Surface ligand influenced free radical protectionof superparamagnetic iron oxide nanoparticles (SPIONs) towardH9c2 cardiac cells

Faruq Mohammad • Nor Azah Yusof

Received: 20 March 2014 / Accepted: 23 May 2014 / Published online: 14 June 2014

� Springer Science+Business Media New York 2014

Abstract There have been a number of studies which

deal with either toxic or non-toxic nature of superpara-

magnetic iron oxide nanoparticles (SPIONs); however,

there is no clear cut information about their exact behavior

and the reasons for its dual action. The objective of the

present study was to investigate the SPIONs having similar

oxidation states, but varying surface ligands and their role

in terms of protecting the iron-mediated toxic responses.

The four different SPIONs includes: (i) SPIONs containing

oleic acid (SPIONs-1), (ii) SPIONs without any surface

ligand (SPIONs-2), (iii) SPIONs containing cysteamine

ligand (SPIONs-3), and (iv) SPIONs having both of oleic

acid and cysteamine ligand. The particle size, surface

functionality, and electronic oxidation states were con-

firmed by the HRTEM, FT-IR, and XPS analysis, respec-

tively. On in vitro testing of all four SPIONs with H9c2

cardiomyocyte cell line, the SPIONs-2 without any surface

ligand found to exhibit significant decrease in the viability

of cells at a concentration of 200 lg mL-1 for 16-h

exposure period. Further investigation of toxicity mecha-

nism resulted in the fact that the SPIONs-2 involved in the

formation of ROS due to the role played by the more

electron deficient Fe3? form of iron, there by decreased the

glutathione release, increased DNA cleavage, and dis-

rupted the mitochondrial transmembrane potential. How-

ever, the presence of unsaturation and/or thiol group (–SH)

containing ligands on other SPIONs protected the cardiac

cells from undergoing ROS-induced oxidative stress. Fur-

ther, the results of the study confirming the importance of

having unsaturated double bonds and/or –SH group pos-

sessing ligands onto the surface of SPIONs by means of

protecting the cells from the influence of electron deficient

Fe3? state of iron.

Abbreviations

SPIONs Superparamagnetic iron oxide nanoparticles

NPs Nanoparticles

HRTEM High-resolution transmission electron

microscopy

FT-IR Fourier transform infrared spectroscopy

XPS X-ray photoelectron spectroscopy

ROS Reactive oxygen species

Introduction

In recent years, the increased awareness of superparamag-

netic iron oxide nanoparticles (SPIONs) are due to their

interesting physico-chemical properties and unique mag-

netic behavior, having made them to be potential in the

current biomedical and bioengineering applications as

probes for imaging-based disease diagnosis, hyperthermia-

based cancer therapy, targeted drug delivery, cell sorting,

tissue repair, and also as biomaterial scaffolds [1, 2]. Since

the properties of iron oxide nanoparticles (NPs) are mostly

surface dependent, and there are a number of ways by which

changing the size, shape, or surface chemistry can bring

about concomitant changes in their physico-chemical and

biological properties. The commonly available methods for

the surface modification of SPIONs such as the biomolecule

F. Mohammad (&) � N. A. Yusof (&)

Institute of Advanced Technology, Universiti Putra Malaysia,

43400 Serdang, Selangor, Malaysia

e-mail: farooqm1983@gmail.com; faruq_m@upm.edu.my

N. A. Yusof

e-mail: azahy@upm.edu.my

123

J Mater Sci (2014) 49:6290–6301

DOI 10.1007/s10853-014-8354-5

conjugation, stabilization by organic surfactants, polymeric

encapsulation, inorganic core–shell formation, and biofab-

rication are employed in order to safe upper limit the uses of

SPIONs for in vivo applications with increased biocom-

patibility [2]. Also in some studies, the delivering agents

such as poly(ethylene glycol) [3], dendrimers [4], HIV-

derived TAT protein [5], dimercaptosuccinic acid [6], etc.

are employed to optimize the delivery of SPIONs into cells

by endocytosis. For in vivo applications, the selection of a

specific coating type or derivatization of SPIONs should be

made only on the basis of the nature of end use, as the

properly sequestered iron oxide or its complexes can sig-

nificantly enhance its role in biomedical field. Despite the

enormous applications of ligand-protected SPIONs, the

toxicity originating from the surface of naked SPIONs are

concerning their use for a majority of biological applica-

tions in vivo [7]. Since vast quantities of engineered iron

oxide nanomaterials are released into the human environ-

ment due to the emerging applications of NPs-based tech-

nology in everyday life and more chances are that the

released NPs may come into contact with lungs, liver,

cardiac, and blood by means of air or water pollution and

can initiate short-/long-term toxicological effects. Also, the

low soluble nature of iron oxide persisting them in the

biological moieties especially in liver, lung, or heart and

therefore it is very important to thoroughly investigate the

intracellular effects of nanosized iron oxide on normal cell

behavior [8].

In general, iron oxide exists in several phases with different

oxidation states and structures that includes Fe0, FeO (wus-

tite), Fe3O4 (magnetite), c-Fe2O3 (maghemite), a-Fe2O3

(hematite), and FeOOH (oxyhydroxide). Among all, magne-

tite and maghemite are the most abundant and stable forms of

iron and both exhibits superparamagnetism (magnetization

values falls in between ferro and paramagnetism) when falls

within a nanosize range (\15 nm). The two different oxida-

tion states of 2? and 3? are exhibited by Fe in magnetite,

while the Fe in maghemite has only one oxidation state of 3?

[9, 10]. For a majority of biomedical applications where the

magnetic properties as well as electron rich surfaces are

required, magnetite is preferred most due to the presence of

Fe2? state as it has a potential to further act as an electron

donor. In addition, the magnetite form is relatively stable in

biological environment while the other forms such as ma-

ghemite and hematite are prone to induce free radical-

dependent oxidative stress and resulting cell viability losses

on various cell types due to the presence of more electron

deficient Fe3? [6, 11, 12]. The toxicity effects from iron oxide

are mostly due to the catalytic property developed at the

nanoscale and the cell lines tested includes the Caco-2 intes-

tinal cells [12], J774 macrophage cells [7], pheochromocy-

toma PC12M cells [6], HL-7702 hepatocyte cells [13], A549

lung epithelial cells [14], cardiac microvascular endothelial

cells [15], etc. It was observed in a recent study that the free

radical production (O2�2, OH�) in the biological environment

is mostly caused due to the catalytic reactions occurring at the

surface of iron oxide NPs, but not due to the dissolved metal

ions. In addition, the catalytic centers on the surface of iron

oxide NPs possess at least 50-fold more effective in OH�radical production than the corresponding dissolved Fe3? ions

[11]. This indicates that the protection of iron oxide NP’s

surface from involving catalytic reactions either directly with

the cells intracellular components or indirectly with biological

media can prevent the cells undergoing free radical-induced

oxidative stress. Also, the normal cells under physiological

conditions continuously produce the reactive oxygen species

(ROS) by metabolic pathways, which are effectively neu-

tralized by the intracellular antioxidants like glutathione and

other specific enzymes. However, the increased toxicity of

ROS-induced oxidative stress by exposure to iron-based

materials is due to the interruption of ROS-antioxidant bal-

ance through Fenton and/or Haber–Weiss cycle [11, 16].

