6-Heteroaryl-pyrazolo[3,4-b]pyridines: Potent and selective inhibitors of Glycogen Synthase Kinase-3 (GSK-3)

Article abstract of DOI:10.1016/S0960-894X(03)00646-2

A series of 6-heteroaryl-pyrazolo[3,4-b]pyridines has been optimised to afford potent inhibitors of Glycogen Synthase Kinase-3 (GSK-3). These analogues display excellent selectivity over the closely related Cyclin Dependant Kinase-2 (CDK-2).

Full text of DOI:10.1016/S0960-894X(03)00646-2

Bioorganic & Medicinal Chemistry Letters 13 (2003) 3059–3062
6-Heteroaryl-pyrazolo[3,4-b]pyridines: Potent and Selective
Inhibitors of Glycogen Synthase Kinase-3 (GSK-3)
Jason Witherington,* Vincent Bordas, Alessandra Gaiba, Antoinette Naylor,
Anthony D. Rawlings, Brian P. Slingsby, David G. Smith,
Andrew K. Takle and Robert W. Ward
Department of Medicinal Chemistry, Neurology & GI Centre of Excellence for Drug Discovery, GlaxoSmithKline Research Limited,
New Frontiers Science Park, Third Avenue, Harlow, Essex CM19 5AW, UK
Received 31 March 2003; accepted 22 April 2003
Abstract—A series of 6-heteroaryl-pyrazolo[3,4-b]pyridines has been optimised to afford potent inhibitors of Glycogen Synthase
Kinase-3 (GSK-3). These analogues display excellent selectivity over the closely related Cyclin Dependant Kinase-2 (CDK-2).
# 2003 Elsevier Ltd. All rights reserved.
Glycogen synthase kinase-3 (GSK-3) a serine/threonine
kinase was first discovered by virtue of its ability to
phosphorylate and inactivate glycogen synthase, the
regulatory enzyme of mammalian glycogen synthesis. A
number of other substrates have since been identified,
implicating GSK-3 in the regulation of several physio-
logical processes.1 As a result, GSK-3 inhibition has
emerged as an attractive therapeutic target for the
treatment of numerous serious pathologies, including
Alzheimer’s disease, stroke, bipolar disorders, chronic
inflammatory processes, cancer and Type II diabetes.2
In the preceding papers, we have discussed the rational
design that led to the identification of a novel series of
GSK-3 inhibitors based upon a 6-aryl-pyrazolo[3,4-
b]pyridine nucleus (Fig. 1).3
flanking the phenol moiety (cf 5 with 6 and 7) proved
unsuccessful. Previously, in the structurally related 5-
aryl-pyrazolo[3,4-b]pyridazine series, we had demon-
strated that incorporation of a tertiary amine moiety
into the amide side chain led to an excellent improve-
ment in CDK-2 selectivity.3b Incorporation of this
modification into the 6-aryl series did indeed afford
inhibitors with improved selectivity (e.g., 9), although
intrinsic CDK-2 inhibitory potency was still high. From
modelling studies the increase in both GSK-3 and
CDK-2 potency by introduction of the hydroxyl group
can be rationalised fromthe finding that Glu97 and
Asp200, the proposed residues interacting with the
hydroxyl moieties, are conserved in all kinases. Indeed
fromprofiling of the phenols 4 and 5 against a panel
of kinases a worsening in the overall selectivity
profile was observed compared with the phenyl
analogue 3 (Table 2).
Due to the high degree of homology between GSK-3
and Cyclin Dependant Kinase-2 (CDK-2) we decided to
profile a range of a 6-aryl-pyrazolo[3,4-b]pyridines
against this kinase (Table 1). Unfortunately whilst
incorporation of a hydroxyl group at either the para or
meta position of the C6-phenyl ring afforded an excel-
lent improvement in GSK-3 potency (cf 3 with 5 and
10), we also observed a concomitant increase in CDK-2
inhibition. Attempts to introduce CDK-2 selectivity by
variation of the group at C-5 (cf 5 with 4 and 8) or by
Although we had demonstrated that selectivity against
CDK-2 was a generic issue for the phenol analogues,
the overall selectivity profile of the phenyl analogue 3
was excellent. We thus sought to identify an alternative
C-6 substituent that would afford comparable potency
to the phenol motif, but which was devoid of CDK-2
selectivity issues. It had been shown previously that
replacement of the pyrazolo[3,4-b]pyridine nucleus with
an indazole, afforded analogues with similar potency3a
and for reasons of synthetic accessibility, we decided to
explore the C-6 SAR further in this series (Table 3). To
*Corresponding author. E-mail: jason_witherington@gsk.com
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00646-2

