Use of a pharmacophore model for the design of EGFR tyrosine kinase inhibitors: Isoflavones and 3-phenyl-4(1H)-quinolones

Article abstract of DOI:10.1021/jm980551o

Using a pharmacophore model for ATP-competitive inhibitors interacting with the active site of the EGFR protein tyrosine kinase together with published X-ray crystal data of quercetin (2) in complex with the Hck tyrosine kinase and of deschloroflavopirido

Full text of DOI:10.1021/jm980551o

1018
J . Med. Chem. 1999, 42, 1018-1026
Use of a P h a r m a cop h or e Mod el for th e Design of EGF R Tyr osin e Kin a se
In h ibitor s: Isofla von es a n d 3-P h en yl-4(1H)-qu in olon es
Peter Traxler,*,† J ennifer Green,‡ Helmut Mett,† Urs Se´quin,‡ and Pascal Furet*,†
NOVARTIS Pharmaceuticals, Therapeutic Area Oncology, NOVARTIS Limited, CH-4002 Basel, Switzerland, and
Institute of Organic Chemistry, University of Basel, CH-4056 Basel, Switzerland
Received September 29, 1998
Using a pharmacophore model for ATP-competitive inhibitors interacting with the active site
of the EGFR protein tyrosine kinase together with published X-ray crystal data of quercetin
(2) in complex with the Hck tyrosine kinase and of deschloroflavopiridol (3b) in complex with
CDK2, a putative binding mode of the isoflavone genistein (1) was proposed. Then, based on
literature data suggesting that a salicylic acid function, which is represented by the 5-hydroxy-
4-keto motif in 1, could serve as a pharmacophore replacement of a pyrimidine ring,
superposition of 1 onto the potent EGFR tyrosine kinase inhibitor 4-(3′-chlorophenylamino)-
6,7-dimethoxyquinazoline (4) led to 3′-chloro-5,7-dihydroxyisoflavone (6) as a target structure
which in fact was 10 times more potent than 1. The putative binding mode of 6 suggests a
sulfur-aromatic interaction of the m-chlorophenyl moiety with Cys 773 in the “sugar pocket”
of the EGFR kinase model. Replacement of the oxygen in the chromenone ring of 6 by a nitrogen
atom further improved the inhibitory activity against the EGFR kinase. With IC50 values of
38 and 8 nM, respectively, the quinolones 11 and 12 were the most potent compounds of the
series. N-Alkylation of 11 did not further improve enzyme inhibitory activity but led to
derivatives with cellular activity in the lower micromolar range.
In tr od u ction
reported and reviewed in the past few years,3,6-10
compounds competing with ATP for binding at the
catalytic domain are considered to be of great interest.
Numerous examples of structurally diverse classes have
proved to be highly potent and selective ATP-competi-
tive tyrosine kinase inhibitors. Interest has focused on
a special group of compounds containing a (phenylami-
no)pyrimidine moiety in their structures such as (phen-
ylamino)quinazolines,11-14 7-amino-4-(phenylamino)py-
rido[4,3-d]pyrimidines,15,16 4-(phenylamino)pyrimido-
[5,4-d]pyrimidines,17 4-(phenylamino)pyrrolo[2,3-d]py-
rimidines,18 and 4-(phenylamino)pyrazolo[3,4-d]pyrim-
idines.19 Meanwhile, the first quinazoline derivatives
have entered clinical trials as anticancer agents (e.g.,
Zeneca ZD 183920 and Pfizer/Oncogene CP 35877421),
and several compounds of the other classes are in late
stage of preclinical development.
Protein tyrosine kinases (PTK) play a fundamental
role in signal transduction pathways. Deregulated PTK
activity has been observed in many proliferative dis-
eases (e.g., cancer, psoriasis, restenosis, etc.).1 A number
of tumor types have dysfunctional growth factor receptor
PTKs which results in inappropriate mitogenic signal-
ing. Tyrosine kinases are therefore attractive targets
for the design of new therapeutic agents against can-
cer.2,3 The EGFR PTK was one of the first tyrosine
kinases described in the literature and was therefore
chosen as a target by many companies to start research
programs in the signal transduction area.
The family of the epidermal growth factor receptor
(EGFR) PTK belongs to the large class of the trans-
membrane growth factor receptor PTKs. This EGFR
family contains four members: the EGFR kinase (c-
erbB1 gene product), the p185erbB2 (c-erbB2 gene prod-
uct), and the c-erbB3 and c-erbB4 gene products. The
EGFR and its ligands (EGF, TGF-R) have been impli-
cated in numerous tumors of epithelial origin (e.g.,
squamous cell carcinoma; breast, ovarian, and NSC lung
cancer)1,4 and proliferative disorders of the epidermis
such as psoriasis.5 The EGFR and c-erbB2 PTKs have
been identified as interesting targets for medicinal
chemistry programs. Inhibitors of the EGFR PTK are
therefore expected to have great therapeutic potential
in the treatment of malignant and nonmalignant epi-
thelial diseases.
Although more than 30 crystal structures of protein
kinases complexed with ATP or an ATP-competitive
inhibitor have been published in the past few years, no
crystal structure of the EGFR has yet appeared. In 1995,
Furet et al. (Novartis) published the first data of their
pharmacophore model for inhibitors competing with the
ATP-binding site of the EGFR PTK.22 This model has
then successfully been used for the design and synthesis
of 4-(phenylamino)pyrrolopyrimidines18 and 4-(phenyl-
amino)pyrazolo[3,4-d]pyrimidines.19 In parallel, Parke-
Davis scientists proposed a model for the binding of
4-(phenylamino)quinazolines, 4-(phenylamino)pyrido-
[4,3-d]pyrimidines,23 or pyrido[2,3-d]pyrimidines24 at
the ATP-binding site of the EGFR tyrosine kinase.
Within the great number of different structural
classes of tyrosine kinase inhibitors which have been
In the present paper, we describe another successful
application of our pharmacophore model leading to a
hypothetical binding mode for genistein and to the
design and synthesis of 3′-chloro-5,7-dihydroxyisofla-
* To whom correspondence should be addressed. Dr. Peter Traxler:
phone, +41 61/696 52 86; fax, +41 61/696 34 29. Dr. Pascal Furet
(molecular modeling): phone, +41 61/696 7990; fax, +41 61/696 2761.
† NOVARTIS Ltd.
‡ University of Basel.
10.1021/jm980551o CCC: $18.00 © 1999 American Chemical Society
Published on Web 02/26/1999

Design of EGFR Tyrosine Kinase Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 6 1019
Ch a r t 1
This pharmacophore model was fully consistent with
a model of the EGFR PTK constructed by homology to
the X-ray crystal structure of the cyclic-AMP-dependent
protein kinase.25
Several flavonoid compounds have been reported to
possess protein kinase inhibitory activity. The most
prominent are genistein (1), quercetin (2), and flavopiri-
dol (3a ) (Chart 1). Due to its substantial potency
(micromolar) in inhibiting the EGFR PTK, we consid-
ered 1 as a possible lead structure, and we were
therefore interested in modeling its kinase binding
mode. The three compounds 1-3 share a 5-hydroxy-
chromenone substructure suggesting similar binding
modes. X-ray crystal structures of quercetin (2) in
complex with the Hck tyrosine kinase26 and of the
deschloro analogue 3b of flavopiridol in complex with
CDK227 have recently been determined. These struc-
tures show that the inhibitors make the same hydrogen
bonds with the hinge region of the kinase as the N1 and
N6 nitrogens of ATP which is in agreement with our
pharmacophore model. In the flavopiridol analogue 3b,
these hydrogen bonds involved the keto function as an
acceptor and the 5-hydroxy group of the chromenone
system as a donor. In quercetin (2), the keto function
also serves as a hydrogen bond acceptor; however, in
this case the donor is the 3-hydroxy group. As a
consequence, the region of the binding site occupied by
the 2-phenyl substituent of the chromenone moiety is
not the same for these two inhibitors (Figure 1). Their
binding modes differ despite the fact that they belong
to the same chemical class. On the basis of this
information, we attempted to dock genistein (1) in the
ATP-binding site of our EGFR kinase model, giving its
5-hydroxychromenone moiety the same two orientations
as those seen in the above complexes. In both cases, we
observed severe steric clashes between the p-hydroxy-
phenyl moiety in the 3-position and protein residues of
the hinge region, suggesting the existence of a third
binding mode specific to genistein.
vone as well as 3-aryl-4(1H)-quinolones as potent and
selective EGFR tyrosine kinase inhibitors.
