Effect of Fe3O4 Nanoparticles on Skin Tumor Cells and Dermal Fibroblasts

Iron oxide (Fe3O4) nanoparticles have been used in many biomedical approaches. The toxicity of Fe3O4 nanoparticles on mammalian cells was published recently. Though, little is known about the viability of human cells after treatment with Fe3O4 nanoparticles. Herein, we examined the toxicity, production of reactive oxygen species, and invasive capacity after treatment of human dermal fibroblasts (HDF) and cells of the squamous tumor cell line (SCL-1) with Fe3O4 nanoparticles. These nanoparticles had an average size of 65 nm. Fe3O4 nanoparticles induced oxidative stress via generation of reactive oxygen species (ROS) and subsequent initiation of lipid peroxidation. Furthermore, the question was addressed of whether Fe3O4 nanoparticles affect myofibroblast formation, known to be involved in tumor invasion. Herein, Fe3O4 nanoparticles prevent the expression alpha-smooth muscle actin and therefore decrease the number of myofibroblastic cells. Moreover, our data show in vitro that concentrations of Fe3O4 nanoparticles, which are nontoxic for normal cells, partially reveal a ROS-triggered cytotoxic but also a pro-invasive effect on the fraction of squamous cancer cells surviving the treatment with Fe3O4 nanoparticles. The data herein show that the Fe3O4 nanoparticles appear not to be adequate for use in therapeutic approaches against cancer cells, in contrast to recently published data with cerium oxide nanoparticles.


Introduction
Besides an anchorage-independent cell proliferation an important, still treatment-limiting characteristic of malignant tumors is their ability for invasive and metastatic growth [1,2]. During the invasion process, interactions of tumor cells with the neighbouring interstitial stroma, which is composed of fibroblastic, myofibroblastic, endothelial, and inflammatory cells, as well as extracellular matrix components, play a pivotal role [3,4]. Molecular mechanisms of tumor-stroma interactions include the secretion of multiple growth factors and cytokines by tumor cells and activated stromal cells which stimulate tumor invasion, tumor development, and neoangiogenesis [5]. Myofibroblasts are modified fibroblasts that express the biomarker alpha-smooth muscle actin ( SMA) [6]. The myofibroblastic cell type was originally described in the physiological process of wound healing where it contracts the stroma thereby facilitating wound closure [7,8]. Meanwhile, it is well known that myofibroblasts are also involved in pathological processes and diseases like fibrosis and cancer [9]. They contribute to tumor progression and are, therefore, often found at the tumor invasion front [10][11][12]. The interaction between myofibroblasts and cancer cells is dependent on proinvasive growth-promoting factors through paracrine effects [13]. The transition of fibroblasts to myofibroblasts is primarily initiated by transforming growth factor 1 (TGF 1) [14] and mediated through Smad proteindependent as well as Smad independent pathways [15]. Previously, we showed that TGF 1 induces a reactive oxygen species-(ROS-) mediated pathway leading to formation of myofibroblasts via involvement of protein kinase C (PKC) [16] and NAD(P)H oxidase [17].

BioMed Research International
Nanoparticles are generally defined as structures with sizes between 1 and 100 nm that have a very large surface-tovolume ratio leading to different, novel properties compared with bulk particles of the same chemical composition [18,19]. Because of their unique features and the fact that such nanoscale materials are small enough to enter cells and organelles [20,21], nanoparticles are used for many biomedical approaches in vitro and in vivo [22]. One example for iron oxide nanoparticle based cancer therapy would be the magnetic fluid hyperthermia therapy (MFH) [23]. Injected magnetic iron oxide nanoparticles are heated by an alternating magnetic field leading to tumor cell death either through apoptosis or necrosis [24,25]. Although iron oxide nanoparticles are increasingly used for medical purposes, the actual intracellular influence of these structures is not clear till now. As consequence of the increased surface-to-volume ratio, nanoparticles exhibit a potentially higher biological activity compared with larger particles which has been linked to prooxidative but also to antioxidative processes [26][27][28][29][30][31]. The aim of this study was to determine cell toxicity, myofibroblast development, and tumor invasion, after treatment with Fe 3 O 4 nanoparticles.

