Exposure of Engineered Nanoparticles to Human Lung Epithelial

Apr 25, 2007 - The chemical and catalytic activity of nanoparticles has strongly contributed to the current tremendous interest in engineered nanomate...
1 downloads 4 Views 291KB Size
Download PDF
Environ. Sci. Technol. 2007, 41, 4158-4163

Exposure of Engineered Nanoparticles to Human Lung Epithelial Cells: Influence of Chemical Composition and Catalytic Activity on Oxidative Stress LUDWIG K. LIMBACH,† PETER WICK,‡ PIUS MANSER,‡ ROBERT N. GRASS,† ARIE BRUININK,‡ AND W E N D E L I N J . S T A R K * ,† Department of Chemistry and Applied Biosciences, Institute for Chemical and Bioengineering, ETH Zurich, CH-8093 Zurich, Switzerland, and Laboratory for Material-Biology Interactions, Empa, Swiss Federal Laboratories for Materials Testing and Research, Lerchenfeldstrasse 5, CH-9014 St. Gallen, Switzerland

The chemical and catalytic activity of nanoparticles has strongly contributed to the current tremendous interest in engineered nanomaterials and often serves as a guiding principle for the design of functional materials. Since it has most recently become evident that such active materials can enter into cells or organisms, the present study investigates the level of intracellular oxidations after exposure to iron-, cobalt-, manganese-, and titania-containing silica nanoparticles and the corresponding pure oxides in vitro. The resulting oxidative stress was quantitatively measured as the release of reactive oxygen species (ROS). The use of thoroughly characterized nanoparticles of the same morphology, comparable size, shape, and degree of agglomeration allowed separation of physical (rate of particle uptake, agglomeration, sedimentation) and chemical effects (oxidations). Three sets of control experiments elucidated the role of nanoparticles as carriers for heavy metal uptake and excluded a potential interference of the biological assay with the nanomaterial. The present results indicate that the particles could efficiently enter the cells by a Trojan-horse type mechanism which provoked an up to eight times higher oxidative stress in the case of cobalt or manganese if compared to reference cultures exposed to aqueous solutions of the same metals. A systematic investigation on iron-containing nanoparticles as used in industrial fine chemical synthesis demonstrated that the presence of catalytic activity could strongly alter the damaging action of a nanomaterial. This indicates that a proactive development of nanomaterials and their risk assessment should consider chemical and catalytic properties of nanomaterials beyond a mere focus on physical properties such as size, shape, and degree of agglomeration. * Corresponding author phone: +41 44 632 09 80; fax: +41 44 633 10 83; e-mail: [email protected]. † ETH Zurich. ‡ Empa, Swiss Federal Laboratories for Materials Testing and Research. 4158

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 11, 2007

Introduction The astonishing physical and chemical properties (1) of engineered nanomaterials have provoked an exponential growth of nanoproducts on the free market. Growing concerns over the impact of such materials on human health and the environment have initiated the first in-depth studies on the effect of nanomaterial exposure to biological systems (2-4) and showed a high mobility of such materials in organisms or cells (5-7). Engineered nanoparticles have resulted in strongly improved chemical (8) and environmental catalysts (9-12). While such improved activity is most advantageous for an industrial application, the same catalytic activity may contribute to a partially new and most aggressive form of long-term toxicity: Following a rapid uptake into cells (5-7), such chemically active particles may interfere with cellular metabolism by catalyzing specific reactions within the cytosol. While a normal toxic agent exerts a doseand mass-related effect on the corresponding organism, a catalytically active material may repeat its chemical interaction with the host over and over again. Catalytic activity therefore greatly enhances the potency of such toxins, especially if the material has a long-term persistence within the host organism. The use of readily measurable physical and chemicals properties for early risk assessments of industrially attractive engineered nanomaterials (13) is therefore particularly interesting for chemically active materials (14). The present work investigates how catalytic activity of a nanoparticulate material may induce oxidations inside living cells. Unspecific oxidative damage is one of the largest concerns in nanoparticle mediated toxicity (15-17) as shown for titania (16, 18-23) or fullerene nanocrystals (24). In order to provide the most relevant sample sets, the present study applies the same materials as currently broadly used for industrial oxidation processes (8, 25, 26) accounting for the production of several tens of megatons of fine chemicals annually (14). More specifically, iron-, manganese-, cobalt-, or titanium-doped silica were tested for the promotion of reactive oxygen species (ROS) within human lung epithelial cells in vitro at nontoxic concentrations. This intracellular activity could then be correlated to catalytic activity or chemical composition and confirmed the presence of catalytic oxidative effects inside living cells.

