Exploring Epidermal Growth Factor Receptor (EGFR) inhibitor features: The role of fused dioxygenated rings on the quinazoline scaffold

Article abstract of DOI:10.1021/jm901338g

A number of dioxolane, dioxane, and dioxepine quinazoline derivatives have been synthetized, and evaluated as EGFR inhibitors. Their cytotoxic activity has been tested against two cell, lines overexpressing and not expressing EGFR. Most derivatives were a

Full text of DOI:10.1021/jm901338g

1862 J. Med. Chem. 2010, 53, 1862–1866
DOI: 10.1021/jm901338g
Exploring Epidermal Growth Factor Receptor (EGFR) Inhibitor Features: The Role of Fused
Dioxygenated Rings on the Quinazoline Scaffold
Adriana Chilin,*,† Maria Teresa Conconi,† Giovanni Marzaro, Adriano Guiotto, Luca Urbani, Francesca Tonus, and
Pierpaolo Parnigotto
Dipartimento di Scienze Farmaceutiche, Universitaꢀ di Padova, via Marzolo 5, 35131 Padova, Italy.
†First two authors contributed equally to this work and should be considered co-first authors.
Received September 9, 2009
A number of dioxolane, dioxane, and dioxepine quinazoline derivatives have been synthetized and
evaluated as EGFR inhibitors. Their cytotoxic activity has been tested against two cell lines over-
expressing and not expressing EGFR. Most derivatives were able to counteract EGF-induced EGFR
phosphorylation, and their potency was comparable to the reference compound PD153035. The size of
the fused dioxygenated ring was crucial for the biological activity, the dioxane derivatives being the most
promising class of this series.
Introduction
determinants that could positively modulate the interaction
with EGFR and consequently the biological activity, we have
modified the molecular structure of PD153035 in two specific
sites: the moiety involved in the interaction with the ATP-
binding site of EGFR, incorporating both the 6- and 7-
methoxy groups into rings of various size (dioxolane, dioxane,
and dioxepine rings), and the 4-aniline moiety responsible for
the interaction with an hydrophobic pocket of EGFR, sub-
stituting the bromine atom with other functional groups of
different lipophilicity and/or steric hindrance (Table 1).
The antiproliferative activity of the synthesized compounds
has been evaluated against A431 and NIH3T3 cells over-
expressing and not expressing EGFR, respectively. Moreover,
the inhibition of ligand-induced receptor phosphorylation
and the abrogation of EGF-mediated growth effect and
reversibility were all assessed on A431 cells.
The epidermal growth factor receptor (EGFRa) is a trans-
membrane tyrosine kinase receptor that belongs to the ErbB
family. In the absence of a ligand EGFR exists in an inactive
monomeric form, whereas after interaction with its growth
factor ligands, the formation of homo- or heterodimers
occurs,1 leading to signal transduction pathways that regulate
cell differentiation, growth, migration, and apoptosis.2 EGFR
is often mutated and/or overexpressed in many tumors,
representing therefore a valuable target for the design of
anticancer agents. A number of ATP-mimetic tyrosine kinase
inhibitors (TKIs) have so far received approval by FDA for
cancer therapy3 as Erlotinib,4 Gefitinib,5 and Lapatinib.6
Among several TKIs, the 4-anilinoquinazoline class of com-
pounds plays a key role and has been extensively studied in the
past few years. These TKIs generally show high potency in
vitro, inhibiting EGFR at nanomolar concentration.7 How-
ever, a major problem of these compounds is their poor in vivo
activity, and many analogues of PD153035,7 the most potent
inhibitor on isolated EGFR (IC50 = 25 pM), have been deve-
loped by inserting solubilizing groups to address this issue.8
Moreover, it has been established that TKIs can bind their
target both in the active and in the inactive forms, leading to
different activity profiles.9 For example, Lapatinib, which
binds only the inactive EGFR form, is a slow off-rate inhibitor,
leading to a longer inhibition effect with respect to Erlotinib or
Gefitinib, which bind only the active EGFR form.6 This is due
to Lapatinib ability to occupy a deep pocket only present in the
inactive form of the receptor. To investigate the structural
Chemistry
All the designed compounds were synthesized according to
a novel synthetic strategy, with a quinazoline oxidation step
leading to the key quinazolinone intermediates10 (Scheme 1).
