Design, synthesis and evaluation of 2-amino-imidazol-4-one derivatives as potent β-site amyloid precursor protein cleaving enzyme 1 (BACE-1) inhibitors

Article abstract of DOI:10.1016/j.bmcl.2019.126772

Inhibition of β-site amyloid precursor protein cleaving enzyme 1 (BACE1) to prevent brain β-amyloid (Aβ) peptide's formation is a potential effective approach to treat Alzheimer's disease. In this report we described a structure-based optimization of a series of BACE1 inhibitors derived from an iminopyrimidinone scaffold W-41 (IC₅₀ = 7.1 μM) by Wyeth, which had good selectivity and brain permeability but low activity. The results showed that occupying the S₃ cavity of BACE1 enzyme could be an effective strategy to increase the biological activity, and five compounds exhibited stronger inhibitory activity and higher liposolubility than W-41, with L-5 was the most potent inhibitor against BACE1 (IC₅₀ = 0.12 μM, logP = 2.49).

Full text of DOI:10.1016/j.bmcl.2019.126772

Bioorganic&MedicinalChemistryLetters29(2019)126772
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design, synthesis and evaluation of 2-amino-imidazol-4-one derivatives as
potent β-site amyloid precursor protein cleaving enzyme 1 (BACE-1)
inhibitors
Tian-Yuan Fana, Wen-Yu Wub, Shao-Peng Yua, Yue Zhonga, Chao Zhaoa, Min Chena, He-Min Lia,
⁎
⁎
Nian-Guang Lia, , Zhi Chena, Sai Chena, Zhi-Hui Suna, Jin-Ao Duana, Zhi-Hao Shic,
a National and Local Collaborative Engineering Center of Chinese Medicinal Resources Industrialization and Formulae Innovative Medicine, Jiangsu Collaborative
Innovation Center of Chinese Medicinal Resources Industrialization, Jiangsu Key Laboratory for High Technology Research of TCM Formulae, Nanjing University of Chinese
Medicine, Nanjing 210023, China
b Department of Nuclear Medicine, Nanjing First Hospital, Nanjing Medical University, Nanjing 21006, China
c Department of Organic Chemistry, China Pharmaceutical University, Nanjing 211198, China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Inhibition of β-site amyloid precursor protein cleaving enzyme 1 (BACE1) to prevent brain β-amyloid (Aβ)
peptide’s formation is a potential effective approach to treat Alzheimer’s disease. In this report we described a
structure-based optimization of a series of BACE1 inhibitors derived from an iminopyrimidinone scaffold W-41
(IC50 = 7.1 μM) by Wyeth, which had good selectivity and brain permeability but low activity. The results
showed that occupying the S3 cavity of BACE1 enzyme could be an effective strategy to increase the biological
activity, and five compounds exhibited stronger inhibitory activity and higher liposolubility than W-41, with L-5
was the most potent inhibitor against BACE1 (IC50 = 0.12 μM, logP = 2.49).
