Identification of substituted 3-[(4,5,6,7-tetrahydro-1H- indol-2-yl)methylene]-1,3-dihydroindol-2-ones as growth factor receptor inhibitors for VEGF-R2 (Flk-1/KDR), FGF-R1, and PDGF-Rβ tyrosine kinases

Article abstract of DOI:10.1021/jm9906116

A series of new 3-substituted indolin-2-ones containing a tetrahydroindole moiety was developed as specific inhibitors of receptor tyrosine kinases associated with VEGF-R, FGF-R, and PDGF-R growth factor receptors. These compounds were evaluated for their

Full text of DOI:10.1021/jm9906116

J . Med. Chem. 2000, 43, 2655-2663
2655
Id en tifica tion of Su bstitu ted 3-[(4,5,6,7-Tetr a h yd r o-1H-in d ol-2-yl)m eth ylen e]-
1,3-d ih yd r oin d ol-2-on es a s Gr ow th F a ctor Recep tor In h ibitor s for VEGF -R2
(F lk -1/KDR), F GF -R1, a n d P DGF -Râ Tyr osin e Kin a ses
Li Sun,*,§ Ngoc Tran,§ Congxing Liang,§ Steve Hubbard,‡ Flora Tang,§ Kenneth Lipson,§ Randall Schreck,§
Yong Zhou,§ Gerald McMahon,§ and Cho Tang§,*
SUGEN, Inc., 230 East Grand Avenue, South San Francisco, California 94080-4811, and Skirball Institute of Biomolecular
Medicine and Department of Pharmacology, New York University Medical Center, New York, New York 10016
Received December 15, 1999
A series of new 3-substituted indolin-2-ones containing a tetrahydroindole moiety was developed
as specific inhibitors of receptor tyrosine kinases associated with VEGF-R, FGF-R, and PDGF-R
growth factor receptors. These compounds were evaluated for their inhibitory properties toward
VEGF-R2 (Flk-1/KDR), FGF-R1, PDGF-Râ, p60c-Src, and EGF-R tyrosine kinases and their
ability to inhibit growth factor-dependent cell proliferation. Structure-activity relationships
of this new pharmacophore have been determined at the level of kinase inhibition. Compounds
containing a propionic acid moiety at the C-3′ position of the tetrahydroindole ring represented
the most potent indolin-2-ones to inactivate the VEGF, FGF, and PDGF receptor kinases. The
inhibitory activities of 9d against VEGF-R2 (Flk-1), 9h against FGF-R1, and 9b against PDGF-
Râ were 4, 80, and 4 nM, respectively. However, all of these compounds were inactive when
tested against the EGF-R tyrosine kinase. Compounds 9a and 9b represented the most potent
inhibitors of these classes to inhibit both biochemical kinase and growth factor-dependent cell
proliferation for these three targets. In addition, compound 9a was cocrystallized with the
catalytic domain of FGF-R1 providing evidence to explain the structure-activity relationship
results. This study has provided evidence to support the potential of these new tyrosine kinase
inhibitors for the treatment of angiogenesis and other growth factor-related diseases including
human cancers.
In tr od u ction
kinases (RTKs) such as FGF (fibroblast growth factor)
and PDGF (platelet-derived growth factor) may play a
direct or indirect role in the tumor angiogenic process.
For instance, FGF was found to be a mitogen of different
cell types including vascular endothelial cells and
fibroblasts and, in some cases, may induce the expres-
sion of VEGF.4 Studies using in vitro angiogenesis
assays have indicated that both VEGF and FGF may
cooperate during new blood vessel development.5 In
contrast, PDGF has been shown to stimulate the growth
of pericytes and fibroblast-like cells which have been
shown to be required for the formation of the capillaries
during angiogenesis. Additionally, PDGF has been
shown to play a role in angiogenesis leading to upregu-
lation of VEGF expression.4 Furthermore, both FGF and
PDGF may be involved in a hypothesized “angiogenic
switch” following VEGF signaling during early stages
of tumor growth.6
Recently, diverse pharmacophores of small molecule
inhibitors have been emerged as antiangiogenic agents
targeting at different RTKs including the VEGF and
FGF receptors.7 Our own previous studies have dem-
onstrated the potential of 3-substituted indolin-2-ones
as antiangiogenic agents.8,9 As shown in Chart 1, two
compounds (SU5416, 1, and SU6668) from the indolin-
2-one pharmacophore have advanced into clinical tri-
als.10,11 The goal of this present study was to further
diversify the indolin-2-one pharmacophore in order to
develop catalytic kinase inhibitors that could block
It is well-accepted that de novo angiogenesis is critical
to the growth of solid tumors.1 In this regard, newly
formed capillaries around the tumor mass have been
suggested to (1) provide nutrients and growth factors
for rapid tumor growth, (2) remove waste generated by
tumor metabolism, and (3) transport tumor cells to
locations far from the primary site to cause tumor
metastasis. Because of these implications, it is not
surprising that novel treatments to inhibit tumor an-
giogenesis have become a means to develop new cancer
therapies. In this regard, large numbers of cellular
components, including growth factors and cytokines,
have been implicated in the tumor angiogenic process.2
One key growth factor is VEGF (vascular endothelial
growth factor) and its cognate receptor tyrosine kinases
(VEGF-R2 (Flk-1) and VEGF-R1 (Flt-1)). VEGF and its
receptors have been shown to play direct role in tumor
angiogenesis by promoting endothelial cell proliferation
and vascular permeability, key processes in angio-
genesis. Importantly, VEGF-R2 (Flk-1) has been shown
to be expressed selectively in endothelial cells.3
In addition, other studies have demonstrated that
other growth factors and their cognate receptor tyrosine
* To whom correspondence should be addressed. Phone: (650)553-
8480. Fax: (650)553-8348. E-mail: connie-sun@sugen.com.
§ SUGEN, Inc.
‡ New York University Medical Center.
10.1021/jm9906116 CCC: $19.00 © 2000 American Chemical Society
Published on Web 06/23/2000

2656 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 14
Ch a r t 1. Indolin-2-ones in Clinical Trials
Sun et al.
then condensed with diethyl aminomalonate hydrochlo-
ride to give 3-(2-ethoxycarbonylethyl)-4,5,6,7-tetrahy-
dro-1H-indole-2-carboxylic acid ethyl ester (compound
6 in Scheme 2). Hydrolysis of 6 followed by decarbox-
ylation resulted in compound 7 that upon formylation
gave 8.
All of the compounds prepared in this report exist in
the Z (cis) conformation due to the intramolecular
hydrogen bonding between the NH-1′ of the tetrahy-
droindole ring and the carbonyl group at the C-2
position of the indolin-2-one core as we described
previously.8
Sch em e 1. Synthesis of Substituted 3-[(4,5,6,7-
Tetrahydro-1H-indol-2-yl)methylene]-1,3-dihydroindol-
2-ones 4a -ia
Resu lts a n d Discu ssion
Compounds were evaluated for their inhibitory activ-
ity toward tyrosine phosphorylation for the VEGF-R2
(Flk-1), FGF-R1, PDGF-Râ, p60c-Src, and EGF-R ty-
rosine kinases. The p60c-Src kinase has been shown to
be activated by a wide variety of RTKs following growth
factor addition.13 It was of importance to include an
analysis of kinase inhibition for p60c-Src in order to fully
interpret results using growth factor-stimulated prolif-
eration in cells since inhibitors of p60c-Src may score as
inhibitors of all proliferation assays where this kinase
may be activated. These data would aid in our under-
standing of the effects of the compound on RTKs or
downstream events. Biochemical tyrosine autophos-
phorylation assays were used to assess the inhibitory
potency of all compounds toward PDGF-Râ and EGF-R
tyrosine kinases, respectively, whereas peptide-based
transphosphorylation assays were performed for VEGF-
R2 (Flk-1), FGF-R1, and p60c-Src kinase activities. In
addition, these compounds were also evaluated in
growth factor-stimulated cell proliferation assays using
either human umbilical vein endothelial cells (HUVECs)
or NIH3T3 mouse fibroblast cells. IC50 values were
defined as the concentration of a compound required to
achieve 50% inhibition of maximal tyrosine autophos-
phorylation, transphosphorylation, or growth factor-
stimulated growth as measured by bromodeoxyuridine
(BrdU) incorporation when compared to vehicle-treated
controls (DMSO). The results for these compounds are
summarized in Tables 1 and 2. The structure-activity
relationships (SAR) are discussed separately for these
two classes of compounds.
a
(a) POCl3, ClCH2CH2Cl, 40 °C, 4 h; (b) substituted indolin-2-
ones, piperidine, ethanol, reflux, 2-4 h.
signal transduction pathways associated with VEGF,
FGF, and PDGF growth factor receptors. These three
receptor types belong to a large family of tyrosine
kinases denoted as “split” kinases due to the presence
of a unique insert region in the catalytic domain. Two
classes of substituted indolin-2-ones containing a tet-
rahydroindole moiety at the C-3 position of the indolin-
2-one core were designed and evaluated for their
potential to inhibit kinase activities (Tables 1 and 2).
Data related to biochemical kinase activity, cocrystal-
lography of a prototype compound in the FGF-R1
catalytic domain, and antiproliferative activities for
growth factors are shown and discussed.
