Inhibiting epidermal growth factor receptor signalling potentiates mesenchymal–epithelial transition of breast cancer stem cells and their responsiveness to anticancer drugs

Article abstract of DOI:10.1111/febs.14084

The recurrence of breast cancer in patients is a persistent challenge to the medical fraternity. Breast tumor contains a small population of cells with high tumor initiating and metastatic potential, known as cancer stem cells (CSCs), which are resistant to existing chemotherapeutics. CSCs contribute to the aggressiveness of triple negative breast cancers (TNBCs), thereby necessitating the identification of molecular targets on breast CSCs. TNBC cell line MDA-MB-231, in comparison with MCF-7, demonstrated a higher expression of epidermal growth factor receptor (EGFR). Thus, the naturally occurring flavanone, chrysin, with limited potential as a chemotherapeutic agent, was structurally modified by designing an analog with EGFR binding affinity using a molecular docking approach and subsequently synthesised. Chrysin analog CHM-09 and known EGFR inhibitors demonstrated a comparable anti-proliferative, anti-migratory activity along with the induction of apoptosis and cell cycle arrest in MDA-MB-231. Furthermore, sorted CD24^?/CD44⁺-breast CSCs and CD24⁺-breast cancer cells from MDA-MB-231 demonstrated a markedly high expression of EGFR in the former than in the latter. CHM-09 and EGFR inhibitors could perturb EGF-induced EGFR signalling of breast CSC proliferation, migration, mammosphere formation and mesenchymal tri-lineage differentiation. CHM-09 or EGFR inhibitors not only led to inactivation of EGFR downstream signalling pathways such as Akt, extracellular signal regulated kinase and signal transducer and activator of transcription 3, but also induction of mesenchymal–epithelial transition as confirmed by decreased N-cadherin and increased E-cadherin expression. Finally, combinatorial treatment of EGFR inhibitors and doxorubicin led to significant increase in breast CSCs responsiveness to a chemotherapeutic drug. The results of the present study suggest that EGFR is a therapeutic target in breast CSCs and that abrogation of EGFR signalling along with chemotherapeutic drugs is an effective approach against breast cancer.

Full text of DOI:10.1111/febs.14084

Inhibiting Epidermal Growth Factor Receptor Signalling Potentiates Mesenchymal-
Epithelial Transition of Breast Cancer Stem Cells and their Responsiveness to
Anticancer Drugs
1
,2
1,2
3
1,2
Kanakaraju Manupati , Neha R Dhoke , Tanusree Debnath , Ragini Yeeravalli , Kalpana
1
1
4
3
1,2
Guguloth , Shahrzad Saeidpour , Utpal Chandra De , Sudhan Debnath , Amitava Das *
1
Centre for Chemical Biology, CSIR-Indian Institute of Chemical Technology, Uppal Road,
Hyderabad – 500 007, India
2
Academy of Scientific & Innovative Research, 2 Rafi Marg, New Delhi – 110 001, India
3
Department of Chemistry, Maharaja Bir Bikram College, Agartala – 799 004, Tripura, India
4
Department of Chemistry, Tripura University, Agartala–799 130, Tripura, India
*
Corresponding Author: Dr. Amitava Das, PhD, Ph No.: +91-40-27191873; E-mail:
amitavadas@iict.res.in; amitavadas.iict@gov.in
Running Title: EGFR inhibitor sensitize breast CSC to chemotherapeutic drug
Article type : Original Articles
Conflict of Interest: “The authors declare no potential conflicts of interest.”
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process, which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1111/febs.14084
This article is protected by copyright. All rights reserved.

-
+
Key words: CD24 /CD44 -breast CSCs, mammospheres, EGFR, MET, chemo-
responsiveness.
Abbreviations: EGFR- Epidermal growth factor receptor, TNBC- Triple negative breast
cancer, EMT- Epithelial to Mesenchymal transition, RTK- Receptor tyrosine kinase, ER-
Estrogen receptor, PR-progesterone receptor, Her2- Human epidermal growth factor
receptor2, CSCs- Cancer Stem Cells, MET- Mesenchymal to Epithelial transition, CD-
Cluster of differentiation.
Abstract
Recurrence of breast cancer in patients is a persisting challenge to medical fraternity. Breast
tumor contains a small population with high tumor initiating and metastatic potential known
as Cancer Stem Cells (CSCs) that are resistant to existing chemotherapeutics. CSCs
contribute to the aggressiveness of Triple Negative Breast Cancers (TNBCs) thereby
necessitates identifying molecular targets on breast CSCs. TNBC cell line MDA-MB-231, as
compared to MCF-7 depicted a higher expression of Epidermal Growth Factor Receptor
(EGFR). Thus naturally occurring flavanone, chrysin with limited potential as
chemotherapeutic agent was structurally modified by designing analog with EGFR binding
affinity using molecular docking approach and synthesised. Chrysin analog, CHM-09 and
known EGFR inhibitors depicted a comparable anti-proliferative, anti-migratory activity
along with induction of apoptosis and cell cycle arrests in MDA-MB-231. Furthermore, sorted
-
+
+
CD24 /CD44 -breast CSCs and CD24 -breast cancer cells from MDA-MB-231 depicted
markedly high expression of EGFR in the former than latter. CHM-09 and EGFR inhibitors
could perturb EGF-induced EGFR signalling of breast CSC’s proliferation, migration,
This article is protected by copyright. All rights reserved.

mammosphere formation and mesenchymal tri-lineage differentiation. CHM-09 or EGFR
inhibitors not only led to inactivation of EGFR downstream signalling pathways like AKT,
ERK and STAT3 but also induction of mesenchymal-epithelial transition as confirmed by
decreased N-cadherin and increased E-cadherin expressions. Finally, combinatorial treatment
of EGFR inhibitors and doxorubicin led to significant increase in breast CSCs responsiveness
to chemotherapeutic drug. This study suggests that EGFR as a therapeutic target in breast
CSCs and abrogation of EGFR signalling along with chemotherapeutic drugs as an effective
approach against breast cancer.
Introduction
Cancer is an imbalance between cell-division and death that favours unlimited growth of
tumor. Various tissue-specific cancers such as breast, lung, and liver, occur due to DNA
mutations caused by physical or chemical mutagens. Among these, breast cancer is the
second most common cause of morbidity and mortality in women [1]. To date, several anti-
cancer therapeutics are targeted against a number of receptor tyrosine kinases (RTK) that are
known to play major role in tumorigenesis by activating downstream signalling [2]. Among
the RTKs, Transforming Growth Factor Receptor-beta has been reported to promote breast
cancer metastasis through induction of epithelial-to-mesenchymal transition (EMT) [3],
Vascular Endothelial Growth Factor Receptor as a mediator of breast tumor angiogenesis [4]
and Platelet-derived Growth Factor Receptor as a predominant signalling pathway in
maintenance of mesenchymal phenotype of breast cancer [5]. Literature suggests a crucial
role of Epidermal Growth Factor Receptor (EGFR) in breast cancer cell proliferation and
migration [6]. EGFR belongs to the ErbB tyrosine kinase family that comprises of four
receptors such as EGFR/ErbB1, HER2/ErbB2, HER3/ErbB3 and HER4/ErbB4 [7]. Among
these, EGFRs are highly expressed in human malignant tumors [8] but is a poor prognostic
This article is protected by copyright. All rights reserved.

