Pyrazole diaminopyrimidines as dual inhibitors of KDR and Aurora B kinases

Article abstract of DOI:10.1016/j.bmcl.2012.05.067

In an effort to identify kinase inhibitors with dual KDR/Aurora B activity and improved aqueous solubility compared to the Abbott dual inhibitor ABT-348, a series of novel pyrazole pyrimidines structurally related to kinase inhibitor AS703569 were prepared. SAR work provided analogs with significant cellular activity, measureable aqueous solubility and moderate antitumor activity in a mouse tumor model after weekly ip dosing. Unfortunately these compounds were pan-kinase inhibitors that suffered from narrow therapeutic indices which prohibited their use as antitumor agents.

Full text of DOI:10.1016/j.bmcl.2012.05.067

Bioorganic & Medicinal Chemistry Letters 22 (2012) 4750–4755
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Pyrazole diaminopyrimidines as dual inhibitors of KDR and Aurora B kinases
⇑
Michael L. Curtin , H. Robin Heyman, Robin R. Frey, Patrick A. Marcotte, Keith B. Glaser,
James R. Jankowski, Terrance J. Magoc, Daniel H. Albert, Amanda M. Olson, David R. Reuter,
Jennifer J. Bouska, Debra A. Montgomery, Joann P. Palma, Cherrie K. Donawho, Kent D. Stewart,
Chris Tse, Michael R. Michaelides
Global Pharmaceutical Research and Development, Abbott Laboratories, 100 Abbott Park Road, Abbott Park, IL 60064-6100, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
In an effort to identify kinase inhibitors with dual KDR/Aurora B activity and improved aqueous solubility
compared to the Abbott dual inhibitor ABT-348, a series of novel pyrazole pyrimidines structurally
related to kinase inhibitor AS703569 were prepared. SAR work provided analogs with significant cellular
activity, measureable aqueous solubility and moderate antitumor activity in a mouse tumor model after
weekly ip dosing. Unfortunately these compounds were pan-kinase inhibitors that suffered from narrow
therapeutic indices which prohibited their use as antitumor agents.
Received 2 May 2012
Revised 16 May 2012
Accepted 17 May 2012
Available online 26 May 2012
Keywords:
Ó 2012 Elsevier Ltd. All rights reserved.
Aurora B kinase inhibitors
KDR kinase inhibitors
Diaminopyrimidines
Aminopyrazoles
Antitumor activity
The Aurora kinases are a family of serine/threonine kinases that
mediate multiple events in cell division.1 Humans have three Aur-
ora kinases, A, B and C, that are differentially localized and mediate
distinct functions during mitosis. Because the Aurora kinases play a
key role in mitosis and are overexpressed in multiple human tumor
types, there has been considerable interest in developing Aurora
kinase inhibitors as antitumor agents. A number of small-molecule
Aurora kinase inhibitors have been reported and there are several
compounds currently in Phase I/II clinical trials for cancer.2 Precli-
nically, it has been shown that pan-Aurora inhibitors reproduce the
phenotype of Aurora B selective agents and it has been suggested
that the clinical behavior of pan-Aurora inhibitors will resemble
compounds selectively targeting Aurora B rather than Aurora A.3
The VEGF receptor family of RTKs, most notably VEGFR2 or KDR,
mediates the biological function of vascular endothelial growth
factor (VEGF) which is a regulator of vascular permeability and
an inducer of endothelial cell proliferation, migration and sur-
vival.4 Accordingly, interruption of the KDR mediated signaling
cascade can provide therapeutic benefit in human cancers as re-
cently demonstrated by the FDA approval of the anti-VEGF anti-
body Avastin™ and four small molecule KDR kinase inhibitors.5
The notion of inhibiting multiple kinases involved in tumor pro-
gression with a single small molecule6 is well established and the
compelling antitumor effects elicited by multi-targeted KDR kinase
inhibitors such as linifanib7 led us to screen Abbott proprietary
KDR kinase inhibitors for activity against a broad panel of kinases.
