X-ray structural analysis, antioxidant and cytotoxic activity of newly synthesized salicylic acid derivatives

Article abstract of DOI:10.1007/s11224-009-9524-y

New salicylic (2-hydroxybenzoic) acid derivatives 1-6 were prepared by conventional heating or microwave irradiation of a mixture consisting of methyl salicylate and the corresponding amino alcohol (2,2′-dihydroxydiethylamine, 2,2′,2″-trihydroxytriethylamine or N-phenyl-2,2′-dihydroxydiethylamine) and metallic sodium as catalyst. For compounds 1, 3, and 5 X-ray structure analysis was performed, as well as molecular mechanics calculations (MMC), to define their conformation in terms of their energy minima. Comparison of crystal and MMC structures for these three compounds (1, 3, and 5) revealed that the intramolecular hydrogen bonds play an important role, stabilizing conformation of the most part of the molecule. The antioxidant activity and cytotoxicity of the synthesized derivatives were evaluated in a series of in vitro tests. The newly synthesized compounds exhibited strong activity against hydroxyl radical, as well as promising lipid peroxidation inhibition. The study showed that the electronic effects of the groups at the N atom are responsible for neutralization of the OH radical, i. e., antioxidant activity. Compounds 1-3 exhibited sub-micromolar cytotoxicity against HeLa S3, whereas compounds 1, 3 and 5 efficiently inhibited the growth of PC3 cells. Springer Science+Business Media, LLC 2009.

Full text of DOI:10.1007/s11224-009-9524-y

Struct Chem (2010) 21:67–78
DOI 10.1007/s11224-009-9524-y
ORIGINAL RESEARCH
X-ray structural analysis, antioxidant and cytotoxic activity
of newly synthesized salicylic acid derivatives
•
Evgenija A. Djurendic Sanja V. Dojcinovic-Vujaskovic Marija N. Sakac
•
•
´
Emilija Dj. Jovin Vesna V. Kojic Gordana M. Bogdanovic
ˇ
´
ˇ
´
ˇ
•
•
•
´
Olivera R. Klisuric Slobodanka M. Stankovic Dusan V. Lazar
´
•
•
•
´
Laslo Fabian Katarina M. Penov-Gasi
´
ˇ
•
ˇ
Received: 3 August 2009 / Accepted: 3 October 2009 / Published online: 22 October 2009
ꢀ Springer Science+Business Media, LLC 2009
Abstract New salicylic (2-hydroxybenzoic) acid deriva-
tives 1–6 were prepared by conventional heating or micro-
wave irradiation of a mixture consisting of methyl salicylate
and the corresponding amino alcohol (2,20-dihydroxydieth-
ylamine, 2,20,200-trihydroxytriethylamine or N-phenyl-2,20-
dihydroxydiethylamine) and metallic sodium as catalyst. For
compounds 1, 3, and 5 X-ray structure analysis was per-
formed, as well as molecular mechanics calculations
(MMC), to define their conformation in terms of their energy
minima. Comparison of crystal and MMC structures for
these three compounds (1, 3, and 5) revealed that the intra-
molecular hydrogen bonds play an important role, stabilizing
conformation of the most part of the molecule. The antiox-
idant activity and cytotoxicity of the synthesized derivatives
were evaluated in a series of in vitro tests. The newly syn-
thesized compounds exhibited strong activity against
hydroxyl radical, as well as promising lipid peroxidation
inhibition. The study showed that the electronic effects of the
groups at the N atom are responsible for neutralization of the
OH radical, i.e., antioxidant activity. Compounds 1–3
exhibited sub-micromolar cytotoxicity against HeLa S3,
whereas compounds 1, 3 and 5 efficiently inhibited the
growth of PC3 cells.
Keywords Salicylic acid derivatives ꢀ Microwave
assisted synthesis ꢀ Antioxidant and cytotoxic activities ꢀ
Molecular mechanics calculations ꢀ X-ray structural
analysis
Introduction
Reactive products of oxygen are among the most potent
and omnipresent threats for living organisms. Intracellular
accumulation of reactive oxygen species such as superox-
ide anion, hydrogen peroxide, hydroxyl radical, and peroxy
radical, can arise from toxic insults or normal metabolic
processes. These species may perturb the cell’s natural
antioxidant defense systems, resulting in damage to all of
the major classes of biological macromolecules (e.g., lip-
ids, nucleic acids, and proteins) [1]. Furthermore, radical
reactions play a significant role in the development of life-
limiting chronic diseases such as cancer, hypertension,
arteriosclerosis, rheumatism, and/or neurodegenerative
diseases [2–5]. For that reason the search for new mole-
cules with antioxidant properties is a very active domain of
research.
´
ˇ
´
ˇ
´
ˇ
E. A. Djurendic ꢀ S. V. Dojcinovic-Vujaskovic ꢀ M. N. Sakac ꢀ
ˇ
E. Dj. Jovin ꢀ K. M. Penov-Gasi
Department of Chemistry, Faculty of Sciences, University of
´
Novi Sad, Trg Dositeja Obradovica 3, 21000 Novi Sad, Serbia
´
´
V. V. Kojic ꢀ G. M. Bogdanovic
Oncology Institute of Vojvodina, Institutski put 4, 21204
Sremska Kamenica, Serbia
Many phenol substances of plant origin [6, 7] and syn-
thetic products [8, 9] belong to powerful antioxidants. They
have also been found to be effective at reducing the inci-
dence of various types of carcinomas [6, 8, 10, 11].
On the other hand, some salicylic acid derivatives
exhibit cytotoxicity against PC-3 (prostate cancer cells)
[12], MCF-7 (human breast adenocarcinoma ER? cells)
´
´
O. R. Klisuric (&) ꢀ S. M. Stankovic ꢀ D. V. Lazar
Department of Physics, University of Novi Sad, Trg Dositeja
´
Obradovica 4, 21000 Novi Sad, Serbia
e-mail: olivia@uns.ac.rs
L. Fabian
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK
123