factor in breast cancer [9]. The activation of EGFR phosphorylates its downstream signalling
molecules AKT, STAT3 that are involved in cellular migration and MAPK signalling which
is implicated in cell proliferation [10]. While breast cancer patients treated with EGFR
inhibitors such as erlotinib and gefitinib have been shown with minimum clinical activity
[
[
11] but with reduced side-effects in comparison to conventional chemotherapeutic drugs
12]. Therefore, it is pertinent to target EGFR signalling with more potent inhibitors which
are naturally derived chemicals as cancer therapeutics.
Chrysin, a flavonoid is a natural phytochemicalabundantly found in many plant extracts. It
shows cytotoxicity against colon, lung, cervical, liver, prostate, and pancreatic cancer [13].
The use of chrysin as a chemotherapeutic agent has limitations such as poor therapeutic
potential and poor absorption, rapid metabolism, and fast systemic elimination [14]. Although
liposomal nanoparticles conjugated with chrysin have been used for efficient anticancer drug
delivery [15], we hypothesize that designing of synthetic analogs of chrysin against a specific
molecular target may improve its efficacy as chemotherapeutic agents.
Inspite of available chemotherapeutic agents against breast cancer, reoccurrence of cancer
often arises due to the presence of a small sub-population of cancer cells known as Cancer
Stem Cells (CSCs). CSCs are tumor-initiating cells with self-renewal ability and high
metastatic potential [16]. CSCs are resistant to existing chemotherapy and radiotherapy due
to higher expression of efflux transporters, enhanced DNA repair mechanism and resistance
to apoptosis [17].
In the present study, a higher expression of EGFR was observed in aggressive triple negative
breast cancer (TNBC) cell line, MDA-MB-231 as compared with luminal breast cancer cell
line, MCF-7. This led us to design and synthesize a chrysin analog, CHM-09 that binds to
EGFRas confirmed by molecular docking studies and cellular thermal shift assays.Inhibition
This article is protected by copyright. All rights reserved.

of EGFR signalling pathway by known EGFR inhibitors or CHM-09 caused significant
perturbation on proliferation and migration of parental MDA-MB-231 cells as well as sorted
+
-
+
breast cancer cells (CD24 ) and breast CSCs (CD24 /CD44 ) populations. This was followed
by aninteresting observation of a higher expression of EGFR in the breast CSCs as compared
to breast cancer cells. Both CHM-09 as well as EGFR inhibitors could inhibit the EGF-
induced increased proliferation, migration, mammosphere formation and tri-lineage
-
+
differentiation (owing to mesenchymal phenotype) of CD24 /CD44 -breast CSCs along with
concomitant increase in epithelial and decrease in mesenchymal gene and protein expression.
A simultaneous treatment with chemotherapeutic drugs and CHM-09 or EGFR inhibitor in
breast CSCs made them increasingly responsive to the anticancer drugs. This study
emphasises the role of EGFR as a therapeutic target for breast CSCs and a combination
therapy of EGFR inhibitors and chemotherapeutic agents as a potential strategy targeting
breast CSCs to prevent cancer recurrence.
Results
Differential expression of EGFR in breast cancer cell lines: EGFR expression in MCF-7
+
+
-
-
-
-
(
ER /PR /Her2 , Luminal type) and MDA-MB-231 (ER /PR /Her2 , Basal type) breast cancer
cells using semi-quantitative RT-PCR (Fig 1A) and immunoblotting (Fig 1B) depicted
relatively high level in MDA-MB-231 as compared to MCF-7. This observation suggests a
crucial role of EGFR in the aggressive triple negative breast cancer (TNBC) types which is
associated with poor prognosis and reduced survival in patients [18].
Molecular docking of chrysin analogs on EGFR: Chrysin (Fig 1C), a naturally occurring
flavanone having a limited clinical potential as an anticancer agent led us to design using
chemical docking approach and synthesize its analog, CHM-09 (Fig 1D) with improved
This article is protected by copyright. All rights reserved.

binding affinity to EGFR (Fig 1E). The low resolution X-ray crystal structures of different
EGFR (PDB ID: 3W32, 3W33, 3POZ and 3W2R) complexed with co-ligands have been
retrieved from RCSB protein data bank (www.rcsb.org). The final selection of EGFR was
performed on the basis of the docking accuracy by calculating root mean square deviation
(
RMSD). The RMSD values were calculated by superimposing best docked poses of co-
ligands on its native co-ligand. The RMSD value has been observed to be the lowest for
W32 (0.8254) (Table 1). The XP glide scores of co-ligands with receptors 3W32, 3W33
3
were −14.297 and −12.624, respectively and the important interacting amino acid residues
were observed to be MET-793, PHE-856, LEU-718 (Fig 1F). The interacting amino acid
residues for CHM-09 were observed to be MET-793, ASP-855 (2), LYS-745 with XP glide
score of −12.187 (Fig 1E, 1F). Similarly, molecular docking of commercially available
EGFR inhibitors AG-1478 and PD-153035 depicted that ASP-855 is the crucial amino acid
residue for inhibition which is also present in case of CHM-09 (data not shown). Other than
the common interactions with EGFR inhibitors, CHM-09 also have interactions with MET-
7
93, LYS-745. Based on the active site amino acid residues and XP glide score of both
CHM-09 and co-ligands (Fig 1G, 1H), chrysin was isolated from the bark of Oroxylum
indicum (Fig 2A), and chrysin analog, CHM-09 was synthesized as described in the scheme
(Fig 2B) and characterized using spectral analysis (Fig 2C, 2D) before subjecting for further
evaluations. To further support the in silico molecular docking of EGFR-CHM-09
interaction, we performed cellular thermal shift assay (CETSA) to elucidate the binding of
CHM-09 to EGFR in MDA-MB-231 cells. CETSA evaluates the binding of a ligand or small
molecule to its target protein that leads to increased target protein stabilization at differential
temperature [19]. For binding assessment, cells were treated with EGF (25ng/mL), CHM-09
and EGFR inhibitors-AG1478 and PD153035 (25μM) for 1h followed byimmunoblotting.
Optimum EGFR stabilization by CHM-09 binding was observed till 46°C (Fig 3A). Cells
This article is protected by copyright. All rights reserved.

treated with CHM-09, PD153035 depicted EGFR expression at 50°C as compared to
unbound protein in DMSO-treated cells that depicted a complete disappearance of EGFR at
the same temperature (Fig 3A). The significant shift in melting temperatures of EGFR and its
stabilization upon addition of the CHM-09 indicated that both the compounds bind to this
protein.
Proliferation of breast cancer cells by EGFR inhibitors: Next, we evaluated the anti-
proliferative potential of our synthesized chrysin analog, CHM-09 and compared it with its
parent molecule, chrysin and EGFR inhibitors, AG1478 and PD153035 using cell viability
assays. The compound CHM-09 depicted a higher efficacy against a panel of TNBC cell
lines, MDA-MB-231 (Fig 3B), MDA-MB-453 (Fig 3C), MDA-MB-468 (Fig 3D) and BT-549
(Fig 3E) with low IC50 values (µM) 4.52±1.4, 8.35±0.24, 7.88±0.2 and 10.17±0.8,
respectively as compared to MCF-7 (Fig 3F; IC (µM): 25.71±0.1). The anti-proliferative
5
0
potential of chrysin against both the cell types luminal and TNBCs was relatively comparable
IC (µM): MCF-7 -32.08±2.1 and MDA-MB-231 - 40.01±6.7, MDA-MB-468 - 36.02±1.2,
(
5
0
MDA-MB-453 - 28.96±1.82 and BT-549 -33.97±2.4). EGFR inhibitors, AG1478 depicted an
equal anti-proliferative effect in MCF-7 (IC : 11.47±1.2µM) and MDA-MB-231 (IC :
5
0
50
1
1.72±0.2µM) cells. In contrast, PD153035 another EGFR-specific inhibitor reduced
proliferation of MDA-MB-231 cells by half of inhibitor concentration (IC : 3.20±0.5µM) as
5
0
compared to MCF-7 (IC : 6.69±0.9µM). However, TNBC cell lines depicted relatively high
5
0
proliferation in presence of PD153035, with IC50 values (MDA-MB-468 –15.4 ± 0.8µM,
MDA-MB-453 –10.53±0.92µM and BT-549 –26.1 ± 1.3µM). Interestingly, CHM-09 depicted
a less toxicity towards HEK293, human primary kidney control cells with 5-10 –fold high
IC values (53.04±0.72µM) as compared to TNBCs (Fig 3G) suggesting a good therapeutic
5
0
window. EGFR inhibitors depicted relatively similar toxicity towards HEK293 control cells
This article is protected by copyright. All rights reserved.

