Discovery of anilino-furo[2,3-d]pyrimidine derivatives as dual inhibitors of EGFR/HER2 tyrosine kinase and their anticancer activity

Article abstract of DOI:10.1016/j.ejmech.2017.12.022

Being responsible for the development of many cancer types, EGFR (Epidermal Growth Factor Receptor) and HER2 (Human Epidermal growth factor Receptor 2) were the focus of this study where a series of novel 4-anilino-furo[2,3-d]pyrimidine derivatives was designed, synthesized and biologically evaluated. Modification of the solvent accessible 5-position side chain greatly affected the in-vitro EGFR/HER2 inhibitory activity. Three derivatives bearing 5-carboxylic acid side chain, namely the 3-chloroanilino derivative (8c), the 3-bromoaniline (8d) and the lapatinib analogue (10) demonstrated the most significant submicromolar EGFR inhibition. Surprisingly, the in-vitro assay of the ester 7h and its acid analogue 10 showed a significant variation of results between the antiproliferative activity against A549 cell line (IC₅₀ 0.5 and 21.4 μM) respectively and EGFR inhibitory activity (18% and 100%) respectively, suggesting that 7h might be a prodrug for 10. This assumption was also affirmed by the in-vivo results, where the in-vivo antitumor assessment against EAC (Ehrlich Ascites Carcinoma) solid tumor model revealed that 7h and 8d (10 mg/kg dose) exhibited antitumor activity comparable to that of gefitinib at the same dose, exhibiting TGI% of 67%, 71% and 70%, respectively. This effect could be explained, at least partly, via activation of apoptosis, where 7h and 8d caused more than 2-fold increase of caspase 3 and cytochrome c expression than the control group which is comparable to that of gefitinib-treated group. Finally, 7h was the most effective apoptotic inducer, resulting in a significant elevation in annexin V–FITC-positive apoptotic cells (both early and late apoptosis) by 25 and 79-folds, respectively, compared to control, which is higher than that of gefitinib (22 and 61-folds, respectively).

Full text of DOI:10.1016/j.ejmech.2017.12.022

Accepted Manuscript
Discovery of anilino-furo[2,3-d]pyrimidine derivatives as dual inhibitors of EGFR/
HER2 tyrosine kinase and their anticancer activity
Monia Hossam, Deena S. Lasheen, Nasser S.M. Ismail, Ahmed Esmat, Ahmed M.
Mansour, Abdel Nasser B. Singab, Khaled A.M. Abouzid
PII:
S0223-5234(17)31028-0
10.1016/j.ejmech.2017.12.022
EJMECH 10000
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 26 September 2017
Revised Date: 5 December 2017
Accepted Date: 6 December 2017
Please cite this article as: M. Hossam, D.S. Lasheen, N.S.M. Ismail, A. Esmat, A.M. Mansour, A.N.B.
Singab, K.A.M. Abouzid, Discovery of anilino-furo[2,3-d]pyrimidine derivatives as dual inhibitors of
EGFR/HER2 tyrosine kinase and their anticancer activity, European Journal of Medicinal Chemistry
(2018), doi: 10.1016/j.ejmech.2017.12.022.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Discovery of anilino-furo[2,3-d]pyrimidine derivatives as dual inhibitors of
EGFR/HER2 tyrosine kinase and their anticancer activity
a*
a,e
a
b,e
c
Monia Hossam , Deena S. Lasheen , Nasser S.M. Ismail , Ahmed Esmat , Ahmed M. Mansour ,
d,e
a,e**
Abdel Nasser B. Singab and Khaled A.M. Abouzid
Pharmaceutical Chemistry Department, Faculty of Pharmacy, Ain Shams University, Abassia, Cairo
a
1
1566, Egypt
b
Department of Pharmacology and Toxicology, Faculty of Pharmacy, Ain Shams University, Abassia,
Cairo 11566, Egypt
c
Department of Pharmacology and Toxicology, Faculty of Pharmacy, Al-Azhar University, Nasr City,
Cairo, Egypt
d
Pharmacognosy Department, Faculty of Pharmacy, Ain Shams University, Abassia, Cairo 11566, Egypt
e
Center for Drug Discovery and Development Research, Faculty of Pharmacy, Ain Shams University,
Abassia, Cairo 11566, Egypt
*
*
Corresponding author: monia_hh@pharma.asu.edu.eg
*Corresponding author: khaled.abouzid@pharma.asu.edu.eg
Being responsible for the development of many cancer types, EGFR (Epidermal Growth
Factor Receptor) and HER2 (Human Epidermal growth factor Receptor 2) were the focus of this
study where a series of novel 4-anilino-furo[2,3-d]pyrimidine derivatives was designed,
synthesized and biologically evaluated. Modification of the solvent accessible 5-position side
chain greatly affected the in-vitro EGFR/HER2 inhibitory activity. Three derivatives bearing 5-
carboxylic acid side chain, namely the 3-chloroanilino derivative (8c), the 3-bromoaniline (8d)
and the lapatinib analogue (10) demonstrated the most significant submicromolar EGFR
inhibition. Surprisingly, the in-vitro assay of the ester 7h and its acid analogue 10 showed a
significant variation of results between the antiproliferative activity against A549 cell line (IC50
0
.5 and 21.4 µM) respectively and EGFR inhibitory activity (18% and 100%) respectively,
suggesting that 7h might be a prodrug for 10. This assumption was also affirmed by the in-vivo
results, where the in-vivo antitumor assessment against EAC (Ehrlich Ascites Carcinoma) solid
tumor model revealed that 7h and 8d (10 mg/kg dose) exhibited antitumor activity comparable to
that of gefitinib at the same dose, exhibiting TGI% of 67%, 71% and 70%, respectively. This
effect could be explained, at least partly, via activation of apoptosis, where 7h and 8d caused
more than 2-fold increase of caspase 3 and cytochrome c expression than the control group
Abbreviations
ADME: Absorption, Distribution, Metabolism and excretion
EAC: Ehrlich Ascites Carcinoma
EGFR: Epidermal Growth Factor Receptor
FITC: Fluorescein Isothiocyanate
HER2: Human Epidermal growth factor Receptor 2
NSCLC: Non-Small Cell Lung Cancer

