Synthesis, structural analysis, and SAR studies of triazine derivatives as potent, selective Tie-2 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2007.02.067

A novel class of selective Tie-2 inhibitors was derived from a multi-kinase inhibitor 1. By reversing the amide connectivity and incorporating aminotriazine or aminopyridine hinge-binding moieties, excellent Tie-2 potency and KDR selectivity could be achi

Full text of DOI:10.1016/j.bmcl.2007.02.067

Bioorganic & Medicinal Chemistry Letters 17 (2007) 2886–2889
Synthesis, structural analysis, and SAR studies of triazine
derivatives as potent, selective Tie-2 inhibitors
Brian L. Hodous,a,* Stephanie D. Geuns-Meyer,a Paul E. Hughes,d Brian K. Albrecht,a
Steve Bellon,c Sean Caenepeel,d Victor J. Cee,a Stuart C. Chaffee,a Maurice Emery,e
Jenne Fretland,e Paul Gallant,b Yan Gu,b Rebecca E. Johnson,a Joseph L. Kim,c
Alexander M. Long,c Michael Morrison,b Philip R. Olivieri,a Vinod F. Patel,a
Anthony Polverino,d Paul Rose,c Ling Wangd and Huilin Zhaoc
aDepartment of Medicinal Chemistry, Amgen Inc., One Kendall Square, Building 1000, Cambridge, MA 02139, USA
bDepartment of Molecular Pharmacology, Amgen Inc., One Kendall Square, Building 1000, Cambridge, MA 02139, USA
cDepartment of Molecular Structure, Amgen Inc., One Kendall Square, Building 1000, Cambridge, MA 02139, USA
dDepartment of Oncology Research, Amgen Inc., One Amgen Center Drive, Thousand Oaks, CA 91320, USA
eDepartment of Pharmacokinetics & Drug Metabolism, Amgen Inc., One Amgen Center Drive, Thousand Oaks, CA 91320, USA
Received 18 December 2006; revised 16 February 2007; accepted 21 February 2007
Available online 25 February 2007
Abstract—A novel class of selective Tie-2 inhibitors was derived from a multi-kinase inhibitor 1. By reversing the amide connectivity
and incorporating aminotriazine or aminopyridine hinge-binding moieties, excellent Tie-2 potency and KDR selectivity could be
achieved with 3-substituted terminal aryl rings. X-ray co-crystal structure analysis aided inhibitor design. This series was evaluated
on the basis of potency, selectivity, and rat pharmacokinetic parameters.
Ó 2007 Elsevier Ltd. All rights reserved.
Inhibition of angiogenesis,1 the formation of new capil-
laries from preexisting blood vessels, is a promising and
clinically proven approach for limiting tumor growth
and survival.2 Angiogenesis is believed to be necessary
for solid tumors to grow larger than 1–2 mm diameter,
the point at which vasculature is required to supply oxy-
gen and nutrients, and remove waste.2,3 Disrupting the
vascular endothelial growth factor (VEGF)/KDR
(VEGFR-2) signaling cascade is a well-validated ap-
proach for inhibiting tumor growth as demonstrated
by the FDA-approved anti-VEGF molecule, bev-
acizumab (AvastinTM),4 and small molecule kinase inhib-
itors, sunitinib (SutentTM)5 and sorafenib (NexavarTM).6
In contrast to the well-studied VEGF/KDR axis, the
role of the competing interaction of angiopoietins
(Ang-1 and Ang-2) with the Tie-2 receptor (tyrosine ki-
nase with immunoglobulin and epidermal growth factor
homology domains-2) in angiogenesis is less well under-
stood.7 The receptor tyrosine kinase Tie-2 is primarily
expressed in the vascular endothelium and is involved
in vessel branching, sprouting, remodeling, maturation,
and stability.7 Angiopoietin-1 binds to Tie-2 and stimu-
lates tyrosine phosphorylation and signal transduction,
resulting in blood vessel formation and/or mainte-
nance.8 The role of Angiopoietin-2 is controversial and
has been shown to act as both an agonist and an antag-
onist of Tie-2 signaling.7 The outcome of the Ang-2/Tie-2
interaction on vasculature depends on the presence/ab-
sence of VEGF and other proangiogenic factors.7,9
In order to evaluate the effect of Tie-2 kinase inhibitors
on angiogenesis and tumor growth, we sought to develop
small molecule inhibitors with high selectivity against ki-
nases involved in angiogenesis, for example KDR. Pre-
vious work revealed pyridinyl pyrimidine 1 to be a
potent, dual Tie-2/KDR inhibitor that served as a
launching point for SAR studies (Fig. 1).10 Herein, we
describe an isomeric ‘reversed-amide’ series (e.g., 2,
where the amide linkage is reversed compared to 1) that
possesses good in vitro potency, selectivity, and unique
Keywords: Tie-2 kinase; Oncology; Small molecule kinase inhibitors;
Triazine; KDR.