(AG1478-22.15±3.09µM, PD153035-18.40±0.13µM) (Fig 3G). These results correlated well
with EGFR expression levels in these breast cancer cell lines indicating that chrysin analog,
CHM-09 with improved EGFR binding ability effectively decreases the cell viability in
TNBC cell lines that expresses relatively high EGFR.
Modulation of aggressive TNBC cell cycle and apoptosis by EGFR inhibitors: To further
substantiate the in vitro anti-proliferative effect of EGFR inhibitors and/or chrysin analog,
CHM-09 at low micromolar range on MDA-MB-231 cells, we evaluated the BrdU
incorporation using cell cycle analysis. CHM-09 treated cells depicted 62.3% of cells in
G /G phase of the cell cycle as compared to 53.3% in vehicle control group (Fig 4A)
0
1
indicating a G /G cell cycle arrest. EGFR inhibitors, AG1478 and PD153035 treatment also
0
1
resulted in G /G cell cycle arrest with 62.4% and 64.5%, respectively. The expression of
0
1
unique cyclins and cyclin-dependent kinases (CDKs) along with CDK inhibitors (CKIs)
occur during the cell cycle progression. The gene expression levels of cyclin D, cyclin E and
their CDK partners, CDK2 that regulates G /G phase was markedly decreased in cells treated
0
1
with inhibitors as well as CHM-09 as compared with vehicle control (Fig 4B).
To further corroborate the cell cycle gene expression of EGFR inhibitors and/or chrysin
analog, CHM-09 on MDA-MB-231 cells, we evaluated the protein expression levels too.
Expression of CDK4 and CDK2 and their partners Cyclin D and Cyclin E regulates G phase
1
of cell cycle was markedly decreased in cells treated with inhibitors indicates that
accumulation of cells at G /G phase (Fig 4C). These results correlated well with BrdU cell
0
1
cycle analysis data. In contrast, cyclin B regulates G /M phase of cell cycle was increased
2
whereas with CDK1 no obvious changes was observed in presence of inhibitors as compared
to vehicle control indicates M phase progression. This observation was further substantiated
This article is protected by copyright. All rights reserved.

by a concomitant increased expression of CKI, p21 and decreased expression of p27 in
presence of inhibitors (Fig 4C) suggesting that major role of p21 in preventing the CDK4 and
CDK2 complexes and their interacting partners thereby arresting cells at the G /G phase of
0
1
cell cycle. These results collectively suggests an apoptotic cell death by these treatments, led
us to evaluate apoptotic populations using flow cytometric analysis. A significantly high
percent (~98.4%) of cells were observed to undergo apoptosis in CHM-09 treated cells as
compared to vehicle control (Fig 5A). Similarly, the EGFR inhibitors AG1478 and
PD153035 induced apoptosis with 84.4% and 94.5%, respectively (Fig 5A). Further
substantiating this observation, pro-apoptotic genes –Bax and Bid (Fig 5B) as well as
proteins–BAX, Caspase3 and Annexin V (Fig 5C) expression levels were significantly high
in cells treated with either CHM-09 or EGFR inhibitors with a concomitant decrease in anti-
apoptotic, Bcl-2 expression at gene (Fig 5B) and protein level (Fig 5C). Also a significant
increase in expression of p53 further substantiated induction of apoptotic signalling.
Attenuation of EGFR downstream signalling in aggressive TNBC cells by CHM-09: EGFR
activation has been associated with multiple biological processes in cancer progression such
as cell proliferation, migration and metastasis through different signalling pathways [20]. To
evaluate the effect of CHM-09 on signal transduction mechanism, we performed immunoblot
analysis of EGFR downstream signalling pathways-AKT, ERK and STAT3. Aggressive
TNBC, MDA-MB-231 cells that express high EGFR were treated with increasing
concentration of CHM-09 depicted a dose-dependent decrease in phosphorylation of EGFR,
AKT, ERK and STAT3 (Fig 6A). Similar CHM-09-mediated dose-dependent decrease in
phosphorylation of these signalling mediators wasalso observed in BT-549 (Fig 6B), MDA-
MB-453 (Fig 6C) and MDA-MB-468 (Fig 6D). β-catenin, a downstream target of AKT
signalling pathway [21], was also downregulated in CHM-09 treated cells indicating that
This article is protected by copyright. All rights reserved.

AKT attenuation leads to downregulation (decrease in phosphorylation) of β-catenin.
Inhibition of EGFR by PD153035 also abolished activation of AKT, STAT3 and ERK
signalling pathways responsible for cell migration and proliferation, respectively (Fig 7A). In
order to confirm if this observation holds true, we performed proliferation assay in MDA-MB-
2
31 cells treated with EGF in presence/absence of EGFR, ERK, AKT and STAT3 signalling
blockers at indicated doses. The EGF-induced proliferation was significantly decreased by
CHM-09 similar to EGFR inhibitors, AG1478 and PD153035 (Fig 7B). Also wortmannin,
PD98059 and static inhibitors of AKT, ERK and STAT3, respectively could significantly
decrease EGF-induced proliferation (Fig 7B). Furthermore, addition of CHM-09 along with
EGFR inhibitors or AKT and ERK inhibitors did not alter the cellular proliferation as
compared with when added separately (Fig 7B). However, combination of CHM-09 and
static significantly reduced proliferation as compared to static alone (Fig 7B) suggesting
unique EGFR-STAT3 signalling pathway to regulate cell proliferation of these aggressive
TNBCs. Next, we analyzed EGF-induced migration potential of MDA-MB-231 cells in
presence/absence of pathway inhibitors by scratch/migration assay. The rate of cell migration
was evaluated temporally at 6, 12 and 24h time-points. EGF-induced migration was
significantly attenuated by these inhibitors at treated concentrations (Fig 7C). Combinations
of CHM-09 with these pathway inhibitors depicted a further significant decrease in cell
migration by wortmannin and static (Fig 7C) suggesting a role of EGFR-AKT and EGFR-
STAT3 signalling-mediated regulation of cell migration in aggressive TNBCs (Fig 7C).
Aggressive property of TNBCs are contributed by a small highly migratory population
known as Breast Cancer Stem Cells (CSCs) that differentially express specific cluster of
differentiation (CD) markers on their cell surface [22,23].
This article is protected by copyright. All rights reserved.

High EGFR expression in breast CSCs renders them sensitive to EGFR inhibitors: Breast
CSCs were sorted separately from a panel of TNBC cell lines – MDA-MB-231, MDA-MB-
4
53, BT-549 and MDA-MB-468 based on presence or absence of CD24 cell surface marker
-
(
Fig 8A). Subsequently CD24 cell populations were further subjected to CD44 selection (Fig
-
+
8
A). The sorted CD24 /CD44 (breast CSCs – percent yield of total: 9.3±1.6) cells expressed
+
CD44 and were negative for CD24 whereas CD24 (breast cancer cells – percent yield of
total: 42.9±6.1) were positive for CD24 cell surface markers as analyzed using confocal
imaging (Fig 8B). Additionally, flow cytometry analysis revealed 77.5% CD44 expressing
cells while negative for CD24 surface marker expression thereby confirming the purity of
-
+
-
+
these CD24 /CD44 sorted fractions (Fig 8C). Upon characterization, CD24 /CD44 depicted
a marked increase in CSC-specific genes expression -ALDH1, MUC1, ABCG2, CD133 (Fig
+
-
8
D) as compared to CD24 sorted populations. CD24 gene expression was lacking in CD24
+
/
CD44 population (Fig 8D). ALDH1, an enzyme which is expressed at high levels in CSCs
-
is known to be major characteristic feature of breast CSCs. Our sorted populations, CD24
+
/
CD44 exhibited a 100% aldefluor fluoroscence as compared to control (DEAB) indicating
+
an ALDH population (Fig 8E). Encouragingly, a markedly high level of EGFR expression
-
+
+
was observed in CD24 /CD44 as compared with CD24 sorted cell populations (Fig S1).
-
+
Next, cell viability of the sorted CD24 /CD44 -breast CSC populations from TNBC cell lines
depicted a markedly decrease in cell proliferation in presence of CHM-09 with low IC50
values (µM) of 1.34±0.37 (MDA-MB-231), 6.1±0.3 (MDA-MB-453), 4.3±0.3 (BT-549) and
+
3
8
.2±0.1 (MDA-MB-468) (Fig 9A) as compared with CD24 populations (IC values (µM) of
50
.8±2.17 (MDA-MB-231), 17.6±1.1 (MDA-MB-453), 18.9±1.2 (BT-549) and 14.9±1.2
(MDA-MB-468) (Fig 9B). The effect of CHM-09 on breast CSCs was 1.7-4.0 –fold higher
than EGFR inhibitor, PD153035 (Fig 9A). This observation suggests a higher expression of
This article is protected by copyright. All rights reserved.

