The discovery of N-(1,3-thiazol-2-yl)pyridin-2-amines as potent inhibitors of KDR kinase

Article abstract of DOI:10.1016/j.bmcl.2004.03.052

An azo-dye lead was modified to a novel N-(1,3-thiazol-2-yl)pyridin-2-amine series of KDR kinase inhibitors through the use of rapid analog libraries. This new class has been found to be potent, selective, and of low molecular weight. Molecular modeling has postulated an interesting conformational preference and binding mode for these compounds in the active site of the enzyme.

Full text of DOI:10.1016/j.bmcl.2004.03.052

Bioorganic & Medicinal Chemistry Letters 14 (2004) 2941–2945
The discovery of N-(1,3-thiazol-2-yl)pyridin-2-amines as potent
inhibitors of KDR kinase
Mark T. Bilodeau,a,* Leonard D. Rodman,a Georgia B. McGaughey,b Kathleen E. Coll,c
Timothy J. Koester,a William F. Hoffman,a Randall W. Hungate,a Richard L. Kendall,c
Rosemary C. McFall,c Keith W. Rickert,c Ruth Z. Rutledgec and Kenneth A. Thomasc
aDepartment of Medicinal Chemistry, Merck Research Laboratories, PO Box 4, West Point, PA 19486, USA
bDepartment of Molecular Systems, Merck Research Laboratories, PO Box 4, West Point, PA 19486, USA
cDepartment of Cancer Research, Merck Research Laboratories, PO Box 4, West Point, PA 19486, USA
Received 27 January 2004; accepted 11 March 2004
Abstract—An azo-dye lead wasmodified to a novel N-(1,3-thiazol-2-yl)pyridin-2-amine series of KDR kinase inhibitors through the
use of rapid analog libraries. This new class has been found to be potent, selective, and of low molecular weight. Molecular modeling
has postulated an interesting conformational preference and binding mode for these compounds in the active site of the enzyme.
Ó 2004 Elsevier Ltd. All rights reserved.
Abnormal angiogeneissisoperative in the progreison
of diseases including diabetic retinopathy, rheumatoid
arthritis, psoriasis, and cancer.1 The growth of solid
tumorsin particular isdependent on the ingrowth of
new blood vessels.2 Vascular endothelial growth factor
(VEGF) expression has been shown to be up-regulated
in a variety of tumorsand hasbeen reported to be a
prognostic indicator of tumor progression and/or
decreased survival in a variety of tumors. KDR, a trans-
membrane tyrosine kinase-linked receptor for VEGF,
mediates a mitogenic response, and is expressed by
endothelial cells. Inhibition of VEGF activity or KDR
kinase has been shown to inhibit tumor angiogenesis
and the growth of tumorsin animal model.s Bev-
In searching for lead compounds that showed a high
degree of selectivity for KDR inhibition over related
kinases, we were intrigued by the potent and selective
screening hit 1 (Table 1).7 However the molecule pos-
sessed problematic functionality from a toxicology per-
spective, containing both an azo function and a
nitroaromatic moiety. We therefore sought to explore
changesin thisesriesin a rapid analog format. In our
initial effort, we explored amide librariesbaesd on
1
where the planar amide function wasutilized to replace
the azo moiety. These efforts resulted in the discovery of
the benzamide 2, demonstrating that both of the prob-
lematic functionalitiesin 1 could be replaced, in thiscase
with amide and phenyl moieties.8 Although good
potency was maintained with these changes, kinase
selectivity was significantly decreased.
acizumab, a VEGF antibody wasreported to be effica-
3
ciousin Phaes III clinical trial.s
Clinical evaluations
have also been undertaken with an antiKDR antibody,4
a soluble VEGF decoy-receptor,5 aswell assmall mol-
ecule inhibitorsof KDR kinaes activity.
We continued to seek alternative templates eliminating
the amide functionality by creating a library of amino-
thiazoles(Fig. 1). We employed a Hantzcsh thiazole
synthesis, reacting 33 thioureas consisting of alkyl,
aromatic, and heteroaromatic exampleswith (1-bromo-
2,2-dimethoxyethyl)benzene9 to produce 5-phen-
yl substituted aminothiazoles. One compound stood
out upon testing, namely the N-(1,3-thiazol-2-yl)pyridin-
2-amine 3 (IC50 ¼ 7 nM). Thiscompound had isgnifi-
cantly greater activity than any other membersof the
library. For the subsequent compounds discussed the
6
The impor-
tance of KDR in the signaling pathway of VEGF makes
thisenzyme a particularly attractive target againts
tumor angiogenesis.
