Synthesis and antitumor evaluation of novel 5-substituted-4-hydroxy-8- nitroquinazolines as EGFR signaling-targeted inhibitors

Article abstract of DOI:10.1016/j.bmc.2005.05.045

The synthesis and biological activity of a series of novel 5-substituted-4-hydroxy-8-nitroquinazolines that may function as inhibitors of EGFR- and/or ErbB-2-related oncogenic signaling are described. These compounds were prepared by S_NAr reaction of 5-chloro-4-hydroxy-8- nitroquinazoline with alkyl or aryl amines, or alkyl alcohol as nucleophiles. Although the enzyme assay showed a weak inhibition effect against both EGFR and ErbB-2 tyrosine kinases, the cell-based antitumor activity turned out promising. Compounds having 5-anilino substituent exhibit high potency with 5-(4-methoxy)anilino-4-hydroxy-8-nitroquinazoline (1h) being the best dual EGFR/ErbB-2 inhibitors, which effectively inhibited the growth of both EGFR (MDA-MB-468, IC₅₀ < 0.01 μM) and ErbB-2 (SK-BR-3, IC ₅₀ = 13 μM) overexpressing human tumor cell lines in vitro. More interestingly, the variation of the substituent(s) at the 3- and/or 4-position of the 5-anilino portion was found to modulate the selectivity and potency dramatically. However, compounds having an alkylamino or alkyloxy group at the 5-position of 4-hydroxy-8-nitroquinazolines are essentially inactive. These results are consistent with molecular modeling observations. This study was the first attempt to identify new structural types of dual EGFR/ErbB-2-related signaling inhibitors by incorporation of the anilino group at the 5-position of 4-hydroxy-8-nitroquinazolines' core structure, providing promising new templates for further development of potent inhibitors targeting both EGFR and ErbB-2 tyrosine kinases.

Full text of DOI:10.1016/j.bmc.2005.05.045

Bioorganic & Medicinal Chemistry 13 (2005) 5613–5622
Synthesis and antitumor evaluation of novel
-substituted-4-hydroxy-8-nitroquinazolines as EGFR
signaling-targeted inhibitors
5
a
Yi Jin, Hui-Yuan Li, Li-Ping Lin, Jinzhi Tan, Jian Ding,
a
b
c
b
c
Xiaomin Luo and Ya-Qiu Long
a,
*
a
State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica, Shanghai Institutes for Biological Sciences,
Graduate School of the Chinese Academy of Sciences, CAS, 555 Zuchongzhi Road, Shanghai 201203, PR China
b
Division of Anti-tumor Pharmacology, Shanghai Institute of Materia Medica, Shanghai Institutes for Biological Sciences,
CAS, 555 Zuchongzhi Road, Shanghai 201203, PR China
c
Drug Discovery and Design Center, Shanghai Institute of Materia Medica, Shanghai Institutes for Biological Sciences,
Graduate School of the Chinese Academy of Sciences, CAS, 555 Zuchongzhi Road, Shanghai 201203, PR China
Received 29 March 2005; revised 25 May 2005; accepted 26 May 2005
Available online 29 June 2005
Abstract—The synthesis and biological activity of a series of novel 5-substituted-4-hydroxy-8-nitroquinazolines that may function as
inhibitors of EGFR- and/or ErbB-2-related oncogenic signaling are described. These compounds were prepared by S Ar reaction of
-chloro-4-hydroxy-8-nitroquinazoline with alkyl or aryl amines, or alkyl alcohol as nucleophiles. Although the enzyme assay
N
5
showed a weak inhibition effect against both EGFR and ErbB-2 tyrosine kinases, the cell-based antitumor activity turned out prom-
ising. Compounds having 5-anilino substituent exhibit high potency with 5-(4-methoxy)anilino-4-hydroxy-8-nitroquinazoline (1h)
being the best dual EGFR/ErbB-2 inhibitors, which effectively inhibited the growth of both EGFR (MDA-MB-468,
IC50 < 0.01 lM) and ErbB-2 (SK-BR-3, IC50 = 13 lM) overexpressing human tumor cell lines in vitro. More interestingly, the var-
iation of the substituent(s) at the 3- and/or 4-position of the 5-anilino portion was found to modulate the selectivity and potency
dramatically. However, compounds having an alkylamino or alkyloxy group at the 5-position of 4-hydroxy-8-nitroquinazolines
are essentially inactive. These results are consistent with molecular modeling observations. This study was the first attempt to iden-
tify new structural types of dual EGFR/ErbB-2-related signaling inhibitors by incorporation of the anilino group at the 5-position of
4
-hydroxy-8-nitroquinazolinesÕ core structure, providing promising new templates for further development of potent inhibitors tar-
geting both EGFR and ErbB-2 tyrosine kinases.
2005 Elsevier Ltd. All rights reserved.
ꢀ
1
. Introduction
exist as inactive monomers, which dimerize after ligand
activation. This causes homodimerization or heterodi-
merization between EGFR and another member of the
ErbB receptor family, with HER-2 being a preferred
Growth factor signaling pathways play a fundamental
role in regulating key cellular functions, including cell
proliferation, differentiation, metastasis, and survival.
An important mediator of growth-factor signaling path-
1
2
heterodimeric partner. After ligand binding, the tyro-
sine kinase intracellular domain of the receptor is acti-
vated, with autophosphorylation of the intracellular
domain, which initiates a cascade of intracellular events
including activation of ras and mitogen-activated pro-
tein kinase followed by the activation of several nuclear
proteins including cyclin D1, a protein required for cell
cycle progression from G1 to S phase. The EGFR sig-
naling is important for tumor cell proliferation, inhibi-
tion of apoptosis, angiogenesis, metastasis, and
sensitivity to chemotherapy and radiotherapy. Overex-
pression of EGFR and ErbB-2 is common in a variety
1
ways is the epidermal growth factor receptor (EGFR).
EGFR is a member of a family of four closely related
receptors: EGFR (or ErbB-1), HER-2/neu (ErbB-2),
HER-3 (ErbB-3), and HER-4 (ErbB-4). The receptors
Keywords: EGFR inhibitor; Quinazolines; ErbB-2; Breast cancer;
Cellular signaling.
*
Corresponding author. Tel.: +86 21 50806876; fax: +86 21
0807088; e-mail: yqlong@mail.shcnc.ac.cn
5
0
968-0896/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.05.045

5614
Y. Jin et al. / Bioorg. Med. Chem. 13 (2005) 5613–5622
of major human solid tumors and has been correlated
3
with poor prognosis in patients. Consequently, inhibi-
para-position of the 5-substituent of the quinazoline
ring to see if it enhanced the interaction with the
ErbB-2 tyrosine kinase. The aim of the study was to find
new structural types of tyrosine kinase inhibitors which
specifically target both EGFR and ErbB-2 TK. Dual
inhibition of EGFR and ErbB-2 may offer increased
activity over agents which target only one of these recep-
tor kinases.
tors of the EGFR and/or HER-2 have emerged as prom-
ising anticancer agents, which was validated by the
recent success in the clinical evaluation of EGFR/
4
HER-2 TK inhibitors.
