Radioiodination of new EGFR inhibitors as potential SPECT agents for molecular imaging of breast cancer

Article abstract of DOI:10.1016/j.bmc.2007.04.008

In our search for the development of novel SPECT radioligands for EGFR positive tumours, new potentially irreversible tyrosine kinase (TK) inhibitors are being explored. The radioiodination of N-{4-[(3-chloro-4-fluorophenyl) amino]quinazoline-6-yl}-3-brom

Full text of DOI:10.1016/j.bmc.2007.04.008

Bioorganic & Medicinal Chemistry 15 (2007) 3974–3980
Radioiodination of new EGFR inhibitors as potential
SPECT agents for molecular imaging of breast cancer
a
a
a
b
´
Celia Fernandes, Cristina Oliveira, Lurdes Gano, Athanasia Bourkoula,
Ioannis Pirmettisb and Isabel Santosa,*
aDepartamento de Quı´mica, ITN, Estrada Nacional 10, 2686-953, Sacave´m Codex, Portugal
bInstitute of Radioisotopes-Radiodiagnostic Products, NCSR ‘‘Demokritos’’, Athens, Greece
Received 15 November 2006; revised 29 March 2007; accepted 5 April 2007
Available online 10 April 2007
Abstract—In our search for the development of novel SPECT radioligands for EGFR positive tumours, new potentially irreversible
tyrosine kinase (TK) inhibitors are being explored. The radioiodination of N-{4-[(3-chloro-4-fluorophenyl) amino]quinazoline-6-yl}-
3-bromopropionamide, a novel EGFR-TK inhibitor synthesised in our laboratory, was accomplished via halogen exchange.
Purification by RP-HPLC gave [125I]-N-{4-[(3-chloro-4-fluorophenyl)amino]quinazoline-6-yl}-3-iodopropionamide with a radio-
chemical purity higher than 95% and a high specific activity. In vitro studies indicate that both iodinated quinazoline and its bromo
precursor inhibit A431 cell growth and also possess higher potency than the parent quinazoline to inhibit the EGFR autophospho-
rylation. In vivo stability studies suggest metabolization of the radioiodinated quinazoline indicating a short biological half-life. The
in vitro results point out that these quinazoline derivatives could be promising candidates for SPECT imaging of EGFR positive
tumours provided that they are selectively modified in order to achieve better in vivo radiochemical stability.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
EGFR. The Gefitinib (ZD1839, Iressaꢁ)2 was approved
by the FDA as a third-line treatment of chemotherapy-
refractory non-small cell lung carcinomas. The Erlotinib
(OSI-774, Tarcevaꢀ)3 is presently undergoing phase 3
clinical trials (Fig. 1).
The epidermal growth factor receptor (EGFR) belongs
to the ErbB family of receptor tyrosine kinases (TK) in-
volved in the proliferation of normal and malignant
cells. EGFR is often overexpressed in many tumour
cells, namely in breast and prostate cancer and non-
small cell lung carcinoma.1 Therefore, the EGFR has
become an attractive target for the design and develop-
ment of potential anticancer drugs. Several compounds
that selectively bind the receptor and inhibit its tyrosine
kinase activity have been reported. The most active tyro-
sine kinase inhibitors are quinazoline-based small mole-
cules that compete with adenosine triphosphate (ATP)
and prevent its binding to the intracellular tyrosine
kinase region. Such compounds can serve for therapeu-
tics and, when labelled, as targeted diagnostic agents.
The two compounds that are at the most advanced stage
of development are the 4-(phenylamino)quinazolines,
Gefitinib2 and Erlotinib,3 both of which reversibly target
A predictive marker that will help to select patients for
treatment with EGFR inhibitors is needed to optimise
the therapeutic potential of EGFR inhibition in cancer.
Recently, there has been a growing interest in the use of
EGFR-TK inhibitors, such as quinazoline derivatives,
as radiotracers for molecular imaging. Several reversible
and irreversible inhibitors, particularly from the 4-anili-
noquinazoline class, were labelled with positron emitters
such as fluorine-18, carbon-11 and iodine-124 and their
F
HN
N
HN
N
Cl
O
O
O
O
N
N
MeO
MeO
N
MeO
Keywords: EGFR; Tyrosine kinase inhibitors; Quinazoline; SPECT;
Cancer.
Tarceva / Erlotinib
Iressa /Gefitinib
*
Figure 1. Some successful examples of EGFR-TK inhibitors that have
progressed to clinical trials and patient care.
