Synthesis of the Vitamin E amino acid esters with an enhanced anticancer activity and in silico screening for new antineoplastic drugs

Article abstract of DOI:10.1016/j.ejps.2016.04.008

Tocopherols and tocotrienols belong to the family of vitamin E (VE) with the well-known antioxidant properties. For certain α-tocopherol and γ-tocotrienol derivatives used as the lead compounds in this study, antitumor activities against various cancer ce

Full text of DOI:10.1016/j.ejps.2016.04.008

Synthesis of the vitamin E amino acid esters with an enhanced anticancer
activity and in silico screening for new antineoplastic drugs
Zarko Gagic, Branka Ivkovic, Tatjana Srdic-Rajic, Jelica Vucicevic, Katarina
Nikolic, Danica Agbaba
PII:
DOI:
Reference:
S0928-0987(16)30109-9
doi: 10.1016/j.ejps.2016.04.008
PHASCI 3532
To appear in:
Received date:
Revised date:
Accepted date:
15 January 2016
5 April 2016
5 April 2016
Please cite this article as: Gagic, Zarko, Ivkovic, Branka, Srdic-Rajic, Tatjana, Vucicevic,
Jelica, Nikolic, Katarina, Agbaba, Danica, Synthesis of the vitamin E amino acid esters
with an enhanced anticancer activity and in silico screening for new antineoplastic drugs,
(2016), doi: 10.1016/j.ejps.2016.04.008
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
Synthesis of the Vitamin E Amino Acid Esters with an Enhanced Anticancer Activity and
In Silico Screening for New Antineoplastic Drugs
Zarko Gagic a*, Branka Ivkovic b, Tatjana Srdic-Rajic c, Jelica Vucicevic b, Katarina Nikolic b,
Danica Agbaba b
a University of Banja Luka, Faculty of Medicine, Department of Pharmacy, Save Mrkalja 14,
78000 Banja Luka, Bosnia and Herzegovina
b University of Belgrade, Faculty of Pharmacy, Institute of Pharmaceutical Chemistry, Vojvode
Stepe 450, 11000 Belgrade, Serbia
c Department of Experimental Oncology, Institute for Oncology and Radiology of Serbia,
Pasterova 14, 11000 Belgrade, Serbia
* Corresponding author:
Zarko Gagic, University of Banja Luka, Faculty of Medicine, Department of Pharmacy, Save
Mrkalja 14, 78000 Banja Luka, Bosnia and Herzegovina;
phone: +387-51-340-106
e-mail: zarko.gagic@unibl.rs
Abstract
Tocopherols and tocotrienols belong to the family of vitamin E (VE) with the well-known
antioxidant properties. For certain α-tocopherol and γ-tocotrienol derivatives used as the lead
compounds in this study, antitumor activities against various cancer cell types have been
1

ACCEPTED MANUSCRIPT
reported. In the course of the last decade, structural analogs of VE (esters, ethers and amides)
with an enhanced antiproliferative and proapoptotic activity against various cancer cells were
synthesized. Within the framework of this study, seven amino acid esters of α-tocopherol (4a-d)
and γ-tocotrienol (6a-c) were prepared using the EDC/DMAP reaction conditions and their
ability to inhibit proliferation of the MCF-7 and MDA-MB-231 breast cancer cells and the A549
lung cancer cells was evaluated. Compound 6a showed an activity against all three cell lines
(IC50: 20.6 µM, 28.6 µM and 19 µM for the MCF-7, MDA-MB-231 and A549 cells,
respectively), while compound 4a inhibited proliferation of the MCF-7 (IC50=8.6 µM) and A549
cells (IC50=8.6 µM). Ester 4d exerted strong antiproliferative activity against the estrogen-
unresponsive, multi-drug resistant MDA-MB-231 breast cancer cell line, with IC50 value of 9.2
µM. Compared with the strong activity of compounds 4a, 4d and 6a, commercial α-tocopheryl
succinate and γ-tocotrienol showed only a limited activity against all three cell lines, with IC50
values greater than 50 µM. Investigation of the cell cycle phase distribution and the cell death
induction confirmed an apoptosis of the MDA-MB-231 cells treated with 4d, as well as a
synergistic effect of 4d with the known anticancer drug doxorubicin. This result suggests a
possibility of a combined therapy of breast cancer in order to improve the therapeutic response
and to lower the toxicity associated with a high dose of doxorubicin. The stability study of 4d in
human plasma showed that ca. 83% initial concentration of this compound remains in plasma in
the course of six hours incubation. The ligand based virtual screening of the ChEMBL database
identified new compounds with a potential antiproliferative activity on MCF-7 and on multi-drug
resistant MDA-MB 231 breast cancer cells.
Keywords: vitamin E, esters, anticancer, synthesis, screening
2

ACCEPTED MANUSCRIPT
1. Introduction
Vitamin E (VE) is a generic term that represents two families of compounds, i.e., tocopherols (T)
and tocotrienols (T3). Both families share similar chemical structure characterized by a saturated
(T) or unsaturated (T3) phytyl chain attached to the chromane ring. Moreover, each family has
four isomers (α, β, γ, δ) that differ in the number and positions of methyl groups in the
chromanol ring (Eitenmiller and Lee, 2004). Although the VE family members are the best
known for their antioxidant activity (Jiang, 2014), recent studies have shown that they exhibit
many antioxidation-independent effects such, as inhibition of protein kinase C, 5-lipoxygenase
and phospholipase A2, activation of protein phosphatase 2A and diacylglycerol kinase, inhibition
of cell proliferation, platelet aggregation and monocyte adhesion (Zingg and Azzi, 2014).
Nowadays, special attention is paid to the antiproliferative effect of VE. Tocopherols and
tocotrienols exert antitumor activity on various types of cancer cells, at the same time being
selective to normal cells (Moya-Camarena and Jiang, 2012; Ling et al., 2012). Direct comparison
between the two VE subclasses showed that tocotrienols are generally more potent than
tocopherols in reducing proliferation of tumor cells, with δ-tocotrienol being the most potent and
α-tocopherol the least potent isomer. (McIntyre et al., 2000; Guan et al., 2012; Campbell et al.,
2006).
Although an exact mechanism by which the VE family members inhibit carcinogenesis remains
unknown, several models are proposed, including apoptosis via the cleavages of poly (ADP-
ribose) polymerase (PARP) (Park et al., 2010; Patacsil et al., 2012), or inhibition of the nuclear
factor kappa-light-chain-enhancer of the activated B cells (NF-κB) signaling pathway (Sun et al.,
2014). Modulation of angiogenesis is now recognized as a strategy for preventing various
angiogenesis-mediated disorders, including solid tumors (Zhao and Adjei, 2015). Recent studies
3

ACCEPTED MANUSCRIPT
confirmed that the palm tocotrienols exhibit anti-angiogenic properties through a decreased
expression of vascular endothelial growth factor in human umbilical vein endothelial cells
(Selvaduray et al., 2012), human hepatocellular carcinoma cells (Siveen et al., 2014) and murine
mammary cancer cells (Selvaduray et al., 2010). Synthetic analogs of tocopherols showed more
potent effects than the natural forms in inducing dephosphorylation of the protein kinase B (Akt)
and thereby in inhibiting the Akt-mediated signals that promote cell metabolism, proliferation,
and motility (Huang et al., 2013). The experimental findings showed that the combined treatment
with γ-T3 and the antihyperlipidemic drugs such, as erlotinib and gefitinib, resulted in a
synergistic inhibition in vitro of highly malignant +SA mouse mammary epithelial cells, due to a
down-regulation of 3-hydroxy-3-methyl-glutaryl-CoA (HMG-CoA) reductase activity, which is
elevated in cancer cells (Goldstein et al., 2006; Sylvester, 2012). Furthermore, a combined
treatment with a low dose of γ-T3 and an anti-inflammatory drug celecoxib can greatly improve a
therapeutic response in the treatment of breast cancer through the inhibition of the
cyclooxygenase-2 (COX-2) activity (Shirode and Sylvester, 2011).
In the course of the last decade, the VE derivatives with a modified hydroxyl group have been
synthesized and tested for their anticancer activity. These redox-silent analogs that lost their
antioxidative potency showed an improved antiproliferative activity and stability relative to the
parent compounds (Behery et al., 2010; Birringer et al., 2003; Djuric et al., 1997; Elnagar et al.,
2010). The most prominent VE analog, α-tocopheryl succinate (α-TOS), was shown to exhibit
anticancer properties in several cancer models, while retaining low toxicity to healthy cells
(Huang et al., 2009; Dong et al., 2008). Selectivity of the VE analogs to cancer cells can be
explained by higher esterase activity in normal cells that cleave α-TOS and similar agents, to
produce the nonapoptogenic α-T (Neuzil et al., 2004). Due to their lesser susceptibility to
4