Design of In h ibitor s
In previous work, we proposed a pharmacophore
model for the inhibitors of the EGFR PTK that are ATP-
competitive.10,18,19,22 Using this model, we showed that
potent inhibitors could be obtained with molecules
presenting the following structural features:
sHydrogen-bonding groups mimicking the N1 and N6
nitrogens of the adenine ring of ATP which are engaged
in hydrogen bonds with the peptide backbone of two
residues (Gln 767 and Met 769) of the hinge region of
the EGFR kinase (dual hydrogen bond donor-acceptor
22
system).
sA chlorophenyl moiety occupying the region of the
protein which is binding the ribose ring of ATP (“sugar
pocket”). In addition, a sulfur-aromatic interaction
between this moiety and residue Cys 773 of the pocket
is assumed.18
sAnd optionally, groups filling a hydrophobic pocket
extending in the direction of the lone pair of the N7
nitrogen of the adenine moiety of ATP. This pocket is
present but not occupied in most protein kinases.
A clue to the possible nature of this third binding
mode came by comparing genistein (1) to the (phenyl-
amino)quinazoline inhibitor class (compound 4) reported
by Zeneca and Parke-Davis.12,13 In the literature, we
found an example of enzyme inhibitors where a
F igu r e 1. Binding modes of deschloroflavopiridol (3b) (white) and quercetin (2) (yellow) as compared to ATP (red). Hydrogen
bonds appear as blue lines.

1020 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 6
Traxler et al.
pyrimidine ring of 4 gives a very good overall superim-
position of the two molecules. In particular, the anilino
moiety of 4 superimposes nicely on the p-hydroxyphenyl
group of genistein. Given the importance of the anilino
m-chloro substituent in conferring high EGFR PTK
inhibitory activity to the quinazoline derivatives,13 as
well as to the other classes of (phenylamino)pyrimidine
inhibitors (e.g., (phenylamino)pyrrolopyrimidines18 and
(phenylamino)pyrazolopyrimidines19), the overlay shown
in Figure 2 immediately suggested that replacing the
p-hydroxyphenyl moiety of genistein by a m-chlorophen-
yl ring would enhance its potency. This idea motivated
the synthesis of compound 6 as a first target structure.
A representation of the putative binding mode of 6
in the ATP-binding site of our EGFR kinase model is
shown in Figure 3. Consistent with our pharmacophore
model, we assumed that the chlorophenyl ring fits in
the “sugar pocket” making a sulfur-aromatic interac-
tion with Cys 773. Mimicking ATP, the 5-hydroxy
substituent accepts a hydrogen bond from the backbone
NH of residue Met 769, while the 7-hydroxy substituent
donates a hydrogen bond to the backbone carbonyl of
Gln 767. It can be seen that in this binding mode, the
orientation of the chromenone moiety of 6 is such that
oxygen O1 approximately occupies the region of the
binding site corresponding to the C8 atom of the adenine
ring of ATP. This oxygen atom is thus facing the part
of the cavity binding the triphosphate chain of ATP.
On the basis of this putative binding mode, we
suggested that replacement of O1 by a nitrogen atom
would allow the attachment of groups that may give rise
to beneficial interactions with amino acids of the tri-
Figu r e 2. Superimposition of quinazoline 4 (red) with genistein
(1) (yellow). The intramolecular hydrogen bond present in
genistein appears in blue.
salicylic group has been reported to be a successful
bioisosteric replacement of a quinazoline moiety.28
Thereby, the pyrimidine ring mimics the pseudo-six-
membered ring resulting from the hydroxy-keto in-
tramolecular hydrogen bond in salicylic acid. A similar
hydroxy-keto function is present in genistein (1). As-
suming the same bioisosteric relationship between the
quinazoline-type EGFR kinase inhibitors and genistein
(1) led to the hypothesis that they share a common
binding mode. In Figure 2 it can be seen that overlap-
ping the 5-hydroxy-keto system of genistein (1) and the
F igu r e 3. Putative binding mode of compound 6: compound 6 (yellow) superimposed on ATP (red). Hydrogen bonds appear as
blue lines.

Design of EGFR Tyrosine Kinase Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 6 1021
Sch em e 1. Synthesis of Isoflavonesa
a
Reagents and conditions: (a) 3-chlorobenzyl cyanide, ZnCl2, HCl, ether, room temperature, 4 h; (b) CH3SO2Cl, DMF, BF3‚O(C2H5)2,
100 °C, 2 h; (c) diazomethane, methanol; (d) 3-chlorophenylacetyl chloride, SnCl4, room temperature, 18 h.
phosphate binding part of the cavity. The quinolone
derivatives 11-15 were synthesized following this
concept.
Biologica l Eva lu a tion a n d Discu ssion
Genistein (1), the isoflavones 6-9, and the quinolones
11-15 and 17 were tested against the intracellular
domain of the EGFR tyrosine according to assay condi-
tions described in detail in the literature (Table 1).31
As predicted, the target compound 6 (IC50 ) 95 nM) was
approximately 10 times more potent than 1. The 7-meth-
oxy derivative 7 (IC50 ) 1.27 µM) was less active than
6 indicating that modifications at this position decrease
potency. In the model, the 7-hydroxy substituent of 6
donates a hydrogen bond to the backbone carbonyl of
residue Gln 767, which is no longer possible with a
methoxy group. Abolishing this interaction seems to be
detrimental for kinase inhibition. The 5,7-dimethoxy
derivative 8 and compound 9, where the 5-hydroxy
group is missing, were inactive (IC50 > 100 µM), thus
proving that the 5-hydroxy function is essential for
biological activity. We assumed that this substituent
accepts an important hydrogen bond from residue Met
769 of the kinase. The loss of activity observed upon its
deletion in compound 9 was thus expected. With regard
to the replacement of the 5-hydroxy group by a methoxy
group, it was expected to have a more dramatic effect
on the activity than in the case of the 7-hydroxyl
substituent. In the 7-position, the CH3 of the methoxy
group can be oriented in the opposite direction to the
7-hydroxy proton of 6 forming the hydrogen bond, thus
avoiding a steric clash with the backbone carbonyl of
Gln 767. In addition, it is conceivable that the methoxy
oxygen atom can still accept a hydrogen bond from the
side chain of the adjacent Thr 766 residue. In contrast,
due to the presence of the proximal keto function that
imposes a steric restriction, a methoxy substituent in
the 5-position must have its methyl group oriented
toward the backbone NH of residue Met 769, therefore
causing a steric clash with the protein.