68
Struct Chem (2010) 21:67–78
[13, 14], MDA-MB 231 (human breast adenocarcinoma
ER- cells) [14], and other tumor cell lines.
sodium in the mole ratios as in the above procedure was
heated to 110 ꢁC. When sodium reaction was completed
the mixture was cooled to room temperature, and then
irradiated using a MW source of 170 W at 150 ꢁC for
10 min. Further treatment was the same as in the previous
procedure. Chromatographic separation of crude product
on silica gel column (100 g, hexane, then hexane:acetone,
9:1) afforded pure products 1 and 3–6.
In order to investigate the antioxidant and cytotoxic
activities we applied conventional heating and microwave
(MW) irradiation methods to synthesize mono-, bis- and
tris-derivatives of salicylic acid 1–6. Besides, we deter-
mined the X-ray crystal structure of compounds 1, 3, and 5,
discussed the hydrogen bonds involved and performed
molecular mechanics calculations (MMC) for their crys-
talline states.
N-(2-hydroxybenzoyl)-2,20-bis(2-hydroxybenzoyloxy)die
thylamine (1). Colorless crystals (10% by conventional
heating, 2.5% by MW, mp 97–98 ꢁC from acetone–
hexane). IR (KBr): 3450–3150, 3140, 2960, 2880, 1675,
1620, 1590, 1490, 1460, 1390, 1370, 1260, 1220, 1170,
1100. 1H NMR (CDCl3): 3.99 (t, 4H, 2COOCH2CH2,
J = 5.3 Hz); 4.59 (t, 4H, 2COOCH2CH2, J = 5.3 Hz);
6.77–7.70 (m, 12H, from H); 8.80 (bs, 1H, OH); 10.56
(s, 2H, 2OH from A,B rings). 13C NMR (CDCl3): 47.27
(2COOCH2CH2); 62.47 (2COOCH2CH2); 111.69 (2C-1
from A,B rings); 117.63 (2C-3 from A,B rings); 117.81
(C-3 from C ring); 118.45 (C-1 from C ring); 119.23
(C-5 from C ring); 119.29 (2C-5 from A,B rings); 127.75
(C-6 from C ring); 129.67 (2C-6 from A,B rings), 132.41
(C-4 from C ring), 136.09 (2C-4 from A,B rings), 156.94
(C-2 from C ring); 161.67 (2C-2 from A,B rings); 169.77
(2COO); 172.63 (N–C = O). MS (m/z, rel %): 465 (M?,
4); 328 (46); 190 (52); 121 (100); 120 (31); 92 (27); 65
(21); 56(19). For C25H23O8N (465.4) calculated: 64.52%
C, 4.95% H 3.01% N; found: 64.99% C, 5.10% H,
3.47% N.
N-methyl-2,20-bis(2-hydroxybenzoyloxy)diethylamine (2).
Colorless oil (3.2% by conventional heating). IR (film): 3350,
3050, 2960, 2880, 1675, 1615, 1590, 1490, 1470, 1300, 1260,
1160, 1080. 1H NMR (CDCl3): 2.47 (s, 3H, CH3); 2.92 (t, 4H,
2COOCH2CH2, J = 5.7 Hz); 4.47 (t, 4H, 2COOCH2CH2,
J = 5.7 Hz); 6.84 (m, 2H, 2H-5 from A,B rings,
J(3,5) = 1.1 Hz, J(4,5) = 7.1 Hz, J(5,6) = 8.1 Hz); 6.99 (m,
2H, 2H-3 from A,B rings, J(3,4) = 8.4 Hz, J(3,5) = 1.1 Hz);
7.44 (m, 2H, 2H-4, J(3,4) = 8.4 Hz, J(4,5) = 7.1 Hz,
Experimental
Chemical synthesis
General
Melting points were determined using a Bu¨chi SMP 20
apparatus and are uncorrected. Infrared spectra (wave
numbers in cm-1) were recorded in KBr pellets (for crystals)
or as film (for oils) on a NEXUS 670 SP-IR spectrometer;
UV/VIS–CECIL CE2021 spectrophotometer. NMR spectra
were taken on a Bruker AC 250E spectrometer operating at
250 MHz (proton) and 62.5 MHz (carbon) with tetrameth-
ylsilane as the internal standard. Chemical shifts are given in
ppm (d-scale). Mass spectra were recorded on a Finnigan
MAT 8230 instrument, using chemical ionization (isobu-
tane) or electron impact (70 eV) technique; the first number
denotes m/e value, and the ion abundances are given in
parentheses. The microwave reactor was a monomode sys-
tem (Microwave Synthesis System—Discover Bench Mate
from CEM) with focused waves. All reagents used were of
analytical grade.
General procedure for preparation of compounds 1–6
Conventional heating A mixture consisting of methyl
2-hydroxy benzoate (30 or 20 mmol), the corresponding
aminoalcohol (2,20-dihydroxydiethylamine, 2,20,200-tri-
hydroxytriethylamine or N-phenyl-2,20-dihydroxydiethyl-
amine; 10 mmol) and sodium (1 mmol) was heated at
150 ꢁC with continuous removal of methanol for 2.5 h.
The reaction mixture was diluted with water (100 mL) and
extracted with dichloromethane (2 9 50 mL) and ethyl
acetate (2 9 50 mL). After drying over anhydrous Na2SO4
and removal of the solvent in vacuum, the result was an
oily mixture of products. The pure compounds 1–6 were
obtained after chromatography on a column of silica gel
(100 g, toluene:ethyl acetate, 9:1).
J(4,6) = 1.8 Hz); 7.75 (dd, 2H, 2H-6 from A,B rings, J(4,6)
=
1.8 Hz, J(5,6) = 8.1 Hz);10.65 (bs, 2H, 2OH from A,B rings).
13C NMR (CDCl3): 42.85 (CH3); 55.81 (2COOCH2CH2);
62.94 (2COOCH2CH2); 112.36 (2C-1 from A,B rings);
117.53 (2C-3 from A,B rings); 119.12 (2C-5 from A,B rings);
129.83 (2C-6 from A,B rings); 135.66 (2C-4 from A,B
rings);161.50(2C-2 fromA,B rings);169.89 (2COO). MS(m/z,
rel %): 359 (M?, 18); 221 (38); 208 (98); 165 (42); 121 (52);
88 (100); 93 (12); 83 (14); 70 (25); 65 (18); 58 (14); 44 (34).
2,20,200-tris(2-hydroxybenzoyloxy)triethylamine (3). Col-
orless crystals (30% by conventional heating, 9% by MW,
mp 89 ꢁC from acetone–hexane). IR (KBr): 3450–3150,
3080, 2960, 2870, 1670, 1625, 1590, 1490, 1470, 1360;
1
Microwave irradiation A mixture consisting of methyl
salicylate, the corresponding amino alcohol and metallic
1300, 1260, 1160, 1100. H NMR (CDCl3): 3.09 (t, 6H,
3COOCH2CH2, J = 5.7 Hz); 4.45 (t, 6H, 3COOCH2CH2,
123