EGFR in breast CSCs as compared with breast cancer cells leads to better inhibition of
EGFRby these inhibitors thereby modulating further downstream signalling-mediated cell
viability.
Attenuation of EGFR downstream signalling in breast CSCs: To further support the
observation of EGFR inhibition led to decrease in physiological phenotype of breast CSCs we
explored the downstream signalling of EGFR in presence of CHM-09 as well as various
receptor blockers using immunoblotting techniques. An EGF-induced increased
phosphorylation of EGFR as well as its downstream kinases –ERK, AKT and STAT3
(Transcription factor) was observed (Fig 9C). Inhibition of EGFR by PD153035 or CHM-09
decreased the phosphorylation of AKT, ERK and STAT3 in presence of EGF, indicating that
EGFR blocking leads to abrogation of downstream signalling pathways (Fig 9C). Use of
specific pathway blockers decreased activation of these signalling mediators. Next to confirm
-
+
if the downstream signalling pathways of EGFR holds true in CD24 /CD44 -breast CSCs, we
evaluated the EGF-induced proliferation in combination of EGFR inhibitor and other pathway
blockers. Indeed, EGF-induced increased proliferation of breast CSCs were significantly
inhibited by these pathways blockers individually (Fig 9D). However, no further decrease in
proliferation was observed when treated along with CHM-09 (Fig 9D). This observation
suggests the role of EGFR signalling in maintenance of stemness (mammospheres, etc.) in
breast CSCs.
EGFR inhibition suppresses EGF-induced breast CSC properties –mammosphere formation
and migration: The stemness property of breast CSCs are often evaluated using an in vitro
assay system, wherein these breast CSCs are enriched in suspension cultures as
‘mammospheres’ that imitate in vivo tumors [24]. The breast CSCs sorted from MDA-MB-231
This article is protected by copyright. All rights reserved.

(Fig 10A, 10B and 10C), BT-549 (Fig 10D, 10E and 10F), MDA-MB-453 and MDA-MB-468
(data not shown) when treated with EGF (10ng/ml) depicted significantly larger size
mammospheres and higher sphere forming efficiency as compared with vehicle control. In
contrast, the sphere forming efficiency was significantly decreased by CHM-09 in dose
dependent manner in presence or absence of EGF as well as PD153035 as depicted
quantitatively (Fig 10C and 10F). These findings indicated that EGF-induced in vitro tumor
formation ability of breast CSCs was hindered by PD153035 and CHM-09, via inhibiting
EGFR signalling. Metastasis, a crucial phenomenon that occurs due to high migratory
potential of breast CSCs was further evaluated. A significant decrease in EGF-induced
migration was observed in presence of EGFR inhibitor, PD153035 and CHM-09 in dose
dependent manner (Fig 11A).
-
EGFR inhibition perturbs the mesenchymal differentiation phenotype of breast CSCs: CD24
+
/
CD44 -breast CSCs possess the mesenchymal phenotype that has inherent property to
differentiate into adipocytes, chondrocytes and osteoblasts when cultured in specific
differentiation mediums. As described in the methods, the cells grown in chondrocyte
differentiation medium stained positively with toluidine blue for detection of collagen type II.
EGF treated cells depicted significantly higher staining by toluidine blue as compared to
vehicle control (Fig 11B) that was inhibited by an increasing dose of CHM-09 (Fig 11B) or
PD153035 (5µM) (Fig 11C) as evaluated quantitatively (Fig 11D). Next, for the detection of
adipocytes, cells were stained with Oil red O stain which specifically stains the lipid droplets
present in the mature adipocytes. Breast CSCs treated with EGF depicted more number of
lipid droplets as compared to vehicle control (Fig 11E) that was reduced by increasing
concentration of CHM-09 (Fig 11E) or PD153035 at 5µM (Fig 11F) as analyzed
quantitatively (Fig 11G). Further, for osteoblast differentiation, breast CSCs were cultured in
This article is protected by copyright. All rights reserved.

osteoblast differentiation medium and stained with 5% silver nitrate (von Kossa staining) to
detect the mineral deposits. EGF treated cells showed more mineral deposits as compared to
vehicle control that was attenuated in presence of CHM-09 in a concentration dependent
manner (Fig 11H) or PD153035 (5µM) (Fig 11I) as assessed quantitatively (Fig 11J). These
results demonstrate an induction of EGFR signalling that maintains the mesenchymal
phenotype of breast CSCs was effectively reduced by CHM-09 or PD153035.
Inhibition of EGFR signalling induces MET and responsiveness to anticancer drugs in breast
CSCs: The epithelial marker, E-cadherin was down-regulated in EGF treated cells as
compared to control (Fig 12A). EGFR inhibition in presence of PD153035 or CHM-09 led to
a significant up-regulation of E-cadherin expression, indicating a gain of epithelial phenotype
(Fig 12A). The EGF-induced mesenchymal markers gene expression such as vimentin and
fibronectin were significantly down-regulated in presence of EGFR inhibition indicating a
loss of mesenchymal phenotype (Fig 12A). Interestingly, EMT-specific transcription factors,
TWIST1, ZEB2 along with adhesion molecules, ALCAM and chemokines, CXCR4 and
CXCR6 depicted a marked reversal of EGF-induced gene expression in presence of EGFR
inhibition (Fig 12A). Similarly at protein expression levels too, EGFR inhibition led to
marked decrease in vimentin and N-cadherin along with concomitant increase in E-cadherin
in breast CSCs sorted from MDA-MB-231 (Fig 12B, left panel) and BT-549 (Fig 12B, right
panel) as well as MDA-MB-453 (Fig 12C, left panel) and MDA-MB-468 (Fig 12C, right
panel). These results demonstrate that inhibition of EGFR signalling leads to reversal of EMT
or acquisition of MET by the breast CSCs. Finally, we evaluated the proliferation of our
sorted breast CSCs in presence of doxorubicin, a chemotherapeutic agent. Doxorubicin alone
could not inhibit the proliferation of the breast CSCs by more than 33-38%. However, when
these breast CSCs isolated from MDA-MB-231 (Fig 13A), BT-549 (Fig 13B), MDA-MB-453
This article is protected by copyright. All rights reserved.

(Fig 13C) and MDA-MB-468 (Fig 13D) were treated with doxorubicin along with EGFR
inhibitors,a significantly high percent of inhibition (~74%) was observed (Fig 13A). These
results suggest that inhibition of EGFR signalling leads to increased responsiveness of breast
CSCs to chemotherapeutics.
Discussion
Breast tumor exhibits different sub-types based on its molecular profiling such as, luminal,
(
type A and type B), HER-2 overexpressing, normal-like and basal-like breast cancer
TNBC) [23]. TNBC is most aggressive form of breast cancer that is associated with
(
increased rate of recurrence and poor patient survival [25]. TNBC cell lines are negative for
receptors - estrogen, progesterone and HER-2 while positive for basal markers -
cytokeratin5/6 and EGFR [26]. Our study correlated well with the existing literature that a
TNBC cell line, MDA-MB-231(basal) depicted high EGFR expression as compared to MCF-7
(luminal). Preclinical breast cancer models and clinical studies suggest that EGFR expression
is concomitant with aggressive phenotype and poor clinical outcome [27]. Furthermore,
basal-like tumors do not respond to conventional cancer treatments such as chemotherapy and
radiotherapy. The four classes of EGFR inhibitors used for cancer therapy are monoclonal
antibodies (cetuximab), antisense oligonucleotides, antibody-based immuno-conjugates and
small molecules (RTKs - erlotinib and gefitinib) [26]. These kinase inhibitors prevent auto-
phosphorylation of the intracellular tyrosine kinase domain of EGFR [28]. Standard cancer
chemotherapy drugs affects replication of cancer cells but EGFR inhibitors target the crucial
pathways of cancer cell growth and survival [29]. Due to lack of efficacious small molecule
inhibitors against EGFR with less side-effects, we have synthesized a chrysin analog, CHM-
0
9 with enhanced binding affinity to EGFR active site confirmed by molecular docking
studies and cellular thermal shift assays. Our observation matched with the previous reports
This article is protected by copyright. All rights reserved.