*
Corresponding author. Tel.: +1 617 444 5090; fax: +1 617 621
3908; e-mail: bhodous@amgen.com
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.02.067

B. L. Hodous et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2886–2889
2887
Figure 1. The potent, non-selective pyridinyl pyrimidine 1a modified
by reversing the amide orientation to arrive at 2. 1b,
R = –(CH2)3N(CH3)2; Tie-2 = 1 nM, KDR = 2 nM.
SAR, but sub-optimal pharmacokinetic properties.
X-ray co-crystallographic studies aided the SAR
investigations.
Figure 2. X-ray co-crystal structures of amide 1b (green) and reversed-
amide 2 (brown) bound to the ATP binding site of Tie-2. (Oxygen
atoms (red), nitrogen atoms (blue), and fluorine atoms (light blue).)
All of the ‘reversed-amide’ analogs were synthesized in
the manner outlined for 5 in Scheme 1. 2-N-Methylami-
notriazine 3 can be synthesized on multi-gram scale as
previously described.10 This building block was reacted
with 5-amino-2-methylphenol under basic conditions
to displace the activated chloride and form penultimate
biaryl ether intermediate 4. Biaryl anilines (e.g., 2) were
not pursued due to poor physical properties (mp of
2 = 192 °C vs mp of 5 = 95 °C).10 HATU-mediated cou-
pling of the substituted benzoic acid to 4 afforded final
compound 5 in 40% yield over two steps. 2-N-Methyl-
aminopyrimidine (9) and pyridine (11, 15, 16, 20) deriv-
atives were synthesized in an analogous fashion.10
DFG-motif and Glu872 of the aC-helix: 1b,
˚
H-bonds = 2.92 and 2.91 A; 2, H-bonds = 3.04 and
˚
2.73 A.
Realizing the similarities in binding mode between 1b
and 2, we initiated SAR studies with two modifications
that were previously determined to be optimal: (1)
N-methylamino-substituted hinge-binding heterocycles
provided improved potency versus hydrogen substitu-
tion; (2) the biaryl ether series provided improved
physical properties versus biaryl anilines (e.g., 9–11,
Table 1).10 Although a potent Tie-2 inhibitor, reversed
amide 9, showed no improvement in selectivity over
KDR versus the corresponding amide 6. However, by
changing the pyrimidine (9) to a triazine (10) or pyridine
(11) hinge-binding moiety, selectivity over KDR was
dramatically improved for the reversed amide without
X-ray co-crystal structures of an amide 1b (Tie-2
IC50 = 1 nM, KDR IC50 = 2 nM; similar selectivity pro-
file to 1a) and ‘reversed-amide’ 2 bound to the ATP-site
of the Tie-2 kinase were solved (Fig. 2).11 In each ligand,
the pyrimidine ring makes identical key hydrogen bonds
to the backbone NH of the linker residue, Ala905
Table 1. Comparison of enzyme potency (IC50, nM) for amides and
reversed-amides
˚
˚
(1b = 2.85 A; 2 = 2.88 A). This interaction situates the
pyridine rings in alignment, and within edge-to-face
˚
p-stacking distance of Phe983 (ꢀ3.5 A) of the DFG-mo-
tif (first three residues of the activation loop). The cen-
tral and terminal aryl rings overlay with only slight
differences in orientation, placing the CF3-groups in
essentially the same space in the extended hydrophobic
pocket (EHP) in each series. Both amide arrangements
engage in strong hydrogen bonds with Asp982 of the
Compound
Ar
A
N
B
Enzyme (nM)
Tie-2
KDR
CH
13
38
6
7
8
N
CH
N
CH
1
3
20
5
9
10
11
N
CH
N
2
1
1
2
N
CH
117
18
Scheme 1. Reagents and conditions: (a) 5-amino-2-methylphenol,
Cs2CO3, DMSO, 130 °C, 62%; (b) 3-(1,1,2,2-tetrafluoroethoxy)benzoic
acid, HATU, i-Pr2NEt, CHCl3, rt, 64%.
