Dual irreversible kinase inhibitors: Quinazoline-based inhibitors incorporating two independent reactive centers with each targeting different cysteine residues in the kinase domains of EGFR and VEGFR-2

Article abstract of DOI:10.1016/j.bmc.2007.03.055

A series of 4-dimethylamino-but-2-enoic acid [4-(3,6-dioxo-cyclohexa-1,4-dienylamino)-7-ethoxy-quinazolin-6-yl]-amide derivatives were prepared. These compounds have two independent reactive centers and were designed to function as dual irreversible inhib

Full text of DOI:10.1016/j.bmc.2007.03.055

Bioorganic & Medicinal Chemistry 15 (2007) 3635–3648
Dual irreversible kinase inhibitors: Quinazoline-based
inhibitors incorporating two independent reactive centers with
each targeting different cysteine residues in the kinase domains
of EGFR and VEGFR-2
Allan Wissner,* Heidi L. Fraser, Charles L. Ingalls, Russell G. Dushin, M. Brawner Floyd,
Kinwang Cheung, Thomas Nittoli, Malini R. Ravi, Xingzhi Tan and Frank Loganzo
Chemical and Screening Sciences and Oncology Research, Wyeth Research, 401 N. Middletown Road,
Pearl River, NY 10965, USA
Received 8 November 2006; revised 14 March 2007; accepted 18 March 2007
Available online 23 March 2007
Abstract—A series of 4-dimethylamino-but-2-enoic acid [4-(3,6-dioxo-cyclohexa-1,4-dienylamino)-7-ethoxy-quinazolin-6-yl]-amide
derivatives were prepared. These compounds have two independent reactive centers and were designed to function as dual irrevers-
ible inhibitors of the kinase domains of both Epidermal Growth Factor Receptor (EGFR) and Vascular Endothelial Growth Factor
Receptor-2 (VEGFR-2) where each reactive center targets a different, non-conserved, cysteine residue located in the ATP binding
pocket of these enzymes. The compounds contain a 6-(4-(dimethylamino) crotonamide) Michael acceptor group that targets Cys-
773 in EGFR and a 4-(amino-[1,4]benzoquinone) moiety that targets Cys-1045 in VEGFR-2. In vitro studies indicated that most of
these compounds are relatively potent inhibitors of each enzyme. These inhibitors were compared with reference compounds that
lack one or both of the reactive centers. The relative dependence of the IC50 values on the concentration of ATP used in the assays
suggests that these compounds appear to function as irreversible inhibitors of each kinase.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
lation. In effect, the duration of action of such a drug
will be a function of the rate of enzyme turnover.
There has been a considerable effort to design small-
molecule irreversible inhibitors of kinase targets for
the treatment of cancer.1,2 Compared to conventional
ATP-competitive inhibitors, irreversible kinase inhibi-
tors have a number of both potential and realized
advantages. It is frequently observed that irreversible ki-
nase inhibitors show enhanced potency when compared
to structurally similar conventional inhibitors. There is
also the expectation that an irreversible inhibitor need
not exhibit prolonged circulating blood levels to achieve
a desired biological effect. Once the target enzyme is
deactivated by covalent bond formation, the biological
effect should persist even after the drug leaves the circu-
Furthermore, it has recently been observed that certain
irreversible inhibitors can inhibit the function of kinases
that have mutated and developed resistance to conven-
tional inhibitors. For example, Kwak and coworkers3
have shown that the compounds 1 (EKB-569) and 2
(HKI-272), both 4-anilinoquinoline-3-carbonitrile-based
irreversible inhibitors, designed to covalently bind to a
conserved cysteine residue present in some erbB kinase
family members, retained activity against Erlotinib-
and Gefitinib-resistant cell lines. Moreover, in contrast
to the conventional reversible inhibitors, it proved extre-
mely difficult to induce resistance in cells with 1 and 2.
Similar results were obtained in a related study by Cater
et al. for the irreversible EGFR kinase inhibitors 1 and 3
(CI-1033).4 Additionally, it was observed that the qui-
nazoline-based irreversible EGFR kinase inhibitor 4
(EKI-785) retained inhibitory activity against resistant
cell lines.5–7 In a clinical study, some patients with
non-small cell lung cancer having an acquired resistance
to Gefitinib responded to treatment with 1.8
Keywords: Dual; Irreversible; Kinase; Inhibitors; Quinazoline-based;
Independent reactive centers; Cysteine residues; EGFR; VEGFR-2;
KDR; Quinone; Michael acceptor; Epidermal; Growth; Factor; Rec-
eptor; Vascular; Endothelial.
*
Corresponding author. Tel.: +1 845 602 3580; fax: +1 845 602
5561; e-mail: wissnea@wyeth.com
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.03.055

3636
A. Wissner et al. / Bioorg. Med. Chem. 15 (2007) 3635–3648
When appropriately designed, irreversible inhibitors can
show enhanced selectivity. Cohen et al. have shown that
covalent binding inhibitors based on a pyrrolo[2,3-
d]pyrimidine scaffold showed remarkable specificity
toward kinases of the RSK family.9 It was postulated
that the selectivity exhibited by these compounds was
a result of covalent binding to a conserved cysteine
residue present in members of the family combined with
functionality on the inhibitors intended to interact with-
in the gatekeeper region of the protein. Finally, it has
been suggested that irreversibly binding EGFR inhibi-
tors might make useful molecular imaging agents for
cancer.10,11
inhibitors of VEGFR-2 as exemplified by the quinazo-
line derivative 5.19 For these compounds, we proposed
that covalent bond formation with the enzyme is the re-
sult of a initial redox reaction between the quinone ring
of the inhibitor and Cys-1045 located within the ATP
binding pocket of the enzyme, resulting in the formation
of a semiquinone species and a sulfhydryl radical. The
combination of these radicals resulted in a reductive
addition of the cysteine sulfhydryl group to the quinone
ring. Here, we note that the targeted cysteine in EGFR
is not conserved in VEGFR-2 and that the VEGFR-2
targeted cysteine is not conserved in EGFR. Since both
EGFR and VEGFR-2 inhibitors having a quinazoline
core structure were used in the past for the construction
of both conventional ATP-competitive inhibitors as well
as irreversible inhibitors, we hypothesized that it might
be possible to design molecules, using this scaffold, that
have two different reactive centers, one designed to form
a covalent interaction with the Cys-773 in EGFR and
the other designed to form a covalent interaction with
the Cys-1045 in VEGFR-2. An inhibitor of this type
would then function as an irreversible inhibitor of each
enzyme, but one that targets a different cysteine in each
case. In this work, we will describe the preparation of
some quinazoline derivatives such as 6 that combined
the Michael acceptor functionality of our earlier series
of EGFR inhibitors with the quinone ring of our VEG-
FR-2 inhibitors and we will present our evidence from
in vitro enzyme studies showing that these compounds
appear to function as we propose. To our knowledge,
the preparation of a molecule that has two separate
reactive groups, each designed to target, via a covalent
interaction, a different cysteine residue in unrelated
enzymes, is unprecedented in the literature.
Contrasting with the potential advantages of these cova-
lent binding kinase inhibitors is a possible liability that
they might indiscriminately react with non-target related
proteins. This could be manifested as an increased toxic-
ity. Therefore, it is important, if these covalent binding
agents are to make useful drugs, that they should not
have excessive reactivity and that they should be appro-
priately constructed to have a reasonably good fit at the
active site of the target with the reactive centers oriented
in the proper manner for a covalent interaction to ensue.
Such a drug, for example, one that is designed to react
with an active site cysteine residue, needs to be more
reactive toward the target enzyme than it is toward other
nucleophilic species (such as glutathione (GSH), or
other proteins), which may be present in high concentra-
tions in the circulation or in the cytosol. In designing an
irreversible inhibitor, one can rely on an anticipated
entropic effect wherein the rate of reaction of such an
inhibitor, when bound at the active site with the correct
orientation, will be enhanced relative to the rate of reac-
tion when in solution, due to the fact that the reactive
centers on both the protein and drug will be held in close
proximity. Under these circumstances, the local concen-
tration of the reacting partners will be much higher
when the drug is bound to its target compared to when
the components are free in solution. This phenomenon
has been measured with the irreversible kinase inhibitor
hypothemycin.12
Moreover, one could imagine that combining EGFR
and VEGFR-2 kinase inhibitory activities in a single
molecule could result in a useful entity having anti-can-
cer applications. In the former case, the inhibitor will be
targeting a growth factor receptor that is deregulated,
either by overexpression or mutation, in a number of
malignancies20–23 while in the latter case, the inhibitor
will be targeting a growth factor receptor which the tu-
mor depends upon to build and maintain its vascula-
ture.24–27
In an earlier project, we successfully produced covalent
binding inhibitors of EGFR and Her2 kinases.13–17
These inhibitors had either a quinazoline or quinoline-
3-carbonitrile core structure. Two members of the latter
series, 1 and 2, have entered clinical trials. Other work-
ers have arrived at some similar compounds.18 These
inhibitors are able to covalently interact with the target
proteins by virtue of the fact that they undergo a
Michael addition reaction with a conserved, solvent
exposed, cysteine residue present in these enzymes
(Cys-773 in EGFR and Cys-805 in Her2). We showed
that the 4-(dimethylamino)crotonamide Michael accep-
tor group present on these inhibitors is a particularly
effective one for this purpose. We demonstrated that
the terminal dimethylamino group of this Michael
acceptor serves as an intramolecular catalyst for
Michael addition reactions in solution and we have
proposed that it has a similar role when the inhibitor
is bound at the active site of these enzymes. More re-
cently, we described our work on a series of irreversible
2. Chemistry
The compounds were prepared as shown in Scheme 2.
Nitration of 7,28 as previously described,29 gave a mix-
ture of isomers from which pure 8 could be obtained
after recrystallization from acetic acid. Replacement of
the fluorine with an ethoxy group was accomplished
by heating with sodium ethoxide in ethanol at reflux.
The resulting intermediate
9 was chlorinated by
heating in thionyl chloride at reflux in the presence of
a catalytic quantity of DMF. Displacement of the 4-
chloro group of 10 with 4-chloro-2,5-dimethoxyaniline
was accomplished by heating in isopropanol at reflux.
Reduction of the nitro group of 11 using iron and
acetic acid in methanol gave intermediate 12. The 4-(dim-
ethylamino)crotonamide Michael acceptor group was