The starting quinazolines 3a-c have been synthesized from
the appropriate aniline derivative 1a-c using the HMTA/
TFA/K3Fe(CN)6 method.11 Compounds 3a-c were oxidized
to quinazolin-4(3H)-ones with CAN in aq AcOH and then
submitted to chlorination with POCl3 in presence of TEA and
finally condensed with the suitable substituted aniline to give
the final products 6a-c, 7a-c, 8a-c, 9a-c, and 10a-c,
directly as hydrochlorides. The final compounds are summar-
ized in Table 1.
*To whom correspondence should be addressed. Phone: þ39
0498275349. Fax: þ39 0498275366. E-mail: adriana.chilin@unipd.it.
a Abbreviations: AP, alkaline phosphatase; CAN, cerium ammonium
nitrate; EGF, epidermal growth factor; EGFR, epidermal growth factor
receptor; ELISA, enzyme-linked immunoassorbent assay; FDA, Food
and Drug Administration; HMTA, hexamethylenetetramine; HRMS,
high resolution mass spectrometry; HRP, horseradish peroxidase;
MAbs: monoclonal antibodies; MOE, Molecular Operating Environ-
ment; PDB, Protein Data Bank; SD, standard deviation; TEA, triethyl-
amine; TFA, trifluoroacetic acid; THF, tetrahydrofuran; TKI, tyrosin
kinase inhibitors; VEGFR, vascular endothelial growth factor receptor.
Biology
The inhibitory effects on cell proliferation were determined
by MTT assay (Table 2 and Figure S1 in the Supporting
Information). The results obtained on A431 cells (over-
expressing EGFR) showed that the dioxane derivatives were
more cytotoxic than the corresponding dioxolane and dioxe-
pinederivatives. IC50 values of compounds 6b, 7b, and 8b were
about 6 times lower than that determined for PD153035.
r
pubs.acs.org/jmc
Published on Web 01/22/2010
2010 American Chemical Society

Brief Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4 1863
Scheme 1a
a Reagents and conditions: (a) ClCOOEt, TEA, THF, rt, 30 min; (b) (i) HMTA, TFA, MW, 110 °C, 10 min, (ii) K3Fe(CN)6, KOH 10% EtOH/H2O 1/1,
reflux, 4 h; (c) CAN, AcOH, H2O, rt, 5 min; (d) POCl3, TEA, reflux, 3 h; (e) ArNH2, iPrOH, MW, 80 °C, 15-30 min. See Table 1 for R specification.
Table 1. Structures of Synthesized Compounds
Table 2. IC50 Values for 72 h Exposure of A431 and NIH3T3 Cells to
the Compounds
IC50 (μM)a
compd
A431
NIH3T3
PD153035
6a
7a
4.40 ( 1.3
7.16 ( 1.11
6.69 ( 0.92
8.46 ( 0.87
>1 0 b
>10b
>10b
>10b
8a
9a
>10b
ndc
>1 0 b
ndc
n = 1
n = 2
n = 3
R
10a
6b
6aa
7aa
8a
6ba
7b
8ba
9ba
10b
6ca
7c
30-Br
0.75 ( 0.09
0.67 ( 0.11
0.77 ( 0.09
>1 0 b
>10b
30-Me
7b
8b
5.80 ( 0.58
7.10 ( 0.60
ndc
8c
9c
30-CF3
30-Cl, 40-F
20-F, 40-Br
9b
9a
10a
10b
6c
6.28 ( 0.46
8.17 ( 0.76
6.29 ( 0.62
4.73 ( 0.56
7.13 ( 0.63
>1 0 b
5.40 ( 0.20
10c
>10b
a These compounds have already been synthesized (see references in
the Supporting Information) but using a different strategy, and only
data about cytotoxicity and inhibition of isolated EGFR have been
published.