Materials and Methods
Cell culture media (Dulbecco's modified Eagle's medium (DMEM)) were purchased from Invitrogen (Karlsruhe, Germany) and the defined fetal calf serum (FCS gold) was from PAA Laboratories (Linz, Austria). All chemicals including protease as well as phosphatase inhibitor cocktail 1 and 2 were obtained from Sigma (Taufkirchen, Germany) or Merck Biosciences (Bad Soden, Germany) unless stated otherwise. The protein assay kit (Bio-Rad DC, detergent compatible) was from Bio-Rad Laboratories (München, Germany). Matrigel and polycarbonate cell culture inserts (6.5 mm diameter, 8 m pore size) were delivered from BD Biosciences (Heidelberg, Germany). The Oxyblot Protein Oxidation Detection Kit was from Millipore (Schwalbach, Germany). The enhanced chemiluminescence system (Super-Signal West Pico/Femto Maximum Sensitivity Substrate) was supplied by Pierce (Bonn, Germany). Monoclonal mouse antibodies raised against human -smooth muscle actin and -tubulin were supplied by Sigma. The following secondary antibodies were used: polyclonal horseradish peroxidase-(HRP-) conjugated rabbit anti-mouse IgG antibody (DAKO, Glostrup, Denmark) and goat anti-rabbit immunoglobulin G antibodies were from Dianova (Hamburg, Germany). Recombinant human TGF 1 (rTGF 1) was from R&D Systems (Wiesbaden, Germany).

Cell Culture.
Human dermal fibroblasts (HDF) were established by outgrowth from foreskin biopsies of healthy human donors with an age of 3-6 years. Cells were used in passages 2-12, corresponding to cumulative population doubling levels of 3-27 [32]. Dermal fibroblasts and the squamous carcinoma cell line SCL-1, originally derived from the face of a 74-year-old woman [33] (generously provided by Professor Dr. Norbert Fusenig, DKFZ Heidelberg, Germany), were cultured as described [34]. Myofibroblasts (MF) were generated by treatment of HDF with recombinant TGF 1 (rTGF 1) for 48 h in conditioned medium from HDF (CM HDF ) [16].

Preparation of Conditioned Medium.
Conditioned medium was obtained from human dermal fibroblasts (CM HDF ) and myofibroblasts (CM MF ). For this, seeded 1.5 × 10 6 HDF cells were grown to subconfluence (∼70% confluence) in 175 cm 2 culture flasks. The serum-containing medium was removed, and after washing in phosphatebuffered saline (PBS) the cells were incubated in serum-free DMEM or treated with rTGF 1 (5 ng/mL) in serum-free DMEM for 48 hours. This medium was removed, and after washing in PBS all cells were incubated in 15 mL serumfree DMEM for further 48 hours before collection of the now called conditioned medium of HDF (CM HDF ) and myofibroblasts (CM HDF,TGF 1 = CM MF ). Conditioned media were used fresh or stored at −20 ∘ C for at the most 2 weeks before use [30].

Synthesis and Stabilization of Fe
The synthesis of magnetite nanoparticles on the gram scale was carried out by alkaline precipitation of iron(III) and iron(II) chloride following a method of Cabuil and Massart as described in detail elsewhere [35]. For stabilization, the freshly synthesized nanoparticles were stirred with 420 mL of 2 N nitric acid for 5 min. After washing with distilled water, 90 mL 0.01 N citric acid (CA) was added to the nanoparticles and stirred for 5 min. The particles were magnetically separated from the supernatant and 15 mL of tetramethylammonium hydroxide aqueous solution was added to obtain 3.32 g magnetic nanoparticles   2.6. Particle Transfer to Water/Buffer. The DMSO-based particle dispersion was added dropwise to diethyl ether (Et 2 O). The precipitate was washed five times with Et 2 O/acetone (1 : 1) and was redispersed in distilled water or buffer to obtain an aqueous magnetic fluid [36].