CH

2888
B. L. Hodous et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2886–2889
Table 2. Tie-2 and KDR potency (IC50, nM) and rat in vivo clearance (mL/h/kg) for central ring variations of triazines and pyridines
Compound
A
X
Enzyme (nM)
KDR
Cellular (nM)
Tie-2
Rat iv CL (mL/h/kg)
Tie-2
10
12
13
14
11
15
16
N
2-Me
H
2-F
1
6
117
456
128
25,000
18
5
85
—
—
11
—
—
3942
4421
1104
—
N
N
4
N
3-CF3
2-Me
H
55
1
CH
CH
CH
3260
—
4
18
9
4-Cl
6
—
compromising potency for Tie-2 enzyme (7 vs 10 and 8
vs 11).
selectivity over KDR and comparable potency on
Tie-2 relative to pyridines (e.g., 10 vs 11 and 12 vs 15).
Encouraged by the low nanomolar potency and 100-fold
selectivity of triazine 10, we explored the SAR of the
central aryl ring (Table 2). Replacement of the 2-methyl
group with a hydrogen resulted in a significant loss in
cellular potency (10 vs 12; Tie-2 autophosphorylation
measured in EA.hy 926 cells). We determined that poor
pharmacokinetic properties were the most significant
issues with this reversed-amide series. Rat clearances
equal to rat liver blood flow were observed for 10, 11,
and 12. Interestingly, a 2-fluoro derivative (13) led to a
much reduced rat iv clearance (CL = 1104 mL/h/kg);
however, this was accompanied by an undesirable half-
life of 0.8 h (mean-residence time = 0.5 h). Other elec-
tron-withdrawing groups such as 3-CF3 (14) resulted
in sub-optimal enzyme inhibition; and 4-chloro lacked
selectivity (16). In general, triazines provided superior
In our SAR studies of the terminal aryl ring, we set out
to reduce the rat in vivo clearance while maintaining po-
tency and selectivity. Reducing the bulk of the hydro-
phobic substituent from isopropyl (10) to CF3 (17)
resulted in a decrease in potency and selectivity
(Table 3). The 3-OCF3 (18) and 3-pyrrolyl (19) groups
imparted very good Tie-2 potency in the cellular auto-
phosphorylation assay (IC50 = 15 and 23 nM, respec-
tively), and impressive cellular selectivity over KDR
(>1000-fold for 19), but these compounds were rapidly
cleared in rats. The highly fluorinated derivative 5, did
lead to a moderate clearance (CL = 889 mL/h/kg) com-
pound that also maintained high potency and selectivity.
The corresponding pyridine derivative 20 led to a high
clearance molecule, although it was very potent in the
Tie-2 cellular assay (IC50 = 13 nM).
Table 3. Tie-2 and KDR potency (IC50, nM) and rat in vivo clearance (mL/h/kg) for terminal ring variation of triazines and pyridines
Compound
A
X
Enzyme (nM)
Cellular (nM)
Tie-2 KDR
Rat iv CL (mL/h/kg)
Tie-2
KDR
17
18
N
N
–CF3
–OCF3
20
8
309
442
—
15
—
—
4980
540
19
N
17
1126
23
25,000
3825
5
20
N
CH
16
9
1002
233
33
13
4300
943
889
3640

B. L. Hodous et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2886–2889
2889
Table 4. Kinase selectivity summary for 5 (IC50, nM)
pharmacokinetics; however, further improvement is re-
quired to enhance oral bioavailability. Modifications
to this promising scaffold that address these parameters
will be reported in subsequent publications.
Enzyme inhibition
Tie-2
p38a
Lck
16
47
cKit
cMet
17,800
25,000
25,000
25,000
25,000
25,000
40,000
146
EGFR
IGFR-1
Zap70
BTK
KDR
Jak2
Src
1002
1746
2195
5360
Supplementary data
FGFR
JNK3
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmcl.
2007.02.067.
Table 5. Rat pharmacokinetic properties for 5a
iv administrationb
po administrationc
t1/2
CL
Vss
1.7 h
889 mL/h/kg
1311 mL/kg
%F
Cmax
4
119 ng/mL
509 ng h/mL
References and notes
AUC0–inf
1. Risau, W. Nature 1997, 386, 671.
2. (a) Folkman, J. Ann. Surg. 1972, 175, 409; (b) Folkman, J.
N. Engl. J. Med. 1971, 285, 1182.
3. (a) Holmgren, L.; O’Reilly, M. S.; Folkman, J. Nat. Med.