Struct Chem (2010) 21:67–78
69
J = 5.7 Hz); 6.79 (m, 3H, 3H-5 from A,B,C rings,
J(3,5) = 1.1 Hz, J(4,5) = 7.1 Hz, J(5,6) = 8.1 Hz); 6.92 (dd,
Ph); 129.72 and 129.91 (C-3 and C-5 from Ph, 2C-6 from A,B
rings); 136.02 (2C-4 from A,B rings); 147.11 (C-1 from Ph);
161.83 (2C-2 from A,B rings); 170.16 (2COO). MS (m/z, rel.
%): 422 (M??1, 100); 284 (5). For C24H23O6N (421.4) cal-
culated: 68.40% C; 5.50% H; 3.30% N; found: 68.19% C;
5.52% H; 3.50% N.
N-phenyl-2-hydroxy-20-(2-hydroxybenzoyloxy)diethyl-
amine (6). Colorless oil (22% by conventional heating,
27% by MW). IR (film): 3450–3100, 2890, 2940, 1675,
1600, 1470, 1390, 1300, 1210, 1160, 980. 1H NMR
(CDCl3): 2.33 (bs, 1H, OH); 3.59 (t, 2H, CH2OH,
J = 5.8 Hz); 3.76–3.86 (m, 4H, CH2N(Ph)CH2); 4.54
(t, 2H, COOCH2, J = 6.1 Hz); 6.75–6.94 (m, 4H, H-5,
H-3 from A ring; H-2, H-4, H-6 from Ph); 7.00 (dd, 1H,
H-3 from A ring, J(3,4) = 8.4 Hz, J(3,5) = 0.9 Hz); 7.28
3H, 3H-3 from A,B,C rings, J(3,4) = 8.4 Hz, J(3,5)
1.1 Hz); 7.41 (m, 3H, 3H-4 from A,B,C rings, J(3,4)
=
=
8.4 Hz, J(4,5) = 7.1 Hz, J(4,6) = 1.8 Hz); 7.75 (dd, 3H, 3H-6
from A,B,C rings, J(5,6) = 8.1 Hz, J(4,6) = 1.8 Hz); 10.69
(s, 3H, 3OH from A,B,C rings). 13C NMR (CDCl3): 53,29
(3COOCH2CH2); 63.19 (3COOCH2CH2); 112.11 (3C-1
from A,B,C rings); 117.52 (3C-3 from A,B,C rings); 119.13
(3C-5 from A,B,C rings); 129.61 (3C-6 from A,B,C rings);
135.73 (3C-4 from A,B,C rings); 161.57 (3C-2 from A,B,C
rings); 169.93 (3COO). MS (m/z, rel. %): 508 (M?-1, 9); 358
(70); 238 (40); 165 (60); 121 (100); 118 (57); 100 (18); 69
(14); 65 (22); 56 (20). For C27H27O9N (509.5) calculated:
63.65% C, 5.34% H, 2.75% N 2.75% found: 63.55% C,
5.53% H, 3.01% N.
(dd, 2H, H-3 and H-5 from Ph, J(2,3) = 8.1 Hz, J(3,4)
7.3 Hz); 7.48 (m, 1H, H-4 from A ring, J(4,5) = 7.8 Hz,
J(4,6) = 1.7 Hz); 7.75 (dd, 1H, H-6 from ring,
=
2-Hydroxy-20,200-bis(2-hydroxybenzoyloxy)triethylamine
(4). Colorless oil (40.5% by conventional heating, 19% by
MW). IR (film): 3450–3150, 2950, 2840, 1670, 1620, 1590,
A
J(5,6) = 8.0 Hz, J(4,6) = 1.7 Hz); 10.76 (s, 1H, OH from
A ring). 13C NMR (CDCl3): 50.21 (COOCH2CH2); 53.61
(CH2OH); 59.75 (NCH2CH2OH); 62.36 (COOCH2CH2);
112.03 (C-1 from A ring); 112.95 (C-2 and C-6 from
Ph); 117.46 (C-4 from Ph); 117.50 (C-3 from A ring);
119.16 (C-6 from A ring); 129.37 (C-3 and C-5 from
Ph); 129.76 (C-6 from A ring); 135.79 (C-4 from A
ring); 147.54 (C-1 from Ph); 161.53 (C-2 from A ring);
170.03 (COO). MS (m/z, rel. %): 302 (M??1, 100); 182
(4); 164 (9); 137 (4).
1
1490, 1300, 1250, 1220, 1160, 1080. H NMR (CDCl3):
2.84 (t, 2H, NCH2CH2OH, J = 5.3 Hz); 3.03 (t, 4H, 4H,
2COOCH2CH2, J = 4.4 Hz); 3.64 (t, 2H, NCH2CH2
OH, J = 5.3 Hz); 3.80 (bs, 1H, OH), 4.43 (t, 4H,
2COOCH2CH2, J = 4.4 Hz); 6.80 (m, 2H, 2H-5 from A,B
rings, J(3,5) = 1.1 Hz, J(4,5) = 7.1 Hz, J(5,6) = 8.1 Hz);
6.91 (dd, 2H, 2H-3 from A,B rings, J(3,4) = 8.4 Hz,
J(3,5) = 1.1 Hz); 7.41 (m, 2H, 2H-4, J(3,4) = 8.4 Hz,
J(4,5) = 7.1 Hz, J(4,6) = 1.8 Hz); 7.75 (dd, 2H, 2H-6 from
A,B rings, J(4,6) = 1.8 Hz, J(5,6) = 8.1 Hz); 10.61 (bs, 2H,
2OH from A,B rings). 13C NMR (CDCl3): 52.86 (2CO-
OCH2CH2); 56.82 (NCH2CH2OH); 58.84 (NCH2CH2OH);
62.90 (2COOCH2CH2); 111.91 (2C-1 from A,B rings);
117.45 (2C-3 from A,B rings); 119.17 (2C-5 from A,B
rings); 129.50 (2C-6 from A,B rings); 135.74 (2C-4 from
A,B rings); 161.50 (2C-2 from A,B rings); 169.94 (2COO).
MS (m/z, rel. %): 389 (M?, 16); 358 (52) 238 (72); 165
(54); 121 (72); 118 (100); 65 (12); 56 (19).
Crystal structure determination
The diffraction data for compounds 1 and 5 were collected
at room temperature on an Oxford Diffraction Gemini S
diffractometer with graphite-monochromated MoKa radi-
˚
ation (k = 0.7107 A) and for compound 3 also at room
temperature using Enraf–Nonius CAD-4 diffractometer
˚
with CuKa radiation (k = 1.5418 A). The data reduction
N-phenyl-2,20-bis(2-hydroxybenzoyloxy)diethylamine (5).
Colorless crystals (49% by conventional heating, 8.5%
by MW, mp 121–122 ꢁC from acetone–hexane). IR (KBr):
3500–3100, 2960, 1670, 1600, 1460, 1400, 1210, 1160, 1060,
1090. 1H NMR (CDCl3): 3.87 (t, 4H, 2COOCH2CH2,
J = 6.1 Hz);4.57(t, 4H, 2COOCH2CH2, J = 6.1 Hz); 6.76–
6.95 (m, 5H, 2H-5 from A,B rings; H-2, H-6, and H-4 from
Ph); 7.01 (dd, 2H, 2H-3 from A,B rings, J(3,4) = 8.4 Hz,
J(3,5) = 0.9 Hz); 7.32 (dd, 2H, H-3 and H-5 from Ph,
J(2,3) = 8.1 Hz, J(3,4) = 7.3 Hz); 7.48 (m, 2H, 2H-4 from
A,B rings, J(3,4) = 8.4 Hz, J(4,6) = 1.7 Hz, J(4,5) = 7.8 Hz);
7.81 (dd, 2H, 2H-6 from A,B rings, J(4,6) = 1.8 Hz,
J(5,6) = 8.0 Hz); 10.74 (s, 2H, 2OH from A,B rings).
13C NMR (CDCl3): 49.94 (2COOCH2CH2); 62.51
(2COOCH2CH2); 112.17 (2C-1 from A,B rings); 112.64 (C-2
and C-6 from Ph); 117.75 (2C-3 from A,B rings and C-4 from
for compounds 1 and 5 were performed with program
package CrysAlis RED [15], and for compound 3 with
program package XCAD4 [16]. The space group determi-
nations were based on an analysis of the Laue class and the
systematically absent reflections. The structures 1 and 5
were solved by direct methods using SIR92 [17], and the
structure 3 was solved using SHELXS97 [18]. All struc-
tures were refined using full-matrix least-squares. For all
three compounds non-hydrogen atoms were refined aniso-
tropically, the C–H hydrogen atoms were included on
calculated positions riding on their attached atoms with
˚
fixed distances 0.93 (CH) or 0.97 A (CH2) and all O–H
hydrogen atoms were identified on difference electron
density maps and isotropically refined. All calculations
were performed using SHELXL97 [18], PARST [19], and
PLATON [20], as implemented in the WINGX [21] system
123

