4-Anilinoquinazolines with Lavendustin A subunit as inhibitors of epidermal growth factor receptor tyrosine kinase: Syntheses, chemical and pharmacological properties

Article abstract of DOI:10.1016/j.ejmech.2004.03.010

4-Anilinoquinazoline derivatives are widely investigated due to their potent epidermal growth factor receptor (EGFR) tyrosine kinase inhibitory activity. Two 4-anilinoquinazolines with Lavendustin A subunit (10a,b) were synthesized and examined for their EGFR tyrosine kinase inhibitory activity as well as their antiproliferative properties on variant human cancer cell lines. Both compounds maintained their EGFR tyrosine kinase inhibitory activity at the 10-7:M level and led to significant growth inhibition in certain leukemia, non-small cell lung cancer (NSCLC), ovarian cancer, renal cancer and breast cancer cell lines with GI₅₀ values at the 10-6 M level. There could not be observed any notable difference between 10a and 10b regarding to their antiproliferative activity. Interestingly, we observed the high tendency of 10a and 10b to include certain solvents, e.g. water, DMF, DMSO, which may be due to the remarkable number of hydrogen accepting/donating groups in 10a and b. An X-ray analysis of 10a including water and DMF illustrates a possible hydrogen bond pattern and could serve as information for preferred receptor (e.g. EGFR tyrosine kinase) binding sites. Finally, we aimed for irreversible EGFR tyrosine kinase inhibitors. The p-quinone derivatives 11a and 11b, which contain a Michael acceptor position according to the irreversible inhibitor CI-1033, could be derived from the p-hydroquinone derivatives 10a or 10b, respectively, by oxidation. However, due to their instability 11a and 11b could not be obtained in a pure form.

Full text of DOI:10.1016/j.ejmech.2004.03.010

European Journal of Medicinal Chemistry 39 (2004) 1001–1011
www.elsevier.com/locate/ejmech
Original article
4-Anilinoquinazolines with Lavendustin A subunit as inhibitors
of epidermal growth factor receptor tyrosine kinase: syntheses, chemical
and pharmacological properties
Rica Albuschat a, Werner Löwe a,*, Manuela Weber b, Peter Luger b, Verena Jendrossek c
a Institute of Pharmacy, Free University of Berlin, Königin-Luise-Strasse 2+4, 14195 Berlin, Germany
b Institute of Crystallography, Free University of Berlin, Takustrasse 6, 14195 Berlin, Germany
c Department of Radiation Oncology, University of Tübingen, Hoppe-Seyler-Straße 3, 72076 Tübingen, Germany
Received 11 September 2003; received in revised form 2 March 2004; accepted 11 March 2004
Abstract
4-Anilinoquinazoline derivatives are widely investigated due to their potent epidermal growth factor receptor (EGFR) tyrosine kinase
inhibitory activity. Two 4-anilinoquinazolines with Lavendustin A subunit (10a,b) were synthesized and examined for their EGFR tyrosine
kinase inhibitory activity as well as their antiproliferative properties on variant human cancer cell lines. Both compounds maintained their
EGFR tyrosine kinase inhibitory activity at the 10–7 M level and led to significant growth inhibition in certain leukemia, non-small cell lung
cancer (NSCLC), ovarian cancer, renal cancer and breast cancer cell lines with GI50 values at the 10–6 M level. There could not be observed
any notable difference between 10a and 10b regarding to their antiproliferative activity. Interestingly, we observed the high tendency of 10a
and 10b to include certain solvents, e.g. water, DMF, DMSO, which may be due to the remarkable number of hydrogen accepting/donating
groups in 10a and b. An X-ray analysis of 10a including water and DMF illustrates a possible hydrogen bond pattern and could serve as
information for preferred receptor (e.g. EGFR tyrosine kinase) binding sites. Finally, we aimed for irreversible EGFR tyrosine kinase
inhibitors. The p-quinone derivatives 11a and 11b, which contain a Michael acceptor position according to the irreversible inhibitor CI-1033,
could be derived from the p-hydroquinone derivatives 10a or 10b, respectively, by oxidation. However, due to their instability 11a and 11b
could not be obtained in a pure form.
© 2004 Elsevier SAS. All rights reserved.
Keywords: 4-Anilinoquinazolines; Salicylanilides; Quinones; Inclusion of solvents; EGFR tyrosine kinase inhibitory activity; Antiproliferative effects
1. Introduction
death pathways. In this context, extracellular growth factors
(e.g. EGF, TGF-a) bind to their respective cell surface recep-
tors (e.g. EGFR) and activate a network of signaling cascades
triggering complex cellular functions such as cell cycle pro-
gression, proliferation, survival, invasion, metastasis and an-
giogenesis [1,2].
Growth of malignant tumours is mostly caused by a dere-
gulation of the balance between proliferation, survival and
Abbreviations: EGFR, epidermal growth factor receptor; PDGFR, plate-
let derived growth factor receptor; RPTK, receptor protein tyrosine kinase;
EGF, epidermal growth factor; TGF-a, transforming growth factor-a; Cys,
cysteine; Tyr, tyrosine; ptyr, phosphorylation rate; r.t., room temperature;
m.p., melting point; calc., calculated; det., determined; n.t., not tested;
NSCLC, Non-small cell lung cancer; IC50, drug concentration resulting in a
50% inhibition of the EGFR tyrosine kinase; GI50, drug concentration
resulting in a 50% growth inhibition; TGI, drug concentration resulting in
total growth inhibition; LC50, drug concentration resulting in a 50% reduc-
tion in the measured protein at the end of drug treatment as compared to that
at the beginning; abl kinase, ableson protein tyrosine kinase.
It has been shown that up-regulated activity of EGFR
correlates with the malignant phenotype of many human
epithelial tumours. Therefore, the EGFR signaling represents
an attractive target for the design of novel anticancer drugs
[3–6]. Therapeutic strategies to inhibit EGFR mediated si-
gnaling include antibodies against the ligand binding extra-
cellular receptor domain (cetuximab) as well as small phar-
macological inhibitors of the receptor tyrosine kinase
Iressa™1 (Gefitinib, ZD1839) and Tarceva™2 (Erlotinib,
OSI-774), which are acting as reversible antagonists, as well
as the irreversible inhibitor CI-1033 (compound 12, Fig. 1).
* Corresponding author. Tel.: +49 30 83855280; fax: +49 30 83853854.
E-mail address: wloewe@zedat.fu-berlin.de (W. Löwe).
0223-5234/$ - see front matter © 2004 Elsevier SAS. All rights reserved.
doi:10.1016/j.ejmech.2004.03.010

1002
R. Albuschat et al. / European Journal of Medicinal Chemistry 39 (2004) 1001–1011
Fig. 1. The 4-anilinoquinazolines Iressa™ (1), Tarceva™ (2) and CI-1033 (12).
Fig. 2. The Lavendustin A pharmacophore (3) and the related salicylanilides 4a (R1=Br, R2=H) and 4b (R1=Cl, R2=F).
Since aberrantly increased EGFR signaling has been
shown to interfere with cytotoxic therapies, inhibition of
EGFR signaling may not only reduce growth, angiogenesis,
invasion and metastasis of tumour cells but also increase
sensitivity to cytotoxic therapies [7–9].
Recently, we designed the salicylanilides 4a and 4b,
which are derivatives of the Lavendustin A pharmacophore 3
(Fig. 2) [10,11]. The anilides 4a and 4b (Fig. 2) showed in
vitro EGFR tyrosine kinase inhibitory activities of 90% and
96% at a concentration of 10 µM. The N-(3-bromophenyl)
derivative was tested at 0.1 µM, too. At this concentration it
still led to an inhibition of 42% [10].
