Depletion of intracellular glutathione and increased lipid peroxidation mediate cytotoxicity of hematite nanoparticles in MRC-5 cells *

1Department of Biochemistry and Molecular Biology, Faculty of Biology, University of Bucharest, Bucharest, Romania; 2Departament of Preclinical Sciences, Faculty of Veterinary Medicine, University of Agricultural Sciences and Veterinary Medicine, Bucharest, Romania; 3Department of Histology, Vasile Goldis Western University of Arad, Arad, Romania; 4Laser Department, National Institute of Laser, Plasma and Radiation Physics, Bucharest-Magurele, Romania


INTRODUCTION
Human lungs are in permanent contact with the environment, being one of the entry points for natural or anthropogenic nanoparticles, which have at least one dimension smaller than one micrometer.Due to their small dimensions, the nanometer-sized particles can penetrate the lung barrier and enter the human circulatory and lymphatic systems, becoming dispersed in different organs, tissues and cells.The effects of nanoparticles on cellular processes mainly depend on their size, chemical composition, crystallinity, aggregation, including a potential contribution to the development of lung and neurodegenerative diseases (Buzea et al., 2007).
Prepared nanoparticles are often used as industrial catalysts (Oberdorster et al., 2005).Their elevated levels at sites surrounding industrial plants and in the gravitation dust sediments of residential agglomeration (Flórián et al., 2003) combined with the dynamics of dust migration across large distances (Buzea et al., 2007) enhance human exposure.
Nanoparticles such as Fe 3 O 4 (Hernández et al., 2009) or Fe 2 O 3 (Sevilla et al., 2009) were used previously for the preparation of efficient magnetic nanocomposites for de-contamination of persistent organic pollutants and arsenic (Giasuddin et al., 2007).All three forms of iron oxide are also used in synthetic pigments in paints, ceramics, and porcelain (Cornell & Schwertmann, 2003).
Unintentional inhalation due to air pollution or application of products containing iron nanoparticles, as well as unavoidable exposure in the workplace, can generate cellular responses in lungs and other organs that may cause significant public health problems (Donaldson et al., 2005;Powell & Kanarek, 2006).
According to Klotz and Sies (2009), nanoparticles can generate ROS by different mechanisms.Consequently, the uptaken metal oxide nanoparticles can lose the me-tallic ions.Iron, being a highly redox-active transition metal, can be involved in the Fenton reaction producing hydroxyl radicals.These nano-sized particles could directly interact with NADPH oxidases from the plasma membrane and/or mitochondria thus disturbing the electron transport chain and generating a superoxide anion.It is also possible that the nanoparticle-induced damage of the mitochondria and endoplasmic reticulum could determine Ca 2+ release into the cytosol, where several Ca 2+ /calmodulin dependent enzymes, such as nitrogen monoxide synthase isoforms, become activated, resulting in production of nitric oxide and peroxynitrite.
Both in vitro and in vivo studies of the health effects of ambient nano-materials have identified the generation of oxidative stress as one of the major mechanisms by which air pollution particles exert adverse biological effects.
Cellular redox homeostasis is carefully maintained by an elaborate antioxidant defense system which includes antioxidant enzymes, proteins, and low-molecular-mass scavengers.Excessive ROS production or weakening of antioxidant defense could lead to oxidative stress.Oxidative stress is a redox disequilibrium in which ROS and lipid radicals attack proteins and nucleic acids, redox-sensitive signaling cascades being activated (Li et al., 2003).The late cellular response is represented by pro-inflammatory effects (Jeng & Swanson, 2006) and cytotoxicity (Nel et al., 2006).
The aim of this study was to evaluate the oxidative stress mechanisms induced by α-Fe 2 O 3 nanoparticles in human lung fibroblast cell line MRC-5.The malondialdehyde (MDA) and reduced glutathione (GSH) levels as well as the activities of antioxidant enzymes, such as superoxide dismutase (SOD), catalase (CAT), glutathione peroxidase (GPX), glutathione reductase (GR) and glutathione-S-transferase (GST) were also analyzed.