Experimental Section General Experimental Design. The study of chemical effects arising from nanoparticle uptake in vitro requires nanomaterials with similar physical properties (27) affording a detailed particle characterization. The formation of ROS following exposure to nanoparticles was studied without (cell-free) or in the presence of cultured human lung epithelial cells (in vitro). Soluble salts of the corresponding heavy metals were used as reference materials. To investigate whether oxygen or hydrogen peroxide served as predominant precursors to ROS formation through catalytical activation (28, 29), the cell-free reference tests were conducted either under oxygen (air) or in the presence of additional hydrogen peroxide. A demonstration of the uptake of oxide nanoparticles by electron microscopy had already been demonstrated in earlier studies (5, 48). Materials. The present study used two sets of materials: (1) silica doped with 0.5 and 1.6 wt % of iron, cobalt, manganese, and titanium and the corresponding pure oxides; and (2) a series of eight iron-containing silica nanoparticles (0-10 wt % Fe/SiO2) to systematically investigate the role of 10.1021/es062629t CCC: $37.00

 2007 American Chemical Society Published on Web 04/25/2007

FIGURE 1. Representative transmission electron micrograph of nanoparticles used in the present study (left) and typical log-normal particle size distributions as measured by X-ray disk centrifugation (right). Phase composition, element sensitive mapping, and X-ray diffraction pattern are given in the Supporting Information. catalytic effects. Prior to exposure, all materials were thoroughly characterized for primary particles size, hydrodynamic agglomerate size distribution, morphology, and chemical composition. Details on the analytical methods and particle production are given in the Supporting Information (SI). Selected materials were further investigated by diffuse reflectance FTIR spectroscopy to confirm the presence of catalytically active sites. In vitro assays. For all exposure experiments, the in vitro assays were either done on diluted nanoparticle suspensions at a concentration of 30 ppm (30 microgram per milliliter of cell culture medium) following the procedure by Brunner et al. (13) or using corresponding amounts of soluble transition metal salts in cell culture medium (see SI Table ST3 for concentrations). Since the lung has been identified as one of the most exposed tissues for nanoparticles uptake (6), lung epithelial cells (A549) were chosen for the present experiments. This cell line has already been used in earlier cytotoxicological studies on hematite and silica (30) and carbon black (31). The oxidative stress was measured using an established fluorescence assay (32-37) (SI, Figure S6). The assay is based on the uptake of an inactivated form (ester) of a reduced dye (2′,7′-dichlorodihydrofluorescein diacetate, HDCF-DA). Once inside the cytosol, intracellular esterases cleave the inactive delivery form to the free, reduced dye (2′,7′-dichlorodihydrofluorescein, HDCF) which is sensitive to radical-mediated oxidation forming a highly fluorescent dye (2′,7′-Dichlorofluorescein, DCF) (32, 38) that can be measured spectroscopically.

Results Nanoparticle Properties and Sample Consistency. Preparation of twelve different oxide nanoparticle samples containing various amounts of transition metals and silica using exactly the same preparation conditions resulted in spherical nanoparticles (Figure 1) of 20-75 nm size. All materials were of similar morphology (see Supporting Information) as tested by electron microscopy and consistently showed a log-normal particle size distribution (Figure 1, right) which stayed in a full agreement with both experimental (39) and theoretical understanding of the production process (40). The phase composition of all silica-based oxides was fully amorphous (SI Figure S3). In contrast to this, pure transition metal oxides exhibited partial crystallinity as previously observed for other oxides (26, 41, 42). The chemical composition was measured after synthesis by quantitative element analysis and showed a homogeneous distribution of elements within the particles (SI Table S1). To prove that the transition-metal-doped silica contained the catalytically active species as previously described by Stark et al. (8) and Hutter et al. (26), diffuse reflectance FTIR spectroscopy was used to confirm the presence of tetrahedrally substituted

FIGURE 2. ROS concentrations in human lung epithelial cells after 4 h nanoparticle exposure (full columns) relative to reference cultures without particle exposition. Empty columns depict cultures only exposed to the corresponding amount of metal salts as aqueous solution. Note the different scale bars. (t ) no water soluble titanium salts available for reference). transition metal species (SI Figure S5) as the so-called active sites in these catalysts. ROS Formation in Lung Epithelial Cells. In a first set of experiments four transition metal oxides were investigated in the form of nanoparticles for their stimulation of ROS formation during exposure to cultures of human lung epithelia cells in vitro. Exposure of 30 ppm (30 µg particles/mL media) iron, titanium, manganese, or cobalt oxide of 20-75 nm size onto preincubated cultures resulted in a promoted increase in reactive oxidizing species after 4 h (Figure 2). To distinguish between effects arising from the presence of transition metals ions and nanoparticle-derived effects, the ROS formation was compared to the corresponding ion concentration (supplied as aqueous solutions) of these transition metals. In the case of pure silica (silica, 30 ppm), no statistically significant increase in ROS formation could be observed if compared to untreated cultures or blind probes using pure buffer. If the silica nanoparticles were doped with additional transition metals (Ti, Fe, Co, Mn) this behavior changed drastically. The addition of as few as 1.6 wt % Mn in silica afforded a ROS increase of 2500% (Figure 2, full bars). Since the total Mn dose (30 µg particles * 1.6 wt % Mn ) 480 ng Mn/mL) is below 1 µg per mL, the high activity of such Mn species is most evident. The addition of Ti and Fe had a much smaller effect but also affected the amount of reactive oxidizing species. Exposure of A549 cells to 30 ppm 1.6 wt % iron/silica (total Fe dose: 480 ng per mL) resulted in an about twice as high VOL. 41, NO. 11, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4159

