibited a
hydrophobic interaction with Tyr299. The piperazine or pyridine ring of the Dꢀregion also
exhibited hydrophobic interactions with Pro326 and Pro324, but the NH in the Arg mimetic part
of the compounds did not show any interactions. We further adopted Glide QMꢀPolarized Ligand
Docking (QPLD) in Maestro, which can improve the docking accuracy over pure Molecular
Mechanics (MM) methods by calculating the partial charges of the ligand using quantum
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
3
mechanics (QM). The resulting docking modes showed that the Arg mimetic Dꢀregion
considerably moved toward the Glu327 of the hQC active site, denoting promising interactions
between them. These encouraging results prompted us to further optimize the proteinꢀligand
complexes.
The local optimization refinement made the side chain of Glu327 significantly move
within the range of a salt bridge interaction with the Dꢀregion of our ligands. To estimate the
global minimum, the proteinꢀligand complexes were further refined using Monte Carlo
ACS Paragon Plus Environment

Page 21 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
minimization by randomly sampling the discrete set of energy minima and accumulating the
4
4
energy contributions. Finally, scrupulous optimization of the proteinꢀligand complexes
produced a remarkable change in the position of the Glu327 side chain, allowing it to interact
with the Dꢀregion of the ligands (Supporting Information).
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
For the most potent compound, 191, the Dꢀregion displayed a salt bridge interaction as
well as Hꢀbonding with the side chain carboxylate of Glu327 (Figure 5). In the case of 212, the
Dꢀregion showed Hꢀbonding interaction in its neutral form and a salt bridge interaction in its
protonated form (Figure 6). In addition, the Dꢀregion of both compounds showed hydrophobic
interactions with Pro326 and Val328. The Aꢀregion maintained docking results such as zinc
chelation, Hꢀbonding, and hydrophobic interactions. The thiourea group in the Bꢀregion formed
an additional Hꢀbonding with Gln304. The phenyl ring in the Cꢀregion presented πꢀπ interactions
with Phe325. Moreover, aside from the previous hydrophobic interaction with Tyr299,
continuous refinements lead to the formation of additional hydrophobic interactions with Pro324
and Phe325.
In summary, our docking and refinement results showed that the designed and synthesized
inhibitors bind very well in the active site of the hQC via the aggregated interactions of the Aꢀ,
Bꢀ, Cꢀ, and Dꢀregions. Notably, the Arg mimetic Dꢀregions of the inhibitors were able to form
strong interactions with the carboxylate group of Glu327, providing a feasible explanation of how
our Arg mimetic compounds enhanced the inhibitory activity against hQC.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 66
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. Docked and refined structure of 191 in hQC. (A) Binding interactions of 191 at the
2
+
active site of hQC. 191 is displayed as sticks with magenta carbon atoms, and the Zn is a purple
ball. The interacted residues are depicted as light blue sticks. Hydrogen bonds are depicted as
black dashed lines. (B) 2D representation of the interactions of 191 with the active site residues
of the hQC. Hydrophobic interactions are marked in light brown. Hydrogen bonds are shown as
redꢀdotted arrows with their directionality. The πꢀπ stacking interaction is marked as a blue disc
and arrow, and the salt bridge interaction is displayed as a blue wedge line.
ACS Paragon Plus Environment

Page 23 of 66
Journal of Medicinal Chemistry
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 6. Docked and refined structures of 212 in hQC. Binding interactions of 212 in (A)
neutral form and (B) protonated form at the active site of the hQC. 212 is represented in sticks
2
+
with cyan carbon atoms, and Zn is a purple ball. The interacted residues are depicted as light
blue sticks. Hydrogen bonds are depicted as black dashed lines. 2D representation of 212
interactions in (C) neutral form and (D) protonated form with the active site of hQC.
Hydrophobic interactions are marked in light brown. Hydrogen bonds are shown as redꢀ and
greenꢀdotted arrows with their directionality. The πꢀπ stacking interaction is marked as a blue
disc and arrow, and the salt bridge interaction is displayed as a blue wedge line.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 66
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
■
CONCLUSION
We developed novel QC inhibitors that have the potential to serve as a novel therapeutic
option for AD. In particular, we aimed to further expand our library of QC inhibitors and
designed compounds with an extended scaffold based on the Nꢀterminal tripeptide (GluꢀPheꢀArg)
of Aβ3Eꢀ42. Inꢀdepth SAR studies identified the eight most potent inhibitors, which demonstrated
excellent QC inhibition, with from 5 to 40ꢀfold increases in potency compared to a known
inhibitor, compound 1. These inhibitors were further tested in acute model mice for Aβꢀlowering
effects in vivo, and compound 212 penetrated the BBB and effectively reduced the brain levels of
Aβ1ꢀ42 and AβN3pEꢀ42. Moreover, we evaluated the in vivo efficacy of 212 in two different
transgenic model mice of AD, APP/PS1 and 5xFAD, to verify the desired therapeutic effects.