According to the results of Hodge and Traxler that revea-
led a bioisosteric relationship between the salicylic group
and the quinazoline (illustrated in Fig. 3), we synthesized the
4-anilinoquinazolines with a 6-[(2,5-dihydroxybenzyl)-
amino] side chain 10a and 10b (Fig. 3) [12,13].
Based on results of other 4-anilinoquinazoline derivatives,
e.g. Iressa™, which emphasized the relevance of the
4-anilinoquinazoline partial structure as an important phar-
macophore group of EGFR tyrosine kinase inhibitors [14],
we suggested increased EGFR tyrosine kinase inhibitory
activities of 10a and 10b in comparison to 4a and 4b. In
addition, the putative improved cellular penetration of 10a
and 10b in comparison to the salicylanilides 4a and 4b
together with promising results obtained with Iressa™ we
hoped for cytotoxic qualities of 10a and 10b [8,15].
A second aim of the present study was to obtain irreversi-
ble EGFR tyrosine kinase inhibitor derivatives from 10a and
10b, since the diphenolic partial structure in the
6-benzylamino side chain provides the opportunity to get to
the corresponding p-quinone derivatives 11a and 11b by a
simple oxidation reaction.
The compounds 11a and 11b (Scheme 2) contain a Mi-
chael acceptor position on C2 of the quinone that is compa-
rable to that in the 6-acrylamido side chain of the irreversible
EGFR tyrosine kinase inhibitor CI-1033 (Fig. 1, Scheme 2).
CI-1033 is believed to irreversibly inhibit the EGFR tyrosine
kinase by forming a covalent bond to Cys 773 in the ATP
binding pocket [16]. CI-1033, which is currently tested in
clinical trials for anticancer treatment, has been shown to
irreversibly inhibit the EGFR tyrosine kinase with a remar-
kable IC50 value of 1.5 nM [17–20].
Thus, because of the putative irreversible activity of the
quinones 11a and 11b we hoped for an increased anti-EGFR
Fig. 3
. Salicylic acid and quinazoline as bioisosters and the compounds 10a
(R1=Br, R2=H) and 10b (R1=Cl, R2=F).

R. Albuschat et al. / European Journal of Medicinal Chemistry 39 (2004) 1001–1011
1003
Scheme 1. Synthesis of 10a (7–10a: R1=Br, R2=H) and 10b (7–10b: R1=Cl, R2=F): (a) dimethylformamide dimethyl acetal, 100 °C, 90 min; (b) 3-bromoaniline
or 3-chloro-4-fluoroaniline, acetic acid, reflux, 60 min; (c) iron, acetic acid/ethanol, reflux, 4 h; (d) 2,5-dihydroxybenzaldehyde, acetic acid/ethanol, reflux 2 h;
(e) dimethylamine borane, acetic acid, 25 °C.
effectiveness of these two compounds in comparison to the
reversible inhibitors 10a and 10b.
lines (see also Table 1). Two molecules of 10a, related by a
pure translation in y-direction are hydrogen bonded via the
O(4)–H group of the hydroquinone group to N(13) of the
quinazoline ring. O(4) is also acceptor of a hydrogen bond
from the water molecule. The opposite phenolic O–H group
is also donor (to the oxygen of a DMF molecule) and accep-
2. Chemistry
2.1. Syntheses of 10a and 10b
Commercial available 5-nitroanthranilo nitrile (5) was
converted into the corresponding formamidine 6 using di-
methylformamide dimethyl acetal. The following reflux boi-
ling of 6 with 3-bromoaniline or 3-chloro-4-fluoroaniline,
respectively, in acetic acid gave the 6-nitro-4-phenyl-
aminoquinazoline derivatives 7a,b. Reduction of the nitro
group of 7a,b with iron in a mixture of acetic acid and
ethanol yielded the intermediate 6-aminoquinazolines 8a,b
(step 1–3: [21]).
The Schiff bases 9a,b were prepared by condensation of
8a,b with 2,5-dihydroxy-benzaldehyde in a refluxing mix-
ture of acetic acid and ethanol. The following reduction of the
imines 9a,b with dimethylamine borane in glacial acetic acid
gave the obtained 6-[(2,5-dihydroxybenzyl)-amino] derivati-
ves 10a and 10b (Scheme 1) [22].
2.2. Inclusion of solvents by 10a and 10b
Surprisingly, we observed a high tendency of 10a,b to
include certain solvents, such as dimethylformamide (DMF),
dimethylsulfoxide (DMSO) and water, and proved that by
1H-NMR-(additional peaks of DMF/DMSO) and mass-
(molecular peaks of DMF/DMSO) spectra and additionally
by X-ray analysis. Fig. 4 shows the X-ray picture of crystals
of 10a including water and DMF.
Fig. 4 shows the molecular structure of 10a in the crystal
co-crystallizing with water and one DMF molecule. All but
one possible N–H and O–H donor groups establish inter-
molecular hydrogen bonds, as illustrated in Fig. 4 by dashed
Fig. 4. Crystal structure of 10a including water and one DMF molecule,
hydrogen bonds are indicated by dashed lines, generated with SCHAKAL
[23].

1004
R. Albuschat et al. / European Journal of Medicinal Chemistry 39 (2004) 1001–1011
Table 1
Hydrogen-bonds in the crystal structure of 10a including water and one DMF molecule [A and deg.]
D-H...A
D(D–H)
0.77(3)
0.95(3)
0.76(3)
0.85
d(H...A)
1.88(4)
1.69(3)
2.25(3)
1.96
d(D...A)
2.641(3)
2.632(3)
2.994(3)
2.802(3)
2.886(3)
<(DHA)
174.(4)
169.(3)
165.(3)
177.
Symmetry
3–x, –y, –z
x,–1 +y, z
x, y, z
O1-H1...O1L
O4-H4...N13
N19-H19...O1W
O1W-H11W...O4
O1W-H12W...O1
2 – x, –y, 1 – z
–1 +x, y, z
0.83
2.06
198.
tor (from the water molecule) of hydrogen bonds. A further
hydrogen bond is formed by the N–H of the 4-[(3-
bromophenyl)-amino] substituent to a water molecule. Sur-
prisingly, the possible donor group N(8)-H is not involved in
hydrogen bonding in this crystal structure. Although the
intermolecular interactions under physiological conditions
may be different, the hydrogen bonding pattern found in this
structure can serve as information for preferred receptor
binding sites.
3. Pharmacology
3.1. EGFR tyrosine kinase inhibitory activity
The determination of the in vitro EGFR tyrosine kinase
inhibitory activity was carried out by MDS pharma services,
Taiwan (R.O.C.). A431 cell membranes served as the en-
zyme source and poly (glu/tyr, 4:1) served as the standardi-
zed substrate. The phosphorylation rate was determined after
incubation of substrate, enzyme and ATP via ELISA in pre-
sence (p-tyrtest) and in absence (p-tyrmax) of inhibitor. The
minimal phosphorylation rate (p-tyrmin) was determined af-
ter incubation of substrate and ATP in absence of enzyme.
The calculation was carried out by the following equation:
2.3. Oxidation of the hydroquinone moiety of 10a and 10b
In order to obtain irreversible inhibitors we undertook the
oxidation of 10a,b. The resulting quinone partial structure
contains a Michael acceptor position (illustrated in Scheme
2).
Inhibition % = 100 % − 100 % ×
͓ ͔
ptyrtest − ptyrmin ⁄ ptyrmax − ptyrmin
͑
͒ ͑
͒
The electron deficient carbon atom is probably able to
form a covalent linkage with the sulfhydryl group of Cys
773 in the ATP binding pocket of the EGFR tyrosine kinase
domain, corresponding to the b-carbon in the acrylamido
side chain of compound 12 (CI-1033), developed by Pfizer
(Scheme 2). CI-1033 was the first irreversible EGFR tyrosine
kinase inhibitor reported to be in clinical trials [17–20].