FIGURE 3. ROS production of 30 ppm iron/silica nanoparticles exposed to A549 cells (left) relative to saline controls exactly follows the activity pattern of FeOx/SiO2 known for heterogeneous catalysts. Exposure to nanoparticles (full columns) results in different ROS levels than exposure to iron ions (empty bars) at the same concentration. A cell-free control experiment in medium (right) showed no statistically relevant dependence of ROS levels on the nanoparticle composition (full columns). Exposure to iron salts at the same iron dose did provoke some ROS above 5 ppm (Fe)aq. Much less ROS was formed with A549 cells exposed to 30 ppm aqueous iron (left, empty bar, about 320%) if compared to the cell-free control (right, empty bar, about 1980%) corroborating the barrier function of cell membranes for ions. ROS level as cultures exposed to pure silica. Use of the corresponding amount of iron in an ionic form (supplied as FeCl3 in water) did not increase the ROS level (Figure 2, open columns). Since iron is considered to be one of the most important transition metals, we investigated its role for an extended set of concentrations (0-10 wt % iron in silica). Figure 3 shows the influence of the chemical composition of iron oxide silica mixtures and the resulting oxidative stress of cultures exposed to 30 ppm of the corresponding particles. A maximal ROS production was found for nanoparticles with 1-3 wt % iron/silica. The effect of transition metal oxide impurities or dopants in silica was further compared to pure transition metal oxide nanoparticles, e.g., titanium, iron, cobalt, and manganese oxide (Figure 2). While pure titania lead to a significant increase in ROS, the exposure of 30 ppm iron oxide did not result in higher ROS level. Very high levels of reactive oxidizing species were observed for cultures exposed to pure cobalt or manganese oxide nanoparticles. There, a 30-50 times higher ROS level after 4 h confirmed the high oxidation capacity of these metals. Exposure of the cell cultures to the same amount of cobalt or manganese in the form of aqueous solutions (salts) provoked a much smaller ROS level (Figure 2, compare empty (salt) and full (nanoparticles) columns). This observation again confirmed a most significant activity of the nanoparticles compared to a mere presence of the metal in its ionic form. Catalytic Activity under Cell-Free Conditions. Since the activation of oxygen and the resulting promotion of ROS formation in cells may proceed by a chemical (cells are in principle not needed for particle-derived ROS) and a biochemical pathway (interaction or promotion of cell internal ROS production or degradation; see SI Figure S1), we measured the ROS formation of cell culture medium without cells but containing exactly the same nanoparticles and using exactly the same measuring method. Since the ester form of the dye (HCDF-DA) must be hydrolyzed prior to the reaction, a fair comparison involved the use of the reactive precursor of the dye (HDCF). Figure 4 summarizes the ROS formation of the same set of nanoparticles in absence of cells and compares them to ROS levels arising from the addition of the corresponding metal ions (salts) in an aqueous form. A high similarity to the ROS levels in cultures (Figure 2) can be observed and the relative ROS levels were predominantly conserved. These observations confirmed a strong role of the chemical pathway in nanoparticle-mediated ROS stimulation (SI Figure S1) inside the cells and proved this statement by reproducing an environment close to the nanoparticle-containing cytosol (cell-free experiments, Fig4160

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 11, 2007

FIGURE 4. Comparison of ROS production in cell-free culture medium for heavy metals at 30 µg/mL medium. Exposure to particles (full bars) stimulates few ROS for iron oxide and titania. Dissolved iron ions (empty bars), however, promote a 20 times higher ROS production than exposure to the same amount of iron in the form of Fe2O3 nanoparticles. This proves that the Fe2O3 particles did not significantly dissolve to iron ions within the duration of the exposure. In the case of cobalt and manganese, similar ROS levels were found for both nanoparticles (full bars) and dissolved ions (empty bars). This clearly shows that the Co3O4 and Mn3O4 nanoparticles can dissolve in the cell culture medium within the duration of the experiment. The dissolved nanoparticles then deliver ionic manganese or cobalt which gives rise to the same ROS levels as medium directly prepared with the corresponding heavy metal ions. (t ) no titanium salt reference was investigated due to its insolubility). ure 4). Some most important differences, however, were visible. Iron-doped silica nanoparticles were more active in the presence of cells than under cell-free reference conditions (medium-only). Iron salts at 30 ppm (empty columns) were much more active for ROS formation without any living cells present. The cells apparently inactivated part of the effects arising from iron. In the case of manganese salts, this suppression reduced the ROS levels by a factor of 7 (compare ROS levels with cells (Figure 2) and without (Figure 4)). In agreement with its physical properties as a semiconductor with photoactivity, titania nanoparticles did not provoke an appreciable ROS level since all experiments were conducted in the dark.

