Syntheses, characterization and antitumor activities of transition metal complexes with isoflavone

Article abstract of DOI:10.1016/j.jinorgbio.2009.11.008

Five new complexes were synthesized by the reaction of 4′-methoxy-5,7-dihydroxy-isoflavone ligand (a) with transition metal ions zinc (Zn²⁺) (complex b), manganese (Mn²⁺) (complex c), copper (Cu²⁺) (complex d), cobalt (Co²⁺) (complex e) and nickel (Ni²⁺) (complex f). The composition of the complexes has been characterized by elemental analysis, IR, mass spectrometry (MS) and ¹H NMR spectrometric techniques. Their antitumor properties were evaluated against five human cancer cell lines using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay and flow cytometry. The results indicated that the complexes possess higher growth inhibitory effects than free isoflavone and corresponding metal ions. Complex c and f showed greater antitumor activity and selectivity than other described complexes, even more effective than the positive control cisplatin against the selected cell lines. In addition, DNA flow cytometric analysis demonstrated that complexes display a significant G₂/M phase arrest, which then progressed to early apoptosis as detected by flow cytometry after double-staining with annexin V and propidium iodide (PI).

Full text of DOI:10.1016/j.jinorgbio.2009.11.008

Journal of Inorganic Biochemistry 104 (2010) 379–384
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
Syntheses, characterization and antitumor activities of transition metal
complexes with isoflavone
*
Xiang Chen, Li-Jun Tang, Yu-Na Sun, Pei-Hong Qiu , Guang Liang
School of Pharmacy, Wenzhou Medical College, Wenzhou 325 035, China
a r t i c l e i n f o
a b s t r a c t
0
Article history:
Received 13 July 2009
Received in revised form 17 November 2009
Accepted 18 November 2009
Available online 26 November 2009
Five new complexes were synthesized by the reaction of 4 -methoxy-5,7-dihydroxy-isoflavone ligand (a)
2+
2+
2+
with transition metal ions zinc (Zn ) (complex b), manganese (Mn ) (complex c), copper (Cu ) (complex
d), cobalt (Co ) (complex e) and nickel (Ni ) (complex f). The composition of the complexes has been
characterized by elemental analysis, IR, mass spectrometry (MS) and H NMR spectrometric techniques.
Their antitumor properties were evaluated against five human cancer cell lines using the 3-(4,5-dimeth-
ylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay and flow cytometry. The results indicated
that the complexes possess higher growth inhibitory effects than free isoflavone and corresponding metal
ions. Complex c and f showed greater antitumor activity and selectivity than other described complexes,
even more effective than the positive control cisplatin against the selected cell lines. In addition, DNA flow
2+
2+
1
Keywords:
Isoflavone
Metal complexes
Antitumor activities
Flow cytometry
2
cytometric analysis demonstrated that complexes display a significant G /M phase arrest, which then
progressed to early apoptosis as detected by flow cytometry after double-staining with annexin V and
propidium iodide (PI).
Ó 2009 Elsevier Inc. All rights reserved.
0
1
. Introduction
With the discovery of cisplatin in 1965 by Rosenberg et al. [1],
Mn2+, Cu2+, Co2+ and Ni2+ complexes (b, c, d, e, f) with the 4 -meth-
oxy-5,7-dihydroxy-isoflavone (a). All compounds were tested
against human lung cancer A549 cells, human adenocarcinoma
Hela cells, human hepatoma carcinoma HepG2 cells, human colon
carcinoma SW620 cells and human breast carcinoma MDA-MB-
435 cells with the aim of assessing the activity and selectivity in
the antitumor action and evaluated whether the complexes pos-
sess higher anti-proliferative activities than the parent isoflavone
to develop promising anti-cancer candidates.
numerous platinum complexes were synthesized and their appli-
cations as antineoplastic agents were established, whereas the
impressive clinical effectiveness of cisplatin is limited by signifi-
cant side effects including nephrotoxicity, emetogenesis, neurotox-
icity and the emergence of drug resistance [2–4]. Since then
numerous non-platinum metal complexes were studied, including
ruthenium, iron, and cobalt complexes with different ligands [5–8].
Flavonoids are important natural anti-oxidant and have been
extensively studied because of their numerous biological activities
2
. Experimental
[
9–13]. Most of flavonoids are good metal chelators which can che-
2.1. Materials and methods
late many metal ions to form different complexes due to their high
superdelocalizability and conjugated system. It has been exten-
sively reported that the interaction of metals with flavonoids con-
tributes to the anti-oxidant activity [14–23].
Zinc acetate, manganese acetate, cupric chloride, nickel chloride
and cobalt chloride were purchased from standard chemical sup-
pliers and used without further purification. All reagents were of
Isoflavones belong to one of subclasses of flavonoids and have
been shown extensive biological activities with low toxicity, such
as antiviral, antiinflammatory, anti-oxidant and antitumoral ac-
tions [24–26]. Several researchers has reported the interaction of
metals with some isoflavones [27,28], while the composition char-
acterized and the biological activities of these complexes have
0
analytical grade. The ligand 4 -methoxy-5,7-dihydroxy-isoflavone,
was prepared as previously reported [29].