3060
J. Witherington et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3059–3062
validate the strategy, the phenol 12 was prepared and,
as predicted, showed increased potency against both
GSK-3 and CDK-2 compared with the corresponding
phenyl analogue 11. Although the indolyl analogue 13
displayed good GSK-3 potency and CDK-2 selectivity,
attempts to identify further phenol isosteres (e.g., 14–
19) or optimise the lipophilic interaction of the phenyl
ring (e.g., 20–25) proved generally unsuccessful. Inter-
estingly, introduction of small heterocyclic substitutents
at the C-6 position consistently afforded analogues (e.g.,
26–30) with greater potency than the corresponding
phenyl analogue 11, although CDK-2 selectivity was
compromised in some cases.
It was envisaged, however, that because the group at C-
6 was not interacting directly with Glu97 or Asp200,
transferral of this promising SAR into the more potent
5-Br pyrazolopyridine series may lead to an increase in
both GSK-3 potency and CDK-2 selectivity. To test this
hypothesis, a range of C-6 heteroaryl pyrazolopyridines
was prepared (Table 4). Gratifyingly, this led to an
improvement in GSK-3 potency, without compromising
CDK-2 selectivity (e.g., 32, 34, and 36). Noteworthy is
the importance of a C-5 Br substituent, which affords
ca. 20-fold increase in GSK-3 potency without affecting
selectivity (cf 33 with 34). Lipophilic groups at C-3 are
also preferred, thus replacing a cyclopropyl with a
cyclopentyl can afford a ca. 5-fold improvement in
potency (cf 31 with 32 and 35 with 36). The rationale for
the observed SAR at C-5 and C-3 has been discussed in
previous communications.3 Importantly, water solubi-
lising groups are tolerated at the C-3 position to afford
potent GSK-3 inhibitors with retention of excellent
CDK-2 selectivity (e.g., 37 and 38).
Cross screening of 6-heteroaryl-pyrazolo[3,4-b]pyridines
against a panel of kinases indicated an excellent overall
selectivity profile (Table 5).
Figure 1. Identification of 6-aryl pyrazolo[3,4-b]pyridine 4.
Table 1. Inhibition of hGSK-3a and hCDK-2 by selected 6-aryl-pyrazolo[3,4-b]pyridine analogues4
No.
R1
R2
R3
GSK-3a, IC50 nM
CDK-2, IC50 nM
3
4
5
6
7
8
9
10
H
4-OH
4-OH
3-Br-4-OH
3-Cl-4-OH
4-OH
Br
Br
H
H
H
Ph
Br
H
cyPr
cyPr
cyPr
cyPr
75Æ10
0.8Æ0.4
8Æ1
>1000
5Æ1
13Æ2
2Æ1
5Æ2
cyPr
7Æ2
2Æ1
cyPr
(CH2)3-4-piperazinyl-N-Et
cyPr
24Æ3
4Æ1
5Æ1
4-OH
3-OH
60Æ1
62Æ3
12Æ3
Table 2. Selectivity profiling of pyrazolopyridines 3 and 4 and 5a
Compd
3
4
5
0
28 31
0
49
15 89 45 19
12 70 29 14 12 65
0
12 10
5
0
20
58 14 59 30 19 17 62
31 80 20 17 83
3
3
0
0
21
1
8
4
79 11
41 59 62 61 80 100 100
91 90
1
0
28
25
87
57 66
56
3
3
5
0
8
41 15 10 58 69 92
aValues are % inhibition @10 mM using 100 mM ATP (see ref 5 for kinases used and assay details).