A. Wissner et al. / Bioorg. Med. Chem. 15 (2007) 3635–3648
3637
N
F
O
HN
N
Cl
HN
N
Cl
H
N
H
N
CN
CN
N
N
O
O
O
O
1 (EKB-569)
2 (HKI-272)
F
O
HN
Br
HN
Cl
H
N
HN
O
N
N
O
N
N
N
O
3 (CI-1033)
4 (EKI-785)
O
OR
O
O
OR
O
HN
N
H
N
HN
N
N
O
O
N
O
O
O
N
5
6
Scheme 1.
incorporated by the reaction of 12 with the acid chloride
derived from (E)-4-(dimethylamino)crotonic acid. Oxi-
dation of 13 to the quinone derivative 14 was achieved
using ceric ammonium nitrate (CAN) in aqueous aceto-
nitrile at room temperature. We found that 14 is a highly
reactive species and is relatively unstable in the absence
of solvent. Therefore, we found it best to isolate 14 in a
chloroform or methylene chloride solution and to use it
in that manner without additional purification. The
reaction of 14 with various alcohols, in the presence of
triethylamine or sodium phenoxide as catalysts, then
furnished inhibitors 15–24 in modest yields.
Therefore, in the present study, many of the quinone
substituents we incorporated are of that type. Addition-
ally, as will be shown by our molecular modeling results,
such inhibitors set up a favorable interaction between
the phenyl ring of the benzyloxy group and a phenylal-
anine residue that is present in both EGFR and VEG-
FR-2. This likely results in better initial binding
affinity and helps position the quinone ring toward
Cys-1045 of VEGFR-2 with the proper orientation to
initiate a redox reaction.
In addition to preparing these dual irreversible inhibi-
tors, we needed to have available some reference com-
pounds with which to compare their inhibitory
properties. Compound 27 was prepared as shown in
Scheme 3. This compound was prepared assuming,
based on its structure, it would function as an irrevers-
ible EGFR inhibitor and a reversible VEGFR-2 inhibi-
tor. The compound incorporates the Michael acceptor
functional group of our earlier EGFR inhibitors but it
also has a 4-substituent on the quinazoline core that is
known to confer ATP-competitive VEGFR-2 inhibitory
activity on this class of compounds.30,31 Accordingly,
the quinazoline derivative 2532 was acylated with the
acid chloride derived from (E)-4-(dimethylamino)cro-
tonic acid. Heating a mixture of 26 and 4-bromo-2-flu-
oro aniline in a mixture of NMP and isopropanol in
the presence of a catalytic amount of pyridine hydro-
chloride furnished 27.
In our earlier study of the irreversible VEGFR-2 inhib-
itors,19 we surveyed a large variety of substituents
attached to the quinone ring. We showed that the reac-
tivity of the resulting molecules correlated with their
LUMO energy and that there was a tendency for the
potency of the inhibitors to be related to their reactivity.
In this study, we decided to incorporate substituents on
the quinone ring having a narrow spectrum of reactivity
based on the earlier calculations of the LUMO energies
and kinetic studies for related compounds. One group of
compounds that were generally potent inhibitors, but
which had relatively moderate reactivity, were those
with 4-benzyloxy groups on the quinone ring. Further-
more, in our EGFR and Her2 projects,17 we observed
that compounds with 4-(4-benzyloxy-phenylamino)
groups were usually potent inhibitors of those enzymes.

3638
A. Wissner et al. / Bioorg. Med. Chem. 15 (2007) 3635–3648
OH OH
OH
O2N
O2N
ii
i
N
N
N
F
N
F
N
O
N
7
8
9
O
Cl
Cl
N
HN
O
O2N
O
R
vi
iii
iv
N
N
O
N
10
11, R = NO2
12, R = NH2
v
O
Cl
O
O
Cl
O
HN
HN
H
H
N
N
vii
N
N
N
N
O
O
O
N
O
N
14
13
15, R = OCH3
16, R = OCH(CH2F)2
17, R = OCH2Ph-3-F
18, R = OCH2Ph-3,4-di-F
19, R = OCH2Ph
20, R = OCH2-2-pyridyl
21, R = OCH2Ph-3-Cl
22, R = OCH2-2-thiophene
23, R = OCH2Ph-3-OCH3
24, R = OCH2Ph-3-CH3
O
R
O
HN
N
H
viii
N
N
N
O
O
Scheme 2. Reagents and conditions: (i) H2SO4, HNO3, 100 ꢁC, 1 h; (ii) NaOEt, EtOH, reflux, 1.5 h; (iii) SOCl2, DMF (cat), reflux, 3.5 h; (iv) 4-Cl-
2,5-dimethoxyaniline, isopropanol, reflux, 3 h; (v) Fe, HOAc, NaOAc, CH3OH, reflux, 3 h; (vi) 4-dimethylaminocrotonic acid hydrochloride,
(COCl)2, CH3CN, NMP, 55 ꢁC, 20 min; (vii) CAN, CH3CN, H2O; (viii) ROH, CHCl3 or CH2Cl2, Et3N or NaOPh 25–40 ꢁC, 3–5 h.
A number of other reference compounds were used in
this study and are shown in Schemes 1 and 3. Com-
pounds 1 and 2 were optimized to be irreversible inhib-
itors of EGFR/Her2 kinases. The quinazoline 28 is a
known ATP-competitive EGFR inhibitor.33 Compound
29 (Gefitinib)34 is a marketed ATP-competitive EGFR
inhibitor. The phthalazine derivative 30 is reported to
be a selective, reversibly binding, VEGFR-2 inhibitor.35
The quinazoline 31 (ZD6474)36 is a reversible binding
VEGFR-2 inhibitor that is known to also have EGFR
inhibitory properties. Compounds 32–34 are from our
earlier work19 and are irreversible VEGFR-2 kinase
inhibitors.
the X-ray crystal structure of the complex between
Erlotinib and the catalytic domain of EGFR.37 The
ligand was removed from this crystal structure and,
with the exception of the water molecule that bridges
the N3 atom of the quinazoline and Thr-766, the water
molecules were removed. The inhibitor was docked
using the GLIDE38 docking algorithm in the XP (extra
precision) mode. The resulting model shows the inhib-
itor oriented in a manner consistent with the way other
quinazoline derivatives have been shown to bind to
different kinases.37,39
In the most favored bound conformation, the oxygen
atom of a quinone carbonyl group makes an intramole-
cular H-bond to the NH substituent on the quinazoline.
This is similar to the low energy conformations found
for the similar unbound inhibitors based on molecular
mechanics calculations. A favorable interaction between
the quinazoline N1 of 19 and the hinge region backbone
NH of Met-769 is found to be significant in our model.
3. Molecular modeling
A binding model for inhibitor 19 at the active site in
EGFR is presented in Figure 1. The initial protein
coordinates used in this study are those reported in

A. Wissner et al. / Bioorg. Med. Chem. 15 (2007) 3635–3648
3639
Cl
Cl
N
H
H2N
N
i
N
N
N
O
N
26
25
F
Br
HN
H
ii
N
N
N
O
N
27
F
HN
N
Br
O
HN
N
Cl
O
O
N
O
N
N
O
28 (PD153035)
29 (Gefitinib)
F
Br
N
HN
N
O
O
Cl
N
N
N
N
H
N
30 (PTK787)
31 (ZD6474)
O
R
O
32: R = OCH2Ph-3-OCH3
33: R = OCH2Ph
HN
O
N
34: R = OCH2Ph-3-F
O
O
N
Scheme 3. Reagents and conditions: (i) 4-dimethylaminocrotonic acid hydrochloride, CH3CN, NMP, (COCl)2, 55 ꢁC, 30 min; (ii) NMP, 4-Br-2-F-
aniline, PyÆHCl, isopropanol, 85 ꢁC, 30 min.
The polar aromatic hydrogen atom on C2 of the quin-
azoline ring is favorably oriented at approximately
coordinates used in this study are those reported in the
X-ray crystal structure of the catalytic domain of the
enzyme.40 The inhibitor was docked into the active site
as described above. The bound conformation of 19 in
VEGFR-2 is very similar to that found in EGFR (see
Fig. 4) as are the specific interactions it makes with
the protein. In this case, N1 of the quinazoline makes
a hinge region interaction with the NH of Cys-919.
The polar aromatic hydrogen atom on C2 of the quinaz-
˚
2.4 A from the backbone carbonyl of Gln-767. The N3
atom of the quinazoline is H-bonded to the hydroxyl
group of Thr-766 via a water bridge. A favorable face-
to-face p–stacking interaction is evident between the
benzyloxy quinone substituent and the aromatic ring
of Phe-832. Finally, the sulfhydryl group of Cys-773 is
˚
located 6.0 A away from the b-carbon atom of the
Michael acceptor. A 10-picosecond molecular dynamics
simulation of the protein–ligand complex revealed that
˚
oline ring is favorably oriented at approximately 2.2 A
from the backbone carbonyl of Glu-917. As with the
EGFR complex, we observe a favorable interaction
between the benzyloxy quinone substituent of 19 and
the aromatic ring of Phe-1047; however, in this case,
the aromatic rings are oriented in an edge-to-face man-
ner. Significantly, the center of the quinone ring is
˚
this distance approached to within 3.0 A, suggesting
the feasibility of covalent bond formation (see Fig. 2).
A binding model for inhibitor 19 at the active site in
VEGFR-2 is presented in Figure 3. The initial protein