7c
8c
>10b
>10b
9c
10c
>10b
ndc
a The values are the mean(SD of at least three independent experi-
ments. b IC50 not determined because less than 50% inhibition was
observed at the highest tested concentration (10 μM). Higher concentra-
tions were not used to avoid precipitation of the compounds in the
culture medium. c nd: not determined. Cell proliferation assay was not
carried out on NIH3T3 cells because the compound was quite ineffective
on A431 cells.
Because the 4-aniline moiety interacts with a hydrophobic
pocket of EGFR,12 the presence of lipophilic groups in speci-
fic positions improves the interaction with this site, leading to
an increase in the affinity toward EGFR. Thus, the com-
pounds bearing a bromine atom or methyl (compounds 6a-c
and 7a-c) or trifluoromethyl (compounds 8a-c) groups at
the 30 position are more active than those substituted with two
halogen atoms.
To assess whether the antiproliferative activity of the com-
pounds could be only due to inhibition of EGFR, the same
test was repeated with NIH3T3 cells, not expressing EGFR
(Table 2). In general, almost all the compounds, including
PD153035, showed a lower but not negligible cytotoxicity in
NIH3T3 cells, demostrating that other cell targets, probably
other kinases, may be affected by these compounds (parti-
cularly in the case of dioxane derivatives). The interaction
with other tyrosine kinase receptors, such as the vascular
endothelial growth factor receptor (VEGFR), could improve
the antitumor activity via an antiangiogenetic mechanism.13 It
is known that the pharmacological inhibition of EGFR can
decrease VEGF expression and consequently angiogenesis in
many tumor types.14 Nevertheless, VEGF production is con-
trolled by many other factors and the prolonged administra-
tion of EGFR inhibitors can lead to resistance. Thus, several
inhibitors of both EGFR and VEGFR have been designed
and studied.15 It is noteworthy that compounds 10a-c bear
two halogen atoms (F and Br) in the 4-aniline moiety like
Vandetanib, a dual inhibitor able to slow down the growth of
tumors nonresponsive to EGFR inhibitors.16 Among these
three compounds, compound 10b was the most effective,
inducing similar antiproliferative effects in both cell lines.
We are planning in vitro studies on endothelial cells to assess if
this compound could interact also with VEGFR and exert
antiangiogenic effects.
On the basis of the inhibitory effect on A431 cell prolifera-
tion, the more active compounds were selected for further
studies. The antiproliferative effects of 4-anilinoquinazoline
derivatives seems to be related to a pro-apoptotic activity. All
the compounds, as well as PD153035, induced strong in-
creases in the apoptotic rate when compared to nontreated
cultures (Figure 1).
The reversibility of the antiproliferative effects of the
compounds was also studied. Almost all the compounds
seems to be cytostatic, like PD153035,17 showing on A431
cells a reversible growth inhibitory activity (see Figure S2 in
the Supporting Information). Only compound 6a partially

1864 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4
Chilin et al.
Figure 1. Pro-apoptotic effects of compounds on A431 cells. A431 cells were treated for 24 h with compounds (10 μM). At the end of the
incubation period, both cell viability (A) and apoptotic rate (B) were determined by MTT assay and assessment of mono- and oligo-
nucleosomes formation, respectively. Results, expressed as percent change from control nontreated cultures (taken as 0), are the mean ( SD of
at least three independent experiments. p<0.05 vs control cultures, Student t test.
form because only preincubation reversed the EGF-induced
growth effects.
A docking study was performed on the two structurally
related compounds 8a and 8b to explain the reason of such a
different behavior for so similar derivatives. The structures of
8a and 8b were docked in the crystal structure of inactive
EGFR form (PDB ID: 2GS7): in this structure, a magnesium
ion, a well-known cofactor for the ATPase activity,18 was
present near the interaction pocket. Only compound 8b was
able to interact with the ion through the oxygen at the 6
position (Figure 4 and Figure S3 in the Supporting In-
formation).
