Discovery of novel 2-(alkylmorpholin-4-yl)-6-(3-fluoropyridin-4-yl)-pyrimidin-4(3H)-ones as orally-active GSK-3β inhibitors for Alzheimer's disease

Article abstract of DOI:10.1016/j.bmcl.2015.01.005

We herein describe the results of further evolution of GSK-3β inhibitors for Alzheimer's disease from our promising compounds with in vivo tau phosphorylation inhibitory activity by oral administration. Introduction of a low alkyl group instead of the phenyl group at the 3-position of the morpholine moiety aiming to improve pharmacokinetic profiles resulted in potent low molecular weight GSK-3β inhibitors with good in vitro pharmacokinetic profiles, which also showed in vivo tau phosphorylation inhibitory activity by oral administration. Effect of the stereochemistry of the alkyl moiety is also discussed using docking models.

Full text of DOI:10.1016/j.bmcl.2015.01.005

Bioorganic & Medicinal Chemistry Letters 25 (2015) 1086–1091
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of novel 2-(alkylmorpholin-4-yl)-6-(3-fluoropyridin-4-yl)-
pyrimidin-4(3H)-ones as orally-active GSK-3b inhibitors for
Alzheimer’s disease
,
⇑
Kenji Fukunaga, Daiki Sakai, Kazutoshi Watanabe , Kazuki Nakayama, Toshiyuki Kohara, Hiroshi Tanaka,
Shinji Sunada, Mika Nabeno, Masako Okamoto , Ken-Ichi Saito à, Jun-ichi Eguchi §, Akiko Mori –,
Shinji Tanaka, Keiko Inazawa, Takashi Horikawa
Research Division, Mitsubishi Tanabe Pharma Corporation, Kamoshida-cho, Aoba-ku, Yokohama 227-0033, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
We herein describe the results of further evolution of GSK-3b inhibitors for Alzheimer’s disease from our
promising compounds with in vivo tau phosphorylation inhibitory activity by oral administration.
Introduction of a low alkyl group instead of the phenyl group at the 3-position of the morpholine moiety
aiming to improve pharmacokinetic profiles resulted in potent low molecular weight GSK-3b inhibitors
with good in vitro pharmacokinetic profiles, which also showed in vivo tau phosphorylation inhibitory
activity by oral administration. Effect of the stereochemistry of the alkyl moiety is also discussed using
docking models.
Received 16 October 2014
Revised 24 December 2014
Accepted 5 January 2015
Available online 13 January 2015
Keywords:
Glycogen synthase kinase 3b
Alzheimer’s disease
Alkylmorpholine
Ó 2015 Elsevier Ltd. All rights reserved.
Alzheimer’s disease (AD) is
a
neurodegenerative disorder
to. GSK-3b is a promising therapeutic target against AD due to the
increase in activity in AD brains and the involvement in the accel-
eration of tau pathology.1–4
Recently we reported the discovery of a GSK-3b inhibitor for AD,
UDA-680 which exhibited potent GSK-3b inhibitory activity with
high kinase selectivity and in vivo tau phosphorylation inhibitory
activity by oral administration in mice. Pharmacokinetic studies
indicated that UDA-680 had high brain/plasma ratio with moder-
ate bioavailability (37%, rat).5
In order to identify further promising compounds for clinical
trials, evolution of UDA-680 was planned. UDA-680 had an accept-
able level of bioavailability and water solubility, which, however,
remained to be improved. It is already known that reducing the
number of carbon atoms of a phenyl group according to a so-called
‘carbon-cutting approach’ may improve bioavailability,6 and we
tried to transform the phenyl group into a small alkyl group to
evolve a new chemical series of GSK-3b inhibitors. A 3-fluoropyri-
din-4-yl group was selected as the substituent at the 6-position of
the pyrimidone moiety in order to compensate the loss of a
reported by Dr. A. Alzheimer in 1907. Patients of AD show progres-
sive and irreversible memory loss and cognitive impairment with
psychiatric symptoms such as psychosis, hallucinations, depres-
sion, anxiety, sleep disorder and aberrant behavior disturbance.