The hydroquinone moiety of 10,b was easily oxidized by
tetra-n-butylammonium periodate yielding the quinone deri-
vates 11a,b. However, due to their instability because of an as
yet unknown rearrangement reaction, the desired quinones
could not be obtained in a pure form, but mixtures of it with
the Schiff bases 9a,b (Scheme 2).
3.2. Cytotoxic/cytostatic activity against human malignant
glioma cell lines
Growth inhibitory effects of 10a,b were evaluated in three
human malignant glioma cell lines in vitro after 24, 48,
72 and 96 h of continuous drug exposure to the cells using the
Wst-1 test [38]. This assay is based on the cleavage of the
tetrazolium salt Wst-1 by mitochondrial enzymes that are
only active in viable cells. Both drugs were tested at
11 concentrations (100, 75, 50, 25, 12.5, 10, 5, 2.5, 1, 0.5 and
0.1 µM) and dose–response curves were established for each
day and each cell line. Growth inhibitory activity of the drugs
Scheme 2. The irreversible EGFR tyrosine kinase inhibitor CI-1033 (12) and the oxidation of the hydroquinone moiety of 10a and 10b to obtain the
quinone-derivatives 11a and 11b as putative irreversible inhibitors (9–11a: R1=Br, R2=H; 9–11b: R1=Cl, R2=F): (a) tetra-n-butyl-ammonium periodate,
dichloromethane/methanol, 25 °C, 2 min.

R. Albuschat et al. / European Journal of Medicinal Chemistry 39 (2004) 1001–1011
1005
Table 2
EGFR tyrosine kinase inhibitory activities of tested salicylanilides and 4-anilinoquinazolines at a 10, 1 or 0.1 µM concentration, respectively
Compound
R1
R2
Inhibitory activity [%]
10 µM
90
1 µM
n.t.
n.t.
65
0.1 µM
42
4a
4b
10a
10b
9a
Br
Cl
Br
Cl
H
F
96
n.t.
H
F
n.t.
45
n.t.
71
22
n.t.
58
n.t.
was calculated from dose–response curves (for further de-
tails, please refer also to Section 6.3.2.)
(Table 2). The imino derivative 9a (Fig. 5) was at a concen-
tration of 1 µM nearly as active as its amino analogue 10a.
The hoped-for further increase of the inhibitory activity in
comparison to the salicylanilides 4a and 4b was not obser-
ved. The salicylanilide 4a is at a concentration of 0.1 µM as
active as the corresponding 4-anilinoquinazoline 10a. This
could be valuated as a proof of the supposed bioisosteric
relationship of the salicylic group to the quinazoline.
3.3. Cytotoxic/cytostatic activity against a panel
of 60 human cell lines
The 4-anilinoquinazolines 10a,b both have been selected
by the Evaluation Committee of the National Cancer Institute
(NCI), Maryland (USA) for anticancer screening on 60 cell
lines.
4.2. Growth inhibitory effects of 10a and 10b on U87MG,
The determination was carried out following the known in
vitro antitumour screening program, which is based upon use
of multiple panel of 60 human cell lines [24,25].
The compounds have been tested at five concentrations:
100, 10, 1, 0.1, and 0.01 µM. A 48 h continuous drug expo-
sure protocol is used. The anticancer activity of each com-
pound is deduced from dose–response curves. The response
parameters are GI50, TGI and LC50, which are presented in
Table 2 according to the data provided by NCI [25].
Further information on the experimental execution is out-
lined in the experimental Section 6.3.3.
A172 and T98G cells
The compounds 10a and 10b showed pronounced inhibi-
tory effects on U87MG, A172 and T98G cells. A 50% reduc-
tion in the number of viable cells could already be observed
upon 24 h of treatment with drug-concentrations ranging
from 11 to 30 µM. In this regard, the extent of growth
inhibitory action was dependent on the cell line and the drug
used.
The highly proliferative U87MG cells were most sensitive
to a treatment with the EGFR tyrosine kinase antagonists
while A172 cells showed intermediate sensitivity and T98G
cells turned out to be most resistant to a treatment with 10a
and 10b (Fig. 6). In addition, a slight tendency of the cells to
be more responsive to a treatment with 10a than to 10b could
be observed: The drug concentrations required for a reduc-
tion of the number of viable cells to 50% of the untreated
control cells upon 24, 48, 72 and 96 h of continuous drug
treatment (DC50) are given in Table 3.
4. Results and discussion
4.1. EGFR tyrosine kinase inhibitory activity of 4a and 4b,
9a, 10a and 10b
Both 4-anilinoquinazolines 10a and 10b (Fig. 5) showed
moderate EGFR tyrosine kinase inhibitory activities
Fig. 5. Tested compounds: 4a (R1=Br, R2=H), 4b (R1=Cl, R2=F), 10a (R1=Br, R2=H), 10b (R1=Cl, R2=F) and 9a.

1006
R. Albuschat et al. / European Journal of Medicinal Chemistry 39 (2004) 1001–1011
Especially, the growth of the cell lines CCRF-CEM,
K-562 and RPMI-8226 (Leukemia), EKVX and NCI-H522
(NSCLC), IGROV1 and SK-OV-3 (Ovarian cancer), A498,
ACHN, CAKI-1 (Renal cancer), BT-549 and F47D (Breast
cancer) was inhibited at concentrations at the 10–6 M level.
The 3-chloro-4-fluoro derivative 10b displayed high acti-
vity against CCRF-CEM (GI50 = 0.04 µM, TGI = 0.09 µM),
EKVX (GI50 = 0.75 µM), A498 (GI50 = 0.23 µM) and ACHN
(GI50 = 0.68 µM) cells.
Additionally, cytotoxic effects—as indicated by the LC50
values—can be observed on some cell lines, e.g. Leukemia
(10b: 33.7–98.7 µM), some NSCLC (10a,b: 51.5–98.7),
some Melanoma (10a,b: 61.4–74.1) and some Renal (A498:
10a,b: 42.4 and 40.8) cells.
The mechanism of the cytotoxic activity is not completely
clear. Cancer cells of NSCLC (40–80%), renal tumours (50–
90%), glioblastoma (40–50%), ovarian (35–70%) and some
breast (14–91%) tumours are known for their EGFR (over)-
expression [26].
It is possible, that both compounds are not only
cytotoxic/cytostatic because of their EGFR tyrosine kinase
inhibitory activity, but act by other mechanisms, e.g. inhibi-
tion of other receptor protein tyrosine kinases like the PD-
GFR tyrosine kinase or of cellular protein kinases like the abl
kinase. The abl kinase or its related oncogenic product, the
bcr-abl kinase (caused by a chromosome rearrangement re-
sulting in the so-called Philadelphia chromosome) plays an
important role in the genesis and progression of leukemias,
e.g. 95% of chronic myeloid leukemias are Philadelphia
chromosome positive [27]. The 2-phenylaminopyrimidine
derivative Glivec used for the treatment of the chronic mye-
loid leukemia has been shown to act as an ATP antagonist at
the abl kinase as well as at the PDGFR tyrosine kinase
[28–30]. Since compound 10b was designed as an ATP
antagonist as well, we suggest, that the excellent activities of
10b on leukemia cell lines could at least be partially due to an
inhibition of the abl kinase and/or of the PDGFR tyrosine
kinase as well.
Fig. 6. Human astrocytoma/glioblastoma cell lines differ in their sensitivity
to EGF-R antagonists. U87MG, A172 and T98G cells were exposed for 48 h
to the indicated concentrations of (a) 10a or (b) 10b. Viability was then
determined after continuous exposure to the drug by the Wst-1 test. Data
represent means of one representative experiment (n = 3).
