Design and discovery of silybin analogues as antiproliferative compounds using a ring disjunctive – Based, natural product lead optimization approach

Article abstract of DOI:10.1016/j.ejmech.2017.03.033

The present study reports the synthesis and anticancer activity evaluation of twelve novel silybin analogues designed using a ring disjunctive-based natural product lead (RDNPL) optimization approach. All twelve compounds were tested against a panel of cancer cells (i.e. breast, prostate, pancreatic, and ovarian) and compared with normal cells. While all of the compounds had significantly greater efficacy than silybin, derivative 15k was found to be highly potent (IC₅₀?

Full text of DOI:10.1016/j.ejmech.2017.03.033

Accepted Manuscript
Design and discovery of silybin analogues as antiproliferative compounds using a ring
disjunctive – Based, natural product lead optimization approach
Elangovan Manivannan, PhD, Haneen Amawi, Noor Hussein, Chandrabose
Karthikeyan, Aubry Fetcenko, N.S. Hari Narayana Moorthy, Piyush Trivedi, Amit K.
Tiwari, PhD
PII:
S0223-5234(17)30191-5
10.1016/j.ejmech.2017.03.033
EJMECH 9293
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 29 December 2016
Revised Date: 10 February 2017
Accepted Date: 15 March 2017
Please cite this article as: E. Manivannan, H. Amawi, N. Hussein, C. Karthikeyan, A. Fetcenko, N.S.H.
Narayana Moorthy, P. Trivedi, A.K. Tiwari, Design and discovery of silybin analogues as antiproliferative
compounds using a ring disjunctive – Based, natural product lead optimization approach, European
Journal of Medicinal Chemistry (2017), doi: 10.1016/j.ejmech.2017.03.033.
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
Graphical Abstract
Described herein is design and synthesis of
novel silybin derivatives using natural lead
optimization and ring disjunctive approach
leading to discovery of a lead compound
15k, which produces apoptosis in ovarian
cancer cells through tubulin inhibition

ACCEPTED MANUSCRIPT
Design and discovery of silybin analogs as antiproliferative compounds using a
ring disjunctive – based, natural product lead optimization approach
Elangovan Manivannana,*, Haneen Amawib,#, Noor Husseinb, Chandrabose
Karthikeyanc, Aubry Fetcenkob, N.S. Hari Narayana Moorthyc, Piyush Trivedic, Amit K.
Tiwarib,*
aSchool of Pharmacy, Devi Ahilya Vishwavidyalaya, Indore, (MP) 452001, India
bDepartment of Pharmacology and Experimental Therapeutics, College of Pharmacy &
Pharmaceutical Sciences, University of Toledo, OH, USA
cSchool of Pharmaceutical Sciences, Rajiv Gandhi Proudyogiki Vishwavidyalaya, Airport
Bypass Road, Gandhi Nagar, Bhopal (MP) 462036, India
Running title: Novel silybin analogs display potent tubulin inhibition
# Note: First two authors contributed equally to this work.
*Corresponding Authors:
Amit K. Tiwari, PhD
Department of Pharmacology and Experimental Therapeutics,
College of Pharmacy & Pharmaceutical Sciences,
University of Toledo, Ohio 43614
Tel: 419-383-1913
Fax: 419-383-1909
E-mail: amit.tiwari@utoledo.edu
Elangovan Manivannan, PhD
Medicinal & Pharmaceutical Chemistry,
School of Pharmacy, Devi Ahilya Vishwavidyalaya,
Indore, M.P., India 452001
Tel: +91-7509114754
E-mail: drmanislab@gmail.com
1

ACCEPTED MANUSCRIPT
Abstract
The present study reports the synthesis and anticancer activity evaluation of twelve
novel silybin analogues designed using a ring disjunctive-based natural product lead
(RDNPL) optimization approach. All twelve compounds were tested against a panel of
cancer cells (i.e. breast, prostate, pancreatic, and ovarian) and compared with normal
cells. While all of the compounds had significantly greater efficacy than silybin,
derivative 15k was found to be highly potent (IC50<1 µM) and selective against ovarian
cancer cell lines, as well as other cancer cell lines, compared to normal cells.
Preliminary mechanistic studies indicated that the antiproliferative efficacy of 15k was
mediated by its induction of apoptosis, loss of mitochondrial membrane potential and
cell cycle arrest at the sub-G1 phase. Furthermore, 15k inhibited cellular microtubules
dynamic and assembly by binding to tubulin and inhibiting its expression and function.
Overall, the results of the study establish 15k as a novel tubulin inhibitor with significant
activity against ovarian cancer cells.
Keywords: Silybin, ring disjunctive natural product lead optimization, (E)-3-(3-
(benzyloxy)phenyl)-1-phenylprop-2-en-1-one,
chemoprevention
tubulin
inhibitor,
2

ACCEPTED MANUSCRIPT
Introduction
Natural products research has help elucidate previously unknown biochemical
mechanisms, while identifying several clinically effective anticancer drugs, such as
paclitaxel, doxorubicin, vinblastine, and vincristine [1]. Silymarin is a polyphenolic
flavonoid complex found in the dried fruits of the milk thistle plant (Silybum marianum)
[2-4]. Over the last few decades, the pharmacological properties of different
components of silymarin i.e. silybin A (1), silybin B (2), isosilybin A (3), isosilybin B (4),
silychristin (5), isosilychristin (6), silydianin (7) and (+)-taxifoline (8) as shown in (Figure
1) have been extensively investigated [5-13]. Of these aforementioned compounds,
silybin was found to be the most potent antiproliferative component of silymarin and it
has been reported to inhibit numerous cancers, including breast [14, 15], colon [16],
liver [17, 18], lung [19], ovary [20], prostate [21-23], and skin cancer [24, 25]. Silybin’s
cytotoxic and chemopreventive efficacy results from concomitantly targeting multiple
important cancer targets [17]. For example, silybin 1) produces cell cycle arrest at both
the G1/ and G2 phase;s 2) activates Erb1 signalling; 3) induces cyclin-dependent
kinase (CDK) inhibition; 4)down-regulates specific anti-apoptotic proteins; , 5) inhibits
kinases required for cell survival and 6) inhibits inflammatory transcription factors [21,
24, 26-31]. Furthermore, silybin produces antimetastatic activity by targeting
dysregulated stemness pathways, such as the Wnt/beta-catenin/epithelial to
mesenchymal transition (EMT). Both in vitro and in vivo studies indicate that silybin has
an excellent safety profile, with no significant induction of toxicity or carcinogenesis at
doses up to 1200 mg daily [32]. Although silybin, a polyphenolic natural product, has
anticancer efficacy in preclinical models, its pharmacokinetic limitations such as poor
absorption, rapid metabolism, and poor bioavailability has hampered its clinical use [33].
3

ACCEPTED MANUSCRIPT
Attempts to overcome these issues with structural modifications of the phenol and
alcohol functions of silybin [5, 34] through esterification [35], phosphorylation [36],
glycosidation [33], and oxidation of the C-23 alcohol [37] have had limited success.
Structural modification of a natural product lead (NPL) is a frequently used approach in
NPL based drug discovery. NPL optimization can be used to attain novel bioactive
molecules with improved pharmacological and PK properties, along with target
selectivity [38, 39]. The strategy of splitting fused ring structures of NPL by following a
'disjunctive approach', more often results in new bioactive molecules with a similar
mechanism of action and improved PD/PK [40, 41]. Herein, we report a NPL
optimization strategy consisting of sequential exclusion of functional groups, and a ring
disjunctive approach coupled with exploration of different functional groups on the
silybin core structure. Simplification of the silybin core structure by removing
functionality, chirality and ring opening, allowed us to identify the minimum
pharmacophore required for efficacy to inhibit cancer cell proliferation. This strategy
also resulted in the identification of an anticancer lead, (E)-3-(3-(benzyloxy) phenyl)-1-
phenylprop-2-en-1-one (15a). We further optimized 15a by substituting advantageous
functional groups to obtain more potent derivatives. The design, synthesis,
identification, SAR, and antiproliferative profile of these derivatives are reported herein.
In addition, additional pharmacodynamic mechanisms (cytotoxicity, anti-apoptosis and
tubulin inhibition) of the optimized molecule 15k are identified and reported herein.
4