This screen revealed a series of dual KDR/Aurora B thienopyridine
pyrazole inhibitors8 that possessed efficacy in murine xenograft
models and provided the clinical candidate ABT-348 (1 in Fig. 1).
To complement the development of this poorly water-soluble
inhibitor which is being developed as an oral agent, we wanted
to identify novel and potent dual KDR/Aurora B inhibitors with
aqueous solubility that would allow their use as intravenous
agents. One approach to accomplish this goal was to use privileged
structural moieties from Abbott proprietary inhibitors to modify
known KDR, Aurora B or KDR/Aurora B inhibitors that possessed
better solubility than the thienopyridine ureas.
One such inhibitor was the Rigel/Serono clinical compound
AS703569 (2)9 which in-house testing indicated was a moderate
to excellent inhibitor of Aurora B, both enzymatically (IC50
=
54 nM)10 and cellularly (IC50 = 1 nM),12 and KDR, enzymatically
(IC50 = 95 nM)10 and cellularly (IC50 = 19 nM);14 this compound
also had better aqueous solubility (83
1 (<1 M). We envisioned using the pyrazole moiety of our thieno-
l
M)15 than thienopyridine
l
pyridine series as a phenyl replacement for 2 which would give no-
vel pyrimidine pyrazoles such as compound 3; modeling of this
compound with Aurora B indicated that the pyrazole ring would
make a unique interaction with the ‘extended hinge’ region similar
to that in the thienopyridine series and perhaps provide inhibitors
with enhanced potency or efficacy versus 2. Herein is described the
effort to prepare the compounds of general structure 4 in order to
⇑
Corresponding author.
E-mail address: mike.curtin@abbott.com (M.L. Curtin).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.05.067

M. L. Curtin et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4750–4755
4751
F
F
F
H
N
X
5
O
H
N
O
R'
NH
O
N
NH
NH2
N
NH
NH2
N
NH
O
N
NH2
S
N
N
N
HN
N
HN
HN
N
2
3
1
4
N
N
N
(AS703569)
(ABT-348)
N
R
N
HO
N
Figure 1.
determine the effect of modification/replacement of the norborn-
ene carboxamide, pyrimidine 5-substituent and pyrazole N-substi-
tuent on potency, solubility and tumor efficacy.16
which was ring-opened with ammonia to give, after amine depro-
tection with TFA, norbornene carboxamide 7 in high yield. Addi-
tion of 7 to 2,4-dichloro-5-fluoropyrimidine in the presence of
sodium bicarbonate regioselectively (>5:1) gave chloropyrimidine
8 which was treated with 1-methyl-4-aminopyrazole and hydro-
chloric acid at 80 °C to afford pyrazole pyrimidine 3. Diversity
could be introduced into this structure by replacement of ammonia
with primary amines in the ring-opening step, replacement of the
5-fluoropyrimidine with other 5-substituted pyrimidines and use
of 4-aminopyrazoles with a variety of N-substituents. Selected
analogs were resynthesized as the single (+)-enantiomer (absolute
configuration as drawn in Scheme 1) using (1S,2S,3R,4R)-carbox-
The preparation of racemic pyrazole pyrimidine 317 is shown in
Scheme 1 and exemplifies the preparation of the majority of the
compounds in this work. Boc-protection of commercially available,
racemic, norbornadiene-based lactam 5 provided carbamate 6
O
TFA
NH2
O
a
b
NH
N
O
amide 718
.
O
H2N
N
O
7
5
6
The typical preparation of the 5-(trifluoromethyl)pyrimidine
analogs is shown in Scheme 2. Because the addition of amine 7
to 2,4-dichloro-5-(trifluoromethyl)pyrimidine was not regioselec-
tive, a route was used to prepare 5-(trifluoromethyl)pyrimidines
that entailed a ZnCl2-mediated, highly regioselective addition of
1-methyl-4-aminopyrazole to the 2-position of pyrimidine 9 to
provide intermediate 10 in modest yield. Addition of amine 7 to
chloropyrimidine 10 in the presence of Hunig’s base at 90 °C gave
the desired 5-(trifluoromethyl)pyrimidine 11 in high yield. This
F
F
N
N
c
d
N
N
H
N
Cl
N
H
N
N
H
H2N
O
H2N
O
8
3
Scheme 1. Reagents and conditions: (a) Boc-anhydride, DMAP, THF, rt, >95%; (b)
NH4OH, EtOAc, rt then TFA, 98%; (c) 2,4-dichloro-5-fluoropyrimidine, NaHCO3,
MeOH, 65%; (d) 1-methyl-4-aminopyrazole, HCl, i-PrOH, 80 °C, 54%.