ACCEPTED MANUSCRIPT
oxidation, the succinate and acetate ester of α-tocopherol became commercially available for use
in the supplements and cosmetics (Zinng, 2007). In order to enhance stability of the VE
derivatives in vivo, the ester bond was replaced by an amide bond or by an even more resistant
ether bond. The amide and ether VE derivatives showed an additional enhancement of the
proapoptotic activity on the MDA-MB-453 breast cancer, Jurkat T lymphoma, the U937
leukemic and the Meso-2 malignant mesothelioma cells (Tomic-Vatic et al., 2005; Nishikawa et
al., 2003).
Derivatization of the phenolic hydroxyl of α-T as an amino acid ester conjugate (Arya et al.,
1998) introduces the NH2 group that is protonated in an acidic medium of the cancer tissue
(Montcourrier et al., 1994). The protonated amino group together with the long alkyl chain of α-
T acts as a cationic detergent and interacts with the lysosomal and mitochondrial membranes in
cancer cells, causing an increased membrane permeability and destabilizing lysosomes (Neuzil et
al., 2001). Contrary to α-T and α-TOS, the α-tocopheryl-lysine ester exerts an antiproliferative
activity against the human MCF-7 breast cancer cell line (Arya et al., 1998).
Within the framework of this study, we synthesized four amino acid esters of α-T (1) with lysine
(4a), proline (4b), glutamine (4c), and asparagine (4d), and three amino acid esters of γ-T3 (2)
with lysine (6a), proline (6b) and glutamine (6c). Based on the results from our previous QSAR
studies (Nikolic and Agbaba, 2009; Gagic et al., 2016), esterification of the phenol group at
position C6 of the chromane ring with at least one hydrogen bond donor and/or hydrogen bond
acceptor present in the ester chain, and an increased polarity of the molecule, exert positive
effect on the antiproliferative activity of the synthesized compounds. This activity was tested on
the human MCF-7 and MDA-MB-231 breast cancer cell lines and the A549 lung cancer cell
lines, and compared with the commercially available α-TOS and γ-T3. Since an exact anticancer
5

ACCEPTED MANUSCRIPT
mechanism and drug targets of these compounds are not known, we could only perform a ligand
based virtual screening (LBVS), employing the ChEMBL database to identify the new
compounds with a potential antiproliferative activity on human MCF-7 and MDA-MB-231
breast cancer cell lines.
2. Materials and methods
2.1. Chemistry
For the chemical synthesis, commercial α-tocopherol, Nα,Nε-di-Cbz-L-lysine, N-Cbz-D-proline,
Nα-Cbz-L-asparagine, Nα-Cbz-L-glutamine, 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide
(EDC), 4-dimethylaminopyridine (DMAP) and 10% Pd on the activated carbon were used. γ-
Tocotrienol was obtained from ExcelVite Sdn Bhd. (Malaysia) company (www.excelvite.com).
The purity of the compounds was checked using the HPLC and TLC methods. TLC was carried
out using the commercially precoated silica gel plates (Merck, Germany). The chromatographic
spots were visualized in UV light at 254 nm, or with potassium permanganate/sulfuric acid as a
visualizing reagent. The HPLC analysis was performed on the Zorbax Eclipse XDS-C18 column
with methanol/water (97:3, v/v) as mobile phase, and the UV detection at 210 nm. Column
chromatography and preparative thin-layer chromatography were performed for the purification
of the compounds using n-hexane/AcOEt, 3:1.2, v/v (1), n-hexane/AcOEt, 3:0.8, v/v (2),
AcOEt/n-hexane: 7:3, v/v (3), n-hexane/AcOEt, 3:0.6, v/v (4), CH2Cl2/MeOH, 100:0 → 99:1,
v/v (5), AcOEt (6) as the eluent systems. The IR spectra were recorded on the Thermo Nicolet
6700 FT-IR spectrophotometer (Thermo Fisher Scientific, USA), using the ATR technique. The
NMR spectra were recorded on the Varian Gemini 200 or Bruker Ascend 400 spectrometer.
Chemical shifts are expressed in the parts per million (δ ppm). The splitting patterns are
6

ACCEPTED MANUSCRIPT
described as singlet (s), doublet (d), triplet (t) and multiplet (m). The mass spectra of the
compound (MS/MS) were recorded with use of the TSQ Quantum Access MAX triple
quadripole spectrometer (Thermo Fisher Scientific, USA) equipped with the electrospray
ionization (ESI) source operating in the positive ionization mode. The high-resolution mass
spectra (ESI-TOF analyses) of the compounds were recorded with use of the Agilent 6210 time-
of-flight LC/MS system (G1969A, Agilent Technologies, USA) and the LTQ Orbitrap XL FT
Mass Spectrometer (Thermo Fisher Scientific, Germany). The purity of the compounds was
established as > 95%.
2.2. General procedure for synthesis of (RS)-α-tocopheryl α-(N-benzyloxycarbonylamino)
carboxylates 3a-d
To a stirred solution of α-tocopherol (2.2 mmol) in 20 mL dry CH2Cl2 (3a-b) or in a mixture of
10 mL dry CH2Cl2 and 10 mL THF (3c-d), DMAP (0.5 mmol), EDC (1.3 mmol) and the Cbz-
protected amino acid (1 mmol) were added. The reaction mixture was protected from the
daylight and it was left for an overnight at room temperature under the nitrogen. Then it was
washed three times with water, dried over the anhydrous Na2SO4 and the solvent was evaporated
under the vacuum. The residue was purified by the preparative TLC using the eluent systems (1),
(2), (3) and (4) to obtain compounds 3a, 3b, 3c and 3d, respectively.
2.2.1.
2,5,7,8‐tetramethyl‐2‐(4,8,12‐trimethyltridecyl)‐3,4‐dihydro‐2H
1‐benzopyran‐6‐yl
1
2,6‐di{[(benzyloxy)carbonyl]amino}hexanoate (3a). Yield:86%; H NMR (CDCl3) δ: 0.84-0.88
(m, 12H), 1.08-2.17 (m, 29H), 1.23 (s, 3H), 1.94 (s, 3H), 1.98 (s, 3H), 2.08 (s, 3H), 2.56-2.58 (m,
13
2H), 3.23 (m, 2H), 4.65 (m, 1H), 5.08-5.12 (m, 4H), 7.30-7.34 (m, 10H); C NMR (CDCl3) δ:
11.81, 12.11, 12.96, 19.68, 19.75, 20.59, 21.02, 22.62, 22.71, 24.44, 24.79, 27.98, 29.40, 29.69,
7