Ch em istr y
The compounds in this report were prepared by aldo-
condensation of the substituted indolin-2-ones and
corresponding aldehydes in the presence of base in
ethanol as described previously.8,9 The two aldehydes
used in this study were prepared as follows.
Tetrahydroindole-2-carboxaldehyde (3) (Scheme 1)
was prepared via a Vilsmeier formylation reaction of
commercially available tetrahydroindole (Scheme 1).
Condensation of this aldehyde with substituted indolin-
2-ones gave the corresponding 4a -i (Scheme 1 and
Table 1).
A. Su bstitu ted 3-[(4,5,6,7-Tetr a h yd r oin d ol-2-yl)-
m eth ylid en yl]in d olin -2-on es. In our effort to diver-
sify substitutions on the pyrrole ring at the C-3 position
of the indolin-2-one core, 3-[(4,5,6,7-tetrahydroindol-2-
yl)methylidenyl]indolin-2-ones (4a analogues in Table
1) were synthesized and showed interesting biological
properties. By comparison to 1, 4a showed a similar
kinase profile to 1 when tested against VEGF-R2 (Flk-
1) and FGF-R1 and was found to be a slightly more
potent than 1 when tested against PDGF-Râ (IC50
)
Compounds 9a -i were synthesized by condensing
substituted indolin-2-ones with 3-(2-carboxyethyl)-
4,5,6,7-tetrahydroindole-2-carboxaldehyde (compound 8
in Scheme 2). Compound 8 was prepared starting from
the 4-cyclohex-1-enylmorpholine which was acylated
with 3-chlorocarbonylpropionic acid ethyl ester to give
4-(2-morpholin-4-ylcyclohex-1-enyl)-4-oxobutyric acid eth-
yl ester (compound 5 in Scheme 2). Compound 5 was
5.46 µM for 4a vs IC50 ) 10.1 µM for 1) (Table 1). Of
particular interest, 4a was 4-fold more potent than 1
when tested against the p60c-Src kinase. Computational
studies based in part on previous cocrystallographic
studies suggested that this might be the result of a
favorable interaction of the tetrahydroindole moiety
with hydrophobic valine present in the adenine binding
pocket of p60c-Src kinase.12 On the other hand, corre-

Growth Factor Receptor Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 14 2657
Sch em e 2. Synthesis of Substituted 3-[2-(2-Oxo-1,2-dihydroindol-3-ylidenemethyl)-4,5,6,7-
tetrahydro-1H-indol-3-yl]propionic Acids 9a -ia
a
(a) (C2H5)3N, ClCH2CH2Cl, reflux, 30 min; (b) diethyl aminomalonate hydrochloride, sodium acetate, acetic acid, reflux, 2 h; (c) NaOH,
reflux, 1.5 h; (d) HCl(aq); (e) POCl3, DMF, ClCH2CH2Cl; (f) substituted indolin-2-ones, piperidine, ethanol, reflux, 4 h.
Ta ble 1. Inhibition of Tyrosine Kinase Activities and Growth Factor-Dependent Cell Proliferation Using Compounds 4a -i
compounds
R
inhibition of tyrosine kinase activity (IC50, µM)a
inhibition of cell proliferation (IC50, µM)a
no.
VEGF-R2
FGF-R1
PDGF-Râ
p60c-Src
EGF-R
VEGF
FGF
PDGF
EGF
1
0.70
0.48
0.07
0.03
0.02
12.1
>20
>20
20
7.08
10.5
13.3
0.22
0.31
>20
>20
17.5
17.7
>20
10.1
5.46
0.92
11.5
1.03
16.4
4.74
4.92
0.08
0.12
>20
>20
>20
>20
>20
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
0.04
0.07
0.11
0.65
0.03
50.0
4.54
3.22
2.01
0.24
9.98
1.08
>50
>50
1.40
>50
>50
4a
4b
4c
4d
4e
4f
4g
4h
4i
H
5-Br
5-SO2NH2
5-COOH
6-OCH3
>50
24.6
2.38
2.80
>50
>12.5
>50
14.64
>12.5
>50
>50
6.00
>50
>50
>50
>50
34.08
>50
3.46
70.6
>50
6-phenyl
>12.5
>50
6-(3-OCH3phenyl)
6-(2-OCH3phenyl)
6-(4-OCH3phenyl)
>100
13.9
94.01
13.0
>20
>12.5
a
IC50 values were determined by at least two separate tests and are reported as mean values.
sponding amino acids in VEGF-R2 (Flk-1), FGF-R1, and
PDGF-Râ are hydrophilic lysine, lysine, and arginine
residues, respectively. These several features may help
explain, in part, the improved inhibitory activity of the
tetrahydroindole-containing indolin-2-one series toward
p60c-Src kinase as compared to VEGF-R2 (Flk-1), FGF-
R1, and PDGF-Râ. In addition, bromination at the C-5
position of the 4a core resulted in retaining the poten-
cies against both FGF-R1 and p60c-Src and resulted in
a 7- and 6-fold increase in potency (4b) toward VEGF-
R2 (Flk-1) and PDGF-Râ, respectively. This suggested
that small lipophilic substituents at the C-5 position on
the indolin-2-one core might favor inhibitory activities
for these targets. Both 4a and 4b were found to be
selective inhibitors for VEGF-R2 (Flk-1).
FGF-R1 kinases. Both of these compounds were found
to be the most potent inhibitors of VEGF-R2 (Flk-1) and
FGF-R1 in this series. This result is in accordance with
our previous studies indicating that hydrophilic substi-
tutions (carboxylic acid and aminosulfonyl) at the C-5
position of the indolin-2-one core could enhance the
inhibitory potency against both VEGF-R2 (Flk-1) and
FGF-R1 kinase.9 Interestingly, compounds 4c and 4d
were also found to be highly potent toward p60c-Src
kinase with IC50 values of 0.08 and 0.12 µM, respec-
tively, compared to the IC50 value of 4.74 µM for 4a
(Table 1). Compound 4c represented one of the most
potent inhibitors of p60c-Src kinase in this study. The
respective crystallographic structures of FGF-R112 and
p60c-Src have indicated that although amino acids in the
phosphate binding region were conserved, residues
Asn391 and Asp404 in p60c-Src protrude significantly
more into the ATP binding site than those found in
Hydrophilic substituents at the C-5 position of 4c
(aminosulfonyl) and 4d (carboxylic acid) greatly im-
proved the potencies toward both VEGF-R2 (Flk-1) and

2658 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 14
Sun et al.
Ta ble 2. Inhibition of Tyrosine Kinase Activities and Growth Factor-Dependent Cell Proliferation Using Compounds 9a -i
compounds
R
inhibition of tyrosine kinase activity (IC50, µM)a
inhibition of cell proliferation (IC50, µM)b
no.
VEGF-R2
FGF-R1
PDGF-Râ
p60c-Src
EGF-R
VEGF
FGF
PDGF
EGF
2
0.02
0.09
0.03
0.60
0.004
0.38
0.05
0.07
0.06
0.03
0.03
0.27
0.27
0.20
0.22
1.08
1.27
1.35
0.08
0.88
0.51
0.24
0.004
<0.78
0.02
19.3
0.04
0.68
0.87
10.6
1.89
1.27
0.68
0.06
1.39
0.11
0.10
0.84
0.06
>100
>100
>100
29.1
0.05
0.002
0.0003
16.5
4.49
0.02
0.09
0.06
2.70
0.18
2.80
0.01
0.10
28.4
7.66
1.38
>50
>50
3.01
0.92
1.36
1.28
1.32
>50
>50
>50
>50
>50
>50
31.6
37.2
41.5
37.8
9a
9b
9c
9d
9e
9f
9g
9h
9i
H
5-Br
5-SO2NH2
5-COOH
6-OCH3
29.0
>25
56.5
>100
>100
8.88
>100
>100
1.60
3.80
0.18
6.91
0.53
6-phenyl
6-(3-OCH3phenyl)
6-(2-OCH3phenyl)
6-(4-OCH3phenyl)
5.43
a
b
IC50 values for VEGF-R2 (Flk-1) and FGF-R1 were determined by at least two separate tests and are reported as mean values. IC50
values were determined by at least two separate tests and are reported as mean values.
Ta ble 3. Data Collection and Refinement Statistics
Data Collection
a
resolution (Å)
observations (N)
completeness (%)
98.2 (96.3)b
redundancy
1.9
Rsym (%)
signal I/σI
30.0-2.2
69057
3.8 (19.7)b
12.4
Refinementc
root-mean-square
angles (deg)
1.2
resolution deviations (Å)
reflections (N)
Rcrystd (%)
bonds (Å)
B-factorse (Å2)
25.0-2.2
33894
22.1 (26.5)f
0.006
1.1
a
b
Rsym ) 100 × ΣhklΣi |Ii(hkl) - I(hkl) |/ΣhklΣi Ii(hkl). Data are from one crystal. Value in parentheses is for the highest resolution
shell. c Atomic model includes 550 residues (two kinase molecules), two 9a molecules, and 223 water molecules (4631 atoms). Rcryst
)
d
100 × Σhkl ||Fo(hkl)| - |Fc(hkl)||/Σhkl |Fo(hkl)|, where Fo and Fc are the observed and calculated structure factors, respectively (Fo > 2σ).
e For bonded protein atoms. f Value in parentheses is the free Rcryst determined from 5% of the data.