ACCEPTED MANUSCRIPT
2. Results & Discussion
2.1 Chemistry of new silybin analogues
General methods for the preparation of chalcones involve the Claisen–Schmidt
condensation of appropriate aryl methyl ketones and aldehydes in the presence of
either an acid or base [42]. The compounds, 15a – 15k, were prepared by reacting 3-
(benzyloxy) benzaldehyde (12) with the appropriate acetophenones (14a – 14k), in
presence of dilute potassium hydroxide at room temperature (Scheme 1). Similar
reaction conditions were used for the synthesis of unsubstituted chalcone, (17) using
unsubstituted benzaldehyde (Scheme 2). Furthermore, all reactions resulted in the
formation of chalcone derivatives 15a – 15k and 17 in good yield as shown in Figure 2.
1
The structures of all the compounds were established by the use of IR, H-NMR, and
mass spectral data. The elemental analysis of the final compounds was found to be
within ± 0.3 % limit. The physico-chemical and spectral data of all the final compounds
15a – 15k have been furnished in materials and method section. In the IR spectra, the
absorption bands due to stretching frequency of olefinic double bonds appeared
between 1603 to 1592 cm-1 whereas the absorption bands due to aromatic stretch
appeared at 1600.8 and 1506.3 cm-1. The absorption bands due to carbonyl stretch
(C=O) appeared between 1679 to 1650 cm-1. The absorption bands due bending
vibrations of aliphatic and aromatic C–H appeared at 1467.7 and 827.4 cm-1,
respectively, in the spectrum which confirms the formation of all synthesized
1
compounds. H NMR spectra showed that typically E-isomers were generated in this
1
reaction.In the H NMR spectra of chalcones (15a – 15k), the protons of α, β-
unsaturated system absorbed as two doublets around δ 7.4 ppm for Ha and δ 7.9 for Hb
−8.5 ppm
with coupling constant J = 15−16 Hz. The aromatic hydrogens appear at δ 6.5
5

ACCEPTED MANUSCRIPT
depending on the substituent on the aromatic (Ar) group. The singlet signals at the
region between δ 7.5 and 8.5 ppm are attributed to the resonance of the 1,4-phenylene
ring hydrogens. A singlet for two protons of Ph–(CH2)–O– appeared downfield at δ5.12.
Furthermore the mass spectrum of all the synthesized compounds (15a – 15k) showed
the molecular ion peak i.e. m/z (M+1), which confirmed their synthesis. The functional
group transformation of –CHO into –CO–CH=CH– in compounds (15a – 15k) was
established on the basis of IR, 1H NMR and mass spectral data.
2.2 Natural product lead optimization of silybin
Silybin’s antiproliferative efficacy (IC50~80-222 µM) has been previously validated in
multiple studies [43-46]. In the present study, we selected silybin as a cytotoxic lead for
medicinal chemistry optimization. First, we characterized the detailed role of ring
structures and the respective hydroxyl groups of silybin. This helped guide the
systematic structural modification and lead optimization of silybin to obtain a more
potent proliferation inhibitor with specificity for neoplastic cells. Silybin is nearly a 1:1
mixture of two diastereomers, namely silybin A (1) and silybin B (2). Structurally, the
isomers of silybin represent a typical flavonolignan scaffold. Silybin A (1) is
characterized by five ring structures, with alternating carbo- and heterocycles (ring A to
ring E) and five hydroxyl groups. The five hydroxyl groups of silybin A (1) are the main
targets for derivatization [5]. These hydroxyl groups can be divided into three categories
according to their nature: three phenolic (two in A and one in E ring), one secondary (in
B Ring) and one primary alcoholic group (in D ring). For example, we considered the
antiproliferative efficacy of silybin and its analogues (9 and 10) in the breast cancer cell
6

ACCEPTED MANUSCRIPT
line, MCF-7, as previously reported by Zhao H et. al., to describe our disjunctive
approach based on a natural product lead optimization strategy [34]. Silybin inhibited
MCF-7 cell proliferation (IC50 = 222.8 ± 3.6 µM). The modification steps are described
in details in Figure 2. In brief, we first removed the two phenolic substituents of ring A in
silybin A 1, resulting in analogue 9 that was approximately 16-fold more potent than
silybin, with an IC50 value of 13.8 ± 3.6 µM in MCF-7 cells. The increase in efficacy
may be due to an increase in solubility. In the second step, we used a disjunctive
approach to open ring B of the silybin analogue 9, followed by removal of a hydroxyl
group and introduction of a double bond. The resulting silybin analogue 10, was only
slightly more potent (IC50 value of 12 ± 0.2 µM) than silybin in MCF-7 cells. In the third
step, we cleaved ring D and removed all hydroxyl and a methoxy groups to derive a
new silybin analogue, 15a, whose IC50 value (9.8 ± 3.0 µM) in MCF-7 cells was lower
than that for analogues 9 and 10. It should also be noted that 15a had differential
inhibitory efficacy on breast cancer proliferation in other breast cancer cell lines, i.e.
ZR-75-1 (IC50 = 8.4 ± 0.9 µM), BT-20 (IC50 = 25.5 ± 0.2 µM), and MDAMB-231 (IC50
=
19.7 ± 1.9 µM). In the fourth step, we cleaved ring E to derive a compound, 17, with an
IC50 value of 12.6 ± 6 µM in MCF-7 cells.
We ended our disjunctive approach at the third step and kept the 3-(3-
(benzyloxy)phenyl)-1-phenylprop-2-en-1-one backbone (15a) and systematically varied
the substituents of ring A of 15a with respect to the hydroxyl positions of silybin A (1).
This resulted in synthesis of compounds 15b – 15k, with some of these compounds
having electron withdrawing groups (EWG), such as 2-chloro, 2,4-dichloro, 2,3,4-
trichloro, 2-bromo, 3-bromo, 4-bromo and 4-nitro, followed by some electron donating
7

ACCEPTED MANUSCRIPT
groups (EDG) (4-methoxy 4-methyl and 4-hydroxy). The introduction of a 2-bromo
substitution on ring A of 3-(3-(benzyloxy)phenyl)-1-phenylprop-2-en-1-one 15h (IC50 =
1.6 ± 0.9 µM) on MCF-7 resulted in a ≈140-fold increase in cytotoxic activity compared
to silybin.
2.3 Antiproliferative efficacy and the structure-activity relationships for the novel
silybin analogues
In the process of optimizing the NPL silybin A (1) and designing novel analogues, we
prepared and evaluated a series of (E)-3-(3-(benzyloxy)phenyl)-1-phenylprop-2-en-1-
one analogues to determine their antiproliferative efficacy in eleven different cell lines:
three normal cell lines (normal epithelial colon cells: CRL-1459, human embryonic
kidney cells: HEK293/pcDNA.3.1, and Chinese hamster ovarian cells: CHO) and eight
cancer cell lines [human breast adenocarcinoma (MCF-7), human metastatic breast
cancer (ZR-75-1), human breast carcinoma (BT-20), human breast adenocarcinoma
(MDAMB-231), human prostate adenocarcinoma (DU-145), human pancreatic
adenocarcinoma (PANC1), human ovarian carcinoma (OV2008) and human ovarian
carcinoma (A2780) cells] using the MTT assay. The antiproliferative efficacy data are
presented as the IC50 values, defined as the concentration of compound that decreases
cell proliferation by 50% (Table 1). The IC50 values were generated from at least three
independent experiments.
Compounds 15a – 15f had IC50 values ranging from ~5 to 81 µM for different cancer
cells. The unsubstituted benzyloxy derivative 15a exhibited significant antiproliferative
8