Ch em istr y
Syn th esis of Isofla von es. 3′-Chloro-5,7-dihydroxy-
isoflavone (6) was synthesized according to a published
procedure (Scheme 1).29 Reaction of phloroglucinol with
3-chlorobenzyl cyanide by the Houben-Hoesch method
gave trihydroxyphenyl benzyl ketone 5a .30 5a was
converted to the isoflavone 6 by Vilsmeier-Haack type
reaction in 95% yield. The monomethyl ether 7 and
dimethyl ether 8 were obtained by reaction of 6 with
diazomethane in methanol. In analogy to 6, 3′-chloro-
7-hydroxyisoflavone (9) was obtained from 5b via Vils-
meier-Haack reaction.
Syn th esis of 3-Ar yl-4(1H)-qu in olon es. The synthe-
sis of 3-aryl-4(1H)-quinolones was based on the Vils-
meier-Haack reaction as used for the synthesis of the
isoflavones. The target compound 3-(3-chlorophenyl)-5-
hydroxy-7-methoxy-4(1H)-quinolone (11) was synthe-
sized in two steps starting from N-(3,5-dimethoxyphenyl)-
acetamide (obtained from 3,5-dimethoxyaniline by
acetylation) (Scheme 2). Friedel-Crafts alkylation of the
acetamide with (3-chlorophenyl)acetyl chloride gave the
ketone intermediate 10 in 65% yield (together with its
regioisomer). Ring closure of the ketone 10 under
modified Vilsmeier-Haack conditions (POCl3, DMF)
directly gave the desired quinolone 11 in 42% yield. 5,7-
Dihydroxy-4(1H)-quinolone 12 was obtained from 11 by
ether cleavage with boron tribromide in toluene.
The N-alkylated quinolones 13-15 were obtained by
reaction of 11 with the corresponding alkyl halide and
potassium carbonate in DMF. Alkylation of 12 under
the same conditions failed. In a similar way as described
for compound 11, 3-(3-chlorophenyl)-4(1H)-quinolone
(17) was synthesized in two steps by Friedel-Crafts
alkylation of aniline with 3-chlorbenzyl cyanide to the
ketone 16, followed by ring closure to the final quinolone
17.
Replacement of the oxygen in the chromane ring
system of compound 6 by a nitrogen (compound 12)
further improved the inhibitory activity against the
EGFR kinase by a factor of more than 10. With an IC50

1022 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 6
Sch em e 2. Synthesis of Quinolonesa
Traxler et al.
a
Reagents and conditions: (a) 3-chlorophenylacetyl chloride, SnCl4, room temperature, 3 h; (b) POCl3, DMF, 72 h, 80 °C; (c) PBr3
toluene, reflux, 18 h; (d) alkyl halide, K2CO3, DMF, room temperature; (e) 3-chlorobenzyl cyanide, BCl3, AlCl3; CH3SO2Cl, DMF,
BF3‚O(C2H5)2, 80 °C, 3 h.
Ta ble 1. EGFR Tyrosine Kinase Activity and in Vitro Selectivity of Derivatives
compd
R1
R2
R3
formula
EGFRa
v-Abla
c-Srca
PKC-Ra
1
4
6
7
8
1.0
0.006
0.095
1.27
500
58.8
0.038
0.008
0.170
4.0
>10
>10
>100
>100
>100
>100
nt
0.125
2.70
OH
OH
C
15H9ClO4
20
>100
nt
>100
>100
nt
nt
>100
8.0
>10
nt
OH
OCH3
H
OH
OH
OH
OH
OH
H
OCH3
OCH3
OH
OCH3
OH
OCH3
OCH3
OCH3
H
C16H11ClO4
C17H13ClO4
9
C
15H9ClO3
C16H12ClNO3
15H10ClNO3
nt
nt
11
12
13
14
15
17
>100
>10
>10
nt
>10
nt
>100
>100
>100
nt
C
CH3
(CH2)2C6H5
CH2COOCH3
C17H14ClNO3
C24H20ClNO3
C19H16ClNO5
0.064
>100
>10
nt
>100
C
15H10ClNO
nt
a
Expressed in IC50 values (µM). Average of at least three determinations with <30% variability.
value of 8 nM, this compound was the most potent one
of this series and was under our assay conditions as
potent as the quinazoline 4. As already observed in the
isoflavone series, the 7-methoxyquinolone 11 (IC50 ) 38
nM) was less active than compound 12. Docking of
compound 12 into the EGFR PTK model according to
our binding mode hypothesis suggests the possibility of
a direct or water-mediated hydrogen bond between the
quinolone NH group of 12 and amino acid Asp 831 of
the protein depending on the rotameric state chosen for
the side chain of the latter. Again, and in accordance
with our model, removal of the 5-hydroxy function
together with the 7-hydroxy group (compound 17) led
to a complete loss of activity (IC50 > 100 µM).

Design of EGFR Tyrosine Kinase Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 6 1023
Ta ble 2. Cellular Activity of Compounds
relationships. In both cases, the gain in binding affinity
produced by a chloro or bromo substituent in the meta
position of the anilino moiety is explained by a sulfur-
aromatic interaction with a cysteine residue of the
protein (Cys 773 in our model and Cys 751 in the Parke-
Davis model). It is our conviction that depending on the
substitution pattern and the nature of the substituents,
the quinazoline and related inhibitors can adopt either
one of the two binding modes. X-ray crystal structures
of CDK2 complexed with different ligands have clearly
shown that molecules belonging to the same chemical
class (adenine derivatives) can bind to the kinase in
different orientations depending on the substitution
pattern.32 For the isoflavone and quinolone inhibitors
described in this work, a better interpretation of the
structure-activity data is obtained assuming that they
adopt the binding mode of our model, but the compounds
can also be docked in the EGFR kinase model according
to Parke-Davis’ binding mode without producing steric
clashes and structural inconsistencies.
inhibition of
inhibition of
EGF-dependent PDGF-dependent inhibition of MK
compd phosphorylationa phosphorylationa cell proliferationa
1
4
6
7
0.12
12.0
>100
>100
6.0
>10
>10
>10
>10
>10
>10
>10
>10
>10
<0.1
16.0
45.2
46.2
11.4
10.2
2.83
4.93
9.55
8
11
12
13
14
15
2.0
3.0
1.5
12.5
a
Expressed in IC50 values (µM). Average of at least three
determinations with <30% variability.
According to the proposed binding mode, the nitrogen
of 12 points into the direction of the triphosphate region
of ATP. Therefore, this nitrogen should be susceptible
for modifications if one assumes a certain flexibility of
Asp 831, the amino acid of the protein most proximal
to this position. Although the activity against the EGFR
kinase could not be improved by N-alkylation (com-
pounds 13-15), the best compound 15 of this series had
an IC50 value of 64 nM and was more potent than the
simple methyl compound 13 (IC50 ) 170 nM), indicating
that there is a chance to further improve the potency
with a more polar substituent at the nitrogen, capable
of forming an additional interaction with an amino acid
in the triphosphate region.
Exp er im en ta l Section
Kin a se Assa ys. Purification of protein kinases and in vitro
enzyme tests were performed as previously described33,34 using
poly(E4Y) (2 µg/mL) and [33P]ATP (0.4 µM) as sustrates. All
compounds were dissolved in DMSO and diluted in buffer,
giving a final DMSO concentration of 1% in the assay. IC50
values represent averages of at least three determinations.
Genistein (IC50 ) 1 µM) and the dianilinophthalimide CGP
52411 (IC50 ) 1 µM) served as an internal standard inhibitor
in all EGFR kinase assays.