ACCEPTED MANUSCRIPT
31.03, 31.92, 32.18, 32.69, 32.79, 37.28, 37.38, 37.46, 37.56, 39.38, 40.53, 53.83, 66.70, 67.10,
75.13, 117.52, 123.20, 124.78, 126.51, 128.10, 128.19, 128.50, 128.53, 136.19, 136.55, 140.14,
149.63, 156.14, 156.53, 171.12; MS m/z: 827.3 [M+H]+.
2.2.2.
1‐benzyl-2‐(2,5,7,8‐tetramethyl‐2‐(4,8,12‐
(3b).
trimethyltridecyl)‐3,4‐dihydro‐2H‐1‐benzopyran‐6‐yl)-pyrrolidine‐1,2‐dicarboxylate
Yield:69%; 1H NMR (CDCl3) δ: 0.83-0.87 (m, 12H), 1.08-1.61 (m, 21H), 1.23 (s, 3H), 1.73-1.92
(m, 4H), 2.05 (s, 3H), 2.09 (s, 3H), 2.17 (s, 3H), 2.29-2.41 (m, 2H), 2.54-2.60 (m, 2H), 3.56-3.70
(m, 2H), 4.67-4.74 (m, 1H), 5.02-5.30 (m, 2H), 7.29-7.38 (m, 5H); 13C NMR (CDCl3) δ: 11.79,
11.83, 12.69, 19.60, 19.70, 20.59, 21.05, 22.64, 22.73, 23.51, 24.46, 24.83, 27.99, 29.29, 30.16,
31.14, 32.72, 32.79, 37.31, 37.41, 37.43, 37.47, 39.40, 47.01, 59.30, 75.10, 117.42, 123.09,
124.85, 126.57, 127.81, 127.99, 128.10, 128.46, 136.76, 140.37, 149.52, 154.89, 171.29; MS
m/z: 662.4 [M+H]+.
2.2.3.
2,5,7,8‐tetramethyl‐2‐(4,8,12‐trimethyltridecyl)‐3,4‐dihydro-2H‐1‐benzopyran‐6‐yl
1
2‐((benzyloxycarbonyl)amino)‐4‐carbamoylbutanoate (3c). Yield:38%; H NMR (CDCl3) δ:
0.84-0.87 (m, 12H), 1.07-1.71 (m, 21H), 1.23 (s, 3H), 1.72-1.82 (m, 2H), 1.95 (s, 3H), 1.99 (s,
3H), 2.08 (s, 3H), 2.16-2.19 (m, 1H), 2.41-2.49 (m, 3H), 2.56-2.59 (m, 2H), 4.67 (m, 1H), 5.09-
5.17 (m, 2H), 7.32-7.36 (m, 5H); 13C NMR (CDCl3) δ: 11.83, 12.15, 12.99, 19.63, 19.77, 20.60,
21.05, 22.64, 22.73, 24.45, 24.83, 27.99, 28.44, 29.70, 31.20, 31.93, 32.72, 32.79, 37.31, 37.37,
37.40, 37.49, 37.58, 39.39, 53.82, 67.20, 75.18, 117.56, 123.24, 124.77, 126.44, 128.13, 128.26,
128.35, 128.57, 136.18, 140.14, 149.70, 156.51, 170.71, 174.26; MS m/z: 693.4 [M+H]+.
2.2.4.
2,5,7,8‐tetramethyl‐2‐(4,8,12‐trimethyltridecyl)‐3,4‐dihydro‐2H‐1‐benzopyran‐6‐yl
1
2‐((benzyloxycarbonyl)amino)‐3‐cyanopropanoate (3d). Yield:31%; H NMR (CDCl3) δ: 0.83-
8

ACCEPTED MANUSCRIPT
0.87 (m, 12H), 1.05-2.01 (m, 21H), 1.24 (s, 3H), 1.73-1.83 (m, 2H), 2.04 (s, 3H), 2.09 (s, 3H),
2.16 (s, 3H), 2.57-2.63 (m, 2H), 3.20-3.24 (m, 2H), 4.85-4.89 (m, 1H), 5.17 (m, 2H), 7.33-7.38
(m, 5H); 13C NMR (CDCl3) δ: 11.83, 12.20, 13.04, 19.69, 19.76, 20.61, 21.03, 21.81, 22.62,
22.71, 24.45, 24.80, 27.98, 30.88, 32.70, 32.80, 37.31, 37.36, 37.40, 37.43, 37.47, 37.57, 39.39,
50.8, 67.68, 75.31, 115.94, 117.74, 123.49, 126.25, 128.21, 128.46, 128.65, 135.65, 140.05,
150.05, 155.55, 167.39; MS m/z: 661.1 [M+H]+.
2.3. General procedure for synthesis of (RS)-α-tocopheryl α-aminocarboxylates 4a-d
To a solution of compound (1 mmol) in anhydrous ethanol (3a-b, 3d) or in anhydrous methanol
(3c), 10% Pd/C catalyst (20% w/w with respect to 3a-d) was added. The reaction mixture was
stirred at ambient temperature in the atmosphere of hydrogen (2 atm H2). Once the reaction was
completed, (4 h for 3a-c and 6 h for 3d), the catalyst was filtered off using the membrane filter
and the solvent was evaporated. Solution of HCl in dry diethyl ether was added and the white
precipitate of the amine salt was filtered off and dried in a vacuum oven for 24 h at room
temperature, to result in compound 4a. Compounds 4b-4d were isolated as free amines by
purification on a chromatographic column, using the eluent systems (5), (6) and (3), to give
compounds 4b, 4c and 4d, respectively.
2.3.1.
2,5,7,8‐tetramethyl‐2‐(4,8,12‐trimethyltridecyl)‐3,4‐dihydro‐2H‐1‐benzopyran‐6‐yl
2,6‐diaminohexanoate hydrochloride (4a). Yield:82%; IR (ATR) cm-1: 1764.4, 3385.3; 1H NMR
(CDCl3) δ: 0.83-0.88 (m, 12H), 1.08-2.16 (m, 29H), 1.18 (s, 3H), 1.90 (s, 3H), 1.93 (s, 3H), 2.03
(s, 3H), 2.52-2.60 (m, 2H), 3.02-3.15 (m, 2H), 3.87 (m, 1H); 13C NMR (CDCl3) δ: 11.74, 12.11,
12.95, 19.63, 20.52, 20.70, 21.01, 22.60, 22.67, 23.74, 24.42, 24.76, 27.58, 27.93, 29.64, 30.93,
32.74, 33.17, 37.25, 37.40, 39.33, 39.77, 53.68, 75.03, 117.44, 123.03, 124.79, 126.49, 140.21,
9

ACCEPTED MANUSCRIPT
149.46, 173.77; MS m/z: 559.5 [M+H]+; HRMS calcd. for C35H62N2O3 [M+H]+ 559.48332,
found 559.48060.
2.3.2.
2,5,7,8‐tetramethyl‐2‐(4,8,12‐trimethyltridecyl)‐3,4‐dihydro‐2H‐1-benzopyran‐6‐yl
pyrrolidine‐2‐carboxylate (4b). Yield:78%; IR (ATR) cm-1:1763.5, 3360.2; 1H NMR (MeOD) δ:
0.87-0.90 (m, 12H), 1.12-2.13 (m, 23H), 1.24 (s, 3H), 2.00 (s, 3H), 2.04 (s, 3H), 2.10 (s, 3H),
2.16-2.25 (m, 2H), 2.37-2.42 (m, 1H), 2.60-2.63 (m, 2H), 2.65-2.70 (m, 1H), 3.47-3.52 (m, 2H),
4.84-4.88 (m, 1H); 13C NMR (MeOD) δ: 10.70, 10.91, 11.78, 18.84, 18.90, 20.15, 20.68, 21.69,
21.78, 23.40, 24.06, 24.52, 27.76, 28.53, 32.39, 32.44, 32.53, 36.89, 36.97, 37.01, 37.09, 39.17,
45.68, 59.44, 75.00, 117.72, 122.84, 126.14, 140.04, 149.65, 168.22; MS m/z: 528.1 [M+H]+;
HRMS calcd. for C34H57NO3 [M+H]+ 528.44112, found 528.44130.
2.3.3.
2,5,7,8‐tetramethyl‐2‐(4,8,12‐trimethyltridecyl)‐3,4‐dihydro‐2H‐1‐benzopyran‐6‐yl
1
2‐amino‐4‐carbamoylbutanoate (4c). Yield:68%; IR (ATR) cm-1: 1644.1, 1734.4, 3321.8; H
NMR (MeOD) δ: 0.87-0.90 (m, 12H), 1.11-1.56 (m, 21H), 1.26 (s, 3H), 1.80-1.83 (m, 2H), 1.99
(s, 3H), 2.02 (s, 3H), 2.10 (s, 3H), 2.17-2.20 (m, 2H), 2.49-2.53 (m, 2H), 2.62-2.65 (m, 2H),
3.86-3.90 (m, 1H); 13C NMR (MeOD) δ: 10.67, 10.97, 11.85, 18.81, 18.86, 20.17, 20.71, 21.66,
21.75, 24.05, 24.51, 27.76, 28.17, 29.31, 29.70, 30.72, 31.24, 32.53, 37.00, 37.05, 37.09, 37.13,
39.16, 54.68, 74.84, 117.51, 122.55, 124.80, 126.34, 140.39, 149.25, 173.31, 176.39; MS m/z:
559.5 [M+H]+; HRMS calcd. for C34H58N2O4 [M+H]+ 559.44693, found 559.44685.
2.3.4.
2,5,7,8‐tetramethyl‐2‐(4,8,12‐trimethyltridecyl)‐3,4‐dihydro‐2H‐1‐benzopyran‐6‐yl
2‐amino‐3‐cyanopropanoate (4d). Yield:28%; IR (ATR) cm-1: 1753.1, 2159.3, 3292.1; 1H NMR
(DMSO-d6) δ: 0.81-0.85 (m, 12H), 1.06-1.53 (m, 21H), 1.19 (s, 3H), 1.73-1.75 (m, 2H), 1.92 (s,
3H), 1.94 (s, 3H), 2.02 (s, 3H), 2.55-2.58 (m, 2H), 2.95-2.98 (m, 2H), 4.03-4.06 (m, 1H); 13C
10