ACCEPTED MANUSCRIPT
efficacy. Compound 15a was efficacious in inhibiting the proliferation of MCF-7, ZR-75-
1, BT-20, MDAMB-231, DU-145, PANC1, OV2008 and A2780 cancer cell lines, with
IC50 values of 9.8 ± 3.0, 8.6 ± 0.9, 25.5 ± 0.2, 20.0 ± 1.9, 42.4 ± 7.1, 23.5 ± 1.7, 4.9 ±
1.4, and 6.7 ± 0.7 µM, respectively. Additionally, the first set of synthesized compounds
(15a – 15f) exhibited moderate to good antiproliferative efficacy against all the cancer
cell lines tested. Also, it should be noted that the potency of most of the compounds
was generally greater for the cancer cells compared to the normal cell lines CRL-1459,
HEK293/pcDNA.3.1 and CHO.
The chalcone derivative 17 (without benzyloxy moiety) showed moderate
antiproliferative efficacy for the breast cancer lines MCF-7(IC50 = 12.6 ± 1.5 µM), ZR-75-
1 (IC50 = 13.7 ± 0.9 µM), BT-20 (IC50 = 21.1 ± 0.9 µM), and MDAMB-231, (IC50 = 6.7 ±
0.9 µM) It showed significantly lower inhibition for DU-145, PANC1 and OV2008, A2780
cancer cells, with IC50 values of 52.5 ± 3.1, 23.5 ± 1.7, 33.4 ± 2.1, and 67.4 ± 4.3
respectively. However, its counterpart benzyloxy derivative 15a, exhibited better
antiproliferative efficacy in both ovarian cancer cell lines OV2008 (IC50 = 4.9 ± 1.4 µM)
and A2780 (IC50 = 6.7 ± 0.7 µM). The structure−activity relationship (SAR) analysis
suggested that the 3-benzyloxy moiety enhanced the antiproliferative efficacy,
particularly in ovarian cancer cell lines. Therefore, it can be concluded that the
benzyloxy group (ring E) attached to 3-position of chalcone (ring D) is crucial for
increasing antiproliferative activity against breast and ovarian cancers. The present
SAR is in good agreement with the previously published reports indicating that the
benzyloxy group augments the antiproliferative or cytotoxic efficacy [47, 48].
9

ACCEPTED MANUSCRIPT
Interestingly, we found that altering the substituents at the 2- and 4-positions of ring A
(phenyl) of the silybin analogues could also significantly affect the antiproliferative
efficacy. By introducing EWGs on ring A, three compounds (15b, 15c, and 15d) were
generated that possessed improved efficacy compared to chalcone (17), except for
breast cancer cell lines. In fact, compounds 15b and 15c with EWGs displayed an
enhanced potency in comparison to that of 15d, whereas compounds 15e and 15f with
EDGs showed a decreased antiproliferative efficacy against all the cancer cell lines.
However, compound 15d, with a nitro substitution at the 4th position, was less
efficacious than the chloro substituted compounds 15b and 15c (Table 4). Similarly,
compound 15f, which has a polar 4-methoxy substitutent, was also lower
antiproliferative efficacy. However, the remaining compounds had significantly
increased potencies, suggesting that a concurrent increase in electronegativity and
hydrophobicity of the compounds increased the efficacy.
When the 4th position of ring A was substituted with small alkyl groups (methyl), the
efficacy was not significantly increased. Compound 15e, with a methyl substitution on
ring A, was moderately efficacious in inhibiting proliferation of MCF-7, ZR-75-1, BT-20,
MDAMB-231, DU-145, PANC1, OV 2008 and A2780 cancer cell lines, with IC50 values
of 21.7 ± 1.1, 25.8 ± 4.2, 61.0 ± 3.5, 15.6 ± 2.3, 25.6 ±3.5, 41.6 ± 1.6, 12.9 ± 0.9 and
20.1 ± 1.9 µM, respectively. However, it was more cytotoxic against HEK293/pcDNA.3.1
cells, with an IC50 value of 8.2 ± 0.3. Interestingly, the alkoxy derivative 15f was more
active on the breast cancer cell lines MCF-7, ZR-75-1, BT-20, MDAMB-231 and with
IC50 values of 6.0 ± 0.9, 19.9 ± 0.9, 38.3 ± 5.1, 12.4 ± 1.4 µM, respectively, than the
alkyl derivative, 15e, suggesting that the electronic influence of ring A significantly
10

ACCEPTED MANUSCRIPT
affects the breast cancer efficacy. Collectively, our result clearly indicate that the
hydrophobicity on ring A at the 2,4-position of the synthesized newer silybin analogs is
beneficial to their antiproliferative efficacy. It should be noted that most of the
compounds tested in this study were less toxic to normal cells compared to cancer cells,
suggesting that the silybin analogues were selectively toxic to cancer cells compared to
the non-malignant, normal cells. Furthermore, compounds 15a−15f displayed poor
antiproliferative efficacy for both prostate (DU-145) and pancreatic (PANC1) cancer cell
lines compared to their efficacy to inhibit breast cancer cells growth. Based on the
above results, the SAR were not immediately apparent because of the limited number of
compounds tested and the positions for the substituents yielded differential
antiproliferative efficacies. Owing to antiproliferative efficacy of 15a and 15f, they were
subjected to further medicinal chemistry optimization and SAR investigations to develop
more potent, selective and broad spectrum anticancer compounds. A set of four
compounds bearing EWGs, i.e. 2,3,4-trichloro 15g, 2-bromo 15h, 3-bromo 15i, 4-bromo
15j and a compound bearing an EDG 4-hydroxy, 15k, were synthesized and assessed
for their efficacy to inhibit tumor cell growth. As anticipated from the previous SAR
experiments, the newly synthesized compounds 15g−15k, had significant
antiproliferative efficacy (ranging from 0.3 – 9 µM) in all of the tested cancer cell lines
(except 15h for PANC1, with an IC50 = 61.2 µM).
The introduction of the chloro, bromo and hydroxyl groups at positions 2, 3, 4 of the ring
A of the lead compound 15a resulted in a significant improvement in antiproliferative
efficacy (compare 15a−15f with 15g−15k). Compound 15g, having 2,3,4-trichloro
substitutions, had significant antiproliferative efficacy in all the breast cancer cell lines
11