that cell viability of TNBC cells significantly reduced in presence of EGFR blockers [30].
Activation of EGFR signalling in TNBC plays a major role in cancer cell proliferation,
migration, EMT and metastasis via activation of downstream signalling cascades [31].
Chrysin analog, CHM-09 treated cells depicted a decreased activation of EGFR and its
downstream signalling as reported in the literature [7, 8].
Aggressive TNBC tumors possess a small population of cells with stem cell-like properties of
self-renewal and differentiation capabilities known as breast CSCs [32]. Apart from these,
literature suggests that breast CSCs also show enhanced migration and invasion properties
that aids to metastasis [33]. Metastatic human breast CSCs owing to stem cell properties
forms mammospheres [24]. The observations depicting large mammosphere forming
efficiency of our sorted breast CSCs suffices this important characteristic feature of self-
renewal capacity [34]. The breast CSCs bearing mesenchymal phenotype exhibit tri-lineage
differentiation ability into –adipocytes, chondrocytes and osteoblasts similar to mesenchymal
stem cells [35]. Cancer cells which are epithelial in nature are sensitive to chemotherapies
[
36]. The loss of epithelial phenotype (E-cadherin) and gain of mesenchymal phenotype
vimentin, N-cadherin and Fibronectin) leads to chemoresistance in breast CSCs due to
(
perturbation of important signal transduction pathways [37]. Thus, molecular targeted
therapies against specific signalling mediator holds high hope for survival of breast cancer
patients against metastatic relapse of disease [38]. Recent literature has been suggested that
chemotherapeutic drugs such as paclitaxel, salinomycin, doxorubicin and etoposide inhibit
tumor growth and metastasis of CSCs with limited efficiency [17] similar to our observation
of very low inhibition of proliferation in our sorted breast CSCs by doxorubicin. EGFR being
an important therapeutic target and expresses high in breast CSCs as compared to cancer cells
makes it a preferred molecular target to inhibit CSC metastasis and recurrence of cancer. Our
This article is protected by copyright. All rights reserved.

data demonstrating attenuation of EGFR signalling led to reversal of EMT or induction of
MET in the breast CSCs along with increased responsiveness to chemotherapeutics when
simultaneously treated with doxorubicin and EGFR inhibitors suggests an effective
therapeutic approach to suppress tumorigenicity and prevent metastasis in EGFR expressing
breast CSCs.
Methods
Isolation of chrysin: The air dried stem bark of Oroxylum indicum (a voucher specimen #
BD-02/07) has been deposited in the National Herbarium, Govt. of India, Shibpur, Howrah)
was crushed in to powder and extracted with cold methanol [39]. The dried methanol extract
was dissolved in 150 ml water and fractionated by solvent extraction using petroleum ether,
chloroform, n-butanol, successively (20 × 3 ml each). Both chloroform and n-butanol fraction
on column chromatography with silica gel (60-120 mesh, Merck) eluting with PE-EtOH (7:3)
gave light yellow crystals of chrysin (1.3g).
Chemical characterization of isolated chrysin: The melting point of the compound was
0
1
13
2
70 C. It was then characterised by H, C NMR spectral analysis and compared with the
spectral data of an authentic sample.
1
Spectral data of chrysin: HN-MR (400 MHZ, DMSO-d ): δ 12.80 (brs, 1H), 10.89 (brs, 1H),
6
8
.04 (dd, J=8.0, 2H), 7.59 (m, 1H), 7.54 (t-like, J= 8.0 Hz, 2H), 6.50 (d, J=2.0 Hz, 1H), 6.93
1
3
(
s, 1H), 6.19 (d, J=2.0 Hz, 1H), C-NMR 400 MHz, DMSO-d6): 181.8, 164.4, 163,2, 161.4,
+
1
57.4, 121.0, 136.9, 129.0 (2C), 126.36 (2C), 105.1, 104.0., EI-MS m/z 254 [M ].
This article is protected by copyright. All rights reserved.

Protein preparation and grid generation: The X-ray crystal structures of EGFR with low
resolution have been retrieved from RCSB protein data bank (www.rcsb.org).The EGF
receptors of PDB ID: 3W32, 3W33, 3POZ and 3W2R [40, 41] complexed with co-ligands
have been prepared using the ‘Protein Preparation Wizard’ [42]. Protein has been pre-
processed, optimized and minimized with force field of OPLS_2005. The prepared protein
was used for grid generation using the ‘Receptor Grid Generation’ panel. The grid box has
been defined with 15Å size from the centre of selected co-crystallized ligand.
Docking validation: The docking accuracy has been evaluated by calculating root mean
square deviation (RMSD). Lower the value of RMSD, higher will be the docking accuracy
and for precise glide docking the value should be < 1.0 [43]. The RMSD values were
calculated by superimposing best docked poses of co-ligands on its native co-ligand are
summarized in Table S1. The RMSD value of 3W32 was 0.8254, which has been found to be
the lowest and chosen for docking study. All the computational work were carried out using
HP Z820 Workstation running over CentOS 6.3.
Chemical synthesis of chrysin analog (CHM-09): For transformation of compound 1 to
compound 3, a mixture of 1 (0.100g, 0.393 mmole), t-butylbromoacetate (0.1228g,0.393
mmole), K CO (0.110g, 0.795mmole), and 20 ml dried acetone was stirred at room
2
3
temperature for 3h and the starting material 1 was disappeared [44]. The reaction mixture was
0
dried in rotary evaporator at 40 C followed by extraction with ether and the residue was
subjected to column chromatography over silica gel eluate PE-EtOH (9:1) gave compound 2
(0.1250g, yield: 86%) [45]. The compound was kept in vacuum desiccator for overnight.
Trifluro acetic acid was added drop by drop to compound 2 and dried in rotary evaporator for
several times and resulted compound 3 (0.0946 g, yield: 88.7%). To a solution of compound
This article is protected by copyright. All rights reserved.

3
(0.090g, 0.288mmol) in 15 ml of dried DMF, DCC (0.090g, 1.5eq) was added and the
solution was heated at 90-95°C. After 5 mins, dopamine (0.045g, 1eq), triethyl amine
0.045g, 1.5eq) and DMAP (1mg) were added to the solution and then continued heating at
0-100°C. After 24h the compound 3 disappeared on TLC from reaction mixture and the
(
9
mixture was dried under reduced pressure to a semi-solid mass. This was then suspended in
minimum volume (10 ml) of distilled water and extracted with ethyl acetate (3 × 10 ml). The
ethyl acetate extracts were mixed together, washed with brine and dried under reduced
pressure. The residue was purified using silica gel column chromatography and eluent
petroleum ether-ethylacetate (8:2 ratio) gave CHM-09(U) (0.065g yield: 43.51%) while
eluent petroleum ether-ethyl acetate (7:3 ratio) obtained CHM-09 (0.030g, yield: 23.28%).
The R value of the compound CHM-09 was 0.402 (petroleum ether:ethyl acetate, 9:1). The
f
purified compound was then characterised by physical and spectroscopic methods and was
consistent with the desired structure.
Spectral data of chrysin analog, CHM-09:N-(3,4-dihydroxyphenethyl)-2-(5-hydroxy-4-oxo-
2
-phenyl-4H-chromen-7-yloxy)acetamide (CHM-09):C H NO Mass (ESI-MS) m/z: 449
25 21 7;
1
[
M+2H]; H-NMR (400 MHz, DMSO-d ): δ 8.10 (2H, d, J= 6.8), 8.03 (1H, brs), 7.63–7.57
6 H
(5H, m), 7.00 (1H, s), 6.57 (1H, brs), 6.24 (1H, brs), 6.00 (brs, 2H), 4.94 (1H, brs), 4.23 (s,
1
3
2
1
1
H), 3.00-2.94 (2H, m), 2.16 (2H, t, J= 7.2, 7.2). C-NMR (300 MHz, DMSO-d ): δ 182.6,
6 C
66.1, 164.5, 164.0, 161.6, 157.3, 154.2, 152.9, 132.6, 132.2, 131.0 (2C), 129.6, 126.9 (2C),
22.3, 116.1, 115.2, 105.9, 105.6, 98.8, 94.1, 67.2, 41.13, 39.35.
Cytotoxic effect on proliferation: 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyl tetrazolium
bromide assay was performed to evaluate the cytotoxicity of chrysin or chrysin analog-
CHM-09, and positive controls such as EGFR inhibitors, AG1478 and PD153035 (Sigma
This article is protected by copyright. All rights reserved.

