Dinuclear zinc bis(thiosemicarbazone) complexes: Synthesis, in vitro anticancer activity, cellular uptake and DNA interaction study

Article abstract of DOI:10.1016/j.ica.2013.09.014

Four dinucleating bis(thiosemicarbazone) ligands and their zinc complexes have been synthesized and characterized by multinuclear NMR (¹H and ¹³C), IR, UV-Vis, ESI-MS and fluorescence spectroscopic techniques. Their purity was assessed by elemental analysis. Cytotoxicity was tested against five human cancer cell lines using the sulphorhodamine B (SRB) assay, where one of the complexes, 1,3-bis{biacetyl-2′-(4″-N- pyrrolidinylthiosemicarbazone)-3′-(4″-N- pyrrolidinylthiosemicarbazone)zinc(II)}propane (6), was found to be quite cytotoxic against MCF-7 (breast cancer) and HepG2 (hepatoma cancer) cell lines, with a potency similar to that of the well known anticancer drug adriamycin. It is evident from the cellular uptake studies that the uptake is same for the active complex 6 and the inactive complex 8 (1,6-bis{biacetyl-2′- (4″-N-pyrrolidinylthiosemicarbazone)-3′-(4″-N- pyrrolidinylthiosemicarbazone)zinc(II)}hexane) in MCF-7 and HepG2 cell lines. In vitro DNA binding and cleavage studies revealed that all complexes bind with DNA through electrostatic interaction, and cause no significant cleavage of DNA.

Full text of DOI:10.1016/j.ica.2013.09.014

Inorganica Chimica Acta 408 (2013) 152–161
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Dinuclear zinc bis(thiosemicarbazone) complexes: Synthesis, in vitro
anticancer activity, cellular uptake and DNA interaction study
⇑
Duraippandi Palanimuthu, Ashoka G. Samuelson
Department of Inorganic and Physical Chemistry, Indian Institute of Science, Bangalore 560012, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 18 July 2013
Received in revised form 2 September 2013
Accepted 4 September 2013
Available online 12 September 2013
Four dinucleating bis(thiosemicarbazone) ligands and their zinc complexes have been synthesized and
characterized by multinuclear NMR (1H and 13C), IR, UV–Vis, ESI-MS and fluorescence spectroscopic tech-
niques. Their purity was assessed by elemental analysis. Cytotoxicity was tested against five human can-
cer cell lines using the sulphorhodamine B (SRB) assay, where one of the complexes, 1,3-bis{biacetyl-20-
(400-N-pyrrolidinylthiosemicarbazone)-30-(400-N-pyrrolidinylthiosemicarbazone)zinc(II)}propane (6), was
found to be quite cytotoxic against MCF-7 (breast cancer) and HepG2 (hepatoma cancer) cell lines, with
a potency similar to that of the well known anticancer drug adriamycin. It is evident from the cellular
uptake studies that the uptake is same for the active complex 6 and the inactive complex 8 (1,6-bis{biace-
tyl-20-(400-N-pyrrolidinylthiosemicarbazone)-30-(400-N-pyrrolidinylthiosemicarbazone)zinc(II)}hexane) in
MCF-7 and HepG2 cell lines. In vitro DNA binding and cleavage studies revealed that all complexes bind
with DNA through electrostatic interaction, and cause no significant cleavage of DNA.
Keywords:
Bis(thiosemicarbazone)
Zinc
Dinuclear complex
Anticancer activity
DNA binding
Cellular uptake
Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction
cytotoxicity, DNA binding and DNA cleavage [13–17]. A typical
example is BBR3464, a trinuclear platinum complex (Fig. 1), which
The use of transition metal complexes in cancer therapy has
gained importance from the time it was discovered that cis-
diamminedichloroplatinum(II) (cisplatin) was anticancer active
[1,2]. A variety of mononuclear metal complexes with different li-
gands have been shown to be effective for cancer therapy both
in vitro and in vivo [3–7]. Some of the metal complexes including
cisplatin, carboplatin and oxaliplatin have been approved by FDA
and are currently being used worldwide to treat various cancers
[8,9]. Nine transition metal complexes with platinum, gold and
ruthenium metal centers are under various stages of clinical trials
and in the last two years two platinum complexes and a palladium
complex have entered clinical trials [10–12].
rendered 10–100 times higher cytotoxicity than the monomeric
analogue cisplatin (Fig. 1) due to its unique DNA binding property
[15]. Recently, Wang and coworkers demonstrated that when two
nontoxic mononuclear ruthenium complexes (B in Fig. 1) bearing
bipyridyl ligands are linked together by a thiophene ligand (A in
Fig. 1), the cytotoxicity is dramatically enhanced by many folds
owing to their strong DNA binding [18]. More recently, Spingler
and coworkers have shown that the dinuclear nickel and copper
complexes containing 1,3-bis(1,5,9-triazacyclododecyl)propane li-
gand (C in Fig. 1) are able to change the conformation of right-
handed B-DNA into left-handed Z-DNA, whereas the corresponding
mononuclear complexes (D in Fig. 1) fail to bring such a conforma-
tional change [19]. Therefore developing dinuclear or multinuclear
complexes and their comparison with mononuclear analogs is
quite interesting due to their enhanced anticancer activity and
interactions with biomolecules.
Mounting evidence in the literature shows that dinuclear and
multinuclear complexes are more efficacious than the correspond-
ing mononuclear complexes in terms of their cellular uptake,
Transition metal bis(thiosemicarbazones), a class of metal com-
plexes, have been of great interest due to the wide range of biolog-
ical activities they exhibit, this includes anti-alzheimer’s, anti-
malarial, anti-bacterial, anti-viral and anti-cancer activity [20–
24]. In recent years, zinc and copper bis(thiosemicarbazone) com-
plexes have attracted attention due to their intracellular fluores-
cence and anticancer activity, respectively [25,26]. A few reports
have shown that zinc complexes of mono- and bis-thiosemicarba-
zone ligands exhibit both anticancer activity and fluorescence
Abbreviations: GTSCH2, glyoxal-bis(4-methyl-4-phenyl-3-thiosemicarbazone);
ATSMH2, biacetyl-bis(4-methyl-3-thiosemicarbazone); [Zn(GTSC)]3, zinc glyoxal-
bis(4-methyl-4-phenyl-3-thiosemicarbazone); HRMS, high-resolution mass spec-
tra; EB, ethidium bromide; CT, calf-thymus; NC, nicked-circular; SC, supercoiled;
EDTA, ethylenediaminetetraacetic acid; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide; SRB, sulphorhodamine B; bpy, 2,20-bipyridine; phen,
1,10-phenanthroline; PBS, phosphate buffered saline; Tris–HCl, tris(hydroxy-
methyl)aminomethane–hydrochloride; HEPES, 4-(2-hydroxyethyl)-1-piperazinee-
thanesulfonic acid.
⇑
Corresponding author. Tel.: +91 80 2293 2663; fax: +91 80 2360 1552.
E-mail address: ashoka@ipc.iisc.ernet.in (A.G. Samuelson).
0020-1693/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ica.2013.09.014

D. Palanimuthu, A.G. Samuelson / Inorganica Chimica Acta 408 (2013) 152–161
153
Fig. 1. Structures of metal complexes and the ligand, GTSCH2, referred to in the introduction.
imaging property [27,28]. For example, Pascu and coworkers doc-
umented that acenaphthenequinone based zinc bis(thiosemicarba-
zone) complexes (E in Fig. 1) exhibited comparable cytotoxicity to
cisplatin in the MCF-7 cell line and emitted fluorescence as well
[28]. Recently, we reported synthesis and biological characteriza-
tion of a range of bis(thiosemicarbazone) ligands and their mono-
nuclear copper and zinc complexes. We showed that one of the
ligands, GTSCH2 [glyoxal-bis(4-methyl-4-phenyl-3-thiosemicarba-
zone)] (Fig. 1), is suitable for imaging cellular zinc and the corre-
sponding zinc complex is suitable for imaging several cancer cell
lines of different tissue origin [29,30]. And interestingly, many of
the copper complexes are found to be anticancer active [31]. Sev-
eral reports including ours have proven that small modifications
in the structure of metal bis(thiosemicarbazone) complexes lead
to significant changes in their anticancer activity and their fluores-
cence property [25,30–32].
potential is also reported. Cytotoxicity has been evaluated in five
cancer cell lines of different tissue origin. The cellular uptake was
investigated by estimating the amount of accumulation of zinc in
two cell lines using ICP-OES. The ability of zinc complexes to fluo-
resce within HepG2 cells has been examined by flow cytometry. Ef-
fects of these complexes on DNA binding and DNA cleavage have
been investigated by the ethidium bromide displacement assay,
viscosity analysis, circular dichroism experiments and agarose gel
electrophoresis. Although most properties of dinuclear complexes
are similar to mononuclear complexes, the cytotoxicity of the pro-
pylenediamine bridged dinuclear zinc complex is as high as adria-
mycin, a potent anticancer drug.
