Cytotoxicity of Cadmium-Free Quantum Dots and ... - ACS Publications

Publication Date (Web): May 28, 2014. Copyright © 2014 American Chemical Society. *(S.J.S.) E-mail: [email protected]. , *(S.C.D.) E-mai...
1 downloads 0 Views 4MB Size
Download PDF
Article pubs.acs.org/crt

Cytotoxicity of Cadmium-Free Quantum Dots and Their Use in Cell Bioimaging Stefaan J. Soenen,*,† Bella B. Manshian,† Tangi Aubert,‡,§ Uwe Himmelreich,† Jo Demeester,⊥ Stefaan C. De Smedt,*,⊥ Zeger Hens,‡,§ and Kevin Braeckmans§,⊥ †

MoSAIC/Biomedical NMR Unit, Department of Medicine, Catholic University of Leuven, Herestraat 49, B3000 Leuven, Belgium Physics and Chemistry of Nanostructures, Ghent University, Krijgslaan S9, 9000 Ghent, Belgium § Center for Nano- and Biophotonics, Ghent University, Sint-Pietersnieuwstraat 25, 9000 Ghent, Belgium ⊥ Lab General Biochemistry and Physical Pharmacy, Ghent University, Harelbekestraat 72, 9000 Ghent, Belgium ‡

S Supporting Information *

ABSTRACT: The use of quantum dots (QDots) as bright and photostable probes for long-term fluorescence imaging is gaining more interest. Thus far, (pre)clinical use of QDots remains limited, which is primarily caused by the potential toxicity of QDots. Most QDots consist of Cd2+ ions, which are known to cause high levels of toxicity. In order to overcome this problem, several strategies have been tested, such as the generation of cadmium-free QDots. In the present study, two types of cadmium-free QDots, composed of ZnSe/ZnS (QDotZnSe) and InP/ZnS (QDotInP), were studied with respect to their cytotoxicity and cellular uptake in a variety of cell types. A multiparametric cytotoxicity approach is used, where the QDots are studied with respect to cell viability, oxidative stress, cell morphology, stem cell differentiation, and neurite outgrowth. The data reveal slight differences in uptake levels for both types of QDots (maximal for QDotZnSe), but clear differences in cytotoxicity and cell functionality effects exist, with highest toxicity for QDotZnSe. Differences between cell types and between both types of QDots can be explained by the intrinsic sensitivity of certain cell types and chemical composition of the QDots. At concentrations at which no toxic effects can be observed, the functionality of the QDots for fluorescence cell visualization is evaluated, revealing that the higher brightness of QDotZnSe overcomes most of the toxicity issues compared to that of QDotInP. Comparing the results obtained with common Cd2+-containing QDots tested under identical conditions, the importance of particle functionality is demonstrated, revealing that cadmium-free QDots tested in this study are not significantly better than Cd2+-containing QDots for long-term cell imaging and that more work needs to be performed in optimizing the brightness and surface chemistry of cadmium-free QDots for them to replace currently used Cd2+-containing QDots.

■

core and enhances the optical properties of the QDots.7 QDots possess exceptional fluorescence properties, such as a high brightness and high photostability and broad excitation spectra combined with very narrow, size-dependent emission spectra enabling multiplexed imaging of cells, tissues, and whole animals using a single excitation source.8,9 In contrast to classical organic fluorophores, these properties have made QDots interesting tools for fluorescence imaging of fixed cells and tissues where multiple markers can be analyzed simultaneously. Additionally, QDots can be used for the continuous follow-up of single particles over longer time periods, allowing one to study the kinetics of mobile cellular markers such as cell membrane receptors.10 The use of cadmium-containing QDots for biomedical purposes has, however, remained quite limited due to the high toxicity of free Cd2+ ions and concerns on the potential toxic effects of these Cd2+-containing NPs.11−13 Studies on the