Keywords: KDR kinase; VEGF.
* Corresponding author. Tel.: +1-215-652-5304; fax: +1-215-652-7310;
e-mail: mark_bilodeau@merck.com
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.03.052

2942
M. T. Bilodeau et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2941–2945
Table 1. Potency and selectivity profile of lead compounds
Me
N
O
N
NO2
N
Ph
S
N
S
N
H
n-Bu
N
1
2
Me
Compd
KDR IC50 (nM)a
Fold-selectivity versus KDRb
Flt-1
Flt-4
FGFR-1
FGFR-2
Src
1
2
151 ꢀ 30
81 ꢀ 23
>130
4
1
1
>130
72
>130
14
>130
>240
a For determinationswhere n P 2, standard deviation is given.
b Ratio of given enzyme IC50 divided by KDR IC50
.
Table 2. Heterocycle substitution of lead
H
N
Br
N
H
N
N
S
(MeO)2HC
Ph
R
N
S
R
S
N
H
NH2
Ph
3
HCl, EtOH
reflux
Ph
R = 33 groups
KDR IC50 = 7 2 nM
Compd
R
KDR IC50 (nM)a
Figure 1. 2-Aminothiazole library.
N
4
23 ꢀ 10
N
N
H
R
R
Cl
N
NH2
N
N
5
6
7
36 ꢀ 8
137
NaH, THF
reflux
N
+
X
N
N
X
R
R
N
Figure 2. Alternate synthesis of aminothiazoles.
N
N
N
240
productswere made according to the chemistry in Fig-
ure 1 or, alternately, according to Figure 2. In thislatter
sequence, aminothiazoles were made by nucleophilic
substitution of aminoheterocycle and 2-chlorothiazoles.
8
9
259
124
The SAR of this new lead class was first explored with
aromatic substitution for the pyridine (Table 2). Het-
eroaromaticsthat place a baisc nitrogen ortho- to the
amino group tend to be required for a high level of
potency. Hence the 2-pyrazine 4 and the 4-pyrimidine 5
had potency moderately lower than 3. An exception was
the 2-pyrimidine (6), which showed significantly less
activity. The 3-pyridyl (7), 4-pyridyl (8), and phenyl
analogs( 9) showed similarly diminished levels of activ-
ity.
a For determinationswhere n P 2, standard deviation is given.
poorly tolerated. Alternatively, we have sought to
replace the phenyl moiety with smaller substituents.
Halides and simple alkyl groups (see 19, 20, and 21)
provided analogswith significantly lower potency than
the corresponding phenyl thiazoles. Remarkably, we
found that replacement with cyano afforded analog 22
with only a moderate loss in potency. We had not
anticipated that cyano would be a good substitution for
a large lipophilic phenyl group but thismodification
The SAR of the pyridine ring wasgenerally tolerant of a
wide variation of substitution (Table 3). Methyl group
substitution was acceptable at every position (com-
pounds 10–13). In addition, electron withdrawing
functionality was also well tolerated based on com-
pounds 14 and 15. An electron donating methoxy sub-
stituent at the 3-position (16) resulted in a drop in
potency. Methoxysubstitution at the 4- or 6-position (17
and 18) wasbetter tolerated.
provided
a
remarkably low molecular weight
(MW ¼ 202) and potent (IC50 ¼ 36 nM) inhibitor of
KDR kinase.
Several changesto the core were made that are impor-
tant to understanding the SAR of the series. Methyl-
ation of the central NH destroys potency. Replacement
of the NH with an ether linkage also abolishes activity.
The most remarkable observation has been that the
We have explored substitution of the thiazoyl-5-phenyl
moiety and found that in all cases substitution leads to
lower potency (data not shown). Even relatively con-
servative changes, such as fluorine substitution are

M. T. Bilodeau et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2941–2945
2943
Table 3. Pyridine substituent SAR
(11 and 12) show levels of selectivity similar to the
parent 3. Electron withdrawing groupssuch as5-Cl ( 14)
and 6-Br (15) tend to show even more dramatic
improvementsin the levelsof eslectivity verussthe
H
N
N
R1
N
S
R2
FGFRsand Src. Interetsingly, the 5-cyano thiazole 22
provided a very different profile, which ismore analo-
gousto the nitrothiazole lead 1. It hasisgnificantly
enhanced levels of selectivity versus two of the most
closely related kinases, Flt-1, and PDGFRb, while still
providing little selectivity versus Flt-4 and maintaining
good selectivity versus the FGFRs and Src.