2. Experimental
Synthesis and characterization of a few dinuclear zinc bis(thio-
semicarbazone) complexes, where the bis(thiosemicarbazone) li-
gands are coupled through an organic linker or where the two
zinc atoms are linked through a bridging sulfur, have been reported
[33–37]. However no anticancer activity has been evaluated for
these dinuclear compounds. Based on these observations, we felt
that dinuclear zinc bis(thiosemicarbazone) complexes would have
interesting biological properties. We have synthesized a series of
dinuclear zinc bis(thiosemicarbazone) complexes by varying sub-
stituents on the thiosemicarbazide part of the ligand and varying
the length of the bridging diaminoalkane. Their characterization
and structural influence on anticancer activity and imaging
2.1. Materials and methods
All reagents and solvents were purchased from either Aldrich
(USA) or Merck (India) or Spectrochem (India) and used without
further purification. 1,3-diaminopropane and CS2 were purified
by standard methods prior to use. 1H and 13C{1H} NMR spectra
were recorded on a Bruker AMX 400 spectrometer operating at
400 MHz for 1H and 100.6 MHz for 13C NMR with tetramethylsil-
ane as an internal standard. Infrared spectra were recorded on a
Bruker ALPHA10 FT-IR spectrometer operating on an ATR mode.
All UV–Vis and fluorescence spectra were recorded on PerkinElmer
(Lambda 750) and Horiba Jobin Yvon (FluoroMax-4) spectrofluo-

154
D. Palanimuthu, A.G. Samuelson / Inorganica Chimica Acta 408 (2013) 152–161
rometer, respectively. Mass spectra were recorded on an Agilent
6538 UHD Accurate-Mass QTOF-LC/MS instrument. Elemental
analyses were carried out using a Thermo Finnigan Flash EA
2000 CHNS analyzer. Molar conductivity measurements were car-
ried out on a Control Dynamics conductivity meter. Zinc standard
(ICP grade) was purchased from Fluka for the Inductively Coupled
Plasma-Optical Emission Spectroscopy (ICP-OES) and the amount
of zinc in cells was estimated by PerkinElmer Optima 2000 DV.
Supercoiled (SC) pBR322 DNA (cesium chloride purified) was pur-
chased from Bangalore Genie (India). Calf-thymus DNA was pro-
cured from Sigma (India). All reagents used for the cell culture,
MTT assay and DNA binding and cleavage experiments were ob-
tained from Sigma (USA).
3156 (m, NH), 1610 (m), 1541 (m, C@N), 1425 (s), 1406 (vs),
1244 (s, thioamide), 1192 (s), 1085 (vs), 837 (m, CS), 487 (m).
UV–Vis (DMSO): kmax(e
) = 338 nm (60800 Mꢀ1 cmꢀ1).
2.2.4. 1,3-Bis{biacetyl-20-(400-N-pyrrolidinethiosemicarbazone)-30-(400-
N-thiosemicarbazone)} propane (2)
Ligand 2 was prepared following the procedure described for
preparing ligand 1 except that 4-pyrrolidinylthiosemicarbazide
(460 mg, 1.28 mmol) was used instead of thiosemicarbazide,
mono-keto-(propane-1,3-dithiosemicarbazone)
(460 mg,
3.45 mmol) and glacial acetic acid (1 mL). The product was isolated
as a yellow powder (500 mg). Yield: 64%. 1H NMR (400 MHz, d6-
DMSO): d 1.87 (10H, bs, 4 ꢂ CH2, pyrrolidine, 1 ꢂ CH2, linker), d
2.14 (6H, s, CH3–C@N), d 2.22 (6H, s, CH3–C@N), d 3.65 (4H, q,
CH2–NH, linker), d 3.71 (8H, bs, CH2–N, pyrrolidine), d 8.63 (2H, t,
NH–CH2, linker), d 9.52 (2H, s, NH), d 10.29 (2H, s, NH). 13C NMR
(100.6 MHz, d6-DMSO): d 12.0 (CH3–C@N), d 12.5 (CH3–C@N), d
24.5–26.9 (CH2, broad, pyrrolidine), d 30.0 (CH2, linker), d 41.8
(CH2–NH, linker), d 52.0–54.3 (CH2-N, broad, pyrrolidine), d 149.4
(C@N), d 149.7 (C@N), d 178.3 (C@S), d 178.7 (C@S). HRMS (+ESI)
2.2. Syntheses
2.2.1. Mono-keto-(propane-1,3-dithiosemicarbazone)
Propane-1,3-dithiosemicarbazide (0.40 g, 1.8 mmol) and 2,3-
butanedione (0.41 g, 4.7 mmol) were added to water (20 mL) con-
taining conc. HCl (0.4 mL) and stirred for 30 min at room temper-
ature. The precipitate formed was collected by filtration, washed
with water and dried in air to obtain 0.42 g of the white solid.
Yield: 67%. 1H NMR (400 MHz, d6-DMSO): d 1.90 (2H, t, CH2), d
1.97 (6H, s, CH3–C@N), d 2.43 (6H, s, CH3–C@O), d 3.66–3.70 (4H,
q, CH2–NH), d 8.82 (2H, t, NH–CH2, linker), d 10.68 (2H, s, NH).
13C NMR (100.6 MHz, d6-DMSO): d 11.0 (CH3–C@N), d 25.8 (CH3–
C@O), d 29.4 (CH2), d 42.1 (CH2–NH), d 146.8 (C@N), d 179.3
(C@S), d 198.3 (C@O). IR Data/cmꢀ1: 3345 (m), 3245 (m, NH),
3236 (m, NH), 2936 (m), 2352 (m), 1675 (s, C@O), 1594 (m),
1539 (s), 1487 (s), 1237 (vs thioamide), 1172 (s), 1141(m), 964 (m).
found mass: 613.2454. Calc. mass for
C23H41N12S4 (100%,
[M+H+]+): 613.2440. Anal. Calc. for C23H40N12S4ꢁ2H2O (%): C,
42.57; H, 6.83; N, 25.90; S, 19.77. Found: C, 42.92; H, 6.34; N,
25.61; S, 20.81%. IR Data: IR Data/cmꢀ1: 2923 (w), 2863 (w),
1625 (w), 1525 (vs C@N),1433 (w, thioamide), 1343 (s), 1277 (s),
1249 (vs), 1188 (s), 1122 (s), 898 (w, CS), 791 (w), 572 (m). UV–
Vis (DMSO): kmax(e
) = 337 nm (38500 Mꢀ1 cmꢀ1).