In h ibition of Cellu la r Tyr osin e P h osp h or yla tion . In-
hibition of EGF- and PDGF-stimulated total cellular tyrosine
phosphorylation in A431 cells and BALB/c 3T3 cells, respec-
tively, was measured using a microtiter ELISA assay as
previously reported.28,29 Under our assay conditions, EGF or
PDGF stimulation caused a 3-4-fold increase in tyrosine or
substrate phosphorylation.
An tip r olifer a tive Assa ys. Assays were performed es-
sentially as previously described.35 Drug (in a final DMSO
concentration of 0.5%) was added 24 h after plating, and
growth was monitored after 3-5 days of incubation using
methylene blue staining. Cell growth of nonadherent cells was
monitored by using the calorimetric MTT assay. IC50 was
defined as the drug concentration causing a 50% reduction of
signal as compared to control cultures which contain indentical
solvent concentrations (100%). IC50 values represent the mean
of at least two independent assays using serial dilutions of
the drug diluted with culture medium.
All tested compounds showed high selectivity against
other tyrosine kinases (v-Abl, c-Src) and against the
serine/threonine kinase PKC-R.
When the inhibition of phosphorylation was tested on
a cellular level, only the quinolone 14 (IC50 ) 1.5 µM)
and to a certain extent compounds 11 (IC50 ) 6 µM), 12
(IC50 ) 2 µM), and 13 (IC50 ) 3 µM) showed satisfactory
but not exciting inhibitory activities. However, all four
compounds did not inhihbit PDGF-dependent phospho-
rylation (IC50 values > 10 µM), thereby indicating
selectivity for the EGF-mediated signal transduction
pathway. None of the compounds were very potent in
the inhibition of the proliferation of the EGF-dependent
mouse kerationcyte MK cell line. The best compounds
13 and 14 had IC50 values of 2.83 and 4.9 µM, respec-
tively (Table 2). On the basis of these results, none of
the compounds were further tested in vivo.
Syn th esis. Elemental analyses were within (0.4% of the
theoretical value. 1H NMR spectra were recorded on a Varian
Gemini 300 spectrometer. The chemical shifts are reported in
parts per million (ppm) downfield from tetramethylsilane
(TMS). Splitting patterns are designated as follows: s, singlet;
brs, broad singlet; d, doublet; t, triplet; q, quartet; m, multiplet.
13C NMR spectra were recorded on a Varian Gemini 300
spectrometer. Chemical shifts are reported in ppm downfield
from TMS. Multiplicities observable with off-resonance decou-
pling were obtained from 1H,13C shift correlated spectra
(HETCOR) and are denoted with s, d, t, q (see above); the
number of hydrogen atoms bound directly to carbon as
obtained from attached proton test experiments is denoted
with e, even; o, odd. Mass spectra (MS) were recorded on a
VG 70-250 mass spectrometer. Analytical thin-layer chroma-
tography (TLC) was carried out on precoated plates (silica gel,
60 F-254, Merck), and spots were visualized with UV light or
iodine. Flash chromatography was carried out with Merck
silica gel 60 (40-63 µm) and Chemische Fabrik Uetikon silica
gel (30-70 µm). HPLC was carried out using a Waters system
(Waters controller 600 and Waters photodiode array detector)
The activity profiles of the isoflavones 6 and 7 and
especially of the quinolones 11 and 12 show that our
pharmacophore model of the ATP-binding site together
with additional rational approaches can successfully be
used for the design of novel classes of highly potent and
selective EGFR tyrosine kinase inhibitors.
It is implicit in what is described above that we
assumed that the quinazoline inhibitor 4 conforms to
our pharmacophore model (its N1 atom accepts a
hydrogen bond from Met 769, while its anilino moiety
occupies the sugar pocket). Parke-Davis scientists have
proposed a different binding mode for the quinazoline
inhibitors.23 In their model the N1 atom still makes the
hydrogen bond with Met 769; however, the anilino
moiety fits in the large hydrophobic pocket extending
in the direction of the lone pair of the N7 atom of the
adenine moiety of ATP. We think that both models can
account for most of the existing structure-activity

1024 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 6
Traxler et al.
with Millenium 2010 software. Melting points were measured
on a Kofler block and are uncorrected.
(DMSO-d6) δ 12.76 (s, 1H), 8.59 (s, 1H), 7.68 (t, 1H), 7.56-
7.46 (m, 3H), 6.70 (d, 1H), 6.44 (d, 1H), 3.88 (s, 3H); 13C NMR
(DMSO-d6) 179.6 (e), 165.4 (e), 161.6 (e), 157.4 (e), 155.9 (o),
132.8 (e), 129.9 (o), 128.5 (o), 126.9 (o), 127.5 (o), 121.0 (e),
105.3 (e), 98.2 (o), 92.6 (o), 56.1 (o); EIMS m/z 302 (M+). Anal.
(C16H11ClO4) C, H, N, O.
2-(3-Ch lor op h en yl)-1-(2,4,6-t r ih yd r oxyp h en yl)et h a -
n on e (5a ). An amount of 9.2 g (72.9 mmol) of phloroglucinol
(1,3,5-trihydroxybenzene) and 11.05 g (72.9 mmol) of 3-chlo-
robenzyl cyanide were dissolved in 70 mL of anhydrous ether.
Zinc chloride was added, and a stream of dry hydrogen chloride
was bubbled through the reaction mixture for 4 h. A viscous
yellowish mass has separated from the slightly brown solution.
The flask was left closed overnight, and then the etheral layer
was removed by decantation; 150 mL of 2 M HCl was added,
and the partly dissolved mass was heated to 100 °C for 2.5 h.
The hot solution was filtered and the yellow solid collected. A
lower second phase formed in the filtrate as a brown oil, which
crystallized upon cooling to give 14.05 g of the crude product
(69% yield). Crystallization from ethyl acetate/pentane gave
5a as light-yellow crystals: mp 182.6-184.7 °C; 1H NMR
(DMSO-d6) δ 12.25 (brs, 2H), 10.52 (brs, 1H), 8.49 (s, 1H),
7.35-7.19 (m, 4H), 5.900 (s, 1H), 4.898 (s, 1H), 4.40 (s, 2H);
13C NMR (DMSO-d6) 201.5 (e), 165.0 (e), 164.3 (e, 2 C), 138.4
(e), 132.7 (e), 129.74 (o), 129.66 (o), 128.4 (o), 126.2 (o), 103.7
(e), 94.8 (o, 2 C), 48.6 (e); CIMS (NH3) m/z 279 (M + H)+. Anal.
(C14H11ClO4) N; C: calcd, 60.34; found, 61.50. H: calcd, 3.98;
found, 3.27.
2 -(3 -Ch lor op h e n yl)-1 -(4,6 -d ih yd r oxyp h e n yl)e t h a -
n on e (5b). To a solution of 1.3 mL (11.20 mmol) of stannic
chloride in 12 mL of 1,2-dichloroethane was added dropwise
at room temperature a solution of 695 mg (3.68 mmol) of
3-chlorophenylacetyl chloride in 12 mL of 1,2-dichloroethane;
385 mg (3.5 mmol) of resorcinol was added in small portions.
The solution was heated until all added resorcinol had dis-
solved and was stirred at room temperature overnight. The
solution was poured onto ice and the organic phase diluted
with methylene chloride. The organic phase was washed with
water, dried, and evaporated. The reddish oily residue was
purified by flash chromatography (toluene/ethyl acetate, 9:1)
giving 366 mg (40% yield) of 5b which was used without
further purification: 1H NMR (DMSO-d6) δ 12.61 (s, 1H), 10.91
(brs, 1H), 8.13 (d, 1H), 7.57-7.43 (m, 4H), 6.63 (dd, 1H), 6.51
(d, 1H), 4.54 (s, 2H); 13C NMR (DMSO-d6) 201.2, 165.1, 164.5,
137.6, 133.4, 132.9, 130.0, 129.6, 128.5, 126.6, 112.3, 108.4,
102.6, 43.6.