INTRODUCTION The ongoing progress in nanotechnology is rapidly opening the doors for a wide number of advanced biomedical applications.1−3 Novel nanosized materials (NMs), each with its own unique properties, are generated at an incredibly fast pace, stimulating the progress of a broad variety of biomedical research areas, including tissue engineering, neural conductance, targeted drug delivery, multimodal whole-body imaging, and combinations of both diagnostics and therapy.4,5 Among the different types of NMs, colloidal nanoparticles (NPs) take an important place either as (multimodal) imaging contrast agents or as therapeutic agents.6 One type of NM that has attracted a lot of attention in particular is colloidal semiconductor quantum dots (QDots). QDots are nanocrystals in the range of 1−10 nm diameter that are most often composed of heavy metals from group 12 and group 16. The most common QDot types with the highest quantum yields are the cadmium-containing QDots such as CdTe or CdSe. For biomedical purposes, the QDot cores will typically be surrounded by a shell layer (e.g., ZnS) that passivates the © 2014 American Chemical Society

Received: March 18, 2014 Published: May 28, 2014 1050

dx.doi.org/10.1021/tx5000975 | Chem. Res. Toxicol. 2014, 27, 1050−1059

Chemical Research in Toxicology

Article

as various types of Cd2+-containing QDots or fluorescently tagged silica NPs.12,29

toxic effects of QDots have generated a lot of disparate data, owing to, among others, differences in (a) size of the QDot core, (b) chemical composition of the QDots, (c) surface coating, (d) presence or absence of a shell layer, or (e) the experimental methodology used to evaluate QDot toxicity.14 The in vivo toxicity of Cd2+-containing QDots remains largely unknown. Although there is one initial study on primates where no toxic effects were observed under the conditions tested, this study did not take into account any possible long-term effects.15,16 In contrast to this, most cell studies found that Cd2+-containing QDots were highly toxic, although large differences exist between different cell types, where under some conditions, particles have been described to be selectively toxic toward cancer cells.17 These differences are likely due to the intrinsic differences between classical in vitro and in vivo methods, highlighting the need for proper experimental toxicity tests.18,19 On the basis of the data obtained, some general conclusions concerning Cd2+-based QDot toxicity can, however, be drawn: (a) the presence of a passivating shell layer typically reduces QDot toxicity, (b) a polymeric coating that envelopes the QDots typically provides the best protection in contrast to short ligands, and (c) toxicity is for the most part caused by the induction of reactive oxygen species and as a result of (intracellular) QDot degradation and release of free Cd2+ ions.11 As the presence of Cd2+ ions has been shown to be a major determinant in the toxicity of Cd2+-containing QDots, several strategies have been worked out to optimize the production and performance of Cd2+-free QDots.20,21 Originally, these QDots displayed inferior optophysical properties compared to that of classical Cd2+-containing QDots, but recent advances have enabled the generation of very bright QDots, mostly In3+based, with high quantum yields that were stable in aqueous media.22 These particles have demonstrated their potential use for biomedical applications in proof-of-concept studies as nearinfrared (NIR) emitters and for in vitro imaging of cancer cells or in vivo imaging of sentinel lymph nodes.22−24 However, for these materials to develop further as biomedical research tools, next to optimizing their optophysical properties and comparing them to Cd2+-containing QDots, their potential toxicity needs to be analyzed. Few studies thus far have looked into the toxicity of these materials, but some promising preliminary results have already been obtained, where toxic levels were found to be lower than those for their Cd2+-containing counterparts.23,25 The present work evaluates the cytotoxicity of two different types of Cd2+-free QDots (ZnSe/ZnS and InP/ ZnS, respectively) according to a previously established multiparametric methodology.26 This methodology enables an in-depth evaluation of cell−NP interactions employing reproducible protocols and will provide a concentration at which no toxic effects can be observed.27 As accurate determination of NP concentrations is a time-consuming and error-prone task, a direct comparison of nontoxic NP concentrations cannot be performed with high certainty. Therefore, at the concentration at which no toxic effects are observed, the functionality of the QDots will be evaluated with regard to their potential for visualization of live cells, as determined by visualizing the number of QDot-containing cells by fluorescence microscopy during extended culture times, as described previously.12,28 Then, their functionality at nontoxic levels can be compared directly to other NPs with similar functionality that have been previously studied under identical conditions using the same multiparametric methodology, such