3640
A. Wissner et al. / Bioorg. Med. Chem. 15 (2007) 3635–3648
Figure 4. Overlap of 19 bound to EGFR (green) and VEGFR-2
(yellow) using the coordinated system defined by the overlapped
proteins.
Figure 1. Proposed binding model for 19 bound to the kinase domain
of EGFR.
perpendicularly oriented toward the sulfhydryl group of
˚
Cys-1045, being located about 4.2 A away. A 10-
picosecond molecular dynamics simulation of the pro-
tein–ligand complex revealed that the lone pair on the
sulfhydryl group of Cys-1045 approached the center of
˚
the quinone ring to within 3.7 A (see Fig. 5). Previous
molecular orbital calculations on related quinone-based
inhibitors showed that the orbital density of the LUMO
is largely localized on the quinone ring. It is evident
from these simulations that the orbital overlap between
the LUMO of 19 and an orbital centered on the sulfhy-
dryl group of Cys-1045 should be sufficient to initiate an
electron transfer reaction resulting in a reductive addi-
tion of the sulfhydryl group to the quinone ring.
Interestingly, there is also a cysteine residue, Cys-751,
located in the quinone binding region in the EGFR
complex. In this case, however, the center of the quinone
˚
ring is located about 6.5 A away from the sulfhydryl
group of Cys-751 with the lone pairs of the sulfhydryl
group pointing away from the quinone ring of the
bound inhibitor. The distance changed minimally during
the dynamics simulations. Therefore, because of a less
favorable orientation for orbital overlap and a longer
Figure 2. A view of the dynamics simulation for 19 bound to EGFR
kinase in the region of the Michael acceptor showing the initial
structure (green), intermediate structures (white), and final structure
(orange). Note that a lone pair on the sulfhydryl group of Cys-733
˚
approached the b-carbon atom of the Michael acceptor to within 3.0 A
during this dynamics run.
Figure 5. A view of the dynamics simulation for 19 bound to VEGFR-
2 kinase in the region of the quinone ring showing the initial structure
(green), intermediate structures (white), and final structure (orange).
Note that a lone pair on the sulfhydryl group of Cys-1045 approached
˚
the center of the quinone ring to within approximately 3.7 A during
this dynamics run.
Figure 3. Proposed binding model for 19 bound to the kinase domain
of VEGFR-2.

A. Wissner et al. / Bioorg. Med. Chem. 15 (2007) 3635–3648
3641
distance, one would not expect a covalent interaction to
occur with this cysteine residue to a significant degree.
Reference compounds, 1 and 2, are known irreversible
inhibitors of EGFR and their potent inhibitory activities
in this respect were confirmed in this study. Neither
compound was a potent inhibitor of VEGFR-2. Com-
pound 29 (Gefitinib)34 is a known ATP-competitive
EGFR inhibitor and its activity in this respect was also
confirmed in this study. It did not inhibit VEGFR-2
kinase significantly in our assay. Both the phthalazine
3035 and the quinazoline 3136 are reported to be ATP-
competitive inhibitors of VEGFR-2 and we confirmed
their activities against this target. They also showed
modest EGFR kinase inhibitory activity.
4. Results and discussion
The results of our biological assays are shown in Table 1.
Our EGFR and VEGFR-2 assays are DELFIAꢂ kinase
assays41 where a recombinant full cytoplasmic domain of
human protein is utilized and where the activity of an
inhibitor is expressed as an IC50. For compounds where
an IC50 could not be determined, the average % inhibi-
tion observed at the highest/lowest concentration tested
is given. In order to assess the dependence of the IC50
values on the ATP concentration used in the assays,
the EGFR and VEGFR-2 kinase assays were conducted
using 1 lM and 1 mM concentrations of ATP. While the
IC50 of a reversibly binding ATP-competitive inhibitor is
a direct measure of the compound’s ability to fit at the
enzyme active site, the situation is expected to be more
complicated for an irreversibly binding inhibitor where
the IC50 will be a measure of both the fit at the active site,
as well as the reactivity of the molecule. In this situation,
one expects that for two molecules that fit the site equally
well, the more reactive molecule will have a lower IC50
value. The IC50 value that we measure for a covalent
binding inhibitor will be time dependent and addition-
ally, depending on the rate of reaction of the inhibitor
with the target enzyme, the IC50 value could simulta-
neously consist of components reflecting both reversible
and irreversible binding.
It is clear from the data presented in Table 1 that each
compound which has a quinone ring is an inhibitor of
VEGFR-2 and each compound which has the 4-(dim-
ethylamino)crotonamide Michael acceptor group is an
inhibitor of EGFR. Compounds that have both types
of reactive groups, 15–24, inhibit both enzymes, many
with good potency. It is instructive to compare com-
pounds having a quinone ring but no Michael acceptor
group, 32–34, with the respective compounds having the
same quinone substituents and also having the Michael
acceptor group, 17, 19, and 23. While the former
compounds are potent VEGFR-2 inhibitors with IC50
values in the range 46.1–53.7 nM at 1 lM ATP, they
lack the ability to inhibit EGFR significantly. In
contrast, 17, 19, and 23 are relatively potent inhibitors
of both enzymes. Similarly, the compounds that have
a Michael acceptor but do not have the quinone ring,
1, 2, and 27, are each potent inhibitors of EGFR, but
Table 1. Inhibition of EGFR and VEGFR-2 kinasesa
Compound
EGFR
VEGFR-2
IC50 (nM)
Avg % IC50 (nM)
Avg % Ratio
IC50 (nM)
Avg % IC50 (nM)
Avg % Ratio
Hi/Lo ATPe
1 lM ATPb inhc
1 mM ATPd inhc
Hi/Lo ATPe 1 lM ATPf inhc
1 mM ATPg inhc
1
2
0.6
34.9
0.18
59.2
2.3
3131
858.2
378.7
109.3
113.4
111.6
102.3
135.0
80.2
>10,000
>10,000
1397
0
>3.2
0.08
799.3
283.3
4.1
À4
>11.7
3.7
2.8
15
16
17
18
19
20
21
22
23
24
27
28
29
30
31
32
33
34
>10,000
2089
43.1
24
>12.5
7.4
303.3
347.6
550.2
783.1
1009
10.6
5.4
3.1
4.9
166.7
18.7
90.8
12.2
47.5
107.3
24.9
894.3
216.2
11.5
8.7
7.7
7.5
791.9
35.6
2.9
6.2
178.8
298.0
1285
2.2
3.9
295.2
346.9
76.2
3.2
7.0
371.5
188.7
4389
1489
>10,000
198.3
27.3
3.5
3.2
173.1
704.0
612.7
>10,000
>10,000
>10,000
6754
<10.0
2.6
5.4
64
>70.4
>38.7
505.3
>21.8
>89.0
>1.3
>1.0
>1.0
12
17
6
>2.3
>6.7
>1.0
34.1
177.1
5.0
>100
2753
5
12
457.7
112.4
7710
>10,000
>10,000
>10,000
>10,000
>10,000
46
18
4834
250
À13
À30
À27
50.2
53.7
>10,000
>10,000
20
34
140
188
2.6
4.1
46.1
a The assays were conducted using the indicated ATP concentration. The IC50 is the concentration in nM needed to inhibit the phosphorylation of the
protein substrate by 50% as determined from the dose–response curve. The IC50 values are averages of multiple determinations.
b The standard errors, on average, were 15% of the IC50 value.
c The % inhibition at the concentration shown in the left column.
d The standard errors, on average, were 18% of the IC50 value.
e The ratio of the IC50 values determined at 1 mM and 1 lM concentrations of ATP.
f The standard errors, on average, were 15% of the IC50 value.
g The standard errors, on average, were 17% of the IC50 value.