Figure 2. Effects on EGFR phosphorylation. A431 cells were
pretreated for 30 min with 1 μM of compounds and then incubated
Conclusion
The syntheses and biological evaluation of three classes of
fused tricyclic quinazolines as EGFR inhibitors have been
described. The cyctotoxic activity of all the compounds was
assessed against two cell lines overexpressing and not expressing
EGFR, respectively. Most derivatives were able to counteract
EGF-induced EGFR phosphorylation and showed better or at
least comparable potency with respect to PD153035, used as
reference compound. The size of the fused dioxygenated ring
was crucial for the cytotoxic activity and for the biological
profile. In particular, the dioxane derivatives showed an inter-
esting profile: among these, the most promising derivative is the
one bearing the 30-trifluoromethylaniline substituent (8b) due to
its preferential binding to the inactive form of EGFR. Further
studies on the VEGFR inhibition ability of this compound are
in progress to investigate its potential dual behavior.
with 50 ng/mL EGF for 5 min. Phosphorylated EGFR was deter-
mined by a sandwich ELISA using a phospho-specific anti-EGFR
antibody and a normalization antibody that recognizes the pan-
protein regardless of phosphorylation status. The results were
expressed as ratio between phosphorylated EGFR and total EGFR.
Δ = p < 0.05 vs cultures treated with EGF.
maintained its antiproliferative activity after the recovery
period.
The ability of the compounds to inhibit EGFR phospho-
rylation was assessed using a cell-based ELISA experiment
able to discriminate EGFR from other tyrosine kinases. As
shown in Figure 2, all the compounds significantly decreased
EGF-induced EGFR phosphorylation.
To evaluate whether the synthesized compounds could be
able to counteract the increase in cell growth induced by
exogenous EGF, A431 cells were pretreated for 24 h with
4-anilinoquinazoline derivatives and then, after removal of
media, incubated for 72 h with either EGF alone (Figure 3A)
or both compounds and EGF added to culture media
(Figure 3B). On the basis of MTT data, the experiments were
carried out by treating the cultures with the highest noncyto-
toxic concentrations of the tested compounds, thus avoiding
direct cytotoxic effects. Although all compounds inhibited
EGFR phosphorylation induced by exogenous EGF, only
PD153035, 8a, 8b, and 10b were able to counteract the
proliferogenic effects of the cytokine. In both experimental
conditions, PD153035 and compound 8a reversed the EGF-
induced cell growth. Compound 10b was able to counteract
the EGF stimulatory effects when it was simultaneously ad-
ministered with the growth factor, suggesting a preferential
interaction of 10b with the active form of EGFR. On the
contrary, compound 8b seems to be effective on the inactive
Experimental Section
Chemistry. See Supporting Information for general synthetic
methods, for synthesis of 1c, and for the analytical details (mp,
NMR, HRMS, elemental analyses) of all compounds.
General Procedure for Carbamates 2a-c. A mixture of 1
(10.0 mmol), ethyl chloroformate (20.0 mmol), and TEA (20.0 mmol)
in anhydrous THF (200 mL) was stirred at room temperature
for 30 min. The solid was filtered off, and the solvent was
evaporated under reduced pressure. The residue was dissolved
in CHCl3 (200 mL), and the organic layer was washed with water
(2 ꢀ 100 mL). The organic phase was evaporated under reduced
pressure to give 2 (yields 97-98%).
General Procedure for Quinazolines 3a-c. A mixture of ethyl
carbamate 2 (10.0 mmol) and HMTA (10.0 mmol) in TFA
(30 mL) was microwave irradiated at 110 °C (power set point 80 W;
ramp time 1 min; hold time 10 min). After cooling, the mix-
ture was diluted with aq ethanolic (water/EtOH: 1/1) 10% KOH

Brief Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4 1865
Figure 3. Effects on EGF-mediated growth stimulation. (A) After 24 h of treatment with the compounds at the highest concentration not
inducing decrease in cell viability (0.01 μM PD153035, 6b, 7b, 10b, 6c, and 8c; 0.1 μM 6a, 7a, 8b, 7c, and 9c; 1 μM 8a), medium from A431 cells
was removed and replaced with fresh one containing 5 ng/mL EGF. Then, cultures were incubated for 72 h. (B) A431 cells were treated for 72 h
with 5 ng/mL EGF and noncytotoxic concentrations of compounds. At the end of the incubation period, cell proliferation was determined by
MTT assay. Results, expressed as percent change from control not treated cultures (taken as 0), are the mean ( SD of at least three independent
experiments. * = p < 0.05 vs EGF-treated cultures, Student t test.