ACCEPTED MANUSCRIPT
which is comparable to that of gefitinib-treated group. Finally, 7h was the most effective
apoptotic inducer, resulting in a significant elevation in annexin V–FITC-positive apoptotic cells
(both early and late apoptosis) by 25 and 79-folds, respectively, compared to control, which is
higher than that of gefitinib (22 and 61-folds, respectively).
Keywords: Furopyrimidine; EGFR/HER2 kinase; In-vivo activity; Flow cytometry; Docking
study
1
. Introduction
The HER family represents a cornerstone in normal physiology and in cancer [1,2]
because of its essential role in intracellular signaling and cell transformation [3]. Deregulation of
HER family signaling results in proliferation, invasion, metastasis, angiogenesis, and tumor cell
survival in addition to being correlated to many cancer types; lung, head and neck, and breast
cancer [4,5]. Among this family, two members are particularly involved in cancer development;
EGFR and HER2. Inappropriate activation of EGFR and HER2 in cancer can occur via different
mechanisms, including overexpression of ligands or receptors, mutations or amplification of the
receptor, or via transactivation by other receptors [6]. Overexpression and/or mutation can lead
to ligand-independent activation of EGFR and HER2, as well as to increased activation fol-
lowing engagement with the ligand [7]. EGFR overexpression is notable in many types of human
cancers and is responsible for poor prognosis [8–10]. In particular, overexpression of EGFR has
been found in 40-80% of NSCLC cases [11]. NSCLC accounts for approximately 85% of
primary malignant lung tumors and it remains the leading cause of cancer-related death
worldwide [12]. Meanwhile, HER2 has been identified as a prognostic and predictive factor in
breast cancer (18–25%) [13]. Evidence is growing that HER2 also contributes to tumorigenesis
in gastric and gastroesophageal junction cancer, with studies suggesting that HER2
amplification or overexpression is quite abundant in these tumor types [14–17]. Therefore, the
HER family represents an attractive class of rational targets for anticancer drug development.
2
. Rationale and Design
Different scaffolds have been investigated for their EGFR/HER2 inhibitory activity for
example 4-anilinoquinazolines, 3-cyano-4-anilinoquinolines, pyrimidines, fused pyrimidines and
pyrrolotriazines [18]. In our study, the furo[2,3-d]pyrimidine scaffold was chosen exclusively as

ACCEPTED MANUSCRIPT
it exhibited potent VEGFR-2 inhibitory activity with IC50 values in the nanomolar range in the
work previously accomplished by our team [19]. Thus, we took the chance to complete our work
and explore its EGFR/HER2 inhibitory activity, since it wasn’t explored extensively compared to
the other scaffolds [20]. Although most of the previously synthesized furopyrimidine based
inhibitors bear bulky aromatic groups at position 5 [21], our molecules bear relatively small
aliphatic side chains in order to tackle the solvent accessible region of the enzyme, starting with
a carboxylate ester in series 6 and series 7 then its further optimization to carboxylic acid in
series 8 and compound 10 or carboxamides in series 9 and compound 11 or hydroxymethyl in
series 12. (Figure 1)

ACCEPTED MANUSCRIPT
Figure 1. Design of inhibitors. The solvent accessible region, the hinge region, the hydrophobic pocket and
allosteric pocket are color coded green, red, blue and purple respectively.
3. Results and Discussion
3
.1 Chemistry
The designed furopyrimidine based inhibitors were synthesized as outlined in schemes (1-
4
). Beginning with Williamson ether synthesis to prepare the nitro derivatives 1a-i by heating p-
nitrophenol or its 2-chloroderivative with the appropriate benzyl chloride derivative in DMF in
presence of K CO [22] Further reduction was achieved by heating the nitro derivative (1a-i)
2
3
.
with iron and NH Cl in 70% ethanol to yield the corresponding 4-(benzyloxy)aniline (2a-i) [23]
4
(
Scheme 1). Stirring malononitrile with ethyl 2-chloroacetoacetate in sodium ethoxide afforded
the precursor 3 which was cyclized by reflux in a mixture of formic acid and acetic anhydride to
give 4. Refluxing 4 in neat POCl yielded the chlorinated derivative 5 [19], which is a key
3
intermediate for the synthesis of the final compounds 6a-i and 7a-i. Nucleophilic substitution of
the chloro derivative 5 with simple anilines provided 6a-i while substitution using the previously
prepared 4-(benzyloxy)aniline derivatives (2a-i) provided 7a-i [24] (Scheme 2). The carboxylic
acid derivatives 8a-e were prepared by the hydrolysis of the corresponding esters 6a-e using
2
LiOH.H O in 50% ethanol [25]. Synthesis of amide derivatives was achieved via two methods.
Refluxing the carboxylic acid derivatives (8a-e) in neat SOCl provided the corresponding active
2
acyl chloride which was stirred with the appropriate amine in methylene chloride to give the
amide derivatives 9a-g (Method 1) [26,27]. Coupling of the carboxylic acid derivative8e) with
the appropriate amine in presence of EDC.HCl/ DMAP in DMF yielded the amide derivatives 9h
and 9i [28] (Method 2)(Scheme 3a). By applying method 1, 11 was prepared via hydrolysis of the
ester 7h to give the acid derivative 10 [25], which was further activated [26] and stirred with
hydroxylamine in methylene chloride [29] (Scheme 3b). The alcohol derivatives 12a-c were
prepared by reduction of their corresponding esters 6c-e using LiAlH in THF [30] (Scheme 4).
4

ACCEPTED MANUSCRIPT
Scheme 1, Reagents & Conditions: (a) K CO , DMF, 80°c, 24 hours, 77-99%, (b) Fe, NH Cl, 70 % ethanol, reflux
2
3
4
1
hr, 54-99%.