Melting points were recorded on a XP4 Electrothermal melting
point apparatus and are uncorrected. Infrared spectra were mea-
sured on a 670 FT-IR spectrophotometer using the KBr pellet tech-
ꢀ1
nique (4000–400 cm ). The electrospray mass spectrum, in
2
+
been not described. The aim of this study was to obtain Zn
,
ꢀ
negative mode (ESI ), of the complexes dissolved in DMSO:metha-
1
nol (1:1) were obtained from an Esquire HCT instrument. H NMR
spectra were recorded on a Bruker Avance-600 using dimethyl
*
Corresponding author. Tel./fax: +86 577 86699218.
E-mail address: wzqph@163.com (P.-H. Qiu).
6 6
sulfoxide-d (DMSO-d ) as solvent. Elemental analyses were
0
162-0134/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.jinorgbio.2009.11.008

3
80
X. Chen et al. / Journal of Inorganic Biochemistry 104 (2010) 379–384
performed on a PE-2400-II Elemental Analyzer. Molar conductivity
) were measured on DDS-11A model digital conductivity meter
based on the measurements designed by Geary [30]. A long-term
UV–visible (UV–VIS) study was carried to verify the stability of
new complexes in solution.
2.3.2. Mn(a)
Yield 0.13 g, 42.1%. m.p. >300 °C. Anal. calcd. for c: C32
(%): C, 61.84; H, 3.54; Mn, 8.86. Found (%): C, 62.16; H, 3.37; Mn,
2
, c
(K
22
H O10Mn
ꢀ1
9.28. IR (KBr)
t
(cm ): 3383 (
t
O–H), 1621 (
tC@O), 1609–1417
1
(tC@C), 1248
(
tC–O–H), 1179
(
t
C–O–C), 532
(t
Mn–O). HNMR
, ppm): 3.71 (6H, s, OCH ), 5.32 (2H, d,
HH = 4.2 Hz, CH-8, CH-
d
(
600 MHz, DMSO-d
6
3
3
3
0
J
HH = 4.2 Hz, CH-6, CH-L6), 5.88 (2H, d,
J
2
.2. Synthesis of ligand 4 -methoxy-5,7-dihydroxy-isoflavone, a
3
0
0
0
0
L8), 6.49 (4H, d,
J
HH = 8.3 Hz, CH-3 , CH-5 , CH-L3 , CH-L5 ), 7.48
3
0
0
0
0
0
(4H, d,
JHH = 8.3 Hz, CH-2 , CH-6 , CH-L2 , CH-L6 ), 8.34 (2H, s,
4
-Methoxy-5,7-dihydroxy-isoflavone (a) was prepared by
CH-2, CH-L2), 11.81 (2H, s, OH-7, OH-L7). EIMS calcd. for c (m/z):
using the reaction previously described [29,31,32]. Phloroglucinol
ꢀ
2
ꢀ1
6
21. Found (m/z): 619.8 [M ]. K = 4.5 S cm mol .
(
(
12.6 g, 0.1 mol) and p-methoxy benzacetonitril (15 mL) in ether
50 mL) were cooled in an ice bath and saturated with a stream
2
.3.3. Cu(a)
Yield 0.25 g, 80.3%. m.p. >300 °C. Anal. calcd. for d: C32
%): C, 60.95; H, 3.49; Cu, 10.16. Found (%): C, 62.46; H, 3.38; Cu,
.78. Although the elemental analysis value for ‘‘C” is not satisfac-
tory, the chemical formula for this compound, Cu(a) , d is well sup-
2
, d
of HCl gas (HCl was produced by reacting NaCl and concentrated
SO ). The reaction mixture was refrigerated for 12 h, saturated
22
H O10Cu
H
2
4
(
9
again with HCl gas, and refrigerated for another 12 h. The precipi-
tate obtained was washed with ether twice and refluxed with 1%
aqueous H SO (250 mL) for 3 h. A yellow precipitate was formed,
2 4
filtered off to give the intermediate deoxybenzoin (12.7 g, 46.3%):
m.p. 194–196 °C [32].
2
1
ported with the other analysis data including IR, H NMR and
ꢀ
1
especially EIMS as shown below. IR (KBr)
626 ( C@O), 1611–1415 ( C@C), 1256 ( C–O–H), 1179 (
Cu–O). H NMR d (600 MHz, DMSO-d , ppm): 3.71 (6H, s, OCH
t
(cm ): 3375 (
t
O–H),
1
t
t
t
tC–O–C), 582
A mixture of deoxybenzoin (6.8 g, 0.025 mol) and boron tri-
1
(t
6
3
),
fluoride ether (BF
dimethylformamide (DMF) (60 mL) was added dropwise. In an-
3
ꢁEt
2
O) (20 mL) was cooled to 10 °C and N,N-
3
3
5
.81 (2H, d, JHH = 4.2 Hz, CH-6, CH-L6), 6.38 (2H, d, JHH = 4.2 Hz,
3
0
0
0
CH-8, CH-L8), 6.98 (4H, d,
JHH = 8.3 Hz, CH-3 , CH-5 , CH-L3 , CH-
JHH = 8.3 Hz, CH-2 , CH-6 , CH-L2 , CH-L6 ), 8.45
other flask, DMF (60 mL) was cooled to 10 °C and methylsulfonyl
chloride (MeSO Cl) (30 mL) was added in small portions. The mix-
2
0
3
0
0
0
0
L5 ), 7.48 (4H, d,
(
2H, s, CH-2, CH-L2), 12.89 (2H, s, OH-7, OH-L7). EIMS calcd. for
ture was then allowed to stand at 55 °C for 20 min and added to
the above reaction mixture slowly. During the addition, the tem-
perature of the reaction mixture was maintained below 27 °C.