FGF-R1. As a result, the aminosulfonyl or carboxylic
acid group in 4c and 4d , respectively, may be able to
form better hydrogen-bond contacts with the side chain
of Asn391 and the peptide backbone of Asp404 in
p60c-Src when compared to FGF-R1. However, these
hydrophilic substituents appeared to have less impact
on the inhibitory activity against PDGF-Râ. In contrast
to this latter finding, lipophilic substituents at the C-6
position of the 4a series (4e-i), in general, greatly
decreased or abolished the inhibitory potencies against
all kinases analyzed in this study. Compound 4e,
containing a methoxy substituent at the C-6 position,
was the exception to this and inhibited PDGF-Râ kinase
with an IC50 value of 3.46 µM. Additionally, this series
of compounds was found to be inactive toward EGF-R
tyrosine kinase.
The kinase inhibitory activities of these inhibitors
generally translated well into inhibition of growth
factor-dependent cell proliferation. In this regard, com-
pounds 4a -d were found to show preferential inhibition
of VEGF-dependent HUVEC proliferation when com-
pared to their inhibitory activities against FGF-, PDGF-,
and EGF-dependent cell proliferation. In contrast, poor
kinase inhibitors such as compounds 4e-i containing
the lipophilic substituents at the C-6 position of the
indolin-2-one core, in general, showed low inhibitory
activity or were inactive in the cell proliferation assays.
However, it is worth noting that 4e inhibited the
PDGF-R tyrosine kinase in both the kinase and cell-
based assays. Of particular interest, compound 4c
inhibited EGF-induced cell proliferation (IC50 ) 6.0 µM),
even though it was found to be inactive toward EGF-R
kinase. This might be, in part, due to its potent
inhibitory activity against the p60c-Src kinase that has
been shown to be required for EGF-dependent signal
transduction pathway.13 It is of interest that compound
4e was found to be 4-fold more potent than 4h against
the PDGF-R tyrosine kinase but of equal potency as 4h
toward PDGF-dependent cell proliferation. Further-
more, 4b was over 10-fold more potent than 4c in the
PDGF-R kinase assay but 8-fold less potent in the
PDGF-stimulated cell growth assay. This could be due
to other factors, including the ability of the inhibitor to
penetrate cell membranes or the chemical stability of
the compound in cells, and might also affect the outcome
of signal transduction events following receptor activa-
tion. Compound 4d (Table 1) of this series was the most
potent inhibitor of the three targets with regard to
inhibiting both kinases and growth factor-dependent
proliferation.

Growth Factor Receptor Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 14 2659
tion. The prototype compound 9a (Table 2) was found
to be 5-, 39-, and 23-fold more potent than 4a (Table 1)
toward VEGF-R2 (Flk-1), FGF-R1, and PDGF-Râ ki-
nases, respectively. Given the crystallographic informa-
tion, the increased potencies may be due to favorable
interactions of the propionic acid functionality of 9a with
Asn568 within the catalytic domains of VEGF-R2 (Flk-
1/KDR) or FGF-R1 kinase and the aspartic acid or
aspartate of PDGF-Râ.12 A comparison of compound 9a
to 2 indicated that compound 9a was 5- and 9-fold less
potent toward VEGF-R2 (Flk-1) and FGF-R1 kinases,
respectively, and showed similar inhibitory activity
toward the PDGF-Râ kinase. This might be the result
of the unfavorable interactions of the lipophilic cyclo-
hexyl ring of tetrahydroindole within the hydrophilic
environment at the opening of the ATP-binding site of
VEGF-R2 (Flk-1/KDR) and FGF-R1 kinases. Interest-
ingly, 9a was found to be 5-fold more potent than 2
when tested against p60c-Src kinase. In this regard, it
is in accordance with the results from the 4a series (4a
compared to 1 in Table 1) which suggested that the
tetrahydroindole-containing indolin-2-ones might favor
p60c-Src inhibition when compared to the corresponding
pyrrole-containing analogues.
F igu r e 1. Crystallographic analysis of compound 9a in the
ATP-binding pocket of FGF-R1 tyrosine kinase. Hydrogen
bonds are shown between compound 9a and FGF-R1 (dotted
lines). The nucleotide-binding loop is colored yellow for visu-
alization and has undergone a conformational change relative
to crystallographic analysis of FGF-R1 in the absence of 9a .
B. Su b st it u t ed 3-[2-(2-Oxo-1,2-d ih yd r oin d ol-3-
ylid en em et h yl)-4,5,6,7-t et r a h yd r o-1H -in d ol-3-yl]-
p r op ion ic Acid s. Previously, we have shown that the
propionic acid functionality at the C-3′ position of the
pyrrole ring could enhance the inhibitory activity against
both VEGF-R2 (Flk-1) and FGF-R1.9,12 Therefore, in this
study, we prepared the 9a series (Table 2) in order to
improve the inhibitory activities against the VEGF-R2
(Flk-1) and FGF-R1 kinases. Additionally, we have
cocrystallized the prototype compound, 9a , within the
catalytic domain of the FGF-R1 tyrosine kinase in order
to guide our understanding of the molecular mechanism
for this type of indolin-2-one series and to direct further
structural modification (Figure 1). Cocrystallographic
methods, X-ray diffraction, and refinement were similar
to that described previously12 and are more specifically
described in the Experimental Section (Table 3).
The crystallographic structure of 9a within the cata-
lytic domain of the FGF-R1 kinase revealed that 9a
binds to the receptor in a manner similar to that of
compound 2 (Table 2) described previously.12 The indo-
lin-2-one core of compound 9a was found to occupy the
adenine-binding site of FGF-R1 through a bidentate
hydrogen-bond donor-acceptor system. In this case, the
NH-1 and CdO of the indolin-2-one core were found to
form hydrogen bonds to the two backbone amino acid
residues, Glu562 and Ala564, in the hinge region of the
FGF-R1 catalytic domain. The tetrahydroindole moiety
of 9a projected to the opening of the ATP-binding cleft.
The NH-1′ of the tetrahydroindole ring was found to
form an intramolecular hydrogen bond to the carbonyl
oxygen at the C-2 position of the indolin-2-one core. In
addition, a conformational change in the nucleotide-
binding loop was found to occur in a manner similar to
that observed previously in the 2-FGF-R1 cocrystal
structure due to an interaction between the propionic
acid moiety at the C-3′ position of the tetrahydroindole
side chain and Asn568 in the hinge region. It is also
worth noting that three methylene moieties of the
tetrahydroindole ring were found to be exposed to the
solvent.
SAR analysis of a bromo substitution at the C-5
position of the 9b was found to improve the inhibitory
activity against both VEGF-R2 (Flk-1) (IC50 ) 0.03 µM)
and PDGF-Râ (IC50 ) 0.004 µM) kinases and retained
the potent inhibitory activities against FGF-R1 (IC50
)
0.27 µM) and p60c-Src (IC50 ) 1.27 µM) kinases when
compared to 9a (Table 2). Compound 9b represented the
most potent inhibitor of PDGF-Râ tyrosine kinase in
this report. The same SAR trend was observed for 4a
series (Table 1) where bromination at the C-5 position
of the 4a resulted in the most potent PDGF-R inhibitor
(compound 4b) of the series.
According to our previous study,9 hydrophilic substi-
tution at the C-5 position of the indolin-2-one core (e.g.
aminosulfonyl or carboxylic acid) was found to increase
the inhibitory potency of indolin-2-ones toward both
VEGF-R2 (Flk-1) and FGF-R1. Thus, in this study, these
same hydrophilic substituents were also introduced into
the core of 9a . Unlike our previous series, a hydrophilic
aminosulfonyl moiety at the C-5 position of 9c decreased
the potency toward VEGF-R2 (Flk-1) and retained the
inhibitory activity against both FGF-R1 and PDGF-R
kinases. On the other hand, compound 9d , containing
a carboxylic acid moiety at the C-5 position of the core,
was found to be 22- and 12-fold more potent than 9a
toward VEGF-R2 (Flk-1) and PDGF-Râ kinases, respec-
tively. This SAR was found to be different from that
deduced from our own previous study where analogues
containing the propionic acid group at the C-4′ position
of the pyrrole ring tended to increase the inhibitory
potencies toward both VEGF-R2 and FGF-R1 kinases
and retained or decreased activity against the PDGF-R
kinase.9 This discrepancy could be the result of different
receptor binding modes for the two series. In this regard,
the receptor conformational change has been observed
within the 9a -FGF-R1 cocrystal structure whereas
such change has not been found in the cocrystal
structure of FGF-R1 and the compound with a propionic
acid at the C-4′ position of the pyrrole ring.11 In addition,
both 9c and 9d were found to be more potent than 9a
It is of interest to interpret the biological activities of
compound 9a in context of the crystallographic informa-

2660 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 14
Sun et al.
when tested toward p60c-Src kinase. This is in ac-
cordance with the SAR of the 4a series where the
hydrophilic substituents greatly enhanced the p60c-Src
kinase inhibitory activity (compounds 4c and 4d com-
pared to 4a ). Phenyl or substituted phenyl substituents
at the C-6 position on the 9a core, in general, were found
to increase the inhibitory potencies toward p60c-Src
kinase, retain the inhibitory activity against VEGF-R2
(Flk-1), and decrease the potency toward FGF-R1 and
PDGF-R kinases. Two exceptions to this general obser-
vation were compounds 9f and 9h , which were more
potent than 9a when tested against PDGF-Râ (9f: IC50
) 0.04 µM) and FGF-R1 (9h : IC50 ) 0.08 µM) kinases,
respectively (Table 2). In addition, most of these 9a
analogues were found to be inactive against the EGF-R
tyrosine kinase.