3′-Ch lor o-5,7-d ih yd r oxyisofla von e (6). An amount of
5.06 g (18.16 mmol) of the ketone 5a was dissolved in 40 mL
of DMF; 14 mL (109 mmol) of boron trifluoride etherate was
carefully added; 4.5 mL (54.5 mmol) of methanesulfonyl
chloride, dissolved in 40 mL of DMF, was added to the ketone-
etherate mixture. The red solution was heated at 100 °C for 2
h and then poured onto ice, thereby forming a yellowish
mixture which gave an off-white precipitate after a few hours.
The precipitate was collected by filtration, dissolved in ethyl
acetate, and washed with water. The organic phase was dried
over sodium sulfate and evaporated to dryness yielding a red
oil which was purified by flash chromatography using toluene/
ethyl acetate (6:4) as eluent; 5.15 g (95% yield) of the isoflavone
6 were obtained which after crystallization from ethanol/water
gave fine off-white needles: mp 182.4-184.5 °C; 1H NMR
(DMSO-d6) δ 12.76 (s, 1H), 10.95 (brs, 1H), 8.49 (s, 1H), 7.67
(s, 1H), 7.57-7.46 (m, 3H), 6.42 (d, 1H), 6.25 (d, 1H); 13C NMR
(DMSO-d6) 179.4 (e), 164.5 (e), 161.9 (e), 157.5 (e), 155.5 (o),
132.9 (e), 132.8 (e), 129.9 (o), 128.5 (o), 127.8 (o), 127.5 (o),
120.8 (e), 104.3 (e), 99.2 (o), 93.8 (o); EIMS m/z 288 (M+). Anal.
(C15H9ClO4) C, H, N, O.
3′-Ch lor o-5-h yd r oxy-7-m et h oxyisofla von e (7) a n d 3′-
Ch lor o-5,7-d im eth oxyisofla von e (8). An amount of 0.12 g
(0.426 mmol) of 6 was dissolved in 6 mL of methanol and cooled
to -20 °C; 10 mL of an etheral solution of diazomethane was
added, and the reaction mixture stirred for 4 days at -14 °C.
The solution was evaporated to dryness and the residue
dissolved in ethyl acetate/water. The organic phase was
washed with water, dried, and evaporated. The crude yellow
solid was purified by flash chromatography (toluene/ethyl
acetate, 6:4) to give 0.11 g (85% yield) of 7. Crystallization from
toluene gave yellowish crystals: mp 147.5-150.2 °C; 1H NMR
A total of 8.4 mg (6% yield) of 8 of mp 107-112 °C was
isolated as a byproduct: EIMS m/z 316 (M+), corresponding
to C17H13ClO4; 1H NMR (DMSO-d6) δ 8.31 (s, 1H), 7.60 (s, 1H),
7,48-7.32 (m, 3H), 6.64 (d, 1H), 6.50 (d, 1H), 3.87 (s, 3H), 3.84
(s, 3H); 13C NMR (DMSO-d6) 173.1 (e), 163.7 (e), 160.8 (e),
159.1 (e), 152.2 (o), 134.3 (e), 132.6 (e) 129.7 (o), 128.7 (o), 127.5
(o), 127.4 (o), 123.5 (e), 108.8 (e), 96.2 (o), 92.9 (o), 56.0 (o) 55.9
(o). Anal. (C17H13ClO4) H, N, O; C: calcd, 64.46; found, 63.64.
3′-Ch lor o-7-h yd r oxyisofla von e (9). An amount of 0.42
mL (3.30 mmol) of boron trifluoride etherate was added
dropwise to a solution of 144 mg (0.55 mmol) of the ketone 5b
in 1.3 mL of DMF. Then, 0.13 mL (1.65 mmol) of methane-
sulfonyl chloride in 1.3 mL of DMF was added dropwise. The
reaction mixture was heated at 100 °C for 2 h and then
hydrolyzed by pouring onto ice. After standing overnight, a
yellow precipitate had formed which was filtered off (Celite).
The residue was dissolved in ethyl acetate; the organic solution
was washed with water, dried, and evaporated. The yellowish
oily residue was purified by flash chromatography (toluene/
ethyl acetate, 9:1) to give 85 mg (57% yield) of 9 as yellowish
crystals: mp 243.5-246.5 °C; 1H NMR (DMSO-d6) δ 10.90 (brs,
1H), 8.49 (s, 1H), 7.99 (d, 1H), 7.68 (s, 1H), 7.57-7.38 (m, 3H),
6.97 (dd, 1H), 6.90 (d, 1H); 13C NMR (DMSO-d6) 174.0, 162.7,
157.3, 154.4, 134.2, 132.6, 129.9, 128.5, 127.5, 127.4, 127.2,
122.0, 116.4, 115.3, 102.1; EIMS m/z 272 (M+), corresponding
to C15H9ClO3.
2-(3-Ch lor op h en yl)-1-(6-a ceta m id o-2,4-d im eth oxyp h e-
n yl)eth a n on e (10). An amount of 0.27 g (1.39 mmol) of N-(3,5-
dimethoxyphenyl)acetamide (obtained from 3,5-dimethoxy-
aniline by acetylation with acetic anhydride) was partially
dissolved in 50 mL of 1,2-dichloroethane and cooled to 0 °C;
0.33 mL (2.77 mmol) of stannic chloride was added. Finally,
0.315 g (1.66 mmol) of (3-chlorophenyl)acetyl chloride was
added over 15 min at room temperature. The yellow-brown
reaction mixture was stirred at room temperature for 3 h, then
poured onto ice, and extracted with methylene chloride. The
organic phase was washed with water, dried, and evaporated
to dryness to yield a brown oil which was purified by flash
chromatography (toluene/ethyl acetate, 6:4) to yield 0.31 g
(65% yield) of crude yellowish product. Crystallization from
toluene gave off-white needles of 10: mp 110-112.5 °C; 1H
NMR (CDC13) δ 11.42 (brs, 1H), 7.97 (d, 1H), 7.27-7.20 (m,
3H), 7.07 (dd, 1H), 6.18 (d, 1H), 4.22 (s, 2H), 3.85 (s, 6H), 1.26
(s, 3H); 13C NMR (CDCl3) 201.5 (e), 169.6 (e), 164.8 (e), 161.9
(e), 143.5 (e), 137.9 (e), 134.0 (e), 129.8 (o), 129.5 (o), 127.8 (o),
126.8 (o), 108.2 (e), 97.1 (o), 94.1 (o), 55.6 (o), 55.5 (o), 50.9 (e),
25.7 (o); EIMS m/z 347 (M+). Anal. (C18H18ClNO4) C, H, N, O.
The isomeric 2-(3-chlorophenyl)-1-(4-acetamido-2,6-dimethoxy-
phenyl)ethanone was isolated as a byproduct (yellow needles
from toluene of mp 154.3-157.1 °C).