J. Witherington et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3059–3062
3061
Scheme 1. Preparation of 6-aryl/heteroaryl indazoles. Reagents: (a) boronic acid, KOAc, Pd(PPh3)4; (b) (i) bispinacolatodiboran, KOAc,
ꢀ
.
PdCl2(dppf)2, DMSO; (ii) arylbromide, KOAc, Pd(PPh3)4 100 C; (c) N2H4 H2O, EtOH, reflux; (d) C3H7COCl, pyridine, reflux.
Chemistry6,7
Table 3. Inhibition of hGSK-3a and hCDK-2 by selected indazole
analogues
The 6-heteroaryl-pyrazolo[3,4-b]pyridines were pre-
pared according to the same general procedure pre-
viously described.3c Synthesis of the 6-aryl/heteroaryl
indazoles were prepared according to the procedures
outlined in Scheme 1. Suzuki cross-coupling of bromide
39 with a range of arylboronic acids afforded the corre-
sponding nitrile analogues, 40. Alternatively, the bro-
mide 39 could be converted to the corresponding
boronate 41, utilising bispinacolatodiboran. Subsequent
Suzuki cross-coupling with a range of aryl bromides
afforded the 6-substituted analogues 40. Treatment of
the nitriles with hydrazine hydrate at reflux afforded the
indazole analogues 42 which underwent selective C-3
acylation with cyclopropyl carbonyl chloride in pyridine
at reflux to give the desired analogues 43 in excellent
overall yield.
No.
R
GSK-3a, IC50 nM
CDK-2, IC50 nM
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Ph
4-OH
5-Indolyl
498Æ30
15Æ1
>1000
5Æ1
42Æ5
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
497Æ33
631Æ55
>1000
341Æ21
484Æ26
6-Quinolyl
Ph-4-SO2NH2
Ph-3-SO2NH2
Ph-3-NHSO2Me
Ph-4-NHSO2Me
5-1H-pyrid-2-one
2-F-Ph
>1000
>1000
481Æ25
>1000
>1000
>1000
>1000
>1000
>1000
>1000
828
2,5-diF-Ph
3,4-diF-Ph
2,3-diF-Ph
3-F-Ph
4-F-Ph
>1000
320Æ61
50Æ15
35Æ2
Profiling
a series of 6-arylpyrazolo[3,4-b]pyridines
2-Pyrrolyl
2-Furanyl
3-Furanyl
2-Thienyl
revealed that introduction of a phenol group into the C-
6 phenyl group afforded an excellent improvement in
GSK-3 potency but also a dramatic reduction in CDK-2
selectivity. This finding is most likely due to the high
215Æ29
329Æ73
3-Thienyl
Table 4. Inhibition of hGSK-3a and hCDK-2 by 6-heteroaryl-pyrazolo[3,4-b]pyridines
No.
R1
R2
R3
GSK-3a, IC50 nM
CDK-2, IC50 nM
3
4
H
Br
Br
Br
Br
H
Br
Br
Br
Br
Br
cyPr
cyPr
cyPr
cyPent
cyPr
cyPr
cyPr
cyPent
75Æ10
0.8Æ0.4
39Æ18
7Æ1
>1000
5Æ1
4-HO-Ph
2-Thienyl
2-Thienyl
2-Furyl
31
32
33
34
35
36
37
38
639Æ78
>1000
>1000
>1000
>1000
>1000
>1000
>1000
141Æ18
7Æ2
2-Furyl
2-Thiazoyl
2-Thiazoyl
2-Thienyl
2-Furyl
99Æ55
16Æ4
CH2-4-Piperidine-N-Et
(Æ)-3-Pyrrolidine-N-Bn
18Æ3
14Æ1
Table 5. Selectivity profiling of pyrazolopyridine 34a
5
8
4
2
8
6
2
19
3
1
8
3
0
1
0
18
13
9
4
7
10
23
31
87
aValues are % inhibition @10 mM using 100 mM ATP (see ref 5 for kinases used and assay details).