In the cell proliferation assays, most of the compounds
in the 9a series were found to inhibit VEGF-, FGF-, and
PDGF-dependent cell proliferations with some com-
pounds with more selectivity toward inhibition of VEGF-
dependent HUVEC proliferation (Table 2). Specifically,
compounds 9a and 9b inhibited VEGF-dependent HU-
VEC proliferation with IC50 values of 2 and 0.3 nM,
respectively. Both compounds were also found to be good
inhibitors of FGF-dependent HUVEC proliferation with
IC50 values of 10 and 100 nM, respectively. However,
compound 9b, the most potent PDGF-Râ kinase inhibi-
tor (IC50 ) 4 nM), inhibited PDGF-dependent cell
proliferation with an IC50 value of 1.38 µM. The lack of
concordance between the two types of assays for the
highly polar compounds 9c and 9d might be due to poor
cell membrane permeability as discussed in our previous
report.9 Other factors might help explain this discor-
dance including chemical stability, solubility, or binding
to serum proteins in the tissue culture media. In
general, the 9a series was found to be more potent than
the 4a series to block cell proliferations.
and 4a that may be due to the presence or absence of
conformational changes upon binding. Third, the aryl
substituents at the C-6 position of the 4a core tended
to decrease, if not abolish, the potency against all three
targets including p60c-Src kinase. On the other hand,
aryl substituents in 9a -related compounds retained
inhibitory potencies toward VEGF-R2 (Flk-1) and PDGF-
Râ kinases, decreased inhibitory activity against FGF-
R1 kinase, and increased activity when tested against
p60c-Src kinase. Finally, all of the compounds in this
study were found to be inactive against the EGF-R
tyrosine kinase.
To date, the indolin-2-one series described in this
report exhibits the most profound potencies to block
growth factor-mediated signal transduction as measured
by cell proliferation. In this regard, these compounds
were found to inhibit signaling due to stimulation of
VEGF, FGF, and PDGF receptors in contrast to the EGF
receptor. We have extended this selectivity to show that
these compounds have no effect to block cell prolifera-
tion following stimulation of the IGF-1 receptor as well
(data not shown). We conclude, therefore, that these
compounds are relatively specific to inhibition of “split”
tyrosine kinases. In contrast, the activities of these
compounds on the p60c-Src kinases would not have been
explained purely on the receptor architecture since the
p60c-Src kinase is a cytoplasmic tyrosine kinase without
a kinase insert domain. Nonetheless, models of p60c-Src
kinase have been constructed and, in the vicinity of the
adenine-binding pocket, help explain the potencies of
these compounds to inhibit this enzyme.
In summary, we have prepared compounds that can
inhibit VEGF-R2 (Flk-1/KDR), FGF-R1, and PDGF-Râ
tyrosine kinases. Since all three RTKs play crucial roles
in tumor growth and the tumor angiogenic process, a
single compound with this type of target profile may be
advantageous as a therapeutic. Further in vivo studies
of those compounds may be useful to test for the
treatment of angiogenesis-related diseases, such as
cancer, diabetic retinopathies, atherosclerosis, psoriasis,
and rheumatoid arthritis. In addition, these compounds
may have utility in the treatment of PDGF and FGF
growth factor-related disorders, including cancer and
tissue injury, where these receptor systems have been
implicated.
Con clu sion s
In this study, we have explored a new indolin-2-one
pharmacophore that contains the tetrahydroindole moi-
ety at the C-3 position of the indolin-2-one core in order
to determine potency and selectivity of these compounds
to inhibit VEGF-R2 (Flk-1/KDR), FGF-R1, and PDGF-
Râ tyrosine kinases. In this regard, these compounds
were evaluated for their relative inhibitory activity
against a broad panel of tyrosine kinases and growth
factor-mediated signal transduction in cells.
The relationships between compounds in regard to the
inhibitory activities of VEGF-R2 (Flk-1), FGF-R1, PDGF-
Râ, p60c-Src, and EGF-R tyrosine kinases were inter-
preted in the context of crystallographic analysis of 9a -
FGF-R1 and protein structural models. First, it appears
that the propionic acid moiety is required for high
potencies against all three target RTKs (9a series
compared to 4a series). Second, acidic substituents (i.e.
aminosulfonyl or carboxyl group) at the C-5 position of
the 4a series resulted in favorable inhibitory activities
against both VEGF-R2 (Flk-1) and FGF-R1 (i.e. 4c and
4d compared to 4a ). In contrast, the hydrophilic sub-
stituent at the C-5 position of the 9a series resulted in
different effects on the inhibitory activity against these
kinases (9c and 9d compared to 9a ). This difference
might be the result of different binding modes for 9a
Exp er im en ta l Section
NMR spectra were recorded by Acorn NMR using a Nicolet
NT300 or Nicolet NT360. Tetramethylsilane (TMS) was used
as an internal standard and chemical shifts are reported in
parts per million (δ) downfield from TMS. Coupling constants
are reported in hertz (Hz). Mass spectra (electron spray) were
recorded by SYNPEP Corp., using an API I PLUS spectrom-
eter. Elemental analyses were performed by Galbraith Labo-
ratories, Inc. Elemental analysis results are within (0.4% of
the theoretical values.
3-(4,5,6,7-Tet r a h yd r o-1H -in d ol-2-ylm et h ylen e)-1,3-d i-
h yd r oin d ol-2-on e (4a ). Dimethylformamide (2.2 mL, 28.5
mmol) and 240 mL of dichloromethane were cooled to -4 °C
in an ice-salt bath maintained at -10 °C. Phosphorus
oxychloride (2.7 mL, 28.5 mmol) was rapidly added via the
dropping funnel. The reaction mixture was further stirred at
room temperature for 30 min and then cooled to 0 °C. 4,5,6,7-
Tetrahydro-1H-indole (2.3 g, 19 mmol) was added portionwise
to the above reaction mixture. The mixture was refluxed for 1
h, cooled to 5 °C, diluted with 2 N potassium hydroxide to pH
10, and further stirred at room temperature for 1 h. The

Growth Factor Receptor Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 14 2661
precipitate was filtered, washed with water and dried to give
2.0 g (70% yield) of 4,5,6,7-tetrahydro-1H-indole-2-carbalde-
hyde (3) as a tan solid: 1H NMR (360 MHz, DMSO-d6) δ 11.56
(s, br, 1H, NH-1), 9.26 (s, 1H, CHO-2), 6.66 (d, J ) 2.12 Hz,
1H, H-3′), 2.54 (t, J ) 6.18 Hz, 2H, CH2CH2CH2CH2), 2.43 (t,
CH2CH2CH2CH2), 1.70-1.79 (m, 4H, CH2CH2CH2CH2); MS
m/z (relative intensity, %) 341 ([M + 1]+, 50). Anal. (C23H20N2
O) C, H, N.
6-(3-Meth oxyp h en yl)-3-(4,5,6,7-tetr a h yd r o-1H-in d ol-2-
ylm eth ylen e)-1,3-d ih yd r oin d ol-2-on e (4g). This compound
was prepared using the same method as for synthesizing 4a
with a yield of 92%: 1H NMR (360 MHz, DMSO-d6) δ 13.10
(s, br, 1H, NH-1′), 10.83 (s, br, 1H, NH-1), 7.64 (d, J ) 7.95
Hz, 1H, H-4), 7.62 (s, 1H, H-vinyl), 7.35 (t, J ) 7.95 Hz, 1H,
H-5′′), 7.28 (dd, J ) 1.58, 7.95 Hz, 1H, H-5), 7.19 (d, J ) 7.95
Hz, 1H, H-4′′), 7.13 (t, J ) 2.25 Hz, 1H, H-2′′), 7.09 (d, J )
1.58 Hz, 1H, H-7), 6.90 (dd, J ) 2.25, 7.95 Hz, 1H, H-6′′), 6.60
(s, br, 1H, H-3′), 3.82 (s, 3H, OCH3-3′′), 2.71 (t, J ) 6.00 Hz,
2H, CH2CH2CH2CH2), 2.52 (t, J ) 6.11 Hz, 2H, CH2CH2-
CH2CH2), 1.70-1.79 (m, 4H, CH2CH2CH2CH2); MS m/z (rela-
tive intensity, %) 371 ([M + 1]+, 100]. Anal. (C24H22N2O2) C,
H, N.