2.2.5. 1,6-Bis{biacetyl-20-(400-N-thiosemicarbazone)-60-(400-N-
thiosemicarbazone)}hexane (3)
To a stirred suspension of mono-keto-(hexane-1,6-dithiosemic-
arbazone) (500 mg, 1.25 mmol) in ethanol (30 mL) thiosemicarba-
zide (240 mg, 2.64 mmol) followed by glacial acetic acid (0.9 mL)
were added and refluxed for 24 h. After cooling the solution to
room temperature, the precipitate that formed was collected by fil-
tration, washed with ethanol and dried in vacuo. A light yellow so-
lid was obtained (430 mg). Yield: 63%. 1H NMR (400 MHz, d6-
DMSO): d 1.31 (4H, bs, CH2), d 1.58 (4H, bs, CH2), d 2.16 (6H, s,
CH3–C@N), d 2.20 (6H, s, CH3–C@N), d 3.59 (4H, q, CH2–NH), d
7.86 (2H, s, NH2), d 8.39–8.42 (4H, m, 2 ꢂ NH–CH2 and NH2), d
2.2.2. Mono-keto-(hexane-1,6-dithiosemicarbazone)
Hexane-1,6-dithiosemicarbazide (2.5 g, 9.46 mmol) was added
to water (50 mL) containing conc. HCl (2.2 mL) and stirred until
it dissolved completely. The addition of 2,3-butanedione (2.4 g,
27.37 mmol) brought about the immediate precipitation of the de-
sired product which was collected by filtration, washed with water
and dried in air to get 3.3 g of the white solid. Yield: 86%. 1H NMR
(400 MHz, d6-DMSO): d 1.34 (4H, t, CH2), d 1.60 (4H, quintet, CH2),
d 1.95 (6H, s, CH3–C@N), d 2.41 (6H, s, CH3–C@O), d 3.59 (4H, q,
CH2–NH), d 8.65 (2H, t, NH–CH2), d 10.59 (2H, t, NH). 13C NMR
(100.6 MHz, d6-DMSO): d 10.9 (CH3–C@N), 25.7 (CH3–C@O), 27.1
(CH2), d 29.3 (CH2), d 44.8 (CH2–NH), d 146.4 (C@N), d 179.1
(C@S), d 198.4 (C@O). IR Data/cmꢀ1: 3249 (m, NH), 2981 (m),
2849 (m), 1620 (m, C@O), 1496 (s), 1262 (m, thioamide), 1178
(s), 1063(m), 945 (m).
10.16 (2H, s, NH, linker),
d
10.22 (2H, s, NH). 13C NMR
(100.6 MHz, d6-DMSO): d 12.4 (CH3–C@N), 12.6 (CH3–C@N), d
27.0 (CH2), d 29.5 (CH2), d 44.6 (CH2–NH), d 148.8 (C@N), d 149.3
(C@N), d 178.5 (C@S), d 179.7 (C@S). HRMS (+ESI) found mass:
547.1982. Calc. mass for C18H35N12S4 (100%, [M+H+]+): 547.1985.
Anal. Calc. for C18H34N12S4ꢁH2O (%): C, 38.27; H, 6.07; N, 29.74; S,
23.69. Found: C, 38.28; H, 6.42; N, 29.76; S, 22.72%. IR Data/
cmꢀ1: 3404 (m), 3183 (m, NH), 3150 (m), 2933 (m), 2858 (w),
1596 (s), 1488 (vs C@N), 1291 (s, thioamide), 1191 (m), 1083 (s),
2.2.3. 1,3-Bis{biacetyl-20-(4’’-N-thiosemicarbazone)-30-(4’’-N-
thiosemicarbazone)}propane (1)
A methanolic solution (15 mL) of mono-keto-(propane-1,3-
dithiosemicarbazone) (200 mg, 0.56 mmol) was refluxed for 1 h
and then thiosemicarbazide (110 mg, 1.2 mmol) and glacial acetic
acid (0.5 mL) were added. The resulting suspension was refluxed
for 4 h. After cooling to room temperature, the product was fil-
tered, washed with methanol and dried in vacuo. The product
was isolated as a cream coloured solid (265 mg). Yield: 94%. 1H
NMR (400 MHz, d6-DMSO): d 1.82 (2H, quintet, CH2), d 2.17 (6H,
s, CH3–C@N), d 2.22 (6H, s, CH3–C@N), d 3.65 (4H, q, CH2–NH), d
7.87 (2H, s, NH2), d 8.43 (2H, s, NH2), d 8.64 (2H, t, NH–CH2, linker),
d 10.25 (2H, s, NH), d 10.29 (2H, s, NH). 13C NMR (100.6 MHz, d6-
DMSO): d 12.6 (CH3–C@N), 12.7 (CH3–C@N), d 29.9 (CH2), d 41.7
(CH2-NH), d 149.1 (C@N), d 149.3 (C@N), d 178.7 (C@S), d 179.8
(C@S). HRMS (+ESI) found mass: 503.1512. Calc. mass for
831 (m, CS), 500 (m). UV–Vis (DMSO): kmax(e) = 338 nm
(52700 Mꢀ1 cmꢀ1).
2.2.6. 1,6-Bis{biacetyl-20-(400-N-pyrrolidinylthiosemicarbazone)-30-
(400-N-pyrrolidinylthiosemi carbazone)}hexane] (4)
Ligand 4 was prepared following the procedure used for prepar-
ing 3, except that 4-pyrrolidinylthiosemicarbazide (0.54 g,
3.72 mmol) was used instead of thiosemicarbazide, conc. HCl
(0.4 mL)
and
mono-keto-(propane-1,3-dithiosemicarbazone)
(0.4 g, 0.95 mmol). The product was isolated as a yellow powder
(500 mg). Yield: 80% (Purity of the compound is 90% as determined
by 1H NMR, and the remaining 8% has the unreacted mono-keto-
(hexane-1,6-dithiosemicarbazone). 1H NMR (400 MHz, d6-DMSO):
d 1.33 (4H, bs, CH2, linker), d 1.59 (4H, bs, CH2, linker), d 1.91
(8H, bs, CH2, pyrrolidine), d 2.13 (6H, s, CH3–C@N), d 2.18 (6H, s,
CH3–C@N), d 3.54 (4H, q, CH2–NH, linker), d 3.68 (8H, t, CH2–NH,
pyrrolidine), d 8.38 (2H, t, NH–CH2, linker), d 9.46 (2H, s, NH), d
C
15H29N12S4 (100%, [M+H+]+): 503.1515. Anal. Calc. for C15H28N12-
S4ꢁ3H2O (%): C, 32.24; H, 6.13; N, 30.08; S, 22.95. Found: C,
32.91; H, 5.76; N, 28.91; S, 22.32%. IR Data/cmꢀ1: 3237 (m, NH),

D. Palanimuthu, A.G. Samuelson / Inorganica Chimica Acta 408 (2013) 152–161
155
10.15 (2H, s, NH). 13C NMR (100.6 MHz, d6-DMSO): d 12.0 (CH3–
C@N), d 12.5 (CH3–C@N), d 25.5 (CH2, pyrrolidine), d 26.1 (CH2, pyr-
rolidine), d 27.1 (CH2, linker), d 29.6 (CH2, linker), d 44.6 (CH2–NH,
linker), d 52.1 (CH2–N, pyrrolidine), d 54.1 (CH2–N, pyrrolidine), d
148.9, (C@N), d 149.7, (C@N), d 178.3 (C@S), d 178.5 (C@S). HRMS
HRMS (+ESI) found mass: 675.0228. Calc. mass for C18H31N12S4Zn2
(100%, [M+H+]+): 675.0206. Anal. Calc. for C18H30N12S4Zn2ꢁ3H2O
(%): C, 29.71; H, 4.98; N, 23.10; S, 17.63. Found: C, 29.61; H,
4.41; N, 22.50; S, 18.24%. IR Data/cmꢀ1: 3223 (w, NH), 3173 (w,
NH), 2928 (m), 2858 (w), 1424 (vs C@N), 1364 (m), 1295 (m, thio-
amide), 1209 (s), 1179 (s), 1136 (m), 831 (m, CS), 710 (m). UV–Vis
(+ESI) found mass: 677.2710. Calc. mass for
C26H46N12S4Na
(100%, [M+Na+]+): 677.2743. Anal. Calc. for C26H46N12S4ꢁ2H2O (%):
C, 45.19; H, 7.29; N, 24.32; S, 18.56. Found: C, 45.50; H, 6.91; N,
24.18; S, 18.85%. IR Data/cmꢀ1: 3316 (m), 3175 (m, NH), 2933
(m), 2933 (m), 2861 (m), 2359 (m), 1487 (vs C@N), 1209 (vs thio-
(DMSO): kmax(e
) = 310, 435 nm (14500, 13300 Mꢀ1 cmꢀ1).
