Imidazo[1,2-b]pyridazines: A potent and selective class of cyclin-dependent kinase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2004.02.008

Modification of imidazo[1,2-a]pyridine CDK inhibitors lead to identification of less lipophilic imidazo[1,2-b]pyridazine series of CDK inhibitors. Although several equivalent compounds from these two series have similar structure and show similar CDK acti

Full text of DOI:10.1016/j.bmcl.2004.02.008

Bioorganic & Medicinal Chemistry Letters 14 (2004) 2249–2252
Imidazo[1,2-b]pyridazines: a potent and selective class of
cyclin-dependent kinase inhibitors
Kate F. Byth,a Nicola Cooper,a Janet D. Culshaw,a David W. Heaton,a Sandra E. Oakes,a
Claire A. Minshull,b Richard A. Norman,b Richard A. Pauptit,b Julie A. Tucker,b
b
b
b
b
^
Jason Breed, Andrew Pannifer, Sian Rowsell, Judith J. Stanway,
Anna L. Valentineb and Andrew P. Thomasa,*
aCancer Research, AstraZeneca, Alderley Park, Macclesfield, Cheshire SK10 4TG, UK
bProtein Structure Laboratory, AstraZeneca, Alderley Park, Macclesfield, Cheshire SK10 4TG, UK
Received 11 November 2003; revised 27 January 2004; accepted 3 February 2004
Abstract—Modification of imidazo[1,2-a]pyridine CDK inhibitors lead to identification of less lipophilic imidazo[1,2-b]pyridazine
series of CDK inhibitors. Although several equivalent compounds from these two series have similar structure and show similar
CDK activity, the SAR of the two series differs significantly. Protein inhibitor structure determination has confirmed differences in
binding mode and given some understanding of these differences in SAR. Potent and selective imidazo[1,2-b]pyridazine inhibitors of
CDK2 have been identified, which show >1 lM plasma levels following a 2 mg/kg oral dose to mice.
Ó 2004 Elsevier Ltd. All rights reserved.
In the preceding paper1 we describe the in vitro SAR
and optimisation of imidazo[1,2-a]pyridines (1) as
cyclin-dependent kinases (CDK) inhibitors. Such
inhibitors are expected to be useful as anti-tumour
agents due to the key role that the CDK enzymes play in
controlling the cell cycle.2 Furthermore, it has been
shown that inhibition of CDK2 activity can lead to the
selective killing of tumour cells with deregulated E2F-1
activity.3 A key question in the optimisation of CDK
inhibitors is whether such agents will be most useful
following an acute single dose or following regular daily
dosing as a chronic therapy. This would be an important
determinant in deciding whether CDK inhibitors should
be optimised as agents for intravenous or for oral
delivery. Our assessment was that regular dosing would
be optimally required and that CDK inhibitors with oral
activity should be sought. The physico-chemical prop-
erties of the imidazo[1,2-a]pyridine CDK inhibitors
reported in the preceding paper1 were not ideally suited
for optimisation as orally active agents, and we sought
to modify these compounds to reduce the lipophilicity of
the core structure.
H
N
N
N
H
N
N
N
N
O
O
N
O
S
N
O
S
N
NH
R
NH
N
R
1
2
We reasoned that introduction of additional hetero-
atoms into the pyridine ring of the imidazo[1,2-a]pyr-
idine system should be undertaken and this work would
lead to the identification of imidazo[1,2-b]pyridazine
CDK inhibitors (2).
Initial examples of the new imidazo[1,2-b]pyridazines
showed similar potency to that shown by corresponding
imidazo[1,2-a]pyridines1 (Table 1).
The preparation of compounds of structure 2 and 3
proceeded via the key methyl ketone intermediate C
(Scheme 1). 3-Amino-6-chloropyridazine was condensed
with DMFÆDMA to give the amidine intermediate A,
which was reacted with chloroacetone through initial
quaternisation on the pyridazine 2-N followed by
deprotonation and attackon the amidine leading to
cyclisation with loss of dimethamine to give imidazo[1,2-
b]pyridazine B.5 The chloro group was removed by
Keywords: CDK2 inhibitor.
* Corresponding author. Tel.: +44-(0)1625-515170; fax: +44-(0)1625-
513910; e-mail: andrew.p.thomas@astrazeneca.com
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.02.008

2250
K. F. Byth et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2249–2252
Table 1. Structures and enzyme activity for imidazo[1,2-b]pyridazines
O
N
Cl
N
Cl
N
R2
N
(b)
(a)
N
N
H
N
N
H
N
N
N
N
NH2
N
E
N
R1
N
D
N
N
(c)
N
N
S
O
R2
N
O
N
O
2
3
Br
O
N
O
N
(d)
R2
N
CDK2 IC50, lMa MCF-7 prolif.