F3C
N
N
N
F3C
Cl
F3C
Cl
b
a
N
N
N
N
N
H
N
N
H
N
Cl
N
N
H
H2N
O
11
9
10
Scheme 2. Reagents and conditions: (a) 1-methyl-4-aminopyrazole, ZnCl2, 0 °C then (i-Pr)2NEt, 50 °C, DCE/t-BuOH, 48%; (b) carboxamide 7, (i-Pr)2NEt, EtOH, 90 °C, 96%.
Table 1
SAR of pyrimidine 5-substitution
R
N
N
N
N
H
N
N
H
H2N
O
12
Compound
R
IC50 (nM)
EC15 (nM) polyploidy
KDR10
KDRcell
Aur B10
14
1(ABT-348)
(+)-2 (AS 703569)
3
2
95
3
7
54
29
294
252
145
197
15
12
1
19
NT
NT
NT
NT
NT
>316
35
9
F
H
329
>12,000
467
922
1540
50
12
13
14
111
93
11
NT
NT
51
3
11
12
13
14
15
16
17
18
CH3
OCH3
CONH2
NO2
CF3
Br
28
42
29
<3
<3
Cl
20
NT, not tested.

4752
M. L. Curtin et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4750–4755
route proved convenient for the initial work on norbornene car-
boxamide replacement.
Aurora B activity, both enzymatically and cellularly. The overall
SAR suggested that a small, electron-withdrawing substituent at
this position was favored and that KDR potency was much more
sensitive to these changes than Aurora B potency.
The prepared analogs were assessed for enzymatic activity
against a panel of kinases including KDR, Aurora B, Flt-1, Flt-3, cKit,
CSF1R and FGFR; compounds with sufficient KDR and Aurora B
enzymatic activity10 (<50 nM) were examined for cellular activity
in a KDR autophosphorylation assay14 and a polyploidy induction
assay12 indicative of Aurora B activity. Selected inhibitors were
then assessed in a xenograft mouse tumor model using the
DoHH-2 (human lymphoma) cancer cell line.
As shown in Table 1, while pyrazole pyrimidine 3 was only a
modest inhibitor of KDR and much less active than thienopyridine
pyrazole 1, it was comparable in potency to diaminopyrimidine 2
in both the KDR and Aurora B enzymatic assays. Variation of the
pyrimidine 5-substituent revealed that substitution at this position
was necessary for acceptable KDR inhibition (11) but that several
5-substituents including methyl (12), methoxyl (13) and carbox-
amide (14) provided only modest to poor KDR activity. A nitro
group (15) gave useful KDR enzymatic potency but a lack of KDR
cellular activity while trifluoromethyl (16), bromide (17) and chlo-
ride (18) substituents at this position provided optimal KDR and
Because of the favorable 5-(trifluoromethyl)pyrimidine potency
(16) and the synthetic pathway to this analog that allowed for
bicyclic carboxamide replacement in the last step, initial SAR work
on the bicyclic moiety was conducted through modification of
inhibitor 16 and the results are shown in Table 2. As was observed
with pyrimidine 5-substitution, modification of the bicyclic car-
boxamide had a much larger effect on KDR potency than Aurora
potency. For example, saturation of the double bond (19), removal
of the carboxamide (20), replacement of the norbornene with a
pinene-based bicycle (21), addition of a carboxamide substituent
(22–24) or replacement of the bicyclic carboxamide with o-benz-
amide (25) gave compounds that were roughly equipotent with
16 against Aurora B but less potent against KDR (5- to 70-fold).
The absence of the carboxamide moiety (20) was detrimental to
the potency against both enzymes.