ACCEPTED MANUSCRIPT
NMR (DMSO-d6) δ: 12.03, 12.30, 13.15, 20.04, 20.10, 20.39, 20.84, 22.93, 23.02, 23.40, 24.16,
24.60, 27.84, 31.08, 32.39, 32.47, 36.98, 37.07, 37.14, 37.18, 37.21, 37.35, 39.23, 51.62, 75.34,
117.92, 118.90, 122.37, 125.37, 126.75, 140.47, 149.24, 171.89; MS m/z: 527.1 [M+H]+; HRMS
calcd. for C33H54N2O3 [M+H]+ 527.42072, found 527.41897.
2.4. General procedure for synthesis of γ-tocotrienyl α-(N-benzyloxycarbonylamino)
carboxylates 5a-c
To a stirred solution of γ -tocotrienol (2.2 mmol) in 20 mL dry CH2Cl2 (5a-b) or in a mixture of
10 mL dry CH2Cl2 and 10 mL THF (5c), DMAP (0.5 mmol), EDC (1.3 mmol) and the Cbz-
protected amino acid (1 mmol) were added. The mixture was protected from the daylight and left
for an overnight at room temperature under the nitrogen. The reaction mixture was washed three
times with water, dried over the anhydrous Na2SO4 and the solvent was evaporated under the
vacuum. The residue was purified by the preparative TLC using the eluent systems (1), (2) and
(3), to obtain compounds 5a, 5b and 5c, respectively.
2.4.1.
2,7,8‐trimethyl‐2‐((3E,7E)‐4,8,12‐trimethyltrideca‐3,7,11‐trien‐1-
yl)‐3,4‐dihydro‐2H‐1‐benzopyran‐6‐yl-2,6‐di((benzyloxycarbonyl)amino)hexanoate
(5a).
1
Yield:83%; H NMR (CDCl3) δ: 1.26 (s, 3H), 1.60 (s, 3H), 1.60 (s, 3H), 1.63 (s, 3H), 1.68 (s,
3H), 1.53-2.13 (m, 20H), 1.99 (s, 3H), 2.10 (s, 3H), 2.69 (m, 2H), 3.21 (m, 2H), 4.59-4.60 (m,
1H), 5.08-5.15 (m, 7H), 6.54 (s, 3H), 7.33-7.34 (m, 10H); 13C NMR (CDCl3) δ: 11.95, 12.67,
15.92, 16.02, 17.69, 22.20, 22.44, 24.08, 25.70, 26.62, 26.78, 29.45, 31.03, 32.27, 39.70, 39.73,
40.05, 40.56, 53.85, 66.70, 67.12, 75.90, 118.54, 118.57, 124.19, 124.22, 124.43, 126.03, 126.87,
128.10, 128.13, 128.21, 128.52, 128.55, 131.25, 134.99, 135.25, 136.23, 136.60, 141.26, 149.74,
156.09, 156.54, 171.54; MS m/z: 807.3 [M+H]+.
11

ACCEPTED MANUSCRIPT
2.4.2.
1‐benzyl-
2‐(2,7,8‐trimethyl‐2‐((3E,7E)‐4,8,12‐trimethyltrideca‐3,7,11‐trien‐1‐yl)‐3,4‐dihydro‐2H‐1‐benzo
1
pyran‐6‐yl)-pyrrolidine‐1,2‐dicarboxylate (5b). Yield:72%; H NMR (CDCl3) δ: 1.26 (s, 3H),
1.59 (s, 3H), 1.60 (s, 3H), 1.68 (s, 3H), 1.68 (s, 3H), 1.63-2.18 (m, 16H), 2.04 (s, 3H), 2.11 (s,
3H), 2.19-2.27 (m, 1H), 2.35-2.39 (m, 1H), 2.63-2.72 (m, 2H), 3.55-3.71 (m, 2H), 4.58-4.65 (m,
13
1H), 5.09-5.21 (m, 5H), 6.60 (s, 1H), 7.30-7.38 (m, 5H); C NMR (CDCl3) δ: 11.91, 12.59,
15.92, 16.02, 17.69, 22.20, 23.59, 24.13, 24.47, 25.70, 26.62, 26.78, 29.29, 31.05, 39.73, 39.95,
47.04, 58.92, 67.22, 75.81, 118.51, 118.75, 124.18, 124.23, 124.42, 125.86, 126.88, 127.82,
127.99, 128.48, 131.25, 134.98, 135.24, 136.76, 141.27, 149.57, 154.91, 171.67; MS m/z: 642.4
[M+H]+.
2.4.3.
2,7,8‐trimethyl‐2‐((3E,7E)‐4,8,12‐trimethyltrideca‐3,7,11‐trien-
1‐yl)‐3,4‐dihydro‐2H‐1‐benzopyran‐6‐yl 2‐((benzyloxycarbonyl)amino)‐4‐carbamoylbutanoate
(5c). Yield:29%; 1H NMR (CDCl3) δ: 1.27 (s, 3H), 1.59 (s, 3H), 1.60 (s, 3H), 1.68 (s, 3H), 1.68
(s, 3H), 1.61-2.18 (m, 16H), 2.00 (s, 3H), 2.10 (s, 3H), 2.39-2.46 (m, 2H), 2.70-2.71 (m, 2H),
4.63 (m, 1H), 5.09-5.17 (m, 5H), 6.56 (s, 1H), 7.31-7.37 (m, 5H); 13C NMR (CDCl3) δ: 11.96,
12.70, 15.92, 16.02, 17.69, 22.19, 22.21, 24.09, 25.70, 26.62, 26.78, 28.54, 31.01, 31.80, 39.70,
39.73, 40.06, 53.74, 67.22, 75.92, 118.53, 118.55, 124.18, 124.21, 124.42, 126.05, 126.84,
128.16, 128.27, 128.58, 131.26, 135.00, 135.27, 136.17, 141.22, 149.82, 171.07, 174.02; MS
m/z: 673.3 [M+H]+.
2.5. General procedure for synthesis of γ-tocotrienyl α-aminocarboxylates 6a-c
To a solution of compound (1 mmol) in anhydrous ethanol (5a-b) or in anhydrous methanol (5c),
10% Pd/C catalyst (20% w/w with respect to 5a-c) was added. The reaction mixture was stirred
12