2.2.10. 1,6-Bis{biacetyl-20-(400-N-pyrrolidinylthiosemicarbazone)-30-
(400-N-pyrrolidinylthiosemi carbazone)zinc(II)} hexane (8)
amide), 1132 (vs), 550 (m). UV–Vis (DMSO): kmax
(
e
) = 338 nm
Complex 8 was prepared according to the general procedure de-
scribed for preparing 5 using zinc acetate (67 mg, 0.31 mmol) and
4 (100 mg, 0.15 mmol) in ethanol (10 mL). The product was ob-
tained as a yellow solid (88 mg). Yield: 75%. 1H NMR (400 MHz,
d6-DMSO): d 1.28 (4H, bs, CH2, linker), d 1.52 (4H, bs, CH2, linker),
d 1.85 (8H, bs, CH2, pyrrolidine), 2.16 (6H, s, CH3–C@N), 2.17 (6H, s,
CH3–C@N), d 3.30 (4H, bs, CH2–NH, linker), d 3.58 (8H, bs, CH2–N,
pyrrolidine), d 7.20 (2H, bs, NH). 13C NMR (100.6 MHz, d6-DMSO):
d 14.6 (CH3–C@N), d 14.8 (CH3–C@N), 25.4 (CH2, linker), d 27.3
(CH2, linker), d 29.9 (CH2, pyrrolidine), d 43.0 (CH2–NH, linker), d
49.4 (CH2–N, pyrrolidine), d 144.9 (C@N), d 175.6 (CS). HRMS
(47500 Mꢀ1 cmꢀ1).
2.2.7. 1,3-Bis{biacetyl-20-(400-N-thiosemicarbazone)-30-(400-N-
thiosemicarbazone)zinc(II)} propane (5)
2.2.7.1. General procedure for synthesizing zinc complexes. Ligand 1
(80 mg, 0.16 mmol) and zinc acetate (75 mg, 0.37 mmol) were sus-
pended in ethanol (10 mL) and refluxed for 8 h. The yellow precip-
itate formed was collected by filtration, washed with ethanol and
diethylether and dried in vacuo. The product 5 was obtained as a
pale yellow solid (52 mg).Yield: 52%. 1H NMR (400 MHz, d6-
DMSO): d 1.76 (2H, bs, CH2), d 2.15 (6H, s, CH3–C@N), d 2.19 (6H,
s, CH3–C@N), d 3.40–3.44 (4H, bs, CH2–NH), d 6.89 (4H, s, NH2), d
7.22 (2H, bs, NH, linker). 13C NMR (100.6 MHz, d6-DMSO): d 14.8
(CH3–C@N), d 15.0 (CH3–C@N), d 29.7 (CH2), d 40.0–41.0 (CH2–N,
merged with DMSO) d 145.0 (C@N), d 146.2 (C@N), d 176.2 (CS),
d 178.9 (CS). HRMS (+ESI) found mass: 632.9725. Calc. mass for C15-
H25N12S4Zn2 (100%, [M+H+]+): 632.9735. Anal. Calc. for C15H24N12-
S4Zn2ꢁ0.5H2O (%): C, 28.12; H, 3.93; N, 26.24; S, 20.02. Found: C,
28.63; H, 4.37; N, 24.18; S, 19.62%. IR Data/cmꢀ1: 3278 (w, NH),
3174 (w, NH), 2358 (w), 1617 (s), 1544 (m), 1488 (s, C@N), 1428
(vs), 1209 (s thioamide), 833 (m, CS), 727 (m). UV–Vis (DMSO):
(+ESI) found mass: 783.1155. Calc. mass for
C26H43N12S4Zn2
(100%, [M+H+]+): 783.1147. Anal. Calc. for C26H42N12S4Zn2ꢁ2.5H2O
(%): C, 37.77; H, 4.84; N, 20.33; S, 15.51. Found: C, 38.39; H,
5.35; N, 19.65; S, 15.67%. IR Data/cmꢀ1: 2928 (m), 2859 (m),
2359 (m), 1435 (vs C@N), 1216 (vs thioamide), 1083 (s), 835 (s,
CS).
UV–Vis
(DMSO):
kmax(e
) = 315,
443 nm
(13800,
13600 Mꢀ1 cmꢀ1).
2.3. Zinc estimation using ICP-OES
Uptake of zinc complexes was determined by measuring the
cellular zinc using ICP-OES. Exponentially growing HepG2 and
MCF-7 cells (10 million cells) were incubated with either 6 or 8
kmax(e
) = 312, 435 nm (17100, 16700 Mꢀ1 cmꢀ1).
2.2.8. 1,3-Bis{biacetyl-20-(400-N-pyrrolidinylthiosemicarbazone)-30-
(400-N-pyrrolidinylthiosemi carbazone)zinc(II)}propane (6)
(50 lM in zinc) for 1 h at 37 °C. After 1 h, the spent media was re-
moved, washed with PBS (2 ꢂ 5 mL) to remove excess complex
from extracellular media, scraped and collected in 5 mL PBS. The
cells were centrifuged at 2500 rpm for 15 min to get cell pellets,
then lysed with 1 M NaOH (0.5 mL) and diluted to 5 mL using 1%
(v/v) HNO3. The solution was subjected to ICP-OES measurement
using an instrument that was pre-calibrated for zinc using stan-
dard solutions containing 10, 50, 100, 500 and 1000 ppb zinc.
Complex 6 was prepared according to the general procedure de-
scribed for 5 using zinc acetate (80 mg, 0.36 mmol) and 2 (100 mg,
0.16 mmol) in ethanol (10 mL) to afford the desired product as a
yellow solid (75 mg).Yield: 63%. 1H NMR (400 MHz, d6-DMSO): d
1.76 (2H, quintet, CH2, linker), 1.85 (8H, bs, CH2, pyrrolidine), d
2.18 (6H, s, CH3–C@N), d 2.20 (6H, s, CH3–C@N), d 3.47 (4H, bs,
CH2–N, linker), d 3.58 (8H, bs, CH2–N, pyrrolidine), d 7.19 (2H, bs,
NH, linker). 13C NMR (100.6 MHz, d6-DMSO): d 14.6 (CH3–C@N), d
14.9 (CH3–C@N), 25.4 (CH2, linker), 29.8 (CH2, pyrrolidine), d
39.6–40.73 (CH2–NH, merged with DMSO, linker), d 49.5 (CH2–N,
pyrrolidine), d 144.9 (C@N), d 175.5 (C@S). HRMS (+ESI) found
mass: 741.0678. Calc. mass for C23H37N12S4Zn2 (100%, [M+H+]+):
741.0677. Anal. Calc. for C23H36N12S4Zn2ꢁH2O (%): C; 36.46; H,
5.06; N, 22.18; S, 16.93. Found: C, 36.51; H, 4.80; N, 21.80; S,
16.79%. IR Data/cmꢀ1: 3296 (w, NH), 2938 (w, NH), 2866 (m),
1487 (w, C@N), 1434 (vs thioamide), 1286 (m), 835 (w, CS), 744
2.4. Ethidium bromide displacement assay
The assay was carried out following previously reported litera-
ture procedure [30]. Purity of CT-DNA was evaluated in Milli-Q
water from the ratio of absorbance at 260 to 280 nm, which was
found to be >1.8, and the concentration was calculated from the
absorbance at 260 nm. DNA (424 lM) was added to an aqueous
solution containing ethidium bromide and sodium perchlorate
(400 mM) until the fluorescence intensity at 602 nm saturated
(kex = 546 nm). Subsequent addition of zinc complexes 5–8 with
(w), 622 (w). UV–Vis (DMSO): kmax(e) = 316,445 nm (15300,
16500 Mꢀ1 cmꢀ1).
increasing concentration (0–161.4
lM) quenched the fluorescence.