ACCEPTED MANUSCRIPT
Scheme 2, Reagents and conditions:(a)NaOEt, rt, 6 hours, 77%, (b) HCOOH, acetic anhydride, reflux, 48 hours,
9
1
0%, (c) POCl , reflux, 3 hours, 87%, (d) aniline, ethanol, reflux, 18-48 hours, 66-98%, (e) 2a-i, ethanol, reflux,
8-48 hours, 45-68%.
3

ACCEPTED MANUSCRIPT
Scheme 3a, Reagents & Conditions: (a)LiOH.H
2
O, 50% ethanol, reflux 2 hours, 66-94%, (b) (method 1 for
derivatives 9a-g) 1) SOCl
derivatives 9h & 9i) Amine, EDC.HCl, DMAP, DMF, under N
2
, reflux 3-4 hours, 2) Amine, CH
2
Cl
2
, rt, 48 hours, 20-76%, or (c) (method 2 for
2
, rt, 24 hours, 46%
Scheme 3b, Reagents & Conditions:(a) LiOH.H
hours, 2) Ethanolamine, CH Cl , rt, 24 hours, 60%.
2 2
O, 50% ethanol, reflux 2 hours, 93%, (b) 1) SOCl , reflux 3
2
2
4
Scheme 4; Reagents & Conditions: (a) LiAlH , THF, -5 to 0°C, 1 hour, 18%.

ACCEPTED MANUSCRIPT
3
.2 Biological Evaluation
3
.2.1 EGFR and HER2 inhibitory assay
3
.2.1.1 Initial screening at a single dose of 10 µM concentration
Compounds bearing small anilines at the 4-position (6a-i, 8a-e, 9a-i and 12a-c)were
assayed for EGFR kinase activity inhibition (Table 1 & 2), while those bearing a bulky
hydrophobic group (7a-i, 10 and 11)were assayed for both EGFR/HER2 kinase activity
inhibition (Table 3). At 10 µM concentration, compound 10 bearing the 3-chloro-4-((3-
fluorobenzyl)oxy)aniline group, similar to lapatinib, and a polar carboxylic acid side chain has
demonstrated potent EGFR inhibition (100%) and significant HER2 inhibition (65%). Series 7
bearing a bulky benzyloxyanilino group showed poor EGFR/HER2 inhibitory activity although
compliant with SAR studies. However, modification of the solvent accessible 5-carboxylate ester
group of 7h to amide group in 11 and carboxylic acid in 10 potentiated the EGFR/HER2
inhibitory activity, 18%, 32% and 100%, respectively against EGFR, 6%, 6% and 65%,
respectively against HER2.
Structure activity relationship among the newly synthesized furo[2,3-d]pyrimidines has
been closely investigated. Meta-substitution on the 4-aniline ring enhanced the EGFR inhibitory
activity as shown by 6c, 6d, 6e and 6g (26-52%) when compared to para-substitution in 6h (6%)
or 3,4-disubstitution in 6i (10%). 3-Cl, 3-Br and 3-CH substituents generally showed enhanced
3
activity over other substituents, but they were still considered as weak inhibitors, thus they were
subjected for further optimization to promote their inhibitory activity.
Optimization of the 5-carboxylate ester group was carried out to increase the
hydrophilicity of the side chain oriented in the solvent accessible region, which resulted in
increase in EGFR/ HER2 inhibitory activity. Three strategies were carried out to render the 5-
carboxylate ester group more hydrophilic: a) the ester group was hydrolyzed to acid in series 8
and compound 10. A major increase in inhibitory activity was observed when comparing an ester
derivative with its acid analogue; 6c compared to 8c, 28% and 85%, respectively, 6d compared
to 8d, 26% and 94%, respectively and 7h compared to 10, 18% (EGFR), 6% (HER2) and 100%
(EGFR), 65% (HER2), respectively. b) Incorporation of an amide linkage in series 9 and
compound 11. Amides with branched or bulky aliphatic side chains demonstrated lower
inhibitory activity compared to those having a terminal hydroxyl group; rendering it more polar.

ACCEPTED MANUSCRIPT
9
g and 9i compared to 9c and 9d, 21%, 30%, 85% and 73%, respectively. c) The ester group was
reduced to methyl alcohol in series 12. A major increase in inhibitory activity was observed
when comparing an ester derivative with its methanol analogue; 6e compared to 12c, 52% and
7
2
2%, respectively, 6d compared to 12b, 26% and 70%, respectively and 6c compared to 12a,
8% and 69%, respectively.
Table 1. Percentage inhibition of EGFR enzymatic activity showed by series 6:
Compound ID
R
EGFR % inhibition at 10 µM
6a
H
22
6
6
6
b
c
d
3-F
3-Cl
3-Br
2
28
26
6
6
e
f
3-CH
3-OCH
3
52
6
3
6
g
3-ethynyl
4-F
3-Cl, 4-F
26
6
10
6
h
6
i
Staurosporine
94 (at 100 nM)
Table 2.Percentage inhibition of EGFR enzymatic activity achieved by series 8, 9 and 12:
Compound ID
R1
R2
EGFR % inhibition at 10 µM
8
a
H
F
Cl
Br
-------------------------
-------------------------
-------------------------
-------------------------
30
49
85
94
8
b
8
c
8
d

ACCEPTED MANUSCRIPT
8
e
CH
H
F
3
-------------------------
58
8
20
85
73
36
65
21
20
30
69
70
72
9
a
-CH
-CH
-CH
-CH
-CH
-CH
2
2
2
2
2
2
CH
CH
CH
CH
CH
CH
2
2
2
2
2
2
3
2
2
OH
OH
OH
OH
OH
9
b
9c
Cl
9
d
Br
CH
Cl
9e
3
9
f
CH
3
9
g
Cl
-CH(CH
)
2
9
h
CH
CH
Cl
3
-CH
-CH
2
CH
CH
CH
N(C
3
9
i
3
2
2
H
4
)
2
O
1
1
1
2a
2b
2c
-------------------------
-------------------------
-------------------------
Br
CH
3
Staurosporine ------------------------------------------
94 (at 100 nM)

