Discovery of imidazole vinyl pyrimidines as a novel class of kinase inhibitors which inhibit Tie-2 and are orally bioavailable

Article abstract of DOI:10.1016/j.bmcl.2008.06.106

Tie-2 is a receptor tyrosine kinase which is involved in angiogenesis and thereby growth of human tumours. The discovery and SAR of a novel class of imidazole-vinyl-pyrimidine kinase inhibitors, which inhibit Tie-2 in vitro is reported. Their synthesis wa

Full text of DOI:10.1016/j.bmcl.2008.06.106

Bioorganic & Medicinal Chemistry Letters 18 (2008) 4723–4726
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of imidazole vinyl pyrimidines as a novel class of kinase
inhibitors which inhibit Tie-2 and are orally bioavailable
David Buttar, Mike Edge, Steve C. Emery, Martina Fitzek, Cheryl Forder, Alison Griffen, Barry Hayter,
*
Chris F. Hayward, Philip J. Hopcroft, Richard W. A. Luke , Ken Page, John Stawpert, Andy Wright
Cancer and Infection Research Area, Mereside, Alderley Park, AstraZeneca, Macclesfield SK10 4TG, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Tie-2 is a receptor tyrosine kinase which is involved in angiogenesis and thereby growth of human
tumours. The discovery and SAR of a novel class of imidazole-vinyl-pyrimidine kinase inhibitors, which
inhibit Tie-2 in vitro is reported. Their synthesis was carried out by condensation of imidazole aldehydes
with methyl pyrimidines. These compounds are lead-like, with low molecular weight, good physical
properties and oral bioavailability.
Received 19 May 2008
Revised 25 June 2008
Accepted 25 June 2008
Available online 23 July 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Angiogenesis
Kinase
Tie-2
Inhibitor
Pyrimidine
Imidazole
Alkene
There is an ongoing need for novel approaches to treating hu-
man tumours and one of the most promising new approaches is
blocking angiogenesis.1 Tumours require blood supply to grow to
a significant size and it is believed that blocking angiogenesis will
prevent the development of this vascular network. The endothelial
cell growth factor VEGF is a potent stimulus for angiogenesis and
recent clinical trials have shown efficacy of the antibody bev-
acizumab,2 which binds to VEGF. There is considerable work
underway directed towards other angiogenesis inhibitors, and
VEGFr inhibitors are the most advanced kinase inhibitors in this
field. Of the other kinases thought to be involved in angiogenesis
in human tumours, Tie-2, an endothelium specific receptor tyro-
sine kinase, promotes tumour angiogenesis through interaction
with the angiopoietins3 and plays an important role in stabilizing
the immature endothelial cell network, attracting pericytes and
maintaining vessel integrity.4 Although the details of Tie-2 recep-
tor biology are still emerging,5 it has attracted considerable inter-
est, although to date only a small number of inhibitors have been
reported outside of patent literature.6 We wish to report here our
work leading to the identification of imidazole vinyl pyrimidines
as a novel class of kinase inhibitors which inhibit Tie-2 in vitro.
A high-throughput screen (HTS) of the AstraZeneca compound
collection identified a singleton hit, 1, an ATP-competitive Tie-2 ki-
target, therefore a phospho-Tie-2 cell-based ELISA assay8 was uti-
lized to test the compound’s cellular potency. In this assay, com-
pound 1 was active with an IC50 of 0.33 lM. This pattern of
greater potency in the cell assay was common to most of the hits
from the HTS, covering more than 20 distinct chemical series,
and we suspect that it reflects a lack of sensitivity in the enzyme
assay.9 Compound 1 showed reasonable selectivity in a large panel
of kinase assays, only inhibiting three other enzymes with an IC50
of less than 10 lM (p38, 1.6 lM; Flt-4, 5.5 lM; KDR, 7.0 lM). In
contrast to Tie-2 the compound was inactive in VEGFr and p38 cel-
lular assays suggestive of a greater drop in potency on going from
enzyme to cell assays for these kinases. Compound 1 showed good
oral exposure in a mouse cassette dosing experiment (Cmax
=
0.29 M following a 1 mg/kg dose) and has a low molecular weight
l
(319) and ClogP (2.8). The novelty of the core structure as well as
its good selectivity profile and lead-like properties motivated us to
initiate hit to lead chemistry to explore the activity of this class of
compounds as Tie-2 inhibitors.