Based on fluorescence quenching, apparent binding constant (Kapp
)
,
2.2.9. 1,6-Bis{biacetyl-20-(400-N-thiosemicarbazone)-60-(400-N-
thiosemicarbazone)zinc(II)} hexane (7)
was calculated from the equation [EB] ꢂ KEB = [complex]50% ꢂ Kapp
where [EB] denotes the concentration of ethidium bromide and
[complex]50% is the concentration that is required to quench the
Complex 7 was prepared according to the general procedure de-
scribed for 5 using zinc acetate (140 mg, 0.63 mmol) and 3
(170 mg, 0.31 mmol) in ethanol (15 mL) to give the product as a
pale yellow solid (175 mg). Yield: 84%. 1H NMR (400 MHz, d6-
DMSO): d 1.28 (4H, bs, CH2), d 1.51 (4H, bs, CH2), d 2.15 (6H, s,
CH3–C@N), d 2.18 (6H, s, CH3–C@N), d 3.27 (4H, CH2–NH), d 6.88
(4H, bs, NH2), d 7.22 (2H, bs, NH, linker). 13C NMR (100.6 MHz,
d6-DMSO): d 14.8 (CH3–C@N), 14.9 (CH3–C@N), d 27.3 (CH2), d
29.8 (CH2), d 43.0 (CH2–NH, linker), d 145.2 (C@N), d 178.9 (CS).
fluorescence of the DNA-EB adduct by 50% (KEB = 1.0 ꢂ 107 Mꢀ1
,
[EB] = 2.6 lM).
2.5. DNA viscosity measurement
Viscometric titrations were carried out on a Ubbelohde viscom-
eter at 37 1 °C by varying concentration of zinc complexes 5–8
and at fixed concentration of CT-DNA (150 lM) in 5 mM Tris–HCl

156
D. Palanimuthu, A.G. Samuelson / Inorganica Chimica Acta 408 (2013) 152–161
(5 mM NaCl) buffer at pH 7.2. Ethidium bromide (EB) and Hoechst
33258 are included for comparison. The flow time was measured
after 5 min of incubation with each addition of the zinc complex.
Relative viscosities were calculated from the following equation:
pBR322 DNA by agarose gel electrophoresis. Stock solutions of
DNA, prepared in Milli-Q water (final DNA concentration of
ꢃ36
l
l
M; L), were incubated with ligand/zinc complex
M; 4 lL) for 4 h in 20 mM HEPES buffer, pH 7.2 at 37 °C.
4
l
(200
g
= (tꢀt0)/t0 where, t is the flow time of DNA with or without com-
Complex [Cu(phen)2](NO3)2 was used as a positive control. After
plex, and t0 is the flow time of buffer. Finally, relative specific vis-
cosity, (
incubation for the indicated time, loading buffer (4 L) containing
l
1/3
g/g0)
was plotted versus [compound]/[DNA], where
g
10 mM Tris–HCl, 0.1 M EDTA, 0.25% bromophenol blue was added
denotes relative viscosity of the DNA after the addition of complex
and g0 refers to the relative viscosity of DNA alone.
and frozen at ꢀ20 °C for 30 min to quench the reaction. Each reac-
tion mixture (7 lL) was loaded on the agarose gel (1%) and elec-
trophoresized for 3 h under dark condition at 50 V in TAE buffer
(40 mM Tris, 20 mM acetic acid, 1 mM EDTA). The gel was stained
2.6. Circular dichroism study
with ethidium bromide in Milli-Q water (1 lg/mL) for 30 min and
destained in Milli-Q water for 4 h and then photographed under
UV light. Percentage of cleaved DNA was estimated by measuring
relative intensities of supercoiled (SC) and nicked-circular (NC)
DNA bands using BioRad Gel Doc XR software.
CD spectra were recorded on a JASCO–J715 spectropolarimeter
at room temperature by keeping the concentration of CT-DNA con-
stant (100
lM) and varying the complex concentration from 0 to
100 M (ri = ([complex]/[DNA] = 0.0, 0.1, 0.5, 1.0) in 5 mM Tris–
l
HCl buffer (5 mM NaCl, pH 7.2) containing 1% DMSO.
2.7. MTT assay
3. Results and discussion
MTT assay was carried out as described earlier to measure cell
3.1. Synthesis and characterization
viability [30]. Approximately 3000 cells in 100
lL of growth media
were seeded in the wells of a 96-well plate. After 24 h, 100
various concentrations of copper bis(thiosemicarbazone) com-
plexes were added and incubated for 48 h at 37 °C in a CO2 incuba-
l
L of
Bis(thiosemicarbazone) ligands 1–4 were synthesized, in fairly
good yields, by using literature procedures after minor modifica-
tions as shown in Scheme 1 [35]. Synthetic routes of the precursor
thiosemicarbazides are depicted in Supplementary material as
Schemes S1 and S2. Dinuclear zinc complexes 5–8 were readily
prepared from the corresponding ligands 1–4 by refluxing with
zinc acetate in ethanol. All bis(thiosemicarbazone) ligands and
their dinuclear zinc complexes were characterized by routine spec-
troscopic techniques, and their purity was assessed by CHN analy-
sis (Fig. 2).
To determine the coordination mode of the ligand with zinc,
infrared spectra were recorded for all the ligands and their zinc
complexes. Upon complexation with zinc the characteristic
stretching frequencies of C@N (1540–1523 cmꢀ1) and C@S (896–
835 cmꢀ1) in the ligands were shifted to lower wave numbers in
the complexes (1491–1486 cmꢀ1 for C@N and 830–797 cmꢀ1 for
tor. At the 48th h of incubation, MTT (20
to the plate. The contents of the plate were pipetted out carefully
and the formazan crystals formed were dissolved in 200 l of
lL of 5 mg/mL) was added
l
DMSO, the absorbance was measured at 550 nm in a microplate
reader (Molecular Devices, Spectramax M5e). A graph of the con-
centration versus percentage cell viability was plotted and the con-
centration at which 50% cell death occurred was used as the IC50
value.
2.8. DNA cleavage experiments
The DNA cleavage activity of the bis(thiosemicarbazone) li-
gands and their zinc complexes were studied using supercoiled
Scheme 1. Synthetic strategy used in the preparation of dinucleating bis(thiosemicarbazone) ligands and their zinc complexes.

D. Palanimuthu, A.G. Samuelson / Inorganica Chimica Acta 408 (2013) 152–161
157
Fig. 2. Proposed molecular structures of bis(thiosemicarbazone) ligands 1–4 and their dinuclear zinc complexes 5–8.
C@S), indicating N2S2 mode of coordination. It is evident from 1H
NMR spectra that all ligands form neutral complexes with zinc
by losing four C(S)NH@C hydrogens that resonate between 10
and 11 ppm. The neutral nature of zinc complexes was verified
by measuring the molar conductance in DMSO (1 mM in zinc) at
room temperature. Complexes 5, 6, 7 and 8 showed low molar con-
ductances of 12, 10, 10 and 11 S m2 Mꢀ1, respectively, whereas
[Cu(phen)2](NO3)2, a known 1:2 electrolyte, showed a molar con-
ductance of 72 S m2 Mꢀ1 under similar conditions, confirming the
neutral nature of zinc complexes. High-resolution electrospray
ionization mass spectra (HR-ESI-MS) showed monoprotonated
([M+H+]+) or monosodiated ([M+Na+]+) peaks as the major species
in all the solutions of ligands and zinc complexes. Their isotopic
distributions conformed to the theoretically calculated isotopic
patterns.
It is interesting to note from 1H NMR spectra of the ligands 1
and 3 that NH2 protons displayed two singlets in the range of d
7.84–7.86 and d 8.40–8.41 ppm in 1:1 ratio, evidencing existence
of partial double bond character between C and N in S@C–NH2 that
restricts the free rotation around C–N [38]. However, such a double
bond character was absent in the zinc complexes 5 and 7 as the
NH2 protons exhibited single peak at d 6.89 ppm.
3.2. UV–Vis and fluorescence studies
The absorption spectra of ligands and their dinuclear zinc com-
plexes were recorded in DMSO (20 lM), and the representative
spectra are shown in Fig. 3A. All ligands displayed an intense band
centered at 337 nm along with a shoulder at 346 nm, assigned to
n ? p⁄ transitions of azomethine and thioamide, which is blue
shifted by ca. 30 nm with hypochromism upon complexation with
Fig. 3. UV–Vis spectra (A) and fluorescence spectra (B) of the ligand 2 (kex = 340 nm) and its zinc complex 6 (kex = 420 nm) in DMSO (20 lM in bis(thiosemicarbazone) or in
zinc).