3642
A. Wissner et al. / Bioorg. Med. Chem. 15 (2007) 3635–3648
are very weak inhibitors of VEGFR-2. It is clear that the
presence of the 4-(dimethylamino)crotonamide Michael
acceptor group has altered the inhibitory profiles of
these compounds resulting in improved EGFR activity,
and that the presence of the quinone ring confers VEG-
FR-2 inhibitory activity.
when the ATP concentration was raised from 1 lM to
1 mM. Likewise, the other reversible EGFR inhibitor
29 (Gefitinib) lost over 500-fold activity on increasing
the ATP concentration. In contrast, for the known
irreversible EGFR/Her2 kinase inhibitors 1 and 2, while
we did see a 58-fold decrease in potency in the EGFR
kinase assay for 1 after raising the ATP concentration,
we observed only about a 1-fold decrease in the
potency for 2. For the compounds 16–24, which we
are proposing function as dual irreversible inhibitors,
we observed that the IC50 values increased on average
only about 6-fold when the ATP concentration was
raised from 1 lM to 1 mM. For several, the increase
was only about 2- to 3-fold. The modest increase in
the IC50 values we observe is consistent with these
compounds functioning as irreversibly binding inhibi-
tors of EGFR kinase.
With respect to the SAR within the group of proposed
dual irreversible inhibitors 15–24, 15 clearly stands out
as being one of the least potent inhibitors of VEGFR-
2. This is consistent with the results of our earlier study
where we found that the 4-methoxy substituted quinone
derivatives were among the less potent inhibitors.19 We
attributed this observation to the lower reactivity of
such compounds as determined by the calculated
LUMO energies and kinetic studies. It also appears that
15 is not a good inhibitor of EGFR despite the fact that
it contains a 4-(dimethylamino)crotonamide Michael
acceptor group. In this case, it is unlikely that the inher-
ent reactivity of the Michael acceptor is significantly dif-
ferent in 15 compared to the other compounds in the
series. One possible explanation for the inferior EGFR
inhibitory activity in this case is that this compound
has a lower initial binding affinity for the EGFR active
site, which could result in a decreased rate of covalent
bond formation. To be sure, this compound will lack
the favorable interaction with the Phe-832 that is ex-
pected to be present with the 4-benzyloxy-type substi-
tuted compounds 17–24. In a related observation, the
1,3-difluoro-2-propoxy derivative 16, which also lacks
this interaction, is a less potent inhibitor than the other
compounds in the series 17–24 which can make such an
interaction. Overall, the remaining compounds that are
proposed to function as dual irreversible inhibitors,
17–24, are each relatively potent inhibitors of both en-
zymes. We also note that there is not much variability
in the potencies of 17–24 in inhibiting EGFR and
VEGF-2, but this is to be expected given that these com-
pounds were purposely designed to have similar struc-
tural features that were optimized in our earlier
work.15–17,19
For the known reversible VEGFR-2 inhibitors 30 and
31, VEGFR-2 kinase inhibitory activity decreased sig-
nificantly with increasing the ATP concentration from
1 lM to 1 mM with a 33- and 176-fold increase in
IC50 values observed for these compounds, respectively.
The compound 27 that has a Michael acceptor and
headpiece designed to confer reversible binding VEG-
FR-2 activity was not very potent in our VEGFR-2 as-
says to begin with and it is therefore difficult to assess
the effects on potency of increasing the ATP concentra-
tion for this analog. For those compounds that have a
quinone ring and no Michael acceptor, 32–34, as
expected for irreversible inhibitors of VEGFR-2, we
observed a less significant increase (3- to 4-fold) in the
IC50 values on raising the ATP concentration. Most
significantly, for the compounds that are proposed to
function as dual irreversible inhibitors of both EGFR
and VEGFR-2 kinases, 15–24, we observed a minimal
increase in the VEGFR-2 IC50 values on changing the
ATP concentration from 1 lM to 1 mM. On average,
the increase was only about 3-fold with the range being
2- to 8-fold. The modest increase in the IC50 values we
see is consistent with these compounds functioning as
irreversibly binding inhibitors of VEGFR-2 kinase.
In our earlier studies with the EGFR/Her2 inhibitors 1
and 2, we demonstrated that these compounds set up a
covalent interaction with the targeted proteins by using
radio-labeled drug.15,16 In our other studies of the qui-
none-based irreversible inhibitors of VEGFR-2, we
demonstrated covalent binding using mass spectral stud-
ies of the drug–protein complex.19 In the present inves-
tigation, our evidence for covalent interaction will rely
on the dependence of inhibitory activities on the concen-
tration of ATP used in the assays. For compounds that
function as conventional reversible inhibitors, we will
expect to see a very significant increase in the IC50 values
when changing the ATP concentration from 1 lM to
1 mM, while for irreversible inhibitors, since they should
be effectively non-ATP-competitive, the changes in IC50
values are expected to be notably less, provided that the
inhibitors are given sufficient time to react with the tar-
get cysteine residue.
In earlier studies15–17 we showed, by unambiguous
means, that quinazoline and 3-cyanoquinoline-based
inhibitors having a 6-(4-(dimethylamino)crotonamide)
Michael acceptor group are capable of functioning as
irreversible EGFR/Her2 inhibitors. We also showed,
again unambiguously, that quinazoline-based inhibitors
having a 4-(amino-[1,4]benzoquinone) moiety can func-
tion as irreversible inhibitors of VEGR-2.19 These earlier
studies, combined with the in vitro studies presented
herein, are consistent with our hypothesis that it would
be possible to construct compounds such as 15–24 that
function as dual irreversible inhibitors of both EGFR
and VEGFR-2 kinases by targeting different, non-con-
served, cysteine residues in each protein using molecules
having two separate reactive centers. The purpose of the
present study was simply to explore this hypothesis.
Clearly, further evaluations such as selectivity studies
and testing in cell-based assays and xenograft models
will be needed to determine if this concept has any prac-
tical utility for the treatment of cancer.
As expected, the known reversible EGFR inhibitor 28
lost essentially all of its EGFR kinase inhibitory activity

A. Wissner et al. / Bioorg. Med. Chem. 15 (2007) 3635–3648
3643
5. Experimental section
5.1. Molecular modeling
BSA, and 2% DMSO. After 40 min at room tempera-
ture, liquid was removed and plates were washed three
times with TBST (TBS with 0.05% Tween 20). The wells
were then incubated for 1 h at room temperature with
75 ll of ꢀ0.1 lg/mL europium-conjugated anti-phos-
photyrosine antibody (PT66; Perkin Elmer Life Sci-
ences, Boston, MA) prepared in Assay Buffer (Perkin
Elmer). Plates were washed three times with TBST and
then incubated for 15 min in the dark with 100 ll of
Enhancement Solution (Perkin Elmer). Plates were read
in a Victor-V multi-label counter (Perkin Elmer) using
the default europium detection protocol. Percent inhibi-
tion or IC50 of compounds were calculated by compari-
son with DMSO-treated control wells using model 201
of XLfit version 4.2 (IDBS).
The inhibitor structures were minimized using the
MMFF94 force field in Sybyl.42 A public domain crystal
structure of EGFR37 and that ofVEGFR-240 were used as
the 3-D representations for molecular docking studies.
The inhibitor structures were docked using the GLIDE38
docking algorithm in the XP (extra precision) mode. De-
tails of the algorithm are found in GLIDE documenta-
tion. Briefly, GLIDEs proprietary conformational
expansion and exhaustive search of the binding site pro-
duces a multitude of ligand poses which undergo an initial
refinement, energy minimization on a pre-computed grid,
and a final scoring and ranking. GLIDE uses proprietary
scoring functions that are variations of the ChemScore43
empirical scoring function and the OPLS-AA44 force field
to compute van der Waals and electrostatic grids for the
receptor. The final ligand binding poses are ranked
according to a computed Emodel score that encompasses
the grid score, the proprietary Glide Score, and the inter-
nal energy strain. The low-energy protein–ligand com-
plexes from these docking studies were used as starting
points for molecular dynamics simulation studies using
the MMFF94 force field in Sybyl.42 Frames from trajecto-
ries of 10 picosecond (ps) simulations carried out at
300 ꢁK at a stepsize of 10 femtoseconds (fs) were selected
to represent the molecular motion of the protein–ligand
complexes (see Figs. 2 and 5). In an effort to conserve com-
putational resources, only the ligand and residues within
In addition to the assay described above, assays were
also conducted in a similar manner in the presence of
1000 lM ATP in order to determine the dependence of
the IC50 values on the ATP concentration.
5.3. Chemistry
1H NMR spectra were determined with a Bruker
DRX400 spectrometer at 400 MHz. Chemical shifts, d,
are in parts per million relative to the internal standard
tetramethylsilane. Electrospray (ES) mass spectra were
recorded in positive mode on a Micromass Platform
spectrometer. Some high-resolution electrospray mass
spectra with higher precision were obtained on a Bruker
9.4T FTMS spectrometer. Chromatographic purifica-
tions were performed by flash chromatography using
Baker 40l silica gel. Analytical HPLC was conducted
on an HP 1090 liquid chromatography system over a
˚
an 8 A radius were allowed to move. The remainder of
the protein was formed into an aggregate. The function
of the aggregate is to accommodate the movement of
the selected region. As the binding site water molecule
of the EGFR crystal structure was retained in these com-
putational studies, the water molecule was included in the
aggregate during the molecular dynamics simulation.
˚
4.6 · 150 mm YMC ODS-A column (5 lm, 120 A) using
multiple wavelength UV detection (typically 230 and
254 nm) unless stated otherwise.
5.4. 7-Ethoxy-6-nitroquinazolin-4-ol (9)
5.2. Biology
To the solution of Na (10.88 g, 473.37 mmol) in 660 mL
EtOH was added 33 g (157.79 mmol) of 7-fluoro-6-ni-
tro-3H-quinazolin-4-one, 8.29 The reaction mixture
was stirred at reflux for 1.5 h. It was then poured into
dilute NH4Cl. The solid was collected and washed with
H2O. The solid was boiled in 1200 mL EtOH and cooled
over the weekend. The resulting solid was collected and
washed with ether to give 31.7 g (85%) of 9 as a light tan
VEGFR-2 and EGFR kinase assays. The cytoplasmic
domain (Val 805–Val 1356) of human VEGFR-2 was
expressed in Sf9 cells and purified to >90% by sequential
chromatography. EGFR kinase was purchased from
Sigma (E3641). Kinase activity was evaluated using a
DELFIAꢂ (dissociation-enhanced lanthanide fluores-
cent immunoassay).41 Assay conditions (enzyme, pep-
tide, ATP, and time) were adjusted for each kinase
such that reactions proceeded under linear kinetics.
Nunc Maxisorb 96-well plates were coated for 1 h with
poly(Glu4-Tyr) peptide (Sigma) at a final concentration
of either 0.5 lg/mL (for VEGFR-2 assay) or 25 lg/mL
(for EGFR assay) in Tris-buffered saline (TBS)
(25 mM Tris, pH 7.2, 150 mM NaCl) and then washed
two times with TBS. VEGFR-2 or EGFR enzyme was
diluted into kinase reaction buffer and incubated with
compounds (prepared in 100% dimethylsulfoxide
(DMSO)) for 10 min (for VEGFR-2) or 20 min (for
EGFR). ATP/MgCl2 was added to each well to initiate
the reaction. Final concentrations of the assay compo-
nents were: 1 lM ATP, 10 mM MgCl2, 1 mM MnCl2,
4 mM HEPES, pH 7.4, 20 lM Na3VO4, 20 lg/mL
solid: MS (ESI) m/z 236.1 (M+H)+1
;
1H NMR
(400 MHz, DMSO-d6) d 1.39 (t, J = 4 Hz, 3H), 4.34 (q,
J = 6.97 Hz, 2H), 7.41 (s, 1H), 8.22 (s, 1H), 8.51
(s, 1H), 12.49 (s, 1H).
5.5. 4-Chloro-7-ethoxy-6-nitroquinazoline (10)
A mixture of 9 (31.7 g, 134.78 mmol), 235 mL SOCl2,
and three drops of DMF was stirred at reflux for
3.5 h. Excess SOCl2 was removed on the rotoevapora-
tor. The reaction was stirred in 600 mL CHCl3 and then
cooled to 0 ꢁC. The mixture was washed with cold NaH-
CO3 and then dried over MgSO4. The solution was
passed through a pad of magnesol and washed with
CHCl3 and EtOAc. The solvent of the filtrate was