ACCEPTED MANUSCRIPT
Table 3.Percentage inhibition of EGFR/HER2 enzymatic activity achieved by series 7 and
compounds 10, 11:
EGFR % inhibition at
HER2 % inhibition at
Compound ID
X
Y
R
1
0 µM
10
34
5
10 µM
7
a
H
H
H
3-F
3-Cl
3
19
7
16
18
20
13
6
7
b
H
H
H
H
H
H
Cl
Cl
H
H
H
H
H
H
H
N
7
c
7
d
3-CH
4-Cl
3
6
6
6
0
18
7
7
e
7f
4-CH
3
7
g
3-Cl, 4-Cl
7
h
3-F
H
7i
10
1
1
0
1
--- --- ----------
--- --- ----------
100
32
65
6
Staurosporine
94 (at 100 nM)
100 (at 100 nM)
3
.2.1.2 Measurement of potential enzyme inhibitory activity (IC )
5
0
Promising candidates, which exhibited EGFR inhibition percent above 75% at 10 µM
concentration (10, 8d, 8c and 9c) were further investigated for their dose-related EGFR
enzymatic inhibition at 5 different concentrations (1 nM - 10 nM - 100 nM - 1 µM - 10 µM) to
subsequently calculate their IC values. 8d bearing a 3-bromoaniline group and a carboxylic
5
0
acid side chain showed the most significant submicromolar EGFR inhibition (IC50 of 0.75 µM)
Table 4).
Table 4. IC values for compounds 10, 8d, 8c and 9c:
(
5
0
Compound ID
EGFR IC50 (µM)
0.951
10
8d
0.752
8c
0.901

ACCEPTED MANUSCRIPT
9c
2.7
Gefitinib [31]
Lapatinib [32]
23 nM
10.8 nM
3
.2.2
In-vitro antiproliferative activity assay against NCI 60-cell line
panel
Sixteen of the final compounds were selected by the National Cancer Institute (NCI),
under the Developmental Therapeutic Program (DTP), namely 6b, 6h, 6i, 7b, 7d, 7e, 7g, 7i, 8b,
a, 9b, 9e, 9f, 9i, 11 and 12a. Compounds are usually selected for screening according to their
9
ability to add diversity to the NCI small molecule compound collection. The submission of novel
heterocyclic ring systems is encouraged. In addition, the submission of compounds with drug-
like properties utilizing the concept of privileged scaffolds is encouraged [33–35]. NCI 60 cancer
cell lines growth percentage results are illustrated in Supplementary Table S1.
The benzyloxyanilino derivative bearing a 5-ethanolamide group (11) demonstrated the
highest cell growth inhibition with mean growth percent inhibition of 39.94%. It exhibited broad
spectrum and good anti-proliferative activity against several NCI cell panel, namely; the
leukemic HL-60(TB) and K-562 cancer cell lines with growth inhibition 57.56% and 66.82%
respectively, the non-small cell lung cancer A549/ATCC, EKVX,HOP-92, NCI-H226, NCI-H23
NCI-H460 and NCI-H522 cell lines with cell growth inhibition 46.52, 52.24%, 58.1%, 59.19%,
4
1.41%, 65.28% and 72.12%, respectively. It showed significant and broad spectrum inhibitory
activity against the CNS cancer SF-268, SF-295, SF-539, SNB-19, SNB-75 and U251 cell lines
with 52.84%, 50.51%, 52.6%, 65.69%, 59.47% and 51.89% inhibition, respectively. It also
illustrated moderate inhibitory activity against the melanoma MALME-3M, SK-MEL-28 and
UACC-62 cell lines with 47.69%, 49.41% and 44.95% inhibition, respectively, It also
demonstrated good inhibitory effect against the ovarian cancer OVCAR-8 cell line with 57.19%
inhibition, the renal cancer ACHN, SN12C and UO-31cell lines with 50.4%, 54.67% and
4
2.35% inhibition, respectively and the prostate cancer PC-3 cell line with 42.45% inhibition.
Finally, it showed significant inhibitory activity against the breast cancer MDA-MB-231/ATCC,
HS 578T and T-47D cell lines with 73.99%, 53.94% and 65.3%, respectively.
Moreover, 9a bearing the same 5-ethanolamide together with a 4-unsubstituted aniline
group, demonstrated selective and potent inhibition against certain leukemic cancer cell lines,
namely; K-562, MOLT-4 and SR with 70.64%, 100.28% and 47.8% inhibition, respectively.

ACCEPTED MANUSCRIPT
Also, 12a showed moderate inhibition against specific leukemic and renal cell lines,
namely; the leukemic HL-60(TB) cell line with inhibition 46%, the renal cancer A498, CAKI-1
and UO-31 cell lines with 48.38%, 46.28% and 43.1%, respectively.
Additionally, 9f exhibited moderate inhibition against specific NSCLC and breast cancer
cell lines, namely; the NSCLC HOP-92 and NCI-H522 cell line with inhibition 42.77% and
4
5.15%, respectively and the breast cancer T-47D cell line with 46.45%.
Finally, 6h, 8b and 9e revealed selective inhibition to certain leukemic cancer cell lines;
h showed 68.7% inhibition against CCRF-CEM cell line, 8b showed 64.83% inhibition against
6
HL-60(TB) cell line and 9e showed 42.59% inhibition against MOLT-4 cell line.
On the contrary, 6b, 6i, 7b, 7d, 7e, 7g, 7i, 9b and 9i showed weak activity against most of
the investigated cell lines.
3
.2.3
In-vitro cytotoxic activity assay against MCF-7 and A549 cancer
cell lines
The cytotoxic activity of the other synthesized compounds not selected by NCI (6a, 6c-g,
7
a, 7c, 7f, 7h, 8a, 8c-e, 9c-d, 9g-h, 10 and 12b-c) was also evaluated against two specific cell
lines, namely the breast cancer MCF-7 cell line and the NSCLC A549 cell line. These cell lines
were chosen specifically as they overexpress EFGR [36,37]. Cytotoxic activity is expressed in
terms of IC and is provided in Table 5.
5
0
Table 5. Cytotoxic activity against MCF-7 and A549 cancer cell lines:
MCF-7
A549
Compound ID
(
IC in µM)
(IC in µM)
5
0
50
6
a
851
>1000
>1000
>1000
>1000
>1000
>1000
>1000
513
6
c
>1000
>1000
>1000
>1000
>1000
>1000
54
6d
6
e
6
f
6
g
a
7
7
c
7
f
>1000
>1000