70
Struct Chem (2010) 21:67–78
Table 1 The crystal data and refinement parameters
(1)
(3)
(5)
Chemical formula
Mr
C25H23NO8
465.44
C27H27NO9
509.50
C24H23NO6
421.43
Cell setting, space group
Temperature (K)
Monoclinic, P21/c
297 (2)
Monoclinic, C2/c
293 (2)
Monoclinic, P21/c
297 (2)
˚
a (A); b (A); c (A)
˚
˚
15.9876 (7); 14.0448 (6);
10.8187 (4)
37.5130 (10); 10.3100 (10); 12.813 (2) 7.8021 (4); 30.9049 (17); 8.9008 (4)
b (ꢁ)
V (A )
105.559 (5)
90.900 (10)
4954.9 (9)
8
106.112 (5)
3
˚
2340.24 (17)
2061.89 (18)
Z
4
4
Dx (mg m-3
Radiation type
l (mm-1
)
1.321
1.366
1.358
MoKa
CuKa
MoKa
)
0.10
0.86
0.10
Crystal form, color
Crystal size (mm)
Diffractometer
Data collection method
Absorption correction
Tmin
Prism, colorless
Prism, colorless
0.53 9 0.35 9 0.20
Enraf–Nonius CAD4
x–2h
Prism, colorless
0.50 9 0.25 9 0.21
Gemini S; Oxford Diffraction
x–h-scans
0.44 9 0.20 9 0.18
Gemini S; Oxford Diffraction
x–h-scans
None
w scan
Multi-scan
–
0.727
0.953
Tmax
–
0.841
0.987
No. of measured,
independent and observed
reflections
14193, 5334, 2460
10514, 5099, 4244
12011, 4738, 2752
Criterion for observed
reflections
I [ 2r(I)
I [ 2r(I)
I [ 2r(I)
Rint
0.023
29.5
F2
0.010
0.025
hmax (ꢁ)
Refinement on
75.0
F2
29.5
F2
R[F2 [ 2r(F2)], wR(F2), S 0.039, 0.099, 0.86
0.045, 0.145, 1.09
5099 reflections
347
0.058, 0.136, 1.04
4738 reflections
288
No. of relections
No. of parameters
H-atom treatment
5334 reflections
319
Mixture of independent and
constrained refinement
Mixture of independent and
constrained refinement
Mixture of independent and
constrained refinement
Weighting scheme
Calculated w = 1/
[r2(F2o) ? (0.0515P)2] where
P = (F2o ? 2F2c)/3
Calculated w = 1/
[r2(F2o) ? (0.0691P)2 ? 1.8743P]
where P = (F2o ? 2Fc2)/3
Calculated w = 1/
[r2(F2o) ? (0.0457P)2 ? 0.6147P]
where P = (F2o ? 2Fc2)/3
(D/r)max
\0.0001
0.13, -0.12
0.004
\0.0001
0.13, -0.16
-3
)
˚
Dqmax, Dqmin (e A
0.27, -0.15
DPPH assay The DPPH-assay was performed as described
before [22], with small modifications. The different aliquots
(0.10–2.00 mL) of 0.01 M sample solution in methanol were
addedto1.00 mLof90lmol/L DPPH• in methanol (Sigma; St.
Louis, MO) and filled up with 95% (v/v) methanol to a final
volume of 4.00 mL. The same reaction mixture without tested
compound was used as the control. Absorbances of the reaction
mixturesand control were recorded at515 nm (CECIL CE2021
spectrophotometer) after 1 h. Commercial synthetic antioxi-
dants, 3,5-di-tert-butyl-4-hydroxytoluene (BHT) (Aldrich;
of programs. The crystal data and refinement parameters
are summarized in Table 1.
Biological methods
Free radical scavenging assays
Free radical scavenging activity of examined compounds
was evaluated by measuring their ability to neutralize
DPPH- and OH-radicals.
123

Struct Chem (2010) 21:67–78
71
Taufkirchen, Germany) and 3-tert-butyl-4-hydroxyanisole
(BHA) (Fluka; Taufkirchen, Germany) were used as positive
controls. For each sample, three replicates were recorded.
DPPH scavenging activity was expressed as radical
scavenging activity (DPPH RSC). Percentage of DPPH
RSC was calculated using the following equation:
commercial preparation in demineralized water (1:10) in an
ultrasonic bath.
In the Fe2?/ascorbate-induced lipids peroxidation, a
60 lL suspension of liposomes was incubated with 20 lL
of 0.01 mol/L FeSO4, 20 lL of 0.01 mol/L ascorbic acid,
and 10 lL of tested compound and filled with 0.05 mol/L
KH2PO4–K2HPO4 buffer, pH 7.4 to a final volume of
3.00 mL. Samples were incubated at 37 ꢁC for 1 h. LP was
terminated by adding 2 mL of TBA reagent, and 0.2 mL of
EDTA, and heating the test tubes at 100 ꢁC for 20 min.
After centrifugation (4000 rpm for 10 min), the content of
the MDA (TBARS) was determined by measuring the
absorbance of the product at 532 nm.
RSC ð%Þ ¼ 100 ꢁ ðAcontrol ꢂ Asample=AcontrolÞ:
IC50 values (the concentration of the tested compound in
the reaction mixture which causes 50% of RSC) were
determined by linear regression analysis from the obtained
RSC values.
Analyses were compared with the commercial synthetic
antioxidants BHT and BHA as a positive control. All
reactions were carried out in triplicate.
Hydroxyl-radical scavenging assay Hydroxyl-radicals
scavenging capacity (HO• RSC) of the tested compounds
was evaluated by measuring the degradation of 2-deoxy-
D-ribose (Aldrich; Taufkirchen, Germany) in the reaction
with OH radicals, generated in situ in Fenton’s reaction [23].
These radicals attack the sugar 2-deoxy-D-ribose and
degrade it into a series of fragments, some or all of which
react on heating with 2-thiobarbituric acid (TBA) (Sigma; St.
Louis, MO) at low pH to give a pink chromogen, which can
be determined spectrophotometrically at 532 nm [12]. Dif-
ferent aliquots (0.005–0.5 mL) of sample solution in meth-
anol were added to test tubes (final concentration ranged
from 0.01 to 8 mmol/L), each containing 0.1 ml of 5 mmol/L
H2O2, 0.1 mL of 10 mmol/L FeSO4, and 0.1 mL of
0.05 mol/L 2-deoxy-D-ribose and 0.067 mol/L KH2PO4-
K2HPO4 buffer pH 7.4 to a final volume of 3.00 mL. The
same reaction mixture without sample was used as the con-
trol. After an incubation period of 1 h at 37 ꢁC, 2 mL of TBA
reagent (10.4 mL of 60% (v/v) HClO4, 3 g TBA and 120 g of
trichloroacetic acid (Sigma; St. Louis, MO)), and 0.2 mL of
0.1 mol/L EDTA (Sigma; St. Louis, MO) were added to the
reaction mixture, and the tubes were heated at 100 ꢁC for
20 min. After cooling, absorbances of the reaction mixtures
and of the control were recorded at 532 nm.
The percentage of LP inhibition was calculated by the
following equation:
Ið%Þ ¼ ðAo ꢂ ꢂ A1Þ=Ao ꢁ 1000
where A0 was the absorbance of the control reaction
(without the test compound) and A1 was the absorbance in
the presence of the inhibitor.
Cytotoxic activity
Cell lines
Six human tumor cell lines and one human non-tumor cell
line were used in the study: human breast adenocarcinoma
ER?, MCF-7, human breast adenocarcinoma ER-, MDA-
MB-231, prostate cancer PC3, cervix epithelioid carci-
noma, HeLa S3, human melanoma, Hs 294T, chronic
myelogenous leukemia, K562, and normal fetal lung
fibroblasts, MRC-5.
The cells were grown in Dulbecco’s modified Eagle’s
medium (DMEM) with 4.5% of glucose (MCF7; MDA-
MB-231; PC3; MRC-5) or in RPMI 1640 (K562). Media
were supplemented with 10% of fetal calf serum (FCS,
NIVNS) and antibiotics: 100 IU/mL of penicillin and
100 lg/mL of streptomycin (ICN Galenika). All cell lines
were cultured in flasks (Costar, 25 cm2) at 37 ꢁC in the
100% humidity atmosphere and 5% of CO2. Only viable
cells were used in the assay. Viability was determined by
dye exclusion assay with trypan blue.
Percentage of HO• RSC was calculated using the fol-
lowing equation:
RSC ð%Þ ¼ 100 ꢁ ðAcontrol ꢂ Asample=AcontrolÞ:
Three replicates were recorded for each sample; BHT
and BHA were used as reference compounds.
Lipid peroxidation (LP) assay The extent of LP was
determined by measuring the absorbance of adduct pro-
duced in the reaction between TBA and malondialdehyde
as an oxidation product in the peroxidation of phospho-
lipids from liposomes, by the TBA assay [23].
SRB assay
Cytotoxicity was evaluated by colorimetric sulforhodamine
B (SRB) assay [24]. Briefly, single cell suspension was
plated into 96-well microtiter plates (Costar, flat bottom):
5 9 103 cells (MCF7; MDA-MB-231; PC3; MRC-5) or
104 (K562) cells per 180 mL of medium. Plates were
The commercial preparation of liposomes ‘‘PRO-LIPO
S’’ (Lucas-Meyer, Hamburg, Germany), pH 5–7, was used as
a model system of biological membranes. The liposomes,
225–250 nm in diameter, were obtained by dispersing the
123