3644
A. Wissner et al. / Bioorg. Med. Chem. 15 (2007) 3635–3648
removed to give 33.2 g (97%) of 10 as a white solid. The
compound was used without further purification: MS
4.29 (dd, J = 12, 8 Hz, 2H), 6.60 (d, J = 16 Hz, 1H),
6.77–6.83 (m, 1H), 7.18–7.26 (m, 2H), 7.60 (s, 1H),
8.40 (s, 1H), 8.91 (s, 1H), 9.15 (s, 1H), 9.48 (s, 1H).
(ESI) m/z 254.0 (M+H)+1
;
1H NMR (400 MHz,
CDCL3) d 1.56 (t, J = 4 Hz, 3H), 4.34 (q, J = 8 Hz,
2H), 7.52 (s, 1H), 8.61 (s, 1H), 9.04 (s, 1H).
5.9. (2E)-4-(Dimethylamino)-N-{7-ethoxy-4-[(4-methoxy-
3,6- dioxocyclohexa-1,4-dien-1-yl)amino]quinazolin-6-
yl}but-2-enamide (15)
5.6. N-(4-Chloro-2,5-dimethoxyphenyl)-7-ethoxy-6-nitro-
quinazolin-4-amine (11)
A sample of 1.57 g (3.23 mmol) of 13 was boiled in
80 mL CH3CN to dissolve most of the solid. To this
was added 36 mL H2O and the mixture was cooled
to room temperature. To the stirred mixture was
added 4.25 g (7.75 mmol) of solid ceric ammonium
nitrate. The reaction mixture was stirred at room
temperature for 2.5 h. The mixture was diluted in
700 mL CHCl3 and 50 mL of saturated aqueous
NaHCO3. The mixture was filtered through Celite.
The solid-Celite mixture was washed several times with
a total volume of 1400 mL CHCl3. The organic layer
was separated and washed with H2O giving a CHCl3
solution of 14.
A mixture of 10 (24.2 g, 95.41 mmol) and 4-chloro-2,5-
dimethoxyaniline (17.9 g, 95.41 mmol) in 1.3 L isopro-
panol was stirred at reflux for 3 h. The solution was
allowed to stand at room temperature overnight. The
solid was collected and washed with ether to give
37.3 g (89%) of 11 as an orange powder. The compound
was used without further purification: MS (ESI) m/z
405.1 (M+H)+1 1H NMR (400 MHz, DMSO-d6) d
;
1.44 (t, J = 6.92 Hz, 3H), 3.77 (s, 3H), 3.79 (s, 3H),
4.39 (q, J = 6.97 Hz, 2H), 7.21 (s, 1H), 7.33(s, 1H),
7.59 (s, 1H), 8.78 (s, 1H), 9.43 (s, 1H), 11.36 (bs, 1H).
5.7. N4-(4-Chloro-2,5-dimethoxyphenyl)-7-ethoxyquinaz-
oline-4,6-diamine (12)
To this was added 300 mL CH3OH. The solution was
dried over MgSO4, filtered, and 50 mL Et3N was
added. The reaction mixture was heated to reflux for
2.75 h. The solvent was removed under vacuum and
the residue was dissolved in CHCl3 and washed with
saturated aqueous Na2CO3. The organic layer was
dried over MgSO4 and filtered through a short column
of magnesol, eluting with CHCl3 and then CHCl3/
EtOAc/MeOH = 500:500:50. The solvent was removed
from the filtrate. The resulting solid was triturated with
EtOAc and collected to give 0.85 g (58%) of 15 as an
orange crystalline solid: MS (ESI) m/z 452.2
A mixture of 11 (37 g, 83.85 mmol), iron (28.1 g,
503.09 mmol), acetic acid (30.21 g, 503.09 mmol), and
sodium acetate (6.88 g, 83.85 mmol) in 1.4 L CH3OH
was mechanically stirred at reflux for 3 h. To this mix-
ture was added 50 mL NH4OH. The boiling mixture
was filtered. The solid was washed with hot CH3OH.
The solvent was removed from filtrate. The resulting
solid was extracted with boiling acetone 4· and filtered.
The acetone extracts were concentrated. The residue was
recrystallized from EtOAc to yield 28.7 g (91%) of 12 as
light tan solid. The compound was used without further
(M+H)+1 1H NMR (400 MHz, CDCl3) d 1.59 (t,
;
1
purification: MS (ESI) m/z 375.1 (M+H)+1; H NMR
J = 6.92 Hz, 3H), 2.34 (s, 6H), 3.19–3.21 (m, 2H),
3.91 (s, 3H), 4.32 (q, J = 6.88 Hz, 2H), 6.00 (s, 1H),
6.25 (d, J = 15.11 Hz, 1H), 7.04–7.15 (m, 1H), 7.30 (s,
1H), 8.07 (s, 1H), 8.14 (s, 1H), 8.81 (s, 1H), 8.98 (s,
1H), 9.30 (s, 1H); Anal. Calcd for C23H25N5O5–(1.5
H2O): C, 57.73; H, 5.90; N, 14.64. Found: C, 57.73;
H, 5.90; N, 14.64.
(400 MHz, DMSO-d6) d 1.44 (t, J = 7.05 Hz, 3H), 3.82
(s, 3H), 3.85 (s, 3H), 4.21 (q, J = 6.97 Hz, 2H), 5.41 (s,
2H), 7.07 (s, 1H), 7.18 (s, 1H), 7.20 (s, 1H), 8.12
(s, 1H), 8.33 (s, 1H), 8.41 (s, 1H).
5.8. (2E)-N-{4-[(4-Chloro-2,5-dimethoxyphenyl)amino]-7-
ethoxyquinazolin-6-yl}-4-(dimethylamino)but-2-enamide (13)
5.10. (2E)-4-(Dimethylamino)-N-[7-ethoxy-4-({4-[2-flu-
oro-1-(fluoromethyl)ethoxy]-3,6-dioxocyclohexa-1,4-dien-
1-yl}amino)quinazolin-6-yl]but-2-enamide (16)
The (E)-4-(dimethylamino)-2-butenoic acid hydrochlo-
ride salt (4.42 g, 26.68 mmol), oxalyl chloride (4.42 g,
26.68 mmol), and a trace of DMF in 57 mL CH3CN
were stirred at 55 ꢁC for 20 min. About half of the sol-
vent was removed at reduced pressure at 50 ꢁC and the
solution was cooled. A solution of 12 (5 g, 13.34 mmol)
in warm NMP (57 mL) was added over 10 min. The
reaction mixture was stirred at 0 ꢁC for 2 h. and diluted
with dilute aqueous NaHCO3. The resulting solid was
collected, dissolved in hot THF, diluted with EtOAc,
dried over MgSO4, and filtered. The filtrate was passed
through a column of silica gel, eluting with EtOAc,
EtOAc/MeOH, and 700:300:10 EtOAc/MeOH/Et3N.
The solvent was removed from product fractions. The
resulting solid was stirred in ether and collected to yield
5 g (77%) of 13 as a white solid: MS (ESI) m/z 486.1
This compound was prepared from 13 (0.86 g,
1.77 mmol) using CAN (2.33 g, 4.25 mmol) as described
below, and then treated with 13.6 g (141.57 mmol) of
1,3-difluoro-2-propanol and 27 mL Et3N using the
procedure described above for 17. In this instance, the
reaction mixture was refluxed for 4 h. and then worked
up in the usual manner, except the product was eluted
from magnesol using CH2Cl2/EtOAc = 1:1. The solvent
of the filtrate was removed. The resulting solid was
swirled in CH3CN-ether and filtered to give 0.38 g
(42%) of 16 as an orange solid: HRMS: Calcd for
C25H27F2N5O5 + H, 516.20530; found (ESI-FTMS,
1
(M+H)+1), 516.20519; H NMR (400 MHz, CDCl3) d
(M+H)+1; MS (ESI) m/z 243.5 (M+2H)+2
;
1H NMR
1.58 (t, J = 7.05 Hz, 3H), 2.39 (s, 6H), 3.26 (d,
J = 4.53 Hz, 2H), 4.32 (m, J = 6.80 Hz, 3H), 4.69-4.81
(m, 4H), 6.14 (s, 1H), 6.29 (d, J = 15.11 Hz, 1H),
(400 MHz, DMSO-d6) d 1.46 (t, J = 4 Hz, 3H), 2.21 (s,
6H), 3.12 (d, J = 4 Hz, 2H), 3.77 (s, 3H), 3.80 (s, 3H),