Corresponding author. Tel.: +351 21 9946201; fax: +351 21 994
6185; e-mail: isantos@itn.pt
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.04.008

C. Fernandes et al. / Bioorg. Med. Chem. 15 (2007) 3974–3980
3975
potential as PET biomarkers was evaluated.4–6 In our
search for the development of novel SPECT radioli-
gands for early detection and staging of EGFR positive
tumours, and having in mind that to date there are no
promising SPECT imaging agents for targeting the
EGFR, novel 125I probes based on quinazoline inhibi-
tors of tyrosine kinase activity are being explored with
the ultimate goal of labelling the molecules with 123I,
the most adequate radioisotope of iodine for SPECT
imaging.
treatment of the 4-hydroxy-6-nitroquinazoline 2 with
PCl5/POCl3 at 160 ꢂC.7 4-[(3-Chloro-4-fluorophe-
nyl)amino]-6-nitroquinazoline, 4, was obtained by cou-
pling 3-chloro-4-fluoroaniline to the chlorinated
compound 3. In the next step, the nitro group at the
6-position of the quinazoline ring in 4 was reduced to
amine by hydrogenation with H2 over Raney/Ni at
room temperature affording 6-amino-4-[(3-chloro-4-flu-
orophenyl)amino]quinazoline 5.8 The reduction of this
1
nitro group was confirmed in the H NMR spectrum
by the presence of a singlet (5.61 ppm) corresponding
to the amine protons. The amine 5 was subsequently
reacted with 3-bromopropanoylchloride to give N-{4-
[(3-chloro-4-fluorophenyl) amino]quinazoline-6-yl}-3-
bromopropionamide 6. The structure of this bromo
In this work, we report the synthesis and characteriza-
tion of the novel bioactive precursor, N-{4-[(3-chloro-
4-fluorophenyl)amino]quinazoline-6-yl-3-bromopropi-
onamide, as well as the synthesis of the corresponding
iodinated and radioiodinated analogues. The in vitro
biological behaviour of the halogenated quinazolines
was studied. In vivo stability studies of the radioiodin-
ated compound,
nyl)amino]quinazoline-6-yl}-3-iodopropionamide, are
also reported in this study.
1
precursor was elucidated throughout its H NMR spec-
trum by absence of the amine protons and by the occur-
rence of two triplets, integrating for two protons each, at
2.96 and 3.93 ppm corresponding to the bromopropi-
onamide chain methylene protons.
[
125I]-N-{4-[(3-chloro-4-fluorophe-
Synthesis of N-{4-[(3-chloro-4-fluorophenyl)amino]qui-
nazoline-6-yl}-3-iodopropionamide 7 was carried out
by a halogen exchange reaction of 6 with sodium iodide
in dry acetone at reflux during 11 h. The iodinated qui-
nazoline 7 was obtained in high chemical purity after
semi-preparative reversed-phase HPLC purification
with CH3CN/TFA 0.1% (35:65) as eluent. This mobile
phase was selected in order to allow an efficient
separation of the iodinated quinazoline from its bromo
precursor. The substitution of bromine by an iodine
atom in the propionamide chain was confirmed in the
1H NMR spectrum of 7. Actually, in this spectrum the
methylene protons are relatively shifted appearing at
3.13 and 3.52 ppm, respectively, while the methylene
protons of 6 appear at 2.96 and 3.93 ppm.
2. Results and discussion
2.1. Chemistry
N-{4-[(3-Chloro-4-fluorophenyl)amino]quinazoline-6-
yl}-3-iodopropionamide 7 was prepared from 4-
hydroxyquinazoline 1 as depicted in Scheme 1.
4-Hydroxy-6-nitroquinazoline 2, prepared by nitration
of 4-hydroxyquinazoline 1, was used as the starting
material for the preparation of 4-chloro-6-nitroquinazo-
line 3. Hence the synthesis of 3 was accomplished
according to a slightly modified reported procedure by
F
F
Cl
Cl
OH
OH
Cl
N
HN
N
O2N
O2N
HNO
conc
O2N
PCl5/POCl3
reflux/2
NH2
N
N
N
N
N
N
i
-propanol
reflux
1
4
2
3
H2/Raney Ni
rt/3.5h
F
F
Cl
F
O
HN
HN
N
Cl
Br
Cl
H
Cl
HN
Na127/125
I
H
N
H2N
N
I
N
Br
N
N
O
(CH3)2CO
reflux
Et3N
(CH3)2CO
N
5
O
N
127I] :7
125I] : [125I]-7
6
[
[
Scheme 1. Synthetic pathway of N-{4-[(3-chloro-4-fluorophenyl)amino]quinazoline-6-yl}-3-iodopropionamide 7 and its radioiodinated analogue
125I]-7.
[

3976
C. Fernandes et al. / Bioorg. Med. Chem. 15 (2007) 3974–3980
2.2. Radiochemistry
[
125I]-N-{4-[(3-Chloro-4-fluorophenyl)amino]quinazo-
line-6-yl}-3-iodopropionamide [125I]-7 was prepared by
halogen exchange reaction of the bromo precursor 6
in the presence of [125I]-NaI at 70 ꢂC for 2–2.5 h
(Scheme 1). The radioiodinated quinazoline [125I]-7,
obtained in 48% radiochemical yield, was purified by
analytical reversed-phase HPLC with simultaneous
UV (254 nm) and radioactivity detection. The experi-
mental HPLC conditions described above for the puri-
fication of the non-radioiodinated quinazoline 7 have
been used to allow an efficient separation of [125I]-7
from other chemical and radiochemical species as well
as from the bromo precursor, leading to compound
[
125I]-7 with a high specific activity (Fig. 2). Since no
UV absorption at the most sensitive detector setting
was observed for the purified radioiodinated quinazo-
line it was then assumed that its specific activity is in
the same range as that of the starting [125I]-NaI,
2200 Ci/mmol. From a biological perspective, high
specific activity is an important requirement to image
receptor positive tumours in order to prevent satura-
tion of targeted receptors.