Therefore, by the incorporation of suitable ligands with anti-

oxidant property onto the surface of iron oxide NPs can pro-

tect iron from mediating catalytic reactions (Fenton) and at

the same time, the surface ligand serves also as a free radi-

cal scavenger to neutralize the ROS formed by any other

means.

In the present study, we aim to investigate and compare the

cellular function and toxicological effects of SPIONs having

no ligands with that of SPIONs possessing variable surface

chemistry. The dose- and time-dependent toxic potential, as

well as the role of oxidative stress in the development of Fe-

induced toxicity were measured using the in vitro cell culture

system. The four different SPIONs with varying surface

ligands were tested against the H9c2 cardiomyoblast cell line

and the reason for choosing this cell type is because of the

increasing number of patients suffering from cardiac diseases

(atherosclerosis) originating in one way by means of exposure

to metal oxide present in pollution [17]. Moreover to the best

of author’s knowledge, this is the first report dealing with the

cardiotoxic effects of iron oxide NPs, as we proposed the free

radical-mediated myocyte damage mechanism to explain the

cardiotoxicity by SPIONs. The four SPIONs with change of

surface chemistry includes: (1) SPIONs coated with oleic acid

(SPIONs-1), (2) SPIONs without any surface coating (SPI-

ONs-2), (3) SPIONs coated with cysteamine (SPIONs-3), and

(4) SPIONs coated with both of oleic acid and cysteamine

(SPIONs-4). The two selected ligands exhibit antioxidative

capacity to some level, i.e., by the presence of unsaturation in

oleic acid and due to the reducing capacity of the –SH (thiol)

group in cysteamine [18, 19]. The SPIONs-induced cardiac

damage was assessed by means of cell viability and prolifer-

ation, intracellular ROS formation, DNA damage, mito-

chondrial transmembrane potential (DWm), and glutathione

(GSH) release.

J Mater Sci (2014) 49:6290–6301 6291

123

Materials and methods

Chemicals and reagents

Iron (III) acetylacetonate (99.9 %), 1,2 hexadecanediol

(90 %), oleic acid (99 %), diphenyl ether (99 %), and

cysteamine hydrochloride (98 %) were purchased from

Sigma-Aldrich (St. Louis, MO, USA). All chemicals were

used as received.

Rat embryonic H9c2 cardiomyoblast cell line (ATCC�

CRL-1446TM), Dulbecco’s modified Eagle’s medium

(DMEM) were obtained from ATCC (Manassas, VA,

USA); penicillin–streptomycin (stabilized solution con-

taining 10,000 units mL-1 penicillin and 10 mg mL-1

streptomycin), phosphate buffered saline (PBS), trypsin–

EDTA (2.5 g trypsin and 0.2 g EDTA-Na4 per liter of

HBSS) from Sigma-Aldrich (St. Louis, MO); fetal bovine

serum (FBS) from Atlanta Biological (Lawrenceville, GA,

USA); tissue culture flasks (T25) and 96-well plates from

Corning (Acton, MA, USA). Kits for cell viability

(CellTiter-Blue�) from Promega (Madison, WI, USA);

MitoCaptureTM for mitochondrial DWm from Calbiochem

(San Diego, CA, USA); cell death detection ELISA kit

from Roche (Indianapolis, IN, USA); 5-(and-6)-chloro-

methyl-2,7-dichlorodihydrofluorescein diacetate (CM-H2-

DCFDA; Molecular Probes, Grand Island, NY, USA);

ApoGSHTM glutathione fluorometric assay kit (BioVision,

Milpitas, CA, USA).

Synthesis of SPIONs with varying surface ligands

The synthesis method to obtain SPIONs is the decompo-

sition of organometallics at higher temperature conditions

in the presence of a suitable reducing agent and surfactant

[20]. Briefly, 0.7 g of iron acetylacetonate was mixed with

a solution of 20 mL of diphenyl ether and 2 mL of oleic

acid under inert atmosphere with vigorous stirring. To this

solution, 2.6 g of 1,2-hexadecanediol was added. The

reaction mixture was heated to around 210 �C with reflux

for 2 h, maintaining oxygen-free conditions. The reaction

mixture was cooled to room temperature, and then,

degassed ethanol was added to precipitate the black-col-

ored product. The precipitate was separated out by centri-

fugation and washed the product first with hexane and then

with ethanol. The product was dispersed in ethanol and

separated magnetically and dried to obtain a black-colored

powder, named as SPIONs-1. The SPIONs-1 contains the

oleic acid groups on their surface as surfactant molecules.

Further, the SPIONs-1 was subjected to heat up to a tem-

perature of 400 �C for 1 h in vacuum oven to remove the

surfactant molecules. Thus the formed product does not

contain any oleic acid surface groups, named as SPIONs-2.

In the following step, the formed SPIONs-2 was used for

conjugation with cysteamine biomolecule to obtain SPI-

ONs-3 [21]. Briefly, 10 mL of toluene was added to 30 mg

of SPIONs-2 and shaken thoroughly. In another flask,

30 mg of cysteamine hydrochloride in 10 mL of distilled

water was prepared. Both of the solutions were added

together and stirred for about 45 min. The SPIONs-3

(cysteamine-bound SPIONs-2) was separated by using a

magnet, washed 2–3 times with ethanol, and dried under

inert atmosphere. Finally, the SPIONs containing both of

oleic acid and cysteamine (SPIONs-4) were obtained by

using the same procedure to form SPIONs-3 with a

replacement of SPIONs-2 by SPIONs-1.