A. Wissner et al. / Bioorg. Med. Chem. 15 (2007) 3635–3648
3645
7.04-7.15 (m, 1H), 7.28 (s, 1H), 8.08 (s, 1H), 8.17 (s, 1H),
8.81 (s, 1H), 8.89 (s, 1H), 9.27(s, 1H); Anal. Calcd for
C25H27F2N5O5–(0.8 H2O): C, 56.65; H, 5.44; N, 13.22.
Found: C, 56.91; H, 5.28; N, 13.51.
5.13. (2E)-N-(4-{[4-(Benzyloxy)-3,6-dioxocyclohexa-1,4-
dien-1-yl]amino}-7-ethoxyquinazolin-6-yl)-4-(dimethyl-
amino)but-2-enamide (19)
A suspension of 13 (0.854 g, 1.76 mmol) and CAN
(2.314 g, 4.22 mmol) in 40 mL CH3CN and 20 mL
H2O was stirred at room temperature for 1 h. The reac-
tion mixture was diluted with 50 mL CH2Cl2 and was
treated with 50 mL of saturated aqueous Na2CO3. The
resulting mixture was stirred for 30 min. It was then fil-
tered through Celite and the Celite was washed with
CH2Cl2. The filtrate was extracted (3·30 mL) with
CH2Cl2. The organic layers were combined and dried
over Na2SO4. To this was added 0.449 g (2.64 mmol)
of NaOPh(3H2O), which was previously melted in ben-
zyl alcohol (1.9 g, 17.6 mmol). The reaction mixture was
concentrated on the rotary evaporator at 30–40 ꢁC over
30 min. It was passed through a magnesol plug, which
was eluted with CH2Cl2-isopropanol mixtures and final-
ly with a CH2Cl2-isopropanol-Et3N mixture. The sol-
vent of the filtrate was concentrated to give 288 mg
(31%) of 19 as a red solid: HRMS: Calcd for
C29H29N5O5 + H, 528.22415; found (ESI-FTMS,
5.11. (2E)-4-(Dimethylamino)-N-[7-ethoxy-4-({4-[(3-fluo-
robenzyl)oxy]-3,6-dioxocyclohexa-1,4-dien-1-yl}amino)qui-
nazolin-6-yl]but-2-enamide (17)
A 0.8 g (1.65 mmol) sample of 13 was boiled in 40 mL
CH3CN. To this solution was added 20 mL H2O and
it was allowed to cool to room temperature. To this
was added 2.17 g (3.95 mmol) of solid ceric ammonium
nitrate. The reaction mixture was stirred at room tem-
perature for 1.75 h. and was then diluted with 700 mL
CHCl3 and 25 mL of saturated aqueous Na2CO3. The
mixture was extracted 4· with a total of 900 mL CHCl3.
3-Fluorobenzyl alcohol (8.31 g, 65.85 mmol) was added.
The mixture was dried over MgSO4, filtered, and 25 mL
Et3N was added. The reaction mixture was refluxed
under N2 for 3.5 h. The reaction mixture was washed
with saturated NaHCO3. The organic layer was dried
over MgSO4. The solvent was removed in vacuum.
The residue was dissolved in CH2Cl2. It was then filtered
through a short column of magnesol and eluted with
CHCl3, then CHCl3/EtOAc/isopropanol = 250:250:75.
The solvent of the filtrate was removed. The resulting
solid was triturated with CH3CN-ether and filtered to
give 0.16 g (18%) of 17 as an orange powder: HRMS:
Calcd for C29H28FN5O5 + H, 546.21472; found (ESI-
FTMS, (M+H)+1), 546.21347; 1H NMR (400 MHz,
CDCl3) d 1.59 (t, J = 6.92 Hz, 3H), 2.35 (s, 6H), 3.22
(d, J = 4.78 Hz, 2H), 4.32 (q, J = 6.80 Hz, 2H), 5.10 (s,
2H), 6.03 (s, 1H), 6.26 (d, J = 15.11 Hz, 1H), 7.03–7.13
(m, 2H), 7.16 (dd, J = 9.32, 2.27 Hz, 1H), 7.21 (d,
J = 7.55 Hz, 1H), 7.30 (s, 1H), 7.35–7.42 (m, 1H), 8.08
(s, 1H), 8.15 (s, 1H), 8.81(s, 1H), 8.94 (s, 1H), 9.29 (s,
1H); Anal. Calcd for C29H28FN5O5–(0.7 H2O): C,
62.39; H, 5.31; N, 12.55. Found: C, 62.61; H, 4.99; N,
12.90.
1
(M+H)+1), 528.22382; H NMR (400 MHz, CDCl3) d
1.59 (t, J = 6.92 Hz, 3H), 2.32 (s, 6H), 3.18 (dd,
J = 5.67, 1.38 Hz, 2H), 4.32 (q, J = 6.97 Hz, 2H), 5.12
(s, 2H), 6.06 (s, 1H), 6.15–6.34 (m, 1H), 7.05–7.14 (m,
1H), 7.30 (s, 1H), 7.33–7.48 (m, 5H), 8.07 (s, 1H), 8.14
(s, 1H), 8.81 (s, 1H), 8.95 (s, 1H), 9.29 (s, 1H); Anal.
Calcd for C29H29N5O5–(0.5 H2O): C, 64.91; H, 5.64;
N, 13.05. Found: C, 64.60; H, 5.54; N, 13.05.
5.14. (2E)-4-(Dimethylamino)-N-(4-{[3,6-dioxo-4-(pyri-
din-2-ylmethoxy)cyclohexa-1,4-dien-1-yl]amino}-7-eth-
oxyquinazolin-6-yl)but-2-enamide (20)
This compound was prepared from 13 (1.084 g,
2.23 mmol), CAN (2.93 g, 5.35 mmol), NaOPh(3H2O)
(0.379 g, 2.23 mmol), and pyridine-2-methanol (2.43 g,
22.3 mmol) using the procedure described above for
19. The reaction mixture was worked up and purified
in the usual manner to give 74 mg (6%) of 20 as a dark
red solid: MS (ESI) m/z 529.1(M+H)+1; MS (ESI) m/z
265 (M+2H)+2; HRMS: Calcd for C28H28N6O5 + H,
529.21940; found (ESI-FTMS, (M+H)+1), 529.21897;
1H NMR (400 MHz, CDCl3) d 1.58 (t, J = 6.92 Hz,
3H), 2.44 (s, 6H), 3.32 (d, J = 5.29 Hz, 2H), 4.31 (q,
J = 7.05 Hz, 2H), 5.24 (s, 2H), 6.13 (s, 1H), 6.33 (d,
J = 15.11 Hz, 1H), 7.05–7.15 (m, 1H), 7.28–7.33 (m,
2H), 7.57 (d, J = 8.06 Hz, 1H), 7.73–7.81 (m, 1H), 8.07
(s, 1H), 8.20 (s, 1H), 8.62 (d, J = 4.28 Hz, 1H), 8.82 (s,
1H), 8.94 (s, 1H), 9.27 (s, 1H); Anal. Calcd for
C28H28N6O5–(3 H2O): C, 57.72; H, 5.88; N, 14.42.
Found: C, 57.54; H, 5.13; N, 14.79.
5.12. (2E)-N-[4-({4-[(3,4-Difluorobenzyl)oxy]-3,6-dioxo-
cyclohexa-1,4-dien-1-yl}amino)-7-ethoxyquinazolin-6-yl]-
4-(dimethylamino)but-2-enamide (18)
This compound was prepared from 13 (0.85 g, 1.75 mmol)
which was oxidized with CAN (2.3 g, 4.2 mmol),
extracted with CHCl3, and then treated with 4.79 g
(33.23 mmol) of 3,4-difluorobenzyl alcohol and 20 mL
Et3N using the procedure described above for 17. In this
instance, the solution was first concentrated to 500 mL
CHCl3 and the reaction mixture was refluxed for 3 h. It
was then worked up and purified in the usual manner.
The resulting solid was swirled in CH3CN-ether and
filtered to give 0.21 g (21%) of 18 as a dark red solid:
MS (ESI) m/z 564.2 (M+H)+1
;
1H NMR (400 MHz,
5.15. (2E)-N-[4-({4-[(3-Chlorobenzyl)oxy]-3,6-dioxocy-
clohexa-1,4-dien-1-yl}amino)-7-ethoxyquinazolin-6-yl]-
4-(dimethylamino)but-2-enamide (21)
CDCl3) d 1.59 (t, J = 6.92 Hz, 3H), 2.34 (s, 6H), 3.20 (d,
J = 5.04 Hz, 2H), 4.32 (q, J = 6.80 Hz, 2H), 5.04 (s, 2H),
6.03 (s, 1H), 6.25 (d, J = 15.36 Hz, 1H), 7.05-7.13 (m,
1H), 7.16-7.23 (m, 2H), 7.28-7.33 (m, 2H), 8.08 (s, 1H),
8.15 (s, 1H), 8.82 (s, 1H), 8.93 (s, 1H), 9.29 (s, 1H); Anal.
Calcd for C29H27F2N5O5–(0.75 H2O): C, 60.36; H, 4.98;
N, 12.14. Found: C, 60.16; H, 4.56; N, 12.48.
To a stirred solution of 13 (0.829 g, 1.71 mmol) in 40 mL
CH3CN and 20 mL H2O was added CAN (2.25 g,
4.1 mmol). After 1 h, 50 mL CH2Cl2 and 75 mL of satu-
rated Na2CO3 were added. The reaction mixture was