ACCEPTED MANUSCRIPT
for 2 h at ambient temperature in the atmosphere of hydrogen (2 atm H2). The catalyst was
filtered off using the membrane filter and the solvent was evaporated. Solution of HCl in dry
diethyl ether was added and the white precipitate of the amine salt was filtered off to result in
compound 6a. Compounds 6b and 6c were isolated as free amines by purification on a
chromatographic column using the eluent system (6).
2.5.1.
2,7,8‐trimethyl‐2‐((3E,7E)‐4,8,12‐trimethyltrideca‐3,7,11‐trien‐1‐yl)‐3,4‐dihydro‐2H‐1‐benzopyr
an‐6‐yl 2,6‐diaminohexanoate hydrochloride (6a). Yield:78%; IR (ATR) cm-1: 1765.9, 3396.4;
1H NMR (DMSO-d6) δ: 1.23 (s, 3H), 1.54 (s, 3H), 1.54 (s, 3H), 1.56 (s, 3H), 1.63 (s, 3H), 1.07-
2.04 (m, 20H), 1.98 (s, 3H), 2.06 (s, 3H), 2.69 (m, 2H), 2.80 (m, 2H), 4.29 (m, 1H), 5.06-5.13
(m, 3H), 6.71 (s, 1H); 13C NMR (DMSO-d6) δ: 12.18, 12.98, 16.08, 16.27, 18.01, 21.87, 22.17,
22.97, 24.21, 25.95, 26.41, 26.66, 28.92, 29.85, 30.87, 38.64, 40.44, 40.65, 40.76, 52.20, 76.27,
119.04, 119.31, 124.35, 124.59, 124.64, 125.44, 127.12, 131.07, 134.81, 134.87, 141.20, 149.68,
172.39; MS m/z: 539.4 [M+H]+; HRMS calcd. for C34H54N2O3 [M+H]+ 539.42072, found
539.41883.
2.5.2.
2,7,8‐trimethyl‐2‐((3E,7E)‐4,8,12‐trimethyltrideca‐3,7,11‐trien‐1‐yl)‐3,4‐dihydro‐2H‐1‐benzopyr
1
an‐6‐yl pyrrolidine‐2‐carboxylate (6b). Yield:55%; IR (ATR) cm-1: 1766.1, 3369.4; H NMR
(CDCl3) δ: 1.27 (s, 3H), 1.58 (s, 3H), 1.59 (s, 3H), 1.60 (s, 3H), 1.68 (s, 3H), 1.69-2.13 (m, 14H),
2.01 (s, 3H), 2.11 (s, 3H), 2.18-2.31 (m, 4H), 2.63-2.73 (m, 2H), 2.96-3.0 (m, 1H), 3.14-3.17 (m,
1H), 4.01-4.04 (m, 1H), 5.09-5.15 (m, 3H), 6.56 (s, 1H); 13C NMR (CDCl3) δ: 11.96, 12.63,
15.91, 16.02, 17.69, 22.20, 22.67, 24.12, 25.55, 25.70, 26.62, 26.78, 30.58, 31.06, 39.71, 39.73,
40.02, 47.03, 59.80, 75.85, 118.47, 118.65, 123.96, 124.22, 124.43, 125.94, 126.95, 131.25,
13

ACCEPTED MANUSCRIPT
134.98, 135.23, 141.45, 149.60, 174.55; MS m/z: 508.4 [M+H]+; HRMS calcd. for C33H49NO3
[M+H]+ 508.37852, found 508.37874.
2.5.3..
2,7,8‐trimethyl‐2‐((3E,7E)‐4,8,12‐trimethyltrideca‐3,7,11‐trien‐1‐yl)‐3,4‐dihydro‐2H‐1‐benzopyr
an‐6‐yl 2‐amino‐4‐carbamoylbutanoate (6c). Yield:68%; IR (ATR) cm-1: 1667.2, 1739.3,
1
3357.5; H NMR (CDCl3) δ: 1.27 (s, 3H), 1.59 (s, 3H), 1.60 (s, 3H), 1.66 (s, 3H), 1.68 (s, 3H),
1.61-2.18 (m, 14H), 2.01 (s, 3H), 2.11 (s, 3H), 2.32-2.43 (m, 2H), 2.47-2.55 (m, 2H), 2.69-2.71
(m, 2H), 3.72-3.77 (m, 1H), 5.10-5.14 (m, 3H), 6.55 (s, 1H); 13C NMR (CDCl3) δ: 11.96, 12.72,
15.90, 17.68, 22.19, 22.23, 24.10, 25.74, 26.61, 26.77, 29.91, 31.02, 32.15, 39.69, 39.73, 39.90,
53.96, 75.90, 118.53, 118.58, 124.17, 124.37, 124.42, 126.01, 126.84, 131.24, 134.96, 135.24,
141.37, 149.68, 170.96, 174.61; MS m/z: 539.3 [M+H]+; HRMS calcd. for C33H50N2O4 [M+H]+
539.38433, found 539.38390.
2.6. Biological assays
2.6.1. Cell culture
All cell lines used in this study were obtained from the American Type Culture Collection
(Rockville, MD), unless specified otherwise. The human lung cancer (A549) cells and human
normal lung tissue (MRC5) cell lines were maintained as a monolayer culture in the Roswell
Park Memorial Institute (RPMI) 1640 medium, while the human breast cancer cell lines (MCF7
and MDA-MB-231) were maintained in the Dulbecco's Modified Eagle medium (DMEM). The
powdered RPMI 1640 medium and DMEM were purchased from Sigma Chemicals Co, USA.
The nutrient medium RPMI 1640 was prepared in sterile deionized water, supplemented with
penicillin (192 U/mL), streptomycin (200 µg/mL), 4-(2-hydroxyethyl) piperazine-1-
14

ACCEPTED MANUSCRIPT
ethanesulfonic acid (HEPES) (25 mM), L-glutamine (3 mM) and 10% heat-inactivated fetal calf
serum (FCS) (pH 7.2). The nutrient medium DMEM was prepared in sterile deionized water
supplemented with penicillin (192 U/mL), streptomycin (200 µg/ mL) and 10% heat-inactivated
FCS. The cells were grown at 37C in a humidified atmosphere of 5% CO2 with twice-weekly
subculture.
2.6.2. MTT assay
The antiproliferative activity of the tested compounds was assessed using the MTT assay
(Mosmann, 1983; Ohno and Abe, 1991). All cell lines were seeded at a density of 7000
cells/well in the 96-well plates and left for 24 h to rest, followed by an addition of the tested
compounds at various concentrations (0.6-50 μM). Stock solutions of the compounds (80 mM)
were prepared in DMSO immediately before the experiment, whereas further dilutions were
performed with the fresh medium. Each concentration was tested in triplicate during the 48 hours
incubation. After the treatment in the 96-well plates, the MTT solution (3-(4, 5-dimethylthiazol-
2-yl)-2, 5-dyphenyl tetrazolium bromide) (Sigma-Aldrich, St. Louis, USA) (20 µL/well) was
added to each well. Samples were incubated for further 4 h, followed by an addition of 100 μL
10% sodium dodecyl sulfate (SDS). The absorbance was measured on the next day at 570 nm
and the cell survival (S %) was calculated as an absorbance (A570 nm) ratio between the treated
and the control cells multiplied by 100. IC50 was defined as the concentration of the agent that
inhibited cell survival by 50%, compared with the vehicle control.
2.6.3. Flow-cytometric analysis of cell cycle phase distribution
Quantitative analysis of the cell cycle phase distribution was performed by the flow-cytometric
analysis of the DNA content in the fixed MDA-MB-231 cells after staining with propidium
15

ACCEPTED MANUSCRIPT
iodide (PI) (Sigma-Aldrich, US) (Ormerod, 1994). The tested cell line in the exponential phase
of growth, plated at a density of 2 x 105 cells/Petri dish (60 x 15 mm, NUNC), was continually
exposed to compound 4d, Dox and their combination at the IC50 concentrations. After 48 hours
continuous treatment, cells were collected by trypsinization, washed twice with the ice-cold PBS
and fixed for 30 min in 70% EtOH. The fixed cells were washed again with PBS and incubated
with RNase A (1 mg/mL) for 30 min at 37 C. The cells were stained with PI (400 µg/mL)
immediately before the analysis. The cell cycle phase distribution was analyzed using the FACS
(Becton Dickinson, USA) flow cytometer and the Cell Quest computer software.
2.6.4. Apoptotic assay
Apoptosis of the MDA-MB-231 cells treated with compound 4d, Dox and their combination was
evaluated by the Annexin V-fluorescein isothiocyanate (FITC) apoptosis detection kit (BD
Biosciences, Pharmingen San Diego, CA, USA). Briefly, after incubation for 48 h with 1x IC50
of the investigated compounds, cells were washed twice with the cold PBS and resuspended in
200 µL of the binding buffer (10 mM HEPES/NaOH, pH 7.4, 140 mM NaCl, 2.5 mM CaCl2) at a
concentration of 1 x 106 cells/mL. A total of 100 µL of the solution (1 x 105 cells) was
transferred to a 5 mL culture tube and 5 L Annexin V–FITC and 5 L PI was added. The cells
were gently vortexed and incubated for 15 min at 25 C in the dark. Finally, 400 L of the
binding buffer was added to each tube and the analysis was carried out using the FACS (Becton
Dickinson, USA) flow cytometer and the Cell Quest computer software.
2.7. Stability in human plasma
Plasma was prepared by centrifugation of human blood obtained from a healthy adult volunteer.
A stability study in human plasma was conducted by the procedure described elsewhere (Behery
16