Two therapeutic approaches have been used in clinical
studies to inhibit the EGFR family members: human-
ized monoclonal antibodies (MAbs), and small molecule
inhibitors of the EGFR tyrosine kinase (TKIs). MAbs
are generally directed at the ectodomain of the EGFR
to block ligand binding and receptor activation; TKIs
prevent the autophosphorylation of the intracellular
tyrosine kinase domain of the EGFR. The most promis-
ing small molecule selective EGFR-TKIs are currently
three series of compounds, which include 4-anilinoqui-
nazolines, 4-[ar(alk)ylamino]pyridopyrimidines, and 4-
In this present work, we would like to report the synthe-
sis and biological activity of a series of novel 5-substitut-
ed quinazoline derivatives represented by the general
formula of 1 in Figure 1. The enzyme inhibitory activity
as well as cellular activity in relevant tumor cell lines will
be discussed to develop the structure–activity relation-
ship of this new series. This work was the first attempt
to explore the effect of 5-substitution on the EGFR/
ErbB-2 kinase activity of the 4-hydroxyquinazoline ser-
ies. Significantly, several of these compounds showed
promising anti-proliferative effect against the EGFR
and ErbB-2-overexpressing tumor cell lines.
5
phenylaminopyrrolo-pyrimidines (Fig. 1). Of these,
the anilinoquinazolines are the most developed class of
inhibitors. Figure 1 includes some examples in the qui-
nazoline series that are currently approved drugs or in
clinical trials.
2. Chemistry
The quinazoline scaffold provides the necessary binding
properties for inhibition of the ErbB family of tyrosine
We developed an efficient and facile synthesis approach
to prepare a variety of quinazoline derivatives with
various C-5 substituents. As depicted in Scheme 1,
the straightforward six-step synthetic route allowed us
to diversify position 5 of the quinazoline moiety via
the key intermediate 6 at a later stage. Beginning with
the commercially available 3-chloro-2-methyl aniline,
the acetylation with acetyl chloride gave the protected
aniline 2 in good yield (90%) using simple filtration iso-
lation. Oxidation of the methyl group of 2 with
KMnO4 and MgSO4 in refluxing water yielded 3.
Deacetylation of 3 in concentrated HCl solution at
the temperature below 90 ꢁC afforded 4 in excellent
yields. The 5-chloroquinazoline 5 was prepared via
Neimentowski synthesis by refluxing 4 in formamide,
6
kinases. However, most of the structure–activity rela-
tionship studies were focused on the substitutions in
the 6- and 7-positions of the quinazoline core structure
0
0
as well as the substitutions at the 3 - and 4 -positions
7
–11
of the 4-anilino portion.
Few data have been pub-
lished describing the antitumor activity of 5-substitut-
ed-4-hydroxyquinazolines. Since small substitution
changes in kinase inhibitors greatly affect the kinase
inhibition and drug properties, we were intrigued to
determine the effect on the EGFR/ErbB-2 kinase activi-
ty resulting from the substitution at the 5-position of 4-
hydroxyquinazoline core structure. Enlightened by the
recently reported small molecule ErbB-2 TK inhibitor,
1
2
namely B-17 (Fig. 1), we put nitro group at the
Figure 1. Representative small-molecule inhibitors of the tyrosine kinase domain of the EGFR. The quinazoline numbering convention is indicated.

Y. Jin et al. / Bioorg. Med. Chem. 13 (2005) 5613–5622
5615
Scheme 1. Reagents and conditions: (i) CH
8%; (iv) formamide, 130 ꢁC for 45 min, 175 ꢁC for 75 min, 67.6%; (v) conc. H
THF, reflux, 12–24 h, 50–80%.
3
COCl/CH
2
Cl
2
, rt, 6 h, 90.2%; (ii) KMnO
4
, MgSO /H
4
2
O, reflux, 5 h, 65.7%; (iii) conc. HCl, 85 ꢁC, 1.5 h,
, 0 ꢁC, 10 min, 51.6%; (vi) RNH or ROH/RONa,
9
2
SO , conc. HNO
4
3
2
followed by an extractive basic workup. Treatment of 5
with the mixture solution of conc. H SO acid and
either ErbB-2 or EGFR to a significant degree. Com-
pounds known to inhibit ErbB-2 or EGFR tyrosine
kinases were used as a positive control for the assays.
The biological results for the 5-substituted-4-hydroxy-
quinazoline inhibitors are shown in Tables 1 and 2.
2
4
fuming HNO3 acid furnished the selective nitration
product 6. Nucleophilic displacement of the key inter-
mediate 6 with the appropriate secondary amine or
alcohol in basic refluxing THF solution generated the
desired final products, 5-substituted-4-hydroxy-8-nitro-
quinazolines (1a–n) in good yields (50–80%). The elec-
tron-withdrawing group nitro at the 8-position of the
quinazoline ring is beneficial for the S Ar displacement
N
reaction, and endows a large tolerance for the substitu-
tion in the 5-position of the quinazoline.
Table 1. Enzyme inhibitory activity of 4-hydroxy-8-nitroquinazolines
with various C-5 substituents
Compound
Enzyme assays, inhibition % at
a
1
0 lM
EGFR
ErbB-2
1
1
a
b
17.7
5.2
17.1
18.3
0
2.4
10.8
0.3
0
3
. Results and discussion
1c
1
1
1
1
1
1
1
1
d
e
f
4.8
10.5
0
Kinase inhibitory activity of the compounds was evalu-
ated using EGFR and ErbB-2 kinase activity assays by
ELISA. The effect of the compounds on cell prolifera-
tion was measured using human tumor cell lines highly
related with the aberrant EGFR signaling by sulforho-
0
g
h
i
3.9
12.2
0
1.4
0
j
21.1
0
7.2
4.1
6.6
3.7
6.8
2.3
1
3
damine B assay. Three human carcinoma cell lines
were chosen for the cell proliferation assay: MDA-
MB-468 (breast), which overexpresses EGFR, was used
to determine the effectiveness of the EGFR TK inhibito-
ry properties; SK-BR-3 (breast) overexpresses ErbB-2
and to a lesser extent EGFR and should be potently
inhibited by a dual ErbB-2/EGFR TK inhibitor;
MDA-MB-435 (breast) which is believed not to express
k
1l
7.9
5.6
6.5
28.0
1m
1n
B17
a
The percent inhibition of the kinase activity was generated by mea-
suring the inhibition of phosphorylation of a peptide substrate added
to enzyme reaction in the presence of 10 lm inhibitor.