We were aware of structural similarities of 1 and the imidazo-
pyrimidine class of p38 inhibitors (e.g., 2, SB203580),10 but these
compounds were not active in our Tie-2 assays, which presumably
reflects different structural requirements for inhibition of Tie-2
and p38. Our initial assumption, based on docking in a Tie-2
homology model (Fig. 1), was that the pyrimidine N1 and NH2
were likely to form a hydrogen bond pair with the hinge domain
of Tie-2 and that the phenyl ring would be directed towards the
selectivity pocket. We wanted to test these assumptions and
nase inhibitor which inhibited the enzyme with an IC50 of 6 l
M.7
Tie-2 inhibitors need to be cell permeable to reach the intracellular
* Corresponding author. Tel.: +44 1625 514495.
E-mail address: richard.luke@astrazeneca.com (R.W.A. Luke).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.06.106

4724
D. Buttar et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4723–4726
A range of imidazole vinyl pyrimidines were synthesized according
11
to the sequence outlined in Scheme 1.
Treatment of the methyl
pyrimidine 3 with the commercially available aldehyde 4 under
either basic or acidic conditions gave the imidazole vinyl pyrimi-
dine 5. In some cases, additional functionality was introduced using
known chemical transformations (vide infra). The compounds were
difficult to separate into the geometric isomers but were predomi-
nantly assigned as the E-isomers by 1H NMR studies.
First, we examined the role of the phenyl ring by replacing it
with Me or Cl. In both cases a significant drop in potency was ob-
served. However, simple substituents such as para-fluoro were
well tolerated (Table 1). These results are consistent with the pro-
posed binding mode. Next, we modified the amino substituent of
the pyrimidine ring. Small substituents such as NHMe or NHAc
are well tolerated, but larger substituents or groups lacking an
NH donor were not (Table 2). Results for isomers of the pyrimidine
showed good potency for the two isomers with a nitrogen para to
the point of attachment of the alkene linker but much reduced
activity for the compound without a nitrogen there (Table 3). Again
all these results are consistent with the proposed binding mode.
Next, we turned our attention to the imidazothiazole ring.
Groups with good H-bond acceptors in a position equivalent to
the N of the imidazothiazole were well tolerated but groups lack-
ing this were not (Table 4). These results are consistent with this
Figure 1. Proposed binding mode for compounds to Tie-2.
explore the importance of other structural features such as the imi-
dazothiazole ring and the alkene linker.
O
S
N
N
S
N
N
N
N
N
NH2
1
2
F
H
S
N
S
N
N
N
a
O
N
N
N
N
+
NH2
NH2
5
3
4
Scheme 1. General synthetic route. Reagents and conditions: (a) (LDA or nBuLi, 0 °C, then TFA) or (H2SO4, AcOH, 50–100 °C).
Table 1
Results for variation of the phenyl position
S
N
N
N
N
NH2
R
a
a
a
Compound
R
Ratio E:Z
Tie-2 IC50
(lM)
Flt-1 IC50
(
lM)
KDR IC50 (lM)
1
6
7
8
Ph–
p-Fluoro-Ph
Me–
>95:5
>95:5
>95:5
>95:5
6.0
13
>100
>100
30
nd
>100
>100
7.0
nd
70
31
Cl–
nd, not determined.
a
Values are geometric means of two or more experiments with a standard deviation of < 0.3 log units.