The mixture was then stirred at room temperature for 4 h. The
workup was carried out by pouring the reaction mixture into
methanolic HCl (0.1 M) followed by heating at 70 °C for 20 min
and extracting the product by acetoacetate after removing the
methanol and most of DMF. The organic layer was washed with
ꢀ
2
ꢀ1
d (m/z): 630. Found (m/z): 628.1 [M ].
K = 3.2 S cm mol .
2
.3.4. Co(a)
Yield 0.21 g, 68.2%. m.p. >300 °C. Anal. calcd. for e: C32
%): C, 61.44; H, 3.52; Co, 9.44. Found (%): C, 61.28; H, 3.68; Co,
2
, e
22
H O10Co
(
9
ꢀ1
.12. IR (KBr)
C@C), 1246 (
t
t
(cm ): 3385 (
C–O–H), 1180 (
, ppm): 3.71 (6H, s, OCH
t
O–H), 1629 (
tC@O), 1610–1437
1
(t
t
C–O–C), 783 ( Co–O). H NMR d
t
(
600 MHz, DMSO-d
J
6
3
), 5.81 (2H, d,
4
water and dried with MgSO . The solvent was removed under vac-
3
3
0
HH = 4.2 Hz, CH-6, CH-L6), 6.08 (4H, d,
JHH = 8.3 Hz, CH-3 , CH-
uum using a rotary evaporator and product was purified by
recrystallization.
0
0
0
3
5
, CH-L3 , CH-L5 ), 6.98 (2H, d, JHH = 4.2 Hz, CH-8, CH-L8), 7.48
3
0
0
0
0
(cmꢀ1): 3388
t
(4H, d,
JHH = 8.3 Hz, CH-2 , CH-6 , CH-L2 , CH-L6 ), 8.65 (2H, s,
a: Yield 6.1 g, 86.1%, m.p. 213–215 °C. IR (KBr)
O–H), 1653 ( C@O), 1609–1440 ( C@C), 1248 ( C–O–H), 1175 (t
C–O–
CH-2, CH-L2), 12.18 (2H, s, OH-7, OH-L7). EIMS calcd. for e (m/z):
(
t
t
t
t
ꢀ
2
ꢀ1
). 1H NMR d (600 MHz, DMSO-d
, ppm, s: singlet, d: doublet, t:
triplet, m: multiplet): 3.77 (3H, s, OCH
625. Found (m/z): 625.7 [M ]. K = 8.9 S cm mol .
C
6
3
3
), 6.21 (1H, d, JHH = 2.1 Hz,
3
3
2.3.5. Ni(a)
Yield 0.24 g, 76.6%. m.p. >300 °C. Anal. calcd. for f: C32
%):C, 61.44; H, 3.52; Ni, 9.44. Found (%): C, 61.96; H, 3.35; Ni, 9.18.
2
, f
CH-6), 6.38 (1H, d,
J
HH = 2.1 Hz, CH-8), 6.98 (2H, d,
J
HH = 8.7 Hz,
0
0
3
0
0
H
22
O10Ni
CH-3 , CH-5 ), 7.48 (2H, d,
JHH = 8.7 Hz, CH-2 , CH-6 ), 8.36 (1H, s,
(
CH-2), 10.90 (1H, s, OH-5), 12.92 (1H, s, OH-7). EIMS m/z: 282.7
ꢀ
1
ꢀ
IR (KBr)
t
(cm ): 3394 (
C–O–H), 1179 (
DMSO-d , ppm): 4.11 (6H, s, OCH
CH-6, CH-L6), 5.48 (2H, d,
tO–H), 1628 (tC@O), 1609–1441 (t
C@C),
[
M ].
1
245 (
t
t
C–O–C), 535 (
t
Ni–O). 1H NMR d (600 MHz,
3
6
3
), 5.31 (2H, d,
J
HH = 4.2 Hz,
2.3. Synthesis of complexes
3
J
HH = 4.2 Hz, CH-8, CH-L8), 7.00 (4H,
3
0
0
0
0
d,
JHH = 8.3 Hz, CH-3 , CH-5 , CH-L3 , CH-L5 ), 7.48 (4H, d,
HH = 8.3 Hz, CH-2 , CH-6 , CH-L2 , CH-L6 ), 8.34 (2H, s, CH-2, CH-
Ligand a (0.284 g, 0.001 mol) was dissolved in ethanol (30 mL)
adjusted to pH 7–8 by addition of triethylamine, stirring for 1 h
at 40 °C followed by hydrated metal (II) salt (acetate or chloride)
3
0
0
0
0
J
L2), 12.36 (2H, s, OH-7, OH-L7). EIMS calcd. for f (m/z): 625. Found
ꢀ
2
ꢀ1
(
m/z): 623.4 [M ].