3-(3-Ch lor op h en yl)-5-h yd r oxy-7-m et h oxy-4(1H )-q u i-
n olon e (11). DMF (1 mL) was cooled to 10 °C, 0.74 mL (9.67
mmol) of phosphorus oxychloride was added, and the reaction
mixture was stirred at room temperature for 30 min, thereby
developing a deep-pink color; 2.80 g (8.06 mmol) of 10 in 2
mL of DMF was added to the DMF-POCl3 complex, upon
which the reaction mixture turned deep red. The mixture was
stirred for l h at room temperature and then for 72 h at 80 °C,
cooled to room temperature, and finally poured onto ice. The
dark-brown residue was purified by flash chromatography
(toluene/ethyl acetate, 6:4) yielding 1.02 g of crude product
(42% yield). Crystallization from DMF gave fine yellow needles
11: mp 268.5-271.6 °C; 1H NMR (DMSO-d6) δ 14.75 (s, 1H),
12.40 (brs, 1H), 8.25 (s, 1H), 7.81 (t, 1H), 7.66 (dd, 1H), 7.47-
7.34 (m, 2H), 6.46 (d, 1H), 6.23 (d, 1H), 3.82 (s, 3H); 13C NMR
(DMSO-d6) 178.8 (s), 163.6 (s), 162.9 (s), 141.4 (s), 139.7 (d),
136.8 (s), 132.6 (s), 129.7 (d), 127.9 (d), 126.9 (d), 126.4 (d),

Design of EGFR Tyrosine Kinase Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 6 1025
116.4 (s,), 108.3 (s), 96.7 (d), 89.9 (d), 55.4 (q); EIMS m/z 301
at 80 °C and extracted with dichloromethane. The organic
phase was washed with 1 M NaOH, dried, and evaporated to
yield 8.4 g of a yellow oil, which consisted of a mixture of
product 16 and residual 3-chlorobenzyl cyanide, which could
not be separated. The crude product was used for the next step
without further purification: 1H NMR (CDCl3) δ 7.79 (dd, 1H),
7.38-7.10 (m, 5H), 6.69-6.62 (m, 2H), 6.3 (brs, 1H); GCMS
m/z 245 (M+).
(M+). Anal. (C16H12ClNO3) C, H, N, O.
3-(3-Ch lor oph en yl)-5,7-dih ydr oxy-4(1H)-qu in olon e (12).
An amount of 0.02 mL (1.03 mmol) of boron tribromide was
added to a suspension of 81.5 mg (0.27 mmol) of the quinolone
11 in 10 mL of toluene and refluxed for 18 h. The dark-brown
suspension was cooled to 0 °C and methanol added. The
solution was evaporated and the residue triturated with
diethyl ether, which was removed by decantation leaving a
crude brown residue. Purification by flash chromatography
(toluene/ethyl acetate, 6:4) gave 11.7 mg (15.5% yield) of 12
as a yellow solid: mp 145-147 °C; 1H NMR (DMSO-d6) δ 14.70
(s, 1H), 12.26 (d, 1H), 10.32 (s, 1H), 8.16 (d, 1H), 7.79 (t, 1H),
7.66-7.60 (m, 1H), 7.46-7.25 (m, 2H), 6.35 (d, 1H), 6.06 (d,
1H); 13C NMR (DMSO-d6) 178.7, 163.0, 162.4, 141.6, 139.3,
137.0, 132.6, 129.7, 128.0, 126.8, 126.3, 115.9, 107.4, 97.8, 91.7;
EIMS m/z 287 (M+), corresponding to C15H10ClNO3.
3-(3-Ch lor op h en yl)-4(1H)-qu in olon e (17). To the mixture
of the ketone 16 and 3-chlorobenzyl cyanide (2.77 g, 4.76 mmol)
was carefully added boron trifluoride etherate (4.16 mL, 33.1
mmol). Methanesulfonyl chloride (1.28 mL, 16.6 mmol) was
added, and the reaction mixture was heated to 80 °C for 3 h
and then poured onto 1 M NaOH. The precipitate so formed
was collected by filtration, washed with water and toluene,
dispersed in hot hexane, filtered, and dried in vacuo to yield
717.4 mg of crude product. This was adsorbed on silica gel,
which then was washed with toluene/ethyl acetate (3:2) for
removal of the starting materials. Desorption of 17 was
achieved by refluxing the silica gel with ethyl acetate. The
adsorbent was removed by filtration and the filtrate dried and
evaporated. The residue obtained was washed with acetone
to give 310 mg (21% yield) of quinolone 17 as a white powder:
mp 300-301 °C; 1H NMR (DMSO-d6) δ 12.1 (brs, 1H), 8.28 (s,
1H), 8.23 (d, 1H), 7.92 (t, 1H), 7.74-7.59 (m, 3H), 7.47-7.30
(m, 3H); 13C NMR (DMSO-d6) 174.5 (e), 139.2 (e), 138.7 (o),
138.3 (e), 132.5 (e), 131.8 (o), 129.6 (o), 127.9 (o), 126.6 (o),
126.1 (o), 125.9 (e), 125.6 (o), 123.5 (o), 118.3 (o), 117.9 (e);
EIMS m/z 255 (M+). Anal. (C15H10ClNO) C, H, N, O.
3-(3-Ch lor op h en yl)-5-h yd r oxy-7-m et h oxy-N-m et h yl-
4(1H)-qu in olon e (13). An amount of 0.123 g (0.41 mmol) of
3-(3-chlorophenyl)-5-hydroxy-7-methoxy-4(1H)-quinolone (11)
was dissolved in 4 mL of DMF. Then 0.03 mL (0.48 mmol) of
methyl iodide and 0.1 g (1 mmol) of potassium carbonate were
added. The reaction mixture was stirred at room temperature
for 30 min, poured into water, and extracted with ethyl acetate.
The organic phase was dried and the solvent removed in vacuo
to yield 0.12 g (93% yield) of a yellow solid. Crystallization
from toluene gave 13 as fine yellow needles: mp 177.4-183.0
1
°C; H NMR (DMSO-d6) δ 15.27 (s, 1H), 8.40 (s, 1H), 7.80 (t,
1H), 7.70-7.66 (m, 1H), 7.46-7.37 (m, 2H), 6.45 (d, 1H), 6.33
(d, 1H), 3.88 (s, 3H), 3.85 (s, 3H); 13C NMR (DMSO-d6) 178.1
(e), 163.9 (e), 163.9 (e), 145.1 (o), 141.9 (e), 136.5 (e), 132.7 (e),
129.7 (o), 127.9 (o), 126.8 (o), 126.5 (o), 116.1 (e), 108.3 (e),
96.9 (o), 89.7 (o), 55.6 (o), 41.0 (o); EIMS m/z 315 (M+). Anal.
(C17H14ClNO3) C, H, N, O.
Ack n ow led gm en t. We gratefully acknowledge the
contributions of Andreas Wicki and Roland K. Sigel for
the synthesis of compounds 5a , 9, 16, and 17 and the
excellent technical assistance of B. Adam, U. Duerler,
I. Oberkirch, R. Reuter, and R. Roth.
3-(3-Ch lor op h en yl)-5-h yd r oxy-7-m eth oxy-N-(2-p h en yl-
eth yl)-4(1H)-qu in olon e (14). Quinolone 11 (60 mg, 0.198
mmol) was dissolved in 5 mL of DMF. Then, 0.1 mL (0.74
mmol) of 2-phenylethyl bromide and 0.1 g (1 mmol) of potas-
sium carbonate were added. The reaction mixture was stirred
at room temperature for 17 h, then poured into water and
extracted with ethyl acetate. The organic phase was dried and
the solvent removed in vacuo to yield 74 mg of a yellow oil.