6-(2-Meth oxyp h en yl)-3-(4,5,6,7-tetr a h yd r o-1H-in d ol-2-
ylm eth ylen e)-1,3-d ih yd r oin d ol-2-on e (4h ). This compound
was prepared using the same method as for synthesizing 4a
with a yield of 85%: 1H NMR (360 MHz, DMSO-d6) δ 13.11
(s, br, 1H, NH-1′), 10.74 (s, br, 1H, NH-1), 7.58 (d, J ) 7.41
Hz, 1H, H-4), 7.58 (s, 1H, H-vinyl), 7.27-7.34 (m, 2H), 6.99-
7.10 (m, 4H), 6.58 (s, br, 1H, H-3′), 3.76 (s, 3H, OCH3-2′′), 2.71
(t, J ) 5.91 Hz, 2H, CH2CH2 CH2CH2), 2.52 (t, J ) 6.01 Hz,
2H, CH2CH2 CH2CH2), 1.7-1.79 (m, 4H, CH2CH2 CH2CH2);
MS m/z (relative intensity, %) 371 ([M + 1]+, 100). Anal.
(C24H22N2O2) C, H, N.
6-(4-Meth oxyp h en yl)-3-(4,5,6,7-tetr a h yd r o-1H-in d ol-2-
ylm eth ylen e)-1,3-d ih yd r oin d ol-2-on e (4i). This compound
was prepared using the same method as for synthesizing 4a
with a yield of 50%: 1H NMR (360 MHz, DMSO-d6) δ 13.07
(s, br, 1H, NH-1′), 10.80 (s, br, 1H, NH-1), 7.61 (d, J ) 8.02
Hz, 1H, H-4), 7.58 (s, 1H, H-vinyl), 7.56 (d, J ) 8.83 Hz, 2H,
H-3′′,5′′), 7.21 (dd, J ) 1.57, 8.02 Hz, 1H, H-5), 7.04 (d, J )
1.57 Hz, 1H, H-7), 7.00 (d, J ) 8.83 Hz, 2H, H-2′′, 6′′), 6.58 (d,
br, J ) 1.05 Hz,1H, H-3′), 3.79 (s, 3H, OCH3-4′′), 2.71 (t, J )
6.10 Hz, 2H, CH2CH2CH2CH2), 2.51 (t, J ) 5.96 Hz, 2H, CH2-
CH2CH2CH2), 1.70-1.79 (m, 4H, CH2CH2CH2CH2); MS m/z
(relative intensity, %) 371 ([M + 1]+, 100). Anal. (C24H22N2O2)
C, H, N.
3-[2-(2-Oxo-1,2-d ih yd r oin d ol-3-ylid en em eth yl)-4,5,6,7-
tetr a h yd r o-1H-in d ol-3-yl]p r op ion ic Acid (9a ). 1-(Morpho-
lin-4-yl)cyclohexene (20.1 g, 120 mmol), 14.4 g (142 mmol) of
triethylamine and 100 mL of dichloromethane were charged
to a 1-L, three-neck round-bottom flask equipped with a reflux
condenser, mechanical stirring, a thermometer and a dropping
funnel. The mixture was refluxed for 15 min and cooled in a
water bath to 15-20 °C. With vigorous stirring 18 g (109
mmol) of ethylsuccinyl chloride dissolved in 40 mL of dichlo-
romethane was added over 5 min via the dropping funnel. At
the end of the addition the mixture contained a large amount
of precipitate and was refluxing. Refluxing was continued for
30 min and the mixture was cooled to ambient temperature
in a water bath. The mixture was extracted twice with 100
mL of water each time and twice with 30 mL of brine each
time, dried over 5 g of anhydrous sodium sulfate and evapo-
rated to give crude 4-(2-morpholin-4-ylcyclohex-1-enyl)-4-oxo-
butyric acid ethyl ester (5) as an oil.
J
) 5.88 Hz, 2H, CH2CH2CH2CH2), 1.62-1.75 (m, 4H,
CH2CH2CH2CH2); MS m/z (relative intensity, %) 150 ([M +
1]+, 100).
A mixture of 3 (90 mg, 0.60 mmol), 67 mg (0.50 mmol) of
oxindole and 1 drop of piperidine in 2 mL of ethanol was heated
to 95 °C overnight. The reaction mixture was cooled and
precipitate was filtered, washed with cold ethanol and hexane,
and dried in a vacuum oven overnight to give 121 mg (92%) of
6-(3-methoxy-phenyl)-3-(4,5,6,7-tetrahydro-1H-indol-2-ylmeth-
ylene)-1,3-dihydroindol-2-one as a brown solid: 1H NMR (300
MHz, DMSO-d6) δ 13.13 (s, 1H, NH-1′), 10.79 (s, br, 1H, NH-
1), 7.59 (s, 1H, H-vinyl), 7.58 (d, J ) 7.95 Hz, 1H, H-4), 7.10
(t, J ) 7.95 Hz, 1H, H-6), 6.97 (t, J ) 7.95 Hz, 1H, H-5), 6.87
(d, J ) 7.95 Hz, 1H, H-7), 6.59 (s, br, 1H, H-3′), 2.71 (t, J )
5.92 Hz, 2H, CH2CH2CH2CH2), 2.52 (t, J ) 3.33 Hz, 2H, CH2-
CH2CH2CH2), 1.70-1.80 (m, 4H, CH2CH2CH2CH2). Anal.
(C17H16N2O) C, H, N.
5-Br om o-3-(4,5,6,7-tetr ah ydr o-1H-in dol-2-ylm eth ylen e)-
1,3-d ih yd r oin d ol-2-on e (4b). This compound was prepared
using the same method as for synthesizing 4a with a yield of
99%: 1H NMR (300 MHz, DMSO-d6) δ 13.09 (s, 1H, NH-1′),
10.84 (s, br, 1H, NH-1), 7.79 (d, J ) 1.96 Hz, 1H, H-4), 7.71
(s, 1H, H-vinyl), 7.22 (dd, J ) 1.96, 8.07 Hz, 1H, H-6), 6.80 (d,
J ) 8.07 Hz, 1H, H-7), 6.59 (s, br, 1H, H-3′), 2.71 (t, J ) 6.12
Hz, 2H, CH2CH2CH2CH2), 2.51 (t, J ) 6.12 Hz, 2H, CH2CH2-
CH2CH2), 1.68-1.79 (m, 4H, CH2CH2CH2CH2). Anal. (C17H15
BrN2O) C, H, N.
-
2-Oxo-3-(4,5,6,7-t et r a h yd r o-1H -in d ol-2-ylm et h ylen e)-
2,3-d ih yd r o-1H-in d ole-5-su lfon ic Acid Am id e (4c). This
compound was prepared using the same method as for
synthesizing 4a with a yield of 79%: 1H NMR (360 MHz,
DMSO-d6) δ 13.04 (s, 1H, NH-1′), 11.10 (s, 1H, NH-1), 8.01 (d,
J ) 1.72 Hz, 1H, H-4), 7.72 (s, 1H, H-vinyl), 7.57 (dd, J ) 1.72,
8.12 Hz, 1H, H-6), 7.09 (s, br, 2H, H2NSO2-5), 6.99 (d, J )
8.12 Hz, 1H, H-7), 6.72 (s, br, 1H, H-3′), 2.72 (t, J ) 5.64 Hz,
2H, CH2CH2CH2CH2), 2.52 (t, J ) 5.64 Hz, 2H, CH2CH2-
CH2CH2), 1.70-1.79 (m, 4H, CH2CH2CH2CH2). Anal. (C17
17N3O3SC‚0.25H2O) C, H, N.
2-Oxo-3-(4,5,6,7-t et r a h yd r o-1H -in d ol-2-ylm et h ylen e)-
-
H
2,3-d ih yd r o-1H-in d ole-5-ca r boxylic Acid (4d ). This com-
pound was prepared using the same method as for synthesiz-
ing 4a with a yield of 60%: 1H NMR (360 MHz, DMSO-d6) δ
13.00 (s, br, 1H, H-1′), 12.49 (s, vbr., 1H, COOH-5), 11.06 (s,
1H, NH-1), 8.16 (d, J ) 1.60 Hz, 1H, H-4), 7.77 (s, 1H, H-vinyl),
7.74 (dd, J ) 1.81, 8.28 Hz, 1H, H-6), 6.94 (d, J ) 8.28 Hz,
1H, H-7), 6.66 (s, br, 1H, H-3′), 2.71 (t, J ) 5.85 Hz, 2H, CH2-
CH2CH2CH2), 2.52 (t, J ) 6.00 Hz, 2H, CH2CH2CH2CH2),
1.70-1.79 (m, 4H, CH2CH2CH2CH2). Anal. (C18H16N2O3‚
0.5H2O) C, H, N.
6-Meth oxy-3-(4,5,6,7-tetr a h yd r o-1H-in d ol-2-ylm eth yl-
en e)-1,3-d ih yd r oin d ol-2-on e (4e). This compound was pre-
pared using the same method as for synthesizing 4a with a
yield of 90%: 1H NMR (300 MHz, DMSO-d6) δ 12.90 (s, br,
1H, NH-1′), 10.72 (s, 1H, NH-1), 7.46 (d, J ) 8.44 Hz, 1H, H-4),
7.42 (s, 1H, H-vinyl), 6.55 (dd, J ) 2.16, 8.44 Hz, 1H, H-5),
6.49 (d, J ) 1.78 Hz, 1H, H-3′), 6.43 (d, J ) 2.16 Hz, 1H, H-7),
3.74 (s, 3H, OCH3-6), 2.68 (t, J ) 5.67 Hz, 2H, CH2CH2CH2-
CH2), 2.48-2.50 (m, 2H, CH2CH2CH2CH2), 1.68-1.78 (m, 4H,
CH2CH2CH2CH2); MS m/z (relative intensity, %) 295 ([M +
1]+, 59). Anal. (C18H18N2O2) C, H, N.