Thus, similar to other cell lines derived from solid human
tumors, human astrocytoma/glioblastoma cell lines showed
intermediate sensitivity to the antineoplastic effects of
4-anilinoquinazolines with inhibitory activity on EGFR tyro-
sine kinases. In this context it has to be taken into account
that the Wst-1 test evaluates the number of viable cells and
does not distinguish between inhibition of proliferation and
cytotoxicity. Therefore, the results reflect a combination of
cytostatic and direct cytotoxic drug effects.
4.3. Cytotoxic/cytostatic activity of 10a and 10b against
a panel of 60 human cell lines
5. Conclusions
The results of the evaluation of anticancer activity on
10a,b are given in Table 4.
Up to 16 cell lines have been shown to be particularly
sensitive to a treatment of 10a,b (GI50 at the 0.01 to 1 µM
level).
The 4-anilinoquinazolines 10a and 10b are relative potent
EGFR tyrosine kinase inhibitors with approximate IC50 va-
lues in the range of 0.1–1 µM. Additionally, both substances
cause antiproliferative to cytotoxic effects on certain human
Table 3
Human astrocytoma/glioblastoma cell lines show a tendency to increased sensitivity to the treatment with 10a. U87MG, A172 and T98G cells were
continuously exposed for 24–96 h to the indicated concentrations of 10a or 10b. Viability was then determined by the Wst-1 test. DC50 values represent the drug
concentration that was required to reduce cell viability as determined by WST-1 reduction to 50% of untreated controls
DC50 values [µM]
T98G
A172
U87MG
10a
10b
36.5
29.5
33.8
32.3
10a
10b
21.5
21.8
19.0
19.0
10a
10.1
9.4
10b
13.6
10.4
8.3
24 h
48 h
72 h
96 h
33.0
24.5
25.3
28.0
19.0
15.5
16.0
15.1
8.8
9.4
8.4

R. Albuschat et al. / European Journal of Medicinal Chemistry 39 (2004) 1001–1011
1007
Table 4
In vitro testing results of 10a,b: GI50, TGI and LC50 [µM]
Panel/cell line
10a
10b
GI50
TGI [µM]
LC50
GI50
TGI [µM]
LC50
Leukemia
CCRF-CEM
K-526
5.26
2.54
4.61
8.32
20.5
20.7
≥100
≥100
≥100
0.04
2.49
2.37
0.09
13.8
6.51
33.7
80.7
98.7
RPMI-8226
NSCLC
A549/ATCC
EKVX
12.8
2.19
17.1
15.0
22.2
26.0
4.14
36.9
13.0
34.5
29.1
59.5
≥100
≥100
≥100
54.3
69.4
56.6
≥100
≥100
≥100
13.0
0.75
19.3
12.3
54.5
46.2
2.76
37.6
17.2
43.7
25.7
≥100
≥100
≥100
≥100
51.5
98.7
53.8
≥100
≥100
≥100
HOP-62
NCI-H226
NCI-H23
NCI-H460
NCI-H522
Colon cancer
COLO 205
HCC-2998
HCT-116
HCT-15
8.67
27.2
21.9
66.8
35.6
21.8
35.8
≥100
≥100
≥100
≥100
≥100
≥100
14.1
28.6
23.3
24.0
29.2
13.3
≥100
≥100
≥100
≥100
≥100
≥100
≥100
≥100
≥100
≥100
≥100
≥100
≥100
≥100
≥100
≥100
89.6
HT29
SW-620
CNS cancer
SF-268
13.8
17.0
11.2
13.0
29.7
16.1
48.8
37.6
31.3
29.7
≥100
52.7
≥100
83.1
88.0
67.9
≥100
≥100
23.2
24.1
10.7
25.3
21.9
11.4
≥100
53.6
30.6
≥100
85.3
43.5
≥100
≥100
87.6
SF-295
SF-539
SNB-19
≥100
≥100
≥100
SNB-75
U251
Melanoma
LOX IMVI
M14
19.2
18.4
29.9
21.0
14.7
14.3
16.9
53.1
40.4
≥100
50.1
31.8
30.0
35.2
≥100
88.9
≥100
≥100
68.4
62.9
73.4
22.3
25.6
42.1
21.5
9.22
14.7
15.4
91.0
≥100
≥100
64.4
27.0
32.7
30.7
≥100
≥100
≥100
≥100
74.1
72.5
61.4
SK-MEL-2
SK-MEL-28
SK-MEL-5
UACC-257
UACC-62
Ovarian cancer
IGROV1
4.84
6.75
31.0
23.4
8.01
4.56
≥100
28.3
≥100
82.1
32.5
24.1
≥100
95.2
4.63
13.3
25.8
36.2
11.9
6.70
≥100
75.0
≥100
≥100
43.4
33.4
≥100
≥100
≥100
≥100
≥100
≥100
OVCAR-3
OVCAR-4
OVCAR-5
OVCAR-8
SK-OV-3
Renal cancer
786-0
≥100
≥100
≥100
≥100
14.8
1.90
1.10
1.75
15.1
19.2
11.1
41.6
11.8
26.5
31.4
41.1
66.0
59.4
≥100
42.4
13.7
0.23
0.68
2.90
16.4
11.1
8.76
40.4
7.57
22.3
30.4
71.9
34.2
≥100
≥100
40.8
A498
ACHN
≥100
≥100
≥100
≥100
≥100
69.6
CAKI-1
≥100
≥100
≥100
≥100
RXF 393
SN12C
UO-31
Prostate cancer
PC-3
10.9
10.5
33.0
34.1
99.8
13.7
10.1
40.8
26.3
≥100
DU-145
≥100
68.4
Breast cancer
MCF7
29.6
25.7
19.6
22.4
13.9
4.64
5.28
≥100
≥100
42.2
65.6
31.2
28.0
41.2
≥100
≥100
90.6
55.9
40.3
10.4
18.4
19.6
9.21
3.42
≥100
≥100
31.5
61.6
44.2
25.5
45.2
≥100
≥100
95.5
≥100
99.5
67.0
≥100
NCI/ADR-RES
MDA-MB-231/ATCC
HS 578T
MDA-MB-435
BT-549
≥100
69.7
≥100
≥100
T-47D

1008
R. Albuschat et al. / European Journal of Medicinal Chemistry 39 (2004) 1001–1011
4
4
cancer cell lines. Worthy of remark is, that 10b showed
excellent cytotoxic activity on the leukemia cell line CCRF-
CEM with GI50 and TGI values at the 0.01 µM levels
(0.04 and 0.09 µM) and a LC50 value of 33.7 µM.
3J = 9.2 Hz, J = 2.2, 1H, quinazolineH7), 8.17 (d, J (H,
3
F) = 4.6 Hz, 1H, anilinoH2), 7.98 (d, J = 9.2 Hz, 1H,
quinazolineH8), 7.90.7.73 (m, 1H, anilinoH6), 7.61.7.34 (m,
1H, anilinoH5). Anal. C14H8ClFN4O2.
6.1.4. N4-(3-Bromo-phenyl)-4,6-quinazolinediamine (8a)
Following a published procedure [21] 8a was prepared by
reduction of 7a with iron: To a heavy stirred mixture of 7a
(11.00 g, 31.9 mmol) in ethanol (55 ml) and acetic acid
(15 ml) was added iron (8.80 g, 157.6 mmol). The resulting
mixture was heated to reflux for 3 h. After cooling to room
temperature 80 ml water was added to the mixture. The solid
was filtered off and washed several times with water. Then
200 ml ethyl acetate was added to the residue. The resulting
mixture was first washed with 25% KHCO3 solution, then
twice with brine, dried over anhydrous Na2SO4, decolouri-
zed with charcoal and filtered. The filtrate was concentrated
in vacuum. The crude product was purified by flash chroma-
tography over silica gel (ethyl acetate) to yield 8a as yellow
crystals (5.52 g, 54.8%), m.p. 203–204 °C (lit. 203.5–
204.5 °C).
