4-Acylamino-6-arylfuro[2,3-d]pyrimidines: Potent and selective glycogen synthase kinase-3 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2004.05.064

Modeling studies of a furo[2,3-d]pyrimidine GSK-3 hit compound 1 superimposed onto the X-ray crystal structure of a legacy pyrazolo[3,4-c] pyridazine GSK-3 inhibitor 2 led to the identification of 4-acylamino-6- arylfuro[2,3-d]pyrimidine template 3. Synth

Full text of DOI:10.1016/j.bmcl.2004.05.064

Bioorganic & Medicinal Chemistry Letters 14 (2004) 3907–3911
4
-Acylamino-6-arylfuro[2,3-d]pyrimidines: potent and
selective glycogen synthase kinase-3 inhibitors
a
Yutaka Maeda, Masato Nakano, Hideyuki Sato, Yasushi Miyazaki,
a,
a
a
*
b
b
Stephanie L. Schweiker, Jeffery L. Smith and Anne T. Truesdale
b
a
Chemistry Department, Tsukuba Research Laboratories, GlaxoSmithKline K.K., Wadai 43, Tsukuba, Ibaraki 300-4247, Japan
b
GlaxoSmithKline Inc., Five Moore Drive, Research Triangle Park, NC 27709, USA
Received 16 April 2004; revised 24 May 2004; accepted 24 May 2004
Available online 17 June 2004
Abstract—Modeling studies of a furo[2,3-d]pyrimidine GSK-3 hit compound 1 superimposed onto the X-ray crystal structure of a
legacy pyrazolo[3,4-c]pyridazine GSK-3 inhibitor 2 led to the identification of 4-acylamino-6-arylfuro[2,3-d]pyrimidine template 3.
Synthesis of analogues based on template 3 has resulted in a number of potent and selective GSK-3b inhibitors. The most potent and
selective compound was the m-pyridyl analogue 24.
Ó 2004 Elsevier Ltd. All rights reserved.
6
1
. Introduction
pyrazolo[3,4-c]pyridazine GSK-3 inhibitor 2 with
GSK-3b suggested that both compounds bind in the
enzyme with similar space requirements. The studies
also suggested the existence of a unique binding inter-
action between 1 and the hinge region in the ATP
Glycogen synthase kinase-3 (GSK-3) is a serine/threo-
nine kinase involved in the regulations of many cell
functions. GSK-3 is implicated in type-2 diabetes be-
cause of its ability to phosphorylate glycogen synthase,
which effectively inhibits the enzyme in catalyzing the
synthesis of glycogen. Recent publications have also
suggested that the b-isomer (GSK-3b) of its two iso-
forms (a and b) plays a pivotal role in the genesis of
1
7
binding site (Figs. 1 and 2).
The outcome of the modeling studies prompted us to
modify 1 with a view to increasing its affinity for GSK-3.
This included replacement of the p-methoxy group by
other groups, the acylation of the 4-amino group, and
removal of the pyridyl moiety, which resulted in the
creation of a new template 3 (Fig. 2).
2
;3
Alzheimer’s disease (AD) and other dementia.
A
selective inhibitor of GSK-3 potentially can be an
effective agent for the treatment of type-2 diabetes and/
or AD, as well as other neurodegenerative disorders. A
number of GSK-3 inhibitors having different structural
classes have been reported, and they showed varying
We first investigated the structure–activity relationship
(SAR) on the 4-N-acyl portion. Thus a set of analogues
of 3 with varying N-acyl groups and with a constant 6-p-
methoxyphenyl group was synthesized. Their kinase
inhibitory activity is summarized in Table 1. The hexa-
noyl analogue 4 exhibited submicromolar inhibitory
activity versus GSK-3b. The closely related methylthio
analogue 5 was equipotent to 4. On the other hand, the
isobutyryl analogue 6 was approximately 4-fold more
potent than 5. This interesting result provided the
impetus for the synthesis of cycloalkyl analogues 7–10.
Unfortunately, 7 and 9 exhibited only slight improve-
4
degrees of efficacy and specificity.