Discussion ROS Formation in the Presence of Ti, Mn, Fe, and Co. The presence of minute levels of transition metals (0.5 or

1.6 wt %, 150 or 480 ng per mL cell culture medium, 0.15 or 0.48 ppm, respectively) in relatively inert silica nanoparticles strongly changed the reaction of cultured cells exposed to these materials if compared to cultures exposed to pure silica or reference cultures without particles. In the case of silica with 1.6 wt % manganese oxide, the presence of the transition metal afforded an up to 25 times increased ROS level (Figure 2). The high activity of these transition metals became even more evident for pure oxides (TiO2, Fe2O3, Mn3O4, Co3O4) where ROS levels were up to 50 times higher than in cultures exposed to silica or pure saline controls. The presence of a non-mass-linear effect could be clearly demonstrated since the ROS levels in all experiments did not linearly follow the amount of introduced transition metals (Figure 2, e.g., manganese group). More specifically, if the ROS production was primarily related to the presence of distinct amounts of a specific metal inside the cells, experiments with pure transition metal oxides (TiO2, Fe2O3, Mn3O4, Co3O4) would be expected to result in a 60-times higher ROS level than cultures exposed to silica with 1.6 wt % of the corresponding dopants. This non compositional effect indicates that, beyond composition, other particle properties are determining the ROS levels. A comparison of experiments with salts (Figure 2, empty columns) or nanoparticles (Figure 2, solid columns) as heavy metal delivery form showed that the latter generated significantly more ROS in vitro. These two results underline the importance of the delivery form (nanoparticle versus aqueous solutions) and composition (pure silica, transition metals in silica, pure transition metals). Furthermore, the investigated transition metals showed individual differences in their response. These individual reactions could be correlated to their chemical behavior and catalytical observations from previous studies within industrial catalyst developments. In contrast to the heavy metals (Fe, Mn, Co), titania was much less active. This relatively low activity agrees well with an earlier in vivo study by Dick et al. (43) where the broncho-alveolar lavage fluids of rats showed no significantly increased free radical damage following exposure to TiO2. Similarly, in vitro experiments with human skin fibroblasts exclusively found increased ROS formation only following exposure to additional UVA light (44). A most recent study by Long et al. found that titania generated ROS in brain microglia (BV2) at similar concentrations (10-20 ppm) after 1 h (18). These partially contrasting observations may be attributed to the most different behavior of the two cell types (brain microglia vs lung epithelial cells). Evidence for Catalytic Effects with Iron-Containing Nanoparticles. Iron-containing silica and pure iron oxide (Figure 2, second group) nanoparticles strongly illustrated the absence of a purely stoichiometric (mass correlated) adverse effect on the cultures. A more detailed view on the role of nanoparticle composition (Figure 3) clearly demonstrated a maximum ROS induction for particles containing 1-5 wt % Fe in silica. Pure iron oxide (20-100 times more iron than in doped silica nanoparticles) even showed a statistical significantly weaker ROS induction than doped silicas and could not be distinguished from untreated cultures. This unexpected, non-mass correlating effect can be correlated to the increased catalytic activity of transition metal sites consisting of single or few transition metal ions incorporated as clusters in an inert matrix. The presence of such sites was repeatedly demonstrated in numerous earlier investigations on heterogeneous catalysis for fine chemical synthesis (10, 26, 45-47) and has been confirmed in the present investigation by Fourier-transformed IR spectroscopy (see Supporting Information, Figure S5). The importance of such catalytically active sites is further supported by the appearance of a distinct composition-dependent ROS induction (see maximum ROS levels for intermediate Fe concentration in the nanoparticles, Figure 4) which follows