ACCEPTED MANUSCRIPT
compare to dichloro (15c) and mono chloro (15b) substituted compounds. We observed
that when the number of chlorine atom increases in ring A of 15a, the anti-cancer
efficacy was found to also increase. Similarly, a significant increase in antiproliferative
efficacy was observed for prostate cancer (DU-145; IC50 = 6.4 ± 0.5) and pancreatic
cancer cells (PANC1; IC50 = 2.4 ± 0.4). In pancreatic cancer cell, the most potent
compounds were 15g and 15j, with IC50 values 10-fold greater than the lead compound,
15a. In particular, in this subset (15g − 15k), the 2-bromo derivative, 15h, had
significant efficacy in MCF-7, ZR-75-1, BT-20, and MDAMB-231, breast cancer cell
lines, with IC50 values of 1.6 ± 0.9, 2.9 ± 0.1, 1.8 ± 0.9, and 1.9 ± 0.4 µM, respectively.
The 4-bromo derivative, 15j, displayed a higher antiproliferative efficacy against all type
of cancers. Its meta counterpart, the 3-bromo compound, 15i, had low efficacy within
this subset (15g − 15k), indicating that a meta substitution on ring A of the lead 15a
minimizes antiproliferative efficacy. A bromo substitution was well tolerated on the 2 and
4 positions of ring A, was bromo substitution was not at the 3-position (compare 15h,
15i, 15j). Compound 15h, had IC50 values up to 6-fold higher than 15a and 142 fold
higher than silybin in MCF-7 cells. Overall, the replacement of a chlorine with bromine
led to the 2-bromo 15h, 3-bromo 15i, and 4-bromo 15j derivatives, which were more
potent in all types of cancer cell lines (IC50 ~ 1 to 9 µM) (except 15h for PANC1, with
an IC50 ~ 61 µM), indicating that the larger bromine is preferred over chlorine. The
exchange of the EWG group for a hydrophilic hydroxyl substituent in the 4-position, i.e.
15k, was found to be beneficial for antiproliferative efficacy in MCF-7, ZR-75-1, BT-20,
and MDAMB-231 breast cancer cell lines with IC50 values of 2.8 ± 0.9, 4.9 ± 0.2, 5.1 ±
2.1, and 2.7 ± 0.6 µM, respectively. In general, our results indicated that derivatives
12

ACCEPTED MANUSCRIPT
carrying the EWGs at the 2nd and 4th positions on ring A of lead 15a were typically the
most efficacious against all of the cancer cell lines. .
2.4 The 15k has significant efficacy and selectively in inhibiting the growth of
ovarian cancer cell lines.
With regard to the antiproliferative efficacy on ovarian cancer cells, compound 15a had
an IC50 value of 4.9 ± 1.4 µM for OV2008 cells and an IC50 value of 6.7 ± 0.7 µM for
A2780 cells. Specifically, compound 15b, with a 4-chloro substitution, has weak
antiproliferative efficacy in both ovarian cell lines (OV2008; IC50 = 20.0 ± 2.3 µM,
A2780; IC50 = 56.41 ± 2.78 µM). However, the 3,4- dichloro derivative, 15c, was almost
equipotent in OV2008 cells (IC50 = 5.2 ± 1.1 µM), but less efficacious for A2780 (IC50 =
21.4 ± 3.2 µM), compared to compound 15a. The presence of three chlorine groups on
the A-ring at the 2, 3, 4 position led to compound 15g (OV2008; IC50 = 1.5 ± 0.9 µM and
A2780; IC50 = 1.52± 0.02 µM), which had cytotoxic efficacy, but did not show any
selectivity for normal cell lines. The insertion of a 4-nitro group on phenyl ring A led to
derivative 15d, that had low efficacy (up to 50 µM) toward both ovarian cell lines,
whereas with an alkyl moiety (compound 15e), the cytotoxic efficacy in the these cell
lines was retained, although its efficacy is somewhat lower (OV2008; IC50 = 12.85 ±
0.93 µM and A2780; IC50 = 20.45 ± 1.90 µM) compared to compound 15c. In
comparison to compounds 15b, 15c (chloro substitution) 15h, 15i, and 15j (bromo
substitution) had greater efficacy in inhibiting ovarian cancer cell growth, The 2-bromo
derivative, 15h, was almost equipotent in both ovarian cancer cell lines (OV2008; IC50 =
1.16 ± 0.42 µM and A2780; IC50 = 2.48± 0.03 µM), compared to the 4-bromo derivative
compound, 15j, but had similar cytotoxicity towards normal cell lines (CRL-1459; IC50 =
13

ACCEPTED MANUSCRIPT
2.4 ± 0.9 µM and HEK293; IC50 = 2.2 ± 0.0 µM). However, the 3-bromo derivative
compound, 15i, was less efficacious in ovarian cancer cell lines. The ortho and para
isomers were 5-fold more efficacious than the meta isomer. Further, compounds without
a benzyloxy substituent on ring E (compound 17) were significantly less cytotoxic in
ovarian cancer cells.
The antiproliferative efficacy data (Table 3) clearly indicates the significance of the
hydroxyl derivative, 15k, which was almost 6-fold more efficacious in OV2008 and
A2780 cells (OV2008; IC50 = 0.83 ± 0.1 µM and A2780; IC50 = 0.97± 0.1 µM) compared
to compound 15a. Compound 15k had a 9-10 fold greater IC50 value for normal cell
lines compared to ovarian cancer cell lines (Figure 3A-3B). The 4-position substituent
of ring A considerably affected the inhibitory efficacy in ovarian cancer cells, with the
hydroxyl and bromo substituents being the most favorable structural feature. The
inhibition of ovarian cancer cells growth is encouraging as few compounds (especially
compound 15k) have achieved both sub micromolar potency and selectivity. Hence,
compound 15k was chosen for mechanism - based experiments on to determine how it
inhibits ovarian cancer cell growth.
Following the incubation of ovarian cancer cells with different concentrations (0, 1, 2, or
4 µM), 15k produced significant morphological changes in all the cells (see Figure 3A
for representative images). We also determined if compound 15k produced a time-
dependent inhibition of ovarian cancer cell growth. Ovarian cancer cells were incubated
with different concentrations of 15k (0, 1, 2, 4, and 8 µM) and pictures were taken every
two hours using the IncuCyte ZOOM® System (Biocompare, San Francisco, CA, USA)
for 48h. The proliferation of ovarian cancer cells and confluence over time are
14

ACCEPTED MANUSCRIPT
presented and quantified in Figures 3C and 3D. These data indicated that the
antiproliferative efficacy of 15k occurs within 12h and continues to inhibit the growth and
proliferation of these cells over time (Figures 3E and 3F).
2.5 The 15k strongly induces apoptosis and induces sub-G1 cell cycle arrest in
ovarian cancer cell lines
The induction of apoptosis, a form of programmed cell death, that produces (cell death,
is one of the most important mechanisms by which several anticancer drugs exert their
therapeutic effects [49].The apoptotogenic potential of a compound can be determined
by its efficacy to produce a loss of the mitochondrial membrane potential, resulting in
the release of several apoptogenic factors, including cytochrome c, among other
mediators, in the apoptosis signalling pathway [49]. For example, during apoptosis,
phosphatidylserine (PS) typically flips over to the extracellular side from its otherwise
normal, intracellular mitochondrial localization [50]. The exposed PS readily binds to the
fluorophore labelled human vascular anticoagulant protein (annexin V), with significant
affinity, indicating the magnitude of apoptosis induction [51]. Compound 15k produced
significant apoptosis in both the OV2008 and A2780 ovarian cancer cell lines (Figure
4). In OV2008 cells, incubation treatment with 15k for 12 h produced a loss of
mitochondrial membrane potential (Figure 4A) and increased the population of cells
undergoing apoptosis compared to control from 20% to 38% and 43% at 2 µM and 4
µM, respectively, (p value <0.001; Figure 4B). Similarly, the incubation of A2780 cells
with 15 k only up to 12 h produced significant apoptosis and a loss of the mitochondrial
membrane potential (p value <0.001, Figure 4A). As shown in Figure 4B, 2 and 4 µM
of 15k, respectively, produced a significant shift in the percentage of apoptosed cells
15