By the progression of the disease, patients are gradually impaired
cognitively and functionally and it is difficult for them to live alone,
and finally caregivers will make all basic activity of daily living of
the patients. These facts make tremendous burdens physically,
economically and mentally to the patients’ family, and also afford
a great economic loss in worldwide. Pathology of AD shows three
major characteristics: (i) massive neuronal loss in hippocampus
as well as frontal and temporal cortices, (ii) neurofibrillary tangles
(NFTs), and (iii) senile plaques (SP). NFTs are mainly composed of
abnormally hyperphosphorylated microtubule-associated protein
tau, the hyperphosphorylation of which GSK-3b mainly contributes
⇑
Corresponding author.
cation–
p interaction between the phenyl group and Arg141 in
Present address: Research Division, Mitsubishi Tanabe Pharma Corporation, 2-2-
50, Kawagishi, Toda-shi, Saitama 335-8505, Japan.
Present address: Life Science Institute, Inc., THE KAITEKI Bldg., 13-4, Uchikanda 1-
chome, Chiyoda-ku, Tokyo 101-0047, Japan.
Present address: The KAITEKI Institute, Inc., 1-1, Marunouchi 1-chome, Chiyoda-
ku, Tokyo 100-8251, Japan.
Present address: Global product strategy Department, Mitsubishi Tanabe Pharma
Corporation, 17–10, Nihonbashi-Koamicho, Chuo-ku, Tokyo 103-8405, Japan.
the hydrophobic site of GSK-3b (Fig. 1).5 We envisioned that this
modification would afford increased hydrophilicity, reduced
molecular weight and resulting improvement of in vivo pharmaco-
logical profiles.
à
§
Preliminary results of the transformation of the phenyl group
–
into alkyl groups are shown in Table 1.7 Both a 2-alkyl group and
http://dx.doi.org/10.1016/j.bmcl.2015.01.005
0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.

K. Fukunaga et al. / Bioorg. Med. Chem. Lett. 25 (2015) 1086–1091
1087
GSK-3b was performed.8 Figure 2 shows that the nitrogen atom
of the 3-fluoropyridyl group makes hydrogen bonding with the
main chain of Val135 and the carbonyl oxygen of the pyrimidone
moiety forms a hydrogen bond with the side chain of Lys85. The
N
N
N
N
F
F
N
N
1
R
R ; small alkyl
N
O
N
N
O
O
Me
O
Me
morpholine moiety of each compound formed a CH–p interaction
with the side chain of Phe67,9 but the binding area of each methyl
substituent is different. The methyl group of 6 is buried in the
hydrophobic pocket of the ceiling of the active site and that of 9
just looks toward the floor presumably with no interaction with
the active site, which resulted in several-fold difference in activity.
A nearly 10-fold increase of activity by introducing a methyl group
to unsubstituted 2 suggests that this methyl group would confer
so-called magic methyl effect by the hydrophobic interaction.10
Docking studies of 18 and 19 were also performed based on these
results. (Fig. 3) The fluoropyridine and pyrimidone moieties make
almost the same interactions as the methyl group of 6, and the
fused-morpholine moieties of 18 and 19 are located at different
binding positions. The morpholine moiety of 18 is positioned closer
to the side chain of Phe67.11 The position of the morpholine moiety
of 18 seems more appropriate to interact with GSK-3b than that of
19, and therefore 18 shows more potent activity than 19. Signifi-
cant difference of activity of 21 and its (+)-isomer was also
analyzed by a docking study. The (4aR,7aS) and (4aS,7aR) isomers
(Fig. 4), which have the same configuration as 18 and 19,
respectively, locate at a similar binding position to 18 and 19,
respectively. In this case, the distance between the CH with
Phe67 and the distance between the morpholine moiety with
Phe67 are significantly different,12 which would afford large differ-
ences in activity between each stereoisomer.
UDA-680
Figure 1. Structure of UDA-680 and its transformation.
a 3-alkyl group increased in vitro activity compared to unsubstitut-
ed morpholine 2. Compounds with 2-alkyl groups showed the
same level of in vitro activity as UDA-680 accompanying potent
CYP1A2 inhibition and poor metabolic stability. In the case of
compounds with a 3-alkyl group, a 3-(R)-isomer also showed the
same level of potency as UDA-680, and in vitro activity of the
(S)-isomer was lower than that of the (R)-isomer. Compound 6
with a 3-(R)-Me group was particularly interesting as it had the
most potent activity and the smallest molecular weight coupled
with better metabolic stability and moderate CYP inhibition com-
parable to UDA-680.