■

EXPERIMENTAL PROCEDURES

Nanoparticles. Two types of Cd2+-free QDots were purchased from Mesolight LLC (Little Rock, Arkansas USA). One type of QDots was ZnSe/ZnS core−shell QDots, and the other QDots were InP/ZnS core−shell QDots. The particles were coated with 3-mercaptopropionic acid, which bestows them with a negative surface charge. The particles were provided as 9 μM QDot stock solutions in alkaline H2O at pH 11. Nanoparticle Characterization. Size Determination by Transmission Electron Microscopy. The size of the QDots was determined by transmission electron microscopy (TEM) using a Cs-corrected JEOL 2200 FS microscope operating at 200 kV. Elemental Analysis. The chemical composition of the QDots was determined by energy dispersive X-ray spectroscopy (EDX) using the same microscope as that for the TEM observations. Absorbance and Emission Spectra. Emission spectra of the QDots were recorded using a Quantifluor fluorometer (Promega, Belgium) with an excitation wavelength of 340 nm. Dynamic Light Scattering and Electrophoretic Mobility Measurements. The hydrodynamic diameter and ζ-potential of both types of QDots were measured using a Nanosizer instrument (Malvern, Worcestershire, UK). For ζ-potential measurements, the QDots were diluted (1/500) in phosphate buffered saline (PBS, 10 mM; pH 7.0) after which the measurements were performed (12 cycles/run) in triplicate. For the determination of hydrodynamic size, the QDots were diluted (1/500) in full culture medium (see the Cell Culture section below) containing 10% fetal calf serum and 5% horse serum after which the measurements were performed (12 cycles/run) in triplicate. As a control, serum-containing media only were measured. Upon addition of the QDots, any peaks that were shifted in size or new peaks that emerged were considered for the calculation of the average size of the particles in order to exclude the pure serum proteins. QDot Sterility Tests. Both types of QDots were provided as sterile stock suspensions. To rule out the effect of any biological contaminants such as endotoxins, the common end point chromogenic QCL-1000 LAL assay (Lonza, Verviers, Belgium) was performed according to the manufacturer’s instructions. QDot-Cell Interaction Studies. Cell Culture. C17.2 neural progenitor cells and PC12 rat pheochromocytoma cells were cultured in high glucose containing Dulbecco’s modified Eagle’s medium (DMEM), supplemented with 10% fetal calf serum, 5% horse serum, 1 mM sodium pyruvate, 2 mM L-glutamine, and 1% penicillin/ streptomycin (Gibco, Invitrogen, Belgium). C17.2 cells were passaged every 48 h and split 1/5. PC12 cells were passaged when reaching near 70% confluency and split 1/5 in tissue culture dishes (Greiner BioOneBA/BV, Wemmel, Belgium) which were coated with synthetic laminin peptide (Synthetic laminin peptide for Rat Neural Stem Cells, Millipore SA/NV, Brussels, Belgium). To establish nonproliferating cell cultures, cells were exposed with 60 mM Apigenin (Sigma-Aldrich, Bornem, Belgium). As particle uptake is linked with cell cycle progression, Apigenin treatment occurred immediately after cells had been incubated with QDots at the desired concentrations. After media removal, fresh media containing 60 mM Apigenin were used, where media were replaced for 50% every other day with fresh Apigenincontaining medium for the duration of the experiments. Human umbilical vein endothelial cells (HUVECs) were maintained in endothelial basal/growth culture medium (EBM-2/EGM-2, Clonetics, San Diego, CA) with medium changes every 48 h. Cells were passaged when reaching near 80% confluency by lifting the cells with 0.05% trypsin (Gibco) and were plated (1/5) onto tissue-culture flasks coated with collagen. To establish nonproliferating HUVEC cultures, cells were given endothelial cell serum-free defined medium (Cell Applications, Tebu-Bio, Le Perray en Yvelines, France) immediately after cells had been exposed to the QDots. Confluent HUVEC monolayers could then be maintained for at least 1 week. 1051