a behavior well described in such materials. If the iron content in silica is increased above a few percent, the surface is saturated and no more additional active sites can be formed. The surplus iron even deactivates these sites by forming individual spots of crystalline iron oxide. This phase segregation was demonstrated by element sensitive mapping in transition electron microscopy images (48) of 10 wt % iron/ silica (see Supporting Information, Figure S2) and stays in line with earlier studies (8, 26). Cobalt and Manganese. The chemistry of cobalt is very different from that of iron which is reflected, among other effects, in its higher catalytic activity for the decomposition of hydrogen peroxide (see SI Figure S7). Cobalt ions are wellknown to generate ROS (49), and oxidize proteins (50), and have been reported to generate free radical damage in broncho-alveolar lavage fluids of rats (43). Compared to iron, this increased activity resulted in an over ten times higher ROS level in the exposed cultures. Moreover, cells exposed to pure Co3O4 displayed a much higher ROS level than exposure to 1.6 wt % cobalt/silica or cobalt salts (Figure 2, third group, empty columns). These observations stay in line with the chemical behavior of cobalt and may be attributed to an increased uptake of cobalt into cells if the metal was present in the form of Co3O4 nanoparticles rather than Co ions. Cell membranes offer an excellent barrier for most ions while nanoparticle uptake was most recently investigated quantitatively using similar in vitro assays (5). The reported rate of uptake agrees with the strongly elevated ROS levels as found in Figure 2. This interpretation is further supported by the over 20-fold increase of ROS levels of pure Co3O4 compared to cultures exposed to the same amount of 1.6 wt % cobalt/silica. Comparing cobalt and iron indicates a completely contrasting mechanism in which cobalt may react predominantly in an ionic form once inside the cells. The Co3O4 nanoparticles therefore serve as Trojan-horse type carriers enabling the transport of high levels of cobalt into the cells. While the ionic cobalt (delivery as an aqueous solution, Figure 2, empty columns) can hardly pass the intact cell membranes, the cobalt oxide nanoparticles rapidly enter the cells (5) and successively dissolve, releasing damaging cobalt within the cell. This mechanism would require a faster dissolution of cobalt ions from the nanoparticle matrix than in the case of iron. Earlier observation on the leaching stability of transition-metal-doped silica prepared by an aerosol method similar to that used here also found a higher transition metal release for cobalt (8) confirming the proposed mechanism. Nanoparticles of Mn3O4 generated the highest ROS levels if compared to the other metal oxides which agrees well with the fact that bulk Mn3O4 was found highly reactive in cultured neuronal phenotype PC-12 cells (20) and, again, in catalytic processes (51). This behavior was reflected in a pronounced activity for 0.5 wt % or 1.6 wt % manganese/silica while the even higher activity of pure manganese oxide may suffer from reactant (dye) limitation. As in the case of the other heavy metals, exposure to manganese in a particulate form had a much stronger effect than exposure to dissolved manganese ions. Control Experiments. Previous in vitro studies on carbon nanotubes have shown that nanomaterials may interfere in unexpected ways with conventional biochemical assays (52). We therefore investigated the interactions of the here used dye in its active form (cleaved ester form, HDCF) with all investigated nanoparticles dispersions (Figure 4). Particles or ions together with the de-esterified dye HDCF could simulate the intracellular interactions along a chemical pathway as shown in SI Figure S1. These control experiments (Figure 4) therefore describe the chemically arising ROS production at 30 ppm nanoparticle or the corresponding VOL. 41, NO. 11, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4161

metal ion concentrations. Such concentrations are reasonable in view of earlier investigations (5). A time-resolved measurement of the concentration of transition metal ions within the cytosol of living cells would obviously be most interesting, but suffers from technical limitations. The interactions of the metals ions or nanoparticles with the cellular metabolism (see SI Figure S1) may either be different (e.g., cobalt) or very similar to the purely chemical environment (control experiments, e.g., Mn, Fe). This comparison allows therefore to distinguish between mainly chemical effects (ROS generation by the heavy metal itself) and interactions with the cellular metabolism or even stimulation of ROS (indirect ROS generation). The cell-free control experiments corroborated the high catalytic activity of transition metal doped silica if compared to free metal ions in the case of Fe. For cobalt and manganese, the ROS levels were not statistically significantly different if either nanoparticles or aqueous salts were used. This control experiment clearly supports the previous interpretation that cobalt and manganese partially dissolve within the time frame of these experiments and that the corresponding oxide nanoparticles mediated the uptake of significant levels of heavy metals into the cells. Iron clearly demonstrated the opposite behavior. Here, the ROS levels for 1.6 wt % iron/silica was higher than those for pure Fe2O3 (Figure 3). If iron was delivered as a solution (empty columns) an about 20-fold ROS level again demonstrated the huge difference between dissolved iron and particulate iron. The different ROS levels arising from the application of aqueous solutions of transition metals (compare Figures 2 and 4, empty columns) both in cells (Figure 2) and in a simulated interaction with the reactive form of the dye in this assay (Figure 4) corroborated the well-known role of an intact cell membrane as an ion barrier. This barrier function obviously fails if the transition metals have been incorporated into a nanoparticle as a carrier. On the other hand, this observation also shows that the cell cultures were not significantly affected (e.g. apoptosis, morphological changes, detachment) during the present investigation. This was also confirmed by cell activity tests run in parallel to the present ROS measurements (see measured data as MTT conversion (13), SI Figure S9). Therefore, the present setup allowed elucidating the ROS formation of exposed cultures without significantly damaging the cells during the study and maintaining a relevant in vitro environment for simulating an exposure/uptake situation. The present investigation has shown that the chemical composition of nanoparticles is a most decisive factor determining the formation of reactive oxygen species in exposed cells. Beyond mass-based chemical effects where a toxic substance enters and damages a tissue or a cell, the size and mobility of nanoparticles give rise to two other effects: Partially soluble materials such as cobalt oxide may be taken up into cells by a Trojan-horse type mechanism which can significantly increase the damaging action of such materials. Catalytically active nanoparticles can give rise to prolonged damaging action in a cell since the material is not degraded during its interference with intracellular constituents.