Additional bicyclic modification work was completed on the 5-
chloropyrimidine analog 18 and the results are shown in Table 3.
Table 2
Bicyclic carboxamide modification and replacement
F3C
R
N
N
N
N
H
N
N
H
12
Compound
R
IC50 (nM)
EC15 (nM) polyploidy
KDR10
12
KDRcell
Aur B10
14
16
35
28
3
H2N
H2N
O
O
19
20
74
61
66
61
10
827
NT
>300
21
264
NT
43
39
H2N
O
O
22
23
24
68
>316
NT
45
51
<1
<1
N
H
292
N
H
O
266
431
NT
NT
99
16
<1
11
HO
N
O
H
25
H2N
O
NT, not tested.

M. L. Curtin et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4750–4755
4753
Table 3
Bicyclic carboxamide modification and replacement with 5-chloropyrimidine
Cl
N
N
N
R
N
H
N
N
H
12
Compound
R
IC50 (nM)
EC15 (nM) polyploidy
14
KDR10
14
KDRcell
Aur B10
18
26
27
28
20
29
<3
1
H2N
H2N
H2N
H2N
H2N
O
O
O
O
O
19
19
84
8
31
>316
NT
15
132
NT
O
693
169
118
135
29
30
NT
246
153
NT
NT
32
44
62
93
H2N
H2N
O
O
31
NT, not tested.
Unlike the modifications on the 5-(trifluoromethyl)pyrimidine in
Table 2, several of the bicycle modifications in Table 3 affected
both KDR and Aurora B inhibition. While saturation of the double
bond gave a compound (26) that was roughly equipotent with
18, use of an endo stereoisomer (27), oxo bicycle (28) or monocy-
cles (29–31) led to significant decreases in KDR activity while 28
and 29 had significant loss of Aurora activity as well. The conclu-
sion from this portion of the SAR work (Tables 2 and 3) was that
the norbornene carboxamide of inhibitor 2 was optimal for the
pyrazole pyrimidine series and that the 5-chloro pyrimidine sub-
stituent gave slightly more robust KDR potency than the 5-(trifluo-
romethyl) substituent.
The effect of modifying the pyrazole N-substitution on enzy-
matic potency and other properties is shown in Table 4. Gratify-
ingly, many analogs from this subseries had aqueous solubility
that was comparable or equal to that of competitor compound 2.
Otherwise, it was observed that a variety of N-substituents pro-
vided potent inhibitors, both enzymatically and cellularly, while
having significant effects on other properties. For example, while
N-methyl inhibitor (+)-18 was potent, moderately stable in the
presence of mouse liver microsomes, reasonably soluble in water
and bioavailable after ip dosing in mice, other N-alkyl substituents
such N-ethyl [(+)-32] or N-benzyl [(+)-33] provided compounds
with lower microsomal stability and/or diminished aqueous solu-
bility. In general, N-substituents bearing alcohol or ether func-
tionalites gave desireable analogs including alcohol (+)-34 while
the carboxylate moiety in 35 did not; this is in contrast to the acet-
amide analogs (+)-36 and (+)-37 which were stable, soluble and
bioavailable. To improve aqueous solubility further, a number of
inhibitors with pyrazole N-substituents possessing secondary and
tertiary amines were prepared and while many of these com-
pounds were potent (38), stable [(+)-39] and bioavailable [(+)-40
and (+)-41], no obvious improvement in aqueous solubility was
observed.
In order to better understand the SAR associated with this series
of compounds, molecular modeling of (+)-38 bound to the active
site of Aurora B was completed (Fig. 2).19 The compound can bind
to the active conformation of the enzyme with canonical hydrogen
bonds to Ala 157 of the hinge. The chloro group projects toward a
small hydrophobic cavity near the gatekeeper, which is Leu 154 in
Aurora B and Val 914 in KDR kinase. This binding mode is in accord
with the large loss in potency of polar analogs 13 and 14 that
would project methoxy and carboxamide groups, respectively, into
this cavity. Importantly, the carboxamide carbonyl of (+)-38 can
accept a hydrogen bond from Lys 106, a residued that is conserved
in almost all kinases. There is volume available near the amino of
the carboxamide to accommodate small substituents, in accord
with potencies within threefold of 16 for analogs 22–24 with N-
methyl, -cyclopropyl, and hydroxyethyl, respectively.