In the course of our exploratory research targeting the
5
VEGFR2 and TIE2 tyrosine kinases, furo[2,3-
d]pyrimidine 1 was identified as a GSK-3 inhibitor
having an IC50 value of 0.081 lM versus GSK-3b.
Modeling studies in which 1 was superimposed onto the
X-ray crystal structure of the complex of the potent
ments in the potency (IC ¼ ca. 0.3 lM) over that of 4
50
Keywords: Glycogen synthase kinase-3 inhibitors.
*
Corresponding author. Tel.: +81-298-64-5541; fax: +81-298-64-5559;
e-mail: masato.nakano@gsk.com
or 5, and the cyclohexyl analogue 10 was much less
potent. Surprisingly, cyclopentyl analogue 8 was 10-fold
0
960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.05.064

3908
Y. Maeda et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3907–3911
8
Table 1. Inhibition of hGSK-3b by 4-N-acyl analogues 4–14
O
R
HN
N
OMe
O
N
No.
R
GSK-3b
CDK2
VEGFR2
IC50, lM
IC50, lM IC50, lM
4
5
6
–(CH
–(CH
2
)
)
4
CH
SCH
3
0.437
0.537
0.132
>19
>20
>21
>48
>47
>48
2
2
3
3 2
–CH(CH )
7
8
0.316
0.032
>20
>19
3.55
1.02
9
0.324
1.82
>19
>20
>20
>48
1
0
>48
>48
Figure 1. Docking study of furo[2,3-d]pyrimidine 1 (yellow) in an
X-ray crystal structure of the complex of pyrazolo[3,4-c]pyridazine 2
O
(light green) and GSK-3b.
11
N
N
>34
1
2
3
1.95
>21
12.3
PRO136
Acylation
1
>35
2.40
>21
>20
21.9
O
Removal
O
O
N
R2
F
H
N
H
N
O
O
HN4
N
5
14
>48
VAL135
NH
6
N
TYR134
ASP133
O
3
O
N
R1
N
1
Figure 2. Proposed binding of a furo[2,3-d]pyrimidine hit 1 with the
hinge region and design of a new series of GSK-3 inhibitor 3.
more potent than 7 or 9 and was by far the most potent
analogue versus GSK-3. According to the modeling
studies of analogue 8 (see Fig. 3), the methine hydrogen
and vertical orientation of the cyclopentyl group ap-
peared to have a unique binding interaction with
PRO136 in the front hydrophobic pocket. To test this
hypothesis, several compounds 11–14, which were void
of the methine hydrogen, were synthesized. These
compounds ranged from morpholino and pyrrolidino
analogues (11 and 12) having in essence a urea-type
functional group to other aryl groups (13 and 14), and
they exhibited weak GSK-3 inhibitory activity, maybe
due to unfavorable affinity of the 4-N-acyl or urea
substituent with PRO136. All the analogues 4–14
showed weak inhibitory activity against CDK-2 and
VEGFR2.
Figure 3. In silico surface model of the GSK-3 ATP-binding pocket
with compound 8.
ment with p-chloro, p-methyl, and p-fluoro groups
(compounds 16–18) gave analogues having reduced
inhibitory activity, suggesting the potentially detrimen-
tal effects of a p-substituent.
In the ensuing SAR studies, we utilized cyclopropyl
analogue 7 as the starting point and explored the
modification of the 6-position of the furo[2,3-d]pyrimi-
dine scaffold. The resultant analogues are listed in Table
Very interestingly, the most potent compound in the
series was the m-pyridyl analogue 19. Potency of 15–19
against VEGFR2 increased somewhat but those were
still in the submicromolar to low micromolar range.
These findings prompted us to prepare analogues 20–24
having identical 6-aryl groups as those in 15–19 but with
2
. As shown in the table, the unsubstituted 15 showed
improved potency against GSK-3b. However, replace-

Y. Maeda et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3907–3911
Table 2. Inhibition of hGSK-3b by 6-aryl analogues 7, 8, and 15–24
3909
8
O
R
HN
N
Aryl
O
N
No.