Cell Viability.
The cytotoxic effect of Fe 3 O 4 nanoparticles was measured by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide) assay [37]. The activity of mitochondrial dehydrogenases, as indicator of cellular viability, results in formation of a purple formazan dye. Briefly, MTT solution (0.5 mg/mL) was added to the cells treated with different concentrations of Fe 3 O 4 nanoparticles after washing with PBS. Cells were incubated for an additional 20 min. The medium was removed and the cells were lysed in dimethyl sulfoxide. The formazan formation was measured at 570 nm. The results were presented as percentage of untreated control which was set at 100%.

Cellular Uptake of Nanoparticles.
Human dermal fibroblasts (HDF) and squamous cancer cells (SCL-1) in serum-free Dulbecco's Modified Eagle Medium (DMEM) were treated with 350 M Fe 3 O 4 nanoparticles for 24 h. Thereafter, cells were harvested and washed with phosphate-buffered saline (PBS) to remove excess media. As the nanoparticles are not detectable by phase contrast microscopy, transmission electron microscopy was used to determine the cellular uptake of Fe 3 O 4 nanoparticles. For electron microscopy, pelleted samples of Fe 3 O 4 nanoparticles-treated cells were fixed for 2 h in 4% paraformaldehyde and 2.5% glutaraldehyde (Serva, Heidelberg, Germany) in 0.1 M phosphate buffer at pH 7.4 at room temperature. Next, the pellets were thoroughly washed with four changes of PBS, followed by a postfixation for 60 min in 1% osmium tetroxide (Serva) in PBS. The specimens were dehydrated in a graded series of acetone and embedded in Spurr's medium (Serva) at 70 ∘ C for 24 h.
Ultrathin sections were cut from the embedded tissue with a Reichert Ultracut (Vienna, Austria) using a diamond knife. The sections were collected on coated copper grids and subsequently stained with uranyl acetate and lead citrate according to earlier published data [38]. The grids were analyzed using a Hitachi H 600 electron microscope (Düsseldorf, Germany). Documentation was carried out by using an optical system and the Digital Micrograph software (Gatan, Munich, Germany). For light microscopical controls semithin sections were cut and stained with 1% Toluidine blue and 1% Borax [30].

SDS-PAGE and Western
Blotting. SDS-PAGE was performed according to the standard protocols published elsewhere [39] with minor modifications. Briefly, cells were lysed after incubation with Fe 3 O 4 nanoparticles in 1% SDS with 1 : 1000 protease inhibitor cocktail (Sigma; Taufkirchen, Germany). After sonication, the protein concentration was determined by using a modified Lowry method (Bio-Rad DC). 4x SDS-PAGE sample buffer (1.5 M Tris-HCl pH 6.8, 6 mL 20% SDS, 30 mL glycerol, 15 mL -mercaptoethanol, and 1.8 mg bromophenol blue) was added, and after heating, the samples (20-30 g total protein/lane) were applied to 8% (w/v) SDSpolyacrylamide gels. After electroblotting, immunodetection was carried out (1 : 1000 dilution) of primary antibodies (rabbit monoclonal anti-HIF1 and mouse monoclonal anti--tubulin), 1 : 20000 dilution of anti-mouse/rabbit antibody conjugated to HRP). Antigen-antibody complexes were visualized by an enhanced chemiluminescence system. -tubulin was used as internal control for equal loading.

Invasion Assay.
Cell culture inserts (transwells) were overlaid with 125 g/mL growth factor-reduced matrigel and placed in a 24-well plate. Tumor cells (5 × 10 4 cells/insert) either untreated or pretreated with Fe 3 O 4 nanoparticles were seeded on top of the matrigel in serum-free DMEM. Conditioned medium of human dermal fibroblasts (CM HDF ) or of myofibroblasts (CM MF ) (see above) was used as chemoattractant in the lower chamber. After 30 h at 37 ∘ C, the melanoma cells were rubbed off the upper side of the filter using cotton swabs, and the tumor cells, which invaded to the lower side of the insert, were stained with Coomassie Blue solution (0.05% Coomassie Blue, 20% MeOH, and 7.5% acetic acid). The number of invaded cells was estimated by counting 25 random microscopic fields/insert [16,30].