Nanoparticles.Nanoparticles of the α form of Fe 2 O 3 (hematite) were obtained by laser reactive ablation in the Laser Department of the National Institute of Laser, Plasma and Radiation Physics (Bucharest-Măgurele, Romania).The particle size assessment was performed with a high resolution transmission electron microscope (Philips CM120 model).The primary nanoparticle size distribution was lognormal in the range 10-120 nm, most of them measuring 40-60 nm.For crystallinity analysis Bruker AXS/D8 ADVANCE X ray diffractometer was used.
Cell lines and treatment.MRC-5 cells were maintained in MEM containing non-essential amino acids, Earle's salts, l-glutamine and 10 % fetal bovine serum.Cells, between passages 11-20, were grown as monolayers in a humidified 5 % CO 2 air atmosphere at 37 o C in 75 cm 2 culture flasks.They were seeded at a density of 2.5 × 10 5 cells/ml.Stock suspensions of α-Fe 2 O 3 nanoparticles were sterilized before use.In each experiment, the stock suspensions were sonicated and freshly diluted to appropriate concentrations in the cell medium.The cells were incubated with hematite nanoparticles at concentrations of 2.5 µg/ml, 6.25 µg/ ml, or 12.5 µg/ml in the culture medium for 24, 48, and 72 h.Controls without treatment were performed for each experiment.
Cell viability assay.The viability of the cells was determined by the tetrazolium salt test (Mosmann, 1983).The medium from each well was removed by aspiration, the cells were washed with 200 μl of phosphate-buffered saline (PBS)/well and then 50 μl (1 mg/ml) of 3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) solution was added to each well.After 2 h of incubation the MTT solution from each well was removed by aspiration.A volume of 50 μl isopropanol was added and the plate was shaken to dissolve the formazan crystals.The absorbance at 595 nm, was then determined using a Tecan multiplate reader (Tecan GENios, Grödic, Germany), for each well.The absorbance for untreated cells was taken to represent the 100 % viability.
Preparation of cell lysate.MRC-5 cells were harvested from culture flasks, after the removal of medium which contained the floating dead cells, washed with PBS and centrifuged at 1 500 × g for 10 min at 4 o C. Cell pellets were re-suspended in 0.5 ml of PBS and then sonicated on ice three times, for 30 s each.The total extract was centrifuged at 3 000 × g for 15 min at 4 o C. Aliquots of the supernatant were used for enzyme assays.
Lipid peroxidation.MDA, as an in vitro marker of lipid peroxidation, was assessed by a method described by Del Rio et al. (2003).To 200 µl of sample with a protein concentration of 2 mg/ml, 700 µl of 0.1 M HCl was added and the mixture was incubated for 20 min at room temperature.Then, 900 µl of 0.025 M thiobarbituric acid (TBA) was added and the mixture was incubated for 65 min at 37 °C.Finally, 400 µL of 10 mM PBS was added.The fluorescence of MDA was recorded using a 520/549 (excitation/emission) filter.A calibration curve with MDA in the range 0.05-5 µM was used to calculate the MDA concentration.The results were expressed as nmoles of MDA/mg protein.
Glutathione assay.The cellular lysate, deproteinized with 5 % sulfosalicylic acid, was analyzed for total glutathione and oxidized glutathione (GSSG) using the Detect X ® Glutathione colorimetric detection kit and following manufacturer's instructions.GSH concentration is obtained by subtracting the GSSG level from the total glutathione.The total and GSH levels were calculated as nmoles/mg protein.
Antioxidant enzymes assay.Total SOD (EC 1.15.1.1)activity was measured according to the spectrophotometric method of Paoletti et al. (1986), based on NADPH oxidation by the superoxide anion generated from molecular oxygen by a purely chemical reaction in the presence of EDTA, manganese (II) chloride and mercaptoethanol.The decrease in absorbance at 340 nm due to NADPH oxidation was followed for 10 min.A control was run with each set of three duplicate samples and the percent inhibition was calculated as (sample rate)/(control rate) × 100.One unit (U) of activity was defined as the amount of enzyme required to inhibit the rate of NADPH oxidation of the control by 50 %.