R
Aryl
GSK-3b IC50, lM
CDK2 IC50, lM
VEGFR2 IC50, lM
CH3
O
7
0.316
>20
3.55
1
1
1
1
1
8
2
2
2
2
2
5
6
7
8
9
0.174
1.00
>21
>20
>21
>19
0.447
>19
>19
>20
>20
>19
0.457
0.832
2.04
0.631
2.04
3.24
1.02
>23
Cl
CH3
F
0.617
0.575
0.005
0.032
0.023
0.550
0.240
0.219
0.005
N
CH3
O
0
1
2
3
4
Cl
CH3
F
>20
>20
>19
>24
N
a cyclopentyl group in the 4-position. As a result,
analogues 20–24 showed comparable potency versus
GSK-3b to that of compounds 15–19. However and
importantly, these compounds showed a lack of inhibi-
tory activity against CDK-2 and VEGFR2.
lation in L6 cells (Table 4). The 6-phenyl analogue 20
did not show potency in such an assay, but both m-
pyridyl analogues 19 and 24 exhibited excellent potency
11
of glycogen accumulation.
Having identified a series of potent GSK-3 inhibitors,
we next profiled the most potent analogue 24 against a
12
panel of more than 20 kinases, including GSK-3b (Table
2. Chemistry
9
3
). The data showed that analogue 24 indeed had
excellent selectivity over all the kinases in the panel,
including CDK-2. Overlaying of docking structures of
The furo[2,3-d]pyrimidine analogues were prepared
according to procedures outlined in Scheme 1. The
chemistry started with alkylation of commercially
available 2-bromoacetophenones 25i–v with malononit-
rile to give benzoylmethylmalononitriles 26i–v, which
were treated with hydrochloric acid in acetic acid at
room temperature to afford 2-amino-3-cyanofurans 27i–
v. Cyclization of furan 27i–v with formamide resulted in
4-amino-6-phenyl-furo[2,3-d]pyrimidine 28i–v, which
10
2
4 in GSK-3b and CDK-2/cyclin A also suggested that
the selectivity of 24 was likely due to a steric clash be-
tween a cyclopentyl group and its interacting site in
CDK-2. Based on the modeling studies, however, m-
pyridyl effect on the high potency was not clear. Finally,
representative three analogues 19, 20, and 24 were
examined in an assay that measured glycogen accumu-

3910
Y. Maeda et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3907–3911
a
Table 3. Selectivity of m-pyridyl analogue 24 for GSK-3b
completed by addition of sodium ethoxide in the same
solvent to give a precursor, 4-amino-6-(m-pyridyl)-
furo[2,3-d]pyrimidine 28vi. Cyclopentylcarbonylation of
this precursor in a usual manner led to analogue 24.
Compd no.
AMPK
Chk1
24
0
0
CKII
JNK
0
0
LCK
MAPK
RSK-2
2
5
3. Conclusion
NT
0
MAPKAP-K2
MEK1
MSK1
Modeling studies of a GSK-3b hit compound 1 by
superimposing onto the X-ray crystal structure of the
complex of 2 with GSK-3b led to the development of a
template represented by structure 3. Synthesis based on
the template has led to the identification of a series of
potent GSK-3b specific analogues. One such compound,
the m-pyridyl analogue 24, was found to exhibit potent
GSK-3b inhibitory activity, possess excellent selectivity
over other kinases including CDK-2 and VEGFR2, and
exhibit excellent potency in cellular assays.
0
10
0
p70S6K
PDK1
0
PHOS.K
PKA
9
0
PKBa
PKCa
PRAK
1
NT
7
ROKa
0
SAPK2a
SAPK2b
SAPK3
SAPK4
SGK
0
0
0
17
8
Acknowledgements
CDK2/Cyclin A
GSK-3b
66
91
The authors acknowledge with gratitude the invaluable
assistance of colleagues from GlaxoSmithKline RTP,
USA and Harlow, UK. Additionally, we are grateful to
the Division of Signal Transduction Therapy, Depart-
ment of Biochemistry, The University of Dundee for
selective screening.
a
Values are % I @ 10 lM using 100 lM ATP (see Ref. 9 for kinases
used and assay details).