5616
Y. Jin et al. / Bioorg. Med. Chem. 13 (2005) 5613–5622
Table 2. Inhibitory effect of compounds 1a–n on the growth of tumor cell lines
a
Compound
R
Tumor cell IC50 (lM)
b
b
b
MDA-MB-468
0.48
SK-BR-3
MDA-MB-435
1a
1b
1c
33.4% inhibition at 10 lM
No inhibition at 10 lM
8.0
10.0% inhibition at 10 lM
18.5% inhibition at 10 lM
42.8% inhibition at 10 lM
12.1% inhibition at 10 lM
32.5% inhibition at 10 lM
1
1
d
e
28.6% inhibition at 10 lM
15.2% inhibition at 10 lM
32.7% inhibition at 10 lM
No inhibition at 10 lM
38.8% inhibition at 10 lM
15.8% inhibition at 10 lM
1
f
35.8% inhibition at 100 lM
32.3% inhibition at 10 lM
13.4% inhibition at 10 lM
No inhibition at 10 lM
17.4% inhibition at 10 lM
No inhibition at 10 lM
1
g
1
h
<0.01 (75.5% inhibition at 0.01 lM)
13.0
<0.01 (76.0% inhibition at 0.01 lM)
1
1
i
j
4.0
4.2
No inhibition at 10 lM
No inhibition at 10 lM
7.0
29.0% inhibition at 10 lM
1
1
k
l
23.3% inhibition at 100 lM
34.6% inhibition at 10 lM
19.5% inhibition at 100 lM
No inhibition at 10 lM
No inhibition at 10 lM
No inhibition at 10 lM
1
1
m
19.9% inhibition at 100 lM
29.2% inhibition at 100 lM
No inhibition at 10 lM
No inhibition at 10 lM
No inhibition at 10 lM
No inhibition at 10 lM
n
a
Cell line SK-BR-3 (breast carcinoma) overexpresses ErbB-2; cell line MDA-MB-468 (breast carcinoma) overexpresses EGFR; cell line MDA-MB-435 (breast
carcinoma) is believed not to express EGFR or ErbB-2 to a significant extent.
b
Dose–response curves were determined at five concentrations. The IC50 values are the concentrations in micromolar needed to inhibit cell growth by 50% as
determined from these curves.

Y. Jin et al. / Bioorg. Med. Chem. 13 (2005) 5613–5622
5617
The enzyme assay data are summarized in Table 1. Dis-
appointingly, most compounds just display low to mod-
erate inhibition against the enzyme activity of EGFR
and ErbB-2 kinases at the concentration of 10 lM. To
some extent, compounds 1a, 1c, 1d, and 1j exhibited a
better ability to inhibit EGFR kinase than ErbB-2 ki-
nase, whereas 1j is a dual inhibitor of EGFR and
ErbB-2 kinases. However, the cellular assay turned out
more encouraging (Table 2). The 5-substituted quinazo-
line compounds show much more effectiveness in inhib-
iting the growth of the tumor cell lines known to express
high levels of EGFR or ErbB-2. The inconsistency be-
tween the enzyme activity and the cellular efficiency
could suggest that the new type of quinazoline com-
pounds might function by inhibiting multiple key pro-
teins involved in the EGFR signaling pathways, not
only targeting the tyrosine kinase, thus leading to the
significant antiproliferation effect against EGFR depen-
dent tumor cell lines, which is highly relevant to the
overexpression of EGFR or ErbB-2. The definite mech-
anism is still under study.
substitution at the 3-position (1e) completely abolished
the activity. With respect to the substituent effect at the
4-position of the anilino portion, the small lipophilic sub-
stituent was found to be detrimental to the activity of our
5-anilinoquinazoline series, e.g., the 4-chloro compound
(1d) demonstrated much less activity in the two cell prolif-
eration assays. Encouragingly, further elaboration of the
4-substitution in the anilino ring with electron-donating
groups showed a remarkable positive response. The lipo-
philic electron-donating groups at the 4-position were
beneficial, with the 4-methoxy and 4-benzyloxy deriva-
tives (1h, 1j) being the most potent dual inhibitors of
EGFR and ErbB-2 signaling. The 4-methoxy-anilino sub-
stitution afforded the best dual inhibitor of the growth of
both target cell lines (1h) (IC50 6 0.01 and 13.0 lM, for
EGFR and ErbB-2, respectively). The large benzyloxy
substitution at the 4-position (1j) conferred a better inhib-
itor for ErbB-2 (IC50 = 7.0 lM) but an inferior inhibitor
for EGFR overexpression cell line (IC50 = 4.2 lM) com-
pared to 1h, suggesting that ErbB-2 inhibition prefers
0
large substitution at the 4 -position of the 5-anilino substi-
0
tuent. In contrast, the small hydrophilic group at the 4 -
position of the 5-aniline ring was disfavored for the
So, we used the cellular efficiency to investigate the SAR
to determine the preferred substitutions of this novel
quinazoline series as inhibitors of the EGFR-related tu-
mors. Conventional quinazoline-based EGFR TK
inhibitor requires an anilino substituent at the 4-position
of the quinazoline scaffold for retaining high potency,
thus 4-anilinoquinazolines are the most developed class
0
ErbB-2 binding, whereby the 4 -amine group greatly re-
duced the activity for the ErbB-2 while retaining moder-
ate potency against EGFR (IC50 = 4.0 lM). So far, it is
very apparent that for the novel 5-anilino-4-hydroxyqui-
nazoline inhibitors of EGFR/ErbB-2-related tumors, the
target affinity and selectivity could be modulated to a
great extent via the substitutions on 5-anilino moiety,
with the position and the nature of the substituent being
the determining factors.
4
of drugs that inhibit EGFR kinase intracellularly. In
our study with the 5-substituted-4-hydroxy-8-nitroqui-
nazoline series, a similar observation was made; namely,
the compounds not containing an anilino moiety at the
5
-position such as 1b, 1f, 1g, and 1k–n, are significantly
It was also evident that the 5-substituted quinazoline
compounds are a much better inhibitor of the MDA-
MB-468 and SK-BR-3 cell lines than the MDA-MB-
435 line, except for compound 1h. This is consistent with
our proposed mechanism of cell growth inhibition being
reliant, to some degree, on EGFR signaling. The obser-
vation that the most potent inhibitor 1h does inhibit the
growth of the MDA-MB-435 could suggest that this line
has some dependence on EGFR or ErbB-2, even though
these receptors are not expressed to a significant extent.
less potent than the compounds having a 5-anilino sub-
stitution, such as 1a, 1c, 1h, 1i, and 1j, with respect to the
inhibition of the proliferation of MDA-MB-468 and
SK-BR-3 cell lines. Either 5-alkoxy (1f, 1g) or 5-alkyla-
mino (including the benzylamine) compounds (1b, 1k–n)
are essentially inactive. The results suggest that the ani-
lino portion may play a strong role in determining the
potency and selectivity of the quinazoline series as ki-
nase inhibitors, regardless of the substitution position
(
4- or 5-) on the quinazoline ring. It is noteworthy that
the substitution in the anilino portion is another impor-
tant factor for the activity.