Compound 212 not only reduced the brain concentrations of pyroform Aβ and total Aβ in
APP/PS1 mice but also restored cognitive functions in 5xFAD mice. This significant
enhancement of the overall activity was achieved by adding the Arg mimetic Dꢀregion, which
created an additional binding interaction. Stepwise computational studies utilizing quantum
mechanicsꢀbased molecular docking and Monte Carlo algorithmꢀbased refinement suggested that
the Arg mimetic Dꢀregions of 191 and 212 formed strong interactions with the carboxylate group
of Glu327, and both compounds demonstrated favorable binding interactions with all four
pharmacophoric regions. Taken together, compound 212 is an effective QC inhibitor for a
possible therapeutic alternative in AD. We believe that this study provides useful insights in
designing potent QC inhibitors in the future.
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 25 of 66
Journal of Medicinal Chemistry
1
2
3
■
EXPERIMENTAL SECTION
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
General. All chemical reagents were commercially available. Melting points were determined on
a melting point Buchi Bꢀ540 apparatus and are uncorrected. Silica gel column chromatography
was performed on silica gel 60, 230–400 mesh, Merck. The PLC plates used PLC silica gel 60
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
13
F , 1 mm, Merck. H and C NMR spectra were recorded on a JEOL JNMꢀLA 300 at 300 MHz
2
54
and 75 MHz, Bruker Analytik, DE/AVANCE Digital 400 at 400 MHz and 100 MHz, Bruker
Analytik, DE/AVANCE Digital 500 at 500 MHz and 125 MHz, JEOL JNMꢀECAꢀ600 at 600
MHz and 150 MHz, Ultrashielded Bruker Avance III HD NMR spectrometer at 800 MHz and
2
00 MHz, respectively. Chemical shifts are reported in ppm units with Me Si as a reference
4
standard. Mass spectra were recorded on a VG Trioꢀ2 GC−MS instrument and a 6460 Triple
Quad LC−MS instrument. All final compounds were purified to > 95% purity, as determined by
highꢀperformance liquid chromatography (HPLC). HPLC was performed on an Agilent 1120
Compact LC (G4288A) instrument using an Agilent TCꢀC18 column (4.6 mm × 250 mm, 5 ꢁm).
General procedure for Cꢀalkylated reaction (Procedure 1). A solution of a picoline derivative
(1 equiv) in anhydrous THF was added dropwise to an nꢀbutyllithium solution 2.5 M in THF (1.5
equiv for compounds 85 and 86; 2.5 equiv for compounds 87ꢀ89, 92; 3.5 equiv for 93) at ꢀ78°C
under nitrogen. The reaction mixture was stirred at room temperature for 30 minutes before being
cooled again to ꢀ78°C. 1ꢀBromoꢀ2ꢀchloroethane (1.1 equiv, compounds 85, 87), 1ꢀbromoꢀ3ꢀ
chloropropane (1.1 equiv, compound 86), (3ꢀbromopropoxy)ꢀ(tertꢀbutyl)dimethylsilane (1.1
equiv, compounds 88, 89) or diethyl carbonate (1.5 equiv, compounds 92, 93) in anhydrous THF
was added slowly to the reaction mixture. The reaction was monitored by TLC, and water was
added when the reaction was finished. The mixture was extracted by EtOAc, washed by water,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 66
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
dried over MgSO , and concentrated. The product was purified by silica gel chromatography with
4
a gradient of 10 to 25% EtOAc in nꢀhexane as the eluting solvent.
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
General procedure for Williamson reaction (Procedure 2). To a suspension of 4ꢀnitroguaiacol
2
(1.0 equiv) and potassium carbonate (compounds 3ꢀ5, 25ꢀ30, 96ꢀ98) or cesium carbonate
(compound 74) (2.0 equiv) in anhydrous DMF, an alkyl halide was added. The reaction mixture
was heated to 100°C for 1 hour and then cooled to room temperature and quenched by the
addition of water. The precipitated was formed, filtered, and washed with water (3 times). The
yellow solid product was collected, dried under vacuum and carried on to the next step without
further purification or the crude was purified by column chromatography.