IC50, lMb
R2
N
Compd R1
R2
N
N
G
2a
2b
2c
2d
2e
2f
3
H
Me
H
H
H
H
H
<0.003
<0.003
F
Scheme 2. Synthesis of imidazo[1,2-b]pyridazine intermediate G.
Reagents and conditions: (a) chloroacetaldehyde, n-BuOH, 120 °C,
reflux, 18 h, 78%; (b) R1 ¼ Me: NaOMe, MeOH, 20 °C, 18 h, 91%;
R1 ¼ CH2CF3: CF3CH2OH, NaH, DMF, 20 °C, 18 h, 71%; (c) NBS,
CHCl3, 20 °C, 2 h, 76–93%; (d) (i) i-PrMgBr, THF, )40 °C, 2 h; (ii) add
MeCON(OMe)Me, )40 to 20 °C, 4 h, 43–62%.
(CH2)2OMe
(CH2)2NMe2
(CH2)3NMe2
(CH2)2OMe
<0.003
0.008
<0.003
0.6
0.39
0.43
Me
>5
1.0
a Average of at least two measurements; enzyme protocol Ref. 4.
b IC50 for inhibition of BrdU incorporation to MCF-7 cells following
3 day exposure to test compound; average of at least two measure-
ments.
N
O
O
(a)
X
N
X
N
N
N
R1
R1
N
N
O
Cl
N
Cl
N
C
G
X =H
X = OR2
(a)
(c)
N
N
N
H
+
R1
O
(b)
N
N
R1
NH
2
A
N
H
N
(b)
N
X
N
O
O
N
N
Cl
N
R1
N
N
R1
R1
(c)
N
N
N
H
N
N
J
C
B
N
X
N
Scheme 1. Synthesis of imidazo[1,2-b]pyridazine intermediate C.
Reagents and conditions: (a) toluene, reflux, 2.5 h, 80–95%; (b)
R1 ¼ H: chloroacetone, KI, DMF, 80 °C, 5 h, 58%; R1 ¼ Me: chloro-
acetone, KBr, EtOH, reflux, 5 h, 79%; (c) H2 1 atm, Pd/C, Et3N, 20 °C,
2 h, 90–92%.
O
N
S
R1
R3
O
N
N
R4
4
Scheme 3. Synthesis of imidazo[1,2-b]pyridazines 3. Reagents and
conditions: (a) DMFÆDMA, 100 °C, 20–44 h, 86–88%; (b) phenyl-
guanidine hydrocarbonate, DMA, 100–160 °C, 5–7 h, 40–80%; (c) (i)
ClSO3H, SOCl2, 0 °C to reflux, 1 h; (ii) R3(R4)NH, MeOH, 0–20 °C,
24 h, 58–88%.
hydrogenolysis to give key methyl ketone C. Introduc-
tion of a methyl substituent in the 2-position of C was
achieved by an analogous procedure but starting with
DMAÆDMA (Scheme 1).
Following the SAR shown by the imidazo[1,2-a]pyridine
series,1 we wanted to take advantage of the reactivity of
the chloropyridazine to introduce substituents to the
5-position of the imidazo[1,2-b]pyridazine ring system.
This was achieved (Scheme 2) by first forming the
5-chloroimidazopyridazine D.
The biological results (Table 1) appeared to support our
initial expectation that the binding mode and SAR of
the imidazo[1,2-b]pyridazine series would mirror that of
the earlier imidazo[1,2-a]pyridine series. However, it was
obvious that the imidazo[1,2-b]pyridazine showed a
significantly greater dependence on the presence of the
4-sulfonamide substituent (SO2NH) on the aniline
(compare compds 2f and 2c, and compds 3 and 2a) than
shown by the imidazo[1,2-a]pyridines.1
The eventual 5-substituent was introduced by displace-
ment of the chloro group of D with alkoxide nucleo-
philes. Introduction of the acetyl substituent was then
carried out by bromination, followed by transmetall-
ation with isopropyl magnesium bromide6 and reaction
with the appropriate acetamide to give the methyl
ketone G. The final conversion of the methyl ketones to
final compounds 4 was performed (Scheme 3) via an
analogous route to that developed for imidazo[1,2-
a]pyridine CDK inhibitors.1
Also based on the SAR shown by the imidazo[1,2-
a]pyridine series,1 it was our expectation that introduc-
tion of 5-substituents to the imidazo[1,2-b]pyridazine
ring system would be at least tolerated, and potentially
beneficial for CDK2 activity. However, the results for
compounds such as 4a (Table 2) show a major loss in

K. F. Byth et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2249–2252
2251
Table 2. Structures and enzyme activity for substituted imidazo[1,2-b]-
pyridazines
N
H
N
N
X
N
H
N
N
S
R1
O
N
O
O
4
Compd
R1
X
CDK2
CDK1
MCF-7
IC50, lMa IC50, lMa prolif. IC50, lMb
2c
4a
4b
H
H
<0.003
0.6
0.003
0.04
0.3
0.39
0.23
H
Me
MeO
H
a Average of at least two measurements; enzyme protocol Ref. 4.
b IC50 for inhibition of BrdU incorporation to MCF-7 cells following
3 day exposure to test compound; average of at least two measure-
ments.