K
= 9.1 S cm mol
.4. Biological materials and methods
All reagents were purchased from Sigma and have an analytic
.
(
0.001 mol) in ethanol (10 mL) which were introduced slowly via
a syringe. The reaction mixture were stirred at 60 °C for 12 h,
standing at room temperature for 2 days. Powered solids were
yielded by filtration, rinsed with ethanol and dried under vacuum.
2
purity.
2
.3.1. Zn(a)
Yield 0.13 g, 40.3%. m.p. >300 °C. Anal. Calcd. for b: C32
%): C, 60.86; H, 3.49; Zn, 10.30. Found (%): C, 60.56; H, 3.57; Zn,
2
, b
H
22
O
10Zn
2.4.1. Cell culture
(
For routine testing, five cancer cell lines were used in this study:
A549 (human lung cancer cell line), Hela (human cervix cancer cell
line), HepG2 (human hepatoma carcinoma cell line), SW620 (hu-
man colon carcinoma cell line) and MDA-MB-435 (human breast
carcinoma cell line), purchased from Chinese Academy of Sciences
cell bank, China. Cells were grown at 37 °C in a humidified atmo-
ꢀ
1
1
0.48. IR (KBr)
C@C), 1280 ( C–O–H), 1187 (
600 MHz, DMSO-d
t
(cm ): 3369 (
t
O–H), 1645 (
t
C@O), 1611–1411
ꢀ
1 1
(t
t
tC–O–C), 562 (
tZn–O) cm
. H NMR d
(
6
, ppm): 3.77 (6H, s, OCH
3
), 6.21 (2H, d,
3
3
J
HH = 4.2 Hz, CH-6, CH-L6), 6.37 (2H, d,
J
HH = 4.2 Hz, CH-8, CH-
3
0
0
0
0
L8), 6.98 (4H, d,
J
HH = 8.3 Hz, CH-3 , CH-5 , CH-L3 , CH-L5 ), 7.47
3
0
0
0
0
(
4H, d,
J
HH = 8.3 Hz, CH-2 , CH-6 , CH-L2 , CH-L6 ), 8.35 (2H, s,
sphere containing 5% CO
with 10% heat inactivated (56 °C for 30 min) fetal bovine serum
(FBS), 100 IU/mL of penicillin and 100 g/mL of streptomycin.
2
, in RPMI 1640 medium supplemented
CH-2, CH-L2), 12.92 (2H, s, OH-7, OH-L7). EIMS calcd. for b (m/z):
ꢀ
2
ꢀ1
6
31. Found (m/z): 631.2 [M ].
K
= 5.3 S cm mol
.
l

X. Chen et al. / Journal of Inorganic Biochemistry 104 (2010) 379–384
381
2
.4.2. 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
3
CH CN
HO
OH
HO
Dried ether
OH
O
(
MTT) assay
MTT assay was carried out based on the method described by
HCL
Mosmann [33]. Briefly, cells were seeded in 96-well plates at initial
seeding densities of 4000 cells per well with 180 L of culture
media and incubated in an atmosphere of 5% CO for 24 h to allow
attachment onto the wells. All compounds were dissolved immedi-
ately in DMSO, and diluted with the growth medium. The final con-
l
OH
OH
OCH
3
OCH
3
2
1
8
HO
7
O
2
BF
3
.Et
2
O/DMF
2
'
6
1'
'
3
3
'
centrations were reached to 600
1
lM, 300
lM, 100
lM, 50
lM,
MeSO Cl
2
5
4
0
l
M, 1 M and 0.1 M, respectively. In addition, the final DMSO
l
l
4'
OH
O
6
OCH
3
concentration in the culture medium, <1%, showed no obvious
cytostatic effect in preliminary tests. Twenty microliters of each
compound dilution was added into the appropriate wells in three
replicates. Culture medium only containing different concentra-
tions of DMSO was added to the cells in the control wells. Culture
medium with the corresponding concentrations of compounds, but
void of cells was used as blank. Following incubation at 37 °C for
5'
a
1
8
HO
7
O
2
3
M+
2'
6
1'
3
'
5
4
O
O
6'
4'
OCH3
5
'
L5'
L2'
M
4
8 h, 20
l
L of 0.1% MTT was added into each well. The plates were
H CO
3
L6'
re-incubated for 4 h and 150
l
L DMSO was introduced to dissolve
O
L4
O
L4'
L3'
the insoluble blue formazan precipitate produced by MTT reduc-
tion. The plates were shaken for 10 min and the optical density
of each well was measured at 570 nm on a microplate spectropho-
tometer (Bio-Tek Instruments, ELx808zu, USA).