We further sought to increase inhibitory activity by introducing
another methyl group at the 2- or 6-position (Table 2). Introduc-
tion of a 6-methyl group (11 and 12) decreased activity regardless
of its configuration, especially for (3R,6R)-isomer 11. In the case of
2-methyl group, no increase of activity was observed in both cis
and trans isomers except 16 which slightly increased the activity.
2,2,3-Trimethyl isomer 17 also maintained the activity. These
results indicated that the configuration of the 3-methyl group
affected the activity and the methyl group at the 2-position made
no significant contribution to increase the activity.
Fused cyclic moieties designed by the connection of the two
methyl groups of 16 were installed on the morpholine moiety to
increase potency by the conformational restriction of the morpho-
line moiety (Table 3). All fused-morpholine isomers, 18, 20 and 21,
possessing a cyclopentane, cyclohexane and tetrahydrofuran moi-
ety, respectively, showed increased activity. Compound 19, the
optical isomer of 18 showed decreased activity, again suggesting
that the configuration of the carbon atom adjacent to the nitrogen
atom of the morpholine ring affected the activity.
Table 4 listed the promising compounds by the ‘carbon-cutting
approach’. All of the compounds had more potent inhibitory activ-
ity than UDA-680 and possessed good metabolic stability, CYP
inhibition profiles and excellent cell permeability. Ligand efficiency
(LE) of these compounds was increased to 0.35 to 0.37 from 0.29 of
UDA-680.13 Lipophilic ligand efficiency (LLE) of these compounds
was between 7.05 and 8.92 which are more preferable scores than
that of UDA-680 (LLE; 7.21) except 20.14 These results showed that
the ‘carbon-cutting approach’ was effective for the evolution of a
new subseries with a well-balanced LE-LLE value as oral drugs
compared to UDA-680.15 Furthermore, the number of aromatic
rings were decreased from three of UDA-680 to two of the promis-
ing compounds in Table 4. This also suggested that the promising
In order to analyze the effects of the configuration of the carbon
atom of the morpholine moiety, docking studies of 6 and 9 with
Table 1
Introduction of small alkyl group
N
F
N
3
R
N
O
N
2O
Me
a
Compd
R
GSK-3b IC50 (nM)
CLint,
(mL/min/mg)
CYP inhibitionb (IC50
2D6
, lM)
human
1A2
3A4
2
3
4
5
6
7
8
9
H
67
17
12
12
7.0
19
14
23
30
12
—
—
—
>50.0
2.3
37.7
>50.0
—
—
>50.0
—
—
2-(RS)-i-Pr
2-(RS)-n-Bu
2-(RS)-cyc-Bu
3-(R)-Me
3-(R)-Et
3-(R)-i-Pr
3-(S)-Me
3-(S)-Et
0.07
0.21
0.20
0.03
—
—
0.02
—
3.3
7.5
1.5
18.7
—
—
5.7
—
>50.0
>50.0
>50.0
>50.0
—
—
>50.0
—
10
UDA-680
0.06
29.4
39.7
>50.0
a
Human liver microsome assay.
Recombinant human CYP450 isoform inhibition for 1A2, 2D6 and 3A4.
b

1088
K. Fukunaga et al. / Bioorg. Med. Chem. Lett. 25 (2015) 1086–1091
Table 2
Introduction of additional methyl group
N
F
N
3
R
N
O
N
2O
Me
Compd
Morpholine
GSK-3b IC50 (nM)
Compd
Morpholine
GSK-3b IC50 (nM)
Me
Me
O
Me
N
O
N
11
453
41
15
19
Me
Me
N
Me Me
O
O
N
12
16
17
4.9
9.7
Me
Me
Me
N
Me
Me
N
Me
O
13
14
16
32
(RS)
(+)-isomer
(-)-isomer
O
Table 3
Fused-morpholine analogues
N
F
Me N
N
Me
N
O
Me
O
16
Compd
Morpholine
GSK-3b IC50 (nM)
Compd
Morpholine
GSK-3b IC50 (nM)
(R)
O
(R)
N
(R)
O
(R)
N
18
4.2
20
3.7
O
(S)
O
(S)
N
19
35
21
3.1a
(-)
O
N
a
We also prepared (+)-isomer, which IC50 of in vitro activity was >1000 nM.