The high-resolution transmission electron microscopy

(HRTEM) studies were carried out on JEOL-2010 HRTEM

instrument with a point to point resolution of 1.94 A,

operated at a 200 kV accelerating voltage. The samples for

HRTEM studies were prepared by taking a droplet of

SPIONs of different type individually in hexane onto a

carbon-coated copper grid and evaporating the solvent at

room temperature. Fourier transform infrared (FT-IR)

spectra were obtained using Nexus 670 FT-IR spectrometer

in the transmission mode and the samples were prepared by

grinding the material with dry KBr powder and then made

into a transparent pellet. X-ray photoelectron spectroscopy

(XPS) analyses were done on AXIS 165 high performance

multi-technique surface analyzer which is based on a

165 mm mean radius hemispherical analyzer, with an eight

channeltron detection system.

Cell culture and maintenance

The H9c2 cardiomyoblast cells were maintained in DMEM

containing 10 % (v/v) FBS and 1 % antibiotics (penicillin,

100 units mL-1; streptomycin, 100 lg mL-1) and incu-

bated at 37 �C, 5 % CO2, and in 95 % humidity. For the

treatment of SPIONs of different type, the H9c2 cells were

maintained in the culture medium containing 2 % FBS

[20].

In vitro cell viability and proliferation assay

Approximately 104 cells well-1 in 100 lL of culture

medium were seeded onto 96-well plates. After 24–36 h of

incubation (cells were *80 % confluent), the medium was

replaced by 100 lL each of DMEM containing 2 % (v/v)

FBS and NPs of SPIONs-1 through SPIONs-4 individually

(concentration range: 25–400 lg mL-1, 24 h). Also, the

SPIONs-induced cell proliferation studies were carried out

for various time intervals in the range of 0–24 h at a

concentration of 200 lg mL-1 (selection of concentration

is based on the IC50 value obtained from the concentration-

dependent cell viability studies). The cells after the incu-

bation period, the supernatants were separated out from the

6292 J Mater Sci (2014) 49:6290–6301

123

wells, cell monolayers were washed with PBS (0.1 M, pH

7.4) and fresh culture medium was added. The metabolic

activity of the cells treated with SPIONs of various types

was assessed by using the CellTiter-Blue� reagent from

Promega. The assay involves the reduction of resazurin, a

non-fluorescent blue dye, into a pink-colored, red-fluores-

cent resorufin, that can be quantified at excitation and

emission wavelengths of 560 and 590 nm (respectively)

using a Spectramax EM Gemini spectrofluorimeter

(Molecular Devices, Sunnyvale, CA, USA). Briefly, 25 lL

of the CellTiter-Blue� reagent added to each well, incu-

bated for an additional 3–4 h and the amount of resorufin

produced during this time was quantified. The background

fluorescence from wells that contained SPIONs in culture

medium but no added cells was subtracted. The percentage

cell viability was calculated with respect to the control,

untreated cells (set at 100 %). The values shown are

mean ± SD (standard deviation) of 3–5 individual exper-

iments [22].

Intracellular ROS measurements

The SPIONs-induced intracellular ROS measurements were

based on the hydrolysis and subsequent oxidation of 5-(and-

6)-chloromethyl-2,7-dichlorodihydrofluorescein diacetate

(CM-H2DCFDA; Molecular Probes). Briefly, the H9c2 cells

grown in 24-well plates were incubated with 10 mM CM-

H2DCFDA in Krebs–Ringer/HEPES (KRH) buffer (115 mM

NaCl, 5 mM KCl, 1 mM KH2PO4, 1.5 mM CaCl2, 1.2 mM

MgSO4, 5 mM glucose, and 25 mM HEPES, pH 7.4) for

30 min at 37 �C. After the incubation, the cells were washed

twice with fresh KRH buffer and kept at room temperature

for another 30 min. Now the cells were exposed to

200 lg mL-1 concentrations each of SPIONs-1, SPIONs-2,

SPIONs-3, and SPIONs-4, and the changes in fluorescence

were measured after 16 h of incubation time using Spectra-

max Gemini EM spectrofluorometer at excitation and emis-

sion wave-lengths of 485 and 538 nm, respectively. Further,

in order to confirm the formation of ROS, the SPIONs-2 was

co-incubated with the known antioxidant Trolox (at 200 and

500 lM concentration). The background fluorescence

obtained from the control wells was subtracted at all points.

Menadione (100 lM) was used as a positive control and the

results of various treatments were expressed as percentage

fluorescence relative to the positive control (100 %) [16].

Intracellular GSH release

The intracellular GSH was measured based on a glutathione-

S-transferase (GST)-catalyzed conjugation of GSH to mono-

chlorobimane by using ApoGSHTM glutathione fluorometric

assay kit (BioVision). After the designated treatment of each

SPIONs type to the H9c2 cells (at 200 lg mL-1, 16 h), the

medium was removed and the cells were lysed, followed by

incubation in the presence of 50 lM monochlorobimane and

0.1 unit mL-1 GST for 30 min at 37 �C (in accordance with

the manufacturer’s protocol). The fluorescence was quantified

by using Spectramax Gemini EM spectrofluorometer at

excitation and emission wavelengths of 380 and 460 nm,

respectively. The results are normalized to cell number and

expressed as percentage change in GSH concentration relative

to the untreated control measurements. The values shown are

the mean ± SD of three individual experiments [23].

DNA fragmentation assay

The H9c2 cells plated in 24-well plates (ca. 5 9 104) when

reaches to confluency, were treated with 200 lg mL-1

concentrations of all four SPIONs for a time period of 16 h.

At the end of each incubation period, the cytoplasmic

histone-associated DNA fragments were measured using

cell death detection ELISA kit from Roche. Briefly, the

cells were lysed in 200 lL of lysis buffer provided by the

manufacturer, and the lysate (20 lL each) was added to

streptavidin-coated 96-well plates, to which a mixture of

biotinylated anti-histone and peroxidase-coupled anti-DNA

antibody was added. This was incubated for another 2 h at

room temperature and then washed. The histone-associated

DNA fragments were quantified based on the activity of

peroxidase measured at 405 nm using a BioTek ELx800

microplate reader (Winooski, VT, USA). Further, the

supplementation of DNA cleavage was tested by co-incu-

bation of SPIONs-2 with Trolox in a concentration-

dependent manner. The results expressed as fold-increase

relative to the untreated control and the values shown are

mean ± SD of three individual experiments [16].