L5
L1'
L3
L6
L2
L7
O
OH
L8
L1
Assays were performed in triplicate in three independent exper-
iments. The concentration required for 50% inhibition of cell viabil-
ity (IC50) was calculated using the software ‘‘Dose–Effect Analysis
with Microcomputers”. Data were analyzed using Student un-
paired t-test.
b
(M=Zn),
c
(Mn),
d
(Cu), (Co), f (Ni)
e
Scheme 1. Synthesis of 4 -methoxy-5,7-dihydroxy-isoflavone and Zn2+, Mn2+, Cu2+
0
,
2+
2+
Co and Ni complexes.
complexes were isolated as powder solids in different yields, the
purity of which was confirmed by thin-layer chromatography
2
.4.3. Cell cycle analysis
Flow cytometry was used to evaluate the number of cells in the
(
TLC) and elemental analysis. The results of molar conductivity
particular phases of the cell cycle. The selected cells (MDA-MB-
35, SW620, HepG2 and A549 cells) were seeded in six-well plates
showed that all complexes were nonelectrolyte in DMSO [30]. In
addition, all the complexes were insoluble in water, slightly solu-
ble in methanol and soluble in DMSO or mixture of water/DMSO.
4
6
containing 5 ꢂ 10 cells per well in their respective culture media
and pre-incubated for 24 h to induce cell cycle in exponential
1
The complexes were characterized by IR, H NMR, mass spectros-
growth. The cells were then treated with 10 lM of compounds a
copies, elemental analysis and melting point determination.
and c that were dissolved in DMSO and diluted with growth med-
ium. Growth medium only containing DMSO was used as control.
After 24 h incubation, the cells were trypsinised, washed twice in
PBS, centrifuged at 800g for 5 min at 4 °C, fixed with ice-cold 70%
ethanol for 30 min and treated with 50
3
stained with 50 lg/mL propidium iodide (PI) in the dark at 4 °C
for 30 min and analyzed using a FACS scan flow cytometer (Elite
ESP, Beckman Coulter, Brea, CA).
3.2. Spectroscopic data
The IR spectra of the ligand, a showed the characteristic absorp-
lg/mL of RNase A at
ꢀ
1
ꢀ1
tion for
t
(C@O) at 1653 cm
.
The broad band at 3380 cm , a
7 °C for 30 min. To determine their DNA contents, cells were
ꢀ1
ꢀ1
strong band at 1248 cm and a medium band around 1175 cm
were assigned to hydrogen bond stretching, phenolic
tions and (C–O–C), respectively. The bands arising from
t
(C–O) vibra-
(C@C)
t
t
ꢀ1
were observed at 1440–1621 cm . Compared with the IR spectra
of ligand a, the absorption bands of complexes b–f exhibited
downward shifts (8–22 cm ) for the C@O groups and showed ex-
tremely slightly shifts for other groups, indicating that coordina-
2
.4.4. Annexin V and PI binding assay
To assess the simultaneous observation of early phase of apop-
ꢀ1
totic and necrotic features, SW620 cells were treated with ligand a
and complex c at 10 M final concentration for 24 h, then cells
were measured using flow cytometry by adding annexin V-FITC
to 106 cells per sample according to the manufacturer’s specifica-
tions (Bender MedSystems, Vienna, Austria). Simultaneously, the
cells were stained with PI.
tion occurred via oxygen atom of the carbonyl oxygen group. For
l
ꢀ1
all the complexes, new bands in the region of 500–600 cm de-
tected were assigned to stretching frequencies of
t (M–O).
1
The H NMR spectrum of synthesized ligand a was consistent
with the reported data in literature [29]. Resonances of hydrogen
atoms belonging to 5,7-hydroxyl groups appeared at 12.92 and
1
0.90 ppm, respectively. For the corresponding complexes b–f,
the resonances of the 7,L7-hydroxyl group protons were found at
2.92, 11.81, 12.89, 12.18 and 12.36 ppm, respectively, and the sig-
3
. Results and discussion
1
3
.1. Synthesis
nals of the benzopyran and aromatic protons displayed extremely
slightly shifts either in comparison with ligand a. However, it is
worth to noted that the complete disappearance of signals arising
from the 5,L5-hydroxyl group protons was expressed, pointing out
that the coordination of oxygen atoms from the 5-hydroxyl group.