158
D. Palanimuthu, A.G. Samuelson / Inorganica Chimica Acta 408 (2013) 152–161
Table 1
Photophysical properties of dinuclear zinc complexes.
Complexes Absorbance kmax (nm)
Emissiona kmax
(nm)
Quantum yieldb
U
Optical brightness (
e
ꢂ
U,
(e
ꢂ 104 Mꢀ1 cmꢀ1
)
ꢂ 10ꢀ3
Mꢀ1 cmꢀ1
)
5
6
7
8
312 (1.71), 435 (1.67)
316 (1.53), 445 (1.65)
310 (1.45), 435 (1.33)
315 (1.30), 443 (1.50)
525
530
525
530
1.5
2.4
1.6
2.4
22.4
31.9
19.4
27.2
a
Emission spectra were measured in DMSO solvent with 420 nm excitation at 5 nm slit width.
b
Quantum yield was calculated using [Ru(bpy)3](PF6)2
(U
= 42 ꢂ 10ꢀ3) as a standard.
zinc, indicating the coordination of azomethine nitrogen and CS
sulfur to zinc [39]. We observed an additional low energy absorp-
tion band at ca. 440 nm in zinc complexes. The new band is as-
cribed to sulfur to zinc charge transfer (LMCT) transition.
control. The GI50 values, corresponding to the cytotoxic activity, are
presented in Table 2. The cytotoxicity in HepG2 cells revealed that
the ligand 2 and its zinc complex 6 are quite cytotoxic, with a GI50
value of <0.1 lM. Their potency is greater than all other analogous
Several zinc bis(thiosemicarbazone) complexes are known to
exhibit weak fluorescence [25,28,30], therefore we measured the
fluorescence spectra of ligands 1–4 and the corresponding zinc
complexes 5–8 in DMSO following excitation at 340 and 420 nm,
respectively, and the representative spectra are shown in Fig. 3B.
The photophysical properties of 5–8 are listed in Table 1. Invari-
ably, all the ligands exhibited an emission at 415 nm, while the
zinc complexes exhibited emission at longer wavelengths, with a
red shift of 115–120 nm. Complexes 6 and 8 exhibited an emission
centered at 530 nm which is 5 nm greater than what is observed
for 5 and 7. The quantum yields for zinc complexes were deter-
mined using [Ru(bpy)3](PF6)2, as a standard for which a quantum
compounds tested, and is similar to that of adriamycin. When the
measurement was extended to four other cell lines, only 6 exhib-
ited significant cytotoxicity (<0.1 lM) against the MCF-7 cell line,
and is comparable to the cytotoxicity caused by adriamycin. We
also measured the cytotoxicity of all compounds by MTT assay
against MCF-7 and HepG2 cell lines. The cytotoxicity values ob-
tained are represented as IC50 and are tabulated in Table S1 (Sup-
plementary material). Complex 6 was found to be the most
cytotoxic complex, with IC50 values of 12.34 4.92
lM against
HepG2 cells and 11.00 4.20 M against MCF-7 cells. Based on
l
the cytotoxicity we selected 6 for further studies to probe its inter-
action with DNA and its cellular uptake.
yield (U
) of 42 ꢂ 10ꢀ3 has been reported [40,41]. Complexes 6
and 8 having a pyrrolidine substituent on the thiosemicarbazide
fragment exhibited a quantum yield of 2.4 ꢂ 10ꢀ3 which is about
1.5 times higher than that of analogous complexes 5 and 7, and
comparable to that of [(Zn(GTSC)]3, whose quantum yield was re-
ported to be 3.0 ꢂ 10ꢀ3 and displayed intracellular fluorescence
[30]. Hence these complexes are anticipated to show fluorescence
in cells.
3.4. Cellular uptake of zinc in cells
To verify the importance of cellular uptake, and its correlation
with cytotoxic activity, exponentially growing MCF-7 and HepG2
cells were treated with zinc complexes 6 and 8 (50 lM in zinc)
for 60 min at 37 °C and the accumulation of zinc concentration in
the cell was estimated using inductively coupled plasma-optical
emission spectrometry (ICP-OES). In this study, DMSO (0.5% in
OptiMEM media) treated cells were used as a vehicle control. To
obtain the detectable amount of zinc by the ICP-OES technique,
cells were incubated with complexes for a short time (1 h), but
3.3. Cytotoxicity studies
Cytotoxicity was screened initially for all compounds against a
range of human cancer cell lines, derived from five different tissue
types that include SiHa (cervical cancer), MCF-7 (breast cancer),
PC-3 (prostate cancer), A-2780 (ovarian cancer) and HepG2 (hepa-
tocellular liver cancer) using a colorimetric based sulphorhod-
amine-B (SRB) assay. These cell lines were treated with
compounds at 10ꢀ4, 10ꢀ5, 10ꢀ6 and 10ꢀ7 M and the growth inhibi-
tion (GI50) was measured at 48 h after staining with SRB. Adriamy-
cin, an anthracycline based anticancer drug, was used as a positive
at much higher concentration (50 lM in zinc) than the observed
IC50. The treatment of complexes 6 and 8 significantly increases
the zinc concentration in cells compared to untreated cells
(Fig. 4) that suggests the facile intake of complexes. Interestingly
both cytotoxic complex 6 and non-cytotoxic complex 8 accumulate
the same amount of zinc within MCF-7 and HepG2 cell lines in
spite of their different cytotoxicity. This result suggests that differ-
ences in cellular uptake are not causing differences in their cyto-
toxicity. The intracellular behavior is likely to be significantly
different for these two complexes.
Table 2
The GI50 of dinucleating bis(thiosemicarbazone) ligands and their zinc complexes as
measured by SRB assay.
a
3.5. Cellular fluorescence studies by flow cytometry
Compounds
GI50
(lM)
SiHa
MCF-7
PC-3
A-2780
HepG2
As the most cytotoxic complex was also the most fluorescent
complex 6, we investigated the ability of the complex to fluoresce
within the cell. We treated HepG2 cells for a short time, 30 min, at
1
2
3
4
5
6
7
8
n.d.
n.d.
n.d.
n.d.
>100
<0.1
>100
>100
>100
<0.1
>100
>100
<0.1
>100
>100
>100
n.d.
41.8
>100
>100
<0.1
36.6
>100
>100
n.d.
<0.1
47.0
>100
<0.1
70.4
>100
>100
n.d.
50.6
>100
>100
<0.1
61.0
>100
>100
n.d.
34.2
75.0
>100
<0.1
room temperature with 20 lM and 50 lM of complex 6 in PBS buf-
fer, pH 7.2 containing 1% DMSO. Cells that were treated with DMSO
(1% DMSO in PBS buffer) were used as a control. Fluorescence from
the cell was measured using a flow cytometer following excitation
at 488 nm. As shown in Fig. 5, the complex 6 treated cells exhibited
only 1.4–1.5-fold fluorescence enhancement compared to auto-
Adriamycinb
n.d. = Not determined.
fluorescence (DMSO treated cells) at both 20 and 50 lM. These re-
Data are presented as a mean of three independent experiments.
a
sults suggest that 6 does not exhibit fluorescence in the cell though
the fluorescence quantum yield in DMSO is comparable with
[Zn(GTSC)]3, a bis(thiosemicarbazone) complex that displayed
Refers to the amount of drug (
lM in bis(thiosemicarbazone) or in zinc) required
to inhibit 50% cell growth in 48 h.
b
Adriamycin was included for comparison.

D. Palanimuthu, A.G. Samuelson / Inorganica Chimica Acta 408 (2013) 152–161
159
Fig. 4. Cellular zinc content was estimated in MCF-7 cells (panel A) and HepG2 cells (panel B) using inductively coupled plasma optical emission spectrometry after treatment
of DMSO (0.5% in OptiMEM media) as a control, 6 and 8 (50
lM in zinc) for 1 h at 37 °C. Error bars indicate standard deviations for three experiments.
Table 3
Binding constants of 5–8 (in zinc) with CT-DNA.a
Complexes
Kapp (ꢂ104 Mꢀ1
)
5
6
7
8
6.4 0.1
7.4 0.2
9.2 0.1
12.8 0.2
a
Each value is the average value from two sets of
experiments.