ACCEPTED MANUSCRIPT
et al., 2010; Konsoula and Jung, 2008). The freshly prepared human plasma (1.5 mL) was
preheated at 37ºC on a shaking water bath. The reaction was initiated by adding the tested
compound to reach final concentration of 200 µg/mL. Samples (75 μL) were taken at 0, 0.5, 1, 2,
4, 6, and 24 h and added to 300 μL acetonitrile in order to deproteinize the plasma. The samples
were vortexed for 1 min and then centrifuged at 4°C for 15 min at 14,000 rpm. Incubations were
conducted in duplicate. Clear supernatants were analyzed by HPLC using the Zorbax Eclipse
XDB-C18 column and acetonitrile/water (95:5, v/v) as mobile phase. The mobile phase flow rate
was set at 2 mL/min and the UV detection wavelength at 215 nm. Chromatographic analyses
were performed at 30ºC. This HPLC method was developed earlier for the analysis of α-
tocopheryl esters (Marra et al., 2009) and then modified for the purpose of this study. Diluted
concentrations (in the range of 1–500 μg/mL) of the stock solutions of the compounds in plasma
were used to construct the calibration curves.
2.8. Ligand based virtual screening
Two LBVS models were developed using the FLAP software (Molecular Discovery Ltd., 2015)
to identify the compounds with a potential antiproliferative activity on the MCF-7 cells (model
1) and the MDA-MB 321 cells (model 2). Compounds 3a and 4d, which have been shown to
exert a strong antiproliferative activity on these cell lines, were used as templates for screening in
model 1 and model 2, respectively. The dataset was generated using the ChEMBL database
(https://www.ebi.ac.uk/chembl), which currently is the most extensive public database
containing 1 463 270 differing bioactive compounds. In order to validate the model, compounds
with the strongest antiproliferative activity on the MCF-7 and MDA-MB 321 cells should be
taken into the account. Thus, we included in the dataset the newly synthesized compounds with
an experimentally determined antiproliferative activity on the cancer cells and 53 compounds
17

ACCEPTED MANUSCRIPT
from our previous QSAR study (Gagic et al., 2016; Djuric et al., 1997; Elnagar et al., 2010; Arya
et al., 1998; Liu et al., 2013).
FLAP calculates the GRID Molecular Interaction Fields (Cross et al., 2010) (MIFs) using H, O,
N1, and the DRY probes to describe the shape, the hydrogen bond acceptor, the hydrogen bond
donor, and the hydrophobic interactions. Each molecule is described as a set of quadruplets
derived by the combination of all atoms, four by four, in an exhaustive way. All combinations of
the quadruplets are then used to generate different superpositions of a test molecule onto a
template molecule and the MIF similarity is calculated. Compounds from the database were
ranked by different similarity scores, i.e., by the probe scores (representing a degree of an
overlap of the MIFs for each probe individually and for the combinations thereof), by the
distance score (representing an overall difference of the probe score between the ligand and the
template), and the two global scores (Glob-Prod and Glob-Sum). The enrichment plot was used
to determine the best similarity score based on the AUC value that holds for the measure of a
successful discrimination between the known active and inactive compounds in a given template.
3. Results and discussion
3.1. Chemistry
The synthetic route for the target compounds is outlined in Scheme 1. The four α-T amino acid
esters (lysine, proline, glutamine and asparagine) and the three γ-T3 amino acid esters (lysine,
proline, and glutamine) were prepared in two steps and their structures were confirmed by the
IR, 1H-NMR, 13C-NMR, MS and HRMS spectroscopic analyses.
The first step in the synthetic route was the coupling of α-T/γ-T3 with the corresponding Cbz-
protected amino acid by employing the Steglich esterification method (Neises and Steglich,
18

ACCEPTED MANUSCRIPT
1978). In this paper, an alternative reaction employing EDC instead of DCC was used, because
of an enhanced solubility of EDC and more particularly, because the urea by-product formed in
the course of the reaction can easily be removed from the reaction mixture by extraction with
water. In an earlier study, the influence of the reactants concentration (acid, alcohol and a
catalyst) on the progress of esterification and on the urea by-product formation was studied
(Tsvetkova et al., 2006). We employed the acid/alcohol/catalyst molar ratio as equal to
0.01:0.022:0.0050 that was found optimal for the preparation of the esters. Asparagine and
glutamine are known to undergo the nitrile formation via dehydration of the carboxamide side-
chain, when using carbodiimides as the activating agents (Katsoyannis et al., 1958). With
asparagine, this reaction course is almost exclusive, so we decided to prepare the α-T-asparagine
ester in the form of the nitrile. With glutamine, dehydration results in a smaller yield, so we were
able to isolate the glutamine esters as the amides.
The Cbz-protecting groups were removed from each coupled product (3a-d and 5a-c) by
catalytic hydrogenation, using 10% Pd/C as a catalyst. Hydrogenation of the γ-T3 esters (5a-c)
was carefully monitored because of the risk of reduction of double bonds. Samples were
collected in the 30-min intervals and analyzed by means of TLC, HPLC and NMR. We found out
that two hours is an optimal period for hydrogenation, which does not cause the reduction of
double bonds and at the same time, it provides satisfactory reaction yields. Compounds 3a and
6a were precipitated as hydrochloride salts after acidification with the HCl solution in dry diethyl
ether and then the precipitate was filtered off. Precipitation did not occur with the other
compounds so they were isolated as free amino derivatives by purification with use of the
preparative TLC or column chromatography.
3.2. In vitro biological evaluation
19

ACCEPTED MANUSCRIPT
The antiproliferative activity of all synthesized compounds was tested against the human cancer
cell lines, i.e., against the estrogen-responsive MCF-7 breast cancer, the estrogen-unresponsive
MDA-MB-231 breast cancer, and the A549 lung cancer cell line, and the results were compared
with those obtained for the commercially available α-TOS and γ-T3. The proliferation inhibitory
activities were evaluated by the MTT assay after 48 h and expressed as the IC50 values (Table 1).
Moreover, all the compounds were screened for the cytotoxic activity against the normal human
fetal lung fibroblast MRC-5 cell line.
It was established that compounds 4a, 4d and 6a exert a good antiproliferative activity, with the
IC50 values lower than 30 µM (Table 1). As already mentioned before, an antiproliferative
activity of the α-T-lysine ester (4a) against the MCF-7 cell line has already been reported
elsewhere (Arya et al., 1998) and our results remain in good agreement with those already
published. Moreover, compound 4a showed very good activity against the A549 cell line.
Compound 6a was the only one active against all three cell lines, but 4a was more potent against
the MCF-7 and A549 cell lines. The proline esters of α-T (4b) and γ-T3 (6b) showed moderate
activity against the MCF-7 cell line, with the cell survival of 67.1% and 66.1% at 20 µM,
respectively (Fig. 1). In contrast with the synthesized compounds 4a, 4d and 6a able to
effectively inhibit proliferation at the selected cancer cell lines, the parent γ-T3 (2) and the
commercial α-TOS showed a limited activity against all three cell lines only, with the IC50 values
higher than 50 µM, in spite of an earlier demonstrated anticancer activity of these two
compounds against various cancer cells (Ling et al., 2012; McIntyre et al., 2000; Guan et al.,
2012; Patacsil et al., 2012; Sun et al., 2014).
Compound 4d was the only one which exerted a strong selective antiproliferative activity against
the estrogen-unresponsive, multi-drug resistant MDA-MB-231 breast cancer cell line. In order to
20