3646
A. Wissner et al. / Bioorg. Med. Chem. 15 (2007) 3635–3648
stirred for 30 min. The suspension was passed through a
pad of Celite and the pad was washed with CH2Cl2. The
filtrate was extracted with CH2Cl2 (3·250 mL). The
combined organic layers (about a total volume of
800 mL CH2Cl2) were dried over Na2SO4. To this was
added 439 mg (2.57 mmol) of NaOPh(3H2O) that was
previously melted in 3-chlorobenzyl alcohol (2 mL,
17.1 mmol). After stirring, the reaction mixture was con-
centrated and poured onto magnesol, which was eluted
with CH2Cl2-isopropanol mixtures and finally with a
CH2Cl2-isopropanol-Et3N mixture. The solvent of the
filtrate was removed to give 302 mg (31%) of 21 as an or-
ange solid: MS (ESI) m/z 562.0 (M+H)+1, 281.5
(M+2H)+2; HRMS: Calcd for C29H28ClN5O5 + H,
562.18517; found (ESI-FTMS, (M+H)+1), 562.18608;
1H NMR (400 MHz, CDCl3) d 1.59 (s, 3H), 2.35 (s,
6H), 3.21 (dd, J = 5.79, 1.26 Hz, 2H), 4.33 (q,
J = 6.97 Hz, 2H), 5.08 (s, 2H), 6.04 (s, 1H), 6.21–6.31
(m, 1H), 7.05–7.16 (m, 1H), 7.28–7.40 (m, 4H), 7.44 (s,
1H), 8.08 (s, 1H), 8.16 (s, 1H), 8.82 (s, 1H), 8.95 (s,
1H), 9.30 (s, 1H); Anal. Calcd for C29H28ClN5O5–(1.7
H2O): C, 58.80; H, 5.35; N, 11.83. Found: C, 58.71;
H, 5.08; N, 11.94.
6.90 (dd, J = 8.31, 2.01 Hz, 1H), 6.94–7.03 (m, 2H),
7.04–7.13 (m, 1H), 7.28–7.35 (m, 2H), 8.06 (s, 1H),
8.15 (s, 1H), 8.80 (s, 1H), 8.94 (s, 1H), 9.28 (s, 1H).
5.18. (2E)-4-(Dimethylamino)-N-[7-ethoxy-4-({4-[(3-meth-
ylbenzyl)oxy]-3,6-dioxocyclohexa-1,4-dien-1-yl}amino)qui-
nazolin-6-yl]but-2-enamide (24)
This was prepared from 13 (0.854 g, 1.76 mmol), CAN
(2.3 g, 4.2 mmol), NaOPh (3H2O) (0.448 g, 2.69 mmol),
and 3-methylbenzyl alcohol (2.15 g, 17.6 mmol) as
described above for 21 to give 193 mg (20%) of 24 as a
red solid: HRMS: Calcd for C30H31N5O5 + H,
542.23980; found (ESI-FTMS, (M+H)+1), 542.23995;
the purity of 24 was evaluated on two HPLC systems
and found to be 98% (MeCN/H2O 30–90, 230 nm) and
97% (MeOH/H2O 70–90, 230 nm); 1H NMR
(400 MHz, CDCl3) d 1.58 (t, J = 7.05 Hz, 3H), 2.32 (s,
6H), 2.38 (s, 3H), 3.18 (dd, J = 5.54, 1.51 Hz, 2H),
4.31 (q, J = 7.05 Hz, 2H), 5.09 (s, 2H), 6.05 (s, 1H),
6.14–6.31 (m, 1H), 6.98–7.13 (m, 1H), 7.14–7.26 (m,
3H), 7.28–7.30 (m, 1H), 8.05 (s, 1H), 8.13 (s, 1H), 8.13
(s, 1H), 8.80 (s, 1H), 8.94 (s, 1H), 9.28 (s, 1H); Anal.
Calcd for C30H31N5O5–(0.8 H2O): C, 64.79; H, 5.91;
N, 12.60. Found: C, 64.93; H, 5.54; N, 12.46.
5.16. (2E)-4-(Dimethylamino)-N-(4-{[3,6-dioxo-4-(2-thie-
nylmethoxy)cyclohexa-1,4-dien-1-yl]amino}-7-ethoxyqui-
nazolin-6-yl)but-2-enamide (22)
5.19. 4-Dimethylamino-but-2-enoic acid(4-chloro-quin-
azolin-6-yl)-amide (26)
This was prepared from 13 (0.804 g, 1.64 mmol), CAN
(2.72 g, 4.96 mmol), NaOPh(3H2O) (731.6 mg, 4.3 mmol),
and 2-thiophenemethanol (3.12 mL, 33.09 mmol) as
described above for 21 to give 86.3 mg (10%) of 22 as a
red solid: MS (ESI) m/z 534.0 (M+H)+1, 267.5
(M+2H)+2; HRMS: Calcd for C27H27N5O5S + H,
To a stirred suspension of 4-dimethylaminocrotonic acid
hydrochloride (1.45 g, 4.4 mmol) in 17.5 mL CH3CN at
25 ꢁC was added 25 mL (0.25 mmol) of NMP followed
by (COCl)2 (0.76 mL, 8.8 mmol) in one portion. The
reaction mixture was stirred at 55 ꢁC for 30 min, evapo-
rated to 1/3 volume, diluted with 7.5 mL NMP, and
cooled to 0 ꢁC. The resulting solution was treated with
a slurry of 2532 (0.90 g, 5.0 mmol) and 10 mL NMP.
The suspension was stirred at 25 ꢁC for 30 min. The
reaction mixture was stirred with 2.4 g (17.5 mmol) of
K2CO3 in 80 mL H2O and extracted with CH2Cl2 in
three portions. The extract was washed with a small vol-
ume of dilute brine, dried over MgSO4, and concen-
trated to give a solution of 26 (4.4 mmol) in NMP that
was used as such in the subsequent step.
1
534.18057; found (ESI-FTMS, [M+H]1+), 534.18094; H
NMR (400 MHz, CDCl3) d 1.59 (t, J = 7.05 Hz, 3H),
2.34 (s, 6H), 3.20 (dd, J = 5.67, 1.38 Hz, 2H), 4.32 (q,
J = 6.97 Hz, 2H), 5.28 (s, 2H), 6.13 (s, 1H), 6.24 (d,
J = 15.11 Hz, 1H), 6.95–7.16 (m, 2H), 7.20 (d,
J = 2.77 Hz, 1H), 7.30 (s, 1H), 7.39 (dd, J = 5.04,
1.26 Hz, 1H), 8.07 (s, 1H), 8.15 (s, 1H), 8.81 (s, 1H),
8.96 (s, 1H), 9.30 (s, 1H); Anal. Calcd for
C27H27N5O5S–(0.5 H2O): C, 59.77; H, 5.20; N, 12.91.
Found: C, 59.34; H, 4.91; N, 12.91.
5.17. (2E)-4-(Dimethylamino)-N-[7-ethoxy-4-({4-[(3-meth-
oxybenzyl)oxy]-3,6-dioxocyclohexa-1,4-dien-1-yl}amino)-
quinazolin-6-yl]but-2-enamide (23)
5.20. (2E)-N-{4-[(4-Bromo-2-fluorophenyl)amino]quin-
azolin-6-yl}-4- (dimethylamino)but-2-enamide (27)
A mixture of 26 (0.53 mmol in NMP solution, prepared
as above), 4-bromo-2-fluoroaniline (114 mg, 0.6 mmol),
pyridine hydrochloride (116 mg, 1.0 mmol), and 10 mL
of isopropanol was stirred at 85 ꢁC for 30 min, cooled,
and concentrated. The residue was partitioned between
aqueous K2CO3 (pH ꢀ 9) and EtOAc/MeOH = 20:1.
The organic layer was washed with H2O and brine, dried
over MgSO4, and concentrated. Chromatography of the
residue on silica gel with CH2Cl2–EtOAc–MeOH–Et3N
gave 82 mg of the title compound as a tan solid: HRMS:
Calcd for C20H19BrFN5O + H, 444.08298; found (ESI-
This was prepared from 13 (0.519 g, 1.06 mmol), CAN
(1.404 g, 2.56 mmol), NaOPh (3H2O) (167.38 mg,
0.985 mmol), and 3-methoxybenzyl alcohol (809 mL,
6.564 mmol) as described above for 21 to give 220 mg
(37%) of 23 as a red solid: HRMS: Calcd for
C30H31N5O6 + H, 558.23471; found (ESI-FTMS,
(M+H)+1), 558.23403; the purity of 23 was evaluated
on two HPLC systems and found to be 100% (MeCN–
H2O, 40–90% gradient, retention time = 5.13 min)
and 94.9% (MeOH–H2O, 70–90% gradient, retention
1
1
time = 13.01 min); H NMR (400 MHz, CDCl3) d 1.58
FTMS, (M+H)+1), 444.08479; H NMR (DMSO-d6): d
(t, J = 7.05 Hz, 3H), 2.33 (s, 6H), 3.20 (d, J = 4.78 Hz,
2H), 3.83 (s, 3H), 4.31 (q, J = 6.88 Hz, 2H), 5.10
(s, 2H), 6.05 (s, 1H), 6.24 (d, J = 15.36 Hz, 1H),
2.29 (s, 6H), 3.20–3.26 (m, 2H), 6.39 (d, J = 15.36 Hz,
1H), 6.76–6.87 (m, 1H), 7.42–7.54 (m, 2H), 7.64 (dd,
J = 9.95, 1.64 Hz, 1H), 7.77–7.81 (d, J = 12 Hz, 1H),