Aldrich, USA) at a concentration of 0.01 – 100 µM for 48h in breast cancer cell lines: MDA-
MB-231, MDA-MB-453, MDA-MB-468, BT-549, MCF-7 as well as control cells: HEK293
(ATCC, USA) cells as previously described [46]. Separately, cells were also treated with
combination of CHM-09 and pathway blockers such as wortmannin, PD98059 or static
(
Sigma Aldrich, USA). IC of all thetreated drugs and compounds were calculated using
50
GraphPad prism6.
Cellular thermal shift assay (CETSA): MDA-MB-231 cells were treated with DMSO, EGF
25ng/ml), CHM-09, PD153035 and AG1478 at a concentration of 25µM for 1h at 37°C.
(
Cells were trypsinized and harvested in 0.5mL PBS supplemented with protease inhibitors.
Samples were then heated at different temperatures such as 38, 42, 46, 50 and 54°C for 3 min
followed by immediate 2 cycles of freeze/thaw in liquid nitrogen and at 25°C as described
earlier [19]. Cell lysates were centrifuged at 20,000g for 20 min at 4°C and supernatant with
soluble protein fraction were subjected to SDS-PAGE and transferred onto a PVDF
membrane, probed with antibodies EGFR and tubulin (loading control).
Effect on gene expression: Total RNA extracted from MCF-7 and MDA-MB-231 cells was
used for cDNA synthesis (Thermo Scientific, USA). Human gene-specific forward and
reverse primers of various growth factor receptors, apoptotic and cell cycle genes were
utilized for semi-quantitative (RT-PCR) and quantitative real-time RT-PCR (qRT-PCR,
Applied Biosystems, USA) analysis as described previously [46]. Amplification of eukaryotic
1
8S rRNA and β-Actin was used as an internal control. All the target gene expression was
normalized to internal control and depicted as the –fold change.
This article is protected by copyright. All rights reserved.

Effect on protein expression: Cells treated with CHM-09 and EGFR inhibitors at different
concentrations for 48h were lysed and total protein isolated was subjected to SDS–PAGE,
transferred to PVDF membrane and probed with various primary antibodies followed by
incubation with HRP-conjugated secondary antibodies (Abcam, USA) as described earlier
[
47]. Visualization was performed by using G: Box (Syngene, USA).
BrdU Cell cycle analysis: MDA-MB-231 cells treated with chrysin, CHM-09, AG1478 and
PD153035 at 10µM concentration for 24h followed by labelling with BrdU at a final
6
concentration of 10µM/10 cells for 24h. Cells were fixed with 5mL of ice cold ethanol and
incubated at 4°C for overnight. Ethanol fixed cells were centrifuged and cell pellet was re-
dissolved in 2N HCL/Triton X-100 and incubated for 30 min to denature the DNA. Next,
cells were spin down and cell pellet was re-dissolved in 0.1M sodium borate (Na B O , pH
2
4
7
8
/
.5) to neutralize the sample followed by re-suspension in 50µl of 0.5% Tween 20 / 1% BSA
6
PBS. Cells were incubated with BrdU antibody (1µg/10 cells) for 1h and then with
secondary anti-mouse Alexa Flour 488 for 30 mins at room temperature. Cells were
resuspeneded in 0.2mL PBS contain propidium iodide (20µg/mL) and samples incubated in
dark for 30 min followed by analysis using flow cytometer (FlowSight, Amnis, USA) as
described earlier [48].
Effect on cellular apoptosis: Apoptosis assay was performed using Annexin-FITC staining
kit (Biolegend, USA) of MDA-MB-231 cells treated with chrysin, CHM-09, AG1478 and
PD153035 at 10µM concentration for 48h as described earlier [46].
Isolation of breast CSCs by Magnetic Activated Cell Sorting (MACS): Breast CSCs were
isolated from MDA-MB-231, MDA-MB-453, BT-549 and MDA-MB-468 using MACS
(Miltenyi Biotec, Singapore) as described previously [49] based on absence and presence of
This article is protected by copyright. All rights reserved.

CD24 and CD44 cell surface markers, respectively. Cells were subjected to immunodepletion
-
by CD24 microbeads according to manufacturer’s protocol. The CD24 population were
-
+
+
further subjected to CD44 selection to isolate CD24 /CD44 (breast CSCs). The sorted CD24
-
+
and CD24 /CD44 cell population were plated for subsequent experiments in presence and
absence of CHM-09 and PD153035 for cell viability assays and gene and protein expression
analysis as mentioned above.
+
-
+
Immunofluoroscence analysis: The sorted CD24 and CD24 /CD44 cell population were
plated on the glass coverslips in a 6-well plate for 24h at 37°C. Cells were then fixed with
methanol for 20 mins and stained with CD24 and CD44 antibodies for 2h at room
temperature. Post incubation, cells were stained with fluorochrome-conjugated secondary
antibodies and images were procured using confocal microscope (Fluoview FV10i, Olympus,
Australia).
Purity check of sorted populations: Expression of CD24 and CD44 in the sorted cell
populations were evaluated using immunofluoroscence analysis under confocal microscopy
+
-
as described earlier [49]. To confirm the purity of sorted CD24 and CD24
+
/
CD44 populations, cells were stained with CD44-PE and CD24-FITC (Biolegend, USA)
monoclonal antibodies for 30 min at 4°C using flow cytometry analysis carried out with BD
FACS Canto (Becton Dickonson, USA) instrument and data was analyzed using FCS Express
6
software as previously described [49].
+
-
Aldefluor assay: Expression of Aldehyde dehydrogenase (ALDH) in CD24 and CD24
+
/
CD44 populations sorted from MDA-MB-231 was assessed using flow cytometry analysis
of ALDH-based cell detection kit according to manufacturer’s protocol (StemCell
This article is protected by copyright. All rights reserved.

+
-
+
Technologies, USA). Briefly, to the sorted CD24 and CD24 /CD44 populations
6
(
1×10 cells/mL), a fluorescent non-toxic substrate for ALDH, BODIPY amino acetaldehyde
was added that diffuses into the intact viable cells. The activity of ALDH is directly
proportional to the amount of fluorescent reaction product. A specific inhibitor of ALDH,
diethyl amino benzaldehyde (DEAB) was used in the flow cytometry analysis as a control
that shows the specificity of the enzymatic assay as well as background fluorescence
correction for the flow cytometry analysis [50].
-
+
Effect on mammosphere forming capacity of breast CSCs: CD24 /CD44 cells were plated at
4
a density of 1×10 cells/well in 1 mL of mammosphere specific medium in pHEMA (poly
hydroxyl ethyl methacrylate, Sigma-Aldrich, USA) coated plates as described previously
[
51]. Cells were treated with PD153035 (5µM) or CHM-09 at a concentration of 0.1, 1,
0µM in presence/absence of EGF (10ng/ml). After 7 days, the images of mammospheres
1
were taken at 10× magnification using a microscope (Olympus IX71, Japan). The size of
spheres was measured using NIH Image J software (USA).
-
+
Effect on breast CSC differentiation: The CD24 /CD44 sorted population was seeded in 24-
well plate at a density of 500 cells/well and treated with PD153035 (5µM) or CHM-09 at a
concentration of 0.1 and 1µM in presence/absence of EGF (10 ng/mL). The cells were
maintained in adipocyte differentiation medium (Dexamethasone - 1µM, IBMX - 0.5µM and
insulin - 1000ng/ml) for 14 days. For Chondrocyte and Osteoblast differentiation, cells were
maintained in Chondrocyte differentiation medium (Dexamethasone – 0.1µM, Ascorbate-2-
phosphate - 50µM, ITS – 500µl, sodium pyruvate - 1mM, TGF-β-3 - 1µl, proline -
0
.25mg/ml) and Osteoblast differentiation medium (Dexamethasone – 0.1µM, Ascorbate-2-
phosphate - 50uM, β-glycerophosphate – 10mM) for 21 and 28 days, respectively. Post
This article is protected by copyright. All rights reserved.