ACCEPTED MANUSCRIPT
7
h
257
245.5
57.5
68
0.5
191
8
a
8
c
95.5
8d
43
8
9
e
c
57.5
>1000
22
100
>1000
17.4
9d
9
g
190.5
>1000
46
>1000
>1000
21.4
9h
10
1
2b
2c
11
102
1
17
24
Doxorubicin
Gefitinib
0.72
12.05 [38]
0.04
13.59 [39]
Interestingly, 7h demonstrated potent inhibitory activity against A549 cell line (IC 0.5
5
0
µM), known to overexpress EGFR [36], despite its poor EGFR enzymatic inhibition (18%).
When compared to its acid analogue 10 which demonstrated potent enzymatic EGFR inhibition
(
100%) while poor cellular inhibitory activity against A549 cell line (IC 21.4 µM). It might be
50
suggested that the ester analogue 7h, being more hydrophobic, may easily penetrate through the
cell membrane and act as a prodrug releasing the acid analogue 10 intracellularly that potently
inhibits EGFR enzyme.
3
.2.4
In-vivo antitumor activity assessment
3
.2.4.1 Antitumor activities of compounds 10, 7h and 8d compared to
gefitinib against EAC solid tumor model
Testing for anticancer activities through in-vivo tumor models is essential for the
development of novel therapeutics. Accordingly, compounds 10, 7h and 8d were selected for
further in-vivo investigation in an established murine EAC solid tumor model. It is worthy noted
that no significant body weight changes among the different animal groups with no signs of

ACCEPTED MANUSCRIPT
evident toxicity were observed during the treatment. Compounds were tested at the doses of 5
and 10 mg/kg which were orally administered once daily for 7 days. As shown in Figure 2A oral
administration of the acid derivative 10 at doses (5 and 10 mg/kg) exhibited a significant
decrease in tumor volume relative to vehicle treated group starting from day 3, but still
significant from gefitinib-treated group. Moreover, the effect of its ester analogue 7h was dose-
related (Figure 2B), where the higher dose (10 mg/kg/d) exhibited antitumor activity comparable
to gefitinib-treated mice. Interestingly, the 3-bromoaniline derivative 8d at both doses (5 and 10
mg/kg/d) significantly reduced tumor volume that was comparable to gefitinib-treated group, as
shown in Figure 2C. In particular, percent tumor growth inhibition ratio (TGI %) of 10 was
about 34% and 49% at (5 and 10 mg/kg/d) doses, respectively (Table 6). Interestingly, the higher
doses of 7h and 8d resulted in TGI% of 67% and 71%, respectively, which were analogous to
gefitinib-treated group.
In order to further assess their antitumor efficacy, tumor indices were calculated by
dividing tumor weight by the total body weight of the same animal (Table 6). Compound 10 was
unable to significantly reduce tumor indices at 5 and 10 mg/kg/d doses, compared to the control
untreated group. On the other hand, 7h and 8d caused significant decrease (p<0.05) of tumor
indices at all dose levels, compared to the control group. Interestingly, their effects were not
significantly different from that of gefitinib. It is worthy noted that no significant changes of
organs’ indices were detected with gefitinib or other tested compounds at all dose levels (data
not shown).

ACCEPTED MANUSCRIPT
8
6
4
2
00
00
00
00
#
A
Vehicle
#
cpd 10 (5 mg/kg)
cpd 10 (10 mg/kg)
Gefitinib (10 mg/kg)
,#
*
#
,#
*
,
#
*
*,#
*,#
*,#
0
1
2
3
4
5
6
7
8
Days following treatment
8
00
00
00
00
#
B
6
4
2
#
,#
*
Vehicle
#
cpd 7h (5 mg/kg)
cpd 7h (10 mg/kg)
,#
*
,#
*
Gefitinib (10 mg/kg)
*
*
*
0
1
2
3
4
5
6
7
8
Days following treatment
8
6
4
2
00
00
00
00
#
C
#
#
Vehicle
cpd 8d (5 mg/kg)
cpd 8d (10 mg/kg)
Gefitinib (10 mg/kg)
*
*
*
0
1
2
3
4
5
6
7
8
Days following treatment
Figure 2. Effects of 10 (Panel A), 7h (Panel B) and 8d (Panel C) together with gefitinib treatment on tumor growth
in EAC solid tumor bearing mice
Values are given as mean ± SD (n = 6).
Statistical analysis was carried out using one-way analysis of variance (ANOVA) followed by Tukey-Kramer as
indicated as a post hoc test
*
#
: Statistically significant difference from vehicle-treated group at p < 0.05
: Statistically significant difference from gefitinib-treated group at p < 0.05

ACCEPTED MANUSCRIPT
Table 6. Antitumor activities of 10, 7h, 8d and gefitinib against EAC solid tumor model:
Group
Control
Tumor Indices
TGI %
--
0.054 ± 0.014
b
1
1
7
7
8
8
0 (5 mg/kg)
0 (10 mg/kg)
h (5 mg/kg)
h (10 mg/kg)
d (5 mg/kg)
d (10 mg/kg)
0.044 ± 0.017
34.93 %
49.05 %
35.08 %
67.13 %
61.35 %
70.77 %
70.02 %
b
0.039 ± 0.013
a
0.032 ± 0.005
a
0.028 ± 0.007
a
0.025 ± 0.009
a
0.024 ± 0.007
a
Gefitinib (10 mg/kg)
0.021 ± 0.007
Table 6. The three compounds were administered to mice once daily orally for 8 consecutive days at the designated
doses. “Tumor Growth Inhibition” is abbreviated as TGI.
Tumor Indices Data are presented as mean ± SD
n=6
Statistical analysis was carried out by one way ANOVA, followed by Tukey-Kramar post-hoc test
a: statistical significance compared to the corresponding control group at p<0.05
b: statistical significance compared to gefitinib-treated group at p<0.05
3
.2.4.2 Histopathological Examination of EAC
As shown in Supplementary Figure S1A, the control untreated group shows intact anaplastic
cells with polarity pleomorphism and hyperchromachia with less few focal cell deaths.
Treatment of mice with 10 at 5 mg/kg dose resulted in a few focal death of Ehrlich tumor cells
(Supplementary Figure S1B) which was progressed to mild degree of death at the higher dose
1
0 mg/kg (Supplementary Figure S1C). Concerning 7h and 8d, they were more efficient in
killing Ehrlich tumor cells as indicated by severe cell death in a wide area of tumor cells at 10
mg/kg doses (Supplementary Figure S1E & S1G) that was comparable and even more than the
effect of gefitinib at the same dose (Supplementary Figure S1H).