72
Struct Chem (2010) 21:67–78
pre-incubated 24 h at 37 ꢁC, 5% CO2. Tested substances at
concentrations ranging from 10-8 to 10-4 mol/L were added
to all wells except for the control ones. After incubation
period (48 h/37 ꢁC/5% CO2) SRB assay was carried out as
follows: 50 lL of 80% trichloroacetic acid (TCA) was added
to all wells; an hour later the plates were washed with dis-
tilled water, and 75 lL of 0.4% SRB was added to all wells;
half an hour later the plates were washed with citric acid (1%)
and dried at room temperature. Finally, 200 lL of 10 mmol
Tris (pH = 10.5) basis was added to all wells. Absorbance
(A) was measured on the microplate reader (Multiscan
MCC340, Labsystems) at 540/690 nm. The wells without
cells, containing complete medium only, acted as blank.
Cytotoxicity was calculated according to the formula:
amino alcohols: 2,20-dihydroxydiethylamine, 2,20,200-trihy-
droxy triethylamine and N-phenyl-2,20-dihydroxydiethyl-
amine. The reactions were carried out at 150 ꢁC for 2.5 h in
the presence of sodium as catalyst (Scheme 1).
When methyl 2-hydroxybenzoate and the corresponding
aminoalcohol 2,20-dihydroxydiethylamine were taken in a
molar ratio 3:1, the tris-derivative of 2-hydroxybenzoic
acid N-(2-hydroxybenzoyl)-2,20-bis(2-hydroxybenzoyloxy)
diethylamine (1), and bis-derivative N-methyl-2,20-bis
(2-hydroxybenzoyloxy)diethylamine (2) were obtained in
the yields of 10 and 3.2%, respectively.
Using 2,20,200-trihydroxytriethylamine under the same
reaction conditions, the tris-derivative 2,20,200-tris(2-hydrox-
ybenzoyloxy)triethylamine (3) (30%) and bis-derivative
2-hydroxy-20,200-bis(2-hydroxybenzoyloxy)triethylamine (4)
(40.5%) were obtained.
ð1 ꢂ Atest= AcontrolÞ100:
and expressed as a percent of cytotoxicity (CI%).
The transesterification using methyl 2-hydroxybenzoate
and N-phenyl-2,20-dihydroxydiethylamine in a molar ratio
2:1, yielded the bis-derivative N-phenyl-2,20-bis(2-hy-
droxybenzoyloxy)diethylamine (5) (49%) and mono
derivative N-phenyl-2-hydroxy-20-(2-hydroxybenzoyloxy)
diethylamine (6) (22%).
Two independent experiments were set out in quadruplicate
for each concentration of the compound. IC50 value defines the
dose of compound that inhibits cell growth by 50%. The IC50
of compounds was determined by median effect analysis.
The preparation of compound 2 can be explained in terms
of the known but still unusual disruption of the alkyl-oxygen
bond in the methyl ester of 2-hydroxybenzoic acid, whereby
the formed methyl cation binds to the nitrogen atom of the
amino alcohol, yielding either N-alkylated amino alcohol or
Results and discussion
Chemistry
Compounds 1–6 were prepared by transesterification of
methyl salicylate (methyl 2-hydroxybenzoate) with selected
the subsequently N-alkylated bis derivative 2 (CH3O-
?
COC6H4OH ? CH3
?
-OCOC6H4OH). The N-methyl
O
CH3
O
OH
OH
H
N
N
HO
OH
HO
OH
N
Na, 150 oC, 2.5h
Na, 150 oC, 2.5h
HO
OH
Na, 150 oC, 2.5h
HO
C
HO
C
O
O
O
O
O
O
O
O
O
O
N
N
O
N
O
O
O
O
A
A
B
B
A
B
A
OH
O
HO
O
OH
O
5
HO
1
OH
O
3
HO
OH
CH3
N
O
N
N
O
O
O
O
O
OH
A
B
B
A
OH
4
HO
OH
HO
OH
6
2
Scheme 1 Synthesis of salicylic acid derivatives 1–6: a Na, 150 ꢁC, 2.5 h; b Na, 110 ꢁC, 5 min, than MW, 170 W, 150 ꢁC, 10 min
123