4. Molecular modeling
It is of interest to compare the inhibitory activities of the
5
the 3 - and 4 -positions of the anilino moiety. The unsub-
stituted 5-aniline compound 1a displayed excellent activ-
ity for the EGFR overexpressing cell line MDA-MB-468
The earlier observations can be nicely rationalized from
the results of molecular modeling experiments. Recently
the X-ray crystal structure of EGFR kinase having a
bound quinazoline-based ligand (GW572106) was report-
ed. GW572016 was a potent dual EGFR/ErbB-2 inhib-
itor, so the published coordinates from this X-ray
structure was used as the starting point for the modeling
studies in the present study. The automated molecular
docking can produce several optional conformation of
the inhibitor. The conformation of the inhibitor 1h corre-
sponding to the lowest binding energy was selected as the
possible binding conformation, shown in Figure 2.
-anilinoquinazoline series with various substitutions at
0
0
1
5
(
IC50 = 0.48 lM) but was much less effective to inhibit the
ErbB-2 overexpressing cell line SK-BR-3 (IC50 > 10 lM).
Interestingly, the selectivity was completely inversed by
the introduction of 3-chloro substituent at the aniline
ring, resulting in a selective ErbB-2 inhibitor (1c) with im-
proved activity for the SKBR3 cell line (IC50 = 8.0 lM)
but poor potency in inhibiting the growth of MDA-MB-
4
groups in the 3-position of the aniline ring was reported
for EGFR inhibition. In our case, 3-substitution seems
more important for ErbB-2 inhibition. But the methyl
68 cell line (IC50 > 10 lM). The utility of small lipophilic
In the final model (Fig. 2), the N1 and N3 atoms of 1h
are hydrogen-bonded to the backbone carbonyl oxygen
of Phe832 and the sidechain carboxyl oxygen of Asp831,
1
4

5618
Y. Jin et al. / Bioorg. Med. Chem. 13 (2005) 5613–5622
Figure 2. Two-dimensional representative of the interaction model of compound 1h with the kinase domain of EGFR. This image was generated
16
with LIGPLOT program.
respectively. And the quinazoline ring forms hydropho-
bic interactions with residues Met742, Leu753, and
Leu834. Of particular significance to the case of 5-
substituted-8-nitroquinazoline series, the 8-nitro group
of 1h is involved in the interaction with the active site
of EGFR via its oxygen atom forming hydrogen bond
with the backbone NH of Leu753, and the 5-(4-methox-
yaniline) moiety lies in a hydrophobic pocket containing
Val702, Thr766, Ala719, Arg752, Leu820, and Leu768.
We believe that this might account for the large differ-
ence in activity we see between the 5-alkoxy(or alkyami-
no-) and 5-anilinoquinazolines.
of the inhibitory activity against the proliferation of two
human carcinoma cell lines known to express high levels
of EGFR and ErbB-2. For the new 5,8-disubstituted
quinazoline derivatives, the 5-anilino substitution is
essential for its activity. Furthermore, the substitution
on the 5-anilino portion has a substantial effect on the
potency and selectivity of the resulting compound with
respect to the inhibition of EGFR and ErbB-2 receptors.
The 5-anilinoquinazoline (1a) was a selective inhibitor of
EGFR, whereas the 5-(3-chloroanilino)-quinazoline (1c)
displayed selective inhibition of ErbB-2. More signifi-
cantly, two potent dual inhibitors (1h and 1j) of EGFR
and ErbB-2 were discovered by the alkoxy substitution
0
This binding mode is not similar to the original ligand
GW-572016, which is much larger than the inhibitor
at the 4 -position of 5-anilino substituent. A binding
mode of these compounds autodocked into the active
site of an EGFR kinase domain is helpful for explaining
the SAR of 5-aniline. This study was the first attempt to
identify new structural types of EGFR/ErbB-2 signaling
inhibitors, by the incorporation of the anilino group at
the 5-position of 4-hydroxy-8-nitroquinazolinesÕ core
structure, providing promising new templates for further
development of potent inhibitors targeting both EGFR
and ErbB-2 tyrosine kinases.
1
h. GW572016 has a bulky 6-anilino substituent reach-
ing deep into an opened back pocket, which may result
in the slow-off rate in dissociation. From the binding
mode of 1h with EGFR, we found that it just occupied
the sub-pocket, suggesting why the 5-substituted-8-
nitroquinazolines are not potent with respect to
EGFR/ErbB-2 tyrosine kinase assay. This model will
be helpful for our further structural elaboration of the
novel quinazoline series to improve the kinase activity.
6. Experimental
5. Conclusions
6
.1. Molecular modeling
We have synthesized and assessed a series of novel 5-
substituted-4-hydroxy-8-nitroquinazolines that may
function as inhibitors of EGFR/ErbB-2-mediated cellu-
lar signaling pathways. The SAR was discussed in terms
The crystal structure of kinase domain of EGFR in
complex with its inhibitor GW572016 (PDB entry code
1
5
1XKK)
was recovered from Brookhaven Protein

Y. Jin et al. / Bioorg. Med. Chem. 13 (2005) 5613–5622
5619
Database (PDB). The missed atoms and residues were
1
(0.03% H O , 2 mg/mL o-phenylenediamine in citrate
2 2
7
modeled by Sybyl 6.8. The kinds of atomic charges
8
buffer 0.1 M, pH 5.5) was added and incubated at room
temperature until color emerged. The reaction was ter-
minated by the addition of 100 lL of 2 M H SO , and
1
were taken as Kollman-united-atom for the macro-
1
molecule and Gasteiger–Marsili for the inhibitor.
9
2
4
A492 was measured using a multiwell spectrophotome-
TM
ter (VERSAmax ). The inhibition rate (%) was calcu-
To find the binding mode of the compound 1h to the ki-
nase domain of EGFR, the advanced docking program
lated using the equation:
The inhibition rateð%Þ ¼ ½1 ꢀ ðA492=A492controlÞꢁ
ꢂ 100%.
2
0,21
Autodock 3.0.3
inhibitor to the enzyme. The Lamarckian genetic algo-
was used to automatically dock the
2
0
rithm (LGA) was applied to deal with the inhibitor–
enzyme interaction. A Solis and Wets local search
performed the energy minimization on a user-specified
proportion of the population. The docked conforma-
tions of the inhibitor were generated after a reasonable
number of evaluations.