6. Experimental
6.1. Chemistry
All reagents were fresh, commercial grade chemicals.
Melting points (m.p.) were determinated on a Lindström-
apparatus and are uncorrected. 1H-NMR spectra were recor-
ded on Bruker AC-300 spectrometer in DMSO-d6 solutions
using tetramethylsilane as an internal reference. Coupling
patterns are described as follows: s, singlet; d, doublet, dd,
doubled doublet; t, triplet; m, multiplet; and 1H, 2H, 3H, etc.
as the number of hydrogen integrated within a given coupling
pattern. The chemical shifts (d) were measured to two deci-
mal points and are given in ppm.
nJ describes a coupling constant with n = numbers of
bonds between the coupling hydrogens. If the coupling par-
tner is not a hydrogen, but fluor, this is indicated in parenthe-
sis behind nJ. The coupling constants were rounded off to one
decimal place.
6.1.5.
amine (8b)
Compound 7b (10.00 g, 31.4 mmol) was reduced simi-
N4-(3-Chloro-4-fluoro-phenyl)-4,6-quinazolinedi-
larly to 7a by iron (8.80 g, 157.6 mmol) to give 8b as yellow
crystals (2.80 g, 30.9%), m.p. 244 °C. 1H-NMR (400 MHz):
9.47 (s, 1H, anilino-NH, exchanged by D2O), 8.35 (s, 1H,
quinazolineH2), 8.20 (dd, 4J = 2.6 Hz, 4J (H, F) = 6.9 Hz, 1H,
anilinoH2), 7.83 (m, 1H, anilinoH6), 7.54 (d, 3J = 8.9 Hz, 1H,
quinazolineH8), 7.41 (t, 3J (H, F) = 9.1 Hz, 1H, anilinoH5),
7.31 (d, 4J = 2.3 Hz, 1H, quinazolineH5), 7.24 (dd,
Elementary analyses were performed on Perkin-Elmer
Elementar-Vario EL. The analytical results for the elements
C, H, N were within 0.4% of the theoretical values.
6.1.1. N′-(2-cyano-4-nitro-phenyl)-N,N-dimethylimidoform-
amide (6)
According to a published procedure [21] 6 was prepared
by reflux boiling of 5-nitroanthranilonitrile 5 (5.00 g,
30.7 mmol) in 20 ml dimethylformamide dimethyl acetal.
After 90 min the mixture was allowed to cool to room
temperature. The solid was filtered off, washed with several
portions of diethyl ether and dried to yield 6 as yellow
crystals (6.50 g, 96.9%), m.p. 151 °C (lit. 153–155 °C).
4
3J = 8.8 Hz, J = 2.2 Hz, 1H, quinazolineH7), 5.62 (s, 1H,
quinazolineNH2, exchanged by D2O). Anal. C14H10ClFN4.
6.1.6. 2-{[4-(3-Bromo-phenylamino)-quinazolin-6-ylimino]-
methyl}-benzene-1,4-diol (9a)
A mixture of 8a (2.00 g, 6.4 mmol) and 2,5-dihydroxy-
benzaldehyde (0.88 g, 6.4 mmol) was refluxed in 45 ml of a
mixture of ethanol and glacial acetic acid (1:1) for 2.5 h. The
ethanol was then removed in vacuum. The residual mixture
was poured onto ice, neutralized with Na2CO3 solution and
extracted with ethyl acetate. The combined extracts were
washed with water, dried over anhydrous Na2SO4 and the
solvent was removed in vacuum. The crude product was
purified by recrystallization from dichloromethane/methanol
(9 + 1) to yield 9a as yellow-orange powder (2.11 g, 76.5%),
m.p. 243–246 °C (dec.). 1H-NMR (400 MHz): 12.09 (s, 1H,
1,4-Di-OH-ArOH1, exchanged by D2O), 9.90 (s, 1H,
anilino-NH, exchanged by D2O), 9.17 (s, 1H, 1,4-Di-OH-
ArOH4, exchanged by D2O), 9.03 (s, 1H, 1,4-Di-OH-ArCH),
8.65 (s, 1H, quinazolineH2), 8.52 (d, 4J = 2.0 Hz, 1H, quina-
6.1.2. 6-Nitro-4-(3-bromo-phenylamino)-quinazoline (7a)
The preparation of 7a was executed according to a pu-
blished procedure [21]. Therefore, a mixture of 6 (5.00 g,
22.9 mmol) and 3-bromoaniline (3.94 g, 22.9 mmol) was
refluxed in glacial acetic acid. After 60 min the mixture was
filtered hot. The solid was washed with diethyl ether and
dried to yield 7a as yellow crystals (6.05 g, 76.5%),
m.p. ≥ 250 °C (dec., lit. 267–270 °C).
6.1.3. 6-Nitro-4-(3-chloro-4-fluoro-phenylamino)-quinazo-
line (7b)
Compound 7b was obtained similarly to 7a from com-
pound 6 (5.00 g, 22.9 mmol) and 3-chloro-4-fluoroaniline
(3.33 g, 22.9 mmol) as yellow crystals (5.65 g, 77.3%),
3
zolineH5), 8.26 (s, 1H, anilinoH2), 8.02 (dd, J = 8.8 Hz,
1
3
m.p. ≥ 239 °C (dec.). H-NMR (400 MHz): 10.49 (s, 1H,
4J = 2.0 Hz, 1H, quinazolineH7), 7.95 (d, J = 8.2 Hz, 1H,
4
anilino-NH, exchanged by D2O), 9.60 (d, J = 1.7 Hz, 1H,
anilinoH4), 7.88 (d, 3J = 8.9 Hz, 1H, quinazolineH8), 7.39 (t,
3
quinazolineH5), 8.75 (s, 1H, quinazolineH2), 8.55 (dd,
3J = 8.0 Hz, 1H, anilinoH5), 7.30 (d, J = 8.2 Hz, 1H,

R. Albuschat et al. / European Journal of Medicinal Chemistry 39 (2004) 1001–1011
1009
3
4
anilinoH6), 7.11 (d, 4J = 2.9 Hz, 1H, 1,4-Di-OH-ArH5), 6.92
(dd, 3J = 8.9 Hz, 4J = 2.9 Hz, 1H, 1,4-Di-OH-ArH3), 6.84 (d,
3J = 8.9 Hz, 1H, 1,4-Di-OH-ArH2). Anal. C21H15BrN4O2.
quinazolineH8), 7.37 (dd, J = 9.1 Hz, J = 2.1 Hz 1H,
3
quinazolineH7), 7.34.7.31 (t, J = 8.1 Hz, 1H, anilinoH5),
7.29 (d, 4J = 1.9 Hz, 1H, quinazolineH5), 7.24 (d, 3J = 8.1 Hz,
1H, anilinoH6), 6.71 (d, 4J =2.8 Hz, 1H, 1,4-Di-OH-ArH5),
6.65 (d, 3J = 8.6 Hz, 1H, 1,4-Di-OH-ArH2), 6.48 (dd,
6.1.7. 2-{[4-(3-Chloro-4-fluoro-phenylamino)-quinazolin-
6-ylimino]-methyl}-benzene-1,4-diol (9b)
3J = 8.5 Hz, J = 2.9 Hz, 1H, 1,4-Di-OH-ArH4), 6.41 (t,
4
3J = 5.6 Hz, 1H, quinazoline-NH, exchanged by D2O), 4.31
Compound 9b was obtained similarly to 9a from 8b
(1.00 g, 3.5 mmol) and 2,5-dihydroxybenzaldehyde (0.48 g,
3.5 mmol). The crude product was purified by recrystalliza-
tion from dichloromethane/methanol (9 + 1) to yield 9b as an
orange powder (1.10 g, 77.5%), m.p. 257 °C. 1H-NMR
(400 MHz): 12.09 (s, 1H, 1,4-Di-OH-ArOH1, exchanged by
D2O), 9.94 (s, 1H, anilino-NH, exchanged by D2O), 9.17 (s,
1H, 1,4-Di-OH-ArOH4, exchanged by D2O), 9.03 (s, 1H,
1,4-Di-OH-ArCH), 8.65 (s, 1H, quinazolineH2), 8.49 (d,
4J = 2.2 Hz, 1H, quinazolineH5), 8.24 (dd, 4J = 2.6 Hz, 4J (H,
F) = 6.9 Hz, 1H, anilinoH2), 8.00 (dd, 3J = 9.2 Hz,
4J = 2.1 Hz, 1H, quinazolineH7), 7.93.7.79 (m, 2H, quinazo-
lineH8, anilinoH6), 7.48 (t, 3J (H, F) = 9.1 Hz, 1H,
(d, 3J
=
5.5 Hz, 2H, 1,4-Di-OH-ArCH2). Anal.
C21H17BrN4O2·H2O.