antiproliferative activity, EGFR-phosphorylation, and effects
on EGF-induced cell growth.
Acknowledgment. The present work has been carried out
with the financial support of the University of Padova “Pro-
getto di Ricerca di Ateneo 2008”.
Supporting Information Available: Detailed characterization
data for all the compounds, experimental procedures for bio-
chemical assays, and computational methodologies. This ma-
terial is available free of charge via the Internet at http://pubs.
acs.org.
References
(1) Bazley, L. A.; Gullick, W. J. The epidermal growth factor receptor
family. Endocr. Relat. Cancer 2005, 12 (Suppl 1), S17–S27.
(2) Citri, A.; Yarden, Y. EGF-ERBB signalling: towards the systems
level. Nat. Rev. Mol. Cell Biol. 2006, 7, 505–516.
(3) Ellis, P. M.; Morzycki, W.; Melosky, B.; Butts, C.; Hirsh, V.;
Krasnoshtein, F.; Murray, N.; Shepherd, F. A.; Soulieres, D.;
Tsao, M. S.; Goss, G. The role of the epidermal growth factor
receptor tyrosine kinase inhibitors as therapy for advanced, meta-
static, and recurrent non-small-cell lung cancer: a Canadian na-
Figure 4. Superposition of docked structure of compounds 8a and
8b in the crystal structure of inactive EGFR (PDB ID: 2GS7).
tional consensus statement. Curr. Oncol. 2009, 16, 27–48.
(300 mL), and K3Fe(CN)6 (80.0 mmol) was added to the
(4) Dowell, J.; Minna, J. D.; Kirkpatrick, P. Erlotinib hydrochloride.
solution, which was refluxed under stirring for 4 h. After cool-
ing, the mixture was diluted with water (300 mL) and extracted
with CHCl3 (3 ꢀ 200 mL). The organic phase was evaporated
under reduced pressure to give 3 (yields 83-89%).
General Procedure for Quinazolinones 4a-c. To a solution of
3 (5.0 mmol) in AcOH (2.0 mL), a solution of CAN (20.0 mmol)
in water (24.0 mL) was added dropwise. The white precipitate
formed was collected to give 4 (yields 65-73%).
General Procedure for Chloroquinazolines 5a-c. A suspension
of 4 (4.0 mmol) in POCl3 (8.0 mL) and TEA (2.0 mL) was
refluxed for 3 h. After cooling, the mixture was concentrated
under reduced pressure and the solid residue was dissolved in
EtOAc (50 mL) and washed with a satd aq NaHCO3 solution
(2 ꢀ 20 mL). The organic phase was evaporated under reduced
pressure to give 5 (yields 70-93%).
Nat. Rev. Drug Discovery 2005, 4, 13–14.
(5) Cohen, M. H.; Williams, G. A.; Sridhara, R.; Chen, G.; Pazdur, R.
FDA drug approval summary: gefitinib (ZD1839) (Iressa) tablets.
Oncologist 2003, 8, 303–306.
(6) 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. A unique structure for epidermal growth factor
receptor bound to GW572016 (Lapatinib): relationships among
protein conformation, inhibitor off-rate, and receptor activity in
tumor cells. Cancer Res. 2004, 64, 6652–6659.
(7) Bridges, A. J.; Zhou, H.; Cody, D. R.; Rewcastle, G. W.; Mc-
Michael, A.; Showalter, H. D.; Fry, D. W.; Kraker, A. J.; Denny,
W. A. Tyrosine kinase inhibitors. 8. An unusually steep structure-
activity relationship for analogues of 4-(3-bromoanilino)-6,7-
dimethoxyquinazoline (PD 153035), a potent inhibitor of the
epidermal growth factor receptor. J. Med. Chem. 1996, 39, 267–
276.