All compounds have been investigated by positive/negative
mass spectrometric measurements, giving valuable structural
information. The major structurally informative molecular ions
Ligand a was produced by the reaction of phloroglucinol and
p-methoxy benzacetonitril, in the presence of BF
MeSO Cl for the synthesis of corresponding Zn , Mn , Cu , Co
2
and Ni complexes (Scheme 1). The complexes were obtained by
titrating an ethanolic solution of hydrated metal(II) salt (chloride
or acetate) to an ethanol solution of ligand a (pH 7–8). The
3
ꢁEt
2
O, DMF and
2+
2+
2+
2+
2
+

3
82
X. Chen et al. / Journal of Inorganic Biochemistry 104 (2010) 379–384
ꢀ
ꢀ
Table 1
625.7 [M] and 623.4 [M] , respectively, indicating that each com-
plex containing one metal ion and two molecule ligands was
formed. The appearance of the characteristic cluster of isotopic
peaks of the corresponding metal ion isotopes demonstrated that
metal ions have been successfully coordinated to the ligand a. In
addition, the predominant peaks corresponding to ions containing
Zn, Mn, Cu, Co and Ni in the complexes are those with metal
bonded to one ligand molecule at m/z 346.8, 336.1, 344.8, 341.7
and 341.6, respectively. The intact negative ion of ligand molecule
as a dominant fragment ion appeared at m/z 283. Complexes c and
f showed ions obtained by loss of a methoxyl, which was observed
at m/z 589.4 and 590.0. It is worth noting that the spectrum of
complex b showed sodium adduct ion based on addition of sodium
ion presumably to oxygen atom from hydroxyl at m/z 593.1. In
complex d, the loss of a methyl yielded a fragment ion at m/z
Main MS data of compounds a–f (m/z).
Compd.
Molecular ions (-MS)
Fragment ions (-MS2)
a
b
c
d
e
f
282.7
631.2
619.8
628.1
625.7
623.4
–
593.1, 346.8, 309.0, 282.9
589.4, 336.1, 282.8
612.8, 344.8, 301.0, 282.9
341.7, 284.9
607.0, 590.0, 341.6, 280.8, 262.8
Table 2
Long-term UV–VIS data (kmax) of compounds a–f in DMSO/water solution during
0 days.
3
k
max (nm)
Time (days)
0
6
12.8 and another important fragment ion was that of small abun-
3
7
15
30
dance obtained by sequential loss of a ligand, a methoxyl as well as
a methenyl from the 2-position at m/z 301.0.
The results of the long-term UV–VIS study are showed in
Table 2. The band of ligand a was observed at 262 nm. Compared
with ligand a, the absorption wavelength of new complexes
a
b
c
d
e
f
262
277
286
271
296
298
263
276
286
269
296
297
262
276
284
269
296
291
262
277
286
268
297
293
262
277
286
268
295
293
(b–f) exhibited blue shift. In addition, it is significant to note that
the kmax (nm) values of new compounds were hardly changed
during one month, meaning that new complexes were stable in
and fragment ions in the negative-ion mode of compounds a–f are
solution.
presented in Table 1. As expected, the molecular ions peak of com-
The data of elemental analysis, IR, Ms and 1H NMR spectrum
ꢀ
ꢀ
ꢀ
plexes b–f was found at m/z 631.2 [M] , 619.8 [M] , 628.1 [M] ,
confirmed the bonding of ligand a to metal ions and the metals
Fig. 1. Five selected human cancer cell lines were incubated in complete cell growth medium with and without different concentrations (600
lM, 300 lM, 100 lM, 50 lM,
1
0
l
M, 1 M and 0.1 M) of new compounds (a–f) for 48 h. The IC50 values were calculated by the software ‘‘Dose–Effect Analysis with Microcomputers”. Horizontal and
l
l
ꢃ
vertical axes indicate relative cell types and IC50 (lM) values, respectively. ( p<0.05, Complexes compared with the ligand a and cisplatin.).
Table 3
Cell cycle distribution analyzed by flow cytometry after 24 h treatment with ligand a and complex c at 10
l
M final concentration and the corresponding DMSO control.
Cell cycle distribution (%)
Cell line
Compd.
Apoptosis rate (%)
G
1
S
2
G /M
MDA-MB-435
Control
Ligand a
Complex c
10.2
18.4
21.2
51.3
63.6
81.0
30.4
22.5
7.2
19.1
14.1
12.1
SW-620
HepG2
A549
Control
Ligand a
Complex c
9.1
28.7
62.2
53.7
46.2
40.9
21.4
22.8
14.2
25.5
31.3
45.3
Control
Ligand a
Complex c
4.9
25.4
68.8
57.9
52.1
42.1
28.1
21.2
19.9
14.3
27.0
38.4
Control
Ligand a
Complex c
11.2
32.4
47.7
57.5
48.2
37.4
32.3
28.4
25.0
10.4
23.6
38.5

X. Chen et al. / Journal of Inorganic Biochemistry 104 (2010) 379–384
383
are bridged by the position of 5-deprotonated phenolic oxygen and
the oxygen atoms of carbonyl groups. The elemental analysis and
mass spectrum agreed well with the formation of 1:2 (metal/li-
gand) stoichiometry.
3.3. Antitumor activities
The ligand a and corresponding metal complexes (b, c, d, e, f)
were tested for antitumor activity using the MTT assay at concen-
trations ranging from 0.1 to 600 M against five human cancer cell
lines, consisting of A549 (human lung carcinoma cell line), Hela
human uterine cervix cancer cell line), HepG2 (human liver cancer
cell line), SW620 (human colon carcinoma cell line) and MDA-MB-
l
(
4
35 (human breast carcinoma cell line).