3062
J. Witherington et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3059–3062
degree of homology between the two kinases and the
suggestion from molecular modelling studies that the
phenol moiety interacts with two conserved residues.
Optimisation of the C-6 position, avoiding functionality
capable of H-bonding to these residues, led to the iden-
tification of a series of 6-hetero-aryl-pyrazolo[3,4-b]pyr-
idines that displayed excellent GSK-3 potency and
CDK-2 selectivity.
burn, A. W.; Gumbiner, B.; Bulacan, F.; Wallace, P.; Henry,
R. R. J. Clin. Invest. 1990, 85, 522. (j) Beck-Nielsen, H.; Vaag,
A.; Damsbo, P.; Handberg, A.; Neilsen, O. H.; Henriksen,
J. E.; Thye-Ronn, P. Diabetes Care 1992, 15, 418. (k) Bogar-
dus, C.; Lillioja, S.; Stone, K.; Mott, D. J. Clin. Invest. 1984,
73, 1185. (l) 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. (m) Pap, M.;
Cooper, G. M. J. Biol. Chem. 1998, 273, 19929. (n) Klein, P. S.;
Melton, D. A. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 8455.
2. Dorronsoro, I.; Castro, A.; Martinez, A. Expert Opin.
Ther. Pat. 2002, 12, 1527.
Acknowledgements
3. (a) Witherington, J.; Bordas, V.; Garland, S. L.; Hickey, D.
M.; Ife, R. J.; Liddle, J.; Saunders, M.; Smith, D. G.; Ward.
R. W. Bioorg. Med. Chem. Lett. In press. (b) Witherington, J.;
Bordas, 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. In press. (c) Witherington, J.; Bordas, V.; Gaiba,
A.; Naylor, A.; Rawlings, A. R.; Slingsby, B. P.; Smith, D. G.;
Takle, A. K.; Ward, R. W. Bioorg. Med. Chem. Lett. See pre-
ceding paper. doi:10.1016/S0960-894X(03)00645-0.
The authors acknowledge the assistance of Glaxo-
SmithKline colleagues from the Department of Bio-
technology and Genetics for the recombinant human
enzymes, the Neurology Centre of Excellence for Drug
Discovery for helpful discussions and Molecular
Screening Technologies for assay of the compounds and
The Division of Signal Transduction Therapy, Depart-
ment of Biochemistry, The University of Dundee for
Selectivity Screening.
4. Results are a mean of at least two determinations run in
duplicate (n=4) and are given as mean. MeanÆSEM is also
quoted for compounds of specific interest.
5. Davies, S. P.; Reddy, H.; Caivano, M.; Cohen, P. Biochem.
J. 2000, 351, 95.
References and Notes
6. Using microtitre plates, GSK-3 was assayed in 50 mM
MOPS buffer, pH 7.0 containing 5% glycerol, 0.01% Tween-
20, 7.5 mM 2-mercaptoethanol, 10 mM magnesium actetate, 8
uM substrate peptide (Biotin-KYRRAAVPPSPSLSRHSSPHQ
(SP)EDEEE, where (SP) is a pre-phosphorylated serine) and
10 uM [g-33P]-ATP. After incubation for 1 h at roomtem-
perature, the reaction was stopped by addition of 50 mM
EDTA solution containing Streptavidin coated SPA beads
(Amersham) to give a final 0.2 mg of beads per assay well in a
384 microtitre plates were counted in a trilux 1450 microbeta
liquid scintillation counter (Wallac).
1. (a) Leloir, L. F.; Olavaria, S. H.; Goldenberg, S. H.; Car-
minatti, H. Arch. Biochem. Biophys. 1959, 81, 508. (b) Man-
darino, L.; Wright, K.; Verity, L.; Nichols, J.; Bell, J.;
Kolterman, O.; Beck-Nielsen, H. J. Clin. Invest. 1987, 80, 655.
(c) Roach, P. FASEB J. 1990, 4, 2961. (d) Skurat, A. V.;
Roach, P. J. Biochem. J. 1996, 313, 45. (e) Zhang, W.;
DePaoli-Roach, A. A.; Roach, P. J. Arch. Biochem. Biophys
1993, 304, 219. (f) Eldar-Finkelmann, H.; Argast, G. M.;
Foord, O.; Fischer, E. H.; Krebs, E. G. Proc. Natl. Acad. Sci.
U.S.A. 1996, 93, 10228. (g) DeFronzo, R. A.; Bonadonna,
R. C.; Ferrannini, E. Diabetes Care 1992, 15, 318. (h) Thor-
7. All novel compounds gave satisfactory analytical data in
full agreement with their proposed structures.