Table 2
Results for variation of the amino position
S
N
N
N
N
R
a
a
a
Compound
R
Ratio E:Z
Tie-2 IC50
(l
M)
Flt-1 IC50
(l
M)
KDR IC50 (lM)
1
9
NH2
>95:5
7:1
3:1
>95:5
7:3
3:2
6.0
14
>100
68
11
41
77
84
>100
>100
100
96
8
17
NHMe
NHCHMePh
NHPh
NHAc
NHCOtBu
Cl
10
11
12
13
14
15
>100
>100
4.5
13
>100
77
>95:5
>95:5
55
27
66
>100
H
a
Values are geometric means of two or more experiments with a standard deviation of < 0.3 log units.

D. Buttar et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4723–4726
4725
Table 3
Results for variation of the pyrimidine ring
X3
S
N
X1
X2
N
NH2
X1/X2/X3
Ratio E:Z
Tie-2 IC50
(l
M)
Flt-1 IC50
(l
M)
KDR IC50 (lM)
a
a
a
Compound
1
16
17
N/N/CH
CH/N/N
N/CH/N
>95:5
3:2
2:1
6.0
>100
5.2
77
>100
>100
8
>100
35
nd, not determined.
a
Values are geometric means of two or more experiments with a standard deviation of < 0.3 log units.
Table 4
Results for variation of the imidazothiazole ring
N
N
NH2
Het
Het =
H
N
S
N
S
N
N
N
N
N
N
S
N
Me
18
19
20
21
22
23
24
a
a
a
a
Compound
Ratio E:Z
Tie-2 IC50
(lM)
Flt-1 IC50
(l
M)
KDR IC50
(
lM)
Tie-2 cell IC50 (lM)
1
18
19
20
21
22
23
24
>95:5
>95:5
>95:5
19:1
4:1
24:1
6.0
83
64
>100
5.6
2.9
>100
8.4
77
nd
>100
>100
79
23
>100
68
8
nd
>100
>100
13
3.2
>100
9.8
0.33
nd
nd
nd
nd
nd
nd
0.37
>95:5
>95:5
nd, not determined.
a
Values are geometric means of two or more experiments with a standard deviation of < 0.3 log units.
nitrogen forming an important H-bond with the kinase as has been
seen with the related p38 inhibitors10 and was predicted to be pos-
sible from our Tie-2 homology model. The good potency of the sim-
ple imidazole analogue 24 was intriguing and we decided to
explore this more thoroughly as it offered the advantage of re-
duced molecular weight and lipophilicity with no loss in potency.
First, we varied substituents on the imidazole ring nitrogens
(Table 5). Methylation of the nitrogen distal to the alkene linkage
25 led to reduced potency, consistent with our proposal that this
nitrogen forms an important H-bond acceptor. The other nitrogen
was reasonably tolerant of substitution with the NMe compound
26 being of particular interest.
Table 5
Results for variation of the imidazole ring
R1
N
N
N
N
NH2
R2
R1
R2
Ratio E:Z
Tie-2 IC50
(
lM)
Flt-1 IC50
(lM)
KDR IC50
(
lM)
Tie-2 cell IC50 (lM)
a
a
a
a
Compound
24
25
26
27
28
29
H
—
—
Me
—
—
—
>95:5
>95:5
>95:5
>95:5
>95:5
>95:5
8.4
>100
3.8
68
21
2.8
68
>100
34
nd
>100
31
9.8
7.8
3.2
nd
32
7.0
0.37
nd
0.34
nd
nd
nd
Me
CH2CO2H
CH2CH2-1-pyrollidine
CH2Ph
—
nd, not determined.
a
Values are geometric means of two or more experiments with a standard deviation of < 0.3 log units.