1995, 1, 149; (b) Folkman, J. J. Nat. Cancer Inst. 1990,
82, 4.
a N = 3 animals/group.
b Dosed intravenously at 2 mg/kg in DMSO to male Sprague–Dawley
rats.
c Dosed orally at 10 mg/kg as a suspension in 2% HPMC/1% Tween 80/
97% water (pH 2.2) to male Sprague–Dawley rats.
4. Mulcahy, M. F.; Benson, A. B., III Expert Opin. Biol.
Ther. 2005, 5, 997.
5. Sun, L.; Liang, C.; Shirazian, S.; Zhou, Y.; Miller, T.; Cui,
J.; Fukuda, J. Y.; Chu, J.-Y.; Nematalla, A.; Wang, X.;
Chen, H.; Sistla, A.; Luu, T. C.; Tang, F.; Wei, J.; Tang,
C. J. Med. Chem. 2003, 46, 1116.
6. Strumberg, D.; Richly, H.; Hilger, R. A.; Schleucher, N.;
Korfee, S.; Tewes, M.; Faghih, M.; Brendel, E.; Voliotis,
D.; Haase, C. G.; Schwartz, B.; Awada, A.; Voigtmann,
R.; Scheulen, M. E.; Seeber, S. J. Clin. Onc. 2005, 5, 965.
7. Yu, Q. Future Med. 2005, 1, 475.
8. Cho, C.-H.; Kammerer, R. A.; Lee, H. J.; Steinmetz, M.
O.; Ryu, Y. S.; Lee, S. H.; Yasunaga, K.; Kim, K.-T.;
Kim, I.; Choi, H.-H.; Kim, W.; Kim, S. H.; Park, S. K.;
Lee, G. M.; Koh, G. Y. Proc. Natl. Acad. Sci. U.S.A.
2004, 101, 5547.
9. (a) Scharpfenecker, M.; Fielder, U.; Reiss, Y.; Augustin,
H. G. J. Cell Sci. 2005, 118, 771; (b) Lobov, I. B.; Brooks,
P. C.; Lang, R. A. Proc. Natl. Acad. Sci. U.S.A. 2002, 99,
11205.
10. Synthetic procedures and conditions for HTRF and
cellular assays are described in: 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.; Whitting-
ton, D. A.; Zhao, H. J. Med. Chem. 2007, 50, 611.
11. The PDB deposition code for 1b is 2OO8 and for 2 is
2OSC.
12. Lin, T. H.; Metzger, A.; Diller, D. J.; Desai, M.;
Henderson, I.; Ahmed, G.; Kimble, E. F.; Quadros, E.;
Webb, M. L. J. Pharmacol. Exp. Ther. 2006, 318, 495.
13. Tan, H.; Semin, D.; Wacker, M.; Cheetham, J. J. Assoc.
Lab. Automat. 2005, 364.
Based on Tie-2 cellular potency (IC50 = 33 nM), excel-
lent selectivity over KDR (63-fold enzyme selectivity;
130-fold cellular selectivity), and promising pharmaco-
kinetics, we chose 5 for further profiling. The selectivity
profile was explored across a range of tyrosine and ser-
ine/threonine kinases (Table 4). The phosphorylation of
serine/threonine kinase p38a (IC50 = 47 nM) and non-
receptor tyrosine kinase Lck (IC50 = 146 nM) was inhib-
ited most strongly. The p38a activity was examined in a
THP-1/TNFa cellular assay,12 resulting in an
IC50 = 194 nM. All of the other tested kinases in the
selectivity panel returned potency values of greater than
one micromolar.
The rat pharmacokinetic parameters for 5 are summa-
rized in Table 5. Data collected from intravenous dosing
were a moderate clearance and volume of distribution,
leading to
a
half-life of 1.7 h (mean-residence
time = 1.5 h). Unfortunately, when 5 was dosed po at
10 mg/kg an oral bioavailability of 4% was achieved.
One possible factor contributing to the low %F could
be the inadequate solubility at pH values of 2 and 7.1
(solubility: 0.01 N HCl = 0.021 mg/mL; phosphate-buf-
fered solution = 0.003 mg/mL).13
In conclusion, a number of highly potent inhibitors of
Tie-2 cellular autophosphorylation were discovered in
this reversed-amide series. Impressive levels of selectivity
over KDR and a panel of other kinases were achieved
within the triazine series. Compound 5 possessed
the best combination of potency, selectivity, and rat