ACCEPTED MANUSCRIPT
obtain more information regarding the events related to the antiproliferative activity of 4d and to
investigate a potential synergistic effect with the anticancer drug doxorubicin (Dox), changes in
the cell cycle progression were evaluated using the fluorescence-activated cell sorting (FACS)
analysis. For this purpose, the MDA-MB-231 cells were treated for 48 hours with 4d, Dox, as
well as with the combination thereof (4d-Dox). Each of these two agents was applied at the IC50
concentration. It needs to be mentioned that the IC50 value of Dox against the MDA-MB-231
cells was 4 µM (data not presented). After 48 hours continuous treatment, 4d and Dox induced a
percentage increase of the apoptotic cells (sub-G1 peak) and a strong percentage decrease of the
cells in the G1 and S phases, and in the S and G2/M phases, respectively (Fig. 2). The highest
percentage (45%) of apoptotic cells was found with the MDA-MB-231 cells upon 48 hours
treatment with the combination of 4d and Dox. This effect was accompanied by the reduction of
cells in both, the S and G2/M phases of the cell cycle (Fig. 2).
Then we performed the FACS analysis of the cell death induction on the MDA-MB-231 cells
using 4d, Dox and the combination thereof. The cells were incubated with the investigated
compounds for 48 hours and stained with Annexin V and propidium iodide (PI) for the
identification of the early and late apoptotic cells. Treatment with 4d, Dox, and especially with
the combination of 4d and Dox resulted in the onset of a massive apoptotic cell death (Fig 3.).
Interestingly, while the percentage of the early apoptotic cell population increased after the 4d-
Dox co-treatment, the late apoptotic and necrotic cell populations were almost lost (the Annexin
V positive/PI positive and the Annexin V negative/PI positive cells, respectively), probably
suggesting a delay in the activation of the apoptotic execution machinery (Fig. 3). These results
suggest that the combined 4d-Dox treatment of the highly metastatic breast cancer could lead to
21

ACCEPTED MANUSCRIPT
an enhanced therapeutic response and lower the toxicity associated with a high dose of the Dox
monotherapy.
3.3. Stability in human plasma
Due to the high levels of non-specific esterases, the VE esters could be hydrolyzed in the liver.
The sensitivity of α-tocopheryl acetate and the α-tocopheryl asparagine ester to hepatic
hydrolysis were previously verified using the porcine liver esterase (Marra et al., 2009). The
half-life of commercial α-tocopheryl acetate was established as equal to ca. 1 h, and the α-
tocopheryl asparagine ester has proved as significantly more stable against hepatic enzymes with
the half-life of 6 h (Marra et al., 2009). In order to avoid an extensive liver metabolism and to
achieve high concentrations of the VE esters in plasma, the parenteral formulations of the VE
esters were proposed (dos Santos et al., 2012; Weber et al., 2002). Recently, the liposomal
parenteral formulations of the VE derivatives were developed to increase the half-life and ensure
their safe delivery to the tumor cells. An anticancer effect of these formulations has been
successfully demonstrated in the pre-clinical models in vivo (Turanek et al., 2009; Koudelka et
al., 2015; Neophytou and Constantinou, 2015).
Since it is possible that the enzymes can hydrolyze the VE esters in blood plasma, we tested the
stability of our most active compound 4d and the commercial α-TOS in the freshly prepared
human plasma. The results depicted on Fig. 4 show that both compounds slowly decomposed in
plasma. Concentration of the examined esters decreased to about 83% of the initial concentration
within the 6 hours incubation. After 24 h, 39% of the initial 4d concentration and 69% of the
initial α-TOS concentration were determined in plasma. These results point out to an acceptable
metabolic stability of 4d and the related VE esters in human plasma. Based on the all above
mentioned facts, it could be concluded that parenteral application of the examined VE esters
22

ACCEPTED MANUSCRIPT
should provide enough stable concentration in plasma to warrant effective distribution and
sufficient therapeutic concentrations of the compounds in tumor tissue.
3.4. Ligand based virtual screening
Virtual screening is a powerful technique to identify potentially active compounds from
molecular databases. To identify the novel hit compounds with potential antiproliferative activity
on the MCF-7 cells (model 1) and the MDA-MB 321 cells (model 2), 3a and 4d, respectively,
were used as the templates for LBVS of the ChEMBL database (https://www.ebi.ac.uk/chembl).
Moreover, the synthesized compounds and the compounds from our previous QSAR study
(Gagic et al., 2016) were added to the dataset. The GRID probes H, DRY, O, and N1 included in
FLAP (Molecular Discovery Ltd., 2015) were used for modeling of the ligands and templates.
The compounds were ranked based on the similarity score. According to the enrichment plot, the
highest AUC value was achieved using the DRY score (AUC= 0.78) for model 1 and the
N1*DRY score (AUC= 0.75) for model 2. Thus, it can be concluded that hydrophobicity and the
presence of the H-bond acceptors are the important factors for discriminating the
antiproliferative activity of the screened compounds. The five top ranked compounds from
model 1 (CHEMBL3105824, CHEMBL477744, CHEMBL1825075, CHEMBL1190700 and
CHEMBL1825079) and the five top ranked compounds from model 2 (CHEMBL81504,
CHEMBL317957, CHEMBL3105824, CHEMBL3183599 and CHEMBL3091854) were
selected as the hit leads with the highest potential to show an antiproliferative activity on the
MCF-7 and MDA-MB 321 breast cancer cell lines, respectively (Fig. 5). The screening identified
CHEMBL1825079 as the potentially active compound against the MCF-7 cell line, and its
antiproliferative activity on this cell line has already been reported (Carvalho et al., 2011), in that
way confirming the validity of the model. Besides, CHEMBL3105824, CHEMBL477744,
23

ACCEPTED MANUSCRIPT
CHEMBL1825075, CHEMBL81504 and CHEMBL317957 have also been tested and they were
shown active against various different cancer cell lines. Both models identified
CHEMBL3105824 as a compound potentially active against the MCF-7 and MDA-MB 321
cells.
4. Conclusions
Seven amino acid esters of α-tocopherol and γ-tocotrienol were synthesized and their in vitro
antiproliferative activity was tested on the human MCF-7 and MDA-MB-231 breast cancer cell
lines, and the A549 lung cancer cell line. Compound 4a effectively inhibited proliferation of the
MCF-7 and A549 cells, while 6a exerted an antiproliferative effect on the three investigated cell
lines. Interestingly, 4d inhibited proliferation of the multi-drug resistant MDA-MB-231 cells at
the doses that had no effect on the normal human MRC-5 cells. In contrast with good activity of
4a, 4d and 6a, the commercial α-TOS and γ-T3 samples showed only a limited activity against
the three investigated cell lines. The cell cycle and the cell death induction studies demonstrated
synergistic effect of the combined treatment of 4d with the anticancer drug doxorubicin,
indicating a possible combination therapy with the lower dose of Dox to reduce toxicity. Based
on an acceptable metabolic stability of 4d and the related VE esters in human plasma, a
conclusion can be drawn that parenteral application of the VE esters should provide stable
enough concentrations thereof in plasma to warrant their effective distribution and high
therapeutic concentrations in tumor tissues, and therefore the VE esters seem very good
candidates for the further in vivo testing.
The virtual screening studies identified new compounds with a potential antiproliferative activity
on the human MCF-7 and MDA-MB-231 cells.
Acknowledgment
24

ACCEPTED MANUSCRIPT
We would like to thank the ExcelVite Sdn Bhd. (Malaysia) company (www.excelvite.com) for
sample of pure γ-tocotrienol for synthesis. This work was supported by the Ministry of
Education and Science of the Republic of Serbia, Contract #172033 and III 41026. KN, JV, and
DA kindly acknowledge the COST CM1406 action.
References
Arya, P., Alibhai, N., Qin, H., Burton, G.W., Batist, G., You, S.X., Alaoui-Jamali, M.A., 1998.
Design and synthesis of analogs of vitamin E: antiproliferative activity against human
breast adenocarcinoma cells. Bioorg. Med. Chem. Lett. 8, 2433-2438.
Behery, F.A., Elnagar, A.Y., Akl, M.R., Wali, V.B., Abuasal, B., Kaddoumi, A., Sylvester, P.W.,
El Sayed, K.A., 2010. Redox-silent tocotrienol esters as breast cancer proliferation and
migration inhibitors. Bioorg. Med. Chem. 18, 8066-8075.
Birringer, M., EyTina, J.H., Salvatore, B.A., Neuzil, J., 2003. Vitamin E analogues as inducers
of apoptosis: structure-function relationship. Br. J. Cancer 88, 1948-1955.
Campbell, S.E., Stone, W.L., Lee, S., Whaley, S., Yang, H., Qui, M., Goforth, P., Sherman, D.,
McHaffie, D., Krishna, K., 2006. Comparative effects of RRR-a and RRR-g-tocopherol
on proliferation and apoptosis in human colon cancer cell lines. BMC Cancer 6, 13-26.
Carvalho, J., Silva, M.M.C., Moreira, J.N., Simões, S., Sá e Melo, M.L., 2011. Selective
cytotoxicity of oxysterols through structural modulation on rings A and B. Synthesis, in
vitro evaluation and SAR. J. Med. Chem. 54, 6375–6393.
25