Determination of Oxidized (Carbonylated) Proteins: Oxyblot Analysis.
Tumor cells were grown to subconfluence on tissue culture dishes. After removal of serum-containing medium, cells were cultured in serum-free medium and either untreated or pretreated for different times with 350 M Fe 3 O 4 nanoparticles. As positive control, the cells were treated with 250 M H 2 O 2 . Thereafter, cells were lysed and carbonyl groups of oxidized proteins were detected with the OxyBlot Protein Oxidation Detection Kit, following the manufacturer's protocol. Briefly, the protein concentration was determined by using a modified Lowry method (Bio-Rad DC). The protein amounts of the samples were aligned. Five g of this cell lysate was incubated with 2,4dinitrophenyl (DNP) hydrazine to form the DNP hydrazone derivatives. Labeled proteins were separated by SDS-PAGE and immunostained using rabbit anti-DNP antiserum (1 : 500) and goat anti-rabbit IgG conjugated to horseradish peroxidase (1 : 2000). Blots were developed by enhanced chemiluminescence.

Enzyme-Linked Immunosorbent Assay (ELISA).
By means of the ELISA method, the content of 8-iso prostaglandin F2 (8-PGF2a isopropyl, 8-isoprostane) was investigated in cell culture supernatants from SCL-1 cells. The assay was performed using the Acetylcholinesterase Competitive Enzyme Immunoassay kit (Cayman Chemical, Michigan, USA) according to the manufacturer's instructions. This assay is based on the competition between 8-isoprostane and an 8-isoprostane-acetylcholinesterase (AChE) conjugate (8isoprostane tracer) for a limited number of 8-isoprostanespecific rabbit antiserum binding sites. Because the concentration of the 8-isoprostane tracer is held constant while the concentration of 8-isoprostane varies, the amount of 8isoprostane tracer that is able to bind to the rabbit antiserum will be inversely proportional to the concentration of 8isoprostane in the well. This rabbit antiserum-8-isoprostane (either free or tracer) complex binds to the rabbit IgG mouse monoclonal antibody that has been previously attached to the well. The plate is washed to remove any unbound reagents and then Ellman's Reagent (which contains the substrate to AChE) is added to the well. The product of this enzymatic reaction has a distinct yellow color and absorbs strongly at 412 nm. The intensity of this color, determined spectrophotometrically, is proportional to the amount of 8-isoprostane.

Determination of Malondialdehyde (MDA).
MDA is a marker of lipid peroxidation and was determined by HPLC [40] after derivatization with 2-thiobarbituric acid [41]. The HPLC system consisted of a Merck Hitachi L-7100 pump connected with a Merck fluorescence detector (Merck Hitachi; FL Detector L-7480) and a data registration system. Analyses were performed isocratically with a mobile phase composed of 60% phosphate buffer (NaH 2 PO 4 /Na 2 HPO 4 buffer; 50 mmol/L; pH 6.5) and 40% methanol (v/v) at a flow rate of 1 mL/min and a reversed-phase column (LiChrospher 100 RP18, 5 m; Merck) protected by a guard column (4.6 × 4.6 mm) of the same stationary phase. Excitation wavelength was 513 nm and emission wavelength 550 nm. MDA levels were calculated by external calibration with 1,1,3,3-tetramethoxypropane, which releases a stoichiometric amount of MDA in an acidic solution. The MDA amount was normalized to the protein content [42].
2.14. Statistical Analysis. Means were calculated from at least three independent experiments, and error bars represent standard error of the mean (s.e.m.). Analysis of statistical significance was done by Student's -test or ANOVA with * < 0.05, ** < 0.01, and *** < 0.001 as levels of significance.