Alzheimer’s disease
BACE1 inhibitors
Design and synthesis
Iminopyrimidinone scaffold
Structure–activity relationship
Alzheimer’s disease (AD), a chronic neurodegenerative disorder
predominantly occurs among the elderly, is the leading cause of de-
mentia.1 Current treatments include acetylcholinesterase inhibitors and
N-methyl-D-aspartate (NMDA) receptor antagonists which just tem-
porarily alleviate the cognitive decline in the initial phase of the disease
and do not significantly affect the underlying progress.2 According to
the β-amyloid (Aβ) hypothesis, the abnormal aggregation and deposi-
tion of Aβ, generated by the hydrolysis of amyloid precursor protein
which could be catalyzed by β-site amyloid precursor protein cleaving
enzyme 1 (BACE1), forms senile plaques with neurotoxicity and triggers
complex cascades,3 meanwhile it leads damage to synapses,4 those are
the main causes of AD. Therefore, processes that limit the production
and accumulation of Aβ by preventing the formation, inhibiting the
aggregation, or enhancing the clearance for Aβ may offer effective
treatments for AD. BACE1, as the first rate-limiting enzyme of the
amyloidogenic processing pathway, is considered to be a prominent
therapeutic target for treating AD by diminishing the formation of Aβ
peptide.5
early clinical trials.6 Regrettably, MK-8931’s clinical trials had been
terminated in 2018 because of its ineffective treatment of mild to
moderate Alzheimer's disease. AZD3293 (Lanabecestat; LY3314814)
(Fig. 1), which was reported by Astrazeneca and Eli Lilly in 2017 as an
orally active potent BACE1 inhibitor,7 was also terminated in phase III
clinical trial for the same reason like MK-8931. CNP520 (Fig. 1), which
was jointly developed by Novartis and Amgen in 2017, had a se-
lectivity, pharmacodynamics and suitable distribution profile for AD
prevention studies,8 is now in clinical phase III trial. JNJ-54861911
(Fig. 1), which was developed by Janssen in 2016, was a potent brain-
penetrant BACE1 inhibitor, achieved high and stable Aβ reductions
after single and multiple dosing in healthy participants.9 It is currently
undergoing phase II/III clinical trials. Elenbecestat (E-2609), which
was developed by Eisai and Biogen in 2014 and the structure was not
public. It was the first substance that had significant effect on Aβ level
in clinical studies of mild to moderate Alzheimer's disease.10 Elenbe-
cestat is currently undergoing phase III clinical trials and has passed
the 8th safety review of late.
Currently, multiple BACE-1 inhibitors have been studied in clinical
research. MK-8931 (Verubecestat) (Fig. 1), reported by Merck in
2016, could reduce the production of Aβ amyloid peptide by 79% in
Recently, Wyeth identified the diphenyl iminohydantoin compound
W-41 as
a potential BACE1 inhibitor with its IC50 was 7.1 μM
(Fig. 2).11–13 Despite W-41 had relatively weaker BACE1 inhibitory
⁎ Corresponding authors.
E-mail addresses: linianguang@njucm.edu.cn (N.-G. Li), sszh163@163.com (Z.-H. Shi).
https://doi.org/10.1016/j.bmcl.2019.126772
Received 15 April 2019; Received in revised form 29 September 2019; Accepted 19 October 2019
Availableonline28October2019
0960-894X/©2019ElsevierLtd.Allrightsreserved.

T.-Y. Fan, et al.
Bioorganic&MedicinalChemistryLetters29(2019)126772
NH
F
S
NH2
F
F
O
CN
F
F3C
Cl
O
O
N
N
N
HN
N
H
N
H
H
N
N
S O
O
N
N
N
NH2
N
N
O
O
F
F
N
F
NH2
JNJ-54861911
MK-8931
AZD3293
CNP520
Fig. 1. The chemical structures of BACE1 inhibitors that have entered clinical trials.
Fig. 2. The design of compounds L-1–L-23 based on the docking study of W-41 with BACE1 (PDB ID: 4DJU).
activity compared to other compounds, it exhibited substantially im-
proved selectivity against cathepsin D,14 which was closely related to
the side effects of BACE1 inhibitors in animal model,15 and improved
rat pharmacokinetic profile with a brain plasma ratio of 1.8 indicating
that W-41 had good brain permeability, so this inhibitor was appro-
priate to be a lead compound to improve its inhibitory activity against
BACE1.
that the right size of the introduced group was significant for occupying
the S3 cavity. Among the single phenyl-substituted derivatives, com-
pounds L-3, L-5, L-10–L-12, L-15, L-19–L-23 performed certain BACE1
inhibitory activity but the derivatives with only alkyl groups mod-
ification on the introduced benzene ring like L-4, L-6–L-9, L-14, L-16
showed opposite results except 2,4-dimethyl substituted derivative L-
17 with its IC50 against BACE1 was 0.42 μM. Meanwhile, compounds L-
5, L-10, L-15, L-20 and L-21, which contained biphenyl structure with
fluorine or methoxy substituted at 2 or 4 position, obviously exhibited
much stronger inhibitory activity against BACE1 than the positive
control LY2811376 and the parent analog W-41. These results in-
dicated that phenyl group might favor the interaction with the S3 cavity
but require assistance of certain groups like fluorine or methoxy which
were substituted at position 2 or 4. Interestingly, when the meta and
para position were both substituted by methoxy group, the obtained
compound L-5 exhibited the most potent inhibitory activity, with its
IC50 against BACE1 was 0.12 μM.