Measurement of mitochondrial transmembrane

potential (DWm)

The mitochondrial transmembrane potential (DWm) in

H9c2 cells following the treatment of SPIONs of different

types, were accessed using a MitoCapture kit. The assay is

based on the accumulation of a cationic dye, 5,50,6,60-tet-

rachloro-1,10,3,30-tetraethyl-benzimidazolyl carbocyanine

iodide (JC-1) in mitochondria and other cytoplasmic frac-

tions. The dye remains monomeric when the value of DWm

is small and exhibits green fluorescence; however, for a

high DWm, a greater accumulation of the dye forming

J-aggregates and the fluorescence shifts to red (Emax =

590 nm). The cells of either type in 24-well plates (80 %

confluent) were treated with 200 lg mL-1 concentrations

of all four SPIONs and incubated for a 16-h period. After

that, the JC-1 dye was added to the cells and incubated for

another 30 min at 37 �C as per the manufacturer’s

instructions. The unbound dye was removed by washing

J Mater Sci (2014) 49:6290–6301 6293

123

with PBS, fresh medium was added and the fluorescence

was quantified at excitation and emission wavelengths of

488 and 590 nm, respectively, using a Spectramax EM

Gemini spectrofluorimeter. Rotenone (10 lM) was used as

positive control and the results are expressed as percentage

decrease in DWm compared to the untreated controls [23].

Statistical analysis

Data are presented as mean ± SD of 3–5 separate experi-

ments. Statistical analyses were performed using a one-way

analysis of variance (ANOVA) and Bonferroni’s method

for multiple comparisons. Probability of p \ 0.05 was

considered as statistically significant and p \ 0.01 as

highly statistical significant.

Results

The SPIONs containing oleic acid surface ligands (SPIONs-

1) are formed first by the decomposition of iron acetylace-

tonate in the presence of oleic acid as surfactant and 1,2-

hexadecanediol as reducing agent at a temperature of 210 �C

and inert atmosphere. The formed SPIONs-1 in the fol-

lowing step were heated in the oven up to 400 �C to remove

the surfactant molecules and the formed SPIONs-2 are free

from the oleic acid ligands. Further, SPIONs-2 and SPIONs-

1 were reacted with cysteamine hydrochloride at room

temperature to form SPIONs-3 and SPIONs-4, respectively.

Thus produced SPIONs-1 through SPIONs-4 were charac-

terized by HRTEM analysis, FT-IR spectroscopy, XPS and

further tested for cytotoxicity by H9c2 cardiomyoblast cells.

Figure 1 shows the HRTEM images (Fig. 1a–d) and the

corresponding particle size distributions (Fig. 1e–h) of

SPIONs-1, SPIONs-2, SPIONs-3, and SPIONs-4, respec-

tively. From the observation of HRTEM images, all the

SPIONs seem to be spherical, uniform and the size distri-

bution analysis gave the information that the mean sizes of

SPIONs-1 through SPIONs-4 are found to be 5.36 ± 0.77,

5.67 ± 0.0.87, 5.75 ± 0.93, and 5.57 ± 0.82 nm, respec-

tively. It can be clearly seen from the mean sizes of all the

SPIONs that there were no changes occurred to the particle

size of SPIONs on either heating in order to remove sur-

factant or bonding with cysteamine biomolecule. However,

the removal of surfactant molecules from SPIONs-1 to form

SPIONs-2, made the particles to be aggregated, i.e., with an

exception to other SPIONs, the SPIONs-2 particles are not

monodispersed which can be seen clearly in Fig. 1b.

The FT-IR spectrums were recorded in order to confirm

the surface functionality in SPIONs-1 through SPIONs-4 and

are shown in Figs. 2 and 3. The comparison of SPIONs-1

having oleic surface ligands (a) with that of pure oleic acid

(b) and SPIONs-2 having no surface ligand (c) were shown

in Fig. 2. Briefly, from the mixed spectrums of (a) and (b),

the two peaks in the range of 2850–3000 cm-1 are due to the

symmetric and asymmetric methylene (CH2) stretching

modes of groups present in oleic acid. The peak around

1630 cm-1 in (a) and (b) is due to the m(C=C) stretching

mode, the broad peak at 3424 cm-1 in (b) is from the free

O–H stretching vibrations present in oleic acid and, however,

the missing of that peak can be due to its participation in

hydrogen bonding. The sharp peak at 1712 cm-1 in (a) is

due to m(C=O) which was observed in (b) at around

1630 cm-1 and the peak at 960 cm-1 in (a) and (b) can be

from the =C–H bending vibrations. Similarly, the appearance

of peaks in (b) and (c) in the range of 530–570 and

520–540 cm-1 are due to the splitting vibrations of m1 and m2

bands, respectively, for the Fe–O of magnetite NPs [24, 25].

Therefore, the matching of peaks between pure oleic acid

and SPIONs-1 confirming the presence of oleic acid surface

ligands in SPIONs-1 and the absence of those peaks con-

firming that the SPIONs-2 are free from the surface ligands.

Similarly, the FT-IR spectrums of SPIONs-1 (a), SPIONs-

4 (b), and SPIONs-3 (d) were compared with that of pure

cysteamine (c) and are shown in Fig. 3. On comparison of the

spectral peaks in (b), (c), and (d), the presence of sharp peak in

(c) around 3354 cm-1 due to –NH2 stretching changing to a

broad peak at 3395 cm-1 as a result of polymeric hydrogen

bonding in (b) and (d). The peak in (c) at 1641 cm-1 due to

strong in pane –NH2 scissoring absorptions appeared at

1630 cm-1 in (b) and (d) and the –CH2 bending vibrations are

observed in the range of 1450–1475 cm-1 in (b) and (d). Also,

the peaks in (b) and (d) around 1000–1260 cm-1 are due to C–

N stretching vibrations and the peaks at 930–950 cm-1 are

from the C–H bending vibrations [26]. The FT-IR comparison

of (b) and (d) with that of (c), confirming the presence of

cysteamine biomolecule on the surface of SPIONs-3 and

SPIONs-4. From the FT-IR comparison of all the spectrums in

Figs. 2 and 3, it proves that SPIONs-1 contains oleic acid

ligands, SPIONs-2 does not contain any surface ligand, SPI-

ONs-3 contain cysteamine, and SPIONs-4 contain both of

cysteamine and oleic acid ligands.