ACCEPTED MANUSCRIPT
3
.2.5
Assessment of apoptosis for compounds 10, 7h and 8d compared
to gefitinib
3
.2.5.1 In-vivo assessment in EAC solid tumor model
In order to assess the anticancer effects of compounds 10, 7h and 8d, the expression of two
apoptotic markers (caspase 3 and cytochrome c) was evaluated in EAC solid tumor model
immunohistochemically. First, it is important to declare that apoptosis has to be tightly regulated
since too little or too much cell death may lead to a variety of pathological conditions. For
instance, faulty apoptosis is a major contributing factor for cancer [40]. Apoptosis is well known
to be mediated by a subfamily of cysteine proteases known as caspases. In mammalian cells,
cytochrome c initiates caspase activation pathway, where a variety of apoptotic stimuli cause
cytochrome c release from mitochondria, which in turn induces a series of biochemical reactions
that result in caspase activation and subsequent cell death [41]. Conspicuously, caspase 3, an
executioner protease enzyme, plays a crucial function in mediating nuclear apoptosis including
processing various procaspases, chromatin condensation and DNA fragmentation as well as cell
blebbing [42].
As shown in Figure 3A, control untreated group has shown mild baseline expression of
caspase 3. Treatment of EAC solid tumor bearing mice with 10 mg/kg/day of 10 significantly
increased caspase 3 expression by about 51%, compared to vehicle-treated group (Figure 3B).
Interestingly, 7h and 8d caused more than 2-fold increase in caspase 3 expression that were
significantly different from the control group and comparable to that of gefitinib-treated group
(Figures 3C, 3D and 3E). A similar pattern of activity was observed with cytochrome c
expression, where control group showed baseline expression of cytochrome c (Figure 4A) while
compound 10 showed moderate effects on expression (Figure 4B). Again, 7h and 8d have
indicated comparable activities to gefitinib-treated group in increasing cytochrome c expression
significantly different from the control group (Figures 4C, 4D and 4E).

ACCEPTED MANUSCRIPT
Figure 3. Effects of compounds 10, 7h and 8d on caspase 3 expression in EAC solid tumor (x 100)
Immunohistochemical staining of caspase 3 in EAC solid tumor of (A): Control group, (B): 10, (C):7h, (D): 8d, (E)
Gefitinib treated EAC bearing mice. All drugs were tested at 10 mg/kg dose level. In addition, quantification of
caspase 3 staining was calculated as optical density (OD) at 10 different fields for each section. Values are presented
as mean ± SD. a and b: Statistically significant difference from control and gefitinib-treated groups respectively at P
<
0.05 using one way ANOVA followed by Tukey-Kramer as a post hoc test.

ACCEPTED MANUSCRIPT
Figure 4. Effects of compounds 10, 7h and 8d on cytochrome c expression in EAC solid tumor (x100)
Immunohistochemical staining of cytochrome c in EAC solid tumor of (A): Control group, (B): 10, (C): 7h, (D): 8d,
(
E) Gefitinib treated EAC-bearing mice. All drugs were tested at 10 mg/kg dose level. In addition, quantification of
caspase 3 staining was calculated as optical density (OD) at 10 different fields for each section. Values are presented
as mean ± SD. a and b: Statistically significant difference from control and gefitinib-treated groups respectively at P
<
0.05 using one way ANOVA followed by Tukey–Kramer as a post hoc test.
3
.2.5.2 In-vitro DNA-Flow Cytometric Analysis
Control A549 cells exhibited proliferative properties with about 69% at G /G phase and 27%
0
1
at S phase and only 0.54% were at Pre-G phase. The tested compounds 10, 7h and 8d induced
1

ACCEPTED MANUSCRIPT
1
accumulation of A549 cells at Pre-G apoptotic phase; as manifested by significant increases in
the percentage of cells at Pre-G by about 17, 32 and 12-folds; respectively, compared to the
1
control. Treatment with gefitinib caused a significant increase in Pre-G phase by about 26-folds,
1
compared to control. In the same manner, Treatment of A549 cells with 10, 7h and 8d resulted in
cell cycle arrest at G /M phase, as revealed by significant rise in the percentage of cells at G /M
2
2
by about 3, 5 and 2-folds; respectively, compared to the control. Again, treatment with gefitinib
caused a significant increase in G /M phase by about 5-folds, compared to A549 control cells
2
(Figure 5).
Cpd 10
Cpd 7h
Cpd 8d
Gefitinib
Control A549
Figure 5. Effect of the synthetic compounds 10, 7h and 8d on DNA-ploidy flow cytometric analysis of A549 cells.
Gefitinib was used as a positive standard therapy
.
3
.2.5.3 Annexin V–FITC Apoptosis Assay
As shown in Figure 6, flow cytometric analysis revealed that A549 cells treated with the
synthetic compounds 10, 7h and 8d showed obvious increases in the percent of annexin V–
FITC-positive apoptotic cells (both early and late apoptosis), compared to control. It is worthy
noted that 7h was the most effective apoptotic inducer, having apoptotic potential higher than
that of the standard drug “Gefitinib”. 7h resulted in a significant elevation in annexin V–FITC-
positive apoptotic cells (both early and late apoptosis) by 25 and 79-folds, respectively,