General procedure for Nꢀalkylation (Procedure 3). To a solution of an amine (1.5 equiv) and
the same amine participating in the reaction (2.0 equiv) as a base (compounds 6ꢀ17) in anhydrous
DMF, alkyl halide (1.0 equiv) was added. The reaction mixture was heated to 60°C until the
starting material was consumed, and then it was cooled to room temperature and quenched by the
addition of water. The precipitate was formed, filtered, and washed with water (3 times) and the
solid product was collected, dried under vacuum and carried on to the next step without further
purification; or the crude was purified by column chromatography.
General procedure for the reduction of the nitro group to an amine group or alkyne
derivatives to alkyl derivatives (Procedure 4). The nitro compound or the alkyne compound
was dissolved in MeOH (or a mixture of MeOH and THF), and then 10% Pd/C was added. The
mixture was stirred at room temperature under hydrogen gas until the starting material was
consumed. The crude mixture was filtered through Celite, washed with methanol, and then
ACS Paragon Plus Environment

Page 27 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
concentrated. The product was carried on to the next step without further purification or was
purified by column chromatography.
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
General procedure for thiourea coupling (Procedure 5). To a solution of 1,1’ꢀ
thiocarbonyldiimidazole (TCDI) (1.02 equiv) in anhydrous DCM was added dropwise a solution
of arylamine (1.0 equiv) in anhydrous DCM under nitrogen gas at room temperature. The
reaction mixture was stirred at room temperature until the starting material was consumed, and
then a solution of 3ꢀ(5ꢀmethylꢀ1Hꢀimidazolꢀ1ꢀyl)propanꢀ1ꢀamine (1.1 equiv) in anhydrous DCM
was added dropwise, followed by the addition of triethylamine (3.0 equiv) and stirring at room
temperature until the reaction was complete (as monitored by TLC). The mixture was washed 2
times by water. The combined organic layer was dried over MgSO , concentrated, and purified
4
by column chromatography.
General procedure for Bocꢀprotection (Procedure 6). To a suspension of starting material
amine (1.0 equiv) in DCM or tertꢀbutyl alcohol (compounds 83, 84) was added triethylamine (1.2
equiv) and diꢀtertꢀbutyl dicarbonate in DCM under ice cooling. The mixture was stirred at room
temperature until the staring material was consumed. To this was added water, and it was
extracted with DCM. The organic layer was washed with 10% aqueous NaHCO solution, water,
3
and brine, dried over MgSO , filtered, and concentrated in vacuo. The residue was purified by
4
column chromatography.
General procedure for Bocꢀdeprotection (Procedure 7). Trifluoroacetic acid (10.0 equiv) was
added to the solution of bocꢀprotected compound (1.0 equiv) in DCM (DCM:TFA = 1:1 (v/v)).
The mixture was stirred at room temperature until the starting material was consumed, and then
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 66
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
the solvent was evaporated. The residue was dissolved in MeOH and purified by an ionꢀexchange
column or water was added to the residue, and then it was basified by 1 N NaOH and extracted
by DCM (2 times). The combined organic layer was washed with brine, dried over MgSO , and
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
purified by PLC or column chromatography to give the desired product.
General procedure for Mitsunobu reaction (Procedure 8). Triphenylphosphine (1.3 equiv)
was added under nitrogen to a solution of 4ꢀnitroguaiacol (2) (1.0 equiv) in DCM, followed by
the addition of a primary alcohol (1.2 equiv) and a solution of diethyl azodicarboxylate (1.3 equiv)
in DCM (2 mL). After the solution was stirred for 30 minutes at ambient temperature, the
reaction mixture was poured onto a column of silica and was eluted with EtOAc/nꢀhexane to give
the desired product.
General procedure for Sonogashira coupling reaction (Procedure 9). To a 3ꢀnecked round
bottom flask of a solution of aryl halide (1.0 equiv) in anhydrous DCM was added copper iodide
(5% mol), tetrakis(triphenylphosphine)palladium (0) (5% mol), triethylamine (3.0 equiv) and a
terminal alkyne (2.0 equiv) under argon, and the reaction was stirred at 50°C for 15 hours. The
reaction was diluted with EtOAc and washed with saturated aqueous NH Cl solution (twice), and
4
the combined aqueous layers were extracted with EtOAc (3 times). The combined organic layers
were washed with brine, dried over MgSO and concentrated. Column chromatography (nꢀ
4
hexane:EtOAc) yielded the desired product.