Compd
R1
R2
IC50 (nM)a
3
H
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Br
7 ꢀ 2
13 ꢀ 5
6 ꢀ 2
10
11
12
13
14
15
16
17
18
19
20
21
22
3-Me
4-Me
5-Me
6-Me
5-Cl
6-Br
3-OMe
4-OMe
6-OMe
H
19 ꢀ 12
10 ꢀ 3
35 ꢀ 25
14 ꢀ 5
46
Although the lead compound 3 hasexcellent activity in
the enzyme assay, poor activity in a cell-based assay of
11 ꢀ 0
3 ꢀ 2
10
KDR inhibition wasobesrved (Fig. 4).
However we
71 ꢀ 26
194
1274
have found that the pyridine moiety can be substituted
with a basic amine, which serves to improve the physical
propertiesand cell potency of the resulting analogs. The
pyrrolidine containing molecule 25, had excellent cell
potency (cell IC50 ¼ 28 nM), while showing no loss of
enzyme activity relative to 3.
H
Cl
H
Me
CN
H
36 ꢀ 8
a For determinationswhere n P 2, standard deviation is given.
We initiated modeling studies to understand the con-
formational preferencesof these moleculesand develop
a theory of their binding modes.11 Computational
analysis resulted in four relative energy minima that are
all close to planar in conformation (see dihedral angles).
The lowest energy conformations for the thiazole series,
regardless of substitution pattern, are conformers A and
B (Table 5). ConformersC and D were found to be
significantly higher in energy. Conformer A is likely to
be the active conformer since 3-pyridine substitution
(compound 10) doesnot affect potency, while it would
be expected to disfavor conformer B but not conformer
A. Thisargument led usto a biasfor A asthe active
conformation. A notable validation of thisview isfound
H
N
H
N
N
N
N
O
O
Ph
Ph
23
24
KDR IC50 = 1880 nM
KDR IC50 = 134 nM
Figure 3. Oxazoles.
oxazole analog 23 is >250-fold less active than the thi-
azole 3 (Fig. 3). However, it isinteresting to note that
the phenyl oxazole 24 isequipotent with the phenyl
thiazole 9.
We have profiled these inhibitors for selectivity versus a
series of closely related kinases (see Table 4). Compound
3 has no selectivity versus Flt-1, Flt-4, or PDGFRb and
displays respectable levels of selectivity versus FGFR1,
FGFR2, and Src. Certain pyridyl analogshave hs own
enhanced selectivity, including the 3-methyl substituted
compound 10, which shows enhanced levels of selectiv-
ity versus the FGFRs and Src. The 6-Me compound 13
has a similar profile, while the 4- and 5-Me substituents
H
H
N
N
N
N
N
N
S
N
S
Ph
Ph
3
25
KDR IC50 = 7 2 nM
Cell IC50 = 297 nM
KDR IC50 = 6 2 nM
Cell IC50 = 28 nM
Figure 4. Activity of thiazoles in a cell-based assay.
Table 4. Kinase fold selectivity versus KDRa
H
N
N
R1
N
S
R2
Compd
R1
R2
Flt-1
Flt-4
PDGFRb
FGFR1
FGFR2
Src
3
H
Ph
Ph
Ph
Ph
Ph
Ph
Ph
CN
3
16
7
2
3
2
1
3
1
3
4
0.3
1
164
820
51
200
50
280
1200
190
10
11
12
13
14
15
22
3-Me
4-Me
5-Me
6-Me
5-Cl
6-Br
H
0.5
0.3
0.3
0.1
0.3
68
220
220
5
70
570
760
8
600
210
>580
400
107
6
>580
1620
214
>580
840
678
12
30
a Ratio of the given enzyme IC50 divided by the KDR IC50
.