Acknowledgments We thank F. Krumeich for TEM analysis and D. Gu ¨ nther and K. Hametner for LA-ICP MS measurements. Financial support by the Swiss Federal Office of Public Health (BAG, decision number 05.001872) is kindly acknowledged.

Supporting Information Available Additional information on oxidative stress and the mechanism of the here used ROS assay and a detailed materials and methods part. Materials characterization in terms of 4162

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 11, 2007

homogeneity, chemical composition, particle morphology, crystallinity, specific surface area, particle number concentrations during exposure, surface charge distributions (zetapotential), and the structure of the transition metal dopants in silica. The role of oxygen and hydrogen peroxide as precursors for ROS formation, cell viability during experiments, and diverse control experiments. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Cushing, B. L.; Kolesnichenko, V. L.; O’Connor, C. J. Recent advances in the liquid-phase syntheses of inorganic nanoparticles. Chem. Rev. 2004, 104, 3893-3946. (2) Oberdorster, G.; Oberdorster, E.; Oberdorster, J. Nanotoxicology: An emerging discipline evolving from studies of ultrafine particles. Environ. Health Perspect. 2005, 113, 823-839. (3) Colvin, V. L. The potential environmental impact of engineered nanomaterials. Nat. Biotechnol. 2003, 21, 1166-1170. (4) Donaldson, K.; Brown, D.; Clouter, A.; Duffin, R.; MacNee, W.; Renwick, L.; Tran, L.; Stone, V. The pulmonary toxicology of ultrafine particles. J. Aerosol Med. 2002, 15, 213-220. (5) Limbach, L. K.; Li, Y. C.; Grass, R. N.; Brunner, T. J.; Hintermann, M. A.; Muller, M.; Gunther, D.; Stark, W. J. Oxide nanoparticle uptake in human lung fibroblasts: Effects of particle size, agglomeration, and diffusion at low concentrations. Environ. Sci. Technol. 2005, 39, 9370-9376. (6) Rothen-Rutishauser, B. M.; Schurch, S.; Haenni, B.; Kapp, N.; Gehr, P. Interaction of fine particles and nanoparticles with red blood cells visualized with advanced microscopic techniques. Environ. Sci. Technol. 2006, 40, 4353-4359. (7) Geiser, M.; Rothen-Rutishauser, B.; Kapp, N.; Schurch, S.; Kreyling, W.; Schulz, H.; Semmler, M.; Hof, V. I.; Heyder, J.; Gehr, P. Ultrafine particles cross cellular membranes by nonphagocytic mechanisms in lungs and in cultured cells. Environ. Health Perspect. 2005, 113, 1555-1560. (8) Stark, W. J.; Strobel, R.; Gunther, D.; Pratsinis, S. E.; Baiker, A. Titania-silica doped with transition metals via flame synthesis: structural properties and catalytic behavior in epoxidation. J. Mater. Chem. 2002, 12, 3620-3625. (9) Sheldon, R. A. Homogeneous and heterogeneous catalytic oxidations with peroxide reagents. Top. Curr. Chem. 1993, 164, 21-43. (10) Fabrizioli, P.; Burgi, T.; Baiker, A. Environmental catalysis on iron oxide-silica aerogels: Selective oxidation of NH3 and reduction of NO by NH3. J. Catal. 2002, 206, 143-154. (11) Stark, W. J.; Wegner, K.; Pratsinis, S. E.; Baiker, A. Flame aerosol synthesis of vanadia-titania nanoparticles: Structural and catalytic properties in the selective catalytic reduction of NO by NH3. J. Catal. 2001, 197, 182-191. (12) Stark, W. J.; Kammler, H. K.; Strobel, R.; Gunther, D.; Baiker, A.; Pratsinis, S. E. Flame-made titania/silica epoxidation catalysts: Toward large-scale production. Ind. Eng. Chem. Res. 2002, 41, 4921-4927. (13) Brunner, T. J.; Wick, P.; Manser, P.; Spohn, P.; Grass, R. N.; Limbach, L. K.; Bruinink, A.; Stark, W. J. In vitro cytotoxicity of Oxide Nanoparticles: Comparison to Asbestos, Silica, and effects of particle solubility. Environ. Sci. Technol. 2006, 44, 43734381. (14) Stark, W. J.; Pratsinis, S. E. Aerosol flame reactors for manufacture of nanoparticles. Powder Technol. 2002, 126, 103-108. (15) Nel, A.; Xia, T.; Madler, L.; Li, N. Toxic potential of materials at the nanolevel. Science 2006, 311, 622-627. (16) Xia, T.; Kovochich, M.; Brant, J.; Hotze, M.; Sempf, J.; Oberley, T.; Sioutas, C.; Yeh, J. I.; Wiesner, M. R.; Nel, A. E. Comparison of the abilities of ambient and manufactured nanoparticles to induce cellular toxicity according to an oxidative stress paradigm. Nano Lett. 2006, 6, 1794-1807. (17) Tsuji, J. S.; Maynard, A. D.; Howard, P. C.; James, J. T.; Lam, C. W.; Warheit, D. B.; Santamaria, A. B. Research strategies for safety evaluation of nanomaterials, part IV: Risk assessment of nanoparticles. Toxicol. Sci. 2006, 89, 42-50. (18) Long, T. C.; Saleh, N.; Tilton, R. D.; Lowry, G. V.; Veronesi, B. Titanium dioxide (P25) produces reactive oxygen species in immortalized brain microglia (BV2): Implications for nanoparticle neurotoxicity. Environ. Sci. Technol. 2006, 40, 43464352. (19) Suh, W. H.; Jang, A. R.; Suh, Y. H.; Suslick, K. S. Porous, hollow, and ball-in-ball metal oxide microspheres: Preparation, endocytosis, and cytotoxicity. Adv. Mater. 2006, 18, 1832-1837.