4726
D. Buttar et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4723–4726
Y.; Hodous, B. L.; Hoffman, D.; Johnson, R. E.; Kendall, R.; Kim, J. L.; Long, A. M.;
Table 6
Results for variation of the linker group
McGowan, D.; Morrison, M.; Olivieri, P. R.; Patel, V. F.; Polverino, A.; Powers, D.;
Rose, P.; Wang, L.; Zhao, H. J. Med. Chem. 2007, 50, 627; (d) Hodous, B. L.;
Geuns-Meyer, S. D.; Hughes, P. D.; Albrecht, B. K.; Bellon, S.; Caenepeel, S.; Cee,
V. J.; Chaffee, S. C.; Emery, M.; Fretland, J.; Gallant, P.; Gu, Y.; Johnson, R. E.; Kim,
J. L.; Long, A. M.; Morrison, M.; Olivieri, P. R.; Patel, V. F.; Polverino, A.; Rose, P.;
Wang, L.; Zhao, H. Bioorg. Med. Chem. Lett. 2007, 17, 2886; (e) Semones, M.;
Feng, Y.; Johnson, N.; Adams, J. L.; Winkler, J.; Hansbury, M. Bioorg. Med. Chem.
Lett. 2007, 17, 4756.
N
L
N
N
N
NH2
7. Enzyme assays (Tie-2/Flt-1/KDR). The recombinant genes encoding the kinase
domain containing fragments were cloned and expressed in standard
baculovirus/Sf21 system. Lysates were prepared from the host insect cells
following protein expression, by treatment with ice-cold lysis buffer (20 mM
Hepes, pH 7.5, 150 mM NaCl, 10% glycerol, 1% Triton X-100, 1.5 mM MgCl2,
1 mM EDTA), plus protease inhibitors, and then cleared by centrifugation. Nunc
a
a
a
Compound
L
Tie-2 IC50
(
l
M) Flt-1 IC50
(l
M) KDR IC50 (lM)
26
30
31
32
33
34
35
–CH@CH–
–NH–
3.4
68
9.8
>100
98
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
–CONH–
–NHCO–
–CH2NH–
NHCH2CH2–
–
Cyclopropyl–
–C–C–
MaxisorbTM 96-well immunoplates were coated with 100 ll of synthetic
peptide Sigma P3899 (1 mg/ml stock solution in PBS diluted 1:500 in PBS prior
to plate coating) and incubated at 4 °C overnight. Plates were washed in 50 mM
Hepes, pH 7.4, at room temperature to remove any excess unbound synthetic
peptide. Tie-2, KDR or Flt1 activities were assessed by incubation of the
appropriate freshly diluted lysates (1:200, 1:400, and 1:1000, respectively) in
peptide coated plates for 60 min (Tie-2) or for 20 min (KDR, Flt) at room
36
6.4
36
2.9
temperature in 100 mM Hepes, pH 7.4, adenosine trisphosphate (ATP) at 5 lM
a
Values are means of two or more experiments with a standard deviation of
for the respective enzyme, 10 mM MnCl2, 0.1 mM Na3VO4, 0.2 mM DL
dithiothreitol (DTT), 0.1% Triton X-100 together with the test compound
dissolved in DMSO (final concentration of 2.5%) with final compound
< 0.3 log units.
Replacements for the alkene linker were poorly tolerated with
the exception of alkyne 36 (Table 6). However no oral exposure
was seen with this compound.
Compound 26 had a particularly attractive overall balance of
properties. In the phospho Tie-2 cell assay it was reasonably potent
(IC50 340 nM). Selectivity was maintained or improved with the
concentrations ranging from 0.05 to 100 lM. Reactions were terminated by
the removal of the liquid components of the assay followed by washing of the
plates with PBS T (phosphate buffered saline with 0.5% Tween 20) or an
alternative equivalent wash buffer. The immobilized phospho peptide product
of the reaction was detected by immunological methods. Firstly, plates were
incubated for 4 h at room temperature with murine monoclonal anti-
phosphotyrosin–HRP (Horseradish Peroxidase) conjugated antibodies (4G10
from Upstate Biotechnology UBI 16-105). Following extensive washing with
PBS-T, HRP activity in each well of the plate was measured colorimetrically
using 220 Azino di [3 ethylbenzthiazoline sulfonate (6)] diammonium salt
crystals ABTS (Sigma P4922) as a substrate incubated for 30–45 min to allow
greatest potency being against p38 (IC50 12 lM) and a clean profile
in an extended panel of 45 kinases (all <50% inhibition at 10
lM).