2944
M. T. Bilodeau et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2941–2945
Table 5. Relative conformational energies(RHF/6-31G**) and dihedral angles
H
N
H
N
H
N
H
N
6
4
5
N
N
1
X
N
N
X
3
2
X
N
X
N
N
N
B
D
A
C
Thiazole (X ¼ S)
s1 (N1–C2–N3–H4)
s2 (H4–N3–C5–C6)
0.00 kcal/mol
0.0°
0.0°
0.03 kcal/mol
ꢁ179.9°
3.15 kcal/mol
11.7°
ꢁ168.8°
2.23 kcal/mol
8.00 kcal/mol
ꢁ148.6°
16.4°
180.0°
Oxazole (X ¼ O)
s1 (N1–C2–N3–H4)
s2 (H4–N3–C5–C6)
5.78 kcal/mol
8.3°
21.6°
0.00 kcal/mol
179.9°
180.0°
7.72 kcal/mol
179.9
0.0°
180.0°
0.0°
Table 6. Relative conformational energies (RHF/6-31G**) for substituted thiazoles
H
H
H
N
H
N
N
S
N
N
N
N
S
N
R
R
N
N
S
N
N
S
D
B
C
A
R
R
R ¼ Ph
0.00 kcal/mol
0.00 kcal/mol
0.065 kcal/mol
1.63 kcal/mol
3.16 kcal/mol
4.30 kcal/mol
8.06 kcal/mol
9.05 kcal/mol
R ¼ CN
in the analysis of the oxazole (Table 5). The oxazole
series conformer A is remarkably higher in energy than
the lowest energy conformer B (by 5.78 kcal/mol). This
conformational argument accountsfor the dramatically
lower potency of the oxazole 23. The equal activity of
the phenyl amino thiazole 9 and the phenylaminooxaz-
ole 24 isfurther evidence that the lack of activity in 23 is
due to a conformational biasaway from the active
conformer rather than an inherent low enzyme affinity
for the oxazole moiety.
itive (data not shown), we have postulated that the
aminothiazole engagesin hydrogen bonding with the
Cys919 in the enzyme active site, with the 5-phenyl
moiety occupying the H1 hydrophobic pocket.12 Only
conformersA and C allow the thiazole nitrogen to
hydrogen bond with the backbone amide proton of
Cys919 and the amine proton to hydrogen bond to the
backbone carbonyl oxygen of Cys919. But conformer C
isisgnificantly higher in energy than A (by 3.15 kcal/
mol). This model also accounts for the loss of potency
upon substitution of the thiazole 5-phenyl as well as for
the ability to substitute the pyridine ring at the 4-, 5-,
and 6-positions without diminishing activity as these
positions point towards solvent.
In examining the effect of the 5-thiazole substitution on
these conclusions, the phenyl moiety creates only slight
changesto the relative energies(Tables5 and 6). How-
ever, the nitrile substituent creates a remarkable bias
toward the single conformer A. The energetic cost to
achieve conformation B in the nitrile case is 1.63 kcal/
mol (Table 6). Thismay in part account for the usr-
prising level in potency for the nitrile 23, asit hasgreater
biastowardsthe postulated active conformer.
From a structurally undesirable lead we have developed
very potent, small molecule inhibitors of KDR kinase.
These compounds exhibit promising levels of selectivity
versus closely related kinases. We have found a very low
molecular weight molecule that surprisingly replaces a
nitrile function for a phenyl moiety. Thiscompound has
been shown to be a potent and selective inhibitor. From
molecular modeling, a consistent picture has emerged
that indicatesan interetsing conformational preference
of these molecules and a preferred binding mode in the
enzyme active site. This model also helps account for
SAR observed in the series. Future communications will
discuss the optimization of this novel lead series.
These considerations lead us to postulate the following
enzyme binding mode (Fig. 5). Based on the SAR of the
series and the fact that the inhibitors are ATP-compet-
References and notes
1. Adamis, A. P.; Shima, D. T.; Yeo, K. T.; Yeo, T. K.;
Brown, L. F.; Berse, B.; D’Amore, P. A.; Folkman
J. Biochem. Biophys. Res. Commun. 1993, 193, 631;
Giatromanolaki, A.; Sivridis, E.; Athanassou, N.; Zois,
E.; Thorpe, P. E.; Brekken, R. A.; Gatter, K. C.; Harris,
A. L.; Koukourakis, I. M.; Koukourakis, M. I. J. Pathol.
2001, 194, 101; Detmar, M. Dermatol. Sci. 2000, 24, S78;
Carmeliet, P.; Jain, R. K. Nature 2000, 407, 249; Folkman,
J. Nature Med. 1995, 1, 27.
Figure 5. Proposed binding mode of 3 in the KDR kinase active site.