The results, expressed as concentrations of the compounds
required to inhibit the tumor cell growth by 50% (IC50), are
showed in Fig. 1. Ligand a, metal ions and cisplatin were given
for comparative purposes. Metal ions had only low activity
(
IC50 > 50 lM) against the investigated cell lines. In general,
SW620 and A549 cells showed higher sensitivities to the anti-
proliferative effect of these complexes, while MDA-MB-435 and
Hela cell lines were found to be more resistant. In SW620 cell
line, it was shown that complexes c and f induced more than
three times stronger response than ligand a and the positive con-
trol cisplatin. Moreover, complexes b, d, and e exhibited slightly
more effective, indicating that the described complexes improved
antitumor activity. In HepG2 cells, the IC50 values of all com-
plexes displayed at least two times more effective compared with
that of the ligand a. It is interesting to note that in A549 cells,
complex f showed two times more effective than cisplatin, three
times more effective than ligand a. As shown in Fig. 1, complex e
and f exhibited the lowest IC50 values against the MDA-MB-435
and SW620 cell lines, respectively. All of these results demon-
strated that this newly synthesized complexes have elevated
anti-proliferative activities in comparision with ligand a and cor-
responding metal ions, and some were even more effective than
the positive control cisplatin.
To verify the possibility that inhibition of cell proliferation is
associated with induction of apoptosis and cell cycle phase ar-
rested after compounds treatment, the cell lines of MDA-MB-435,
SW620, HepG2 and A549 were treated with ligand a and complex
c at 10
complex c induced blocks in the G
the SW620, HepG2 and A549 cell lines (Table 3), with concomitant
reductions in the percentages of cells in the G phase of the
cell cycle. However, the majority of the cell population was
arrested at the G phase accompanying reduction in the S and
/M phase following the complex c treatment in the MDA-MB-
35 cells (Table 3). Furthermore, complex c was more effective
lM final concentration for 24 h. The results revealed that
2
/M phase of the cell cycle in
1
1
G
2
4
than the ligand a.
Fluorescein isothiocyanate (FITC)-conjugated annexin V has
been utilized to detect the externalization of phosphatidylserine
that occurs at an early stage of apoptosis. Propidium iodide (PI)
is used as a marker of necrosis due to cell membrane destruction
Fig. 2. Results of flow cytometric cellular apoptosis of untreated SW620 control
cells and cells treated with ligand a and complex c at 10
lM final concentration for
24 h. The Q4 region represents signals from cells undergoing apoptosis, and the Q2
[
34]. To elucidate whether complexes-induced cell death in-
region corresponds to necrotic cells. Scattergram of the treated cells were compared
with those of the untreated control cells.
volved apoptosis or necrosis, a biparametric cytofluorimetric
analysis using annexin V and PI double-staining was performed.
The cells were treated with ligand a and complex c at 10 lM
final concentration for 24 h. As showed in Table 3, the apoptosis
rate of complex c treated were substantially higher on SW620,
HepG2 and A549 cell lines, and slightly higher on the MDA-
MB-435 cells, relative to their respective controls (DMSO) and
the ligand a. The representative pictures of compound a and c
treated on SW620 cells and the control were presented in
to necrotic cells. As showed in Fig. 2, both ligand a and complex
c could induced cell apoptosis compared with that of control.
In addition, complex c provoked two times more apoptotic
cells in comparision with ligand a when treating on SW620
cells at 10 lM final concentration for 24 h. In contrast, there
was no obvious change of necrotic cells between these two
+
ꢀ
Fig. 2. Annexin V /PI (Q
4
) population represented cells undergo-
+
+
compounds.
ing apoptosis, and annexin V /PI (Q
2
) population corresponded

3
84
X. Chen et al. / Journal of Inorganic Biochemistry 104 (2010) 379–384
[
[
7] I. Ott, B. Kircher, P. Schumacher, K. Schmidt, R. Gust, J. Med. Chem. 48 (2005)
22–629.
8] M.A. Yakupec, Dalton Trans. (2008) 183–194.
4
. Conclusion
6
0
Ligand 4 -methoxy-5,7-dihydroxy-isoflavone (a) and five new
[9] M. Leopoldini, T. Marino, N. Russo, M. Toscano, J. Phys. Chem. A 108 (2004)
4916–4923.
[10] Y.J. Kim, Y.C. Bae, K.T. Suh, J.S. Jung, Biochem. Pharmacol. 72 (2006) 1268–
complexes (b–f) were synthesized. New compounds were charac-
terized by IR, H NMR, Mass spectroscopy and elemental analysis.
1
1276.