Purification by flash chromatography (toluene/ethyl acetate,
9:1) and crystallzation from toluene gave 43 mg (63% yield) of
14 (mp 142.6-144.5 °C) in the form of yellow needles: 1H NMR
(CDCl3) δ 15.11 (s, 1H), 7.34-7.02 (m, 10H), 6.38 (d, 1H), 6.31
(d, 1H), 4.27 (t, 2H), 3.88 (s, 3H), 3.14 (t, 2H); 13C NMR (CDCl3)
179.2 (e), 165.7 (e), 164.6 (e), 142.8 (o), 140.9 (e), 136.9 (e),
135.9 (e), 134.0 (e), 129.3 (o), 129.2 (o, 2 C), 129.0 (o, 2 C),
128.4 (o), 127.6 (o), 127.1, 126.8 (o), 125.3 (o), 118.1 (e), 109.2
(e), 96.4 (o), 89.7 (o), 55.6 (o), 55.5 (e), 34.3 (e); EIMS m/z 405
(M+). Anal. (C24H20ClNO3) C, H, N, O.
Met h yl [3-(3-Ch lor op h en yl)-5-h yd r oxy-7-m et h oxy-4-
oxo-4H-qu in olin -1-yl]a ceta te (15). Prepared in analogy to
13 and 14. From 0.1 g (0.332 mmol) of the quinolone 11 and
0.032 mL (0.332 mmol) of bromoacetic acid methyl ester, 69
mg (52% yield) of 15 was obtained as yellow needles: mp
151.4-156.5 °C (crystallization from toluene); 1H NMR (CDCl3)
δ 14.82 (s, 1H), 7.57-7.40 (m, 3H), 7.22-7.32 (m, 2H), 6.33
(d, 1H), 5.97 (d, 1H), 4.72 (s, 2H), 3.82 (s, 3H), 3.80 (s, 3H);
13C NMR (CDCl3) 179.6 (e), 167.3 (e), 165.2 (e), 164.7 (e), 142.8
(o), 141.6 (e), 135.6 (e), 134.1 (e), 129.5 (o), 128.5 (o), 127.5 (o),
126.8 (o), 119.4 (e), 108.8 (e), 96.7 (o), 89.4 (o), 55.6 (o), 54.4
(e), 53.2 (o); EIMS m/z 373 (M+). Anal. (C19H16ClNO5) C, H,
N, O.
Refer en ces
(1) Aaronson, S. A. Growth Factors and Cancer. Science 1991, 254,
1146-1152.
(2) Fry, D. W.; Kraker, A. J .; Conners, R. C.; Elliott, W. L.; Nelson,
J . M.; Showalter, H. D.; Leopold, W. R. Strategies for the
discovery of novel tyrosine kinase inhibitors with anticancer
activity. Anti-Cancer Drug Des. 1994, 9, 331-351.
(3) Levitzki, A.; Gazit, A. Tyrosine Kinase Inhibition: An Approach
to Drug Development. Science 1995, 267, 1782-88.
(4) Ullrich, A.; Schlessinger, J . Signal Transduction by Receptors
with Tyrosine Kinase Activity. Cell 1990, 61, 203-212.
(5) Elder, J . T.; Fisher, G. J .; Lindquist, P. B.; Bennett, G. L.;
Pittelkow, M. R.; Coffey, R. J .; Ellingsworth, L.; Derynck, R.;
Voorhees, J . J . Overexpression of transforming growth factor R
in psoriatic epidermis. Science 1989, 243, 811-814.
(6) Fry, D. W. Protein tyrosine kinases as therapeutic targets in
cancer chemotherapy and recent advances in the development
of new inhibitors. Exp. Opin. Invest. Drugs 1994, 3, 577-595.
(7) Traxler, P.; Lydon, N. Recent Advances in Protein Tyrosine
Kinase Inhibitors. Drugs Future 1995, 20, 1261-1274.
(8) Spada, A. P.; Myers, M. R. Small molecule inhibitors of tyrosine
kinase activity. Exp. Opin. Ther. Patents 1995, 5, 805-817.
(9) Bridges, A. J . The current status of tyrosine kinase inhibitors:
do the diarylamine inhibitors of the EGF receptor represent a
new beginning? Exp. Opin. Ther. Patents 1995, 5, 1245-1257.
(10) Traxler, P. Protein tyrosine kinase inhibitors in cancer treat-
ment. Exp. Opin. Ther. Patents 1997, 7, 571-588.
(11) Ward, W. H. J .; Cook, P. N.; Slater, A. M.; Davies, D. H.;
Holdgate, G. A.; Green, L. R. Epidermal growth factor receptor
tyrosine kinase. Investigation of catalytic mechanism, structure-
based searching and discovery of a potent inhibitor. Biochem.
Pharmacol. 1994, 48, 659-666.
(12) Fry, D. W.; Kraker, A. J .; McMichael, A.; Ambroso, L. A.; Nelson,
J . M.; Leopold, W. R.; Connors, R. W.; Bridges, A. J . A Specific
Inhibitor of the Epidermal Growth Factor Receptor Tyrosine
Kinase. Science 1994, 265, 1093-1095.
(13) Rewcastle, G. W.; Denny, W. A.; Bridges, A. J .; Zhou, H.; Cody,
D. R.; McMichael, A.; Fry, D. W. Tyrosine Kinase Inhibitors. 5.
Synthesis and Structure-Activity Relationships for 4-(Phenyl-
methyl-amino]-and 4-(Phenylamino)quinazolines as Potent Ad-
enosine 5′-Triphosphate Binding Site Inhibitors of the Tyrosine
Kinase Domain of the Epidermal Growth Factor Receptor. J .
Med. Chem. 1995, 38, 3482-3487.
1-(2-Am in op h en y1)-2-(3-ch lor op h en yl)eth a n on e (16).
Freshly distilled aniline (2.2 mL, 23 mmol) dissolved in 25 mL
of 1,2-dichloroethane was added dropwise to a solution of 25.3
mL (25.3 mmol) of BCl3 in dichloromethane cooled to 0 °C; 5.9
mL (46 mmol) of 3-chlorobenzyl cyanide and 3.37 g (25.3 mmol)
of AlCl3 were added to the white suspension, and the reaction
mixture was stirred at 80 °C for 20 h and cooled to 0 °C; 2 N
HCl was added to the brown mixture, upon which a yellow
precipitate formed. The mixture was then refluxed for 30 min

1026 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 6
Traxler et al.
(14) Rewcastle, G. W.; Palmer, B. D.; Bridges, A. J .; Showalter, H.
D.; Sun, L.; Nelson, J . M.; McMichael, A.; Kraker, A. J .; Fry, D.
W.; Denny, W. A. Tyrosine kinase inhibitors. 9. Synthesis and
evaluation of fused tricyclic quinazolines analogues as ATP site
inhibitors of the tyrosine kinase activity of the epidermal growth
factor receptor. J . Med. Chem. 1996, 39, 918-928.
(15) Thompson, A. M.; Bridges, A. J .; Fry, D. W.; Kraker, A. J .; Denny,
W. A. Tyrosine Kinase Inhibitors. 7. 7-Amino-4-(phenylamino)-
and 7-Amino-4-[(phenyl-methyl)amino-lpyrido[4,3-d]pyrimidines:
A New Class of Inhibitors of the Tyrosine Kinase Activity of
the Epidermal Growth Factor Receptor. J . Med. Chem 1995, 38,
3780-3788.
(16) Rewcastle, G. W.; Palmer, B. D.; Thompson, A. M.; Bridges, A.
J .; Cody, D. R.; Zhou, H.; Fry, D. W.; McMichael, A.; Kraker, A.