ACCEPTED MANUSCRIPT
compared to control, whereas gefitinib caused 22 and 61-folds increase, respectively, compared
to control.
Cpd 8d
Gefitinib
Control A549
Cpd 10
Cpd 7h
Sample data
Apoptosis
Necrosis %
%
9.36
17.04
6.49
14.23
0.54
Total
% Early
3.72
6.59
2.78
5.84
% Late
Cpd 10/ A549
Cpd 7h/ A549
Cpd 8d/ A549
Gefitinib/A549
Control A549
4.84
8.69
3.05
6.72
0.11
0.8
1.76
0.66
1.67
0.17
0.26
Figure 6. Effect of the synthetic compounds (10, 7h and 8d) on the percentage of annexin V–FITC-positive staining
in A549 cells. Gefitinib was used as a positive standard therapy.
3
.3 Molecular modeling study
3
.3.1
Docking study
Docking study of the target compounds and analysis of their binding modes were
performed to interpret the biological results and to determine further information about the
binding orientations and interactions. The core scaffold adopted same orientation as the lead
compounds. Being docked in the active site of EGFR, series 6, 8, 9 and 12 fulfilled the key
interactions, where the hydrogen bonds with Met769 and Thr766, via water bridging, were
established, as well as the Pi-cation interaction with Lys721 residue (Figure 7A). Lapatinib
analogues (series 7, compounds 10 and 11) were docked into both EGFR (DFG-out
conformation) and HER2 active sites. Although they adopted same orientations as the lead
compounds in both EGFR/HER2 active sites, they showed poor EGFR/HER2 inhibitory activity.
In the EGFR (DFG-out conformation) the hydrogen bonds with Met793 and Thr854, via water
bridging, were maintained, as well as the Pi-cation interaction with Lys745 residue. However,

ACCEPTED MANUSCRIPT
(10) formed an additional hydrogen bond with Phe856 residue, which may interpret its EGFR
potency, (IC 951 nM) (Figure 7B). Meanwhile, 7c, 7e, 7f and 7g missed the crucial hydrogen
5
0
bond with Thr854, interpreting their lack of inhibitory activity; 5%, 6%, 6% and 0% respectively
Figure 7C). Additionally, in the HER2 active site the hydrogen bond with Met801 and the Pi-
(
cation interaction with Lys753 residue were established by all of the compounds despite their
poor inhibitory activity (Figure 7D).

ACCEPTED MANUSCRIPT
(
B)
(
A)
(
C)
(
D)
Figure 7. Docking pose of the target compounds (A) 8d to the active site of EGFR (B)10 to the active site of EGFR
DFG-out conformation) (C)10 to the active site of HER2 (D)7g to the active site of HER2, missing the crucial
hydrogen bond with Thr854.
(

ACCEPTED MANUSCRIPT
3
.3.2
In silico ADME study
For further understanding for the difference in activity between the esters (series 6), the
acids (series 8), the amides (series 9) and the alcohols (series 12) computer aided ADME study
was performed using the Accelrys Discovery studio 2.5 software. The ultimate goal of in silico
ADME is to predict the disposition behavior of the compounds in the whole body by assembling
all kinetic processes in one global model and hence it is expected to reduce the risk of late-stage
attrition of drug development and to optimize screening and testing by looking only at the
promising compounds. This study is based on the chemical structure of the molecule and
involves the calculation of certain descriptors including; aqueous solubility level (Aq Sol),
human intestinal absorption level (HIA), blood brain barrier penetration Level (BBB),
cytochrome P450 2D6 inhibition (CYP2D6), plasma protein binding level (PPB) and
hepatotoxicity (HepaTox). The amide derivatives having the terminal hydroxyl group 9a-e, the
acid derivatives 8a, 8b and the alcohol derivatives 12a, 12c showed good predicted aqueous
solubility level, while the ester derivatives (series 6), the amide derivatives having simple (9f,
9
h) or branched (9g) or bulky (9i) hydrophobic chains, the acid derivatives 8c-e and the alcohol
derivative 12b showed low ADME aqueous solubility level. These results were consistent with
the enzyme inhibition assay results which revealed that increasing the polarity of the solvent
accessible side chain increases the enzyme inhibition. All of the compounds showed HIA level 0
which means that these compounds are expected to be well absorbed. Blood brain barrier
penetration level of all of the compounds was found to be 2 or 3 which indicates medium to low
BBB penetration; hence the chances of CNS side effects are predicted to be low. The CYP2D6
score predicts the inhibitory and non-inhibitory character of the given chemical structure on
cytochrome P450 2D6 enzyme. More than half of the compounds were predicted as non-
inhibitors; hence no drug-drug interactions would be expected upon administration of these
compounds. Most of the compounds had plasma protein binding levels >95%, thus these drugs
would exhibit longer t1/2 and hence less frequent drug administration. The hepatotoxicity level of
all compounds was 1 except for (9i). Hence the compounds are predicted to possess
hepatotoxicity. Further experimental studies are required to determine the hepatotoxic dose
levels. The calculated descriptors from the ADME study are illustrated in Supplementary Table
S2.

ACCEPTED MANUSCRIPT
4
. Conclusion
Four series of furo[2,3-
d
]pyrimidine derivatives were designed, synthesized and evaluated for
their in-vitro EGFR/HER2 inhibitory activity as well as their anti-proliferative activity against
cell line panel. Modification of the solvent accessible 5-position side chain greatly affected their
in-vitro EGFR/HER2 inhibitory activity. At 10 µM, lipophilic ethyl esters, series 6 and series 7
demonstrated weak EGFR inhibition (0-52%) and even poor HER2 inhibition (3-20%), while
their acid analogues series 8 and compound 10, showed the best inhibitory activity against
EGFR (30-100%) and moderate HER2 inhibition (65%), meanwhile the amide analogues series
9
and compound 11 covered a wide EGFR inhibitory activity range (8-85%) and finally, the
alcohol analogues series 12 revealed enhanced EGFR inhibition (69-72%). The variation in
inhibitory activity within the amide derivatives series 9 seemed to depend on the hydrophilicity
of the side chain, where incorporation of a terminal hydroxyl group imparted higher inhibitory
activity. These results were further explained using ADME study which revealed that the amide
derivatives having the terminal hydroxyl group, the acid derivatives and the alcohol derivatives
showed better predicted aqueous solubility levels than their ester analogues and their amide
analogues having simple or branched or bulky hydrophobic side chains. The in-vivo antitumor
activity assessment against EAC solid tumor model suggests that 7h and 8d (10 mg/kg dose)
exhibit comparable antitumor activity to that of gefitinib at the same dose and this effect could be
explained, at least partly, via activation of apoptosis. Additionally, 7h revealed apoptotic
potential higher than that of gefitinib during the in-vitro flow cytometric analysis. Moreover, the
variation in the in-vivo antitumor activity between 7h and 10 supports the assumption, observed through
the antiproliferative activity against A549 cell line, that 7h may be regarded as a prodrug for its acid
analogue 10.