6.1.9. 2-{[4-(3-chloro-4-fluoro-phenylamino)-quinazolin-
6-ylamino]-methyl}-benzene-1,4-diol (10b)
Compound 9b(DMF) (1.50 g, 3.0 mmol) was reduced
similarly to 9a by dimethylamine borane (0.20 g, 3.4 mmol)
to yield 10b(DMF), which contained 1 mol water and 1 mol
dimethylformamide per mol 10b. Compound 10b(DMF)
was obtained as bride yellow crystals (1.10 g, 70.5%), m.p.
230 °C (dec.). 1H-NMR (400 MHz): 9.44 (s, 1H, anilino-NH,
exchanged by D2O), 8.88 (s, 1H, 1,4-Di-OH-ArOH1, ex-
changed by D2O), 8.62 (s, 1H, 1,4-Di-OH-ArOH4, exchan-
4
anilinoH5), 7.11 (d, J = 2.9 Hz, 1H, 1,4-Di-OH-ArH5),
3
4
4
6.92 (dd, J = 8.7 Hz, J = 2.9 Hz, 1H, 1,4-Di-OH-ArH3),
6.84 (d, 3J = 8.9 Hz, 1H, 1,4-Di-OH-ArH2). Anal.
C21H14ClFN4O2·0.5H2O.
ged by D2O), 8.36 (s, 1H, quinazolineH2), 8.13 (dd, J (H,
F) = 6.9 Hz, J = 2.6 Hz, 1H, anilinoH2), 7.95 (s, 1H,
4
DMF-CH), 7.85.7.76 (m, 1H, anilinoH6), 7.55 (d,
3J = 9.0 Hz, 1H, quinazolineH8), 7.46.7.34 (m, 2H, anili-
When the crude product was dissolved in dimethylforma-
mide and then purified by flash chromatography on silica gel
(ethyl acetate/n-hexane, 9:1), 9b(DMF) was obtained, which
contained 1 mol water and 1 mol dimethylformamide per mol
9b. 12.08 (s, 1H, 1,4-Di-OH-ArOH1, exchanged by D2O),
9.94 (s, 1H, anilino-NH, exchanged by D2O), 9.17 (s, 1H,
1,4-Di-OH-ArOH4, exchanged by D2O), 9.03 (s, 1H, 1,4-Di-
OH-ArCH), 8.65 (s, 1H, quinazolineH2), 8.49 (d, 4J = 1.7 Hz,
1H, quinazolineH5), 8.22 (dd, 4J = 2.5 Hz, 4J (H, F) = 6.8 Hz,
4
noH5, quinazolineH7), 7.27 (d, J = 2.2 Hz, 1H, quinazoli-
4
neH5), 6.71 (d, J =2.9 Hz, 1H, 1,4-Di-OH-ArH5), 6.65 (d,
3J = 8.5 Hz, 1H, 1,4-Di-OH-ArH2), 6.49 (dd, J = 8.5 Hz,
3
4J = 2.9 Hz, 1H, 1,4-Di-OH-ArH3), 6.41 (t, 3J = 5.6 Hz, 1H,
3
quinazoline-NH, exchanged by D2O), 4.31 (d, J = 5.6 Hz,
2H, 1,4-Di-OH-CH2), 2.89 (s, 3H, DMF-CH3), 2.73 (s, 3H,
DMF-CH3). Anal. C21H16ClFN4O2·C3H7NO·H2O.
Removal of the included solvents is possible by recrystal-
lization from ethanol/petrolether. Pale yellow powder, m.p.
214 °C. H-NMR (400 MHz): 9.44 (s, 1H, anilino-NH,
exchanged by D2O), 8.88 (s, 1H, 1,4-Di-OH-ArOH1, ex-
changed by D2O), 8.62 (s, 1H, 1,4-Di-OH-ArOH4,
exchanged by D2O), 8.36 (s, 1H, quinazolineH2), 8.13 (dd, 4J
3
4
1H, anilinoH2), 8.01 (dd, J = 8.9 Hz, J = 1.9 Hz, 1H,
quinazolineH7), 7.95 (s, 1H, DMF-CH) 7.92.7.82 (m, 2H,
1
3
quinazolineH8, anilinoH6), 7.48 (t, J (H, F) = 9.1 Hz, 1H,
anilinoH5), 7.12 (d, 4J = 2.8 Hz, 1H, 1,4-Di-OH-ArH5), 6.91
(dd, 3J = 8.8 Hz, 4J = 2.8 Hz, 1H, 1,4-Di-OH-ArH3), 6.85 (d,
3J = 8.8 Hz, 1H, 1,4-Di-OH-ArH2), 2.89 (s, 3H, DMF-CH3),
2.73 (s, 3H, DMF-CH3).
4
(H, F) = 6.9 Hz, J = 2.6 Hz, 1H, anilinoH2), 7.86.7.75
(m, 1H, anilinoH6), 7.55 (d, 3J = 9.0 Hz, 1H, quinazolineH8),
7.43 (t, 3J (H, F) = 9.2 Hz, 1H, anilinoH5), 7.36 (dd,
3J = 9.0 Hz, J = 2.2 Hz, 1H, quinazolineH7), 7.27 (d,
4
6.1.8. 2-{[4-(3-bromo-phenylamino)-quinazolin-6-ylamino]-
methyl}-benzene-1,4-diol (10a)
4J = 2.2 Hz, 1H, quinazolineH5), 6.71 (d, J =2.9 Hz, 1H,
4
3
1,4-Di-OH-ArH5), 6.65 (d, J = 8.5 Hz, 1H, 1,4-Di-OH-
Dimethylamine borane (0.14 g, 2.4 mmol) was dissolved
in 10 ml glacial acetic acid and added dropwise to a cooled
(0–5 °C) suspension of 9a (1.0 g, 2.3 mmol) in glacial acetic
acid. The mixture was stirred at room temperature for 60 min,
then poured onto ice and neutralized with 5N NaOH. The
solid was collected, washed with water and dried. The resul-
ting crude product was recrystallized from dichloro-
methane/methanol (9+1) to yield 10a as yellow powder
3
4
ArH2), 6.49 (dd, J = 8.5 Hz, J = 2.9 Hz, 1H, 1,4-Di-OH-
ArH3), 6.41 (t, 3J = 5.6 Hz, 1H, quinazoline-NH, exchanged
by D2O), 4.31 (d, 3J = 5.6 Hz, 2H, 1,4-Di-OH-ArCH2). Anal.
C21H16ClFN4O2.