Figure 2. Docking model of 6 (cyan carbon) and 9 (orange carbon) with GSK-3b (PDB code; 3F88, gray carbon). Left figure; interactions between 6, 9 and GSK-3b. Right figure;
binding position of 6 and 9 in GSK-3b seen from the apertural area.
compounds by the ‘carbon-cutting approach’ possess preferable
profiles as oral drug candidates.16
one hour after administration at a dose of 10 mg/kg or higher doses
and that its ED50 value was 6.8 mg/kg p.o., 1 h.
Among the compounds in the table, 6 was selected to confirm
the effect of the simple methyl group by the ‘carbon-cutting
approach’, and we evaluated its inhibitory effects on in vivo
Ser396 tau phosphorylation in mice.17,18 (Fig. 5) A dose response
study showed that 6 significantly decreased tau phosphorylation
2-(Alkyl-substituted morpholin-4-yl)-pyrimidin-4-ones 3–21
were synthesized by the condensation of appropriately substituted
morpholines and 2-chloropyrimidin-4-one 26 in the presence of tri-
ethylamine (Scheme 1). 3-Flurorpyridin-4-carboxylic acid (23) was
prepared from 3-fluoropyridine (22) by ortho-lithiation using

K. Fukunaga et al. / Bioorg. Med. Chem. Lett. 25 (2015) 1086–1091
1089
Preparation of di- and trimethylmorpholines is shown in
Schemes 3–5. 2,5-Dimethylmorpholine 35 was prepared according
to Scheme 3 using 2-chloropropionyl chloride as the acyl halide.
Acylation of aminoalcohol 32 with 2-chloropropionyl chloride
followed by intramolecular alkylation afforded lactam 33 as a
mixture of diastereomers. Reduction of 33 and separation of each
isomer by silica gel column chromatography yielded N-protected
morpholine 34,22 which was deprotected to give 2,5-dimethylmor-
pholine 35.
Preparation of trans-2,3-dimethylmorpholine was depicted in
Scheme 4. Ring opening of epoxide 36 by (S)-1-phenethylamine
in the presence of lithium perchlorate afforded aminoalcohol 37
as a mixture of stereoisomers. After cyclization, the resulting lac-
tam was washed with hexane and diethyl ether, and recrystalliza-
tion from dichloromethane/hexane afforded lactam 38 with
(2R,3R) stereochemistry.23 Reduction and successive deprotection
of 1-phenethyl moiety yielded trans-(2R,3R)-dimethylmorpholine
40. Trans-(2S,3S)-isomer was also prepared by this route by using
(S)-1-phenethylamine instead of its (R)-isomer.
Figure 3. Docking model of 18 (cyan carbon) and 19 (orange carbon) with GSK-3b
(gray carbon).
Scheme 5 shows the preparation of cis-2,3-dimethylmorpho-
line. The sequence began with a ring opening of (RS)-epoxide 41
by (S)-1-phenethylamine and the resulting aminoalcohol 42 was
elaborated to morpholine 44, which was separated to each iso-
mer24 by silica gel column chromatography. Deprotection of the
1-methybenzyl group afforded cis-2,3-dimethylmorpholine 45.
2,2,3-Trimethylmorpholine was also synthesized by this sequence
using an amino alcohol derived from 2,2,3-trimethyloxirane and
benzylamine.
N
N
F
F
O
O
N
N
N
N
N
O
N
O
Me
Me
O
O
(4aR, 7aS)
(4aS, 7aR)
Figure 4. Chemical structure of 21 and its stereoisomer.