[30]. It was noticed that the pyrrolidinylthiosemicarbazide based
complexes possess marginally superior binding propensities over
the thiosemicarbazide based complexes. Overall, the DNA interac-
tion studies suggest that the zinc complexes interact with DNA via
electrostatic interaction similar to that of mononuclear zinc diace-
tylbis(4-methyl-3-thiosemicarbazone) [Zn(ATSM)] [25].
Fig. 5. HepG2 cells were incubated either with 6 at 20 and 50 lM (in zinc) or with
DMSO (1% in PBS) for 30 min at room temperature and fluorescence was recorded
using a flow cytometer following excitation at 488 nm.
good fluorescence in several cancer cell lines [30]. The poor fluo-
rescence response in the cell is possibly due to the quenching of
fluorescence in aqueous environments. To validate this hypothesis,
3.6.2. Viscosity measurements
fluorescence spectra was recorded for 6 at 20 lM and 50 lM in the
To understand the nature of interaction between zinc com-
plexes 5–8 and DNA, viscosity experiments were carried out on a
Ubbelohde viscometer at 37 °C by varying concentration of zinc
complexes versus CT-DNA. Hydrodynamic experiments such as
viscosity and sedimentation methods are sensitive to DNA length.
Classical mode of interactions like intercalation increases the vis-
cosity by increasing the distance between base pairs, which in turn
increases the overall length of DNA. Conversely, non classical mode
of interactions such as electrostatic and groove binding causes the
reduction or no change in the DNA viscosity with different concen-
trations. The effect of zinc complexes or EB or Hoechst 33258 on
the relative viscosities of CT-DNA solution is shown in Fig. 7. As ex-
pected, the EB, a classical intercalator, dramatically increases the
contour length of the DNA and thereby its viscosity. On the other
hand, Hoechst 33258, a known minor groove binder, does not elon-
gate DNA and causes a slight decrease in the viscosity (Fig. 7).
However, the addition of increasing amounts of zinc complexes
to DNA solution does not increase the relative viscosity signifi-
cantly, indicating that the zinc complexes interact with DNA nei-
ther by intercalation nor is it a minor groove binder like Hoechst
33258. This result supports our DNA binding studies that weak
electrostatic interactions between the negatively charged DNA
and the positively polarized zinc, which has a vacant coordination
site in the complex, is preferred over other interactions.
cuvette at various compositions of DMSO and PBS buffer (0–99%)
following excitation at 420 and 488 nm (Fig. S1, Supplementary
material). We observed significant quenching (84–94%) in the fluo-
rescence intensity of 6 (kex = 488 nm) when the percentage of PBS
buffer increased to 98% in DMSO. This suggests that fluorescence
quenching is associated with aqueous quenching, which would re-
sult in poor fluorescence inside the cell.
3.6. DNA interaction studies
3.6.1. DNA binding
DNA is one of the major targets for several anticancer active
compounds including metal complexes [42–45]. The increased
activity of dinuclear or multinuclear platinum complexes com-
pared to cisplatin could be traced to the ability of the molecule
to bind DNA in a specific fashion [46]. Therefore, we investigated
the ability of metal complexes 5–8 to interact with calf-thymus
DNA by ethidium bromide displacement assay using fluorescence
spectroscopy. Incremental addition of complexes to a solution con-
taining EB-DNA mixture (fluorescent adduct) quenches the fluores-
cence gradually by displacing EB from DNA. The apparent binding
constants (Kapp) were estimated from the concentration needed to
50% fluorescence quenching. Kapp values are presented in Table 3
and a representative diagram is depicted in Fig. 6. All complexes
exhibited good binding efficacy to DNA, with Kapp varying in the
narrow range of 6.4 ꢂ 104 Mꢀ1 to 12.8 ꢂ 104 Mꢀ1. These binding
constants are in the same range of mononuclear zinc complexes
3.6.3. Circular dichroism studies
To examine if binding of zinc complexes to DNA causes changes
in the conformation of DNA, circular dichroism (CD) studies were

160
D. Palanimuthu, A.G. Samuelson / Inorganica Chimica Acta 408 (2013) 152–161
Fig. 6. DNA binding behavior of dinuclear zinc complexes as measured by ethidium bromide displacement assay. (A) Addition of increasing concentrations 5 (0 to 161.4
l
M)
to DNA-EB mixture (DNA = 424 lM; EB = 2.6 lM) in aqueous NaClO4 (400 mM) solution quenches the fluorescence. (B) A plot of F/F0 vs. complex concentration at 602 nm
(kex = 546 nm).
Table 4
In vitro pBR322 DNA cleavage by bis(thiosemicarbazone)
ligands 1–4 and their zinc complexes 5–8 at 200
lM.
Compounds
% DNA cleavagea
1
2
3
4
5
6
7
8
38
39
39
b
b
35
33
33
a
DNA alone rendered an average cleavage of 25%.
Percentage of cleavage could not be determined
b
due to precipitation.
Fig. 7. Effect of complexes 5–8 on the relative viscosity of CT-DNA. Complexes 5–8
were added at increasing concentrations (0, 30, 60, 90, 120, 150 M in zinc) to CT-
DNA (150 M) at 37 °C in 5 mM Tris–HCl buffer (5 mM NaCl) at pH 7.2. Data shown
is an average of three experiments SD.
l
l
[complex]/[CT-DNA] ratios (0.0, 0.1, 0.5 and 1.0) do not show any
significant changes in base stacking or in helicity. This type of a
behavior is typical of electrostatic interaction between metal com-
plexes and DNA [48]. Moreover this result is consistent with our
DNA binding and viscosity studies.
carried out with CT-DNA in the absence or in the presence of
increasing concentrations of zinc complexes 6 and 8 (Fig. 8). CD
spectrum of CT-DNA in Tris–HCl buffer (5 mM) displayed a positive
band centered at 275 nm due to base stacking and a negative band
centered at 248 nm due to right-handed helicity of DNA [47]. These
bands are characteristic for DNA in the right-handed B form.
Apparently the addition of zinc complexes to CT-DNA at different
3.6.4. DNA cleavage studies
We also probed the ability of the complexes to cleave DNA by
agarose gel electrophoresis after incubating compounds 1–8 with
supercoiled pBR322 plasmid DNA for 4 h at 37 °C (Fig. S2 and S3,
Fig. 8. CD spectra of CT-DNA (100
lM) in 5 mM Tris–HCl buffer (5 mM NaCl, pH 7.2), in the absence and in the presence of increasing concentrations (0, 10, 50 and 100 lM) of
6 (A) and 8 (B) to DNA (ri = [complex]/[DNA] = 0.0, 0.1, 0.5 and 1.0).

D. Palanimuthu, A.G. Samuelson / Inorganica Chimica Acta 408 (2013) 152–161
161
[8] S. van Zutphen, J. Reedijk, Coord. Chem. Rev. 249 (2005) 2845.
Supplementary material). Apparently, neither the ligand nor the
zinc complex cleaves even 50% of DNA at 200 M (Table 4). In con-
trast, [Cu(phen)2](NO3)2, a positive control, cleaves ꢃ80% DNA at
20 M. These results suggest that both ligands and their zinc com-
[9] N.J. Wheate, S. Walker, G.E. Craig, R. Oun, Dalton Trans. 39 (2010) 8113.
[10] N.P.E. Barry, P.J. Sadler, Chem. Commun. 49 (2013) 5106.
[11] Available from: <http://www.clinicaltrials.gov/>.
[12] Available from: <http://www.cancer.gov/>.
[13] J.D. Roberts, J. Peroutka, N. Farrell, J. Inorg. Biochem. 77 (1999) 51.
[14] M.-G. Mendoza-Ferri, C.G. Hartinger, R.E. Eichinger, N. Stolyarova, K. Severin,
M.A. Jakupec, A.A. Nazarov, B.K. Keppler, Organometallics 27 (2008) 2405.
[15] C. Manzotti, G. Pratesi, E. Menta, R. Di Domenico, E. Cavalletti, H.H. Fiebig, L.R.
Kelland, N. Farrell, D. Polizzi, R. Supino, G. Pezzoni, F. Zunino, Clin. Cancer Res.