Determining and understanding the exact oxidation

states of Fe and S for all of the SPIONs is very much

important in the investigation of toxicity mechanism, as it

plays a crucial role in either forming the ROS or neutral-

izing the formed ROS in SPIONs-treated cells. Figure 4

shows the XPS of Fe 2p spectrum in SPIONs-1 through

SPIONs-4 (Fig. 4a) and S 2p spectrum in SPIONs-3 and

SPIONs-4 (Fig. 4b). From the Fig. 4a, the Fe 2p spectrum

for all the SPIONs shows a doublet containing a lower

energy band (2p3/2) and a high energy band (2p1/2) at 710.3

and 724.6 eV for SPIONs-1, 710.9 and 723.9 eV for SPI-

ONs-2, 710.5 and 723.4 eV for SPIONs-3, and 709.3 and

724.5 eV for SPIONs-4, respectively. The observed Fe 2p3/2

6294 J Mater Sci (2014) 49:6290–6301

123

and Fe 2p1/2 binding energy positions for all of the SPIONs

are in good agreement with the literature data for Fe, i.e.,

the peak positions indicating the presence of Fe in the

metallic states of Fe (2?) and Fe (3?), which confirm that

the iron oxide in SPIONs is magnetite (Fe3O4) but not

maghemite (c-Fe2O3) [27]. Similarly, the S 2p1/2 spectrum

shown in Fig. 4b represents the peak positions at 161.8 eV

for SPIONs-3 and 162.0 eV for SPIONs-4, which corre-

sponds that the presence of sulfur to be in either elemental

state or monovalent [28].

Figure 5a and b shows the SPIONs-induced cytotoxicity

studies of H9c2 cells, the concentration-dependent

(Fig. 5a) and time-dependent (Fig. 5b). From the concen-

tration-dependent cytotoxicity studies, it was observed that

the SPIONs-2 particles exhibited significant reduction

(40 %) in the percentage viability of cells at a concentra-

tion of 200 lg mL-1. However, the observation of a non-

toxic behavior in SPIONs-1, SPIONs-3, and SPIONs-4

treated cells can be due to the presence of biomolecules of

oleic acid and cysteamine surface ligands and the absence

of any such biomolecule on SPIONs-2 caused for a sig-

nificant decrease in cell viability. Also from the time-

dependent cytotoxicity studies for a 200 lg mL-1 con-

centration of SPIONs-1 through SPIONs-4, the toxicity

Fig. 1 HRTEM images of a SPIONs-1 (scale = 20 nm), b SPIONs-2 (scale = 100 nm), c SPIONs-3 (scale = 20 nm), and d SPIONs-4

(scale = nm). The particle size distributions of SPIONs-1, SPIONs-2, SPIONs-3, and SPIONs-4 are shown in e, f, g, and h, respectively

Fig. 2 FT-IR comparison of the spectrums of pure oleic acid (a) with

that of SPIONs-1 (b) and SPIONs-2 (c)

Fig. 3 FT-IR comparison of the spectrums of SPIONs-1 (a),

SPIONs-4 (b), pure cysteamine (c), and SPIONs-3 (d)

J Mater Sci (2014) 49:6290–6301 6295

123

effects seems to be prominent (45 %) after 16 h of expo-

sure period. Further, an in-depth mechanism for the

observation of toxicity in SPIONs-2 treated cells and the

absence of such cell death in SPIONs-1, SPIONs-2, and

SPIONs-4, the role played by oleic acid and cysteamine

were investigated by the assays of intracellular ROS, for-

mation of DNA fragments, mitochondrial damage, and

glutathione release.

The comparison of intracellular ROS formed in the cells

treated with SPIONs-1 through SPIONs-4 at 200 lg mL-1

concentration for a 16-h exposure period were shown in

Fig. 6. One can see clearly from the figure that when

compared with the positive reference (Menadione, 100 lM)

and untreated control measurements, the SPIONs-2 treated

cells exhibited significant amount of ROS production

(76 %). However, no such ROS production was witnessed

when the same SPIONs-2 particles were co-incubated with

the well-known antioxidant, Trolox (a water soluble analog

of a-tocopherol). Also, we see from the figure that a dose-

dependent reduction in the amount of ROS on co-incubation

of SPIONs-2 with Trolox and the reason can be due to the

direct scavenging of the formed ROS, thereby making it less

available for reaction with the probe CM-H2DCF [23].

Moreover, the other SPIONs such as SPIONs-1, SPIONs-3,

and SPIONs-4 did not seem to form much ROS production

as compared against untreated control and the reference. The

reason can be the fact that the presence of oleic acid or

cysteamine biomolecule on the surface of these SPIONs

Fig. 4 a XPS of Fe

2p spectrum in SPIONs-1

through SPIONs-4 and b S

2p spectrum in SPIONs-3 and

SPIONs-4

Fig. 5 Cell viability studies of SPIONs (1–4) treated H9c2 cells

a concentration-dependent (24 h) and b time-dependent. The values

shown are the mean ± SD of 3–5 individual experiments. A positive

control of phorbol myristate acetate (PMA; 10 lg mL-1) was used

(not shown in figure). From the figure, asterisk and double asterisk

indicate the significance at p \ 0.05 and p \ 0.01 versus untreated

controls

6296 J Mater Sci (2014) 49:6290–6301

123

must have either protected the formation of ROS from the

surface of iron oxide or neutralized the ROS as soon as they

are formed.

The comparison of the amount of DNA fragmentation

occurred in SPIONs-1 through SPIONs-4 treated H9c2 cells

(200 lg mL-1, 16 h) were shown in Fig. 7. From the figure,

it can be observed clearly that the SPIONs-2 treated cells

exhibited drastic increase (fivefold) in the amount of DNA

cleavage and this fragmentation seems to be reduced on co-

incubation with Trolox in a concentration-dependent man-

ner. Also, when compared with the untreated control mea-

surements, the SPIONs-1, SPIONs-3, and SPIONs-4 treated

cells does not seem to involve in the production of much

DNA fragments due to the protective action of oleic acid

and cysteamine surface ligands. Within SPIONs-1, SPIONs-

3, and SPIONs-4, a very less amount of DNA fragments are

observed in SPIONs-4 treated cardiac cells due to the pre-

sence of both the groups of oleic acid and cysteamine.

Similarly, the SPIONs-1 through SPIONs-4 treated

H9c2 cells and their changes in GSH at 200 lg mL-1

concentration for a 16-h exposure period are shown in

Fig. 8. From the analysis of results, it was observed that the

SPIONs-2 treated cells exhibited significant reduction

(42 %) in the total GSH (%) release when compared

against the untreated control measurements. Also, the

effect seems to be restored on supplementation of the

SPIONs-2 with the antioxidant Trolox (500 lM) and,

however, the other SPIONs do not seem to involve in the

loss of total GSH release as compared to the controls.