Preparation of a fused-morpholine such as 49 was depicted in
Scheme 6. Aminoalcohol 4625 prepared from cyclopenteneoxide
and (R)-1-phenethylamine was acylated, cyclized, reduced and
deprotected to yield morpholine 49. Morpholines fused with
cyclohexane and tetrahydrofuran were also prepared from amino-
alcohols derived from (R)-1-phenethylamine and cyclohexene
oxide or dihydrofuran oxide. In the case of tetrahydrofuran-fused
morpholine, each stereoisomer was separated after reduction of
lactam intermediate.
lithium diisopropylamide and addition of dry ice. Compound 23 was
transformed into the corresponding b-keto ester 24 with potassium
monoethylmalonate, carbonyl diimidazole and magnesium chlo-
ride.19 Condensation of 24 and N-methylthiourea afforded pyrimi-
dine-4(3H)-one 25, which was converted to chloropyrimidone 26
by using phosphorous oxychloride and N,N-dimethylformamide.
Preparation of 3-alkylmorpholines 31 is depicted in Scheme 2.
N-protected aminoalcohol 28, prepared by reductive alkylation of
optically active aminoalcohol 27, was acylated by chloroacetyl
chloride and successive intramolecular alkylation yielded lactam
29. Reduction of 29 by diborane generated in situ20 and deprotec-
tion of benzyl group of 30 by 1-chloroethyl chlorformate21 afforded
3-alkylmorpholine 31 as hydrochloric acid salts.
In conclusion, transformation of the phenyl group of morpho-
line moiety of UDA-680 into a small alkyl group was well tolerated
and led to the discovery of a novel alkylmorpholine series of potent
GSK-3b inhibitors. Extending these findings identified
a
conformationally restrained fused-morpholine series. Several
Table 4
Promising compounds
a
Compd
Morpholine
GSK-3b IC50 (nM)
CLint,
0.03
(mL/min/mg)
CYP inhibitionb (IC50
,
l
M)
Caco-2 (10À7 cm/s)
clogP
LE
LLE
human
1A2
2D6
3A4
Me
6
7.0
19
>50
>50
546
0.47
0.37
7.68
N
O
Me
Me
N
16
18
20
4.9
4.2
3.7
n.t.
16
21
14
>50
>50
>50
>50
>50
>50
485
445
418
0.99
0.83
1.38
0.36
0.35
0.34
7.32
7.55
7.05
O
0.04
0.20
O
N
O
N
O
21
3.1
0.02
>50
>50
>50
545
À0.41
0.35
8.92
(-)
O
N
n.t.; not tested.
LE; ligand efficiency.
LLE; lipophilic ligand efficiency.
a
Human liver microsome assay.
Recombinant human CYP450 isoform inhibition for 1A2, 2D6 and 3A4.
b

1090
K. Fukunaga et al. / Bioorg. Med. Chem. Lett. 25 (2015) 1086–1091
Figure 5. Dose–response of 6 for decrease of phosphorylated tau protein 1 h after administration. Data represents mean SE (N = 6). v: vehicle. 680: UDA-680 100 mg/kg p.o.
treated as a positive control. ###p <0.001 versus vehicle treated group by Student’s t-test. ⁄⁄⁄p <0.001 versus vehicle treated group by Dunnett’s test.
N
Me Me
Ph
Me Me
Me
O
36
a
b
c
Me
*
Me
HO
Me
Me
*
N
COOH
F
O
F
O
N
Ph
F
a
F
b
c
O
H
O
OEt
N
37
N
N
38
N
HS
N
O
Me
22
23
24
Me Me
Ph
Me
NH
25
Me
d
N
N
N
O
O
HCl
F
F
39
40
d
e
N
N
R
O
N
N O
Cl
N
O
Scheme 4. Reagents and conditions: (a) (S)-1-phenethylamine, LiClO4, MeCN,
reflux, 12 h; (b) (i) ClCH2COCl, Et3N, CH2Cl2, 0 °C, 1 h, (ii) KOH, i-PrOH, 0 °C to rt, 3 h,
(iii) wash with hexane and Et2O then recryst. from CH2Cl2/hexane; 19%, 4 steps (c)
LiBH4, Me3SiCl, THF, 0 °C to rt, 3 h, 62%; (d) (i) ClCOOCH(Cl)CH3, ClCH2CH2Cl, reflux,
10 h, (ii) MeOH, reflux, 7 h, 69%.