6.2.2. Cell growth inhibition assay. Two human carci-
noma cell lines, SKBR3 (breast carcinoma) and MDA-
MB-468 (breast carcinoma), that were obtained from
American Type Culture Collection (Rockville, MD),
were used for the cell proliferation assay. Cells were
maintained in RPMI-1640 medium supplemented with
The whole docking operation in this study could be stat-
ed as follows. First, the kinase domain of EGFR was
checked for polar hydrogen and assigned for partial
atomic charges, then a PDBQ file was created, and then
the atomic solvation parameters were also assigned for
the macromolecule. Meanwhile, some of the torsion an-
gles of the inhibitor that would be explored during
molecular docking stage were defined, allowing the con-
formation search for the ligand during the docking pro-
cess. Second, the grid map with 60 · 60 · 60 points and
1
1
0% (v/v) fetal bovine serum, 2 mmol/L glutamine,
00 U/mL penicillin, and 100 lg/mL streptomycin (Gib-
co, Grand Island, NY, USA) in a highly humidified
atmosphere of 95% air with 5% CO2 at 37 ꢁC. The
growth inhibition was analyzed by the sulforhodamine
1
2
B (SRB, Sigma) assay. Briefly, the cells were seeded
at 6000 cells/well in 96-well plates (Falcon, CA, USA),
and allowed to attach overnight. The cells were treated
in triplicate with grade concentrations of compounds
at 37 ꢁC for 72 h. Then, they were fixed with 10% tri-
chloroacetic acid and incubated for 60 min at 4 ꢁC.
Then, the plates were washed and dried. SRB solution
˚
a spacing of 0.375 A was calculated using the AutoGrid
2
0
program, to evaluate the binding energies between the
inhibitor and the macromolecule. Third, some impor-
tant parameters for LGA calculations were reasonably
set on. Not only the atom types, but also the generations
and the number of runs for the LGA algorithm were
edited and properly assigned according to the require-
ment of the Amber force field. The maximum number
of generations, energy evaluations, and docking runs
(0.4% w/v in 1% acetic acid) was added and the culture
was incubated for an additional 15 min. After the plates
were washed and dried, bound stain was solubilized with
Tris buffer, and the optical densities were read on the
plate reader (model VERSA Max, Molecular Devices)
at 515 nm. The growth inhibitory rate of treated cells
5
7
were set to 3.0 · 10 , 1.5 · 10 , and 30, respectively.
Finally, the docked complex of the inhibitor–enzyme
for the inhibitor was selected according to the criterion
of interaction energy combined with geometrical match-
ing quality.
was calculated by the following formula: [1 ꢀ (A
515
treated/A control)] · 100%. The results were also ex-
515
pressed as IC50 (the compound concentration required
for 50% growth inhibition of tumor cells), which was
calculated by the Logit method. The mean IC50 was
determined from the results of three independent tests.
6
.2. Biology
6.3. Chemistry
6
.2.1. EGFR and ErbB-2 kinase activity assays by
ELISA. The assay was performed in 96-well plates
pre-coated with 20 lg/mL Poly(Glu, Tyr)4:1 (Sigma) as
a substrate. In each well, 85 lL of 8 lM ATP solution
and 10 lL of the title compound were added at varying
concentrations. AG825 and PD153035 were used as a
positive control for ErbB-2 and EGFR kinase, respec-
tively, and 0.1% (v/v) DMSO was the negative control.
Experiments at each concentration were performed in
triplicate. The reaction was initiated by adding 5 lL of
ErbB-2-CD or EGFR kinase. After incubation for 1 h
at 37 ꢁC, the plate was washed three times with PBS con-
taining 0.1% Tween 20 (T-PBS). Next, 100 lL anti-
phosphotyrosine (PY99) (1:500 dilution) antibody was
added. After 1 h of incubation at room temperature,
the plate was washed three times. Goat anti-mouse
IgG horseradish peroxidase (100 lL) (1:2000 dilution)
diluted in T-PBS containing 5 mg/mL BSA was added.
The plate was reincubated at room temperature for
1
Unless otherwise stated, the H NMR spectra were
recorded on a Varian 400-MHz spectrometer. The data
are reported in parts per million relative to TMS and
referenced to the solvent in which they were run. Ele-
mental analyses were obtained using a Vario EL spec-
trometer.
Melting
points
(uncorrected)
were
determined on a Buchi-510 capillary apparatus. Low
resolution mass spectra (EI) was determined on Finni-
gan MAT-95 mass spectrometer. The solvent was re-
moved by rotary evaporation under reduced pressure,
and flash column chromatography was performed on
silica gel H (10–40 lm). Anhydrous solvents were ob-
tained by redistillation over sodium wire.
6.3.1. N-(3-Chloro-2-methylphenyl)-acetamide (2). To the
solution of 3-chloro-2-methyl aniline (2.34 g,
16.50 mmol) in CH Cl (6 mL) was added acetic anhy-
dride (2.53 g, 24.75 mmol) with stirring. The resultant
2
2
1
h, and washed as before. Finally, a 100 lL solution

5620
Y. Jin et al. / Bioorg. Med. Chem. 13 (2005) 5613–5622
large gray precipitate of crude acetylated product was
filtered with saction and washed with water. Recrystal-
ization from EtOAc/MeOH = 10:1 afforded a pure white
6.3.5. 5-Anilino-8-nitroquinazolin-4-ol (1a). To the solu-
tion of compound 6 (100 mg, 0.44 mmol) and 18 mL
THF was added aniline (82 mg, 0.88 mmol). The mix-
ture was refluxed for 24 h at 80–90 ꢁC. The solution
was concentrated and diluted with water (15 mL). After
adding the HCl solution (20%) until pH 7, the resultant
solution was extracted with EtOAc (2 · 20 mL), dried
1
crystal (2.73 g, 90.2%). H NMR (DMSO-d , 400 MHz),
6
d 9.70 (s, 1H), 7.40 (d, J = 8.40 Hz, 1H), 7.37 (t,
J = 8.01 Hz, 1H), 1.99 (s, 3H); MS (EI) m/z 183 (M );
+
Anal. Calcd. for (C H ClNO) C, H, N: 58.86, 5.49,
9
10
7
.63.
over MgSO , and then evaporated. The residue was
4
purified by column chromatography (EtOAc/PE = 1:1)
affording brown solid (90 mg, 72.5%). mp: 233–235 ꢁC;
6.3.2. 2-(Acetylamino)-6-chlorobenzoic acid (3). A mix-
ture of compound 2 (3.3 g, 18.0 mmol), magnesium sul-
fate heptahydrate (4.3 g, 38.8 mmol), and water
1
H NMR (DMSO-d , 300 MHz), d 9.88 (br, 1H), 8.81
6
(s, 1H), 8.80 (d, J = 9.08 Hz, 1H), 7.34 (d, J = 8.9 Hz,
1H), 7.0–7.2 (m, 5H); MS (EI) m/z 282 (M ); Anal.
+
(
80 mL) was heated to reflux (135 ꢁC). A solution of
potassium permanganate (11.4 g, 72.2 mmol) and water
100 mL) was added portionwise over 3 h. After the fi-
Calcd. for (C H N O ) C, H, N: 59.57, 3.57, 19.85.