In contrast to its Tie-2 cell potency, it was inactive (IC50 > 10
lM)
colour development, before 100 ll of 1 M H2SO4 was added to stop the
in a HUVEC proliferation assay12 in response to VEGF indicating
no inhibition of VEGFRs in cells and it was also inactive in a p38
reaction. Quantification of colour development and thus enzyme activity was
achieved by the measurement of absorbance at 405nm on a Molecular Devices
ThermoMax microplate reader.
cell assay.13 It had good physical properties: solubility 280
l
M at
pH 7.4, 12.5% free in mouse, 9.9% free in rat. It did not inhibit
CyP450’s significantly (IC50’s > 5 M) and showed only moderate
8. Cellular Tie-2 autophosphorylation assay. Tie-2 cell potency was determined
using CHOK1 cells stably transfected with human Tie-2. Cells were seeded at
l
6 ꢀ 104 cells/well in 250
ll DMEM, G418, 10% FCS into 96-well plates and
grown for 3 days prior to assaying. On the day of assay media was removed
and cells were dosed with compound in DMEM plus 1% FCS for 45 min at
turnover in in vitro rat and mouse microsome metabolism studies.
The stability in microsomes was in stark contrast to the initial hit
(1), which had shown high turnover. In a cassette dosed mouse PK
experiment a 2 mg/kg dose gave a Cmax of 0.46
0.58 M h. However this compound and many of these compounds
suffered from photo instability with respect to cis–trans isomeriza-
tion and also reacted with glutathione in the presence of glutathi-
one S-transferase. Due to these stability issues we sought more
stable analogues and this work will be described in a future paper.
In conclusion, imidazole alkene pyrimidines such as 26 are a
new class of kinase inhibitors with good potency in a Tie-2 cellular
assay and oral bioavailability.
37 °C. Cells were washed and lysed in 100 ll lysis buffer (20 mM Tris, pH 7.6,
150 mM NaCl, 50 mM NaF, 0.1% SDS, 1% NP40, 0.5% DOC, 1 mM
orthovanadate, 1 mM EDTA, 1 mM PMSF, 30 ml/ml Aprotinin, 10 mg/ml
Pepstatin, 10 mg/ml Leupeptin) on ice for 5 min. Lysates were transferred to
Quantikine Immunoassay kit for human Tie-2, (R&D systems), and shaken for
2 h at room temperature. Unbound cell lysate was removed by washing four
times with the supplied wash buffer, prior to addition of anti-phosphorlyated
Tie-2 antibody (4221B Cell Signalling Technologies). Plates were placed on a
shaker for 2 h at room temperature. Unbound secondary antibody was
lM and an AUC of
l
removed by washing four times with the supplied wash buffer.
antibody was detected using goat anti-rabbit HRP conjugated antibody
(P0448 Dako). Following 2 h of incubation at RT, plates were washed
P Tie-2
a
further four times and the degree of Tie-2 phosphorylation was determined
by addition of the colour reagent supplied by R&D Systems. The reaction was
stopped after 30 min by addition of stop reagent and optical density was read
at 450 nm.
References and notes
9. Competitor compounds were also less potent in our Tie-2 enzyme assay than
reported in the literature, for example, CP-547,632 has a reported Tie-2 enzyme
1. Folkman, J. Annu. Rev. Med. 2006, 57, 1.
IC50 of 0.048 l lM.
M,14 but in our Tie-2 enzyme assay it had an IC50 of 0.62
2. Morgensztern, D.; Govindan, R. Exp. Rev. Anticancer Therapy 2006, 6, 545.
3. Kim, I.; Kim, H. G.; Moon, S.-O.; Chac, S. W.; So, J.-N.; Koh, K. N.; Ahn, B. C.; Koh,
G. Y. Circ. Res. 2000, 86, 952.
Protein digestion and mass spectroscopy studies indicated that the enzyme
was phosphorylated on Y897, perhaps due to its expression in insect cells,
which has been suggested to lead to low specific activity.15
4. Thurston, G. Cell Tissue Res. 2003, 314, 61.
10. Lee, J. C.; Kumar, S.; Griswold, D. E.; Underwood, D. C.; Votta, B. L.; Adams, J. L.
Immunopharmacology 2000, 47, 185.