Figure 9 shows the comparison of % change in the

mitochondrial transmembrane potential of H9c2 cells fol-

lowed by the treatment of SPIONs-1 through SPIONs-4 at

200 lg mL-1 concentration for a 16-h exposure period.

From the figure, it is clearly evident that as similar to ROS

production and DNA cleavage, a much change (63 %) in the

membrane potential of cardiac cells treated with SPIONs-2

are observed when compared against the untreated controls.

The decreased transmembrane potential due to lower dye

accumulation in the mitochondria confirms to an apoptotic

pathway of cell death in SPIONs-2 treated cells. It can also

be observed from the figure that the supplementation of

SPIONs-2 with Trolox (500 lM) does not allow the cardiac

cells to undergo loss of membrane potential due to the

scavenging of ROS by Trolox. However, the other cell

treatments of SPIONs-1, SPIONs-3, and SPIONs-4 does not

seem to involve any changes to the membrane potential as

compared against the SPIONs-untreated controls.

Discussion

The physical characterization of SPIONs-1 through SPI-

ONs-4 provided the information that all the particles are in

spherical shape, nanosize with a size distribution of around

5 nm (Fig. 1), the presence of surface functionality

(Figs. 2, 3), and the corresponding Fe, S oxidation states

(Fig. 4). The understanding of Fe oxidation states in all of

the SPIONs is extremely important as it helps to correlate

the toxicity effects and other intracellular responses

induced by Fe. Since, among different forms of iron oxide,

some forms in particular are established to be more toxic to

the cells due to the deficiency of electrons, while others are

comparatively stable as they do not allow toxic mediated

responses. For example, iron oxide, two of its forms

Fig. 6 Comparison of

intracellular ROS formation (%)

in cardiac cells on treatment of

SPIONs-1 through SPIONs-4

(200 lg mL-1, 16 h) and the

effect of supplementation of

Trolox. A positive control of

Menadione (100 lM) was used

and the H9c2 cells without any

treatment were used as negative

control. The values shown are

mean ± SD of three separate

experiments and asterisk

indicates significance at

p \ 0.05 versus the control

measurements

J Mater Sci (2014) 49:6290–6301 6297

123

Fig. 7 Comparison of the DNA

fragmentation of SPIONs-1,

SPIONs-2, SPIONs-3, and

SPIONs-4 treated H9c2 cells at

200 lg mL-1 concentration for

a 16-h exposure period and the

protective action on co-

incubation with Trolox. The

values shown are the fold-

increase relative to the untreated

control measurements. The

results are expressed as the

mean ± SD of three individual

experiments; asterisk and

double asterisk indicate the

significance at p \ 0.05 and

p \ 0.01 versus untreated

controls

Fig. 8 % change in the total

GSH release of SPIONs-1

through SPIONs-4 treated H9c2

cells at 200 lg mL-1

concentration for a 16-h period.

The data shown are the

mean ± SD of three individual

experiments and the cells of no

SPIONs treatment were used

control measurements; double

asterisk indicate the significance

at p \ 0.01 versus untreated

controls

Fig. 9 Comparison of %

change in the DWm of H9c2

cells treated with SPIONs-1

through SPIONs-4 of

200 lg mL-1 concentration for

a 16-h exposure period. A

positive control of rotenone

(10 lM) and a negative control

of H9c2 cells without any

treatment were used. The data

shown are the mean ± SD of

three individual experiments

and from the figure, double

asterisk indicate the significance

at p \ 0.01 versus untreated

controls

6298 J Mater Sci (2014) 49:6290–6301

123

(among five), the hematite (a-Fe2O3), and maghemite

(c-Fe2O3) are found to be toxic relative to the magnetite

form (Fe3O4) due to the presence of Fe in 2?, 3? oxidation

states. However, the presence of iron in 3? oxidation state

in hematite and maghemite, which is a more electron

deficient state compared to Fe2? and hence it tries to

absorb electrons from the electron-rich intracellular com-

ponents, thereby initiating the free radical species forma-

tion [6, 11, 12, 20]. Thus formed free radicals if not

neutralized, can result in ROS-induced oxidative stress

responses such as the DNA strand cleavage and its frag-

mentation, affecting the mitochondrial function by red-ox

cycling mediated oxidation, etc. The other means of pro-

tecting the iron surface from involving cytotoxic responses

is by either forming a biocompatible metallic shell (Au,

Ag, or Si) or conjugation/encapsulation into biomolecules/

polymers having antioxidative property [20]. In our case, it

was confirmed from the XPS studies (Fig. 4) that the

formed iron oxide in all of the four SPIONs is magnetite

and the existence of Fe in both of its Fe2? and Fe3? oxi-

dation states. Also, when compared to hematite and ma-

ghemite, the amount of Fe3? form (a more electron

deficient) is less in the magnetite form and therefore, pri-

marily we can say that our SPIONs are stable.

On testing of SPIONs-1 through SPIONs-4 with H9c2

cardiac cells in our case, all the three SPIONs (SPIONs-1,

SPIONs-3, and SPIONs-4) did not showed much changes

in the percentage viability of cells up to 500 lg mL-1

concentration, while the SPIONs-2 only exhibited signifi-

cant fall in the viability at 200 lg mL-1 after 16 h (Fig. 5).

Since from the literature data, it was established that the

observation of iron toxicity in one way is due to the gen-

eration of cellular oxidative stress and therefore, we

directly proceeded to test the influence of ROS on exposure

of SPIONs-1 to SPIONs-4 (200 lg mL-1, 16 h) [11, 12].

When tested, the SPIONs-2 exposure to H9c2 cells only

resulted in a drastic increase in the amount of ROS pro-

duction as compared to the untreated controls, while other

three SPIONs did not seem to undergo any ROS formation.