Figure 3. HPLC analysis of radioiodinated quinazoline [125I]-7 co-
injected with its unlabelled analogue 7. Analysis was carried out on a
Nucleosil C18 reversed phase column (250 · 4 mm, 10 lm) eluted with
a mixture of CH3CN and 0.1% TFA (35:65) at 1 mL/min. The eluent
was simultaneously monitored by a UV detector (254 nm) and a
radioactivity detector.
2.3. In vitro assays
The authenticity of the radioiodinated quinazoline
[
125I] -7 was established by comparing its chromato-
graphic behaviour with that of the unlabelled analogue
7 (Fig. 3).
2.3.1. Biological evaluation. As quinazoline is a relatively
small biomolecule its radioiodination should be
performed in such a way that its affinity for the EGFR-
associated tyrosine kinase (EGFR-TK) will not be
disturbed. In order to assess the EGFR affinity of the
novel quinazoline analogues both the cell proliferation
inhibition and the EGFR autophosphorylation inhibi-
tion were evaluated in this study.
As shown in the radiochromatogram of Figure 3 the
radiolabelled compound was obtained in radiochemical
purity higher than 98%. High radiochemical stability in
the elution solvent was found up to two weeks. The
in vitro stability of a radioiodinated compound under
physiological conditions is also a very important factor
to be taken into account in the evaluation of its clinical
potential, since dehalogenation leads to free iodide,
which may result in undesirable biodistribution of radio-
activity. In vitro stability studies have demonstrated that
the radioiodinated quinazoline [125I]-7 is radiochemi-
cally stable in saline solution up to 4 days at 6 ꢂC and
37 ꢂC and also in fresh human serum up to 4 h of incu-
bation at 37 ꢂC. However, 28% of free iodine was ob-
served after 21 h incubation time.
2.3.1.1. Inhibition of cell proliferation by MTT assay.
To evaluate the effect of introducing an b-halopropion-
amide moiety at the 6-position of the quinazoline ring
the ability of the halogenated analogues 6 and 7 to inhi-
bit intact A431 cell growth was evaluated by means of a
MTT assay. The results were compared to the parent
quinazoline 5. Both halogenated quinazoline derivatives
6 and 7 inhibit cell growth at lower concentrations (IC50
values 3–12 lM and 2–7 lM, respectively) compared to
the parent compound 5 (IC50 values 25–30 lM).
2.3.1.2. Inhibition of EGFR autophosphorylation. The
inhibitory effect of the halogenated quinazolines 6 and 7
and the parent quinazoline 5 upon EGFR autophospho-
rylation was assessed by incubation of the compounds,
at various concentrations ranging from 0.1 to 5 lM,
with intact A431 cells stimulated for 2 h with EGFR,
as described in the experimental section. Cell lysates
were loaded onto 8% SDS–PAGE and antiphosphotyro-
sine content was measured. Controls were run in the
same experimental conditions without compounds but
with EGF (+) or without EGF (ꢀ). Band quantification
has been performed using Band leader ver. 3.00 program
and the percentage of EGFR autophosphorylation inhi-
bition at 1 lM has been found to be 95%, 100% and
100% for compounds 5, 6 and 7, respectively. As an
example, the results obtained for the parent quinazoline
Figure 2. HPLC purification of radioiodinated quinazoline [125I]-7.
Isolation was carried out on a Nucleosil C18 reversed phase column
(250 · 4 mm, 10 lm) eluted with a mixture of CH3CN and 0.1% TFA
(35:65) at 1 mL/min. The eluent was simultaneously monitored by a
UV detector (254 nm) and a radioactivity detector. The peaks eluting
between 6 and 10 min correspond to unidentified chemical impurities.

C. Fernandes et al. / Bioorg. Med. Chem. 15 (2007) 3974–3980
3977
iodine, based on HPLC analysis. The low in vivo stabil-
ity of compound [125I]-7, which has not been anticipated
by in vitro studies, may be related with the low chemical
stability of the aliphatic carbon–iodine bond.10 This
finding has discouraged us to proceed with in vivo stud-
ies in tumour bearing animal models.
+EGF
-EGF
7
5
Figure 4. Inhibitory effect of compounds 5 and 7 upon EGFR
autophosphorylation.
3. Conclusions
In order to find a potential SPECT biomarker for
molecular imaging of EGFR positive tumours the novel
5 and for the iodinated quinazoline 7 are presented in
Figure 4. At the lowest concentration tested (0.1 lM)
the halogenated derivatives (6 and 7) showed almost
complete inhibition of phosphorylation, suggesting that
the IC50 values are in the low nM range.
quinazoline derivative
ophenyl)amino]quinazoline-6-yl}-3-iodopropionamide,
125I]-7, has been synthesised and characterized by com-
[
125I]-N-{4-[(3-chloro-4-fluor-
[
paring its chromatographic behaviour with that of its
non-radioiodinated analogue 7 which has been fully
characterized by multinuclear NMR spectroscopy, ele-
mental analysis and HPLC.