In this report, we firstly analyzed the X-ray structure of W-41
binding to the active site of BACE1 using DS 4.0. The results (Fig.2)
revealed that H-bonding interactions were formed between the terminal
amino group on guanidine moiety and Asp93, Asp289, which were
crucial for BACE1 binding.16 Meanwhile, the two phenyl groups occu-
pied S1 and S2’ pockets respectively. However, the adjacent S3 pocket
showed no interaction with lead compound W-41. We assumed that
introducing an additional R group at P3 section might facilitate the
binding ability between W-41 and BACE1 by occupying the S3 pocket.
Based on that attaching rigid linear fragments at P3 site was beneficial
for improving the BACE1 inhibitory activity,11 and our docking study
indicated that 3-biaryl substitution could occupy the contiguous S1-S3
pockets, we tried to link a rigid biaryl substitution at P3 position of W-
41 to occupy the S3 pocket. Therefore, a series of aminohydantoin de-
rivatives L-1–L-23 with aryl group substituted at the C-3 position were
designed and synthesized to enhance their binding ability with BACE1.
The synthetic routes for aminohydantoins L-1–L-23 were shown in
Scheme 1. The Sonogashira coupling of 1-bromo-3-iodobenzene (1)
with ethynylbenzene (2) in the presence of Pd(PPh3)4, CuI and PPh3 in
trimethylamine (TEA) afforded 1-bromo-3-(phenylethynyl)benzene (3)
in 96% yield. Then, it was oxidized by heating in DMSO at 140 °C with
Pd(PPh3)2Cl2 as a catalyst to produce the important intermediate T-1,
which could be further converted into the disubstituted diketones T-
2–T-23 through Suzuki coupling with different arylboronic acids, cat-
alyzed by Pd(PPh3)2Cl2 and K2CO3, in the mixed solvent of 1,4-dioxane
and H2O (10:1) at 100 °C. At last, the treatment of diketones T-1–T-23
with N-methylguanidine hydrochloride in EtOH, using TEA as a base at
80 °C, produced the desired products aminohydantoins L-1–L-23 in
excellent yields.
Subsequently, the logP values of all compounds were measured,
considering that they are closely related to the process of drug entering
the human body,17 and the topological polar surface area (tPSA) of all
the compounds were also calculated. Generally, central nervous system
drugs usually have higher liposolubility, with their logP values are
between 2 and 5,17 and small tPSA value which is lower than 90.18 The
results (Table 1) showed that all the compounds had good logP values
which were between 2.04 (L-1) and 3.18 (L-2), and small tPSA which
were below 77.15, indicating that they were suitable as central nervous
drugs for further research.
To help rationalize the aforementioned structure–activity relation-
ships, a model of BACE1 was constructed and the docking study was
carried out. The most potent compound L-5 was selected to perform the
molecular docking study (Fig. 3) with the X-ray structure of BACE1
enzyme (PDB ID: 4DJU), which was obtained from the Protein Data
Bank (RCSB PDB).
As expected, in the docking mode with BACE1 (Fig. 3), compound L-
5 deeply extended into the cavity of BACE1, and excellently combined
with the hydrophilic and hydrophobic surface. Firstly, the amino group
in amidazolidone moiety engaged in hydrogen bonding interactions
with Asp93 and Asp289, which were crucial for BACE1 binding.16
Secondly, the introduced phenyl structure successfully occupied the S3
hydrophobic cavity, through the hydrogen-bonding formed by Arg189
residue and the methoxy group on it. Thirdly, the docking mode
All the compounds were evaluated for their enzymatic inhibition
against BACE1 determined by fluorescence resonance energy transfer
(FRET) method. As shown in Table 1, the bromine substituted deriva-
tive L-1 showed no inhibitory activity against the enzyme, as the same
as multi phenyl substituted derivatives L-2, L-13 and L-18, indicating
2

T.-Y. Fan, et al.