ACCEPTED MANUSCRIPT
(37% and 38%, respectively) in quadrant II compared to 0% in the control group.
Collectively, these results confirm the ability of this silybin derivative to reduce the
mitochondrial membrane potential and induce apoptotic cell death.
To further asses the efficacy of compound 15k to inhibit ovarian cancer cells
proliferation, cell cycle analysis was conducted. The flow cytometry results indicated
that compound 15k increases the percentage of cells in sub G1 phase in a
concentration - dependent manner (Figure 4C and 4D). The percentage of cells in the
sub G1 phase increased from 11.3% control cells, to 25.5% and 34% with 2 and 4 µM
of 15k,respectively (p value <0.01 and 0.001, respectively). Additionally, 15k
significantly (p <0.001) decreased the percentage of cells in the G1 phase from 79% in
the control cells to 58% and 45% after incubation with 2 and 4 µM, respectively, of 15k
(Figures 4C and 4D). However, 15k at both concentrations (2 and 4 µM) did not
significantly affect the percentage of cells in the S or G2 phases (Figures 4C and 4D).
Thus, the antiproliferative efficacy of 15k on ovarian cancer cells results, in part, from
inducing cell cycle arrest at the sub-G1 phase.
2.6 15k inhibits α- tubulin expression in ovarian cancer cells
Tubulin is a major cellular protein and an essential component of the cellular
cytoskeleton that polymerizes to form microtubules [4]. These microtubules are
essential for cellular division, mitosis, intracellular transport, cell shape and polarity [52].
Both α- and β-tubulin polymerize into microtubules to form the mitotic spindle [53].
Accordingly, cell division takes place when these spindles move the chromosomes to
the polar side of the cell [54]. Thus, alterations in the structure and/or function of
microtubules can affect cell proliferation and survival [5]. Carcinogenesis has been
16

ACCEPTED MANUSCRIPT
reported to be correlated with an increase in the polymerization and de-polymerization
of tubulin [8]. Furthermore, a number of anticancer drugs have been developed to
target and disrupt microtubular dynamics and prevent mitotic spindle formation, thereby
inhibiting cell division and inducing cell cycle arrest and apoptosis [55]. Our silybin
derivatives have a chalcone moiety in their structure which is known for tubulin inhibition
and anticancer efficacy [56]. Thus, we determined the effect of 15k on α- tubulin
dynamic.
When 15k was incubated overnight with OV2008 cancer cells at different concentrations
(0, 2 µM and 4 µM), α-tubulin expression was significantly decreased (p <0.001 for both
concentrations) compared to control OV2008 cancer cells (Figure 5A). Also, using
immunofluorescence, we validated the efficacy of 15k on microtubule dynamics in
ovarian OV2008 cancer cells. As shown in Figure 5B, in the control OV2008 cancer
cells, there was a significant magnitude of green fluorescence, indicative of well –
organized, spindle like microtubules arrayed along the long axes in the cytoplasm
around the nucleus. In contrast, incubation of OV2008 cancer cells with 15k (2 and 4
µM) produced a significant disturbance in cellular microtubule shape and integrity. For
example, the cells shrunk and became rounded and the microtubules no longer had a
long spindle shape. Additionally, 15k, at 2 and 4 uM, induced a significant reduction (p
<0.001) in α- tubulin fluorescence intensity compared to control cells.
Molecular docking studies were performed in order to determine if 15k was binding with
a-tubulin. The colchicine binding pocket of tubulin (PDB ID: 1SA0, 3.58 Å) was used as
a binding site for docking calculations as several chalcones that inhibit the assembly of
tubulin are known to bind to this site [55, 57]. The ligands were docked into the
17

ACCEPTED MANUSCRIPT
colchicine binding site of a-tubulin using Glide (Grid-Based Ligand Docking with
Energetics) program of the Schrödinger molecular modeling suite (Schrödinger, Inc.,
New York, NY, 2012) [58]. Colchicine and 15k were used for docking (Figure 5C).
Initially, the docking protocol was validated through redocking DAMA-colchicine
extracted from the X-ray structure to give a docked conformation very similar to the
observed X-ray crystallographic conformation (root mean square deviation (rmsd)
<0.69). Subsequently, the conformational library of compound 15k was docked at the
colchicine binding site of tubulin and the best fit conformation was selected on the basis
of Glide score and visual inspection. Figure 5D shows the docked conformation of 15k
in the colchicine-binding site of tubulin. In this pose, the chalcone moiety aligns planar
to the ring A and C of colchicine in the colchicine-binding site of tubulin Figure 5E. The
phenyl ring B of chalcone overlaps ring A of colchicine, forming hydrophobic contacts
with Lys352β, Val181α, Asn258β, Met259β and Ala316β, whereas the phenyl ring A of
the chalcone moiety is stabilized in the binding cavity through hydrophobic interactions
with Ile378β, Val238β, Thr239β, Leu242β, Ala250β and Leu255β. Furthermore, the
carbonyl oxygen of the enone group in chalcone moiety forms a hydrogen bond with the
sulfhydryl group of Cys241β. This is significant because colchicine shows a similar
interaction with the tubulin[59]. Another hydrogen bond was also observed between the
phenolic 'OH' in the para position phenyl ring B and nitrogen of Leu255β. These two
hydrogen bonds play a crucial role in stabilizing the orientation of chalcone moiety of
15k in the colchicine-binding site. The unsaturated intermediate chain between rings A
and B of chalcone moiety is also positioned in the vicinity of the side-chain of Ala316β
and Leu255β. The phenyl ring C of 15k forms a 'cis' like configuration relative to ring B
of the chalcone moiety, establishing hydrophobic contacts with Ala250β, Lys254β and
18

ACCEPTED MANUSCRIPT
Leu255β. Collectively, Western blot analysis (Figure 5A), immunofluorescence (Figure
5B) and molecular docking analysis (Figure 5C-E) indicate that 15k interacts with
tubulin with high affinity and that it might exert its anticancer activity through lowering
the expression and function of α-tubulin thus producing inhibition of microtubule
assembly and cellular mitosis in the tested ovarian cancer cell lines. Thus, α-tubulin
could be a major target for novel silybin derivative 15k.
3. Conclusion
Here we report the design and synthesize of twelve novel derivatives of silybin using
ring disjunctive based natural product lead optimization approach. All the compounds
synthesized were found to be more effective anticancer agents compared to silybin on
the cancer cell lines tested, hence validating our approach. The lead compound 15k,
that was identified in this study was found to be effective (IC50 < 1µM, for OV2008 and
A2780 cell lines) that’s over 30-200 fold compared to silybin (IC50 ~ 30-200 µM). Even
more interesting are the findings that 15k is 6-30 folds more selective to ovarian cancer
cells compared to other cancer and normal cells. In the preliminary investigation of
15k’s mechanism of action, we found that 15k arrests the cell cycle at sub-G1 phase,
resulting in loss of mitochondrial membrane potential and thus inducing apoptosis.
Further, 15k binds effectively to the tubulin binding pocket , decreasing the expression
and inhibiting tubulin protein function, suggesting that 15k could be a promising novel
inhibitor of tubulin. Further mechanistic studies in cell lines followed by appropriate
animal models as proposed here are required to determine the exact mechanisms of
action of 15k. Particularly, the antiproliferative effects of 15k on the advanced cancers,
its ability to reverse metastasis as well as multidrug resistance and its effectiveness as
19