2.3.2. Lipophilicity. The lipophilicity is a fundamental
physicochemical property that plays an important role
in the tissue permeability of a ligand affecting its ability
to enter target tissues. The lipophilicity of [125I]-7 was
assessed using the octanol/PBS distribution coefficients
(logPo/w) defined as the concentration ratio of the com-
pounds in octanol phase and in aqueous phase at phys-
iological pH. The logPo/w found was 2.25 at pH 7.4.
This value is in the range (1.5–3) usually described as
optimal for a receptor binding radiotracer to cross the
cellular membrane by passive diffusion and reach the
target tissue without leading to high nonspecific binding,
which could compromise image contrast.9
The radiolabelled quinazoline [125I]-7 was obtained with
high radiochemical purity (>98%) and high specific
activity. In vitro assays indicate that the halogenated
quinazolines 6 and 7 inhibit intact A431 cell growth
and also inhibit completely autophosphorylation of the
receptor in intact A431 cells at the concentration of
0.1 lM. These results demonstrate that the introduction
of a b-halopropionamide chain at the 6 position of the
quinazoline moiety has improved the capability of inhi-
bition of A431 cell growth and autophosphorylation of
EGFR when compared to the parent quinazoline (5).
However, preliminary in vivo studies have shown a
low stability for [125I]-7, due to the deiodination of the
compound, certainly due to the low chemical stability
of the aliphatic carbon-iodine bond. These results may
indicate that [125I]-7 is not a good candidate to image
EGFR in breast cancers. Some other quinazoline deriv-
atives labelled in different positions are being explored
as radiotracers for imaging EGFR positive tumours.
2.3.3. Human serum protein binding. Protein binding af-
fects the tissue distribution and blood clearance of a
radiopharmaceutical and its uptake by the target organ.
Determination of the extent of this non-specific binding
at various time intervals (15 min, 1, 2 and 4 h) has been
accomplished in vitro by incubation of the radiolabelled
quinazoline [125I]-7 in fresh human serum at 37 ꢂC fol-
lowed by precipitation of the plasmatic proteins with
ethanol and comparison of the radioactivity in the pre-
cipitate with the radioactivity in the supernatant to give
the percentage of radiolabelled compound bound to
proteins. Data from these studies indicated a low per-
centage of radiolabelled compound bound to the plas-
matic proteins (6–7%) suggesting a low non-specific
binding. By HPLC analysis only the radiolabelled qui-
nazoline was detected in the supernatant confirming its
high in vitro stability.
4. Experimental
4.1. General
All commercial reagents and solvents were of analytical
grade and used as supplied from Sigma–Aldrich, Merck
or Lancaster. Carrier-free [125I]-NaI was obtained from
Amersham Biosciences, UK.
2.4. In vivo stability studies
Proton and carbon nuclear magnetic resonance spectra
(1H and 13C) were performed on a Varian Unity
300 MHz spectrometer using DMSO or CD3OD as sol-
vents. Infrared analyses were performed on a Bruker
Tensor 27 spectrometer. Elemental analyses were per-
formed on an EA-1110, CE Instruments Equipment.
High performance liquid chromatography (HPLC) anal-
yses were performed on a Perkin-Elmer system equipped
with a biocompatible quaternary pump (series 200), a
UV/vis detector (LC 290, Perkin-Elmer; UV detection
at 254 nm) and a radioactivity detector (LB 509, Bert-
hold). Purification of iodinated compound 7 was carried
out by preparative HPLC on a BondaPack C18 column
In vivo studies of compound [125I]-7 were performed in
CD-1 normal female mice, to evaluate its stability. From
the tissue distribution a general decrease of the radio-
tracer over time in all organs was found, excluding intes-
tine and thyroid. The rapid decrease of the radioactivity
in the liver (t = 1 h, 6.1% ID/organ; t = 4 h, 2.3% ID/or-
gan) and stomach (t = 1 h, 35.2% ID/organ; t = 4 h,
3.9% ID/organ) may indicate that this compound is
deiodinated in these organs. The free iodide formed jus-
tifies the increase in thyroid uptake (t = 1 h, 5.7% ID/or-
gan; t = 4 h, 13.2% ID/organ) and the radiochemical
species found in urine which has been identified as

3978
C. Fernandes et al. / Bioorg. Med. Chem. 15 (2007) 3974–3980
(150 · 19 mm, 10 lm) with a flow rate of 5 mL/min
using acetonitrile/0.1% TFA in water (35/65) as eluent.
Analyses of the labelled and unlabelled compounds as
well as the purification of radioiodinated compound
(approximately 1.0 g) flushing the solution with H2 at
room temperature. After 3.5 h, the mixture was filtered
through Celite and the solvent removed under reduced
pressure to give 5 as an orange solid and in quantitative
1
[
125I]-7 were achieved on a reverse phase EC-Nucleosil
yield. H NMR [d, ppm, (CD3)2SO]: 9.6 (s, 1H, NH),
C18 column (250 · 4 mm, 10 lm) eluted with a flow rate
of 1.0 mL/min and using the same binary isocratic
system.