All of the investigated compounds were tested for in vitro anti-
tumor activities against five human cancer cell lines consisting of
A549, Hela, HepG2, SW620 and MDA-MB-435. Complex c and f
possessed greater activity and selectivity than the other described
complexes in the antitumor action against the selected cancer cell
lines. The possible mechanisms by which complex c can exert an
anti-profliferative effect on MDA-MB-435, SW620, HepG2 and
A549 cells were also investigated. Flow cytometry analysis showed
that complex c acted on non-specific phase of the cell cycle arrest
and induced cell apoptosis in MDA-MB-435, SW620, HepG2 and
[
[
11] A.I. Morales, C. Vicente-Sánchez, J.M. Santiago Sandoval, J. Egido, P. Mayoral,
M.A. Arévalo, M. Fernández-Tagarro, J.M. López-Novoa, F. Pérez-Barriocanal,
Food Chem. Toxicol. 44 (2006) 2092–2100.
12] S. Luangaram, U. Kukongviriyapan, P. Pakdeechote, V. Kukongviriyapan, P.
Pannangpetch, Food Chem. Toxicol. 45 (2007) 448–456.
[13] A. Torreggiani, M. Tamba, A. Trinchero, S. Bonora, J. Mol. Struct. 759 (2005)
44–751.
14] S.B. Bukhari, S. Memon, M.M. Tahir, M.I. Bhanger, J. Mol. Struct. 892 (2008) 39–
6.
[15] S.B. Bukhari, S. Memon, M.M. Tahir, M.I. Bhanger, Spectrochim. Acta Part A 71
2009) 1901–1906.
16] W.J. Chen, S.F. Sun, W. Cao, Y. Liang, J.R. Song, J. Mol. Struct. 918 (2009) 194–
97.
[17] F.B.A. El Amrani, L. Perello, JA. Real, J. Inorg. Biochem. 100 (2006) 1208–1218.
7
[
4
(
[
1
A549 cells treated at the concentration of 10 lM. The findings indi-
[
[
[
18] P. Ryan, M.J. Hynes, J. Inorg. Biochem. 102 (2008) 127–136.
19] Y.N. Ni, S. Du, S. Kokot, Anal. Chim. Acta 584 (2007) 19–27.
20] J. Tan, B.C. Wang, L.C. Zhu, Colloids Surf. B Biointerfaces 55 (2007) 149–152.
cated that the anti-proliferative effect of complex c is associated
with cellular apoptosis and cell cycle arrest and profits to the
development of a new anti-cancer drug.
[21] J.P. Cornard, L. Dangleterre, C. Lapouge, J. Phys. Chem. A 109 (2005) 10044–
0051.
22] F.V. De Souza Rubens, F. De Giovani Wagner, Redox Rep. 9 (2004) 97–104.
[23] M. Leopoldini, N. Russo, S. Chiodo, M. Toscano, J. Agric. Food Chem. 54 (2006)
343–6351.
24] Y. Sanchez, D. Amran, E. de Blas, P. Aller, Biochem. Pharmacol. 77 (2009) 384–
96.
1
[
Acknowledgement
6
[
This work was supported by Zhejiang Province Extremely Key
Subject Building Funding ‘‘Pharmacology and Biochemical Pharma-
ceutics 2008 ”.
3
[25] M. Privat, C. Aubel, S. Arnould, Y. Communal, M. Ferrara, Y. Bignon, Biochem.
Biophy. Res. Commun. 379 (2009) 785–789.
[
26] A. Yokosuka, M. Haraguchi, T. Usui, S. Kazami, V.C. Blank, C. Poli, M. Marder,
L.P. Roguin, Bioorg. Med. Chem. Lett. 17 (2007) 3091–3094.
[
[
[
[
[
27] G.M. Saxena, T.R. Seshadri, Proc. Math. Sci. 46 (1957) 218–223.
28] R.R. Priya, K.G. Kumar, M.C. Narayanan, Asian J. Sci. Res. 1 (2008) 176–179.
29] S. Balasubramanian, M.G. Nair, Syn. Commun. 30 (2000) 469–484.
30] W.J. Geary, Coordin. Chem. Rev. 7 (1971) 81–122.
References
[
[
[
1] B. Rosenberg, L. Van Camp, T. Krigas, Nature 205 (1965) 698–699.
2] A. Alama, B. Tasso, F. Novelli, Drug Discov. Today 14 (2009) 500–508.
3] M.A. Jakupec, M. Galanski, B.K. Keppler, Rev. Physiol. Biochem. Pharmacol. 146
31] Y.C. Chang, M.G. Nair, R.C. Santell, W.G. Helferich, J. Agric. Food Chem. 42
(
1994) 1869–1871.
(
2003) 1–53.
[
[
[
32] K. Wahala, T.A. Hase, J. Chem. Soc. Perkin Trans. 1 (1991) 3005–3008.
33] T. Mosmann, Immunol. J. Meth. 65 (1983) 55–63.
34] S. Pervaiz, M.A. Seyed, J.L. Hirpara, M.V. Clement, K.W. Loh, Blood 93 (1999)
[
[
[
4] D. Wang, S.J. Lippard, Nature Rev. Drug Discov. 4 (2005) 307–320.
5] I. Kostova, Curr. Med. Chem. 13 (2006) 1085–1107.
6] P. Köpf-Maier, H. Kopf, E.W. Neuse, J. Cancer Res. Clin. Oncol. 108 (1984) 336–
4096–4108.
3
40.