Me
Me
26
1
Scheme 1. Reagents and conditions: (a) THF, LDA, À78 °C, 1 h then add s.CO2,
À78 °C to rt, 1 h, 70%; (b) carbonyldiimidazole, THF, reflux, 1 h then add
EtOOCCH2COOK, MgCl2, 60 °C, 66%; (c) N-methylthiourea, DBU, toluene, 110 °C,
5 h, 68%; (d) POCl3, DMF, 60 °C, 2 h, 73%; (e) morpholines, Et3N, THF, 37–75%.
Me
O
Me Me
Me Me
a
b
c
*
H
*
Me
Me
Me
*
OH
*
O
N
Ph
N
Ph
R
O
R
R
(RS)
41
a
b
c
42
43
N
H
NH2
N
OH
OH
27
O
OMe
R
O
OMe
28
Me Me
Me
29
*
Me
d
*
*
O
Me
*
N
Ph
NH
O
HCl
R
44
d
45
N
NH
O
O
OMe
HCl
Scheme 5. Reagents and conditions: (a) (S)-1-phenethylamine, LiClO4, MeCN,
reflux, 10 h; (b) (i) ClCH2COCl, Et3N, THF, 0 °C, 1 h, (ii) KOH, i-PrOH, 0 °C to rt, 4 h,
77% (3 steps, mixture of isomer); (c) LiBH4, Me3SiCl, THF, 0 °C to rt, 6 h, then
separation of isomer, 18% (cis, upper), 11% (cis, lower); (d) (i) ClCOOCH(Cl)CH3,
130 °C (neat), (ii) MeOH, reflux, 7 h.
30
31
Scheme 2. Reagents and conditions: (a) 4-methoxybenzaldehyde, NaBH4, MeOH,
0 °C to rt, 1 h, 100% (R = Me); (b) ClCH2COCl, Et3N, THF, 1 h then add NaH, 0 °C, 63%
(R = Me), 82% (R = Et, 2 steps), 59% (R = i-Pr, 2 steps); (c) LiBH4, Me3SiCl, THF, rt, 1 h,
96% (R = Me), 69% (R = Et), 75% (R = i-Pr); (d) (i) ClCOOCH(Cl)CH3, ClCH2CH2Cl, 80 °C,
6 h, (ii) MeOH, reflux, 1 h, 74% (R = Me), 89% (R = Et), 58% (R = i-Pr).
Me
Ph
O
Me
Ph
Me
Ph
a
b
c
N
N
NH
Me
N
Me
N
N
H
46
Me
O
O
O
a
b
HO
HCl
HO
N
O
O
47
48
49
H
O
OMe
OMe
*
OMe
Me
Me
32
34
33
Scheme 6. Reagents and conditions: (a) (i) ClCH2COCl, Et3N, THF, 0 °C, 30 min then
add NaOMe in MeOH, 0 °C to rt, overnight, 88%; (b) LiBH4, Me3SiCl, THF, 0 °C to rt,
2 h, 91%; (c) (i) ClCOOCH(Cl)CH3, ClCH2CH2Cl, reflux, 10 h, (ii) MeOH, reflux, 1 h,
58%.
Me
c
NH
O
*
HCl
Me
35
potent compounds exhibited good in vitro pharmacokinetic pro-
files, and 6 inhibited in vivo tau phosphorylation in mice by oral
administration. LE and LLE score of the promising compounds are
better than those of UDA-680, which indicated that this novel
new chemical series would bind to GSK-3b more effectively than
Scheme 3. Reagents and conditions: (a) MeCH(Cl)COCl, Et3N, THF then add NaOMe
in MeOH, 0 °C to rt, overnight, 93%; (b) LiBH4, Me3SiCl, THF, 0 °C to rt, 2 h,
separation of isomer, 13% (cis), 29% (trans); (c) (i) ClCOOCH(Cl)CH3, ClCH2CH2Cl,
reflux, 2 h, (ii) MeOH, reflux, 2 h, 52% (cis), 21% (trans).