differentiation, media was removed and stained with 300µl of Oil red ‘O’ stain for Adipocyte
differentiation, Toludine blue for Chondrocyte differentiation and Von Kossa staining for
Osteoblast differentiation as described earlier [49]. Images were taken at 10× magnification.
Statistical analysis: Results depicted in mean ± standard error of the mean of three or more
independent experiments, which were performed intriplicates. Photomicrographs represent
typical experiments reproduced at least thrice with similar results. Quantitative analysis
performed using NIH image J software (USA). Statistical analyses were performed to
determine the difference between treatment group and their respective controls, using one-
way ANOVA in GraphPad prism version 6.05 and student’s paired-t testwith p values ≤0.05
considered significant.
Acknowledgements
AD acknowledges the funding provided by DBT, Govt. of India, Cancer Pilot Project “GAP-
0
514, Grant Sanction No. 6242-P65/RGCB/PMD/DBT/AMTD/2015” and CSIR, Ministry of
th
Science and Technology, Govt of India, XII Five-year Plan Project # CSC-0111.SD
acknowledges the funding by DBT, Govt. of India,F.No.BT/327/NE/TBP/2012. Fellowships
provided by DBT, ICMR and CSIR are gratefully acknowledged by KM (DBT-SRF), ND
(
ICMR-SRF) and RY (CSIR-JRF). We also acknowledge the technical assistance provided
by Mr. Suresh Y in assisting flow cytometric analysis and confocal imaging at CSIR-IICT.
This work has pending patent application.
Author Contributions
AD – Conceived, designed and conceptualised the study, performed data analysis, manuscript
writing and figures; SD –Chemical structure design and synthesis and manuscript writing;
This article is protected by copyright. All rights reserved.

KM – Performed the experiments, data analysis and manuscript writing; NRD, RY, KG, SS –
Performed biological experiments; TD – Performed synthesis of compounds and UCD –
Analysis of synthesized compounds.
References
1
. Davis NM, Sokolosky M, Stadelman K, Abrams SL, Libra M, and Candido S,
Nicoletti F, Polesel J, Maestro R, Assoro AD, Drobot L, Rakus D, Gizak A, Laidler P,
Litewka JD, Basecke J, Mijatovic S, Ivannic DM, Montalto G, Cervello M, Fitzgerald
T, Demidenko Z, Martelli AM, Cocco L, Steelman LS & McCubrey JA (2014)
Deregulation of the EGFR/PI3K/PTEN/AKT/mTORC1 pathway in breast cancer:
possibilities for therapeutic intervention. Oncotarget 5, 4603-4650.
2
3
. Takeuchi K & Ito F (2011) Receptor tyrosine kinases and targeted cancer
therapeutics. Biol Pharm Bull 34, 1774-1780.
. Wendt MK, Jason A, Smith BS & Schiemann WP (2010) Transforming Growth
Factor-β-induced Epithelial-Mesenchymal Transition facilitates epidermal growth
factor-dependent breast cancer progression. Oncogene 29, 6485-6498.
4
5
. Hicklin DJ & Ellis LM (2005) Role of the vascular endothelial growth factor pathway
in tumor growth and angiogenesis. J Clin Oncol 23, 1011-1027.
. Ball SG, Shuttleworth A & Kielty C (2012) Inhibition of platelet-derived growth
factor receptor signalling regulates oct4 and nanog expression, cell shape, and
mesenchymal stem cell potency. Stem Cells 30, 548-560.
6
7
. Wang J, Barnes RO, West NR, Olson M, Chu JE & Watson PH (2008) Jab1 is a target
of EGFR signalling in ERα- negative breast cancer. Breast Cancer Res 10, 1-11.
. Christopher P, Michael D, Michael S, Settleman J, James FL & David J R (2011)
ErbB2 is necessary for ErbB4 ligands to stimulate oncogenic activities in models of
human breast cancer. Genes & Cancer 2, 792-804.
8
9
. Corkery B, Crown J, Clynes M & Donovan N (2009) Epidermal growth factor
receptor as a potential therapeutic target in triple-negative breast cancer. Annals Oncol
2
0, 862–867.
. Rimawi F, Shetty B, Weiss H, Schiff R, Osborne K, Chamness G & Elledge M (2010)
EGFR expression in breast cancer association with biologic phenotype and clinical
outcomes. Cancer 116, 1234-1242.
This article is protected by copyright. All rights reserved.

1
1
0. John T, Tony T, Agarwal A, Djakiew D & Thompson W (1999) Epidermal growth
factor promotes MDA-MB-231 breast cancer cell migration through a
phosphatidylinositol 3-kinase and phospholipase c-dependent mechanism. Cancer Res
5
9, 5475–5478.
1. Zhang D, Tari A, Akar U, Arun B, LaFortune A, Nieves-Alicea R & Ueno NT (2010)
Silencing kinase interacting stathmin gene enhances erlotinib sensitivity by inhibiting
serine 10 p27 phosphorylation in EGFR expressing breast cancer. Mol Cancer Ther 9,
3
090-3099.
1
1
1
1
2. Choowongkomon K, Sawatdichaikul O, Songtawee N & Limtrakul J (2010)
Receptor-based virtual screening of EGFR kinase inhibitors from the NCI diversity
database. Molecules 15, 4041-4054.
3. Kasala E, Boddulure BL, Madana RM, Athira K V, Gogoi R & Barua C(2015)
Chemopreventive and therapeutic potential of chrysin in cancer: mechanistic
perspectives. Toxicol Lett 233, 214-225.
4. Walle T, Otake Y, Brubaker JA, Walle UK & Halushka PV (2001) Disposition and
metabolism of the flavonoid chrysin in normal volunteers. Br J Clin Pharmacol 51,
1
43–146.
5. Zheng H, Li S, Pu Y, Lai Y, He B & Gu Z (2014) Nanoparticles generated by PEG-
chrysin conjugates for efficient anticancer drug delivery. Eur J Pharm Biopharm 7,
4
54-460.
6. Liu S & Wicha MS (2010) Targeting breast cancer stem cells. J Clin Oncol 28, 4006-
012.
1
1
4
7. Gupta B, Tamer T, Jiang, G, Tao K, Kuperwasser C, Weinberg A & Lander E (2009)
Identification of selective inhibitors of cancer stem cells by high-throughput
screening. Cell 138, 645-659.
1
1
2
8. Switzer CH, Glynn SA, Cheng RYS, Ridnour LA, Green JE, Ambs S & Wink DA
(2012) S-Nitrosylation of EGFR and Src activates an oncogenic signalling network in
human basal-like breast cancer. Mol Cancer Res 10, 1203-1215.
9. Jafari R, Almqvist H, Axelsson H, Ignatushchenko M, Lundback T, Nordlund P &
Molina DM (2014) The cellular thermal shift assay for evaluating drug target
interactions in cells. Nature Protocol 9, 2100-2122.
0. Phuc V Pham PV, Phan NLC, Nguyen N, Truong NH, Duong TT, Le DV, Truong
KD & Phan NK (2011) Differentiation of breast cancer stem cells by knockdown of
CD44: promising differentiation therapy. J Translational Med 9, 1-13.
This article is protected by copyright. All rights reserved.