6.2. X-ray crystal structure analysis
1
(0.77 g, 76.5%), m.p. 231 °C (dec). H-NMR (400 MHz):
After several attempts with different solvent combinations
small crystals of 10a were successfully obtained by vapour
diffusion from DMF/acetone. Diffraction data of a crystal
with dimension 0.24 × 0.20 × 0.10 mm were collected at
100 K on a Bruker SMART 1000 difractometer with MoKa-
radiation (graphite monochromator, k = 0.7107 Å) and CCD
9.41 (s, 1H, anilinoNH, exchanged by D2O), 8.88 (s, 1H,
1,4-Di-OH-ArOH1, exchanged by D2O), 8.61 (s, 1H, 1,4-Di-
OH-ArOH4, exchanged by D2O), 8.38 (s, 1H, quinazoli-
neH2), 8.16 (d, 4J = 1.8 Hz, 1H, anilinoH2), 7.88 (d,
3
3J = 8.2 Hz, 1H, anilinoH4), 7.54 (d, J = 9.0 Hz, 1H,

1010
R. Albuschat et al. / European Journal of Medicinal Chemistry 39 (2004) 1001–1011
area detector. The crystal lattice is triclinic, space group
P1bar, and unit cell parameters a = 8.572(1), b = 11.249(1),
c = 12.800(1) Å, ␣ = 82.78(2), W = 84.10(2), c = 76.19(2)°,
V = 1185.7(14) Å3, Z = 2, dcalc = 1.480 g × cm–3,
µ = 1.775 mm–1.
Karlsruhe, Germany). All cell lines were grown as mono-
layers in tissue culture flasks in a humidified atmosphere (5%
CO2/95% air) at 37 °C.
Determination of cell viability: Drug sensitivity assess-
ment was performed with cells by theWst-1 cell proliferation
reagent (Roche Molecular Biochemicals, Mannheim, Ger-
many) as described elsewhere [38]. In brief, cells were see-
ded in 96-well plates. Twenty-four hours after plating, the
drugs were diluted in culture medium and added to the cells.
For evaluation of the antiproliferative and cytotoxic effects,
human astrocytoma/glioblastoma cell lines (U87MG,
A172 and T98G) were exposed for 24–96 h to 0.1–100 µM
10a or 10b. Viability was determined after continuous expo-
sure to the drug of interest for the indicated times by addition
of Wst-1 solution. Absorbance at 450 nm (k reference:
620 nm) was determined upon incubation of the cells at 37 °C
in the presence of the Wst-1 solution for 60 min using a
Anthos 2010 plate reader (Anthos, Krefeld, Germany). The
DC50 represents the drug concentration causing a reduction
in the number of viable cells to 50% of untreated control
cells.
The 14636 reflections were collected up to 2hmax = 62°
and reduced to 7155 independent reflections (Rint = 0.034).
The phase problem was solved with direct methods (SIR92)
[31], the structure was refined on F2 with SHELXL with
anisotropic displacement parameters for the non H atoms
[32]. The hydrogens which were located from difference
syntheses were given isotropic displacement parameters. The
asymmetric unit was found to consist of one molecule 10a,
one DMF molecule and one water molecule. An absorption
correction was also applied (SADDABS) [33]. After conver-
gence of the refinements discrepancy factors were
R1 = 0.0444, wR2 = 0.0983 for I > 2r(I) and R1 = 0.0823,
wR2 = 0.1076 for all data, Gof = 0.987.
6.3. Pharmacology
6.3.1. In vitro EGFR tyrosine kinase activity
The EGFR tyrosine kinase inhibitory activity assay was
carried out by MDS Pharma Services, Taiwan (R.O.C.).
Enzyme source: Human A431 cells
6.3.3. Evaluation for cytotoxic activity against a panel
of 60 human cancer cell lines
Evaluation of anticancer activity was performed at the
National Cancer Institute (Bethesda, MD, USA).
Substrate:
10
µg/ml
Polyglutamic
acid/tyrosine
[poly(glu/ptyr, 4:1)]
The compounds were tested in an in-vitro 60-cell line
anticancer assay over a 5 log dose range. The cell lines are
derived from nine human cancer types: leukemia, NSCLC,
colon, CNS, melanoma, ovarian, renal, prostate and breast.
The tumour cell lines are grown in RPMI 1640 medium
containing 5% fetal bovine serum and 2 ml L-glutamine. For
a typical screening experiment, cells are inoculated into
96 well microtiter plates in 100 µM at plating densities
ranging from 5000 to 40 000 cells/well depending on the
doubling time of individual cell lines. After cell inoculation,
the microtiter plates are incubated at 37 °C, 5% CO2, 95% air
and 100% relative humidity for 24 h prior to addition of
experimental drugs.
Vehicle: 1% dimethylsulfoxide
Pre-incubation time/temp.: None
Incubation time/temp.: 60 min/25 °C
Incubation buffer: 50 mM Hepes, 20 mM MgCl2, 0.2 mM
Na3VO4, pH 7.4
Quantitation Method: ELISA quantitation of Poly (glu: ptyr)
Significance Criteria: ≥50% of maximal stimulation or inhi-
bition
The epidermal growth factor receptor was purchased from
Sigma (prod. number: E2645). The enzyme assay was per-
formed in wells coated with poly(glu/tyr) substrate. Incuba-
tion buffer and test compound were added into the wells.
Finally, enzyme was added and thenATP to start the reaction.
For the calculation of the inhibitory activities, each assay
contained two wells for determination of the maximum phos-
phorylation rate (ptyrmax: in the absence of inhibitor) and two
wells for determination of the minimum phosphorylation rate
(ptyrmin: in the absence of enzyme), as well. Quantitation of
phospho-tyrosine was determined via ELISA [34–37].
After 24 h, two plates of each cell line are fixed in situ with
TCA, to represent a measurement of the cell population for
each cell line at the time of drug addition (Tz).
Experimental drugs are solubilized in DMSO at 400-fold
the desired final maximum test concentration. At the time of
drug addition, an aliquot of concentrate is thawed and diluted
to twice the desired final maximum test concentration with
complete medium containing 50 µg/ml gentamicin. Additio-
nal four, 10-fold or 1/2 log serial dilutions are made to
provide a total of five drug concentrations plus control. Ali-
quots of 100 µl of these different drug dilutions are added to
the appropriate microtiter wells already containing 100 µl of
medium, resulting in the required final drug concentration.
Following drug addition, the plates are incubated for an
additional 48 h at the above-mentioned conditions. For adhe-
rant cells, the assay is terminated by the addition of cold
TCA.
6.3.2. Determination of cytostatic/cytotoxic activity
in human malignant glioma cell lines
Cell lines: The human astrocytoma/glioblastoma cell lines
A172, U87MG and T98G were obtained from ATCC (Roc-
kville, MD, USA). All cell lines were adapted to
RPMI1640 medium containing 2 mM glutamine and 2 g/l
sodium bicarbonate (PAA, Linz, Austria) supplemented with
10% (v/v) heat-inactivated fetal bovine serum and 100 U/ml
penicillin and 100 µg/ml streptomycin (Life Technologies,

R. Albuschat et al. / European Journal of Medicinal Chemistry 39 (2004) 1001–1011
1011
[3] D. Salomon, W. Gullick, Signal J. 2 (3) (2001) 4–11.
[4] M.W. Pederson, H.S. Poulson, Sci. Med. (2002) 32–43.
Cells are fixed in situ by the gentle addition of 50 µl of cold
50% (w/v) TCA (final concentration, 10% TCA) and incuba-
ted for 60 min at 4 °C. The supernatant is discarded, and the
plates are washed five times with tap water and air dried.
Sulforhodamine B (SRB) solution (100 µl) at 0.4% (w/v) in
1% acetic acid is added to each well, and plates are incubated
for 10 min at room temperature. After staining, unbound dye
is removed by washing five times with 1% acetic acid and the
plates are air dried. Bound stain is subsequently solubilized
with 10 mM trizma base, and the absorbance is read on an
automatic plate reader at a wavelength of 515 nm.