Selected analogs were assessed in a xenograft mouse tumor
model using the DoHH-2 (non-Hodgkin’s B-cell) cancer line, dosing
intraperitoneally once a week for three weeks. Figure 3 shows the

4754
M. L. Curtin et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4750–4755
Table 4
SAR of pyrazole N-substitution
Cl
N
N
N
R
N
H
N
N
H
H2N
O
Microsomal stability (%)a
Solubility15
(
lM)
AUC, ip, 10 mpk (l
M hr)b
12
Compound
R
IC50 (nM)
EC15 (nM) polyploidy
KDR10
KDRcell
Aur B10
14
(+)-18
(+)-32
CH3
CH2CH3
4
3
10
8
15
8
<1
<1
44
13
78
70
11.1
2.54
(+)-33
(+)-34
7
6
42
48
12
19
<1
<1
<1
45
4
NT
5.0
70
OH
OH
O
O
O
35
1
>316
167
32
25
28
19
>95
68
71
68
NT
(+)-36
18
15.0
NH2
(+)-37
6
89
26
11
88
72
6.68
N
H
N
38
9
2
78
55
27
12
<3
1
4
85
68
NT
2.0
O
(+)-39
76
NH
(+)-40
(+)-41
2
44
49
12
15
1
48
48
70
86
15.7
12.7
N
H
HN
15
11
NT, not tested.
a
Remaining after 30 min in the presence of mouse liver microsomes.
PK studies in CD1(ICR) males.
b
tumor growth inhibition of several pyrazole pyrimidines along
with that of thienopyridine urea 1 which was used as an internal
control. It was found that tumor efficacy varied significantly within
the series despite similar structures, potencies and exposures.
More importantly, all analogs tested from this series provided only
modest to poor efficacy and suffered from narrow therapeutic indi-
ces (1–2 times). For example, while compounds 17 and (+)-18 were
efficacious at 10 mg/kg, they elicited toxicity (i.e., weight loss,
death) at doses of 30 mg/kg and 15 mg/kg, respectively.
One potential explanation for the narrow therapeutic indices
emerged after kinome activity profiling of selected pyrazole pyrim-
idines revealed that they are extraordinarily promiscuous kinase
inhibitors. A partial inhibitory profile of compound 17 is shown
in Table 5 and indicates that this compound is a potent inhibitor
of many RTK and serine/threonine kinases. In fact, a broader,
HTRF-based screen revealed that this compound has a Ki <5 nM
for 38 kinases, a Ki of 5–10 nM for 11 kinases and a Ki of 10–
20 nM for 16 kinases. This is in contrast to the Rigel/Serono inhib-
itor 2 which has better kinome selectivity (Ki <5 nM, 12 kinases; Ki
5–10 nM, 11 kinases) and less toxicity (MTD >100 mg/kg, ip). It is
not clear why the pyrazole pyrimidine series (e.g., 17) is less ki-
nome-selective than compound 2; attempts at modulating kinome
selectivity through manipulation of the pyrazole N-substituent,
including the use of basic moieties such as piperidine or piperizine,
failed to increase selectivity.
Figure 2. Model of pyrazole pyrimidine (+)-38 bound to active site of Aurora B
(active conformation, homology model) with hinge hydrogen bonds to Ala157 C@O
and N–H. The pyrimidine ring makes Van der Waals contact with the side chain of
Leu 207. The chloro group projects toward a small hydrophobic volume near the
gatekeeper residue Leu 154. The carboxamide carbonyl makes a weak intramolec-
ular hydrogen bond with the 4-amino group on the pyrimidine and accepts a
hydrogen bond from Lys 106. The pyrazole fits into a narrow solvent accessible
region near Gly 160 analogous to that reported for compound 1 in Ref. 8 (not shown
here) and the morpholino group is fully solvated. All intermolecular hydrogen
bonds were 3.0 0.3 Å.