Bioorganic&MedicinalChemistryLetters29(2019)126772
H2N
N
Br
O
N
O
Br
Br
a
b
d
Br
O
3
I
T-1
1
2
L-1
c
L-2: R=4-(C6H5)Ph
L-3: R=3,5-(OCH3)2Ph
L-4: R=4-(C2H5)Ph
L-13: R=(4-OBn)Ph
H2N
L-14: R=2,6-(CH3)2Ph
L-15: R=2,5-(OCH3)2Ph
L-16: R=3,5-(CH3)2Ph
L-17: R=3,4-(CH3)2Ph
L-18: R=3-(C6H5)Ph
L-19: R=2,3-(OCH3)2Ph
L-20: R=(5-F-2-OCH3)Ph
L-21: R=(2-F-3-OCH3)Ph
L-22: R=(5-F-2-CH3)Ph
L-23: R=(4-F-2-CH3)Ph
N
O
N
O
L-5: R=3,4-(OCH3)2Ph
L-6: R=2,4-(CH3)2Ph
L-7: R=4-(C(CH3)3)Ph
L-8: R=2,5-(CH3)2Ph
L-9: R=2,4,6-(CH3)3Ph
L-10: R=(2-F-4-CH3)Ph
L-11: R=2,4-(OCH3)2Ph
L-12: R=2,6-(OCH3)2Ph
R
d
R
O
T-2~T-23
L-2~L-23
Scheme 1. Reagents and conditions: a) Pd(PPh3)4 (4 mol %), CuI (2 mol %), PPh3 (8 mol %), TEA, 50 °C, 16 h, 96%; b) Pd(PPh3)2Cl2 (5 mol %), DMSO, 140 °C, 5 h,
60%; c) RB(OH)2 (2 equiv.), Pd(PPh3)2Cl2 (10 mol %), K2CO3 (3 equiv.), 1,4-dioxane, H2O, 100 °C, 3 h, 79%–93%; d) N-methylguanidine hydrochloride (1 equiv.),
TEA (3.5 equiv.), EtOH, 80 °C, 10 h, 64%-74%.
confirmed that the biphenyl motif could occupy S1 and S3 pocket re-
spectively.
modified with appropriate groups like fluorine or methoxy was effec-
tive to improve the BACE1 inhibitory activity. Furthermore, all the
compounds showed good logP values and tPSA. These compounds
could be used as potential BACE1 inhibitors for AD treatment and
precursors for further modification. Subsequent reports will detail fur-
ther optimized efforts to obtain inhibitors with improved cell and in vivo
potency.
In summary, using W-41 as a lead compound, a novel class of
aminohydantoins has been designed and synthesized in order to im-
prove the BACE1 inhibitory activities, five analogs L-5, L-10, L-15, L-20
and L-21 showed excellent inhibitory activity against BACE1, with L-5
was the most potent inhibitor (IC50 = 0.12 μM). The results proved that
occupying the S3 pocket by incorporating phenyl group at 3-position
Table 1
The inhibitory activity against BACE1 and physical property of compounds L-1–L-23.
Compound
R
Inhibition rate (%)
(at 10 μM)
Activity (IC50, μM)
logP
tPSAa
L-1
Br
33.13
43.90
80.74
43.53
97.45
40.86
34.21
42.07
45.73
96.99
75.49
82.11
37.70
34.15
95.05
47.16
88.31
20.54
78.63
95.97
94.67
92.03
87.70
62.19
94.22
0.36
1.37
0.56
0.26
0.54
0.82
2.27
5.08
3.86
0.06
0.86
0.05
0.93
0.62
0.61
1.42
0.50
3.72
0.48
0.31
2.39
0.49
0.75
0.22
2.46
> 10
> 10
2.04
3.18
2.41
2.56
2.49
2.51
3.10
2.52
2.55
2.97
2.46
2.02
2.99
2.42
2.54
2.56
2.49
3.14
2.15
2.86
2.72
2.35
2.34
1.90
–
58.69
58.69
77.15
58.69
77.15
58.69
58.69
58.69
58.69
58.69
77.15
77.15
67.92
58.69
77.15
58.69
58.69
58.69
77.15
67.92
67.92
58.69
58.69
58.69
–
L-2
4-(C6H5)Ph
3,5-(OCH3)2Ph
4-(C2H5)Ph
3,4-(OCH3)2Ph
2,4-(CH3)2Ph
4-(C(CH3)3)Ph
2,5-(CH3)2Ph
2,4,6-(CH3)3Ph
(2-F-4-CH3)Ph
2,4-(OCH3)2Ph