1
4
10
4
3
(
Found: 59.45, 3.44, 19.98.
nal addition, the mixture was refluxed for 2 h (160 ꢁC),
and the cooled mixture was filtered. The filtrate was
acidified with hydrochloric acid (6 M) to pH 1.0. The
crude product was precipitated as a white crystal that
was collected by filtration and dried under high vacu-
um. Recrystalization from MeOH afforded a pure
6.3.6. 5-(Benzylamino)-8-nitroquinazolin-4-ol (1b). The
compound 1b was prepared in a similar procedure for
1a, except for using benzylamine (94 mg, 0.88 mmol),
affording orange solid (100 mg, 76.9%). mp: 182–
1
184 ꢁC; H NMR (DMSO-d , 300 MHz), d 10.00 (br,
6
1
white solid product (2.5 g, 65.7%). H NMR (DMSO-
d6, 400 MHz), d 9.70 (s, 1 H), 7.40 (d, J = 8.04 Hz,
1H), 8.68 (d, J = 9.1 Hz, 1H), 7.2–7.3 (m, 5H), 6.77 (d,
J = 8.99 Hz, 1H), 4.25 (s, 2H); MS (EI) m/z 296 (M );
+
1
H), 7.37 (t, J = 8.01 Hz, 1H), 7.31 (d, J = 8.06 Hz,
Anal. Calcd. for (C H N O ) C, H, N: 60.81, 4.08,
1
18.91. Found: 60.72, 4.11, 18.87.
5
12
4
3
+
H), 1.99 (s, 3H); MS (EI) m/z 213 (M ); Anal. Calcd.
1
for (C H ClNO ) C, H, N: 50.60, 3.77, 6.56. Found:
9
8
3
5
0.42, 3.54, 6.45.
6.3.7. 5-[(3-Chlorophenyl)amino]-8-nitroquinazolin-4-ol
(1c). The compound 1c was prepared as described for
1a, except for using 3-chlorobenzenamine (112 mg,
0.88 mmol), providing an yellow solid (110 mg, 79.5%).
6
1
.3.3. 5-Chloroquinazolin-4-ol (5). Compound 3 (250 mg,
.17 mmol) with 4 mL concentrated hydrochloric acid
1
was heated for 1.5 h in an oil-bath at 80–90 ꢁC, not high-
er than 90 ꢁC, providing a white crystal on cooling. The
solid was filtrated and recrystalized from MeOH to pro-
vide the 2-amino-6-chlorobenzoic acid (4), which was
used directly for the next step. The mixture of 4 (1.2 g,
mp: 235–237 ꢁC; H NMR (DMSO-d , 300 MHz), d
6
12.90 (br, 1H), 11.45 (s, 1H), 8.27 (s, 1H), 8.23 (d,
J = 9.07 Hz, 1H), 7.32 (t, J = 7.93 Hz, 1H), 7.16 (d,
J = 9.07 Hz, 1 H), 7.16–7.07 (m, 2 H), 6.95 (d,
+
J = 7.96 Hz, 1 H); MS (EI) m/e 316 (M ); Anal. Calcd.
for (C H ClN O ) C, H, N: 53.09, 2.86, 17.69. Found:
53.24, 2.97, 17.57.
5
.8 mmol) and 1.5 mL formamide was heated in an oil
1
4
9
4
3
bath at 130 ꢁC for 45 min and at 175 ꢁC for 75 min.
The cooled semi-solid mass was slurried with 1.5 mL
of methyl-cellosove and dumped in 5 mL of cold water
to yield a crude product. Purification from silica gel
chromatography using EtOAc/PE = 1:1 afforded a white
solid (0.75 g, 67.6%). H NMR (DMSO-d , 400 MHz), d
6.3.8. 5-[(4-Chlorophenyl)amino]-8-nitroquinazolin-4-ol
(1d). The compound 1d was prepared as described for
1a, except for using 4-chloroaniline (112 mg,
1
0.88 mmol), yielding an yellow solid (106 mg, 76.8%).
6
1
1
1
1
2.09 (br, 1H), 8.07 (s, 1H), 7.71 (t, J = 8.25 Hz, 7.7,
H), 7.59 (d, J = 8.0 Hz, 1H), 7.51 (d, J = 7.43 Hz,
mp: 241–243 ꢁC; H NMR (DMSO-d , 300 MHz), d
6
11.48 (br, 1H), 8.28 (s, 1H), 8.20 (d, J = 9.08 Hz, 1H),
7.32 (d, J = 8.51 Hz, 2H), 7.10 (d, J = 9.06 Hz, 1H),
6.99 (d, J = 8.52 Hz, 2H); MS (EI): 317 (M + 1); Anal.
+
H); MS (EI) m/z 180 (M ); Anal. Calcd. for
+
(
C H ClN O) C, H, N: 53.21, 2.79, 15.51. Found:
5
8
2
5
3.14, 2.72, 15,36.
Calcd. for (C H ClN O ) C, H, N: 53.09, 2.86, 17.69.
1
4
9
4
3
Found: 52.86, 3.21, 17.35.
6
.3.4. 5-Chloro-8-nitroquinazoline-4-ol (6). The solution
of compound 5 (155 mg, 0.86 mmol) in concentrated
H SO (3 mL) cooled to 0 ꢁC was treated dropwisely
6.3.9. 5-[(3-Methylphenyl)amino]-8-nitroquinazolin-4-ol
(1e). The compound 1e was prepared analogously to
1a, except for using m-toluidine (95 mg, 0.88 mmol),
2
4
(
10 min) with fuming nitric acid (HNO , 3 mL). The
3
reaction mixture was stirred for 10 min (0 ꢁC) before
being poured into crushed ice. The mixture was neutral-
ized with 20% KOH with ice cooling. The product was
affording a black brown solid (185 mg, 71.2%). mp:
1
215–217 ꢁC; H NMR (DMSO-d , 300 MHz), d 12.85
6
(br, 1H), 11.42 (s, 1H), 8.26 (s, 1H), 8.18 (d,
J = 9.89 Hz, 1H), 7.15 (t, J = 7.69 Hz, 1H), 7.08 (d,
J = 8.79 Hz, 1H), 6.93 (d, J = 9.89 Hz, 1H), 6.78 (s,
1H), 6.76 (d, J = 8.79 Hz, 1H), 2.23 (s, 3H); MS (EI)
collected by filtration and washed with H O
2
(
1
(
2 · 6 mL). Recrystalization from EtOH–H O gave
2
00 mg of pure compound 6 as a light yellow solid
1
+
193.5 mg, 51.6%). H NMR (DMSO-d , 400 MHz), d
m/e = 296 (M ); Anal. Calcd. for (C H N O ) C, H,
15 12
6
4
3
1
2.68 (br, 1 H), 8.27 (d, J = 8.86 Hz, 1H), 8.22 (s, 1H),
N: 60.81, 4.08, 18.91. Found: 60.92, 4.11, 18.76.