J .; Denny, W. A. Tyrosine Kinase Inhibitors. 10. Isomeric 4-[(3-
bromophenyl)-amino]pyrido[d]-pyrimidines are potent ATP bind-
ing site inhibitors of the tyrosine kinase function of the epider-
mal growth factor receptor. J . Med. Chem. 1996, 39, 1823-1835.
(17) Rewcastle, G. W.; Bridges, A. J .; Fry, D. W.; Rubin, J . A.; Denny,
W. A. Tyrosine kinase inhibitors. 12. Synthesis and structure-
activity relationships for 6-substituted- 4-(phenylamino)py-
rimido[5,4-d]pyrimidines designed as inhibitors of the epidermal
growth factor receptor. J . Med. Chem. 1997, 40, 1820-1826.
(18) Traxler, P.; Furet, P.; Mett, H.; Buchdunger, E.; Meyer, T.;
Lydon, N. 4-(Phenylamino)pyrrolopyrimidines: Potent and Se-
lective, ATP Site Directed Inhibitors of the EGF-Receptor
Protein Tyrosine Kinase. J . Med. Chem. 1996, 38, 2285-2292.
(19) Traxler, P.; Bold, G.; Frei, J .; Lang, M.; Lydon, N.; Mett, H.;
Buchdunger, E.; Meyer, T.; Mueller, M.; Furet, P. Use of a
Pharmacophore Model for the Design of EGF-R Tyrosine Kinase
Inhibitors: 4-(Phenylamino)Pyrazolo- Pyrimidines. J . Med. Chem.
1997, 40, 3601-3616.
(24) Trumpp-Kallmeyer, S.; Rubin, J . R.; Humblet, C.; Hamby, J . M.;
Showalter, H. D. H. Development of a Binding Model to Protein
Tyrosine Kinases for Substituted Pyrido[2,3-d]pyrimidine In-
hibitors. J . Med. Chem. 1998, 41, 1752-1763.
(25) Zheng, J .; Knighton, D. R.; Ten Eyck, L. F.; Karlsson, R.; Xuong,
N. H.; Taylor, S. S.; Sowadski, J . M. Crystal Structure of the
Catalytic Subunit of cAMP-Dependent Protein Kinase Com-
plexed with MgATP and Peptide Inhibitor. Biochemistry 1993,
32, 2154-2161.
(26) Sicheri, F.; Moarefi, I.; Kuriyan, J . Crystal structure of the Src
family tyrosine kinase Hck. Nature 1997, 385, 602-609.
(27) Filgueira de Azevedo, W., Jr.; Mueller-Dieckmann, H. J.; Schulze-
Gahmen, U.; Worland, P. J .; Sausville, E.; Kim, S.-H. Structural
basis for specificity and potency of a flavanoid inhibitor of human
CDK2, a cell cycle kinase. Proc. Natl. Acad. Sci. U.S.A. 1996,
93, 2735-2740.
(28) Hodge, C. N.; Pierce, J . A diazine heterocycle replaces a six-
membered hydrogen-bonded array in the active site of scytalone
dehydratase. Bioorg. Med. Chem. Lett. 1993, 3, 1605-1608.
(29) Bass, R. J . Verfahren zur Herstellung von 4-Oxo-4H-benzopy-
randerivaten, 4- Oxo-4H-benzopyranderivate und ihre Verwen-
dung in Futtermitteln. Deutsches Patentamt, Offenlegungss-
chrift 26 40 617, 1977.
(30) Chapman, E.; Stephen, H. Di- and Tri-hydroxydeoxybenzoins.
J . Chem. Soc. 1923, 404-409.
(31) Buchdunger, E.; Trinks, U.; Mett, H.; Regenass, U.; Mu¨ller, M.;
Meyer, Th.; McGlynn, E.; Pinna, L. A.; Traxler, P.; Lydon, N. B.
4,5-Dianilino-phthalimide: A protein-tyrosine kinase inhibitor
with selectivity for the epidermal growth factor receptor signal
transduction pathway and potentin vivo antitumor activity. Proc.
Natl. Acad. Sci. U.S.A. 1994, 91, 2334-2338.
(32) Schulze-Gahmen, U.; Brandsen, J .; J ones, H. D.; Morgan, D. O.;
Meijer, L.; Vesely, J .; Kim, S. H. Multiple Modes of Ligand
Recognition: Crystal Structures of Cyclin-Dependent Protein
Kinase 2 in Complex with ATP and Two Inhibitors, Olomoucine
and Isopentenyladenine. Proteins 1995, 22, 378-391.
(33) Buchdunger, E.; Trinks, U.; Mett, H.; Regenass, U.; Mu¨ller, M.;
Meyer, Th.; Beilstein, P.; Wirz, B.; Schneider, P.; Traxler, P.;
Lydon, N. B. 4,5-Bis(4-fluoro-anilino)-phthalimide: a selective
inhibitor of the EGF receptor signal transduction pathway with
potent in vivo antitumor activity. Clin. Cancer Res. 1995, 1, 813-
821.
(34) Buchdunger, E.; Zimmermann, J .; Mett, H.; Meyer, Th.; Mu¨ller,
M.; Regenass, U.; Lydon, N. B. Selective inhibition of the
platelet-derived growth factor signal transduction pathway by
a protein-tyrosine kinase inhibitor of the 2-phenylaminopyri-
midine class. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 2558-2562.
(35) Meyer, T.; Regenass, U.; Fabbro, D.; Alteri, E.; Roesel, J .;
Mueller, M.; Caravatti, G.; Matter, A. A derivative of stauro-
sporine (CGP 41 251) shows selectivity for protein kinase C
inhibition and in vitro proliferative as well as in vivo antitumor
activity. Int. J . Cancer 1989, 43, 851-856.
(20) Woodburn, J . R.; Barker, A. J .; Gibson, K. H.; Ashton, S. E.;
Wakeling, A. E.; Curry, B. J .; Scarlett, L.; Henthorn, L. R.
ZD1839, an epidermal growth factor tyrosine kinase inhibitor
selected for clinical development. Proc. Am. Cancer Res. 1997,
38, 6633.
(21) Moyer, J . D.; Barbacci, E. G.; Iwata, K. K.; Arnold, L.; Boman,
B.; Cunningham, A.; Diorio, C.; Doty, J .; Morin, M. J .; Moyer,
M. P.; Neveu, M.; Pollack, V. A.; Pustilnik, L. R.; Reynolds, M.
M.; Sloan, D.; Theleman, A.; Miller, P. Induction of Apopotosis
and Cell Cycle Arrest by CP-358 774, an Inhibitor of Epidermal
Growth Factor Receptor Tyrosine Kinase. Cancer Res. 1997, 57,
4838-4848.
(22) Furet, P.; Caravatti, G.; Lydon, N.; Priestle, J .; Sowadski, J .;
Trinks, U.; Traxler, P. Modeling Study of Protein Kinase
Inhibitors: Binding Mode of Staurosporine - Origin of the
Selectivity of CGP 52 411. J Comput.-Aided Mol. Des. 1995, 9,
465-471.
(23) Palmer, B. D.; Trumpp-Kallmeyer, S.; Fry, D. W.; Nelson, J . M.;
Showalter, H. D. H.; Denny, W. A. Tyrosine Kinase Inhibitors.
11. Soluble Analogues of Pyrrolo- and Pyrazoloquinazolines as
Epidermal Growth Factor Receptor Inhibitors: Synthesis, Bio-
logical Evaluation, and Modeling of the Mode of Binding. J . Med.
Chem. 1997, 40, 1519-1529.
J M980551O