1ꢀ(4ꢀ(2ꢀAminoethoxy)ꢀ3ꢀmethoxyphenyl)ꢀ3ꢀ(3ꢀ(5ꢀmethylꢀ1Hꢀimidazolꢀ1ꢀyl)propyl)thiourea
o
1
(
169). From compound 146, procedure 7. yield 54%, white solid, mp 84ꢀ85 C. H NMR (300
MHz, CD OD) δ 7.59 (d, J = 1.11 Hz, 1H), 6.98 (d, J = 8.61 Hz, 1H), 6.94 (d, J = 2.22 Hz, 1H),
3
ACS Paragon Plus Environment

Page 29 of 66
Journal of Medicinal Chemistry
1
2
3
6.78 (dd, J = 8.43, 2.40 Hz, 1H), 6.66 (s, 1H), 4.04 (t, J = 5.31 Hz, 2H), 3.99 (t, J = 7.32 Hz, 2H),
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
.83 (s, 3H), 3.61 (t, J = 6.96 Hz, 2H), 3.00 (t, J = 5.31 Hz, 2H), 2.22 (d, J = 0.93 Hz, 3H), 2.07
1
3
(quintet, J = 7.14 Hz, 2H). C NMR (200 MHz, CD OD) δ 183.3, 152.0, 148.6, 138.8, 133.8,
3
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
29.8, 127.3, 119.9, 116.3, 112.0, 71.8, 57.3, 44.1, 43.8, 42.3, 32.0, 9.9. MS(FAB) m/z 364
+
+
[
M+H] . HRMS (FAB) calc. for C H N O S [M + H] 364.1807, found 364.1805.
17 25 5 2
1ꢀ(4ꢀ(3ꢀAminopropoxy)ꢀ3ꢀmethoxyphenyl)ꢀ3ꢀ(3ꢀ(5ꢀmethylꢀ1Hꢀimidazolꢀ1ꢀyl)propyl)ꢀ
thiourea (170). Compound 167 (1.0 equiv) was dissolved in ethanol and hydrazine monohydrate
5.0 equiv) was added dropwise. The mixture was stirred at room temperature until the starting
(
material consumed. The formed precipitate was filtered off and washed with ethanol. The filtrate
was collected and concentrated by rotary evaporation to afford the desired product as an off white
o
1
solid (48 mg, 42%), mp 78ꢀ79 C. H NMR (300 MHz, CD OD) δ 7.59 (s, 1H), 6.97 (d, J = 8.61
3
Hz, 1H), 6.91 (d, J = 2.19 Hz, 1H), 6.77 (dd, J = 8.43, 2.37 Hz, 1H), 6.66 (s, 1H), 4.09 (t, J =
6.06 Hz, 2H), 4.00 (t, J = 6.96 Hz, 2H), 3.81 (s, 3H), 3.62 (t, J = 6.57 Hz, 2H), 2.86 (t, J = 6.78
Hz, 2H), 2.22 (d, J = 0.93 Hz, 3H), 2.08 (quintet, J = 7.14 Hz, 2H), 1.96 (quintet, J = 6.24 Hz,
1
3
2H). C NMR (200 MHz, CD OD) δ 183.4, 151.9, 148.6, 138.7, 133.4, 129.8, 127.3, 119.9,
3
+
116.4, 112.0, 70.1, 57.2, 44.1, 43.8, 42.0, 32.0, 27.5, 9.9. MS(FAB) m/z 378 [M+H] . HRMS
+
(FAB) calc. for C H N O S [M + H] 378.1958, found 378.1965.
1
8
27
5
2
1ꢀ(4ꢀ(4ꢀAminobutoxy)ꢀ3ꢀmethoxyphenyl)ꢀ3ꢀ(3ꢀ(5ꢀmethylꢀ1Hꢀimidazolꢀ1ꢀyl)propyl)ꢀthiourea
(
171). Prepared from compound 168 by following the experiment procedure used for compound
70 to afford the desired product as a white solid (38 mg, 40%), the compound was decomposed
1
o
1
at 199 C. H NMR (300 MHz, CD OD) δ 7.59 (d, J = 0.93 Hz, 1H), 6.95ꢀ6.92 (m, 2H), 6.77 (dd,
3
ACS Paragon Plus Environment