6 (2000) 2626.
l
l
plexes do not manifest their cytotoxicity through DNA cleavage.
4. Conclusion
We have synthesized a family of interesting dinuclear zinc
bis(thiosemicarbazone) complexes with varying length of the lin-
ker connecting the two bis(thiosemicarbazone) moieties and the
nature of the substituents on the thiosemicarbazide part of the li-
gand. These zinc complexes displayed green fluorescence in DMSO
but not in the cell or in aqueous media. Cytotoxicity studies against
a range of cancer cell lines reveal that only complex 6 is signifi-
cantly anticancer active, suggesting that the cytotoxicity is depen-
dent on several factors including the length of the carbon chain
linker and substituents on the thiosemicarbazide. In MCF-7 and
HepG2 cells, both complexes 6 and 8 are taken up in similar
amounts. Based on DNA binding assessed by ethidium bromide
displacement assay, viscosity analysis, CD spectral analysis and
DNA cleavage studies we conclude that these complexes effectively
interact with DNA in an electrostatic manner without causing any
DNA cleavage. Mechanistic studies are underway to elucidate the
root of cytotoxicity at the cellular level.
[16] K. Suntharalingam, A.J.P. White, R. Vilar, Inorg. Chem. 49 (2010) 8371.
[17] P. Molenveld, J.F.J. Engbersen, H. Kooijman, A.L. Speck, D.N. Reinhoudt, J. Am.
Chem. Soc. 120 (1998) 6726.
[18] X.-L. Zhao, Z.-S. Li, Z.-B. Zheng, A.-G. Zhang, K.-Z. Wang, Dalton Trans. 42
(2013) 5764.
[19] A. Medina-Molner, B. Spingler, Chem. Commun. 48 (2012) 1961.
[20] J.R. Dilworth, R. Hueting, Inorg. Chim. Acta 389 (2012) 3.
[21] A.K. Nandi, S. Chaudhuri, S.K. Mazumdar, S. Ghosh, Inorg. Chim. Acta 92 (1984)
235.
[22] D.X. West, A.E. Liberta, S.B. Padhye, R.C. Chikkate, P.B. Sonawane, A.S.
Kumbhar, R.G. Yerande, Coord. Chem. Rev. 123 (1993) 49.
[23] E. Lopez-Torres, M.A. Mendiola, Inorg. Chim. Acta 363 (2010) 1735.
[24] B.M. Paterson, P.S. Donnelly, Chem. Soc. Rev. 40 (2011) 3005.
[25] A.R. Cowley, J. Davis, J.R. Dilworth, P.S. Donnelly, R. Dobson, A. Nightingale, J.M.
Peach, B. Shore, D. Kerr, L. Seymour, Chem. Commun. (2005) 845.
[26] I.H. Hall, C.B. Lackey, T.Y. Kistler, J.S. Ives, H. Beraldo, L.J. Ackerman, D.X. West,
Arch. Pharm. Pharm. Med. Chem. 333 (2000) 217.
[27] C.R. Kowol, R. Trondl, V.B. Arion, M.A. Jakupec, I. Lichtscheidl, B.K. Keppler,
Dalton Trans. 39 (2010) 704.
[28] S.I. Pascu, P.A. Waghorn, T.D. Conry, B. Lin, H.M. Betts, J.R. Dilworth, R.B. Sim,
G.C. Churchill, F.I. Aigbirhio, J.E. Warren, Dalton Trans. 16 (2008) 2107.
[29] D. Palanimuthu, S.V. Shinde, D. Dayal, K. Somasundaram, A.G. Samuelson, Eur.
J. Inorg. Chem. (2013) 3542.
[30] D. Dayal, D. Palanimuthu, S.V. Shinde, K. Somasundaram, A.G. Samuelson, J.
Biol. Inorg. Chem. 16 (2011) 621.
Acknowledgements
[31] D. Palanimuthu, S.V. Shinde, K. Somasundaram, A.G. Samuelson, J. Med. Chem.
56 (2013) 722.
[32] S.I. Pascu, P.A. Waghorn, T.D. Conry, H.M. Betts, J.R. Dilworth, G.C. Churchill, T.
Pokrovska, M. Christlieb, F.I. Aigbirhio, J.E. Warren, Dalton Trans. (2007) 4988.
[33] A.I. Matesanz, I. Cuadrado, C. Pastor, P. Souza, Z. Anorg. Allg. Chem. 631 (2005)
780.
[34] M. Christlieb, A.R. Cowley, J.R. Dilworth, P.S. Donnelly, B.M. Paterson, H.S.R.
Struthers, J.M. White, Dalton Trans. (2007) 327.
[35] M. Christlieb, H.S.R. Struthers, P.B. Bonnitcha, A.R. Cowley, J.R. Dilworth, Dalton
Trans. (2007) 5043.
D. P. thanks Council of Scientific and Industrial Research (CSIR),
New Delhi for senior research fellowship. Authors acknowledge
Department of Science and Technology (DST), New Delhi for re-
search grant. We thank Prof. K. Somasundaram and Ms. V. S. Sride-
vi, Department of Microbiology and Cell Biology, Indian Institute of
Science (IISc) for MTT and flow cytometry data. We thank Prof. S. P.
Sharma, Department of Molecular Biophysics Unit (MBU), IISc for
help in obtaining CD data.
[36] M.L. Duran, A. Sousa, J. Romero, A. Castineiras, E. Bermejo, D.X. West, Inorg.
Chim. Acta 294 (1999) 79.
[37] E.L. Torres, M.A. Mendiola, C.J. Pastor, B.S. Perez, Inorg. Chem. 43 (2004) 5222.
[38] P. Tarasconi, S. Capacchi, G. Pelosi, M. Cornia, R. Albertini, A. Bonati, P.P.
Dall’Aglio, P. Lunghi, S. Pinelli, Bioorg. Med. Chem. 8 (2000) 157.
[39] A. Castineiras, E. Bermejo, D.X. West, L.J. Ackerman, Polyhedron 18 (1999)
1463.
[40] J.V. Caspar, T.J. Meyer, J. Am. Chem. Soc. 105 (1983) 5583.
[41] A.T.R. Williams, S.A. Winfield, J.N. Miller, Analyst 108 (1983) 1067.
[42] A.C. Komor, J.K. Barton, Chem. Commun. 49 (2013) 3617.
[43] L.H. Hurley, Nat. Rev. Cancer 2 (2002) 188.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ica.2013.09.014.
References
[44] E. Gallori, C. Vettori, E. Alessio, F.G. Vilchez, R. Vilaplana, P. Orioli, A. Casini, L.
Messori, Arch. Biochem. Biophys. 376 (2000) 156.
[45] K.E. Erkkila, D.T. Odom, J.K. Barton, Chem. Rev. 99 (1999) 2777.
[46] N.J. Wheate, J.G. Collins, Coord. Chem. Rev. 241 (2003) 133.
[47] C.-Y. Gao, X. Qiao, Z.-Y. Ma, Z.-G. Wang, J. Lu, J.-L. Tian, J.-Y. Xu, S.-P. Yan,
Dalton Trans. 41 (2012) 12220.
[1] B. Rosenberg, L. VanCamp, J.E. Trosko, V.H. Mansour, Nature 222 (5191) (1969)
385.
[2] R.A. Alderden, M.D. Hall, T.W. Hambley, J. Chem. Educ. 83 (2006) 728.
[3] K.J. Kilpin, P.J. Dyson, Chem. Sci. 4 (2013) 1410.
[4] C.G. Hartinger, N. Metzler-Nolte, P.J. Dyson, Organometallics 31 (2012) 5677.
[5] J. Zhang, F. Zhang, H. Li, C. Liu, J. Xia, L. Ma, W. Chu, Z. Zhang, C. Chen, S. Li, S.
Wang, Curr. Med. Chem. 19 (2012) 2957.
[48] N. Shahabadi, M. Falsafi, N.H. Moghadam, J. Photochem. Photobiol. B: Biol. 122
(2013) 45.
[6] P.C.A. Bruijnincx, P.J. Sadler, Curr. Opin. Chem. Biol. 12 (2008) 197.
[7] M. Galanski, Curr. Med. Chem. 12 (2005) 2075.