11. Luke, R. W. A. WO Patent 2004058776, 2004.
5. (a) Kim, K. L.; Shin, I.-S.; Kim, J.-M.; Choi, J.-H.; Byun, J.; Jeon, E.-S.; Suh, W.; Kim,
D.-K. Cardiovascular Res. 2006, 72, 394; (b) Miyazaki, Y.; Matsunaga, S.; Tang, J.;
Maeda, Y.; Nakano, M.; Philippe, R. J.; Shibahara, M.; Liu, W.; Sato, H.; Wang, L.;
Nolte, R. T. Bioorg. Med. Chem. Lett. 2005, 15, 2203.
12. Hennequin, L. F.; Thomas, A. P.; Johnstone, C.; Stokes, E. S. E.; Plé, P. A.;
Lohmann, J.-J. M.; Ogilvie, D. A.; Dukes, M.; Wedge, S. R.; Curwen, J. O.;
Kendrew, J.; Lambert-van der Brempt, C. J. Med. Chem. 1999, 42, 5369.
13. Brown, D. S.; Belfield, A. J.; Brown, G. R.; Campbell, D.; Foubister, A.; Masters, D.
J.; Pike, K. G.; Snelson, W. L.; Wells, S. L. Bioorg. Med. Chem. Lett. 2004, 14, 5383.
14. Beebe, J. S.; Jani, J. P.; Knauth, E.; Goodwin, P.; Higdon, C.; Rossi, A. M.; Emerson,
E.; Finkelstein, M.; Floyd, E.; Harriman, S.; Atherton, J.; Hillerman, S.;
Soderstrom, C.; Kou, K.; Gant, T.; Noe, M. C.; Foster, B.; Rastinejad, F.; Marx,
M. A.; Schaeffer, T.; Whalen, P. M.; Roberts, W. G. Cancer Res. 2003, 63, 7301.
15. Shewchuk, L. M.; Hassell, A. M.; Ellis, B.; Holmes, W. D.; Davis, R.; Horne, E. L.;
Kadwell, S. H.; McKee, D. D.; Moore, J. T. Structure 2000, 8, 1105.
6. (a) Ji, Z.; Ahmed, A. A.; Albert, D. H.; Bouska, J. J.; Bousquet, P. F.; Cunha, G. A.;
Glaser, K. B.; Guo, J.; Li, J.; Marcotte, P. A.; Moskey, M. D.; Pease, L. J.; Stewart, K.
D.; Yates, M.; Davidsen, S. L.; Michaelides, M. R. Bioorg. Med. Chem. Lett. 2006,
16, 4326; (b) Hodous, B. L.; Geuns-Meyer, S. D.; Hughes, P. E.; Albrecht, B. K.;
Bellon, S.; Bready, J.; Caenepeel, S.; Cee, V. J.; Chaffee, S. C.; Coxon, A.; Emery,
M.; Fretland, J.; Gallant, P.; Gu, Y.; Hoffman, D.; Johnson, R. E.; Kendall, R.; Kim,
J. L.; Long, A. M.; Morrison, M.; Olivieri, P. R.; Patel, V. F.; Polverino, A.; Rose, P.;
Tempest, P.; Wang, L.; Whittington, D. A.; Zhao, H. J. Med. Chem. 2007, 50, 611;
(c) Cee, V. J.; Albrecht, B. K.; Geuns-Meyer, S.; Hughes, P.; Bellon, S.; Bready, J.;
Caenepeel, S.; Chaffee, S. C.; Coxon, A.; Emery, M.; Fretland, J.; Gallant, P.; Gu,