ACCEPTED MANUSCRIPT
Cross, S., Baroni, M., Carosati, E., Benedetti, P., Clementi, S., 2010. FLAP: GRID Molecular
Interaction Fields in Virtual Screening. Validation using the DUD Data Set. J. Chem. Inf.
Model. 50, 1442–1450.
Djuric, Z., Heilbrun, L.K., Lababidi, S., Everett-Bauer, C.K., Fariss, M.W., 1997. Growth
inhibition of MCF-7 and MCF-10A human breast cells by alpha-tocopheryl
hemisuccinate, cholesteryl hemisuccinate and their ether analogs, Cancer Lett. 111, 133-
139.
Dong, L.F., Low, P., Dyason, J.C., Wang, X.F., Prochazka, L., Witting, P.K., Freeman, R.,
Swettenham, E., Valis, K., Liu, J., Zobalova, R., Turanek, J., Spitz, D.R., Domann, F.E.,
Scheffler, I.E., Ralph, S.J., Neuzil, J., 2008. α-Tocopheryl succinate induces apoptosis by
targeting ubiquinone-binding sites in mitochondrial respiratory complex II. Oncogene 27,
4324-4335.
dos Santos, G.A., Abreu e Lima, R.S., Pestana, C.R., Lima, A.S., Scheucher, P.S., Thomé, C.H.,
Gimenes-Teixeira, H.L., Santana-Lemos, B.A., Lucena-Araujo, A.R., Rodrigues, F.P.,
Nasr, R., Uyemura, S.A., Falcão, R.P., de Thé, H., Pandolfi, P.P., Curti, C., Rego, E.M.,
2012. (+)α-Tocopheryl succinate inhibits the mitochondrial respiratory chain complex I
and is as effective as arsenic trioxide or ATRA against acute promyelocytic leukemia in
vivo. Leukemia 26, 451–460.
Eitenmiller, T., Lee, J., 2004. Vitamin E: Food Chemistry, Composition, and Analysis. Marcel
Dekker Inc, New York.
26

ACCEPTED MANUSCRIPT
Elnagar, A.Y., Wali, V.B., Sylvester, P.W., El Sayed, K.A., 2010. Design and preliminary
structure-activity relationship of redox-silent semisynthetic tocotrienol analogues as
inhibitors for breast cancer proliferation and invasion. Bioorg. Med. Chem. 18, 755-768.
Gagic, Z., Nikolic, K., Ivkovic, B., Filipic, S., Agbaba, D., 2016. QSAR studies and design of
new analogs of vitamin E with enhanced antiproliferative activity on MCF-7 breast
cancer cells. JTICE 59, 33-44.
Goldstein, J.L., DeBose-Boyd, R.A., Brown, M.S., 2006. Protein sensors for membrane sterols.
Cell 124, 35-46.
Guan, F., Li, G., Liu, A.B., Lee, M.J., Yang, Z., Chen, Y.K., Lin, Y., Shih, W., Yang, C.S., 2012.
δ- and γ-tocopherols, but not α-tocopherol, inhibit colon carcinogenesis in azoxymethane-
treated F344 rats. Cancer Prev. Res. 5, 644-654.
Huang, P.H., Chuang, H.C., Chou, C.C., Wang, H., Lee, S.L., Yang, H.C., Chiu, H.C., Kapuriya,
N., Wang, D., Kulp, S.K., Chen, C.S., 2013. Vitamin E facilitates the inactivation of the
kinase Akt by the phosphatase PHLPP1. Sci. Signal. 6, ra19.
Huang, P.H., Wang, D., Chuang, H.C., Wei, S., Kulp, S.K., Chen, C.S., 2009. α-Tocopheryl
succinate and derivatives mediate the transcriptional repression of androgen receptor in
prostate cancer cells by targeting the PP2A-JNK-Sp1-signaling axis. Carcinogenesis 30,
1125–1131.
Jiang, Q., 2014. Natural forms of vitamin E: metabolism, antioxidant, and anti-inflammatory
activities and their role in disease prevention and therapy. Free Radic. Biol. Med. 72, 76-
90.
27

ACCEPTED MANUSCRIPT
Katsoyannis, P.G., Gish, D.T., Hess, G.P., du Vigneaud, V., 1958. Synthesis of two protected
hexapeptides containing the N-terminal and C-terminal sequence of arginine-vasopressin.
J. Am. Chem. Soc. 80, 2558-2562.
Konsoula, R., Jung, M., 2008. In vitro plasma stability, permeability and solubility of
mercaptoacetamide histone deacetylase inhibitors. Int. J. Pharm. 361, 19–25.
Koudelka, S., Turanek Knotigova, P., Masek, J., Prochazka, L., Lukac, R., Miller, A.D., Neuzil,
J., Turanek, J., 2015. Liposomal delivery systems for anti-cancer analogues of vitamin E.
J. Control. Release 207, 59–69.
Ling, T.M., Luk, S.U., Al-Ejeh, F., Khanna, K.K., 2012. Tocotrienol as a potential anticancer
agent. Carcinogenesis 33, 233-239.
Liu, L.Y., Li, Z.H., Ding, Z.H., Dong, Z.J., Li, G.T., Li, Y., Liu, J.K., 2013. Meroterpenoid
pigments from the basidiomycete Albatrellus ovinus. J. Nat. Prod. 76, 79-84.
Marra, F., Ostacolo, C., Laneri, S., Bernardi, A., Sacchi, A., Padula, C., Nicoli, S., Santi, P.,
2009. Synthesis, hydrolysis, and skin retention of amino acid esters of alpha-tocopherol.
J. Pharm. Sci. 98, 2364-2376.
McIntyre, B.S., Brisk, K.P., Gapor, A., Sylvester, P.W., 2000. Antiproliferative and apoptotic
effects of tocopherols and tocotrienols on preneoplastic and neoplastic mouse mammary
epithelial cells. Proc. Soc. Exp. Biol. Med. 224, 292-301.
Molecular Discovery Ltd., 2015. FLAP, Version 2.1.0. Perugia, Italy, 2015.
28

ACCEPTED MANUSCRIPT
Montcourrier, P., Mangeat, P.H., Valembois, C., Salazar, G., Sahuquet, A., Duperray, C.,
Rochefort, H., 1994. Characterization of very acidic phagosomes in breast cancer cells
and their association with invasion. J. Cell. Sci. 107, 2381–2391.
Mosmann, T., 1983. Rapid colorimetric assay for cellular growth and survival: application to
proliferation and cytotoxicity assays, J. Immunol. Methods 65, 55-63.
Moya-Camarena, S.Y., Jiang, Q., 2012. The role of vitamin E forms in cancer prevention and
therapy-studies in human intervention trials and animal models, in: Sarkar, F.H. (Ed.),
Nutraceuticals and Cancer. Springer, New York, pp. 323-354.
Neises, B., Steglich, W., 1978. Simple Method for the Esterification of Carboxylic Acids.
Angew. Chem. Int. Ed. 17, 522–524.
Neophytou, C.M., Constantinou, A.I., 2015. Drug delivery innovations for enhancing the
anticancer potential of Vitamin E isoforms and their derivatives. Biomed Res. Int. 2015
584862 doi:10.1155/2015/584862.
Neuzil, J., Tomasetti, M., Mellick, A.S., Alleva, R., Salvatore, B.A., Birringer, M., Fariss, M.W.,
2004. Vitamin E analogues: a new class of inducers of apoptosis with selective anti-
cancer effect. Curr. Cancer Drug Targets 4, 267–284.
Neuzil, J., Weber, T., Schroder, A., Lu, M., Ostermann, G., Gellert, N., Mayne, G.C., Olejnicka,
B., Nègre-Salvayre, A., Stícha, M., Coffey, R.J., Weber, C., 2001. Induction of cancer
cell apoptosis by alpha-tocopheryl succinate: molecular pathways and structural
requirements. FASEB J. 15, 403-415.
29