In addition, the observation of dose-dependent fall in the

production of ROS on co-incubation of SPIONs-2 with

Trolox, confirming the free radical-induced pathway of cell

death in SPIONs-2 treated cardiac cells (Fig. 6). Further

observation of toxic responses by means of DNA frag-

mentation (Fig. 7), amount of GSH release (Fig. 8), and

mitochondrial membrane potential changes (Fig. 9) for

only SPIONs-2 treated cells and the simultaneous dimin-

ishing effect with Trolox ratifying the apoptotic pathway of

cell death mediated by ROS. Because of not having any

protecting ligand in SPIONs-2 while the presence of

ligands in other three SPIONs protected the cells from

undergoing apoptosis. In general, for the cells undergoing

apoptosis, the glutathione synthesis is blocked at a certain

point of time and the observation of a slight higher level of

glutathione in SPIONs-4 treated cells (compared with the

controls), however, might be due to the sparing effect from

the two coated ligands on the newly synthesized GSH by

the cardiac cells (Fig. 8). The two coated ligands with an

unsaturated double bond and thiol functionality might have

contributed for an enhanced production of GSH by

diminishing the free radical effects and thereby facilitated

for an environment that is supporting for the healthy

growth of cardiac cells. Similarly, the observation of DNA

fragments (Fig. 7) and the physical damage of mitochon-

drial membrane (Fig. 9) are due to the localization of

SPIONs-2 into the mitochondria and the corresponding

oxidative damage of organelle, disrupting its function and

causing the cells to death. Since the organelle is more

susceptible to ROS-induced oxidative stress as its inner

membrane is composed of unsaturated fatty acids, densely

packed proteins, and DNA molecules that are essential for

mitochondrial function [29].

In the current study, the observed toxicity differences

between SPIONs-2 treated cells with that of other SPIONs

are attributed to the protective action against ROS provided

by the oleic acid and cysteamine ligands. For those two, the

antioxidant capacity supplied by the unsaturation (C=C

bond) in oleic acid and the –SH group in cysteamine, i.e.,

the p electrons in C=C or the two lone pairs in S of thiol

group reduces the Fe3? in iron oxide by itself getting the

ligand groups oxidized. This was supported by a similar

study that the persistence of antioxidative nature of gluta-

thione is contributed by its –SH moiety, as it can be oxi-

dized or reduced reversibly at a high rate [30]. In general,

the Fe present in iron oxide possesses some pro-oxidant

properties due to the electron deficient Fe3? state and the

availability of total amount of such Fe3? state decides the

extent of toxicity, as the Fe in iron oxide exists in two

different states of Fe2? and Fe3?. Whenever the Fe3? form

gets in contact with the electron-rich cell surfaces, they

readily absorb electrons and responsible for the generation

of ROS. The formed ROS if not controlled, can finally lead

to cell death, however in our case, the loss of cardiac

function is observed by the formed ROS. The presence of a

suitable ligand on the surface of iron oxide protects the

intracellular components from undergoing oxidation and

furthers the ROS formation. For example, the synthetic

polymeric coatings of poly(ethylene-co-vinyl acetate),

poly(vinylpyrrolidone), poly(lactic-co-glycolic acid),

poly(ethyleneglycol), poly(vinyl alcohol), etc. or the nat-

ural polymers of gelatin, dextran, chitosan, pullulan, etc. or

the surfactants like sodium oleate, dodecylamine, sodium

carboxymethylcellulose are proved to enhance the bio-

compatibility of iron oxide NPs [2]. The properly seques-

tered iron-based complexes can also be used at the same

time for destroying the free radicals produced by means of

J Mater Sci (2014) 49:6290–6301 6299

123

other chemotherapeutic agents. For example, in some

studies, the ability of Fe to catalyze red-ox reactions was

used for scavenging the ROS generated in the cardiac cells

by the anthracycline antibiotics during cancer treatment

[20]. Also, for most of the biomedical applications, the iron

oxide NPs should possess sufficient hydrophilicity; how-

ever in our case, the SPIONs-1 are not hydrophilic due to

the lipophilic oleic acid groups and the SPIONs-4 are more

suitable due to the existence of both groups of oleic acid

and cysteamine and are very well dispersed in water, in

addition to enhanced antioxidant properties.

In summary, the iron oxide-induced intracellular oxi-

dative stress associated with depletion of GSH and mito-

chondrial function in H9c2 cardiomyocytes were

demonstrated for the first time. The interfacial and surface

properties of iron oxide are thought to play a major role in

the interaction of naked SPIONs (SPIONs-2) with the

biological systems. It is likely in our case that the increased

amounts of intracellular ROS formation by the naked

SPIONs led to further oxidation of critical targets in cells,

which finally leading to the loss of cardiac function. We

believe that these studies may have relevance in the context

of cardiac diseases in general and atherosclerosis in par-

ticular, as ROS-induced oxidative stress also a major

contributor for the initiation and progression of such dis-

eases. In depth understanding of the mechanism of iron

oxide-induced oxidative stress and the pathways involved

in cardiomyocytes are expected to facilitate better under-

standing of these disease processes.

Conclusion

In conclusion, we report the necessity of employing iron

oxide NPs coated with free radical-scavenging ligands in

order to avoid the unwanted toxicity especially in cardiac

cells. In this report, we assessed and compared the dose-

dependent cytotoxicity of ‘‘naked’’ SPIONs with that of

other SPIONs containing ligands of unsaturation and/or

thiol groups, their oxidative stress and the possible toxic

mechanism in cardiomyocytes were discussed. The

observed cell death retained the key features of apoptosis

mechanisms such as the decreased GSH release, DNA

fragmentation, and mitochondrial depletion by the

increased levels of ROS. The presence of oleic acid/cys-

teamine ligands or the supplementation of cardiomyocytes

with Trolox offered significant, protection against the Fe3?

induced cytotoxicity, presumably due to scavenging of

ROS in situ. The results of the study provide experimental

evidence for the importance of having both of unsaturation

and thiol group as protective ligand for reducing the oxi-

dative behavior and increasing the biocompatibility of

SPIONs toward cardiac cells.

Acknowledgements The authors would like to thank Universiti

Putra Malaysia for the support and funding.

References

1. Kumar CSSR, Mohammad F (2011) Magnetic nanomaterials for

hyperthermia-based therapy and controlled drug delivery. Adv

Drug Deliv Rev 63:789–808

2. Gupta AK, Gupta M (2005) Synthesis and surface engineering of

iron oxide nanoparticles for biomedical applications. Biomateri-

als 26:3995–4021

3. Gupta AK, Curtis AS (2004) Surface modified superparamagnetic

nanoparticles for drug delivery: interaction studies with human

fibroblasts in culture. J Mater Sci Mater Med 15:493–496. doi:10.