2,6-(OCH3)2Ph
(4-OBn)Ph
L-3
0.8512
> 10
0.3445
0.0007
L-4
L-5
0.1194
> 10
L-6
L-7
> 10
L-8
> 10
L-9
> 10
L-10
L-11
L-12
L-13
L-14
L-15
L-16
L-17
L-18
L-19
L-20
L-21
L-22
L-23
W-41
LY2811376
0.1414
1.5750
0.7205
> 10
0.0046
0.0070
0.0766
2,6-(CH3)2Ph
2,5-(OCH3)2Ph
3,5-(CH3)2Ph
3,4-(CH3)2Ph
3-(C6H5)Ph
2,3-(OCH3)2Ph
(5-F-2-OCH3)Ph
(2-F-3-OCH3)Ph
(5-F-2-CH3)Ph
(4-F-2-CH3)Ph
–
> 10
0.1953
> 10
0.0241
0.0091
0.4249
> 10
1.1090
0.1735
0.2204
0.3281
0.4562
3.4840
0.2693
0.0121
0.0049
0.0005
0.0180
0.0104
0.0060
0.0060
–
a
tPSA were calculated by ChemProp.
3

T.-Y. Fan, et al.
Bioorganic&MedicinalChemistryLetters29(2019)126772
Project Funded by Jiangsu Province (BRA2016387), Project Funded by
the Priority Academic Program Development of Jiangsu Higher
Education Institutions, Project Funded by the Flagship Major
Asp289
Development
of
Jiangsu
Higher
Education
Institutions
(PPZY2015A070), State Key Laboratory Cultivation Base for TCM
Quality and Efficacy, Nanjing University of Chinese Medicine.
Asp93
Appendix A. Supplementary data
S3
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bmcl.2019.126772.
S1
S2'
Arg189
References
Tyr132
1. Wu WY, Dai YC, Li NG, et al. J Enzyme Inhib Med Chem. 2017;32:572.
2. Blennow K, Jde ME, Zetterberg H. Lancet. 2006;368:387.
3. Hardy JA, Higgins GA. Science. 1992;256:184.
Fig. 3. Interactions between compound L-5 and BACE1 (PDB ID: 4DJU).
4. Lovella MA, Robertsona JD, Teesdale WJ, Campbell JL, Markesbery WR. J Neurol Sci.
1998;158:47.
5. Vassar R, Kandalepas PC. Alzheimer's Res Ther. 2011;3:20.
6. Kennedy ME, Stamford AW, Chen X, et al. Sci Transl Med. 2016;8:363ra150.
7. Sakamoto K, Matsuki S, Matsuguma K, et al. J Clin Pharmacol. 2017;57:1460.
8. Panza F, Lozupone M, Solfrizzi V, et al. Expert Rev Neurother. 2018;18:847.
9. Timmers M, Van BB, Ramael S, et al. Alzheimer's Dementia. 2016;2:202.
10. Mason VL. Drugs Today. 2015;51:447.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
11. Zhu Z, Sun ZY, Ye Y, et al. J Med Chem. 2010;53:951.
12. Malamas MS, Robichaud A, Erdei J, et al. Bioorg Med Chem Lett. 2010;20:6597.
13. Wang YS, Strickland C, Voigt JH, et al. J Med Chem. 2010;53:942.
14. Yasuda Y, Kageyama T, Akamine A, et al. J Biochem (Tokyo). 1999;125:1137.
15. Andrea MZ, Charles EN, Michael AB, et al. Nat Commun. 2016;7:13042.
16. Cumming JN, Smith EM, Wang L, et al. Bioorg Med Chem Lett. 2012;22:2444.
17. Hitchcock SA, Pennington LD. J Med Chem. 2006;49:7559.
18. Ghose AK, Herbertz T, Hudkins RL, Dorsey BD, Mallamo JP. ACS Chem Neurosci.
2012;3:50.
Acknowledgement
This work was financially supported by National Natural Science
Foundation of China (81973171), Natural Science Foundation of
Jiangsu Province (BK20151563), 333 High Level Talents Training
4