(20) Hussain, S. M.; Javorina, A. K.; Schrand, A. M.; Duhart, H. M.; Ali, S. F.; Schlager, J. J. The interaction of manganese nanoparticles with PC-12 cells induces dopamine depletion. Toxicol. Sci. 2006, 92, 456-463. (21) Karakoti, A. S.; Hench, L. L.; Seal, S. The potential toxicity of nanomaterials - The role of surfaces. JOM 2006, 58, 77-82. (22) Sayes, C. M.; Wahi, R.; Kurian, P. A.; Liu, Y. P.; West, J. L.; Ausman, K. D.; Warheit, D. B.; Colvin, V. L. Correlating nanoscale titania structure with toxicity: A cytotoxicity and inflammatory response study with human dermal fibroblasts and human lung epithelial cells. Toxicol. Sci. 2006, 92, 174-185. (23) Gurr, J. R.; Wang, A. S. S.; Chen, C. H.; Jan, K. Y. Ultrafine titanium dioxide particles in the absence of photoactivation can induce oxidative damage to human bronchial epithelial cells. Toxicology 2005, 213, 66-73. (24) Sayes, C. M.; Gobin, A. M.; Ausman, K. D.; Mendez, J.; West, J. L.; Colvin, V. L. Nano-C-60 cytotoxicity is due to lipid peroxidation. Biomaterials 2005, 26, 7587-7595. (25) Dutoit, D. C. M.; Schneider, M.; Baiker, A. Titania-silica mixed oxides. 1. Influence of sol-gel and drying conditions on structural-properties. J. Catal. 1995, 153, 165-176. (26) Hutter, R.; Mallat, T.; Baiker, A. Titania-silica mixed oxides. 2. Catalytic behavior in olefin epoxidation. J. Catal. 1995, 153, 177-189. (27) Donaldson, K.; Li, X. Y.; MacNee, W. Ultrafine (nanometre) particle mediated lung injury. J. Aerosol. Sci. 1998, 29, 553-560. (28) Salem, I. A.; El-Maazawi, M.; Zaki, A. B. Kinetics and mechanisms of decomposition reaction of hydrogen peroxide in presence of metal complexes. Int. J. Chem. Kinet. 2000, 32, 643-666. (29) Beck, C.; Mallat, T.; Burgi, T.; Baiker, A. Nature of active sites in sol-gel TiO2-SiO2 epoxidation catalysts. J. Catal. 2001, 204, 428-439. (30) Wottrich, R.; Diabate, S.; Krug, H. F. Biological effects of ultrafine model particles in human macrophages and epithelial cells in mono- and co-culture. Int. J. Hyg. Environ. Health 2004, 207, 353-361. (31) Stone, V.; Shaw, J.; Brown, D. M.; MacNee, W.; Faux, S. P.; Donaldson, K. The role of oxidative stress in the prolonged inhibitory effect of ultrafine carbon black on epithelial cell function. Toxicol. in Vitro 1998, 12, 649-659. (32) Halliwell, B.; Whiteman, M. Measuring reactive species and oxidative damage in vivo and in cell culture: how should you do it and what do the results mean? Br. J. Pharmacol. 2004, 142, 231-255. (33) Jakubowski, W.; Bartosz, G. 2,7-dichlorofluorescin oxidation and reactive oxygen species: What does it measure? Cell Biol. Int. 2000, 24, 757-760. (34) Bass, D. A.; Parce, J. W.; Dechatelet, L. R.; Szejda, P.; Seeds, M. C.; Thomas, M. Flow cytometric studies of oxidative product formation by neutrophils - a graded response to membrane stimulation. J. Immunol. 1983, 130, 1910-1917. (35) Wang, H.; Joseph, J. A. Quantifying cellular oxidative stress by dichlorofluorescein assay using microplate reader. Free Radical Biol. Med. 1999, 27, 612-616. (36) Keller, A.; Mohamed, A.; Drose, S.; Brandt, U.; Fleming, I.; Brandes, R. P. Analysis of dichlorodihydrofluorescein and dihydrocalcein as probes for the detection of intracellular reactive oxygen species. Free Radical Res. 2004, 38, 1257-1267.