Figure 1. Crystal structures of CDK2 complexed with 2e:11 Final
2Fo À Fc electron density for the inhibitor 2e (1:3r level). The figures
were prepared using Bobscript and Raster3D.18;19
potency for CDK2 inhibition. Also contrary to the SAR
shown by analogous imidazo[1,2-a]pyridine inhibitors,
was the equivalent if not improved CDK2 activity
shown when a methyl group was introduced to the
2-position of the imidazo[1,2-b]pyridazine ring system
(compd 4b) compared to the 2-H derivative (compd 2c)
(Table 2).
From the data generated (Tables 1 and 2) it is clear that
significant differences in SAR exist between the imid-
azo[1,2-b]pyridazine series reported here and the imid-
azo[1,2-a]pyridine series reported in the preceding
paper.1 The inverted tolerability shown by the two series
to substitution at the 2- and 5-positions of the imidazo-
pyrid(az)ine rings systems indicated some likely change
in conformation and binding mode of the 6–5 ring sys-
tem in the two series. To help explain and explore these
differences we undertookstructural studies, based on
7–9
previous workwith CDK2 X-ray crystallography,
to
investigate binding mode of the complex with the rep-
resentative imidazo[1,2-b]pyridazine 2e. This was com-
pared to the structure generated for the parent
imidazo[1,2-a]pyridine 5.10
N
H
N
N
H
N
Figure 2. Crystal structures of CDK2 complexed with 5:10 Final
2Fo À Fc electron density for the inhibitor 5 (1:3r level). The figures
were prepared using Bobscript and Raster3D.18;19
N
N
N
N
H
N
NH2
N
S
N
S
N
O
O
N
O
O
2e
5
the imidazo[1,2-a]pyridine ring of 5. The lackof toler-
ance of 5-substituents on the imidazo[1,2-b]pyridazine
shown by compd 4a is clearly understood as this sub-
stituent is directed towards Phe80 at the closed end of
the binding pocket, while the 2-substituent tolerated in
the imidazo[1,2-b]pyridazine (compd 4b) is directed
towards the open end of the binding pocket.
The crystal structures of CDK2 complexed with 2e (Fig.
1) and with 5 (Fig. 2) show that the key hydrogen-
bonding interactions between the pyrimidine N1 and
Leu83 backbone NH and between the anilino NH and
Leu83 backbone carbonyl O are maintained in both
complexes. However, as suggested by the SAR discussed
above, the orientation of the imidazo[1,2-b]pyridazine
ring of 2e is indeed in an inverted orientation relative to
The binding mode of imidazo[1,2-b]pyridazine 2e shows
loss of the H-bonding interaction between the N-1 of the
imidazole ring and the Lys33 amino group. SAR from

2252
K. F. Byth et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2249–2252
Table 3. PK summary of 2c and 4c following 2 mg/kg po to micea
References and notes
Compd
AUC 0–6 h, Cmax
,
T
max, h
Approx. T1=2, h
1. Byth, K. F.; Culshaw, J. D.; Green, S.; Oakes, S.; Thomas,
A. P. Bioorg. Med. Chem. Lett. 2004, 14, preceding paper.
doi:10.1016/j.bmcl.2004.02.015.
2. Sherr, C. J. Science 1996, 274, 1672–1677.
3. Chen, Y.-N. P.; Sharma, S. K.; Ramsey, T. M.; Jiang, L.;
Martin, M. S.; Baker, K.; Adams, P. D.; Bair, K. W.;
Kaelin, W. G., Jr. Proc. Natl. Acad. Sci. U.S.A. 1999, 96,
4325–4329.
4. Thomas, A. P. PCT Int. Application WO 2002-066481.
5. Podergajs, S.; Stanovnik, B.; Tisler, M. Synthesis 1984,
263–265.
6. Abarbri, M.; Thibonnet, J.; Berillon, L.; Dehmel, F.;
Mario Rottlander, M.; Knochel, P. J. Org. Chem. 2000,
65, 4618–4634.
7. Schulze-Gahmen, U.; DeBondt, H. L.; Kim, S.-H. J. Med.
Chem. 1996, 39, 4540–4546.
8. Davies, T. G.; Tunnah, P.; Meijer, L.; Marko, D.;
Eisenbrand, G.; Endicott, J. A.; Noble, M. E. M.
Structure 2001, 9, 389–397.