ACCEPTED MANUSCRIPT
chemoadjuvant are currently being evaluated in our laboratory and the findings will be
reported in due course.
4. Experimental
4.1. Chemistry
All reagents and solvents were obtained from commercial sources and used without
further purification unless otherwise specified. The progress of reactions was monitored
by thin layer chromatography (TLC) on precoated silica gel F254 plates (Merck) and the
spots were visualized by exposure to UV light and/or Iodine vapor. All compounds were
obtained in moderate to good yields and purified by column chromatography using
Merck 60–120 mesh silica gel. Melting points were determined using an Electrothermal
Melting point apparatus and are uncorrected. IR spectra were recorded on a Perkin-
1
Elmer FT-IR spectrometer. H NMR spectra were obtained with a Bruker AM 300 and
400 MHz spectrometer. The chemical shifts are reported in parts per million (δ)
downfield using tetramethylsilane as internal standard and CDCl3 or deturiated DMSO
as the solvent. Spin multiplicities are given as s (singlet), d (doublet), m (multiplet) and q
(quartet). Coupling constants (J values) were measured in hertz (Hz). ESI MS spectra
were recorded on Micromass Quattro II triple quadrupole mass spectrometer.
Combustion analyses were performed with Carlo-Erba CHNS-O EA 1108 Elemental
Analyser. All the compounds gave satisfactory combustion analysis results (C, H, N
within 0.3% of calculated values). The starting material 3-Benzyloxy benzaldehyde was
prepared according to the procedure reported in the literature.
20

ACCEPTED MANUSCRIPT
4.1.1 Synthesis of 3-(benzyloxy)benzaldehyde (13)
A suspension of 3-hydroxy benzaldehyde 11, benzyl chloride 12, K2CO3 and KI in 50 ml
of EtOH was refluxed for 4.5 hrs. After cooling the solid thus obtained was filtered and
washed several times with H2O. The crude product was purified by column
chromatography.
Yield, 79%; dull white solid; m.p. 77–78˚C; Rf = 0.61.
4.1.2. General Procedure for synthesis of various substituted (E)-3-(3-
(benzyloxy)phenyl)-1-phenylprop-2-en-1-ones (15a–15k)
Equimolar mixture of 3-(benzyloxy)benzaldehyde (13) and substituted acetophenones
(14a–14k) was reacted in the presence of 50% (w/v) KOH/H2O using ethanol as a
solvent. The reaction mixture was stirred at room temperature for 8–24 hr. The reaction
mixture was poured into crushed ice (250 g) and left for few hours. To this
solution/suspension 1N HCl (10 mL), was added until the solution was acidic. The
resulting crude solid precipitate was collected by filtration and dried. Purification was
done by recrystallization in CH3OH or a mixture of CH3OH and EtOAc with acceptable
yields ranging from 70–85%.
(E)-3-(3-(benzyloxy)phenyl)-1-phenylprop-2-en-1-one (15a)
Yield, 85%; dull white solid; mp.81–82˚C; IR νmax (KBr, cm-1) 1664 (C=O), 1596 (C=C),
1249, 1023 (C–O), 847;1H NMR (400 MHz, CDCl3) δ 8.02 (d, J = 7.2 Hz, 2H), 7.78 (d, J
= 16 Hz, 1H), 7.74 (d, J = 16 Hz, 1H), 7.66 (t, J = 7.2 Hz, 1H), 7.51–7.61 (m, 5H), 7.32-
7.49 (m, 5H), 7.05 (d, J = 7.6 Hz, 1H), 5.12 (s, 2H); 13C NMR (100 MHz, CDCl3) δ
190.51, 159.17, 144.68, 138.20, 136.67, 136.35, 132.83, 130.01, 128.69, 128.65,
128.52, 128.14, 127.52, 122.44, 121.44, 117.12, 114.49, 70.19; ESIMS (m/z): 315
(M+1); Anal. Calculated for C22H18O2: C, 84.05; H, 5.77; Found: C, 84.08; H, 5.75.
. (E)-3-(3-(benzyloxy)phenyl)-1-(4-chlorophenyl)prop-2-en-1-one (15b)
21

ACCEPTED MANUSCRIPT
Yield, 73%; white solid; mp.79–80˚C; IRνmax (KBr, cm-1) 1679 (C=O), 1592 (C=C), 1258,
1037 (C–O), 828 (Ar);1H NMR (400 MHz, CDCl3) δ 7.96 (d, J = 7.2 Hz, 2H), 7.85 (d, J =
16 Hz, 1H), 7.74 (d, J = 16 Hz, 1H), 7.66 (t, J = 7.2 Hz, 1H), 7.32–7.51 (m, 4H), 7.23-
7.25 (m, 5H), 7.06 (d, J = 7.6 Hz, 1H), 5.12 (s, 2H); 13C NMR (100 MHz, CDCl3) δ
189.12, 159.19, 145.18, 139.27, 136.62, 136.48, 136.14, 130.06, 129.93, 128.97,
128.70, 128.16, 127.52, 121.84, 121.49, 117.27, 114.58, 70.21; ESIMS (m/z): 349
(M+1); Anal. Calculated for C22H17ClO2: C, 75.75; H, 4.91; Found: C, 75.77; H, 4.95.
(E)-3-(3-(benzyloxy)phenyl)-1-(2,4-dichlorophenyl)prop-2-en-1-one (15c)
Yield, 76%; white solid; mp.64–65˚C; IRνmax (KBr, cm-1) 1669 (C=O), 1594 (C=C), 1250,
1032 (C–O), 826 (Ar);1H NMR (400 MHz, CDCl3) δ 7.51 (d, J = 7.2 Hz, 2H), 7.48 (d, J =
16 Hz, 1H), 7.45 (d, J = 16 Hz, 1H), 7.36 (t, J = 7.2 Hz, 1H), 7.30–7.36 (m, 4H), 7.03-
7.17 (m, 4H), 6.99 (d, J = 7.6 Hz, 1H), 5.09 (s, 2H); 13C NMR (100 MHz, CDCl3) δ
192.49, 159.20, 146.36, 137.44, 137.00, 136.55, 135.69, 132.41, 130.47, 130.23,
130.10, 128.69, 128.61, 127.48, 127.33, 126.16, 121.65, 117.80, 114.45, 70.16; ESIMS
(m/z): 383 (M+1); Anal. Calculated for C22H16Cl2O2: C, 68.94; H, 4.21; Found: C, 68.97;
H, 4.24.
(E)-3-(3-(benzyloxy)phenyl)-1-(4-nitrophenyl)prop-2-en-1-one (15d)
Yield, 70%; pale yellow solid; mp.127–128˚C; IRνmax (KBr, cm-1) 1668 (C=O), 1595
1
(C=C), 1248, 1026 (C–O), 841 (Ar); H NMR (400 MHz, CDCl3) δ 8.34 (d, J = 7.2 Hz,
2H), 8.12 (d, J = 7.2 Hz, 2H), 7.83 (d, J = 16 Hz, 1H), 7.46 (d, J = 16 Hz, 1H), 7.34-7.39
(m, 3H), 7.26–7.31 (m, 5H), 7.07 (d, J = 7.6 Hz, 1H), 5.12; 13C NMR (100 MHz, CDCl3)
δ 188.97, 159.24, 150.11, 146.66, 143.01, 136.53, 135.72, 130.18, 129.42, 128.71,
128.21, 127.50, 123.88, 121.67, 121.63, 117.74, 114.76, 70.24; (s, 2H);ESIMS (m/z):
360 (M+1); Anal. Calculated for C22H17NO4: C,73.53; H, 4.77; N, 3.90; Found: C, 73.52;
H, 4.79; N, 3.93.
(E)-3-(3-(benzyloxy)phenyl)-1-p-tolylprop-2-en-1-one (15e)
Yield, 80%; white solid; mp.102–104˚C; IRνmax (KBr, cm-1) 1651 (C=O), 1603 (C=C),
1248, 1026 (C–O), 807 (Ar); 1H NMR (400 MHz, CDCl3) δ 7.94 (d, J = 16 Hz, 1H), 7.85
(d, J = 7.2 Hz, 2H), 7.73 (d, J = 7.2 Hz, 2H), 7.52 (d, J = 16 Hz, 1H), 7.26–7.47 (m, 8H),
7.02 (d, J = 7.6 Hz, 1H), 5.11 (s, 2H); 2.44 (s, 3H); 13C NMR (100 MHz, CDCl3) δ
189.94, 159.17, 144.22, 143.70, 136.70, 136.47, 135.62, 129.99, 129.36, 128.69,
128.14, 127.54, 122.45, 121.40, 116.99, 114.48, 70.19, 21.71; ESIMS (m/z): 329 (M+1);
Anal. Calculated for C23H20O2: C,84.12; H, 6.14; Found: C, 84.13; H, 6.18.
(E)-3-(3-(benzyloxy)phenyl)-1-(4-methoxyphenyl)prop-2-en-1-one (15f)
Yield, 81%; white solid; mp.136–138˚C; IRνmax (KBr, cm-1) 1650 (C=O), 1599 (C=C),
1251, 1023 (C–O), 801 (Ar); 1H NMR (400 MHz, CDCl3) δ 8.02 (d, J = 7.2 Hz, 2H), 7.73
22