Table 4. Potency of 19, 20, and 24 inglycogen accumulation in L6 cells
No.
Potency of glycogen accumulation (EC50, lM)
19
20
24
0.74
>10
0.39
References and notes
1
. (a) Cohen, P.; Frame, S. Nat. Rev. Mol. Cell Biol. 2001, 2,
69; (b) Coghlan, M. P.; Culbert, A. A.; Cross, D. A. E.;
7
were reacted with acylchlorides to afford the desired
products. With regard to the synthesis of a m-pyridyl
analogues 19 and 24, the precursor 28vi was obtained by
a different method. Obtained 2-amino-3-cyano-furan
Corcoran, S. L.; Yates, J. W.; Pearce, N. J.; Rausch, O. L.;
Murphy, G. J.; Carter, P. S.; Cox, L. R.; Mills, D.; Brown,
M. J.; Haigh, D.; Ward, R. W.; Smith, D. G.; Murray, K.
J.; Reith, A. D.; Holder, J. C. Chem. Biol. (London) 2000,
2
7vi was converted into an ethoxyimino derivative, fol-
7, 793; (c) Cohen, P.; Yellowlees, D.; Aitken, A.; Donella-
lowed by treatment with ammonia in a solution of eth-
anol and tetrahydrofuran. Subsequently, cyclization was
Deana, A.; Hemmings, B. A.; Parker, P. J. Eur. J.
Biochem. 1982, 124, 21.
b ( for i-v )
X
R
1
NC
H N
X
R
1
or
a
c ( for vi )
NC
Br
O
X
2
O
CN
O
R1
2
5 i-vi
2
6 i-vi
27 i-vi
R2
d ( for i-v )
or
e ( for vi )
O
NH2
N
HN
X
f
X
N
N
R1
R1
27 i-vi
O
8 i-vi
O
N
2
4
- 24
1
1
1
1
1
1
(
i) X=CH, R =OCH
3
; (ii) X=CH, R =H; (iii) X=CH, R =Cl; (iv) X=CH, R =CH
3
; (v) X=CH, R =F; (vi) X=N, R =H
Scheme 1. Reagents and conditions: (a) malononitrile, NaOEt; (b) HCl–AcOH; (c) piperidine, reflux EtOH; (d) formamide, 200 °C; (e) 1.
EtO) CHOAc, 2. NH , EtOH–THF, 3. NaOEt, EtOH–THF; (f) acylchloride, NaH, DMF.
(
2
3

Y. Maeda et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3907–3911
3911
2
. GSK-3b is known to phosphorylate the microtubule
associated protein tau in mammalian cells. Lovestone,
S.; Reynolds, C. H.; Latimer, D.; Davis, D. R.; Anderton,
B. H.; Gallo, J.-M.; Hanger, D.; Mulot, S.; Marquardt, B.
Curr. Biol. 1994, 4, 1077.
American Chemical Society: Washington, DC, 2003,
MEDI22; (h) Olesen, P. H.; Sorensen, A. R.; Urso, B.;
Kurtzhals, P.; Bowler, U. E.; Hansen, B. F. [Triazole
derivative]. J. Med. Chem. 2003, 46, 3333; (i) Gallet, T.;
Lardenois, P.; Lochead, A. W.; Nedelec, A.; Marguerie,
S.; Saady, M.; Yaiche, P. [6,7,8,9-tetrahydropyrimido(1,2-
3
. GSK-3b binds to presenilin 1 and its binding complex
mediates b-amyloid-induced neurotoxicity. Planel, E.;
Sun, X.; Takashima, A. Drug Develop. Res. 2002, 56, 491.
. (a) Leclerc, S.; Garnier, M.; Hoessel, R.; Marko, D.; Bibb,
J. A.; Snyder, G. L.; Greengard, P.; Biernat, J.; Wu, Y. Z.;
Mandelkow, E. M.; Eisenbrand, G.; Meijer, L. [Indiru-
bin]. J. Biol. Chem. 2001, 276, 251; (b) Leost, M.; Schultz,
C.; Link, A.; Wu, Y. Z.; Biernat, J.; Mandelkow, E. M.;
Bibb, J. A.; Snyder, G. L.; Greengard, P.; Zaharevitz, D.