K. Fukunaga et al. / Bioorg. Med. Chem. Lett. 25 (2015) 1086–1091
1091
10. (a) Schönherr, H.; Cernak, T. Angew. Chem., Int. Ed. 2013, 52, 12256; (b) Leung,
C. S.; Leung, S. S. F.; Tirado-Rives, J.; Jorgensen, W. L. J. Med. Chem. 2012, 55,
4489.
UDA-680. Further optimization of this series will be published
elsewhere.
11. The distances between centers of the morpholine moiety and the benzene ring
of Phe67 of 18 and 19 are 4.7 Å and 5.1 Å, respectively. In this case the distance
Acknowledgments
between the CH–p and Phe67 is the same between 18 and 19 (4.2 Å for 18 and
4.2 Å for 19).
We gratefully thank Sanofi-Aventis R&D CNS Research Depart-
ment (formerly Sanofi-Synthelabo R&D, Bagneux, France) for a
fruitful collaboration.
12. The distances between the CH with Phe67 of the (4aR,7aS) and (4aS,7aR)
isomers are 3.7 Å and 4.3 Å, respectively, and the distances between centers of
the morpholine moiety and the benzene ring of Phe67 of the (4aR,7aS) and
(4aS,7aR) isomers are 4.5 Å and 5.1 Å, respectively.
13. Hopkins, A. L.; Groom, C. R.; Alex, A. Drug Discovery Today 2004, 9, 430.
14. Leeson, P. D.; Springthorpe, B. Nat. Rev. Drug Disc. 2007, 6, 881.
15. Hopkins, A. L.; Keserü, G. M.; Leeson, P. D.; Rees, D. C.; Reynolds, C. H. Nat. Rev.
Drug Disc. 2014, 13, 105.
16. Ritchie, T. J.; Macdonald, S. J. F. Drug Discovery Today 2009, 14, 1011.
17. Inhibitory activity on tau phosphorylation in vivo; Test compound was
administrated to male CD-1 mice of 5–6 weeks weighing 25–35 g (Charles
River Japan, Inc.) at 1, 3, 10, 30 mg/kg po (0.5% Tween/H2O suspension) and
after 1 h, mice were decapitated and cortex was promptly removed, followed
by being frozen in liquid N2. Cortex was directly homogenized with 2.3% SDS
homogenization buffer (62.5 mM Tris–HCl, 2.3% SDS, 1 mM each of EDTA, EGTA
References and notes
1. Brunden, K. R.; Trojanowski, J. Q.; Lee, V. M.-Y. Nat. Rev. Drug Disc. 2009, 8, 783.
2. Liu, H.-C.; Leu, S.-J.; Chuang, D.-M. J. Exp. Clin. Med. 2012, 4, 135.
3. Martin, L.; Lapypova, X.; Wilson, C. M.; Magnaudeix, A.; Perrin, M.-L.; Yardin,
C.; Terro, F. Ageing Res. Rev. 2013, 12, 289.
4. Cappell, J.; Herrmann, N.; Cornish, S.; Lanctôt, K. L. CNS Drugs 2010, 24, 909.
5. Fukunaga, K.; Uehara, F.; Aritomo, K.; Shoda, A.; Hiki, S.; Okuyama, M.; Usui, Y.;
Watanabe, K.; Yamakoshi, K.; Kohara, T.; Hanano, T.; Tanaka, H.; Tsuchiya, S.;
Sunada, S.; Saito, K.; Eguchi, J.; Yuki, S.; Asano, S.; Tanaka, S.; Mori, A.;
Yamagami, K.; Baba, H.; Horikawa, T.; Fujimura, M. Bioorg. Med. Chem. Lett.