ACCEPTED MANUSCRIPT
(d, J = 16 Hz, 1H), 7.52 (d, J = 16 Hz, 1H), 7.24-7.46 (m, 8H), 6.9–7.06 (m, 3H), 5.11 (s,
2H); 3.89 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 188.67, 163.47, 159.15, 143.79,
136.70, 136.55, 131.09, 130.84, 129.96, 128.68, 128.13, 127.53, 122.25, 121.34,
116.86, 114.44, 113.87, 70.19, 55.52; ESIMS (m/z): 345 (M+1); Anal. Calculated for
C23H20O3: C,80.21; H, 5.85; Found: C, 84.13; H, 6.18.
(E)-3-(3-(benzyloxy)phenyl)-1-(2,3,4-trichlorophenyl)prop-2-en-1-one (15g)
Yield, 86%;yellow solid; mp.134–136˚C; IRνmax (KBr, cm-1)1669 (C=O),1661 (C=C),
1253, 1028 (C–O),804 (Ar); 1H NMR (400 MHz, CDCl3)δ 7.50 (d, J = 7.2 Hz, 1H), 7.34-
7.44 (m, 5H), 7.26–7.33 (m, 3H), 7.15–7.16 (m, 2H), 7.00–7.06 (m, 2H), 5.09 (s, 2H);
13C NMR (100 MHz, CDCl3) δ 192.16, 159.21, 147.05, 139.33, 136.50, 136.05, 135.47,
132.92, 131.40, 128.69, 128.65, 128.17, 127.47, 126.98, 126.01, 121.71, 118.03,
114.45, 70.17; ESIMS (m/z): 417 (M+1); Anal. Calculated for C22H15Cl3O2: C, 63.26; H,
3.62; Found: C, 63.29; H, 3.66.
(E)-3-(3-(benzyloxy)phenyl)-1-(2-bromophenyl)prop-2-en-1-one (15h)
Yield, 89%;white solid; mp.84–86˚C; IRνmax (KBr, cm-1)1672 (C=O), 1659 (C=C),1258,
1
1021 (C–O),807 (Ar); H NMR (400 MHz, CDCl3) δ 7.64 (d, J = 7.2 Hz, 2H), 7.28-7.43
(m, 8H), 7.08–7.16 (m, 3H), 7.01 (d, J = 7.6 Hz, 2H), 5.07 (s, 2H); 13C NMR (100 MHz,
CDCl3) δ 194.65, 159.18, 146.45, 141.13, 136.58, 135.84, 133.47, 131.43, 130.06,
129.22, 128.68, 128.15, 127.49, 127.39, 126.43, 121.63, 119.53, 117.69, 114.39, 70.14;
ESIMS (m/z): 393 (M+1); Anal. Calculated forC22H17BrO2:C, 67.19; H, 4.36;Found: C,
67.15; H, 4.39.
(E)-3-(3-(benzyloxy)phenyl)-1-(3-bromophenyl)prop-2-en-1-one (15i)
Yield, 78%;dull white solid; mp.110–112˚C; IRνmax (KBr, cm-1)1668 (C=O), 1639
(C=C),1254, 1021 (C–O),805 (Ar);1H NMR (400 MHz, CDCl3) δ8.12 (t, 1H), 7.92-7.89
(d, J = 7.2 Hz, 1H), 7.74-7.78 (d, J = 7.2 Hz, 1H), 7.63-7.66 (dd, 1H), 7.19-7.45 (m,
10H), 7.02–7.05 (dd, 1H), 5.10 (s, 2H); 13C NMR (100 MHz, CDCl3) δ 197.04, 159.19,
145.56, 139.98, 136.94, 136.05, 135.67, 131.53, 131.22, 130.26, 130.09, 129.81,
128.60, 128.19, 127.60, 127.54, 126.88, 123.01, 120.09, 117.44, 114.59, 70.22; ESIMS
(m/z): 393 (M+1); Anal. Calculated forC22H17BrO2:C, 67.19; H, 4.36;Found: C, 67.17; H,
4.33.
(E)-3-(3-(benzyloxy)phenyl)-1-(4-bromophenyl)prop-2-en-1-one (15j)
Yield, 76%; dull white solid; mp.112–114˚C; IRνmax (KBr, cm-1)1658 (C=O),1599
(C=C),1251, 1026 (C–O),807 (Ar);1H NMR (400 MHz, CDCl3) δ7.88 (d, J = 7.2 Hz, 2H),
δ7.78 (d, J = 16 Hz, 1H), δ7.64 (d, J = 7.2 Hz, 2H),7.22–7.45 (m, 9H), 7.05 (d, J = 7.6
Hz, 1H), 5.11 (s, 2H); 13C NMR (100 MHz, CDCl3) δ 189.32, 159.19, 145.25, 136.90,
136.63, 136.13, 131.96, 130.05, 128.70, 128.17, 127.95, 127.52, 121.81, 121.50,
23

ACCEPTED MANUSCRIPT
117.30, 114.58, 70.21; ESIMS (m/z): 393 (M+1);Anal. Calculated for C22H17BrO2: C,
67.19; H, 4.36; Found: C, 67.22; H, 4.38.
(E)-3-(3-(benzyloxy)phenyl)-1-(4-hydroxyphenyl)prop-2-en-1-one (15k)
Yield, 84%; yellow solid; mp.129–131˚C; IRνmax (KBr, cm-1)1659 (C=O),1598 (C=C),
1
1254,1022 (C–O),808 (Ar); H NMR (400 MHz, DMSO) δ 8.01 (d, J = 7.2 Hz, 2H),7.75
(d, J = 16 Hz, 1H),7.67 (d, J = 16 Hz, 1H), 7.28-7.48 (m, 7H), 6.9–7.04 (m, 3H), 5.14 (s,
2H); 5.02 (s, 2H); 13C NMR (100 MHz, DMSO-d6) δ 191.89, 158.75, 142.44, 137.42,
136.56, 136.14, 130.79, 129.95, 128.92, 128.17, 127.70, 127.49, 127.30, 122.70,
121.39, 116.68, 69.52 ESIMS (m/z): 331 (M+1); Anal. Calculated for C22H18O3: C,
79.98; H, 5.49; Found: C, 79.95; H, 5.52.
4.1.3. Procedure for synthesis of unsubstituted (E)-Chalcone (17)
Equimolar mixture of benzaldehyde (16) and substituted acetophenone (14a) was
reacted in the presence of 50% (w/v) NaOH/H2O using ethanol as a solvent. The
reaction mixture was stirred at room temperature for 2 hr. The reaction mixture was
poured into crushed ice (250 g) and left for few hours. To this solution/suspension 1N
HCl (10 mL), was added until the solution was acidic. The resulting crude solid
precipitate was collected by filtration and dried. Purification was done by
recrystallization in CH3OH with excellent yields.
Yield, 85%; Yellow solid; mp.56–58˚C; IRνmax (KBr, cm-1) 1661 (C=O), 1605 (C=C),
1248, 1020 (C–O), 746 (Ar); 1H NMR (400 MHz, CDCl3) δ 8.01 (m, 2H), 7.92 (d, J = 16
Hz, 1H), 7.77 (d, J = 16 Hz, 1H), 7.55 (m, 8H).ESIMS (m/z): 209 (M+1); Anal. Calcd for
C15H12O: C,86.51; H, 5.81; Found: C, 86.50; H, 5.84.
4.2. Biological studies
4.2.1 Reagents:
The antibodies α- tubulin and β- actin were purchased from Cell Signalling Technology
(Danvers, MA, USA). A Mitochondrial Membrane Potential/Annexin V apoptosis Kit,
Propidium iodide dye were purchased from life technologies (Eugene, Oregon, USA).
24