8.38 (s, 1H, H-2), 8.20 (dd, 1H, H-20), 7.95-7.80
(m, 1H, H-60), 7.53 (d, 1H, H-8), 7.39 (t, 1H, H-50),
7.31 (d, 1H, H-5), 7.24 (dd, 1H, H-7), 5.61 (s, 2H,
NH2). 13C NMR ((CD3)2SO) d (ppm): 156.9, 153.9 (d,
JF-C = 243.2 Hz), 148.6, 147.8, 135.9, 124.9, 124.8,
124.6, 123.4 (d, JF-C = 6,7 Hz), 119.0, 118.8, 116.7 (d,
JF-C = 21,7 Hz), 115.9, 101.1.
4.2. Chemistry
4.2.1. 6-Nitro-4-hydroxyquinazoline (2). Preparation of
this compound followed the general literature proce-
dure.7 4-Hydroxyquinazoline 1 (2.0 g, 0.014 mol) was
added during 30 min to a mixture of concentrated
H2SO4 (4 mL) and fuming HNO3 (4 mL) and heated
for 1 h at 90–95 ꢂC. After cooling the solution was
poured into ice-water (60 mL) to give 2 in almost pure
form (2.3 g, 88%). Pure 4-hydroxyquinazoline was
obtained after purification by silica gel column chroma-
tography eluted with CH3CN/CH2Cl2 (1/3). mp
276–277 ꢂC; 1H NMR [d, ppm, (CD3)2SO]: 12.8 (s,
1 H, OH), 8.80 (d, 1H, H-5), 8.54 (dd, 1H, H-7), 8.32
(s, 1H, H-2), 7.86 (d, 1H, H-8); 13C NMR [d, ppm,
(CD3)2SO]:160.24, 152.95, 148.98, 145.11, 129.14,
128.43, 122.77, 122.02.
4.2.5. N-{4-[(3-Chloro-4-fluorophenyl)amino]quinazoline-
6-yl}-3-bromopropionamide (6). 3-Bromopropanoyl chlo-
ride (446 mg, 2.6 mmol) was added slowly to a solution
of 6-amino-4-[(3-chloro-4-fluorophenyl)amino]quinazo-
line (5) (342 mg, 1.2 mmol) in dry acetone (15 mL). A yel-
low precipitate was formed and the reaction mixture was
left under stirring for 2 h. The precipitate was separated
by filtration and dried under reduced pressure to give 6
as a yellow solid. Yield: (394 mg, 79%). 1H NMR
((CD3)2SO) d (ppm): 3.09 (t, 2H, –CH2–CH2–Br); 3.78
(t, 2H, –CH2–CH2–Br); 7.55 (t, 1H, H-50, J = 8.7 Hz);
7.65 (m, 1H, H-60); 7.91 (d, 1H, H-8); 7.96 (m, 1H+1H,
H-20, H-7); 8.88 (s, 1H, H-2); 9.02 (s, 1H, H-5); 10.76
1
(s, 1H, NH), 11.46 (s, 1H, CONH). H NMR (CD3OD)
4.2.2. 4-Chloro-6-nitroquinazoline (3). To a mixture of
6-nitro-4-hydroxyquinazoline (0.4 g, 2.1 mmol) and pen-
tachlorophosphorane (1 g) was added phosphoric tri-
chloride (2 mL). The suspension was heated under
reflux for 3 h. After cooling, the excess of PCl5 and
POCl3 was removed under reduced pressure and the res-
idue was washed with ligroin. The residue was dissolved
in aqueous Na2CO3 solution and extracted with
CH2Cl2. The organic layer was dried under MgSO4, fil-
tered and the solvent removed to give 3 (0.35 g, 80%):
1H NMR [d, ppm, (CD3)2SO]: 8.80 (d, 1H, H-5), 8.54
(dd, 1H, H-7), 8.34 (s, 1H, H-2), 7.87 (d, 1H, H-8);
13C NMR [d, ppm, (CD3)2SO]: 160.05, 152.53, 149.04,
145.10,128.80, 128.39, 122.68, 121.70.
d (ppm): 2.96 (t, 2H, –CH2–CH2–Br); 3.93 (t, 2H, –CH2–
CH2–Br); 7.37 (t, 1H, H-50); 7.71–7.66 (m, 1H, H-60);
8.04–7.84 (m, 3H, H-8, H-20, H-7); 8.76 (s, 1H, H-2);
9.01 (d, 1H, H-5). 13C NMR (CD3OD) d (ppm): 171.5
(CONH), 161.7, 157.9 (d, JF-C, 247 Hz), 150.8, 140.7,
136.3, 134.7, 130.7, 128.4, 126.4, 126.5, 117.9 (d, JF-C
,
22.5 Hz), 115.6, 113.5, 40.8 (CH2), 27.8 (CH2). IR
(KBr) (cmꢀ1): 1681.32 (vs); 1498.64 (s); 1446.54 (s);
1205.33 (vs). HPLC rt = 10.6 min using the conditions
described above.