dx.doi.org/10.1021/tx5000975 | Chem. Res. Toxicol. 2014, 27, 1050−1059

Chemical Research in Toxicology

Article

Figure 1. Cellular uptake of QDots. (A) Representative confocal images of HUVEC cells transiently expressing Lamp-1-GFP (marker for late endosomes and lysosomes; green channel) incubated with 80 nM QDotInP (top row; red channel) or QDotZnSe (bottom row; blue channel) for 24 h. A merged image of both the QDots and the GFP-expressing endosomes is shown in the right column. The area indicated by the white rectangle in the merged images is shown in high resolution on the right. Scale bars: 40 μm. (B) The number of QDot clusters per cell as quantified for 30 cells. (C) The total number of QDots per HUVEC cell as a function of the QDot concentration for both QDotInP (light gray) and QDotZnSe (dark gray) as quantified by measuring total fluorescence intensity levels. Data are shown as the mean ± SEM (n = 4). Cell-QDot Studies. A detailed experimental methodology can be found in the Supporting Information. Statistical Analysis. All data are expressed as the mean ± SEM unless indicated otherwise and analyzed using one-way analysis of variance (ANOVA). Comparisons between particle-treated groups and untreated control groups were analyzed using the Dunnett posthoc analysis method. In all cases, the degree of significance is indicated when appropriate (*, p < 0.05; **, p < 0.01; ***, p < 0.001).

■

S2). These data reveal that the MPA coating results in relatively stable particles at alkaline pH, where in cell culture media, some agglomeration is seen, and small clusters of nanoparticles and surface-bound molecules are formed. The QDotsZnSe have a maximal emission at 425 nm (Supporting Information, Figure S3B) and a photoluminescence quantum yield (PLQY) of 36%. The QDotsInP have maximal emission at 598 nm (Supporting Information, Figure S3A) and a PLQY of 27%. Additionally, the stock suspensions were shown not to contain any endotoxins as evaluated by the chromogenic QCL-1000 LAL assay (data not shown). Cellular QDot Uptake. In the present study, cell experiments were performed on three different cell types: primary human umbilical vein endothelial cells (HUVEC), murine neural progenitor cells (C17.2), and rat pheochromocytoma cells (PC12). These cells have been shown to be well suited for nanotoxicity studies12,29,30 as they have widely differing characteristics and therefore provide a nice and general overview of nanoparticle impact on cultured cells. Additionally, the HUVEC and C17.2 cells are frequently used in transplantation experiments for which cell labeling by nanomaterials for cell tracking purposes is a major application.31 Furthermore, the cells have been used in various studies, enabling a direct comparison of the results obtained in the current study to those obtained in previous studies for similar particles investigated under identical conditions.28 For biomedical applications, the particles will finally be internalized by the cultured cells, which normally results in their enclosure in the endosomal compartments of the cell.32 Figure 1A reveals a high level of colocalization between either type of QDot and green fluorescent protein (GFP)-expressing endosomal compartments within the cells, revealing that both types of QDots were indeed intraendosomally located. On the