ACCEPTED MANUSCRIPT
5
. Experimental
5
.1 Chemistry
The used starting materials and reagents were purchased from Sigma-Aldrich
(USA) or Alfa-Aesar Organics and used without prior purification. Melting points were
determined using Stuart Scientific apparatus and were uncorrected. Analytical thin layer
chromatography (TLC) purchased from Merck (Merck, Darmstadt, Germany) was used
to monitor the reactions and performed on silica gel 60 packed on Aluminum sheets, with
visualization under U.V. light(254 nm). FT-IR spectra were recorded on a Thermo
Scientific Nicolet iS10 spectrometer. 1HNMR spectra were run at Bruker 400 MHz
spectrophotometers and 13C NMR spectra were run at Bruker 400 MHz spectrometer, in
δ
scale (ppm), J (Hz) using TMS as reference at Center for Drug Discovery and
Development Research, Ain Shams University. 1HNMR spectra were run at Joel 300
MHz spectrophotometers in scale (ppm), J (Hz) using TMS as reference at
δ
Microanalytical center, Cairo University. Mass spectra were recorded on Thermo
Scientific ISQLT mass spectrometer at the Regional Center for Mycology and
Biotechnology, Al-Azhar University. Elemental analyses were performed on a Thermo
Scientific Flash 2000 elemental analyzer at the Regional Center for Mycology and
Biotechnology, Al-Azhar University. Intermediates ( 1a-1c [43], 1e-1i [43], 2a-2c [23],
2
e-2i [23], 3 [44], 4-5 [19]) were prepared according to the reported procedures.
5
.1.1 1-Methyl-3-((4-nitrophenoxy)methyl)benzene (1d)
A mixture of
chloride (1 equiv., 6.47 mmol, 0.91 gm) and anhydrous K CO (1.66 equiv., 10.74 mmol,
p-nitrophenol (1 equiv., 6.47 mmol, 0.9gm), 3-methylbenzyl
2
3
1
.48gm) in dry DMF (10 mL) was heated at 80°c for 24 hours. The mixture was then
added to ice/water with continued stirring. The resulting solid was filtered off, washed
with water and recrystallized from ethanol/water to afford the titled compound as off-
1
white crystals (1.54gm, 98%); m.p. 90-92°C; H NMR (400 MHz, DMSO-
d
6
)
δ
8.21 (d, J
=
9.2 Hz, 2H, ArH), 7.32 – 7.25 (m, 3H, ArH), 7.21 (d, J = 9.2 Hz, 2H, ArH), 7.16 (d, J =
7
.1 Hz, 1H, ArH), 5.21 (s, 2H, -OCH Ar), 2.32 (s, 3H, CH ).
2
3

ACCEPTED MANUSCRIPT
5
.1.2 4-((3-Methylbenzyl)oxy)aniline (2d)
A mixture of (1d) (1 equiv., 6mmol, 1.46 gm), reduced iron (3 equiv., 18 mmol, 1
gm) and NH Cl (9 equiv, 54 mmol, 2.89 gm) in ethanol (70%, 70 mL) was heated at
0°C for 1 hour. The reaction was allowed to cool to room temp, and then filtered over
4
7
celite. The filtrate was evaporated under vacuum giving a buff-orange solid which was
stirred with EtOAc, and then the resulting mixture was filtered. The filtrate was dried
over anhydrous Na SO and evaporated under vacuum then recrystallized from ethyl
2
4
acetate/petroleum ether to give the titled compound as brown crystals (0.84gm, 66%);
1
m.p. 72-74°C ; H NMR (400 MHz, DMSO-d6
)
δ
7.22 (m, 3H, ArH), 7.11 (d,
= 8.8 Hz, 2H, ArH), 5.66 (s, 2H, -
OCH Ar), 4.92 (s, 2H, NH D O exchangeable), 2.31 (s, 3H, CH ).
J = 7.3 Hz,
1
H, ArH), 6.77 (d, J = 8.8 Hz, 2H, ArH), 6.62 (d, J
2
2
3
5
.1.3 General procedure for the preparation of compounds (6a-i)
To a solution of (5) (1 equiv., 1mmol, 0.24 gm) in ethanol (15 mL), the respective
aniline derivative (viz; aniline, 3-fluoroaniline, 3-chloroaniline, 3-bromoaniline, m-
toluidine, m-anisidine, 3-ethynylaniline, 4-fluoroaniline, 3-chloro-4-fluoroaniline) (1.1
equiv., 1.1mmol) and TEA (2 equiv., 2mmol, 0.3 mL) were added. The mixture was
heated under reflux for 24-48 hours. The mixture was allowed to cool to room
temperature. The precipitated solid was collected by filtration, washed with ethanol and
recrystallized from ethanol to give the titled compounds (6a-i).
5
.1.3.1
Ethyl 6-methyl-4-(phenylamino)furo[2,3-d]pyrimidine-5-carboxylate (6a).
1
White crystals (0.2g, 66%); m.p. 176-178°C; H NMR (400 MHz, DMSO-d6
)
δ
10.56
= 7.9 Hz, 2H,
= 7.4 Hz, 1H, ArH), 4.44 (q, = 7.1
(
s, 1H, NH D O exchangeable ), 8.46 (s, 1H, pyrimidine H), 7.78 (d, J
2
ArH), 7.39 (t,
J
= 7.8 Hz, 2H, ArH), 7.10 (t,
J
J
13
Hz, 2H, -CH CH ), 2.74 (s, 3H, CH ), 1.39 (t,
J
= 7.1 Hz, 3H, -CH CH ); C NMR
2 3
2
3
3
(101 MHz, CDCl3)
δ
165.79, 165.36, 159.87, 155.02, 154.34, 139.25, 129.00, 123.43,
+
1
20.58, 108.50, 101.13, 62.10, 15.25, 14.35; MS : (Mwt. : 297) : m/z , 297 [M ,
(
91%)] , 296 (100%); Anal. Calcd for C H N O : C, 64.64; H, 5.09; N, 14.13;
16 15 3 3
Found: C, 64.64; H, 5.09; N, 14.23.
5
.1.3.2
Ethyl 4-((3-fluorophenyl)amino)-6-methylfuro[2,3-d]pyrimidine-5-
carboxylate (6b).