2013, 23, 6933.
and DTT, protease inhibitor cocktail (sigma P2714) containing 0.2 M 4-(2-
l
aminoethyl)benzenesulfonylfluoride (AEBSF), 13 bestatin, 1.4 lM E-64,
lM
0.1 mM leupeptin, 30 nM aprotinin, pH 6.8) and centrifuged at 15000g for
15 min at 4 °C. Protein concentrations were determined using DC protein assay
kit (BIO-RAD). Supernatants were diluted with sample buffer (62.5 mM Tris–
HCl, 25% glycerol, 2% SDS, 0.01% Bromophenol Blue, pH 6.8) to adjust the
6. Link, J. O., private communication.
7. Human GSK-3b cell free enzyme inhibition assay; 7.5
GS1 peptide and 10
32P]ATP were incubated in 50 mM N-(2-
hydroxyethyl)piperazine-N-(2-ethanesulfonic acid) (HEPES)-sodium
lM of prephosphorylated
l
M [c-
protein concentrations around 0.5–2 mg/mg and then boiled for 5 min. 10 lg
hydroxide (pH 7.2), 1 mM dithiothreitol (DTT), 1 mM magnesium chloride,
0.02% Tween-20 buffer for 1 h at room temperature in the presence of human
recombinant GSK-3b. The reaction was stopped with 0.1 volume of 21%
perchloric acid. An aliquot of the reaction mixture was then transferred onto
Whatman P81 cation exchange filters and the filters were washed 3 times with
75 mM phosphoric acid solution, once in water and once in acetone.
Incorporated 32P radioactivity was determined by liquid scintillation
spectrometry. The prephosphorylated GS1 peptide had the following
sequence; NH2-YRRAAVPPSPSLSRHSSPHQS(P)EDEE-COOH. IC50 values are the
mean of at least two experiments.
of samples were applied on 10% SDS–PAGE mini slab gels and transferred onto
PVDF membranes. Membranes were incubated with PBS containing 5% non-fat
milk for 1 h at room temperature and then probed with pS396 anti-body
(BIOSOURCE) overnight at 4 °C. Anti-rabbit IgG HRP-conjugated anti-body
(Promega) was used as secondary anti-body. Membranes were visualized by
ECL kit (Amersham Bioscience) and detected by LAS 1000 (Fuji Photo Film).
18. Analytical data of 6: 1H NMR (400 MHz, CDCl3) d 1.27 (t, J = 7.2 Hz, 3H), 3.10–
3.17 (m, 1H), 3.40–3.46 (m, 1H), 3.55 (s, 3H), 3.60–3.63 (m, 1H), 3.68–3.72 (m,
1H), 3.83–3.91 (m, 3H), 6.90 (s, 1H), 7.97 (dd, J = 1.2 Hz, 4.2 Hz, 1H), 8.52 (d,
J = 4.2 Hz, 1H), 8.54 (d, J = 1.2 Hz, 1H); MS m/z 305 (M+H+).
8. The X-ray crystal structure of GSK-3b (PDB code: 3F88) was utilized in the
docking calculations after restrained minimization by applying the constraint
to converge the non-hydrogen atoms to an RMSD of 0.3 Å using OPLS 2005
force field. The compounds were docked into GSK-3b using Glide 5.8 (XP)
which was available from Schrodinger, LLC, New York, NY. 2012 (http://www.
schrodinger.com/). After above docking study, the atomic charges of each
compound with the docked conformation were calculated. The charges were
obtained by electrostatic potential (ESP) fitting at the single point energy, using
Jaguar 7.9 at the B3LYP/6-31G⁄⁄ level. Jaguar was available from Schrodinger,
LLC, New York, NY. 2011 (http://www.schrodinger.com/).
19. Mansour, T. S.; Evans, C. A. Synth. Commun. 1990, 20, 773.
20. Giannis, A.; Sandhoff, K. Angew. Chem., Int. Ed. Engl. 1989, 28, 218.
21. Olofson, R. A.; Martz, J. M.; Senet, J.-P.; Piteau, M.; Malfroot, T. J. Org. Chem.
1984, 49, 2081.
22. Relative stereochemistry of the isomer was determined by 1H NMR.
23. No stereoisomer was observed by 1H NMR spectroscopy (400 MHz). Absolute
stereochemistry was determined by X-ray crystallography using lactam
derived from (S)-1-phenethylamine (optical isomer of 38).
24. The absolute configuration of each isomer was not determined.
25. Overman, L. E.; Sugai, S. J. Org. Chem. 1985, 50, 4154.
9. The distance between the CH with Phe67 is 4.0 Å for 6 and 3.6 Å for 9, and the
distance between centers of the morpholine moiety and the benzene ring of
Phe67 is 4.7 Å for 6 and 4.6 Å for 9.