+
.75 (d, J = 8.85 Hz, 1H); MS (EI) m/z 225 (M ); Anal.
7
Calcd. for (C H ClN O ) C, H, N: 42.59, 1.79, 18.63.
Found: 42.47, 1.68, 18.60.
6.3.10. 5-Methoxy-8-nitroquinazolin-4-ol (1f). To the
solution of compound 6 (100 mg, 0.44 mmol) in
8
4
3
3

Y. Jin et al. / Bioorg. Med. Chem. 13 (2005) 5613–5622
5621
1
(
2
0 mL of methanol, sodium methanol was added
100 mg, 1.85 mmol). The mixture was refluxed for
4 h at 70 ꢁC. The solution was concentrated and
diluted with water (15 mL). HCl solution was added
20%) until pH 7, the resultant solution was extracted
6.3.15. 8-Nitro-5-pyrrolidin-1-ylquinazolin-4-ol (1k). The
compound 1k was prepared as described for 1a, except
using pyrrolidine (63 mg, 0.88 mmol), affording a brown
1
solid (90 mg, 78.5%). mp: 208–210 ꢁC; H NMR
(
(DMSO-d , 300 MHz), d 12.30 (br, 1H), 8.12 (s, 1H),
6
with EtOAc (2 · 20 mL). Afer usual work-up, the res-
idue was purified by column chromatography
8.09 (d, J = 9.07 Hz, 1H), 6.72 (d, J = 9.07 Hz, 1H),
3.29 (dt, 4H), 1.89 (dt, 4H); MS (EI) m/e 260 (M );
+
(
6
EtOAc/PE = 1:1) to afford a white solid (65 mg,
6.8%). mp: 209–210 ꢁC,
Anal. Calcd. for (C H N O ) C, H, N: 55.38, 4.65,
3
21.53. Found: 55.41, 4.35, 21.78.
1
2
12
4
1
H
MNR (DMSO-d6,
3
00 MHz), d 11.28 (br, 1 H), 9.13 (s, 1H), 8.67 (d,
J = 8.79 Hz, 1 H), 6.83 (d, J = 8.66 Hz, 1 H), 3.90
6.3.16.
8-Nitro-5-(2H-pyrrol-1(5H)-yl)quinazolin-4-ol
+
s, 3 H); MS (EI) m/z 221 (M ); Anal. Calcd. for
(
(
4
(1l). The compound 1l was prepared as described for
1a, except using 2,5-dihydro-1H-pyrrole (61 mg,
0.88 mmol), providing a black brown solid (79 mg,
C H N O ) C, H, N: 48.87, 3.19, 19.00. Found:
3
9
7
4
8.77, 3.23, 18.89.
1
6
9.8%). mp: 154–160 ꢁC;
H
NMR (DMSO-d6,
6
1
.3.11. 5-Ethoxy-8-nitroquinazolin-4-ol (1g). Compound
g was prepared similar to 1f, except using sodium eth-
300 MHz), d 12.41 (br, 1H), 8.11 (s, 1H), 7.99 (d,
J = 8.99 Hz, 1H), 6.79 (d, J = 8.99 Hz, 1H), 6.79 (d,
J = 9.06 Hz, 1H), 5.91 (s, 2H), 3.98 (s, 4H); MS (EI)
anol (125 mg, 1.85 mmol) and 10 mL ethanol as a sol-
vent. Purification by column chromatography (EtOAc/
PE = 1:1) provided a white solid (80 mg, 77.6%). mp:
1
(
(
J = 6.9 Hz, 3H); MS (EI) m/z 235 (M ); Anal. Calcd.
for (C H N O ) C, H, N: 51.07, 3.86, 17.87. Found:
5
+
m/e 257 (M ꢀ1); Anal. Calcd. for (C H N O ) C, H,
1
2
10
4
3
N: 55.81, 3.90, 21.70. Found: 55.79, 3.86, 21.92.
1
95–198 ꢁC; H NMR (DMSO-d , 300 MHz), d 11.28
6
br, 1H), 9.14 (s, 1H), 8.63 (d, J = 8.20 Hz, 1H), 6.81
d, J = 8.40 Hz, 1H), 4.12 (m, J = 6.9 Hz, 2H), 1.48 (t,
6.3.17. 2-[(4-Hydroxy-8-nitroquinazolin-5-yl)amino]-4-
(methylthio) butanoic acid (1m). The compound 1m
was prepared as described for 1a, except using 2-ami-
no-4- (methylthio)butanoic acid (131 mg, 0.88 mmol).
Purification by column chromatography (CHCl3/
+
1
0
9
3
4
1.13, 3.68, 17.74.
MeOH = 2:1) afforded a brown solid (76 mg, 51.2%).
1
6.3.12. 5-[(4-Methoxyphenyl)amino]-8-nitroquinazolin-4-
ol (1h). The compound 1h was prepared as described for
mp: 214–217 ꢁC; H NMR (DMSO-d , 300 MHz), d
6
12.78 (br, 1H), 10.34 (br, 1H), 8.23 (s, 1H), 8.11 (d,
1H, J = 9.07 Hz), 6.91 (d, 1H, J = 9.06 Hz), 4.01 (s,
1H), 2.37 (s, 2H), 2.2–1.8 (m, 5H); MS (EI) m/e 338
(M ); Anal. Calcd. for (C H N O S) C, H, N: 46.15,
4
4.17, 16.56. Found: 46.23, 4.25, 16.43.
1
0
5
3
1
1
2
a, except for using 4-methoxybenzenamine (108 mg,
.88 mmol), affording a black green solid (81 mg,
1
+
8.9%). mp: 200–203 ꢁC;
H
NMR (DMSO-d6,
13
14
5
00 MHz), d 12.65 (br, 1H), d 11.40 (s, 1H), 8.24 (s,
H), 8.14 (d, J = 9.07 Hz, 1H), 7.02 (d, J = 9.07 Hz,
H), 6.95 (d, J = 9.06 Hz, 2H), 6.85 (d, J = 9.06 Hz,
6.3.18. [(4-Hydroxy-8-nitroquinazolin-5-yl)amino](phen-
yl)acetic acid (1n). The compound 1n was prepared as
described for 1a, except for using 2-amino-2-phenylace-
tic acid (133 mg, 0.88 mmol). Purification by column
+
H), 3.73 (s, 3H); MS (EI) m/e 312 (M ); Anal. Calcd.
for (C H N O ) C, H, N: 57.69, 3.87, 17.94. Found:
4
1
5
12
4
5
7.79, 3.98, 17.67.
chromatography (CHCl /MeOH = 2:1) provided
3
a
1
6
.3.13. 5-[(4-Aminophenyl)amino]-8-nitroquinazolin-4-ol
brown solid (74 mg, 49.7%). mp: >250 ꢁC; H NMR
(
1i). The compound 1i was prepared as described
for 1a, except using 4-amino-phenylamine (48 mg,
.44 mmol). Purification from column chromatography
(DMSO-d , 300 MHz), d 12.80 (br, 1H), 11.25 (br,
6
1H), 8.22 (s, 1H), 7.91 (d, J = 9.34 Hz, 1H), 7.25–7.12
(d, 5H), 5.12 (s, 1H); MS (EI) m/e 340 (M ); Anal.