4.2.6. N-{4-[(3-Chloro-4-fluorophenyl)amino]quinazoline-
6-yl}-3-iodopropionamide (7). To a solution of N-{4-
[(3-chloro-4-fluorophenyl) amino]quinazoline-6-yl}-3-
bromopropionamide (100 mg, 0.24 mmol) in dry
acetone (20 mL) was added sodium iodide (71 mg,
0.47 mmol). The mixture was stirred at 80 ꢂC for 11 h.
The solvent was evaporated and the residue was purified
by preparative reverse-phase HPLC according to the
conditions described above. The collected fractions were
analysed by analytical HPLC and evaporated. The resi-
due was washed several times with water and lyophilized
to give 7 as a white powder (46.8 mg, 32%).
4.2.3. 4-[(3-Chloro-4-fluorophenyl)amino]-6-nitroquinazo-
line (4). A solution of 3-chloro-4-fluoroaniline (482 mg,
3.32 mmol) in isopropanol (5 mL) was added to a sus-
pension of 4-chloro-6-nitroquinazoline (348 mg,
1.66 mmol) in the same solvent. Concentrated hydro-
chloric acid (5 lL) was added and the mixture refluxed
during 30 min. A yellow precipitate was obtained. After
cooling, a few drops of NH3 were added to basify the
solution and the precipitate was isolated by filtration,
washed with isopropanol and dried affording 4 as a yel-
1H NMR ((CD3)2SO) d (ppm): 3.09 (t, 2H, –CH2–CH2–
I); 3.47 (t, 2H, –CH2–CH2–I); 7.44 (t, 1H, H-50,
J = 9.1 Hz); 7.82 (m, 1H, H-60); 8.09 (m, 3H, arom);
8.57 (s, 1H, arom); 8.72 (s, 1H, arom); 10.08 (s, 1H,
–NH2); 10.40 (s, 1H, –NH). 1H NMR (CD3OD) d
(ppm): 3.13 (t, 2H, –CH2–CH2–Br); 3.52 (t, 2H,
–CH2–CH2–Br); 7.30 (t, 1H, H-50); 7.73–7.67 (m, 1H,
H-60); 8.01–7.78 (m, 3H, H-8, H-7, H-20); 8.76 (s, 1H,
H-2); 9.01 (d, 1H, H-5). Anal. (C17H13ClFN4OIÆCF3-
COOH) calcd: C, 39.03; H, 2.41; N, 9.58. Found: C,
39.25; H, 2.51; N, 9.80. IR (KBr) (cmꢀ1): 1678.7 (vs);
1498.4 (s); 1204.2(s). HPLC: rt = 12.6 min.
1
low solid (450 mg, 85%). H NMR [d, ppm, (CD3)2SO]:
10.6 (s, 1H, NH), 9.63 (d, 1H, H-5), 8.76 (s, 1H, H-2),
8.56 (dd, 1H, H-7), 8.15 (dd, 1H, H-8), 7.95 (d, 1H,
H-20), 7.85–7.80 (m, 1H, H-60), 7.49 (t, 1H, H-50).
4.2.4. 6-Amino-4-[(3-chloro-4-fluorophenyl)amino]quinazo-
line (5). Preparation of this compound followed a
slightly modified method described in the literature.8
The free base of 4-[(3-chloro-4-fluorophenyl)amino]-6-
nitroquinazoline (620 mg, 1.95 mmol) in THF:MeOH
(2:1, 80 mL) was hydrogenated over Raney nickel

C. Fernandes et al. / Bioorg. Med. Chem. 15 (2007) 3974–3980
3979
4.3. Radiochemistry
membrane. For visualization of molecular weight
bands, the membrane was immersed in Ponceau re-
agent (0.5% Ponceau in 1% acetic acid) for 5 min.
The membrane was washed in H2O, blocked overnight
in TTN with 5% milk (1% fat) and incubated for 2 h in
antiphosphotyrosine antibody diluted 1/2000 (PY20,
Santa Cruz Biotechnology). Then, the membrane was
washed 3 times with TTN and incubated in a horse-
radish peroxidase-conjugated secondary anti mouse
antibody diluted 1/3000 (Sigma) for 2 h and finally
washed 3 times in TTN. Detection was performed
using chemiluminescent detection system according to
manufacturer’s instructions (ECL kit, Amersham).
Band quantification has been performed using Band
leader ver. 3.00 programm.
4.3.1. [125I]-N-{4-[(3-chloro-4-fluorophenyl)amino]quinaz-
oline-6-yl}-3-iodopropionamide ([125I]-7). To a solution
of N-{4-[(3-chloro-4-fluorophenyl) amino]quinazoline-
6-yl}-3-bromopropionamide (200 lg, 0.47 lmol) in dry
acetone was added [125I]-NaI (25 MBq) and the mixture
was stirred for 2–2.5 h at 70 ꢂC. The radioiodinated qui-
nazoline [125I]-7 was then isolated by analytical reverse
phase HPLC, with simultaneous UV (254 nm) and
radioactivity detection. The radiochemical yield of
[
125I]-7 was 48%. HPLC: rt = 12.6 min.