The CAT (EC 1.11.1.6)activity was assayed by monitoring the disappearance of H 2 O 2 at 240 nm, according to the method of Aebi (1984).The CAT activity was calculated in terms of U/mg protein, where one unit (U) is the amount of enzyme catalyzing the conversion of one μmole of H 2 O 2 in a minute under standard condition of temperature, optimal pH and optimal substrate concentration.
Total GPX (EC 1.11.1.9)was assayed by the Beutler method (1971), using tert-butyl hydroperoxide and NADPH as substrates.The conversion of NADPH to NADP + was followed by recording the changes in absorbance at 340 nm, and the concentration of NADPH was calculated using a molar extinction coefficient of 6.22 × 10 3 M -1 · cm -1 .The activity was expressed as U/mg.One unit of activity was defined as the amount of enzyme that catalyzes the conversion of one μmole of NADPH per minute under standard conditions.
The GR (EC 1.6.4.2) activity was measured according to the method of Goldberg and Spooner (1983), in 0.1 M phosphate buffer, pH 7.4 with 0.66 mM GSSG and 0.1 mM NADPH by recording the decrease of absorbance at 340 nm.The activity of this enzyme was expressed as U/mg, one unit of GR activity been defined as one μmole of NADPH per minute under standard conditions.
The GST (EC 2.5.1.18)activity was assayed spectrophotometrically at 340 nm by measuring the rate of CDNB conjugation with GSH, according to the method of Habig et al. (1974), and calculated as U/mg.One unit of GST activity was defined as the amount of enzyme that catalyzed the transformation of one μmole of CDNB in conjugated product per minute.The extinction coefficient 9.6 × 10 3 M -1 · cm -1 was used for the calculation of CDNB concentration.
All enzymatic activities, calculated as specific activities (units/mg protein) are expressed as percentage of controls.
Protein concentration.The protein concentration (mg/ml) was determined by the method of Bradford (1976), using bovine serum albumine as a standard.
Statistical analysis.The differences between control and α-Fe 2 O 3 nanoparticles-treated cells were compared by Student's t-test using standard statistical packages.All data were expressed as means (n=4) ± standard deviation (S.D.) and differences were considered significant from control at p < 0.05, highly significant at p < 0.01 and extremely significant at p < 0.001.

Cell viability
The hematite nanoparticles decreased the number of viable MRC-5 cells in a time-and dose-dependent manner (Fig. 1).The incubation of cells with 2.5 µg/ml of α-Fe 2 O 3 did not cause a significant loss of viability after 24 h or 48 h, and a decrease by 22 % after 72 h.The decrease in cell viability induced by the exposure to 6.25 µg/ml α-Fe 2 O 3 was between 29 % after 48 h and 31.5 % after 72 h.After the treatment of MRC-5 cells with the highest α-Fe 2 O 3 concentration of 12.5 μg/ml, the viability decreased by 10 % after 24 h, and by 30 % and 44 % after 48 h and 72 h, respectively.
Taking into account that the most pronounced effects were noticed at the highest concentration tested (12.5 µg/ ml), this was used in all subsequent experiments.

lipid peroxidation
The exposure of MRC-5 cells to hematite nanoparticles increased the MDA concentration in a time-dependent manner by 81 %, 189 % and 110 % after 24, 48 and 72 h, respectively (Table 1).As it can be seen in Table 1, the MDA level continuously increased in the first 48 h of exposure and started decreasing after 72 h, but still remained high in comparison with control.

Total and reduced glutathione depletion
The results shown in Table 1 indicate that exposure of MRC-5 cells to Fe 2 O 3 nano-sized particles resulted in a slight diminution in the total glutathione level by 17.7 % after 48 h, followed by a decline by 51.4 % after 72 h.The same tendency was noticed for the GSH con-  tent which decreased by 23.2 % and 51.4 % after 48 h and 72 h of treatment, respectively.

The antioxidant scavenging enzymes
Figures 2a and 2b show the effects of α-Fe 2 O 3 nanoparticles on the activities of antioxidant enzymes, SOD and CAT in MRC-5 cells.The SOD activity increased by 21.2 % and 35.1 % after 48 h and 72 h, and for the same time points, the activity of CAT was up-regulated by 29.5 % and 34.7 %, respectively; however, no significant changes were noticed after 24 h.