ACCEPTED MANUSCRIPT
Dulbecco’s modified Eagle medium (DMEM) was purchased from GE Healthcare Life
Sciences, HyClone Laboratories (Logan, Utah, USA).
4.2.2. Cell lines and cell culture
A panel of different cancer cell lines, including breast (MCF7, ZR-75-1, MDAMB231,
BT20), ovarian (A2789, OV2008), pancreatic (PANC1), and prostate (DU-145) cancer
cell lines, as well as normal cell lines (Chinese hamster ovarian: CHO; human epithelial
normal colon: CRL 1459; human embryonic kidney: HEK293), were grown as adherent
monolayers in flasks with DMEM supplemented with 10% fetal bovine serum (FBS) in a
humidified incubator with 5% CO2 at 37°C.
4.2.3 Cell cytotoxicity by MTT assays and morphological analysis:
The (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) (MTT) assay was
used to determine the cytotoxicity of the twelve silybin derivative compounds in the
above mentioned cell lines. Briefly, cells were harvested with 0.25% trypsin, 2.21 mM
EDTA, 1X from Corning (Corning, NY 14831USA) and suspended at a final
concentration of 5x103 cells/well.
Cells were seeded (180µL/well) into 96-well
multiplates. To the seeded cells eight different concentrations of each derivative in the
series (20 µL) were added at 0, 0.1, 0.3, 1, 3, 10, 30, and 100 µM. After 68h of
incubation, 20 µL of MTT solution (4 mg/mL) was added to each well, and the plates
were incubated for 4h. This allowed viable cells to convert the yellow-colored MTT into
dark-blue formazan crystals. Subsequently, the medium was discarded, and 100 µL of
25

ACCEPTED MANUSCRIPT
DMSO was added into each well to dissolve the formazan crystals. The absorbance
was determined at 570 nm with a DTX 880 multimode detector (Beckman Coulter life
sciences, IN, USA). The IC50 ± SD concentrations were calculated from three
experiments performed in triplicate. The IC50 values were calculated from the cell
survival percentages obtained for each compound tested at eight different
concentrations. Similarly, the cytotoxicity of the test compounds was compared to the
normal cell lines (CHO, CRL 1459 and HEK293).
4.2.4 Confluence with time analysis
OV2008, A2780 cells were seeded in low density (1x103) cell/ well in 96 well plates.
Compound 15k was added in different concentrations (0, 2, 4, and 8) µM and incubated
in the IncuCyte ZOOM® System (Essen BioScience, Inc., Ann Arbor, MI, USA). This
live cell imaging system allowed the imaging of the cells every two hours for up to 48 h.
The cells confluences at each time point were analyzed using the IncuCyte associated
analysis system. A blue confluence mask was added to the cells and timeline curves
were drawn for each concentration. The data is presented as means ± SD of three
independent experiments.
4.2.5 Cell cycle analysis:
Cell cycle analysis was performed as previously described [4]. Briefly, OV2008 cells
were plated into 6 well plates at 1 x 106 cells/ well. The cells were incubated with 0, 2 or
4 µM of 15k compound and incubated for 12 h. Next, the cells were trypsinzed with
26

ACCEPTED MANUSCRIPT
.25% trypsin, 2.21 mM EDTA, 1X, washed, counted and resuspended in 0.5ml ice cold
PBS. The cells then were stained with propidium iodide (PI) dye and incubated for at
least 15 minutes. The distribution of the cells in each cell cycle phase for the different
compounds was measured using BD Accuri™ C6 flow cytometer from BD Biosciences
(Becton-Dickinson, San Jose, CA, USA) and analysed using FCS express 5 plus De
Novo software (Glendale, CA, USA).
4.2.6 Apoptosis and mitochondrial membrane potential analysis:
MitoTracker Red and Alexa Fluor 488 annexin V kits for flow cytometry (Molecular
Probes Inc., Invitrogen, Eugene, OR) were used to measure mitochondrial membrane
potential and apoptosis, respectively, in OV2008 cells. Briefly, cells were seeded in 6
well plates and treated with 0, 2, or 4 µM of compound 15k for 12 h. The cells were then
trypsinyzed 0.25% trypsin, 2.21 mM EDTA, 1X, counted, and 4 µl of 10 µM MitoTracker
Red working solution was added to 1 ml of the harvested cells. The cells were
incubated at 37 °C with 5% CO2 for 30 minutes. Next, PBS washing was followed by
cells resuspension in 100 µl of annexin binding buffer. The cell suspensions were then
incubated with 5µl of Alexa Fluor 488 annexin V for 15 minutes. This was followed by
adding 400 µl of annexin-binding buffer. Finally, flow cytometry was used to detect the
stained cellsand fluorescence emission at 530 nm and 585 nm was used with the BD
Accuri™ C6 flow cytometer from BD Biosciences (Becton-Dickinson, San Jose, CA,
USA) and analysed using FCS express 5 plus De Novo software (Glendale, CA, USA).
4.2.7 Protein expression analysis using Western blot:
27

ACCEPTED MANUSCRIPT
To measure the expression of α-tubulin, β-actin proteins, Western blotting was
performed by lysing OV2008 cell lines using a lysis buffer (50 mMTris–HCl, 150
mMNaCl, 1 mM EDTA, 0.5% NP-40, 1% Triton, 0.1% SDS) containing a protease
inhibitor
cocktail
from
sigma-Aldrich
life
science
(St.
Louis,
MO,
USA). Bicinchoninic acid (BCA) quantification assay was used to determine protein
amount in the cell extracts (G-BIOSCIENCES, St. Louis, MO, USA). The extracted
proteins were loaded onto a 10-20% tris-glycine gel. After separation, proteins were
transferred from the gel onto a PVDF membrane. Membranes were blocked using 5%
milk in Tris-buffered saline Tween 20 for 30 minutes. Rabbit α-tubulin (Cell signalling
technology; 1:3000 dilution and rabbit actin antibody (Covance; 1:5000 dilutions) were
added to the blocking buffer, and the membrane was incubated overnight at 4 °C.
Membranes were washed and incubated with horseradish peroxidase-labelled (HRP)
anti-rabbit secondary antibody (1:5000 dilutions). The membrane was incubated with
the antibody for additional 1 hour. Subsequently, membranes were washed and
developed by Clarity Western ECL substrate (Bio-Rad; Hercules, California, USA).
Protein was detected using ChemiDoc Imaging System (Bio-Rad). Densitometry
analyses of the blots for the detected protein were quantified using the image J
software. Data was calculated as ratios of α-tubulin / β-actin.
4.2.8 Immunofluorescence imaging analysis:
To measure the expression of α-tubulin, OV2008 cells were seeded in a density of
4×104 cell/ well and allowed to grow overnight. Cells were then incubated with different
concentrations of the synthetic derivatives for 24h at 37°C. The cells that had been
28