(8) Liu, G.; Hu, D. Y.; Jin, L. H.; Song, B. A.; Yang, S.; Liu, P. S.;
Bhadury, P. S.; Ma, Y.; Luo, H.; Zhou, X. Synthesis and bioactivi-
ties of 6,7,8-trimethoxy-N-aryl-4-aminoquinazoline derivatives.
Bioorg. Med. Chem. 2007, 15, 6608–6617.
General Procedure for Anilinoquinazolines 6-10a-c. A mix-
ture of 5 (1.0 mmol) and substituted aniline (1.0 mmol) in iPrOH
(2 mL) was microwave irradiated at 80 °C (power set point 60 W;
ramp time 1 min; hold time 15 min). After cooling, further
aniline (1.0 mmol) was added and the mixture was microwave
irradiated at 80 °C for another cycle. After cooling, the obtained
precipitate was collected by filtration to give 6-10a-c as
hydrochlorides (yields 72-86%).
(9) Liao, J. J. Molecular recognition of protein kinase binding pockets
for design of potent and selective kinase inhibitors. J. Med. Chem.
2007, 50, 409–424.
(10) Marzaro, G.; Guiotto, A.; Pastorini, G.; Chilin, A. A novel
approach to quinazolin-4(3H)-one via quinazoline oxidation: an
improved synthesis of 4-anilinoquinazolines. Tetrahedron 2010,
66, 962–968.
Biology. See Supporting Information for evaluation of
antiproliferative and pro-apoptotic activity, reversibility of the

1866 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4
Chilin et al.
(11) Chilin, A.; Marzaro, G.; Zanatta, S.; Guiotto, A. A microwave
improvement in the synthesis of the quinazoline scaffold. Tetra-
hedron Lett. 2007, 48, 3229–3231.
(12) Roskoski, R. The ErbB/HER receptor protein-tyrosine kinases
and cancer. Biochem. Biophys. Res. Commun. 2004, 319, 1–11.
(13) Cao, Y. H. Antiangiogenic cancer therapy. Semin. Cancer Biol.
2004, 14, 139–145.
(14) Dutta, P. R.; Malty, A. Cellular responses to EGFR inhibitors and
their relevance to cancer therapy. Cancer Lett. 2007, 254, 165–177.
(15) Tabernero, J. The role of VEGF and EGFR inhibition: Implica-
tions for combining anti-VEGF and anti-EGFR agents. Mol.
Cancer Res. 2007, 5, 203–220.
Janne, P.; Kobayashi, S.; Halmos, B.; Tenen, D.; Tang, X. M.;
Engelman, J.; Yeap, B.; Folkman, J.; Johnson, B. E.; Heymach,
J. V. Combined vascular endothelial growth factor receptor and
epidermal growth factor receptor (EGFR) blockade inhibits tumor
growth in xenograft models of EGFR inhibitor resistance. Clin.
Cancer Res. 2009, 15, 3484–3494.
(17) Bos, M.; Mendelsohn, J.; Kim, Y. M.; Albanell, J.; Fry, D. W.;
Baselga, J. PD153035, a tyrosine kinase inhibitor, prevents epider-
mal growth factor receptor activation and inhibits growth of cancer
cells in a receptor number-dependent manner. Clin. Cancer Res.
1997, 3, 2099–2106.
(18) Saylor, P.; Wang, C.; Hirai, T. J.; Adams, J. A. A second magne-
sium ion is critical for ATP binding in the kinase domain of the
oncoprotein v-Fps. Biochemistry 1998, 37, 12624–12630.
(16) Naumov, G. N.; Nilsson, M. B.; Cascone, T.; Briggs, A.; Straume,
O.; Akslen, L. A.; Lifshits, E.; Byers, L. A.; Xu, L.; Wu, H. K.;