9. Anderson, M.; Beattie, J. F.; Breault, G. A.; Breed, J.;
Byth, K. F.; Culshaw, J. D.; Ellston, R. P. A.; Green, S.;
Minshull, C. A.; Norman, R. A.; Pauptit, R. A.; Stanway,
J.; Thomas, A. P.; Jewsbury, P. J. Bioorg. Med. Chem.
Lett. 2003, 13, 3021–3026.
lM h
lM
2c
4b
9.4
20
3.2
9.9
0.5
0.5
3.0
3.3
the imidazo[1,2-a]pyridine series1 indicated this H-bond
interaction to be important for CDK2 activity, however
the edge-to-face interaction of the pyridazine ring of 2e
with the phenyl group of Phe80 that is gained must
provide compensatory binding free energy. The change
in binding conformation of the imidazo[1,2-b]pyridazine
2e compared to the imidazo[1,2-a]pyridine 5 can be
understood in terms of the electrostatic repulsion that
would exist between the N4 of the imidazo[1,2-b]pyrid-
azine and the N3 of the pyrimidine ring of 2e if it were to
adopt the binding mode of imidazo[1,2-a]pyridine such
as 5. While the absence of a peri-hydrogen at the 5-
position of the 5,6-ring system in imidazo[1,2-b]pyrid-
azines, such as 2e, also makes the adopted conformation
(Fig. 1) more favourable.
For both the imidazo[1,2-b]pyridazine 2e and imida-
zo[1,2-a]pyridine 5, the sulfonamide group forms
hydrogen bonds with the Asp86 backbone NH and
its carboxylic side chain and also with the side chain
NH of Lys89 (Figs. 1 and 2). Though some subtle dif-
ferences in binding and orientation are apparent, the
cause of the difference in dependence on sulfonamide
group for binding potency shown by the two series is not
clear.
10. For details of methods and crystallographic statistics for
structure of CDK2 complex with compd 5, see Ref. 9 and
references cited therein.
11. Protein and crystals were obtained according to estab-
lished procedures.12;13 Diffraction data were collected on
beamline PX9.6 at SRS, Daresbury, at 100 K. Data
processing, data reduction and structure solution by
molecular replacement were carried out using programs
from the CCP4 suite.14 Compound 2e was modeled into
electron density using QUANTA.15 The protein complex
model was refined using CNX,16 and the final structure17
have been deposited in the Protein Data Bankwith
deposition code 1urw together with structure factors and
detailed experimental conditions.
12. Lawrie, A. M.; Noble, M. E.; Tunnah, P.; Brown, N. R.;
Johnson, L. N.; Endicott, J. A. Nat. Struct. Biol. 1997, 4,
796–801.
13. Legraverend, M.; Tunnah, P.; Noble, M.; Ducrot, P.;
Ludwig, O.; Grierson, D. S.; Leost, M.; Meijer, L.;
Endicott, J. J. Med. Chem. 2000, 43, 1282–1292.
14. CCP4 Acta Crystallogr. 1994, D50, 760–763.
15. Quanta2000, Accelrys.
The properties and activities of compounds 2c and 4b
have been characterised in more detail. Both compounds
show selectivity for CDK2 over CDK1, though this
appears to be greater for 4b (Table 2). Examining
activity against a broader range of kinases, 2c gave an
IC50 of >10 lM against the following kinases: Csk,
EGFR, FAK, FGFR-1, IGF-1R, JAK3, Src, vAbl,
Zap70, p38a, JNK1 and PKA. Significantly, compounds
from the imidazo[1,2-b]pyridazine series show improved
blood levels following oral dosing over those shown by
the imidazo[1,2-a]pyridine series. Both compounds 2c
and 4b show significant blood levels in mice following
oral dosing (Table 3).
16. CNX version 2000.1, Accelrys.
17. Crystallographic statistics for compd 2e are: space group
ꢀ
ꢀ
P212121, unit cell 54.2, 72.9, 73.3 A, resolution 1.6 A,
33,057 reflections from 72,225 observations give 85%
completeness with Rmerge ¼ 5:4% and mean I=rðIÞ of 13.2.
The final model containing 2153 protein, 212 water and 32
inhibitor atoms has an R-factor of 22.0% (Rfree using 5% of
the data 25.5%). Mean temperature factors for the protein
In conclusion, imidazo[1,2-b]pyridazines CDK2 inhibi-
tors have been discovered and optimised. Compounds
2c and 4b have been characterised as potent and selec-
tive inhibitors of CDK enzymes, which show significant
plasma levels following oral dosing. These compounds
provide useful leads for the discovery and development
of orally active CDK inhibitors.
2
ꢀ
and the ligand are 23.7 and 20.4 A , respectively.
18. Esnouf, R. J. Mol. Graphics 1997, 156, 132–134.
19. Merritt, E. A.; Murphy, M. E. P. Acta Crystallogr. 1994,
D50, 869–873.