A. Wissner et al. / Bioorg. Med. Chem. 15 (2007) 3635–3648
3647
14. Torrance, C. J.; Jackson, P. E.; Montgomery, E.; Kinzler,
K. W.; Vogelstein, B.; Wissner, A.; Nunes, M.; Frost, P.;
Discafani, C. M. Nat. Med. 2000, 6, 1024–1028.
15. Wissner, A.; Overbeek, E.; Reich, M. F.; Floyd, M. B.;
Johnson, B. D.; Mamuya, N.; Rosfjord, E. C.; Discafani,
C.; Davis, R.; Shi, X.; Rabindran, S. K.; Gruber, B. C.;
Ye, F.; Hallett, W. A.; Nilakantan, R.; Shen, R.; Wang,
Y.-F.; Greenberger, L. M.; Tsou, H.-R. J. Med. Chem.
2003, 46, 49–63.
7.86 (dd, J = 8.81, 1.76 Hz, 1H), 8.41–8.45 (s, 1H), 8.83
(s, 1H), 9.86 (s, 1H), 10.47 (s, 1H); Anal. Calcd for
C20H19BrFN5O–(1.5 H2O): C, 50.98; H, 4.83; N,
14.69. Found: C, 51.05; H, 4.72; N, 14.89.
Acknowledgments
16. Rabindran, S. K.; Discafani, C. M.; Rosfjord, E. C.;
Baxter, M.; Floyd, M. B.; Golas, J.; Hallett, W. A.;
Johnson, B. D.; Nilakantan, R.; Overbeek, E.; Reich, M.
F.; Shen, R.; Shi, X.; Tsou, H.-R.; Wang, Y.-F.; Wissner,
A. Cancer Res. 2004, 64, 3958–3965.
17. Tsou, H.-R.; Overbeek-Klumpers, E. G.; Hallett, W. A.;
Reich, M. F.; Floyd, M. B.; Johnson, B. D.; Michalak, R.
S.; Nilakantan, R.; Discafani, C.; Golas, J.; Rabindran, S.
K.; Shen, R.; Shi, X.; Wang, Y.-F.; Upeslacis, J.; Wissner,
A. J. Med. Chem. 2005, 48, 1107–1131.
The authors wish to thank Drs. Tarek Mansour, Lee
Greenberger, and Janis Upeslacis for their support and
encouragement. We also thank the members of the
Wyeth Chemical Technologies group for analytical
and spectral determinations.
References and notes
18. Klutchko, S. R.; Zhou, H.; Winters, R. T.; Tran, T. P.;
Bridges, A. J.; Althaus, I. W.; Amato, D. M.; Elliott, W.
L.; Ellis, P. A.; Meade, M.; Roberts, B. J.; Fry, D. W.;
Gonzales, A. J.; Harvey, P. J.; Nelson, J. M.; Sherwood,
V.; Han, H.-K.; Pace, G.; Smaill, J. B.; Denny, W. A.;
Hollis, S. H. D. J. Med. Chem. 2006, 49, 1475–1485.
19. Wissner, A.; Floyd, M. B.; Johnson, B. D.; Fraser, H.;
Ingalls, C.; Nittoli, T.; Dushin, R. G.; Discafani, C.;
Nilakantan, R.; Marini, J.; Ravi, M.; Cheung, K.; Tan,
X.; Musto, S.; Annable, T.; Siegel, M. M.; Loganzo, F.
J. Med. Chem. 2005, 48, 7560–7581.
20. Kari, C.; Chan, T. O.; Rocha de Quadros, M.; Rodeck, U.
Cancer Res. 2003, 63, 1–5.
21. Salomon, D. S.; Brandt, R.; Ciardiello, F.; Normanno, N.
Crit. Rev. Oncol. Haematol. 1995, 19, 183–232.
22. Gullick, W. J. Br. Med. Bull. 1991, 47, 87–98.
23. Woodburn, J. R. Pharmacol. Ther. 1999, 82, 241–250.
24. Carmeliet, P.; Jain, R. K. Nature 2000, 407, 249–257.
25. Folkman, J. Semin. Oncol. 2002, 29, 15–18.
26. Ferrara, N. Oncologist 2004, 9, 2–10.
27. Hicklin, D. J.; Ellis, L. M. J. Clin. Oncol. 2005, 23, 1011–
1027.
28. Pandey, A.; Volkots, D. L.; Seroogy, J. M.; Rose, J. W.;
Yu, J.-C.; Lambing, J. L.; Hutchaleelaha, A.; Hollenbach,
S. J.; Abe, K.; Giese, N. A.; Scarborough, R. M. J. Med.
Chem. 2002, 45, 3772–3793.
29. Rewcastle, G. W.; Palmer, B. D.; Bridges, A. J.; Show-
alter, H. D. H.; Sun, L.; Nelson, J.; McMichael, A.;
Kraker, A. J.; Fry, D. W.; Denny, W. A. J. Med. Chem.
1996, 39, 918–928.
30. Wedge, S. R.; Kendrew, J.; Hennequin, L. F.; Valentine,
P. J.; Barry, S. T.; Brave, S. R.; Smith, N. R.; James, N.
H.; Dukes, M.; Curwen, J. O.; Chester, R.; Jackson, J. A.;
Boffey, S. J.; Kilburn, L. L.; Barnett, S.; Richmond, G. H.
P.; Wadsworth, P. F.; Walker, M.; Bigley, A. L.; Taylor,
S. T.; Cooper, L.; Beck, S.; Juergensmeier, J. M.; Ogilvie,
D. J. Cancer Res. 2005, 65, 4389–4400.
1. Denny, W. A. Pharmacol. Ther. 2002, 93, 253–261.
2. Bridges, A. J.; Denny, W. A.; Dobrusin, E. M.; Doherty,
A. M.; Elliott, W. L.; Fry, D. W.; Hook, K.; Leopold, W.
R.; McNamara, D. J.; Nelson, J. W.; Palmer, B. D.;
Patmore, S.; Rewcastle, G. W.; Roberts, B. J.; Showalter,
H. D. H.; Slintak, V.; Smaill, J. B.; Thompson, A. M.;
Trumpp-Kallmeyer, S.; Vincent, P. W.; Winters, R. T.;
Zhou, H. R. Soc. Chem. 2001, 264, 151–162.
3. Kwak, E. L.; Sordella, R.; Bell, D. W.; Godin-Heymann,
N.; Okimoto, R. A.; Brannigan, B. W.; Harris, P. L.;
Driscoll, D. R.; Fidias, P.; Lynch, T. J.; Rabindran, S. K.;
McGinnis, J. P.; Wissner, A.; Sharma, S. V.; Isselbacher,
K. J.; Settleman, J.; Haber, D. A. Proc. Natl. Acad. Sci.
U.S.A. 2005, 102, 7665–7670.
4. Carter, T. A.; Wodicka, L. M.; Shah, N. P.; Velasco, A.
M.; Fabian, M. A.; Treiber, D. K.; Milanov, Z. V.;
Atteridge, C. E.; Biggs, W. H., Jr.; Edeen, P. T.; Floyd,
M.; Ford, J. M.; Grotzfeld, R. M.; Herrgard, S.; Insko, D.
E.; Mehta, S. A.; Patel, H. K.; Pao, W.; Sawyers, C. L.;
Varmus, H.; Zarrinkar, P. P.; Lockhart, D. J. Proc. Natl.
Acad. Sci. U.S.A. 2005, 102, 11011–11016.
5. Kobayashi, S.; Boggon, T. J.; Dayaram, T.; Jaenne, P. A.;
Kocher, O.; Meyerson, M.; Johnson, B. E.; Eck, M. J.;
Tenen, D. G.;Balizs, H. N.Engl. J. Med. 2005, 352, 786–792.
6. Kobayashi, S.; Ji, H.; Yuza, Y.; Meyerson, M.; Wong, K.-
K.; Tenen, D. G.; Halmos, B. Cancer Res. 2005, 65, 7096–
7101.
7. Greulich, H.; Chen, T. H.; Feng, W.; Janne, P. A.;
Alvarez, J. V.; Zappaterra, M.; Bulme, r S. E.; Frank, D.
A.; Hahn, W. C.; Sellers, W. R.; Meyerson, M. PLoS
Med. 2005, 2, e313.
8. Yoshimura, N.; Kudoh, S.; Kimura, T.; Mitsuoka, S.;
Matsuura, K.; Hirata, K.; Matsui, K.; Negoro, S.;
Nakagawa, K.; Fukuoka, M. Lung Cancer 2006, 51,
363–368.
9. Cohen, M. S.; Zhang, C.; Shokat, K. M.; Taunton, J.
Science 2005, 308, 1318–1321.
31. Rich, J. N.; Sathornsumetee, S.; Keir, S. T.; Kieran, M.
W.; Laforme, A.; Kaipainen, A.; McLendon, R. E.;
Graner, M. W.; Rasheed, B. K. A.; Wang, L.; Reardon,
D. A.; Ryan, A. J.; Wheeler, C.; Dimery, I.; Bigner, D. D.;
Friedman, H. S. Clin. Cancer Res. 2005, 11, 8145–8157.
32. Wissner, A.; Johnson, B. D.; Floyd, M. B., Jr.; Kitchen,
D. B. U.S. PAT. 5760041, 1998.
33. Fry, D. W.; Kraker, A. J.; McMichael, A.; Ambroso, L.
A.; Nelson, J. M.; Leopold, W. R.; Connors, R. W.;
Bridges, A. J. Science 1994, 265, 1093–1095.
10. Shaul, M.; Abourbeh, G.; Jacobson, O.; Rozen, Y.; Laky,
D.; Levitzki, A.; Mishani, E. Bioorg. Med. Chem. 2004, 12,
3421–3429.
11. Mishani, E.; Abourbeh, G.; Jacobson, O.; Dissoki, S.;
Revital Ben, D.; Rozen, Y.; Shaul, M.; Levitzki, A.
J. Med. Chem. 2005, 48, 5337–5348.
12. Andreas, S.; Jonathan, K.; Sumati, M.; Ralph, R.; Santi,
D. V. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 4234–4239.
13. Tsou, H.-R.; Mamuya, N.; Johnson, B. D.; Reich, M. F.;
Gruber, B. C.; Ye, F.; Nilakantan, R.; Shen, R.; Disca-
fani, C.; DeBlanc, R.; Davis, R.; Koehn, F. E.; Green-
berger, L. M.; Wang, Y.-F.; Wissner, A. J. Med. Chem.
2001, 2719–2734.
34. Wakeling, A. E.; Guy, S. P.; Woodburn, J. R.; Ashton, S.
E.; Curry, B. J.; Barker, A. J.; Gibson, K. H. Cancer Res.
2002, 62, 5749–5754.

3648
A. Wissner et al. / Bioorg. Med. Chem. 15 (2007) 3635–3648
35. Bold, G.; Altmann, K.-H.; Frei, J.; Lang, M.; Manley, P.
W.; Traxler, P.; Wietfeld, B.; Brueggen, J.; Buchdunger, E.;
Cozens, R.; Ferrari, S.; Furet, P.; Hofmann, F.; Martiny-
Baron, G.; Mestan, J.; Roesel, J.; Sills, M.; Stover, D.;
Acemoglu, F.; Boss, E.; Emmenegger, R.; Lasser, L.;
Masso, E.; Roth, R.; Schlachter, C.; Vetterli, W.; Wyss,
D.; Wood, J. M. J. Med. Chem. 2000, 43, 2310–2323.
39. Shewchuk, L.; Hassell, A.; Wisely, B.; Rocque, W.;
Holmes, W.; Veal, J.; Kuyper, L. F. J. Med. Chem.
2000, 43, 133–138.
40. McTigue, M. A.; Wickersham, J. A.; Pinko, C.;
Showalter, R. E.; Parast, C. V.; Tempczyk-Russell,
A.; Gehring, M. R.; Mroczkowski, B.; Kan, C.-C.;
Villafranca, J. E.; Appelt, K. Struct. Fold. Des. 1999,
7, 319–330.
41. Loganzo, F.; Hardy, C. A. Am. Biotech. Lab. 1998, 16, 26–
28.
42. Sybyl 7.2; Tripos: St. Louis, MO.
43. Eldridge, M. D.; Murray, C. W.; Auton, T. R.; Paolini, G.
V.; Mee, R. P. J. Comput. Aided Mol. Des. 1997, 11, 425–
445.
36. Hennequin, L. F.; Thomas, A. P.; Johnstone, C.; Stokes,
´
E. S. E.; Ple, P. A.; Lohmann, J.-J. M.; Ogilvie, D. J.;
Dukes, M.; Wedge, S. R.; Curwen, J. O.; Kendrew, J.;
Lambert-van der Brempt, C. J. Med. Chem. 1999, 42,
5369–5389.
37. Stamos, J.; Sliwkowski, M. X.; Eigenbrot, C. J. Biol.
Chem. 2002, 277, 46265–46272.
38. GLIDE;
Schrodinger
Software:
Portland,
OR,
44. Jorgensen, W. L.; Maxwell, D. S.; Tirado-Rives, J. J. Am.
Chem. Soc. 1996, 118, 11225–11236.
www.schrodinger.co.