[5] E. Raymond, S. Faivre, J.P. Armand, Drugs 60 (1) (2000) 15–23.
[6] A. Levitzki, A. Gazit, Science 267 (1995) 1782–1788.
[7] C.L. Arteaga, J. Clin. Oncol. 19 (15) (2001) 32S–40S.
[8] F. Ciardiello, R. Caputo, R. Bianco, D. Vincenzo, G. Somatico, S. De
Placido, R.A. Bianco, G. Tortora, Clin. Cancer Res. 6 (2000) 2053–
2063.
[9] V. Pollack, D.M. Savage, D.A. Baker, et al., J. Pharmacol. Exp. Ther.
291 (2) (1999) 739–748.
[10] R. Albuschat, in: Doctoral Thesis, Free University of Berlin, 2003,
pp. 90–93.
[11] T. Onoda, H. Iinuma,Y. Sasaki, M. Hamada, K. Isshiki, H. Naganawa,
T.J. Takeuchi, J. Nat. Prod. 42 (6) (1989) 1252–1257.
[12] C.N. Hodge, J.A. Pierce, Bioorg. Med. Chem. Lett 3 (1993) 1605–
1608.
[13] P. Traxler, J. Green, H. Mett, U. Sequin, P. Furet, J. Med. Chem. 42 (6)
(1999) 1018–1026.
For suspension cells, the methodology is the same except
that the assay is terminated by fixing settled cells at the
bottom of the wells by gently adding 50 µl of 80% TCA (final
concentration, 16% TCA).
[14] G.W. Rewcastle, et al., J. Med. Chem. 38 (18) (1995) 3482–3487.
[15] J. Baselga, S.D. Averbruch, Drugs 60 (1) (2000) 33–40.
[16] V. Sherwood, A.J. Bridges, W.A. Denny, G.W. Rewcastle, J.B. Smaill,
D.W. Fry, Proc. Am. Assoc. Cancer Res. 41 (2000) 3076.
[17] J.B. Smaill, et al., J. Med. Chem. 42 (10) (1999) 1803–1815.
[18] M.J. Morin, Oncogene 19 (2000) 6574–6583.
[19] P.W. Vincent, S.J. Patmore, B.E. Atkinson, A.J. Bridges, L.S. Kirkish,
R.C. Dudeck, W.R. Leopold, H. Zhou, W.L. Elliott, Proc. Am. Assoc.
Cancer Res. 40 (1999) 117.
Using the seven absorbance measurements [time zero,
(Tz), control growth, (C), and test growth in the presence of
drug at the five concentration levels, (Ti)], the percentage
growth is calculated at each of the drug concentrations levels.
Percentage growth is calculated as:
T − T ⁄ C −
͒͑
i z
͓
͑
T
×100 for concentrations for which Ti ≥ Tz and T −
͓͑
i
͒
͔
z
T ⁄T ×100 for concentrations for which T < T .
͒
͔
i
z
z
z
[20] D.W. Fry, A.J. Bridges, A. Doherty, K. Greis, J.L. Hicks, K.E. Hook,
P.R. Keller, W.R. Leopold, J. Loo, D.J. McNamara, J.M. Nelson,
V. Sherwood, J.B. Smaill, S. Trumpp-Kallmeyer, E. Dobrusin, Proc.
Natl. Acad. Sci. USA 95 (1998) 12022–12027.
[21] H.R. Tsou, et al., J. Med. Chem. 44 (3) (2001) 2719–2734.
[22] J.H. Billmann, J.W. McDowell, J. Org. Chem. 26 (1961) 1437–1440.
[23] E. Keller, SCHAKAL 88 A Fortran Program for the Graphical Repre-
sentation of Molecular and Crystallographic Models, University of
Freiburg, Germany, 1988.
[24] M.R. Grever, B.A. Schepartz, Life Sci. 57 (1995) 131–141.
[25] M.R. Boyd, K.D. Paul, Drug Dev. Res. 34 (1995) 91.
[26] W.K. Hong, A. Ullrich, Oncol Biother. 1 (1) (2000) 2–6.
[27] C.L. Sawyers, N. Engl. J. Med. 340 (1999) 1330–1340.
[28] E. Buchdunger, J. Zimmermann, H. Mett, et al., Cancer Res. 56 (1996)
100–104.
Three dose response parameters are calculated for each
experimental agent. Growth inhibition of 50% (GI50) is cal-
culated from T − T ⁄ C − T ×100 = 50, which is the
͓
͑
͒͑
͔͒
z
z
drug concentratioin resulting in a 50% reduction in the net
protein increase.
The drug concentration resulting in total growth inhibition
(TGI) is calculated from Ti = Tz. The LC50 (concentration of
drug resulting in a 50% reduction in the measured protein at
the end of the drug treatment is calculated from
T −
͓
͑
i
T ⁄T ×100 = − 50 .
͒
͔
͑
͒
z
z
Values are calculated for each of these three parameters if
the level of activity is reached; however, if the effect is not
reached or exceeded, the value for that parameter is ex-
pressed as greater or less than the maximum or minimum
concentration tested [39,40].
[29] N.N. Glivec, Summary of Product Characteristics, Novartis Pharma
AG (2001) (imatinib).
[30] M. Carroll, S. Ohno-Jones, S. Tamura, et al., Blood 90 (1997) 4947–
4952.
[31] A. Altomare, G. Cascarono, C. Giacovazzo, A. Guagliardi,
M.C. Burla, G. Polidori, M. Camalli, J. Appl. Crystallogr. 27 (1994)
435.
[32] G.M. Sheldrick, SHELXL-97 A Program for Refinement of Crystal
Structures Technical Report, University of Göttingen, Germany, 1997.
[33] Bruker, SMART, SAINT-Plus and SADABS, in: Data Collection and
Processing Software for the SMART System, Bruker AXS, Madison,
Wisconsin, USA, 2002.
[34] J.F. Geissler, P. Traxler, U. Regenass, B.J. Murray, J.L. Roesel,
T. Meyer, E. McGlynn, A. Storni, N.B. Lydon, J. Biol. Chem. 265
(1990) 22255–22261.
[35] P.S. Wedegartner, G.N. Gill, J. Biol. Chem. 264 (1989) 11346–11353.
[36] P.Yaish, A. Gazit, C. Golin, A. Levitzki, Science 242 (1988) 933–935.
[37] A. Lin, MDS Pharma Services, Taiwan, R.O.C., correspondence.
[38] V. Jendrossek, B. Erdlenbruch, H. Hunold, W. Kugler, H. Eibl,
M. Lakomek, Intern. J. Oncol. 14 (1999) 15–22.
Acknowledgments
The excellent technical support by Ilka Müller is highly
appreciated. The work was supported by a grant from the
Federal Ministry of Education and Research (Fö. 01KS9602)
and the Interdisciplinary Center of Clinical Research Tübin-
gen (IZKF) to V.J.
The authors thank the Antitumour Evaluation Branch and
Biological Testing Branch of the NCI for performing biolo-
gical evaluations.
[39] M.C. Alley, D.A. Scudiero, P.A. Monks, M.L. Hursey, M.J. Czerwin-
ski, D.L. Fine, B.J. Abbott, J.G. Mayo, R.H. Shoemaker, M.R. Boyd,
Cancer Res. 48 (1988) 589–601.
References
[1] S.A. Aaronson, Science 254 (1991) 1146–1152.
[2] Y. Yarden, M.X. Sliwkowski, Nat. Rev. Mol. Cell Biol. 2 (2001) 127.
[40] M.R. Grever, S.A. Schepartz, B.A. Chabner, Sem. Oncol. 19 (6)
(1992) 622–638.