In summary, a series of pyrazole pyrimidines with potent activ-
ity against both KDR and Aurora B has been identified. SAR work
provided analogs with significant cellular activity, measureable

M. L. Curtin et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4750–4755
4755
5. For a review of the most recently approved KDR inhibitor, see: Commander, H.;
Whiteside, G.; Perry, C. Drugs 2011, 71, 1355.
6. Fojo, T. Oncologist 2008, 13, 277.
2500
2000
1500
1000
500
Vehicle
Cmpd 1 (50 mg/kg, po)
Cmpd 17 (10 mg/kg, ip)
Cmpd (+)-18 (10 mg/kg, ip)
Cmpd (+)-32 (10 mg/kg, ip)
Cmpd 37 (10 mg/kg, ip)
Cmpd 39 (10 mg/kg, ip)
7. Dai, Y.; Hartandi, K.; Ji, Z.; Ahmed, A. A.; Albert, D. H.; Bauch, J. L.; Bouska, J. J.;
Bousquet, P. F.; Cunha, G. A.; Glaser, K. B.; Harris, C. M.; Hickman, D.; Guo, J.; Li,
J.; Marcotte, P. A.; Marsh, K. C.; Moskey, M. D.; Martin, R. L.; Olson, A. M.;
Osterling, D. J.; Pease, L. P.; Soni, N. B.; Stewart, K. D.; Stoll, V. S.; Tapang, P.;
Reuter, D. R.; Davidsen, S. K.; Michaelides, M. R. J. Med. Chem. 2007, 50, 1584.
8. Curtin, M. L.; Heyman, H. R.; Frey, R. R.; Soni, N. B.; Marcotte, P. A.; Pease, L. J.;
Glaser, K. B.; Magoc, T. J.; Tapang, P.; Albert, D. H.; Osterling, D. J.; Olson, A. M.;
Bouska, J. J.; Guan, Z.; Martin, R. L.; Preusser, L. C.; Polakowsi, J. S.; Stewart, K.
D.; Tse, C.; Davidsen, S. K.; Michaelides, M. R. Bioorg. Med. Chem. Lett. 2012, 22,
3208.
9. (a) McLaughlin, J.; Markovtsov, V.; Li, H.; Wong, S.; Gelman, M.; Zhu, Y.; Franci,
C.; Lang, D. W.; Pali, E.; Lasaga, J.; Low, C.; Zhao, F.; Chang, B.; Gururaja, T. L.; Xu,
W.; Baluom, M.; Sweeney, D.; Carroll, D.; Sran, A.; Thota, S.; Parmer, M.;
Romane, A.; Clemens, G.; Grossbard, E.; Qu, K.; Jenkins, Y.; Kinoshita, T.; Taylor,
V.; Holland, S. J.; Argade, A.; Singh, R.; Pine, P.; Payan, D. G.; Hitoshi, Y. J. Cancer
Res. Clin. Oncol. 2010, 136, 99; (b) To our knowledge, SAR data for AS703569 has
not been disclosed.
0
/ /
20
25
30
35
Days
Dosing schedule
10. KDR and Aurora
B enzymatic IC50 values were determined using a
Figure 3. DoHH-2 mouse tumor growth model. All compounds dosed q7dx3
(n = 10, SCID).
homogeneous time-resolved fluorescence (HTRF) kinase assay with an ATP
concentration of 1 mM as described in Ref. 11.
11. Dai, Y.; Guo, Y.; Frey, R. R.; Ji, Z.; Curtin, M. L.; Ahmed, A. A.; Albert, D. H.;
Arnold, L.; Barlozzari, T.; Bauch, J. L.; Bouska, J. J.; Bousquet, P. F.; Cunha, G. A.;
Glaser, K. B.; Guo, J.; Li, J.; Marcotte, P. A.; Marsh, K. C.; Moskey, M. D.; Pease, L.
J.; Stewart, K. D.; Stoll, V. S.; Tapang, P.; Wishart, N.; Davidsen, S. K.;
Michaelides, M. R. J. Med. Chem. 2005, 48, 6066.