After 24 h of exposure to hematite nanoparticles, no significant changes of the specific activities of GPX, GR and GST were visible.
Beginning at 48 h of exposure, an increase in GPX, GR and GST activities was noticed (Figs.3a, 3b, 3c): by 52 % and 37 % for GPX activity, by 21 % and 32 % for GST, and by 13 % and 21 % for GST after 48 h and 72 h of treatment, respectively.

DIsCUssION
This study was initiated in order to evaluate the antioxidative capacity of MRC-5 lung fibroblasts exposed to hematite nanoparticles, taking into account that human occupational and environmental exposures to these is common.In our study, nanoparticles between 40 nm and 60 nm were dominated in the preparation used.Such particles have high coefficient of diffusion leading to efficient deposition probability at airways bifurcation and centers of lung acini (Churg & Vedal, 1996), before they are transported to the deep lung (Maynard & Kuempel, 2005).
Hematite is insoluble at a physiological temperature and pH (Diakonov et al., 1999), but interfacial hydration of anhydrous hematite results in higher solubility (Jang et al., 2007).On the other hand, αFe 2 O 3 nanoparticles could release some ferric ions once they enter the cells, as previously noticed for other types of ultrafine particles (Wang et al., 2008).Subsequently, these ions would be able to interact with hydrogen peroxide physiologically formed by different types of enzyme-catalyzed reactions, generating hydroperoxyl radical (Hurd & Murphy, 2009) according to the reaction: (1) Ferrous ions also participate in the Fenton reaction, generating a hydroxyl radical.
Three possible mechanisms have been put forward to explain superoxide anion generation in MRC-5 cells, af- ter nano-sized hematite enters the cells.First, the uptaken nanoparticles could interact with NADPH oxidase in the plasma membrane, thus generating superoxide in the extracellular space (Dhaunsi et al., 2004), which would enter the cell through an anion channel (Bedard & Krause, 2007).Second, these nanoparticles could interact with a recently identified NADPH oxidase isoform located in the endoplasmatic reticulum (Chen et al., 2008), forming this oxygen radical endogenously.And third, small quantities of Fe 3+ ions released from these nanoparticles in the cytosol, could be reduced to Fe 2+ according to reaction (1) and imported into the mitochondrial matrix through orthologs of Mrs3/Mrs4 transporters (Froschauer et al., 2009) in vertebrates.The mitochondrial iron accumulation appears to be ensued by oxidative damage (Crichton et al., 2002).The labile iron pool present within mitochondria (Kruszewski, 2004) together with the decrease of reduced glutathione level could impair the biogenesis of iron-sulfur clusters within the matrix (Hausmann et al., 2005).These iron-sulfur clusters are present in different mitochondrial proteins including some of the mitochondrial electron transport chain.At the ultrastructural level, Complexes I and III are the main sites of mitochondrial superoxide production.It has been demonstrated that Complex I-dependent superoxide is exclusively released into the matrix whereas Complex III releases superoxide to both sides of the inner mitochondrial membrane (Muller et al., 2004).In Complex I, the most likely sites of electron leakage are the iron sulfur clusters (Barja, 1999).
Several enzymatic defense mechanisms attempt to minimize the production and the action of harmful oxidants, such as SOD, CAT and GPX.
SOD dismutates the free radical superoxide by converting it to hydrogen peroxide, which in turn is decomposed by CAT at high concentration, and by GPX at low concentration.In our experiment, the activities of SOD and CAT increased after 48 h and remained at almost same levels for the next 24 h.Most hydrogen peroxide in the cell is generated through the dismutation of superoxide by SOD action, even though it can be produced by other superoxidegenerating enzymes (Hauptmann et al., 1996;Arimoto et al., 2005).It was proved that lung edema was markedly supressed by pretreatment with polyethylene glycol-modified superoxide dismutase (Sagai et al., 1993).Probably, the generation of superoxide becomes critical in a time-dependent manner, and so the induction of SOD activity is important only after 48 h of exposure to hematite nanoparticles.Despite the good correlation between the SOD and CAT activities in MRC-5 cells, some superoxide and hydrogen peroxide could remain in the cells.