Struct Chem (2010) 21:67–78
73
cation formed as an intermediate can be stabilized by de-
protonation only in the case of 2,20-dihydroxydiethylamine.
Compounds 1 and 3–6 were also obtained by applying
MW irradiation (170 W) of the mixture of methyl
2-hydroxybenzoate, the corresponding amino alcohol and
metallic sodium as catalyst at 150 ꢁC for 10 min. Under
these conditions, compound 2 could not be obtained,
compounds 1, 3, 4, and 5 were obtained in a lower yield
(2.5–19%), whereas compound 6 was obtained in a
somewhat higher yield (27%) than by conventional method
(22%).
Crystal structures analyses and molecular mechanics
calculations
The perspective views of molecules 1, 3, and 5 are shown
in Fig. 1 [25]. Selected bond lengths, bond angles, and
torsion angles are given in Table 2.
Since we wanted to investigate the antioxidant behavior
and cytotoxicity of compounds 1, 3, and 5, we were
interested in the conformation of molecules released from
the influence of crystalline field. The next step, after
determining the three-dimensional structures of these
compounds in crystalline state, was to define the con-
formation of the molecules 1, 3, and 5 in terms of
energy minima. To do that, we performed the MMC using
PCMODEL [26].
The conformation of molecules remains stable in their
energy minima in spite of the significant number of inter-
atomic single bonds. This is confirmed by comparing
selected torsion angles of molecules, obtained in crystal
structure analyzes and after MMC (Table 2).
Figure 2 illustrates the very good overlapping of the
molecular structures in crystalline state and after MMC for
all three molecules (1, 3, and 5). It seems that the presence
of intramolecular hydrogen bonds in all three molecules
stabilizes their conformations in spite of the possible
influence of crystal packing (Fig. 1; Table 3).
Fig. 1 ORTEP drawings of molecular structures of compound 1 (a),
3 (b), and 5 (c) with the labeling of non-H atoms. Displacement
ellipsoids are shown at the 50% probability level and H atoms are
drawn as spheres of arbitrary radii. Intramolecular hydrogen bonds
are shown as dashed lines
Comparing the structures of three molecules (1, 3, and
5) in crystalline state, we noticed the difference in N–C1c
bond length. Molecules 1 and 5 have shorter N–C1c dis-
tance than molecule 3. This is due to the resonance effect
of the free (lone) electron pair on the N atom and p-elec-
trons of the C=O bond in 1 and benzene ring in 5, which
results in partly double bond character of the N–C1c. This
effect does not appear in compound 3. In other words,
hybridization on the N atom in molecules 1 and 5 is partly
sp2 and in molecule 3 is sp3.
and C1c atoms in compounds 1 and 5 are 359.99 in both
cases (Table 2). This is also the confirmation of partial sp2
hybridization on the N atom in molecules 1 and 5. On the
other hand, the sum of the same bond angles for compound
3 is 342.68ꢁ (Table 2), confirming the sp3 hybridization on
N atom.
The best evidence of planarity around an atom is to
consider the sum of the three bond angles because a per-
fectly planar system would have a sum equal to 360ꢁ. The
sums of bond angles, which N atom forms with C1a, C1b,
After the MMC for molecule 3, the values of the same
bond angles decreased (Table 2) and improved the trigonal
configuration around N atom.
123

74
Struct Chem (2010) 21:67–78
˚
Table 2 Selected bond lengths (A), angles (ꢁ), and torsion angles (ꢁ)
(1)
(3)
(5)
Crystal structure
MMC structure
Crystal structure
MMC structure
Crystal structure
MMC structure
Bond lengths
N–C1a
1.464 (2)
1.466 (2)
1.5
1.5
1.3
1.458 (2)
1.452 (2)
1.451 (2)
1.5
1.5
1.4
1.449 (2)
1.456 (2)
1.384 (2)
1.5
1.5
1.4
N–C1b
N–C1c
1.3431 (17)
Bond angles
C1a–N–C1b
117.23 (11)
124.96 (12)
117.80 (11)
113.38 (11)
113.20 (14)
117.6
123.4
119.1
113.3
112.8
113.05 (13)
115.65 (14)
113.98 (14)
111.78 (14)
118.06 (14)
112.8
112.9
112.8
111.0
117.8
115.16 (16)
122.23 (16)
122.61 (16)
115.17 (19)
114.48 (17)
116.4
121.7
121.9
114.4
114.3
C1c–N–C1a
C1c–N–C1b
N–C1a–C2a
N–C1b–C2b
Torsion angles
C1c–N–C1a–C2a
C1c–N–C1b–C2b
C1a–N–C1b–C2b
C1b–N–C1a–C2a
N–C1a–C2a–O1a
N–C1b–C2b–O1b
C1a–C2a–O1a–C3a
C1b–C2b–O1b–C3b
C2a–O1a–C3a–C4a
C2b–O1b–C3b–C4b
C2a–O1a–C3a–O2a
C2b–O1b–C3b–O2b
O1a–C3a–C4a–C5a
O1b–C3b–C4b–C5b
-103.87 (16)
-72.34 (16)
108.52 (15)
75.21 (17)
-103.8
-70.2
111.3
74.7
-134.79 (15)
-67.8 (2)
-141.7
-60.6
68.8
-99.4 (2)
-100.5 (2)
79.3 (2)
-101.4
-100.9
79.1
66.95 (19)
91.25 (17)
71.59 (18)
67.07 (19)
-172.41 (14)
175.18 (13)
-177.68 (13)
176.76 (13)
0.9 (2)
89.0
80.8 (2)
78.6
72.59 (16)
68.0
68.2
66.3 (2)
59.3
-52.17 (17)
-104.73 (15)
176.42 (14)
175.50 (12)
-171.17 (13)
-4.2 (2)
-53.0
-97.9
176.0
-179.1
-177.0
1.2
69.7
67.6 (2)
59.0
175.8
-173.9
-178.2
176.7
1.8
-178.37 (18)
-177.42 (17)
174.78 (17)
175.70 (17)
-5.0 (3)
-177.5
-178.9
-179.1
-179.7
0.7
6.6 (3)
2.8
-3.0 (2)
-2.8
-4.2 (3)
0.3
-175.94 (14)
-179.01 (17)
-169.4
-173.0
176.24 (14)
-178.37 (14)
-167.0
171.3
-176.31 (18)
-179.18 (18)
-178.0
-175.2
The differences between the conformations (in the
region of atoms labeled with A and B, where the structural
formulas are the same for all three compounds) of mole-
cules 3 and 5 on one side and molecule 1 on the other, are
shown by the values of selected torsion angles N–C1b–
C2b–O1b [-52.17 (17)ꢁ in 1, 67.07 (19)ꢁ in 3, and 67.6
(2)ꢁ in 5] and C1a–C2a–O1a–C3a [-104.73 (15)ꢁ in 1,
-172.41 (14)ꢁ in 3, and -178.37 (18)ꢁ in 5] (Table 2).
These differences in conformations originate from the
interaction of free electron pair of the N atom with O1c
atom from the C=O group in molecule 1 [O1cꢀꢀꢀ
certain predominant conformation, and, consequently,
biological activity). As can be seen from Fig. 2, all three
compounds (1, 3, and 5) contain intramolecular O–HꢀꢀꢀO
hydrogen bonds. The hydrogen bond parameters are given
in Table 3.
Interaction of free electron pair of the N atom with the
C=O group in molecule 1 and the steric overcrowding around
the O1c atom in molecule 1 results in the position of O1c
atom that is not energy predominant for construction of
expected hydrogen bond with O2c atom. Instead of intra-, the
O1c and O2c atoms form strong inter-molecular hydrogen
˚
˚
N = 2.2313 (15) A]. Moreover, the environment around
bond O2c–H2CꢀꢀꢀO1C [DꢀꢀꢀA = 2.674 (1) A; D–HꢀꢀꢀA =
the N and O1c atoms in compound 1 is characterized by the
steric overcrowding.
179 (2) o; symmetry code: x, 1/2 - y, -1/2 ? z], which is
illustrated in Fig. 3. Crystal packings of all three compounds
(1, 3, and 5) are illustrated in Fig. 3.
Hydrogen bonding analyses
As illustrated in Fig. 3, the intramolecular O–HꢀꢀꢀO
hydrogen bonds in all three compounds (1, 3, and 5) are
also preserved in the crystalline state after the MMC. This
is the confirmation that intramolecular hydrogen bonds in
all three compounds stabilize molecular conformation.
The information on intramolecular hydrogen bonding, in
particular, is very useful to understand various molecular
properties (the molecular geometries, the stability of a
123