1023/B:JMSM.0000021126.32934.20

4. Bulte JW, Douglas T, Witwer B, Zhang SC, Strable E, Lewis BK

et al (2001) Magnetodendrimers allow endosomal magnetic

labeling and in vivo tracking of stem cells. Nat Biotechnol

19:1141–1147

5. Kircher MF, Allport JR, Graves EE, Love V, Josephson L,

Lichtman AH, Weissleder R (2003) In vivo high resolution three-

dimensional imaging of antigen-specific cytotoxic T-lymphocyte

trafficking to tumors. Cancer Res 63:6838–6846

6. Pisanic TR II, Blackwell JD, Shubayev VI, Finones RR, Jin S

(2007) Nanotoxicity of iron oxide nanoparticle internalization in

growing neurons. Biomaterials 28:2572–2581

7. Saba N, Mohammad S, Abdin MZ, Ahmed FJ, Maitra AN,

Prashant CK, Dinda AK (2010) Concentration-dependent toxicity

of iron oxide nanoparticles mediated by increased oxidative

stress. Int J Nanomed 5:983–989

8. Jain TK, Reddy MK, Morales MA, Leslie-Pelecky DL, Labha-

setwar V (2008) Biodistribution, clearance, and biocompatibility

of iron oxide magnetic nanoparticles in rats. Mol Pharm

5:316–327

9. Mohammad F, du Plessis L, Arfin T (2013) X-ray analysis of

metal oxide-metal core-shell nanoparticles. In: Shih K (ed) X-ray

diffraction: structure, principles and applications. Nova Science,

New York, pp 161–181

10. Espinosa A, Serrano A, Llavona A, Morena JJ et al (2012) On the

discrimination between magnetite and maghemite by XANES

measurements in fluorescence mode. Meas Sci Technol

23:015602 (6 pp)

11. Voinov MA, Pagan JOS, Morrison E, Smirnova TI, Smirnov AI

(2011) Surface-mediated production of hydroxyl radicals as a

mechanism of iron oxide nanoparticle biotoxicity. J Am Chem

Soc 133:35–41

12. Zhang W, Kalive M, Capco DG, Chen Y (2010) Adsorption of

hematite nanoparticles onto Caco-2 cells and the cellular impair-

ments: effect of particle size. Nanotechnology 21:355103 (9 pp)

13. Lin XL, Zhao SH, Zhang L, Hu HQ, Sun ZW, Yang WS (2012)

Dose-dependent cytotoxicity and oxidative stress induced by

‘‘naked’’ Fe3O4 nanoparticles in human hepatocyte. Chem Res

Chin Univ 28:114–118

14. Mahmoudi M, Simchi A, Imani M, Milani AS, Stroeve P (2009)

An in vitro study of bare and poly(ethylene glycol)-co-fumarate

coated superparamagnetic iron oxide nanoparticles: a new tox-

icity identification procedure. Nanotechnology 20:225104

15. Sun J, Wang S, Zhao D, Hun FH, Weng L, Liu H (2011) Cyto-

toxicity, permeability, and inflammation of metal oxide nano-

particles in human cardiac microvascular endothelial cells. Cell

Biol Toxicol 27:333–342

16. Mohammad F, Arfin T (2013) Cytotoxic effects of polystyrene–

titanium–arsenate composite in cultured H9c2 cardiomyoblasts.

Bull Environ Contam Toxicol 91:689–696

6300 J Mater Sci (2014) 49:6290–6301

123

17. Brook RD, Rajagopalan S, Pope CA III, Brook JR et al (2010)

Particulate matter air pollution and cardiovascular disease: an

update to the scientific statement from the american heart asso-

ciation. Circulation 121:2331–2378

18. Lee C, Barnett J, Reaven PD (1998) Liposomes enriched in oleic

acid are less susceptible to oxidation and have less proinflam-

matory activity when exposed to oxidizing conditions. J Lipid

Res 39:1239–1247

19. Kessler A, Biasibetti M, da Silva Melo DA, Wajner M et al

(2008) Antioxidant effect of cysteamine in brain cortex of young

rats. Neurochem Res 33:737–744

20. Mohammad F, Balaji G, Weber A, Uppu RM, Kumar CSSR

(2010) Influence of gold nanoshell on hyperthermia of super

paramagnetic iron oxide nanoparticles (SPIONs). J Phys Chem C

114:19194–19201

21. Kumar CSSR, Mohammad F (2010) Magnetic gold nanoshells:

stepwise changing of magnetism through stepwise biofunction-

alization. J Phys Chem Lett 1:3141–3146

22. Mohammad F, Raghavamenon AC, Claville MO, Kumar CSSR,

Uppu RM (2014) Targeted hyperthermia-induced cancer cell

death by superparamagnetic iron oxide nanoparticles conjugated

to luteinizing hormone-releasing hormone. Nanotechnol Rev.

doi:10.1515/ntrev-2013-0041

23. Satishkumar K, Rangan V, Gao X, Uppu RM (2007) Methyl vinyl

ketone induces apoptosis in murine GT1-7 hypothalamic neurons

through glutathione depletion and the generation of reactive

oxygen species. Free Radic Res 41:469–477

24. Yoo MK, Kim IY, Kim EM, Jeong HJ, Lee CM, Jeong YY,

Akaike T, Cho CS (2007) Superparamagnetic iron oxide nano-

particles coated with galactose-carrying polymer for hepatocyte

targeting. J Biomed Biotechnol 2007:1–9

25. Shukla N, Liu C, Jones PM, Weller D (2003) FTIR study of

surfactant bonding to FePt nanoparticles. J Magn Magn Mater

266:178–184

26. Kuang R, Kuang X, Pan S, Zheng X, Duan J, Dua Y (2010)

Synthesis of cysteamine-coated CdTe quantum dots for the

detection of bisphenol A. Microchim Acta 169:109–115

27. Tan BJ, Klabunde KJ, Sherwood PMA (1990) X-ray photoelec-

tron spectroscopy studies of solvated metal atom dispersed cat-

alysts. Monometallic iron and bimetallic iron-cobalt particles on

alumina. Chem Mater 2:186–191

28. Roberts JT, Friend CM (1988) Desulfurization of ethylene sulfide

on Mo(100): the roles of ring size and strain in adsorbate reaction

selectivity. Surf Sci 202:405–432

29. Smith RAJ, Hartley RC, Cocheme HM, Murphy MP (2012)

Mitochondrial pharmacology. Trends Pharm Sci 33:341–352

30. Simunek T, Stırba M, Popelova O, Adamcova M, Hrdina R, Gers

V (2009) Anthracycline-induced cardiotoxicity: overview of

studies examining the roles of oxidative stress and free cellular

iron. Pharm Rep 61:154–171

J Mater Sci (2014) 49:6290–6301 6301

123