The superoxide anions that escape dismutation can react with hydrogen peroxide according to Haber-Weiss reaction to form hydroxyl radicals or become protonated to hydroperoxyl radicals.Hydroperoxyl and hydroxyl radicals are able to abstract hydrogen atom from a methylene group adjacent to double bonds of polyunsaturated fatty acids forming carbon centered radicals that react with molecular oxygen to form lipid peroxides (Antunes et al., 1996).The significant increase of MDA concentration, by 189 % and 110 % after 48 h and 72 h, respectively, suggests that the antioxidative system adaptation was not sufficient to prevent damage of membrane lipids; in turn, the lipid peroxidation products could affect the structure of DNA bases (Valko et al., 2006), proteins and carbohydrates.Similar results were obtained in human bronchial epithelial BEAS-2B cells exposed to TiO 2 nanoparticles (Gurr et al., 2005) and ZnO ones (Yang et al., 2009) whereas in A549 human lung epithelial cells treated with maghemite nano-sized particles, the change in lipid peroxidation variation was statistically insignificant (Guo et al., 2009).
The enzymes of the glutathione redox cycle, comprising GPX, GR and GST, are another source of protection against oxidative stress.GPX and GST, the enzymes that reduce the levels of peroxides thereby protecting the cell from peroxidative damages, showed similar responses during exposure to α-Fe 2 O 3 nanoparticles.An activation of GPX was observed, with the maximum at 48 h of about 152 % compared to the time-interval control; subsequently, the activity of this enzyme decreased to 137 % of the control at 72 h.The decrease in GPX activity could suggest inactivation by ROS, whose level increased in MRC-5 cells after 72 h.Superoxide anions have been shown to inhibit GPX (Blum & Fridovich, 1985).Accordingly, the antioxidant defense system of MRC-5 cells reacted positively to α-Fe 2 O 3 nanoparticles toxicity by increasing the GST activity.The family of GSTs contains enzymes that are capable of multiple reactions with a multitude of substrates in order to detoxify the endogenous compounds, such as peroxidised lipids, as well as the metabolism of xenobiotics.
The increase of the GPX and GST activities may be correlated with the observed decrease in GSH content of MRC-5 cells.Thus the decrease in the level of GSH after 48 h of α-Fe 2 O 3 nanoparticle treatment may be explained by its use as a substrate of GPX.The increased participation of GSH in conjugation reactions mediated by increased GST activity seems to be a plausible explanation for the reduced GSH content (51.4 % from time interval control) after exposure to α-Fe 2 O 3 nanoparticles for 72 h.Other noticeable changes concerning the enzymatic antioxidant response to the nanoparticles-linked oxidative stress were registered in the specific activity of GR, an enzyme involved in the recycling of GSSG to GSH.Nevertheless, this enzymatic adaptation was insufficient to prevent the significant reduction of the GSH concentration in MRC-5 cells after 72 h of treatment, which suggested an oxidative stress condition.On the other hand, ferric ions at high concentrations, being electrophilic, can deplete intracellular reduced glutathione by interacting with sulfhydryl groups (Brodie et al., 1982) or by inhibition of several cellular activities including the cystine uptake (Bannai, 1984).Also, the long exposure to hematite nanoparticles could induce the loss of cellular GSH by glutathionylation of several proteins (Hare & Stamler, 2005).
In conclusion, taking into account that the GSH concentration decreased and MDA level remained highly increased up to 72 h of treatment, it appears that the MRC-5 antioxidant defense mechanisms do not efficiently counteract the oxidative stress induced by exposure to hematite nanoparticles and possible damaging effects could occur.

acknowledgement
This study was financially supported by the National Research Council of Higher Education, Romania, Grant number 340/2007 and Grant POSDRU 8/1.5/S/61150/2010 co-financed from European Social Fund by the Sectorial Operational Program for Development of Human Resources 2007-2013.The authors are grateful to COST B35/2006 Action for the opportunity to exchange ideas with experts in lipid peroxidation and oxidative stress.