+
0
(
(
CHCl /MeOH = 5:1) afforded a black-brown solid
3
68 mg, 51.4%). mp: 223–226 ꢁC; H NMR (DMSO-
6
Calcd. for (C H N O ) C, H, N: 56.47, 3.55, 16.46.
16 12
Found: 56.22, 3.31, 16.28.
4
5
1
d , 300 MHz), d 11.28 (br, 1H), 8.21 (s, 1H), 8.08
(
(
(
d, J = 9.06 Hz, 1H), 6.90 (d, J = 9.07 Hz, 1H), 6.68
d, J = 8.52 Hz, 2H), 6.47 (d, J = 8.79 Hz, 2H); MS
EI) m/z M = 297; Anal. Calcd. for (C H N O )
Acknowledgments
+
1
4
11
5
3
C, H, N: 56.56, 3.73, 23.56. Found: 56.70, 3.46,
2
Chinese Academy of Sciences (No. KSCX1-SW-11) and
Ministry of Science and Technology of China (No.
3.44.
2
004CB518903) are greatly appreciated for the financial
6
.3.14. 5-[4-(Benzyloxy)phenyl]amino8-nitroquinazolin-4-
supports. The authors thank Prof. Hualiang Jiang of
DDDC at SIMM for his helpful discussion in molecular
modeling.
ol (1j). The compound 1j was prepared in the same pro-
cedure as described for 1a, except using 4-(benzyl-
oxy)benzenamine (175 mg, 0.88 mmol), affording a
yellow brown solid (110 mg, 64.7%). mp: 220–222 ꢁC;
1
H NMR (DMSO-d , 300 MHz), d 11.40 (br, 1H),
6
References and notes
8
5
1
.25 (s, 1H), 8.15 (d, J = 9.07 Hz, 1H), 7.46–7.35 (m,
H), 7.02 (d, J = 9.07 Hz, 1H), 6.94 (s, 4H), 5.05 (s,
H); MS (EI) m/e 388 (M ); Anal. Calcd. for
1
. Yarden, Y.; Sliwkowski, M. X. Nat. Rev. Mol. Cell Biol.
2001, 2, 127.
+
(
6
C H N O ) C, H, N: 64.94, 4.15, 14.43. Found:
16
4.82, 4.45, 14.56.
2. Graus-Porta, D.; Beerli, R. R.; Daly, J. M. EMBO J.
1997, 16, 1647.
2
1
4
4

5622
Y. Jin et al. / Bioorg. Med. Chem. 13 (2005) 5613–5622
3
4
5
. Traxler, P. Expert Opin. Ther. Targets 2003, 7, 215. 13. Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.; McMa-
. Ranson, M. Br. J. Cancer 2004, 90, 2250.
. Supuran, C. T.; Scozzafava, A. Expert Opin. Ther. Patents
hon, J.; Vistica, D.; Warren, J. T.; Bokesch, H.; Kenney, S.;
Boyd, M. R. J. Natl. Cancer Inst. 1990, 82, 1107.
14. Brignola, P. S.; Lackey, K.; Kadwell, S. H.; Hoffman, C.;
Horne, E.; Carter, H. L.; Stuart, J. D.; Blackburn, K.;
Moyer, M. B.; Alligood, K. J.; Knight, W. B.; Wood, E.
R. J. Biol. Chem. 2002, 277, 1576.
2004, 14, 35.
6. Bridges, A. J. Chem. Rev. 2001, 101, 2541.
7. Rewcastle, G. W.; Denny, W. A.; Bridges, A. J.; Zhou, H.;
Cody, D. R.; McMichael, A.; Fry, D. W. J. Med. Chem.
1
995, 38, 3482.
15. Wood, E. R.; Truesdale, A. T.; McDonald, O. B.; Yuan,
D.; Hassell, A.; Dickerson, S. H.; Ellis, B.; Pennisi, C.;
Horne, E.; Lackey, K.; Alligood, K. J.; Rusnak, D. W.;
Gilmer, T. M.; Shewchuk, L. Cancer Res. 2004, 64, 6652.
16. Wallace, A. C.; Laskowski, R. A.; Thornton, J. M. Protein
Eng. 1995, 8, 127.
17. Sybyl, Version 6.8; Tripos Associates: St. Louis, MO,
2000.
18. Weiner, S. J.; Kollman, P. A.; Case, D. A.; Singh, U. C.;
Ghio, C.; Alagona, G.; Profeta, S.; Weiner, P. J. Am.
Chem. Soc. 1984, 106, 765.
8
. Barker, A. J.; Gibson, K. H.; Grundy, W.; Godfrey, A. A.;
Barlow, J. J.; Healy, M. P.; Woodburn, J. R.; Ashton, S.
E.; Curry, B. J.; Scarlett, L.; Henthorn, L.; Richards, L.
Bioorg. Med. Chem. Lett. 2001, 11, 1911.
. Gaul, M. D.; Guo, Y.; Affleck, K.; Cockerill, G. S.;
Gilmer, T. M.; Griffin, R. J.; Guntrip, S.; Keith, B. R.;
Knight, W. B.; Mullin, R. J.; Murray, D. M.; Rusnak, D.
W.; Smith, K.; Tadepalli, S.; Wood, E. R.; Lackey, K.
Bioorg. Med. Chem. Lett. 2003, 13, 637.
0. 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.
9
1
19. Gasteiger, J.; Marsili, M. Tetrahedron 1980, 36, 3219.
20. Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.;
Hart, W. E.; Belew, R. K.; Olson, A. J. J. Comput. Chem.
1998, 19, 1639.
2001, 44, 2719.
1
1. Zhang, Y.-M.; Cockerill, S.; Guntrip, S. B.; Rusnak, D.;
Smith, K.; Vanderwall, D.; Wood, E.; Lackey, K. Bioog.
Med. Chem. Lett. 2004, 14, 111.
2. Wang, S.; Yang, D.; Enyedy, I. US Patent US2004023957-
A1, 2004.
21. Morris, G. M.; Goodsell, D. S.; Huey, R.; Hart, W. E.;
Halliday, R. S.; Belew, R. K.; Olson, A. J. AUTODOCK,
M. Version 3.0.3. The Scripps Research Institute, Molec-
ular Graphics Laboratory, Department of Molecular
Biology; 1999.
1