4.4. In vitro assays
4.4.1. Biological evaluation. The human epidermoid
carcinoma A431 cell line, kindly provided by Dr. E.
Mishani (Hadassah Hebrew University, Israel), was
maintained in D-MEM (high glucose) supplemented
with 10% foetal calf serum and penicillin/streptomycin
100 UI/100 lg/mL, in 5% CO2 incubator at 37 ꢂC.
4.4.2. Lipophilicity. LogPo/w was determined by the
‘shakeflask’ method using the back-extraction tech-
nique.11 The radiolabelled compound was added to a
mixture of octanol (1 mL) and PBS, pH 7.4, (1 mL) pre-
viously saturated in each other by stirring the mixture.
The mixture was vortexed and centrifuged (3000 rpm,
10 min, rt) to allow for phase separation. Aliquots of
both octanol and PBS, pH 7.4, were counted for radio-
activity in a gamma counter. A 0.5 mL aliquot of the
octanol layer was withdrawn and added to an equal vol-
ume of PBS, pH 7.4, and the extraction and counting re-
peated as above. The logPo/w was calculated by dividing
the counts in the octanol by those in the buffer and the
4.4.1.1. Inhibition of cell proliferation by MTT assay.
The A431 cells (3 · 103 cells/well) were seeded in 96-
well plates and incubated for 24 h. Then the exponen-
tially growing cells were incubated with various con-
centrations of quinazoline analogues 5, 6 and 7,
respectively, (ranging from 1 to 50 lM in 4 replicates)
during 24, 48 and 72 h. Controls consisted of wells
without drugs. The medium was removed and the cells
were incubated for 4 h in the presence of 1 mg/mL
MTT in RPMI without phenol red at 37 ꢂC. The
MTT solution was removed and 100 lL of isopropa-
nol per well was added. After thorough mixing, absor-
bance of the wells was read in an ELISA reader at test
and at reference wavelengths of 540 and 620 nm,
respectively. The mean of the optical density of differ-
ent replicates of the same sample and the percentage
of each value were calculated (mean of the OD of var-
ious replicates/OD of the control). The percentage of
the optical density against drug concentration was
plotted in a semi-log chart and the IC50 from the
dose–response curve was determined.
results expressed as logPo/w
.
4.4.3. Human serum protein binding. The radioiodinated
quinazoline [125I]-7 (100 lL, ꢁ370 kBq) was incubated
at 37 ꢂC with human serum (1 mL). At appropriate peri-
ods of time (15 min, 1 h, 2 h, 4 h and 21 h) 100 lL ali-
quots (in duplicate) were sampled and treated with
200 lL of ethanol to precipitate the proteins.12 Samples
were then cooled at 4 ꢂC and centrifuged at 3000 rpm
for 15 min at 4 ꢂC. The supernatant was separated from
the precipitate, the sediment washed with ethanol
(2 · 1 mL) and counted in a gamma counter. The activ-
ity in the sediment was compared with the total activity
used and the percentage of [125I]-7 bound to proteins
was calculated. The supernatant was analysed by
HPLC, using the experimental conditions described
above.
4.4.1.2. Inhibition of autophosphorylation of receptor.
A431 cells (5 · 105) were seeded in 6-well plates and
grown to about 70% confluence, and then incubated
in fresh medium without foetal bovine serum for
24 h. The day after, cells were treated for 2 h with var-
ious concentrations of the compounds ranging from
0.1 to 5 lM. The cells were then stimulated for 5 min
with EGF (20 ng/mL). Then, the medium was re-
moved, cells were washed with PBS and cell lysis buffer
(50 mM Tris–HCl, pH 7.4, 150 mM NaCl, 1% Triton
X-100, 1 mM PMSF, 50 lg/mL aprotinin, 2 mM so-
dium orthovanadate, 50 lg/mL leupeptine and 5 mM
EDTA) was added. Controls consisted of wells without
drugs and with or without EGF. The total amount of
protein was estimated by a BCA assay in ELISA
plates, using a BSA standard curve. A volume of lysis
mixture, corresponding to 30 mg of total protein, were
loaded onto polyacrilamide gel (8%), proteins were
separated by electrophoresis and transferred to PVDF
4.4.4. Stability studies. The radiochemical stability of
[
125I]-7 was evaluated by HPLC analysis at several
time points after incubation in saline at 6 ꢂC and
37 ꢂC. Radiochemical stability in physiological med-
ium was also determined at 37 ꢂC at 15 min, 1 h,
2 h, 4 h and 21 h after incubation in human serum.
For this analysis the supernatants collected in the hu-
man serum protein binding assay were examined by
HPLC for radiochemical purity evaluation at the same
time intervals.
In order to assess the radiochemical stability in physio-
logical medium the supernatants collected in the human
serum protein binding assay were examined by HPLC
for radiochemical purity evaluation at the same time
intervals.