(37) Voelkel, K.; Krug, H. F.; Diabate, S. Formation of reactive oxygen species in rat epithelial cells upon stimulation with fly ash. J. Biosci. 2003, 28, 51-55. (38) LeBel, C. P.; Ischiropoulos, H. C. B. S. Evaluation of the probe 2′,7′-dichlorofluorescin as an indicator of reactive oxygen species formation and oxidative stress. Chem. Res. Toxicol. 1992, 5, 227231. (39) Grass, R. N.; Stark, W. J. Flame synthesis of calcium-, strontium-, barium fluoride nanoparticles and sodium chloride. Chem. Commun. 2005, 1767-1769. (40) Grass, R. N.; Tsantilis, S.; Pratsinis, S. E. Design of hightemperature, gas-phase synthesis of hard or soft TiO2 agglomerates. AIChE J. 2006, 52, 1318-1325. (41) Akurati, K. K.; Dittmann, R.; Vital, A.; Klotz, U.; Hug, P.; Graule, T.; Winterer, M. Silica-based composite and mixed-oxide nanoparticles from atmospheric pressure flame synthesis. J. Nanopart. Res. 2006, 8, 379-393. (42) Ehrman, S. H.; Friedlander, S. K.; Zachariah, M. R. Characteristics of SiO2/TiO2 nanocomposite particles formed in a premixed flat flame. J. Aerosol. Sci. 1998, 29, 687-706. (43) Dick, C. A. J.; Brown, D. M.; Donaldson, K.; Stone, V. The role of free radicals in the toxic and inflammatory effects of four different ultrafine particle types. Inhal. Toxicol. 2003, 15, 39-52. (44) Shen, B.; Scaiano, J. C.; English, A. M. Zeolite encapsulation decreases TiO2-photosensitized ROS generation in cultured human skin fibroblasts. Photochem. Photobiol. 2006, 82, 5-12. (45) Ananieva, E.; Reitzmann, A. Direct gas-phase epoxidation of propene with nitrous oxide over modified silica supported FeOx catalysts. Chem. Eng. Sci. 2004, 59, 5509-5517. (46) Fabrizioli, P.; Burgi, T.; Burgener, M.; van Doorslaer, S.; Baiker, A. Synthesis, structural and chemical properties of iron oxidesilica aerogels. J. Mater. Chem. 2002, 12, 619-630. (47) Ratnasamy, P.; Kumar, R. Ferrisilicate analogs of zeolites. Catal. Today 1991, 9, 329-416. (48) Jasinski, J.; Pinkerton, K. E.; Kennedy, I. M.; Leppert, V. J. Spatially resolved energy electron loss spectroscopy studies of iron oxide nanoparticles. Microsc. Microanal. 2006, 12, 424-431. (49) Zou, W. G.; Yan, M. D.; Xu, W. J.; Huo, H. R.; Sun, L. Y.; Zheng, Z. C.; Liu, X. Y. Cobalt chloride induces PC12 cells apoptosis through reactive oxygen species and accompanied by AP-1 activation. J. Neurosci. Res. 2001, 64, 646-653. (50) Petit, A.; Mwale, F.; Tkaczyk, C.; Antoniou, J.; Zukor, D. J.; Huk, O. L. Induction of protein oxidation by cobalt and chromium ions in human U937 macrophages. Biomaterials 2005, 26, 44164422. (51) Zhou, H.; Shen, Y. F.; Wang, J. Y.; Chen, X.; O’Young, C. L.; Suib, S. L. Studies of decomposition of H2O2 over manganese oxide octahedral molecular sieve materials. J. Catal. 1998, 176, 321328. (52) Worle-Knirsch, J. M.; Pulskamp, K.; Krug, H. F. Oops they did it again! Carbon nanotubes hoax scientists in viability assays. Nano Lett. 2006, 6, 1261-1268.

Received for review November 2, 2006. Revised manuscript received March 10, 2007. Accepted March 26, 2007. ES062629T

VOL. 41, NO. 11, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4163