6-P h en yl-3-(4,5,6,7-tetr ah ydr o-1H-in dol-2-ylm eth ylen e)-
1,3-d ih yd r oin d ol-2-on e (4f). This compound was prepared
using the same method as for synthesizing 4a with a yield of
70%: 1H NMR (360 MHz, DMSO-d6) δ 13.10 (s, br, 1H, NH-
1′), 10.85 (s, br, 1H, NH-1), 7.61-7.66 (m, 4H, H-4, H-vinyl,
and H-3′′,5′′), 7.44 (t, br, J ) 7.64 Hz, 2H, H-2′′,6′′), 7.33 (t,
br, J ) 7.64, 1H, H-4′), 7.27 (dd, J ) 1.60, 8.06 Hz, 1H, H-5),
7.10 (d, J ) 1.60 Hz, 1H, H-7), 6.60 (s, br, 1H, H-3′), 2.71 (t, J
) 6.02 Hz, 2H, CH2CH2CH2CH2), 2.52 (t, J ) 6.05 Hz, 2H,
Crude 5 (30 g, 102 mmol), 26.7 g (126 mmol) of diethyl
aminomalonate hydrochloride, 10.8 g (132 mmol) of sodium
acetate and 28 mL of glacial acetic acid were charged to a 1-L,
three-neck round-bottom flask equipped with mechanical
stirring, a reflux condenser and heated in an oil bath. The
mixture was heated to 108 °C over 30 min accompanied by
the evolution of carbon dioxide and the formation of a
precipitate of sodium chloride. The mixture was held at 100-
108 °C for 2 h and cooled to about 50 °C in a water bath. Water
(160 mL) and 160 mL of ethyl acetate were added. The
precipitate dissolved. The ethyl acetate layer was separated
and washed twice with 100 mL of water each time, once with
70 mL of saturated sodium bicarbonate solution, once with 50
mL of brine, dried over 10 g of anhydrous sodium sulfate and
rotary evaporated to give 30 g (85% yield) of crude 2-ethoxy-

2662 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 14
Sun et al.
carbonyl-3-(2-ethoxycarbonylethyl)-4,5,6,7-tetrahydroindole (6)
as an oil: 1H NMR (400 MHz, DMSO-d6) δ 11.09 (s, 1H, NH-
1), 4.18 (q, J ) 6.9 Hz, 2H, COOCH2CH3-2), 4.03 (q, J ) 7.2
Hz, 2H, CH2CH2COOCH2CH3-3), 2.84 (t, J ) 8.0 Hz, 2H, CH2-
CH2COOC2H5-3), 2.49 (t, J ) 5.4 Hz, 2H, CH2CH2CH2CH2),
2.42 (t, J ) 8.0 Hz, 2H, CH2CH2COOC2H5), 2.34 (t, J ) 5.4
Hz, 2H, CH2CH2CH2CH2), 1.66 (t, J ) 5.4 Hz, 4H, CH2CH2CH2-
CH2), 1.26 (t, J ) 6.9 Hz, 3H, COOCH2CH3), 1.16 (t, J ) 7.2
Hz, 3H, COOCH2CH3).
CH2CH2COOH-3′, CH2CH2CH2CH2), 1.70-1.77 (m, 4H, CH2-
CH2CH2CH2). Anal. (C20H20N2O3) C, H, N.
3-[2-(5-Br om o-2-oxo-1,2-dih ydr oin dol-3-yliden em eth yl)-
4,5,6,7-tetr ah ydr o-1H-in dol-3-yl]pr opion ic Acid (9b). Com-
pound 8 (15.9 g, 72 mmol), 12.7 g (60 mmol) of 5-bromo-2-
indolin-2-one and 150 mL of ethanol were heated to near reflux
in a 500-mL, three-neck round-bottom flask equipped with a
reflux condenser and magnetic stirring. Piperidine (6.4 g, 76
mmol) was slowly added and the mixture refluxed for 1 h at
which time a large amount of reddish precipitate was present.
Thin-layer chromatography (silica gel, ethyl acetate) showed
no starting materials remaining. Acetic acid (20 mL) was
slowly added causing a voluminous precipitate. The mixture
was refluxed for 5 min and cooled to ambient temperature.
The precipitate was collected by vacuum filtration and washed
with 30 mL of ethanol. The solids were slurry-washed by
refluxing in 40 mL of ethanol, cooled, collected by vacuum
filtration, washed with 30 mL of ethanol and dried under high
vacuum to give 22.3 g (89% yield) of 3-[2-(5-bromo-2-oxo-1,2-
dihydroindol-3-ylidenemethyl)-4,5,6,7-tetrahydro-1H-indol-3-
yl]propionic acid (9b) as a red-orange solid: 1H NMR (360
MHz, DMSO-d6) δ 13.32 (s, 1H, NH-1′), 12.12 (s, br, 1H, CH2-
CH2COOH-3′), 10.86 (s, 1H, NH-1), 7.96 (d, J ) 1.91 Hz, 1H,
H-4) 7.71 (s, 1H, H-vinyl), 7.19 (dd, J ) 1.91, 8.19 Hz, 1H,
H-6), 6.79 (d, J ) 8.19 Hz, 1H, H-7), 2.92 (t, J ) 7.55 Hz, 2H,
CH2CH2COOH-3′), 2.66 (t, J ) 5.25 Hz, 2H, CH2CH2CH2CH2),
2.40-2.44 (m, 4H, CH2CH2COOH-3′, CH2CH2CH2CH2), 1.70-
1.77 (m, 4H, CH2CH2CH2CH2). Anal. (C20H19BrN2O3) C, H, N.
Crude 6 (30 g, 102 mmol) and 80 mL (400 mmol) of 5 N
sodium hydroxide were refluxed for 80 min in a round-bottom
flask heated in an oil bath and equipped with magnetic
stirring. The heater was turned off but the flask left in the
hot bath and about 44 mL (440 mmol) of 10 N hydrochloric
acid was cautiously added via a dropping funnel through the
reflux condenser with vigorous stirring. When approximately
80% of the acid had been added, a large amount of carbon
dioxide was evolved. The addition was continued until the pH
was 2-3. The mixture was cooled in a water bath and 200
mL of ethyl acetate was added to dissolve the oil that was
present. The ethyl acetate layer was isolated, washed 3 times
with 50 mL of water each time and twice with 30 mL of brine
each time, dried over 5 g of anhydrous sodium sulfate and
evaporated to give 15.5 g (80%) yield of crude 3-(2-carboxy-
ethyl)-4,5,6,7-tetrahydroindole (7) as
a very dark sticky
syrup: 1H NMR (400 MHz, DMSO-d6) δ 12.0 (s, 1H COOH),
9.92 (br, 1H, NH-1), 6.26 (d, J ) 1.6 Hz, 1H, H-2), 2.28-2.51
(m, 8H, CH2×4), 1.64-1.78 (m, 4H, CH2×2).
3-[2-(2-Oxo-5-su lfa m oyl-1,2-d ih yd r oin d ol-3-ylid en e-
m eth yl)-4,5,6,7-tetr a h yd r o-1H-in d ol-3-yl]p r op ion ic Acid
(9c). This compound was prepared using the same procedure
as for preparation of 9b with a yield of 60%: 1H NMR (300
MHz, DMSO-d6) δ 13.33 (s, br, 1H, NH-1′), 12.15 (s, br, 1H,
CH2CH2COOH-3′), 11.12 (s, 1H, NH-1), 8.11 (s, 1H, H-vinyl),
7.71 (s, 1H, H-4), 7.55 (d, J ) 8.26 Hz, 1H, H-6), 7.13 (s, br,
2H, SO2NH2-5), 6.98 (d, J ) 8.26 Hz, 1H, H-7), 2.93 (t, J )
6.92 Hz, 2H, CH2CH2COOH-3′), 2.69 (s, br, CH2CH2CH2CH2),
2.41-2.48 (m, 4H, CH2CH2COOH-3′ and CH2CH2CH2CH2),
1.61-1.73 (m, 4H, CH2CH2CH2CH2); MS m/z (relative inten-
sity, %) 415 (M+•, 100). Anal. (C20H21N3O5S‚1.25H2O) C, H, N.
Dimethylformamide (11.7 g, 160 mmol) and 240 mL of
dichloromethane in a 1-L, three-neck round-bottom flask
equipped with magnetic stirring, a reflux condenser and a
dropping funnel were cooled to -4 °C in an ice-salt bath
maintained at -10 °C. Phosphorus oxychloride (24.5 g, 160
mmol) was rapidly added via the dropping funnel. The tem-
perature increased to -3 °C and then, with further stirring,
decreased to -7 °C. Crude 7 (15.5 g, 80 mmol) dissolved in
160 mL of dichloromethane was added via the dropping funnel
over 15 min keeping the temperature below 1 °C. The mixture
was refluxed for 10 min, cooled to 5 °C, and diluted with 300
mL of water. The aqueous layer was evaporated and saved
and the organic layer extracted with another 30 mL of water.