2
2
1. Fang D, Hawke D, Zheng Y, Xia Y, Meisenhelder J, Nika H, Mills GB, Kobayashi R,
Hunter T & Lu Z (2007) Phosphorylation of β-catenin by AKT promotes β-catenin
transcriptional activity. J Biol Chem 282, 11221-11229.
2. Marotta LLC, Almendro V, Marusyk A, Shipitsin M, Schemme J, Walker SR,
Qimron NB, Kim JJ, Choudhury SA, Maruyama R, Wu Z, Gonen M, Mulvey LA,
Bessarabova MO, Huh SJ, Silver SJ, Kim SY, Park SY, Lee HE, Anderson KS,
Richardson AL, Nikolskaya T, Nikolsky Y, Liu S, Root DE, Hahn WC, Frank DA &
Polyak K (2011) The JAK2/STAT3 signalling pathway is required for growth of
+
-
CD44 CD24 stem cell-like breast cancer cells in human tumors. J Clin Invest 121,
723-2735.
2
2
2
3. Tischkowitz M, Brunet JS, Begin LR, Huntsman DG, Cheang MCU, Akslen LA,
Nielsen TO & Foulkes WD (2007) Use of immunohistochemical markers can refine
prognosis in triple negative breast cancer. BMC Cancer 7, 1-11.
4. Grimshaw M, Cooper L, Papazisis K, Coleman JL, Bohenkamp HR, Stanke LC,
Papadimitriou JT & Burchell JM (2008) Mammosphere culture of metastaticbreast
cancer cells enriches for tumorigenic breast cancer cells. Breast Cancer Res 10, 1-10.
5. Ovcaricek T, Frkovic SG, Matos E, Mozina B & Borstnar S (2011) Triple negative
breast cancer -prognostic factors and survival. Radiol Oncol 45, 46-52.
2
2
6. Bauer KR, Brown M, Cress RD, Parise CA & Caggiano V (2007) Descriptive
analysis of estrogen receptor (ER)-negative, progesterone receptor (PR)-negative, and
her2-negative invasive breast cancer, the so-called triple-negative phenotype. Cancer
1
09, 1721-1728.
7. Teng YHF, Tan WJ, Thike AA, Cheok PY, Tse GMK, Wong NS, Yip GW, Bay BH
Tan PH (2011) Mutations in the epidermal growth factorreceptor (EGFR) gene in
2
2
&
triple negative breast cancer: possible implications for targeted therapy. Breast
Cancer Res 13, 1-9.
8. Bachawal SV, Wali VB & Sylvester PW (2010) Enhanced antiproliferative and
apoptotic response to combined treatment of gamma-tocotrienol with erlotinib or
gefitinib in mammary tumor cells. BMC Cancer 10, 1-13.
2
3
9. Ciardiello F & Tortora G (2001) A novel approach in the treatment of cancer:
targeting the epidermal growth factor receptor. Clin Cancer Res 7, 2958-2970.
0. Lacouture ME (2006) Mechanisms of cutaneous toxicities to EGFR inhibitors.
Cancer 6, 803–812.
This article is protected by copyright. All rights reserved.

3
3
1. Yang L, Li Y, Ding Y, Choi KS, Kazim AL & Zhang Y (2013) Prolidase directly
binds and activates epidermal growth factor receptor and stimulates downstream
signalling. J Biol Chem 288, 2365-75.
2. Jin W, Chen B, Li J, Zhu H, Huang M, Gu SM, Wang QQ, Chen JY, Yu S, Wu J &
Shao ZM (2012) TIEG1 inhibits breast cancer invasion and metastasis by inhibition of
epidermal growth factor receptor (EGFR) transcription and the egfr signalling
pathway. Mol Cell Biol 32, 50-63.
3
3. Stratford A, Habibi G, Astanehe A, Jiang H, Hu K, Park E, Shadeo A, Buys TP, Lam
W, Pugh T, Marra M, Nielsen TO, Klinge U, Mertens PR, Aparicio S & Dunn SE
(2007) Epidermal growth factor receptor (EGFR) is transcriptionally induced by the
Y-box binding protein-1 (YB-1) and can be inhibited with Iressa in basal-like breast
cancer, providing a potential target for therapy. Breast Cancer Res 9, 1-14.
3
3
4. Alam M, Rajabi H, Ahmed R, Jin C & Kufe D (2014) Targeting the MUC1-C
oncoprotein inhibits self-renewal capacity of breast cancer cells. Oncotarget 5, 2622-
2
634.
5. Battula VL, Evas KW, Hollier BG, Shi Y, Marini FC, Ayyanan A, Wang RY, Brisken
C, Guerra R, Andreeff M & Mani SA (2010) Epithelial-mesenchymal transition-
derived cells exhibit multi-lineage differentiation potential similar to mesenchymal
stem cells. Stem Cells 28, 1435-1445.
3
3
6. Sajithlal G, Rothermund K, Zhang F, Dabbs DJ, Latimer JJ, Grant SG & Prochownik
EV (2010) Permanently blocked stem cells derived from breast cancer cell lines. Stem
Cells 28, 1008-1018.
7. Fillmore CM & Kuperwasser C (2008) Human breast cancer cell lines contain stem-
like cells that self-renew, give rise to phenotypically diverse progeny and survive
chemotherapy. Breast Cancer Res 10, 1-13.
3
3
4
8. Qin L, Liu Z, Chen H & Xu J (2009) The steroid receptor coactivator-1 regulates
twist expression and promotes breast cancer metastasis. Cancer Res 69, 3819-3827.
9. Dinda B, Mohanta BC, Arima S, Sato N & Harigaya Y (2007) Flavonoids from the
stem-bark of Oroxylum indicum. Nat Prod Sci 13,190-194.
0. Aertgeerts K, Skene R, Yano J, Sang BC, Zou H, Snell G, Jennings A, Iwamoto K,
Habuka N, Hirokawa A, Ishikawa T, Tanaka T, Miki H, Ohta Y & Sogabe S(2011)
Structural analysis of the mechanism of inhibition and allosteric activation of the
kinase domain of HER2 protein. J Biol Chem 286, 18756-18765.
This article is protected by copyright. All rights reserved.

4
1. Kawakita Y, Seto M, Ohashi T, Tamura T, Yusa T, Miki H, Iwata H, Kamiguchi
H, Tanaka T, Sogabe S, Ohta Y & Ishikawa T (2013) Design and synthesis of novel
pyrimido [4,5-b]azepine derivatives as HER2/EGFR dual inhibitors. Bioorg. Med.
Chem 21, 2250–2261.
4
4
2. Schrödinger Suite (2013) Protein Preparation Wizard; Epik version 2.5, Schrödinger,
LLC, New York, NY.
3. Loving K, Salam NK & Sherman W (2009) Energetic analysis of fragment docking
and application to structure-based pharmacophore hypothesis generation. J Comp
Aided Mol Des 23, 541–554.
4
4
4
4. Christian AGN, Montalbetti & Falque V (2005) Amide bond formation and peptide
coupling. Tetrahedron 61, 10827–10852.
5. Hu K, Wang W, Cheng H, Pan S & Ren J, (2011) Synthesis and cytotoxicity of novel
chrysin derivatives. Med Chem Res 20, 838-846.
6. Geesala R, Gangasani JK, Budde M, Balasubramanian S, Vaidya JR & Das A (2016)
2
Azetidinones: Synthesis and biological evaluation as potential anti-breast cancer
agents. Eur J Med Chem 124, 554-558.
4
7. Das A, Fernandez-zapico ME, Cao S, Yao J, Fiorucci S, Hebbel RP, Urrutia R &
Shah V (2006) Disruption of a sp2/klf6 repression complex by SHP is required for
farnesoid x receptor-induced endothelial cell migration. J Biol Chem 281, 39105-
3
9113.
4
4
5
8. Wei Lin D, Chung BP & Kaiser P (2014) S-adenosylmethionine limitation induces
p38 mitogen-activated protein kinase and triggers cell cycle arrest in G1. J Cell Sci
1
27, 50-59.
9. Dhoke NR, Kalabathula E, Kaushik K, Geesala R, Sravani B & Das A (2016) Histone
deacetylases differentially regulate the proliferative phenotype of mouse bone marrow
stromal and hematopoietic stem/progenitor cells. Stem Cell Res 17, 170-180.
0. Jauffret E, Ginestier C, Iovino F, Tarpin C, Diebel M, Esterni B, Houvenaeghel G,
Extra JM, Bertucci F, Jacquemier J, Xerri L, Dontu G, Stassi G, Xiao Yi, Barsky S,
Birnbaum D, Viens P & Wicha MS (2010) ALDH1-positive cancer stem cells
mediate metastasis and poor clinical outcome in inflammatory breast cancer. Clin
Cancer Res 16, 45-55.
5
1. Vinnedge LMP, Mcclaine R, Wagh PK, Brokamp KAW, Waltz SE & Wells SI (2011)
The human DEK oncogene stimulates beta catenin signalling, invasion and
mammosphere formation in breast cancer. Oncogene 30, 2741-2752.
This article is protected by copyright. All rights reserved.

Table 1. The RMSD values of different EGFR
Superposition of docked co-ligand on its originally
bound native conformation of co-ligand in X-ray
ligand-enzyme complex
Sl.
No
PDB
ID
Resolution
[Å]
RMSD
1.4900
1
3POZ
1.50
2
3
4
3W2R
3W32
3W33
2.05
1.80
1.70
2.3100
0.8254
1.1633
This article is protected by copyright. All rights reserved.