Struct Chem (2010) 21:67–78
75
Fig. 2 Superimposed fit of the
molecules after MMC (lighter
line) and the molecule in the
crystalline state (darker line):
(a) compound 1, (b) compound
3, and (c) compound 5
˚
test solution and prevent the sugar degradation. Generally,
all of the examined compounds inhibited the degradation of
deoxyribose more than BHT and BHA. The highest activity
exhibited compounds 3 and 5, with respective IC50 values
being about 60- and 70-fold lower than those observed for
the BHT and BHA. Of bis-derivatives, significantly lower
activity showed compound 4, while compound 2 was the
least active. It can be presumed that electronic effects of
the groups at the N atom are responsible for the neutral-
ization of OH radical by examined bis- and tris-derivatives.
Namely, compound 5 contains phenyl group at the nitrogen
atom, compound 4 has 2-hydroxyethyl group, and com-
pound 2 bears methyl group. That is the order of decreasing
of their scavenging activities 5 * 3 [ 4 [ 6 * 1 [ 2.
The tris-derivative 3, with the 2-salicyloyloxyethyl group
at the N atom expressed stronger activity than tris-deriva-
tive 1 with salicyloyl group. It can also be presumed that
the differences between conformations of molecules 3 and
5 on one side and molecule 1 on the other, shown by X-ray
structure analyses, lead to the differences in their antioxi-
dant activity. Besides, it can be presumed that the different
conformation in the region of the N atom in molecule 1,
comparing with the conformations of the same region in
the molecules 3 and 5, changes the positions and distances
Table 3 Intramolecular O-HꢀꢀꢀO hydrogen-bond parameters (A, ꢁ)
D–HꢀꢀꢀA
D–H
HꢀꢀꢀA
DꢀꢀꢀA
D–HꢀꢀꢀA
(1)
O3A–H3AꢀꢀꢀO2A
O3B–H3BꢀꢀꢀO2B
(3)
0.96 (2)
0.82 (2)
1.76 (2)
1.89 (3)
2.598 (2)
2.589 (3)
144 (2)
143 (3)
O3A–H3AꢀꢀꢀO2A
O3B–H3BꢀꢀꢀO2B
O3C–H3CꢀꢀꢀO2C
(5)
0.95 (3)
0.88 (3)
0.90 (3)
1.74 (3)
1.82 (3)
1.74 (3)
2.616 (2)
2.605 (2)
2.607 (2)
150 (3)
148 (3)
161 (2)
O3A–H3AꢀꢀꢀO2A
O3B–H3BꢀꢀꢀO2B
1.00 (3)
0.91 (4)
1.66 (4)
1.78 (4)
2.577 (3)
2.615 (3)
151 (3)
151 (3)
Antioxidant activity
The antioxidant activities of the synthesized derivatives of
salicylic acid were evaluated in a series of in vitro tests.
The hydroxyl radical scavenging activity of the exam-
ined compounds was measured by the deoxyribose assay
[23]. The protective effects of the tested compounds on
2-deoxy-D-ribose were assessed as their ability to remove
hydroxyl radicals (formed in the Fenton reaction) from the
123

76
Struct Chem (2010) 21:67–78
Fig. 3 PLATON drawings
showing the crystal packing of:
a compound 1; Intermolecular
hydrogen bond O2C–
H2CꢀꢀꢀO1C in compound 1 is
shown as dashed line,
b compound 3, and
c compound 5
among the active centers in the molecules (compound 1 has
10 times higher IC50 value than compounds 3 and 5
(Table 4)).
colored radical DPPH• into yellow-colored DPPH-H form.
The strongest scavenging activity of DPPH radicals was
expressed by mono derivative 6 (Table 4). Among bis-
derivatives, the highest DPPH–RSC exhibited compound 2,
somewhat lower derivative 4, whereas compound 5
exhibited no activity at all. Of tris-derivatives, only com-
pound 1 exhibited scavenging activity, while derivative 3
appeared to be inactive. Commercial synthetic antioxidants
Furthermore, in the DPPH assay, the ability of tested
compounds to act as donors of hydrogen atoms or electrons
in transforming DPPH• into its reduced form, DPPH-H,
was measured by spectrophotometric method [22]. Most of
tested compounds were able to reduce the stable, purple-
123

Struct Chem (2010) 21:67–78
77
tested compounds for 48 h [24]. The results are presented
in Table 5.
Table 4 Radical scavenging activities of the tested salicylic acid
derivatives and commercial antioxidants (BHA, BHT)
Compound 4 was active against four out of six cell lines,
but its activity significantly depended on the cell line. HeLa
S3 and PC3 cells were the most sensitive cell line-pro-
nounced cytotoxicity was achieved by all tested com-
pounds. Four (1–3 and 6) out of six compounds were active
at nanomolar concentrations against HeLa S3 cells. PC3
cells were highly sensitive to compounds 1, 3, and 5.
Estrogen receptor negative MDA-MB-231 cells were more
sensitive than MCF7 cells.
Compound
IC50 (mmol/L)
HO•
DPPH
LP
1
0.376
0.705
0.032
0.198
0.028
0.338
1.940
2.130
7.1
6.5
–
0.52
1.76
0.51
0.38
0.22
0.82
0.210
0.048
2
3
4
8.0
–
5
6
6.2
0.040
0.012
Both of tris-derivatives (1 and 3) and only one bis-
derivative (2) exhibited potent cytotoxicity against the
HeLa S3 cells. The other two derivatives (4 and 5)
exhibited a lower, but still potent, activity against the same
cell lines. Both of tris-derivatives (1 and 3) and only one
bis-derivative (5) exhibited strong cytotoxycity against the
PC3 cells. Among four compounds that exhibited activity
against the MDA-MB-231 cells, the most potent was tris-
derivative 3, but only compound 4 showed activity against
the MCF7 cells. Only compound 2 showed a moderate
cytotoxicity against the normal fetal lung fibroblasts
MRC-5.
BHT
BHA
BHT and BHA exhibited higher DPPH scavenging prop-
erties than those observed for the examined compounds.
The inhibition of LP was determined by measuring the
formation of malondialdehyde, using liposomes as an
oxidizable substrate [22]. All examined compounds
showed notable inhibition of the Fe2?/ascorbate induced
LP in liposomes, compound 5 expressing the highest
inhibition. The bis- and tris-derivatives showed similar
order of decreasing activities as in the case of the neu-
tralization of OH radicals 5 [ 4 [ 3 * 1 [ 6 [ 2. Com-
parison of the IC50 values for tested compounds with the
IC50 values for BHT and BHA shows that compounds 4
and 5 could be classified as promising inhibitors of LP.
The bis derivative 5, which showed strong activity as a
scavenger of OH radicals and inhibitor of LP, exhibited
also a potent cytotoxic activity against the PC3 cells. The
tris-derivative 3, which exhibited a similar antioxidative
activity, apart from showing potent activity against the PC3
cells, showed also a strong cytotoxic activity against the
HeLa S3 line.
Cytotoxicity
If compared with doxorubicin (DOX), BHT and BHA,
compounds 1–6 showed a stronger cytotoxicity against PC3
cell lines, e.g., the activity of compound 5 was 340 times
higher than that of DOX. Compounds 1–3 and 6 exhibited
stronger cytotoxicity against HeLa S3 compared to DOX;
compound 1 was 30 times more active than DOX. Com-
pound 4 exhibited cytotoxicity against MCF7 cells, in con-
trast to BHT and BHA, which were not toxic to these cells.
Compounds 1, 3, 4, and 5 were toxic to MDA-MB-231 cells,
The newly synthesized compounds 1–6 were evaluated for
their in vitro cytotoxicity against MCF7, human breast
adenocarcinoma ER?, MDA-MB-231, human breast ade-
nocarcinoma ER-, PC3, prostate cancer, HeLa S3, cervix
epithelioid carcinoma, Hs 294T, human melanoma, K562,
chronic myelogenous leukemia, as well as MRC-5, normal
fetal lung fibroblasts. Cytotoxic activity was determined by
using the standard SRB assay, after exposure of cells to the
Table 5 In vitro cytotoxicity of
the tested compounds
Compound
IC50 (lmol/L)
MCF-7
MDA-MB-231
PC-3
0.59
HeLa S3
Hs 294T
K562
MRC-5
1
[100
[100
[100
11.68
[100
[100
0.75
41.81
[100
12.08
61.07
26.84
[100
0.12
0.04
0.05
0.06
4.67
9.87
0.21
1.17
–
[100
[100
[100
[100
54.90
[100
15.39
–
[100
45.16
[100
[100
134.04
[100
0.36
–
[100
32.77
[100
[100
[100
[100
0.12
2
24.58
0.58
3
4
23.33
0.28
5
6
37.82
95.61
138.21
212.89
DOX
BHT
BHA
[100
[100
[100
[100
[100
[100
–
–
–
123