The aqueous layers were combined, washed with 30 mL of
dichloromethane and cooled to 5 °C. The aqueous layer was
adjusted to pH 10 with about 48 mL of 10 N sodium hydroxide
accompanied by a temperature increase to 15 °C. The mixture
was then cooled to 10 °C and adjusted to pH 2-3 with about
48 mL of 10 N hydrochloric acid. The oil which formed
solidified and was collected by vacuum filtration, washed three
times with 20 mL of water each time and dried under high
vacuum to give 5.5 g (31% yield) of 3-(2-formyl-4,5,6,7-
tetrahydro-1H-indol-3-yl)propionic acid (8) as a brown solid:
1H NMR (360 MHz, DMSO-d6) δ 11.4 (s, 1H, NH-1), 9.24 (s,
1H, CHO-2), 2.80 (t, J ) 9.2 Hz, 2H, CH2CH2COOH), 2.35-
2.50 (m, 6H, CH2×3), 1.64-1.89 (m, 4H, CH2×2).
3-[3-(2-Ca r boxyeth yl)-4,5,6,7-tetr a h yd r o-1H-in d ol-2-yl-
m eth ylen e]-2-oxo-2,3-dih ydr o-1H-in dole-5-car boxylic Acid
(9d ). This compound was prepared using the same procedure
as for preparation of 9b with a yield of 13%: 1H NMR (300
MHz, DMSO-d6) δ 13.27 (s, br, 1H, NH-1′), 12.34 (s, br, 2H,
CH2CH2COOH-3′ and COOH-5), 11.08 (s, 1H, NH-1), 8.27 (s,
1H, H-vinyl), 7.76 (s, 1H, H-4), 7.72 (d, J ) 8.05 Hz, 1H, H-6),
6.93 (d, J ) 8.05 Hz, 1H, H-7), 2.94 (t, J ) 7.19 Hz, 2H, CH2-
CH2COOH-3′), 2.68 (s, br, CH2CH2CH2CH2), 2.34-2.49 (m, 4H,
CH2CH2COOH-3′ and CH2CH2CH2CH2), 1.64-1.73 (m, 4H,
CH2CH2CH2CH2); MS m/z (relative intensity, %) 380 (M+•
100). Anal. (C21H20N2O5‚0.25H2O) C, H, N.
,
3-[2-(6-Me t h oxy-2-oxo-1,2-d ih yd r oin d ol-3-ylid e n e -
m eth yl)-4,5,6,7-tetr a h yd r o-1H-in d ol-3-yl]p r op ion ic Acid
(9e). This compound was prepared using the same procedure
as for preparation of 9b with a yield of 50%: 1H NMR (360
MHz, DMSO-d6) δ 13.27 (s, br, 1H, NH-1′), 12.06 (s, vbr, 1H,
CH2CH2COOH-3′), 10.80 (s, br, 1H, NH-1), 7.74 (d, J ) 7.97
Hz, 1H, H-4), 7.64 (s, 1H, H-vinyl), 7.63 (d, J ) 7.83 Hz, 2H,
H-2′′,6′′), 7.44 (t, J ) 7.83 Hz, 2H, H-3′′,5′′), 7.33 (dd, J ) 7.83
Hz, 1H, H-4′′), 7.27 (dd, J ) 1.11, 7.97 Hz 1H, H-5), 7.10 (d, J
) 1.11 Hz, 1H, H-7), 2.92 (t, J ) 7.41 Hz, 2H, CH2CH2COOH-
3′), 2.67 (t, J ) 5.51 Hz, 2H, CH2CH2CH2CH2), 2.41-2.46 (m,
4H, CH2CH2COOH-3′ and CH2CH2CH2CH2), 1.73-1.76 (m,
Compound 8 (5.4 g, 24 mmol), 3.57 g (27 mmol) of indolin-
2-one and 25 mL of ethanol were heated to near reflux in a
250-mL, three-neck round-bottom flask equipped with a reflux
condenser and mechanical stirring. Piperidine (2.7 g, 310
mmol) was slowly added and the mixture refluxed for 4 h.
Acetic acid (8 mL) was slowly added causing a voluminous
precipitate. The mixture was refluxed for 5 min and cooled to
ambient temperature. The precipitate was collected by vacuum
filtration and washed with 20 mL of ethanol. The solids were
slurry-washed by refluxing in 30 mL of ethanol, cooled,
collected by vacuum filtration, washed with 30 mL of ethanol
and dried under high vacuum to give 5.5 g (68% yield) of 3-[2-
(2-oxo-1,2-dihydroindol-3-ylidenemethyl)-4,5,6,7-tetrahydro-
1H-indol-3-yl]propionic acid (9a ) as an orange solid: 1H NMR
(360 MHz, DMSO-d6) δ 13.27 (s, 1H, NH-1′), 12.09 (s, br, 1H,
CH2CH2COOH-3′), 10.71 (s, 1H, NH-1), 7.65 (d, J ) 7.52 Hz,
1H, H-4) 7.60 (s, 1H, H-vinyl), 7.07 (dt J ) 1.05, 7.52 Hz, 1H,
H-5), 6.95 (dt, J ) 1.05, 7.52 Hz, 1H, H-6), 6.85 (d, J ) 7.52
Hz, 1H, H-7), 2.90 (t, J ) 7.65 Hz, 2H, CH2CH2COOH-3′), 2.66
(t, J ) 5.59 Hz, 2H, CH2CH2CH2CH2), 2.39-2.47 (m, 4H,
4H, CH2CH2CH2CH2); MS m/z (relative intensity, %) 411 (M+•
65). Anal. (C26H24N2O3‚0.5H2O) C, H, N.
,
3-[2-(2-Oxo-6-ph en yl-1,2-dih ydr oin dol-3-yliden em eth yl)-
4,5,6,7-tetr a h yd r o-1H-in d ol-3-yl]p r op ion ic Acid (9f). This
compound was prepared using the same procedure as for
preparation of 9b with a yield of 31%: 1H NMR (360 MHz,
DMSO-d6) δ 13.27 (s, br, 1H, NH-1′), 12.06 (s, vbr, 1H, CH2-
CH2COOH-3′), 10.80 (s, br, 1H, NH-1), 7.74 (d, J ) 7.97 Hz,
1H, H-4), 7.64 (s, 1H, H-vinyl), 7.63 (d, J ) 7.83 Hz, 2H,
H-2′′,6′′), 7.44 (t, J ) 7.83 Hz, 2H, H-3′′,5′′), 7.33 (dd, J ) 7.83

Growth Factor Receptor Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 14 2663
Hz, 1H, H-4′′), 7.27 (dd, J ) 1.11, 7.97 Hz 1H, H-5), 7.10 (d, J
) 1.11 Hz, 1H, H-7), 2.92 (t, J ) 7.41 Hz, 2H, CH2CH2COOH-
3′), 2.67 (t, J ) 5.51 Hz, 2H, CH2CH2CH2CH2), 2.41-2.46 (m,
4H, CH2CH2COOH-3′ and CH2CH2CH2CH2), 1.73-1.76 (m,
CNS was used for simulated annealing and positional/B-factor
refinement,16 and O was used for model building.17 Bulk
solvent and anisotropic B-factor corrections were applied
during refinement. The average B-factor is 32.5 Å2 for all
atoms, 32.5 Å2 for protein atoms, 33.7 Å2 for water molecules,
and 31.4 Å2 for 9a atoms.
4H, CH2CH2CH2CH2); MS m/z (relative intensity, %) 411 (M+•
65). Anal. (C26H24N2O3‚0.5H2O) C, H, N.
,
3-{2-[6-(3-Met h oxyp h en yl)-2-oxo-1,2-d ih yd r oin d ol-3-
y li d e n e m e t h y l]-4,5,6,7-t e t r a h y d r o -1H -i n d o l-3-y l}-
p r op ion ic Acid (9g). This compound was prepared using the
same procedure as for preparation of 9b with a yield of 82%:
1H NMR (360 MHz, DMSO-d6) δ 13.26 (s, br, 1H, NH-1′), 10.79
(s, br, 1H, NH-1), 7.72 (d, J ) 8.09 Hz, 1H, H-4), 7.64 (s, 1H,
H-vinyl), 7.35 (t, J ) 7.94 Hz, 1H, H-5′′), 7.27 (dd, J ) 1.35,
8.09 Hz, 1H, H-5), 7.19 (d, br, J ) 7.94 Hz 1H, H-6′′), 7.13 (t,
J ) 2.02 Hz, 1H, H-2′′), 7.09 (d, J ) 1.35 Hz, 1H, H-7), 6.90
(dd, J ) 2.02, 7.94 Hz, 1H, H-4′′), 3.80 (s, 3H, OCH3-3′′), 2.91
(t, J ) 7.35 Hz, 2H, CH2CH2COOH-3′), 2.66 (t, J ) 5.88 Hz,
2H, CH2CH2CH2CH2), 2.38-2.45 (m, 4H, CH2CH2COOH-3′
and CH2CH2CH2CH2), 1.68-1.76 (m, 4H, CH2CH2CH2CH2);
MS m/z (relative intensity, %) 443 (M+, 100). Anal. (C21H22N2O4)
C, H, N.
Refer en ces
(1) Folkman, J . What is the evidence that tumors are angiogenesis
dependent? J . Natl. Cancer Inst. 1990, 82, 4-6.