M. T. Bilodeau et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2941–2945
2945
2. (a) Veikkola, T.; Karkkainen, M.; Claesson-Welsh, L.;
Alitalo, K. Cancer Res. 2000, 60, 203; (b) Thomas, K. A.
J. Biol. Chem. 1996, 271, 603.
3. Yang, J. C.; Haworth, L.; Sherry, R. M.; Hwu, P.;
Schwartzentruber, D. J.; Topalian, S. L.; Steinberg, S. M.;
Chen, H. X.; Rosenberg, S. A. N. Engl. J. Med. 2003, 349,
427; McCarty, M. Lancet 2003, 361, 1959.
9. Bellesia, F.; Boni, M.; Ghelfi, F.; Pagnoni, U. M. Gazz.
Chim. Ital. 1993, 123, 629.
10. The Cell IC50 value represents the inhibition of VEGF-
stimulated mitogenesis as determined in human umbilical
vein endothelial cells.
11. Moleculeswere minimized uisng the AM1. Hamiltonian
(Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart,
J. J. P. J. Am. Chem. Soc. 1985, 107, 3902) asimplemented
in Ampac 6.55 (distributed by Semichem, 7204 Mullen,
Shawnee, KS 66216 (info@semichem.com). The AM1
minimized structures were used as the initial conformation
used in additional quantum mechanical calculations
(Gaussian 98 (Revision A.9) Frisch, M. J.; Trucks,
G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.,
Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam,
J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas,
O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.;
Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.;
Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghav-
achari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.;
Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill,
P. M. W.; Johnson, B. G.; Chen, W.; Wong, M. W.;
Andres, J. L.; Head-Gordon, M.; Replogle E. S.; Pople,
J. A. Gaussian, Inc., Pittsburgh PA, 1998). All bonds,
angles, and torsions were allowed to optimize at the HF/6-
31G** level of theory. Frequencieswere calculated to
ensure that a minimum was achieved.
4. Lu, D.; Jimenez, X.; Zhang, H.; Bohlen, P.; Witte, L.;
Zhu, Z. Int. J. Cancer 2002, 97, 393.
5. Holash, J.; Davis, S.; Papadopoulos, N.; Croll, S. D.; Ho,
L.; Russell, M.; Boland, P.; Leidich, R.; Hylton, D.;
Burova, E.; Ioffe, E.; Huang, T.; Radziejewski, C.; Bailey,
K.; Fandl, J. P.; Daly, T.; Wiegand, S. J.; Yancopoulos, G.
D.; Rudge, J. S. Proc. Natl. Acad. Sci. U.S.A. 2002, 99,
11393.
6. For recent reviews, see: (a) Bilodeau, M. T.; Fraley, M. E.;
Hartman, G. D. Expert Opin. Investig. Drugs 2002, 11,
737; (b) Boyer, S. J. Curr. Top. Med. Chem. 2002, 2, 973.
7. The KDR IC50 value represents biochemical inhibition of
phosphorylation of a poly-Glu/Tyr (4:1) peptide substrate
by isolated KDR kinase (cloned and expressed as a GST-
fusion protein): see, Kendall, R. L.; Rutledge, R. Z.; Mao,
X.; Tebben, A. L.; Hungate, R. W.; Thomas, K. A. J. Biol.
Chem. 1999, 274, 6453.
8. Aminothiazole amideshave been decsribesasCDK
inhibitors: Kim, K. S.; Kimball, S. D.; Misra, R. N.;
Rawlins, D. B.; Hunt, J. T.; Xiao, H. Y.; Lu, S. F.; Qian,
L. G.; Han, W. C.; Shan, W. F.; Mitt, T.; Cai, Z. W.; Poss,
M. A.; Zhu, H.; Sack, J. S.; Tokarski, J. S.; Chang, C. Y.;
Pavletich, N.; Kamath, A.; Humphreys, W. G.; Marathe,
P.; Bursuker, I.; Kellar, K. A.; Roongta, U.; Batorsky, R.;
Mulheron, J. G.; Bol, D.; Fairchild, C. R.; Lee, F. Y.;
Webster, K. R. J. Med. Chem. 2002, 45, 3905.
12. Traxler, P.; Furet, P. Pharmacol. Ther. 1999, 82, 195.