78
Struct Chem (2010) 21:67–78
References
while BHT and BHA were not. Tested compounds did not
exhibit cytotoxicity toward Hs 294T and K562. It should be
pointed out that all compounds but 2were not toxic to healthy
MRC-5 cells; compound 2 was significantly less toxic
compared to DOX.
¨
1. Droge W (2002) Physiol Rev 82:47–95
2. Soler-Rivas C, Espin JC, Wichers H (2000) Phytochem Anal
11:330–338
3. Valko M, Izakovic M, Mazur M, Rhodes CJ, Telser J (2004) Mol
Cell Biochem 266:37–56
4. Valko M, Rhodes CJ, Moncol J, Izakovic M, Mazur M (2006)
Chem Biol Interact 160:1–14
5. Valko M, Leibfritz D, Moncol J, Cronin MTD, Mazur M, Telser J
(2007) Int J Biochem Cell Biol 39:44–84
Conclusions
6. Al-Dabbas MM, Suganuma T, Kitahara K, Hou D, Fuujii M
(2006) J Ethnopharmacol 108:287–293
7. Ljubincic P, Azaizeh H, Portnaya I, Cogan U, Said O, Abu Saleh
K, Bomzon AJ (2005) J Ethnopharmacol 99:43–47
8. Fujisawa S, Atsumi T, Kadoma Y, Sakagami H (2002) Toxi-
cology 177:39–54
9. Sestili P, Diamantini G, Bedini A, Cerioni L, Tommasini I, Tarzia
G, Cantoni O (2002) Biochem J 364:121–128
10. Simpson J, Forrester R, Tisdale MJ, Billington DC, Rathbone DL
(2003) Bioorg Med Chem Lett 13:2435–2439
Comparison of crystal and MMC structures for newly
synthesized compounds 1, 3, and 5 revealed that the
intramolecular hydrogen bonds play an important role,
stabilizing conformation of the most part of molecule. The
influence of crystal packing on molecular conformation,
including the strong O2C–H2CꢀꢀꢀO1C intermolecular
hydrogen bond in compound 1, is not predominant.
The study showed that the electronic effects of the
groups at the N atom are responsible for neutralization of
OH radical, i.e., antioxidant activity. This electronic effect
in molecule of compound 1 between O1c atom from the
C=O group and N atom resulted in conformational changes
in the region of atom N (comparing with the conformations
of the same region in the molecules 3 and 5). It seems that
it is the reason for reduced antioxidant activity of com-
pound 1. The results of studying the biological effects show
that the investigated compounds 1–6, possessing salicyloyl
group, exhibited strong activities as scavengers of OH
radicals and inhibitors of LP, as well as a strong cytotoxic
activity against the HeLa S3 and PC-3 cells.
11. Zheng LF, Wei QY, Cai YJ, Fang JG, Zhou B, Yang L, Liu ZL
(2006) Free Radic Biol Med 41:1807–1816
12. Rotem R, Tzivony Y, Flescher E (2000) Prostate 42:172–180
13. Logrado LPL, Silveira D, Romeiro LAS, de Moraes MO, Cav-
´
alcanti BC, Costa-Lotufo LV, do O Pessoa C, dos Santos MLJ
(2005) Braz Chem Soc 16:1217–1225
14. Ott I, Schmidt K, Kircher B, Schumacher P, Wiglenda T, Gust RJ
(2005) Med Chem 48:622–629
15. Oxford Diffraction CrysAlis CCD and CrysAlis RED. Versions
1.171. (2008) Oxford Diffraction Ltd, Abingdon, Oxfordshire,
England
16. Harms K (1996) XCAD4, data reduction program for CAD4
diffractometers. Philipps-Universitat Marburg, Germany
17. Altomare A, Cascarano G, Giacovazzo C, Guagliardi A (1993)
J Appl Cryst 26:343–350
18. Sheldrick GM (1997) SHELX97, programs for crystal structure
¨
analysis. University of Gottingrn, Germany
19. Nardelli MJ (1995) Appl Cryst 28:659–662
20. Spek AL (1998) PLATON. A multipurpose crystallographic tool.
University of Utrecht, The Netherlands
Supplementary data
21. Farrugia LJ (1999) J Appl Cryst 32:876–881
´
CCDC 637570-637572 contains the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge at www.ccdc.cam.ac.uk/conts/retrieving.
html [or from the Cambridge Crystallographic Data Centre
(CCDC), 12 Union Road, Cambridge CB2 1EZ, UK; fax:
?44(0)1223-336033; email: deposit@ccdc.cam.ac.uk].
ˇ
´
22. Mimica-Dukic N, Bozin B, Sokolovic M, Simin N (2004) J Agric
Food Chem 52:2485–2489
23. Halliwell B (1990) Free Radical Res Commun 9:1–32
24. Skehan P, Storeng R, Scudiero D, Monks A, McMahon J, Vistica
D, Warren TJ, Bokesch H, Kenney S, Boyd RM (1990) J Natl
Cancer Inst 82:1107–1112
25. Farrugia LJ (1997) J Appl Cryst 30:565
26. Serena Software (1989) PCMODEL, version 4.0. Serena Soft-
ware Bloomington, Indiana, USA
Acknowledgments The authors would like to thank the Ministry of
Science and Technological Development of the Republic of Serbia for
financial support (Grant No. 142052).
123