(2) Folkman, J .; Shing, Y. Angiogenesis. J . Biol. Chem. 1992, 267,
10931-10934.
(3) Strawn, L. M; McMahon, G.; App, H.; Schreck, R.; Kuchler, W.
R.; Longhi, M. P.; Hui, T. H., Tang, C.; Levitzki, A., Gazit, A.;
Chen, I.; Keri, G.; Orfi, L.; Risau, W.; Flamme, I.; Ullrich, A.;
Hirth, K. P.; Shawver, L. K. Flk-1 as a target for tumor growth
inhibition. Cancer Res. 1996, 56, 3540-3545.
(4) Shawver, L. K.; Lipson, K. E.; Fong, T. A. T.; McMahon, G.;
Plowman, G. D.; Strawn, L. M. Receptor tyrosine kinases as
targets for inhibition of angiogenesis. Drug Discovery Today
1997, 2 (2), 50-63.
(5) Pepper, M. S.; Ferrara, N.; Orci, L.; Montessano, R. Potent
synergism between vascular endothelial growth factor and basic
fibroblast growth factor in the induction of angiogenesis in vitro.
Biochem. Biophys. Res. Commun. 1992, 189, 824-831.
(6) Hanahan, D.; Folkman, J . Patterns and emerging mechanisms
of the angiogenic switch during tumorigenesis. Cell 1996, 86,
353-364.
(7) Hennequin, L. F.; Thomas, A. P.; J ohnstone, C.; Ple, P.; Stokes,
E. S. E.; Ogilvie, D. J .; Dukes, M.; Wedge, S. R. 90th Annual
Meeting of American Association for Cancer Research, Phila-
delphia, PA, Apr 10-14, 1999; Abstract 457.
(8) Sun, L.; Tran, N.; Tang, T.; App, H.; Hirth, P.; McMahon, G.;
Tang, C. Synthesis and biological evaluations of 3-substituted
indolin-2-ones: a novel class of tyrosine kinase inhibitors that
exhibit selectivity towards particular receptor tyrosine kinases.
J . Med. Chem. 1998, 41 (14), 2588-2603.
(9) Sun, L.; Tran, N.; Liang, C.; Tang, F.; Rice, A.; Schreck, R.;
Waltz, K.; Shawver, L. K.; McMahon, G.; Tang, C. Design,
synthesis, and evaluations of substituted 3-[(3- or 4-carboxyeth-
ylpyrrol-2-yl)methylidenyl]indolin-2-ones as inhibitors of VEGF,
FGF, and PDGF receptor tyrosine kinases. J . Med. Chem. 1999,
42 (25), 5120-5130.
(10) Fong, A. T. T.; Shawver, L. K.; Sun, L.; Tang, C.; App. H.; Powell,
T. J .; Kim, Y. H.; Schreck, R.; Wang, X. Y.; Risau, W.; Ullrich,
A.; Hirth, K. P.; McMahon, G. SU5416 is a potent and selective
inhibitor of the vascular endothelial growth factor receptor (Flk-
1/KDR) that inhibits tyrosine kinase catalysis, tumor vascular-
ization, and growth of multiple tumor types. Cancer Res. 1999,
59, 99-106.
(11) Laird, A. D.; Vajkoczy, P.; Shawver, L. K.; Thurnher, A.; Liang,
C.; Mohammadi, M.; Schlessinger, J .; Ullrich, A.; Hubbard, S.
R.; Blake, R. A.; Fong, T. A. T.; Strawn, L. M.; Sun, L.; Tang,
C.; Hawtin, R.; Tang, F.; Shenoy, N.; Hirth, K. P.; McMahon,
G.; Cherrington, J . M. SU6668 is a potent anti-angiogenic and
antitumor agent which induces regression of established tumors.
Cancer Res., in press.
(12) Mohammadi, M.; McMahon, G., Sun, L.; Tang, P. C.; Hirth, P.;
Yeh, B. K.; Hubbard, S. R.; Schlessinger, J . Structures of the
tyrosine kinase domain of fibroblast growth factor receptor in
complex with inhibitors. Science 1997, 276, 955-960.
(13) Abram, C. L.; Courtneidge, S. A. Src family tyrosine kinases and
growth factor signaling. In preparation.
3-{2-[6-(2-Met h oxyp h en yl)-2-oxo-1,2-d ih yd r oin d ol-3-
y li d e n e m e t h y l]-4,5,6,7-t e t r a h y d r o -1H -i n d o l-3-y l}-
p r op ion ic Acid (9h ). This compound was prepared using the
same procedure as for preparation of 9b with a yield 35%: 1H
NMR (360 MHz, DMSO-d6) δ 13.26 (s, br, 1H, NH-1′), 12.06
(s, br, 1H, COOH), 10.70 (s, br, 1H, NH-1), 7.67 (d, J ) 7.72
Hz, 1H, H-4), 7.61 (s, 1H, H-vinyl), 7.27-7.34 (m, 2H, H-4′′,5′′),
7.09 (d, br, J ) 7.72 Hz, H-6′′) 7.06 (dd, J ) 1.35, 7.72 Hz, 1H,
H-5), 7.02 (dd, J ) 0.94, 7.54, Hz, 1H, H-3′′), 6.99 (d, J ) 1.35
Hz, 1H, H-7), 3.76 (s, 3H, OCH3-2′′), 2.91 (t, J ) 7.54 Hz, 2H,
CH2CH2COOH), 2.67 (t, J ) 5.70 Hz, 2H, CH2CH2CH2CH2),
2.40-2.46 (m, 4H, CH2CH2COOH and CH2CH2CH2CH2),
1.71-1.78 (m, 4H, CH2CH2CH2CH2); MS m/z (relative inten-
sity, %) 441 (M+•, 100). Anal. (C27H26N2O4‚0.5H2O) C, H, N.
3-{2-[6-(4-Met h oxyp h en yl)-2-oxo-1,2-d ih yd r oin d ol-3-
y li d e n e m e t h y l]-4,5,6,7-t e t r a h y d r o -1H -i n d o l-3-y l}-
p r op ion ic Acid (9i). This compound was prepared using the
same procedure as for preparation of 9b with a yield of 30%;
1H NMR (360 MHz, DMSO-d6) δ 13.24 (s, br, 1H, NH-1), 11.61
(s, br, 1H, CH2CH2COOH-3′), 10.76 (s, br, 1H, NH-1), 7.77 (d,
J ) 8.09 Hz, 1H, H-4), 7.61 (s, 1H, H-vinyl), 7.58 (d, J ) 8.83
Hz, 2H, H-2′′,6′′) 7.21 (dd, J ) 1.47, 8.09 Hz, 1H, H-5), 7.04
(d, J ) 1.47 Hz, 1H, H-7), 7.01 (d, J ) 8.83 Hz, 2H, H-3′′,5′′),
3.79 (s, 3H, OCH3-4′′), 2.91 (t, J ) 7.35 Hz, 2H, CH2CH2COOH-
3′), 2.67 (t, J ) 5.88 Hz, 2H, CH2CH2CH2CH2), 2.40-2.47 (m,
4H, CH2CH2COOH-3′ and CH2CH2CH2CH2), 1.72-1.78 (m,
4H, CH2CH2CH2CH2); MS m/z (relative intensity, %) 441 (M+,
100). Anal. (C27H26N2O4‚H2O) C, H, N.
Cell P r olifer a tion a n d Kin a se Assa ys. These assays were
described previously.9
X-r a y Da t a Collect ion a n d St r u ct u r e R efin em en t .
Expression, purification and crystallization of FGF-R1 were
performed as described.14 Crystals of unliganded FGF-R1 grow
in space group C2 with two molecules in the asymmetric unit
and unit cell parameters when frozen of a ) 208.9 Å, b ) 57.5
Å, c ) 65.7 Å, and â ) 107.6. Unliganded crystals were soaked
in 500 mL of stabilizing solution [25% poly(ethylene glycol)
10000, 0.3 M (NH4)2SO4, 0.1 M bis-Tris (pH 6.5), 5% ethylene
glycol, 2% DMSO] containing 2 mM 9a at 4 °C for 1 week.
Data were collected on a Rigaku RU-200 rotating anode (Cu
KR) operating at 50 kV and 100 mA and equipped with double-
focusing mirrors and an R-AXIS IIC image plate detector.
Crystals were flash-cooled in a dry nitrogen stream at -175
°C. Data were processed using DENZO and SCALEPACK.15
Difference Fourier electron density maps were computed using
phases calculated from the structure of unliganded FGF-R1.14
(14) Mohammadi, M.; Schlessinger, J .; Hubbard, S. R. Structure of
the FGF receptor tyrosine kinase domain reveals
autoinhibitory mechanism. Cell 1996, 86, 577-587.
(15) Otwinowski, Z.; Minor, W. Processing of X-ray diffraction data
collected in oscillation mode. Methods Enzymol. 1997, 276, 307-
326.
(16) Brunger, A. T.; et al. Crystallography & NMR system: A new
software suite for macromolecular structure determination. Acta
Crystallogr. D 1998, 54, 905-921.
(17) J ones, T. A.; Zou, J . Y.; Cowan, S. W.; Kjeldgaard, M. Improved
methods for building protein models in electron density maps
and the location of errors in these models. Acta Crystallogr. A
1991, 47, 110-119.
a novel
J M9906116