W.; Gussio, R.; Senderowicz, A. M.; Sausville, E. A.;
Kunick, C.; Meijer, L. [Paullone]. Eur. J. Biochem. 2000,
a)pyrimidin-4-one].
European
Patent
Application
EP1295885, 2003; (j) Kuo, G.-H.; Prouty, C.; DeAngelis,
A.; Shen, L.; O’Neil, D. J.; Shah, C.; Connolly, P. J.;
Murray, W. V.; Conway, B. R.; Cheung, P.; Westover, L.;
Xu, J. Z.; Look, R. A.; Demarest, K. T.; Emanuel, S.;
Middleton, S. A.; Jolliffe, L.; Beavers, M. P.; Chen, X.
4
[Macrocyclic
imides]. J. Med. Chem. 2003, 46, 4021.
polyoxy-genated
bis-7-azaindolylmale-
5. Adams, J. J.; Bryan, D. D.; Feng, Y.; Matsunaga, S.;
Miyazaki, Y.; Nakano, M.; Rocher, J.-P.; Sato, H.;
Semones, M.; Silvia, D. J.; Tang, J. PCT International
Application WO 03/022852, 2003.
6. Witherington, J.; Boardas, V.; Haigh, D.; Hickey, D. M.;
Ife, R. J.; Rawlings, A. D.; Slingsby, B. P.; Smith, D. G.;
Ward, R. W. Bioorg. Med. Chem. Lett. 2003, 13, 1581.
7. Docking model of 1 was made by using Quanta/
CHARMm (Accelrys Inc., San Diego, CA, USA).
8. hGSK-3b enzyme assay was carried out by scintillation
proximity assay (SPA).
2
67, 5983; (c) Smith, D. G.; Buffet, M.; Fenwick, A. E.;
Haigh, D.; Ife, R. J.; Saunders, M.; Slingsby, B. P.; Stacey,
R.; Ward, R. W. [Maleimide, SB-415286]. Bioorg. Med.
Chem. Lett. 2001, 11, 635; (d) Meijer, L.; Thunnissen, A.
M. W. H.; White, A. W.; Garnier, M.; Nikolic, M.; Tsai,
L. H.; Walter, J.; Cleverley, K. E.; Salinas, P. C.; Wu, Y.
Z.; Biernat, J.; Mandelkow, E. M.; Kim, S. H.; Pettit, G.
R. [Hymenialdisine]. Chem. Biol. 2000, 7, 51; (e) Naerum,
L.; Norskov-lauritsen, L.; Olesen, P. H. [5-(1-pyridyl)-
[
(
1,3,4]-oxadiaol]. Bioorg. Med. Chem. Lett. 2002, 12, 1525;
f) Martinez, A.; Aloso, M.; Castro, A.; Perez, C.;
Moreno, F. J. [Thiadiazolidinone]. J. Med. Chem. 2002,
5, 1292; (g) Wagman, A. S.; Harrison, S. D.; Johnson, K.;
9. Davis, S. P.; Reddy, H.; Caivano, M.; Cohen, P. Biochem.
J. 2000, 351, 95.
10. In-house X-ray crystal structure data.
11. The method of the glycogen accumulation assay in L6 cells
was described by: Peat, A. J.; Garrido, D.; Boucheron, J.
A.; Schweiker, S. L.; Dickerson, S. H.; Wilson, J. R.;
Wang, T. Y.; Thomson, S. A. Bioorg. Med. Chem. Lett.
2004, 14, 2127.
4
Ring, D. B.; Bussiere, D. E.; Nuss, J. M.; Pfister, K. B.;
Goff, D.; Boyce, R. S.; Ramurthy, S.; Brown, S. P.;
Renhowe, P. A.; Subramanian, S.; Zhou, X. A.
[CHIR73911]. In 225th National Meeting of the American
Chemical Society, New Orleans, LA, March 23–27, 2003;
12. Several experimental conditions are available in Ref. 5.