Results
Herein, the effect of Fe 3 O 4 nanoparticles in tumor-stroma interaction was studied. We investigated the influence of Fe 3 O 4 nanoparticles in cultured human dermal fibroblasts and on human squamous carcinoma cells (SCL-1). Fe 3 O 4 nanoparticles are nontoxic on stromal cells (e.g., fibroblasts) but the cell viability in tumor cells was significantly lowered.
Oxidative stress parameters, for example, total reactive oxygen species, carbonylated proteins, and formation of malondialdehyde, were investigated.

Cell Viability.
The potential toxicity of Fe 3 O 4 nanoparticles on human dermal fibroblasts (HDF) was tested. The fibroblasts were incubated with 65 nm-sized polymer-coated Fe 3 O 4 nanoparticles for 72 h. MTT assays were used to analyze the viability of the cells. Cell viability was evidently not altered after 72 h for these cells (Figure 1(a)).   (Figure 2(a)). These data correlated with the SMA protein amount (Figure 2(b)). The SMA protein level was lowered by 50% after preincubation with 350 M Fe 3 O 4 nanoparticles (Figure 2(b)).  [16]. Antioxidants inhibit the expression of alpha-smooth muscle actin resulting in prevention of myofibroblast formation [16]. Herein, we checked if Fe 3 O 4 nanoparticles modulate the invasive capacity of tumor cells.  (Figure 3(b)). The invasion of the squamous tumor cells was 30-50% lowered by

Discussion
An increasing number of different types of nanoparticles are used for applications in the biomedical field, from use as contrast agent to potential carriers for drug delivery. The possible toxic properties of nanoparticles on human health are controversially discussed [18,47] and further studies are needed to understand and evaluate their function and more specific their toxicity.
Many publications have evaluated the biocompatibility of super paramagnetic iron oxide nanoparticles in different cell types, that is, macrophages [48], endothelial cells [49], and fibroblasts [50,51]. Experiments on the influence of iron oxide nanoparticles in human dermal fibroblasts, representing stromal cells, and squamous cancer cells are limited.
In comparison with dextran-coated cerium oxide nanoparticles [30], the question was addressed of whether Fe 3 O 4 nanoparticles have the same bifunctional character like, namely, an antioxidant effect on human dermal fibroblasts and a prooxidative effect in tumor cells.
In this study, we showed that Fe 3 O 4 nanoparticles prevented TGF 1-triggered and ROS-initiated formation of myofibroblasts. The treatment of fibroblasts with the iron nanoparticles is speculated to inhibit the secretion of proinvasive soluble factors and resulted in a significantly lowered invasion of SCL-1 cells. This data correlates with the results obtained with classical antioxidants [16] or redox-active cerium oxide nanoparticles [30,47]. However, the direct treatment of the tumor cells with the iron nanoparticles increased the invasiveness of a fraction of that cells.
ROS production by Fe 3 O 4 nanoparticles causes the cytotoxic effect in several cell types [52]. Fe 3 O 4 is unstable and can easily be oxidized to yield -Fe 2 O 3 + Fe 2+ [53][54][55]. The free Fe 2+ ions are able to produce highly reactive hydroxyl radicals (HO ⋅ , Fenton reaction) by reaction with H 2 O 2 or O 2 and Fe 3+ ions [56] that can modify proteins, lipids, and DNA [52]. Earlier studies described that Fe 3 O 4 caused an increase in oxidative stress and lipid peroxidation in tumor cells, for example, skin epithelial A431 and lung epithelial A549 [57].

Conclusion
Fe 3 O 4 particles with a mean diameter of 65 nm generated reactive oxygen species and, as a consequence, being toxic as well as proinvasive on the fraction of squamous cancer cells surviving the treatment with Fe 3 O 4 nanoparticles whereas the same concentration does not alter the viability of human dermal fibroblasts which were used as model for stromal cells in skin cancer. These data are in contrast to the recently described effect of cerium oxide nanoparticles on tumor cells [30] indicating that the Fe 3 O 4 nanoparticles appear not to be adequate for use in therapeutic approaches against cancer cells.