3980
C. Fernandes et al. / Bioorg. Med. Chem. 15 (2007) 3974–3980
3. Akita, R. W.; Sliwkowski, M. X. Semin. Oncol. 2003,
15.
4.5. In vivo stability studies
4. (a) Bonasera, T. A.; Ortu, G.; Rozen, Y.; Krais, R.;
Freedman, N. M.; Chisin, R.; Gazit, A.; Levitzki, A.;
Mishani, E. Nucl. Med. Biol. 2001, 28, 359; (b) Dissoki, S.;
Laky, D.; Mishani, E. J. Labelled Compd. Radiopharm.
2006, 49, 533.
5. (a) Johnstro¨m, P.; Fredriksson, A.; Thorell, J.-O.; Stone-
Elander, S. J. Labelled Compd. Radiopharm. 1998, 41,
623; (b) Ben-David, I.; Rozen, Y.; Ortu, G.; Mishani, E.
Appl. Radiat. Isot. 2003, 58, 209; (c) Mishani, E.;
Abourbeh, G.; Rozen, Y.; Jacobson, O.; Laky, D.; Ben,
D. I.; Levitzki, A.; Shaul, M. Nucl. Med. Biol. 2004, 31,
469; (d) Mishani, E.; Abourbeh, G.; Rozen, Y.; Jacob-
son, O.; Dissoki, S.; Daniel, R. B.; Rozen, Y.; Shaul,
M.; Levitzki, A. J. Med. Chem. 2005, 48, 5337; (e) Ortu,
G.; Ben-David, B.; Rozen, Y.; Freedman, N. M. T.;
Chisin, R.; Levitski, A.; Mishani, E. Int. J. Cancer 2002,
101, 360.
6. Shaul, M.; Abourbeh, G.; Jacobson, O.; Rozen, Y.; Laky,
D.; Levitzki, A.; Mishani, E. Bioorg. Med. Chem. 2004, 12,
3421.
7. Morley, J. S.; Simpson, J. C. E. J. Chem. Soc. 1948, 360.
8. Smaill, J. B.; Palmer, B. D.; Rewcastle, G. W.; Denny, W.
A.; McNamara, D. J.; Dobrusin, E. M.; Bridges, A. J.;
Zhou, H.; Showalter, H. D.; Winters, R. T.; Leopold, W.
R.; Fry, D. W.; Nelson, J. M.; Slintak, V.; Elliot, W. L.;
Roberts, B. J.; Vincent, P. W.; Patmore, S. J. J. Med.
Chem. 1999, 42, 1803.
In vivo studies were carried out in groups of 5 female
CD-1 mice (randomly bred, obtained from IFFA, CRE-
DO, Spain) weighing approximately 20–25 g according
the EU guidelines for Animal Care and Ethic for Ani-
mal Experiments. Animals were injected into the tail
vein with 175–200 kBq of the radioiodinated compound
in saline (100 lL) containing 0.1% Tween 20. Injected
dose (ID) was assumed to be the difference between
the measured radioactivity in the syringe before and
after injection. Animals were maintained on normal diet
ad libitum and were sacrificed by cervical dislocation at 1
and 4 h post-injection with the radiotracer. Tissue sam-
ples of the main organs were removed, weighed and the
activity measured in a gamma counter (Berthold
LB2111, Germany). Results were expressed as percent
of injected dose per total organ (% ID/organ) and pre-
sented as mean values SD. For blood, bone and mus-
cle, this value was calculated assuming that these organs
constitute 6%, 10% and 40% of the total weight, respec-
tively. Whole animal body radioactivity excretion was
not quantified. The in vivo stability was evaluated by
RP-HPLC analysis of urine samples collected at sacrifice
in the experimental conditions described above for the
radiochemical purity determination.
9. VanBrocklin, H. F.; Lim, J. K.; Coffing, S. L.; Hom, D.
L.; Negash, K.; Ono, M. Y.; Gilmore, J. L.; Bryant, I.;
Riese, D. J. J. Med. Chem. 2005, 48, 7445.
References and notes
10. Adam, M. J.; Wilbur, D. S. Chem. Soc. Rev. 2004, 34,
153.
1. (a) Arteaga, C. Semin. Oncol. 2003, 30, 3; (b) Sridhar, S.
S.; Seymour, L.; Shepherd, F. A. Lancet Oncol. 2003, 4,
397; (c) Laskin, J. J.; Sandler, A. B. Cancer Treat. Rev.
2004, 30, 1; (d) Ritter, C. A.; Arteaga, C. Semin. Oncol.
2003, 3.
11. Troutner, D. E.; Volkert, W. A.; Hoffman, T. J.; Holmes,
R. A. Int. J. Appl. Radiat. Isot. 1984, 35, 467.
12. Eisenwiener, K.-P.; Prata, M. I. M.; Buschmann, I.;
Zhang, H.-W.; Santos, A. C.; Wenger, S.; Reubi, J. C.;
Macke, H. R. Bioconjugate Chem. 2002, 13, 530.
2. Ranson, M. The Oncologist 2002, 7, 16.