12. Aurora B cellular activity was assessed by measuring induction of polyploidy in
the NCI-H1229 (NSCLC) cell line as described in Ref. 13 Initial dose–response
data obtained with this assay indicated that maximal polyploid response to
inhibitors typically did not exceed 50% and was associated with apoptosis and
loss in cell number. To avoid this complexity, a threshold potency, defined as
the concentration necessary to produce DNA content >4 N in 15% of the cell
Table 5
Partial kinase inhibition profile of 17
Kinase
IC50 (nM)a
Kinase
IC50 (nM)a
KDR
AurA
Flt3
GSK3b
PDGFR
6
3
1
34
1
Lck
FGFR2
Syk
PKC
Wee1
7
2
44
82
2800
population (EC15), was chosen rather than the more common EC50
.
e
13. McClellan, W. J.; Dai, Y.; Abad-Zapatero, C.; Albert, D. H.; Bouska, J. J.; Glaser, K.
B.; Magoc, T. J.; Marcotte, P. A.; Osterling, D. J.; Stewart, K. D.; Davidsen, S. K.;
Michaelides, M. R. Bioorg. Med. Chem. Lett. 2011, 21, 5620.
14. The KDR cellular IC50 values represent inhibition of KDR phosphorylation in
NIH3T3 cells stably transfected with full length human KDR. The procedure
was followed as described in Ref. 11.
a
a
TR-FRET assay.
15. Compounds were dissolved in DMSO (10 mM), added to pH 7.2 buffer (1:100
dilution) and equilibrated for 24 h with vigorous shaking. The samples were
centrifuged and compound concentration in the supernatant was determined
using chemiluminescent nitrogen detection (CLND).
16. After this work was completed, the FDA approved the first dianilinopyrimidine
RTK inhibitor, pazopanib, for the treatment of renal cell carcinoma. For an
account of the discovery of this compound, see: Harris, P. A.; Boloor, A.;
Cheung, M.; Kumar, R.; Crosby, R. M.; Davis-Ward, R. G.; Epperly, A. H.; Hinkle,
K. W.; Hunter, R. N.; Johnson, J. H.; Knick, V. B.; Laudeman, C. P.; Luttrell, D. K.;
Mook, R. A.; Nolte, R. T.; Rudolph, S. K.; Szewczyk, J. R.; Truesdale, A. T.; Veal, J.
M.; Wang, L.; Stafford, J. A. J. Med. Chem. 2008, 51, 4632.
17. All intermediates and final compounds are racemates unless otherwise noted.
18. (1S,2S,3R,4R)-Carboxamide 7 was prepared using enzymatic chiral resolution
of lactam 6 as described in WO2006/055561.
19. A homology model of human Aurora B kinase was constructed using template
PDB code 1ATP for protein kinase A. The ligand orientation of (+)-38 was
guided by using template 2NP8 for Aurora A Kinase according to the procedure
described in Ref. 11.
aqueous solubility and moderate antitumor activity in a mouse tu-
mor model after weekly ip dosing. Unfortunately these compounds
were pan-kinase inhibitors that suffered from narrow therapeutic
indices which prohibited their use as antitumor agents.
References and notes
1. Marumoto, T.; Zhang, D.; Saya, H. Nat. Rev. Cancer 2005, 5, 42.
2. Matthews, N.; Visintin, C.; Hartzoulakis, B.; Jarvis, A.; Selwood, D. L. Expert Rev.
Anticancer Ther. 2006, 6, 109.
3. (a) Yang, H.; Burke, T.; Dempsey, J.; Diaz, B.; Collins, E.; Toth, J.; Beckmann, R.;
Ye, X. FEBS Lett. 2005, 579, 3385; (b) Girdler, F.; Gascoigne, K. E.; Eyers, P. A.;
Hartmuth, S.; Crafter, C.; Foote, K. M.; Keen, N. J.; Taylor, S. S. J. Cell Sci. 2006,
119, 3664.
4. Ferrara, N.; Gerber, H. P.; LeCouter, J. Nat. Med. 2003, 9, 669.