RESULTS AND DISCUSSION

Nanoparticle Characterization. Two types of core−shell QDots were used, either InP/ZnS QDots (QDotsInP) or ZnSe/ ZnS QDots (QDotsZnSe). The chemical composition of the two types of QDots was evaluated by elemental analysis (Supporting Information, Figure S1), revealing that the QDotsInP consist of approximately a 13% InP core and an 87% ZnS shell, whereas the QDotsZnSe consist of about a 48% ZnSe core and a 52% ZnS shell, which is in line with the descriptions given by the commercial supplier. These QDots both had a coating of 3-mercaptopropionic acid (MPA), which results in a negative surface charge of −32 ± 2 and −27 ± 2 mV for QDotsInP and QDotsZnSe, respectively. Particle diameters were assessed by transmission electron microscopy (TEM; Supporting Information, Figure S1), revealing similar sizes for both types of QDots of 5.9 ± 0.7 nm and 5.6 ± 0.6 nm for QDotsInP and QDotsZnSe, respectively. The apparent aggregation of the QDots seen in the TEM images may be an artifact derived from the high concentration of NPs in the stock suspension and the TEM procedure itself which uses a vacuum system. However, the MPA coating in itself does not provide excellent colloidal stability, which is in line with the measured hydrodynamic diameters of the QDots exposed for 24 h in cell media (DMEM): 115 ± 12 nm for QDotsInP and 147 ± 14 nm for QDotsZnSe (Supporting Information, Figure 1052

dx.doi.org/10.1021/tx5000975 | Chem. Res. Toxicol. 2014, 27, 1050−1059

Chemical Research in Toxicology

Article

basis of fluorescence images, the number of distinct dots (= QDot clusters) was counted, as described previously,12 revealing slightly but nonsignificantly higher uptake levels for the QDotsZnSe in all three cell types (Figure 1B and Supporting Information, Figure S4). To carefully evaluate nanoparticle toxicity, it is essential to determine the number of QDots per cell in order to link the toxicity of the particles to their functional (= intracellular) concentration.28 As Zn2+ ions are present within the cells, the use of inductively coupled plasma spectroscopy for determining cellular QDotZnSe levels is not that straightforward. Therefore, we used a previously validated methodology,12,28 in which cellular QDot levels are determined. To this end, the cells were lysed in a low molarity buffer with alkaline pH (9.8), after which the concentration of the QDots was determined by comparing fluorescence intensity levels with the fluorescence of known QDot stock suspensions in the same solvent. Figure1C (Supporting Information, Figure S4) shows the same trend as for the number of clusters per cell, where the number of QDotsZnSe is slightly higher than the number of QDotsInP, especially in the case of the PC12 cells. The differences in cellular uptake are somewhat surprising, given the similar hydrodynamic size and ζ-potential of the QDots that have an identical surface coating. This suggests that the actual core of the particles influences “cell vision”, which is a parameter that represents how the cell handles foreign materials when they are subjected to the biological environment (such as serum components) that surrounds the cell.33 As the difference in surface charge and (hydrodynamic) size are so minimal and considering that both QDots consist of an outer shell of identical chemical composition (ZnS), the difference in the chemical composition of the core seems to play an important role. Possibly, the different core structures influence the crystal structure of the shell layer, which in turn influences the formation of the protein corona and how these particles will be processed by the cell. More in-depth research is needed in the future to evaluate this hypothesis. Overall, these data suggest that both types of QDots are taken up efficiently by all cell types and are finally located within the endosomal compartments of the cell. The QDotsZnSe are internalized slightly more efficiently than the QDotsInP, which likely reflects minor differences in cell vision depending on the QDot core. Effect of Endosomal Environment on QDot Properties. One of the consequences of being located inside endosomes is that the QDots will be exposed to low pH levels (4.5 for lysosomes; 5.5 for late endosomes). To verify the effect of this low pH on QDot stability, both types of QDots were subjected to endosomal buffer systems of pH 7.4, 5.5, or 4.5, as described previously,12 after which the effect of the pH on their fluorescence intensity was determined. Figure 2 reveals high optophysical stability of the particles, where almost no decrease in fluorescence intensity could be observed over several days, even when the QDots were exposed to the lowest pH levels. Compared to previous studies on various types of Cd2+containing QDots studied under identical conditions, the stability of the Cd2+-free QDots is substantially higher, where the fluorescence intensity of Cd2+-containing QDots dropped by, on average, 60% after 5 days exposure to pH 4.5. The higher optical stability of the Cd2+-free QDots likely also reflects a higher chemical stability of the particles compared to that of the Cd2+-containing QDots. For the latter particles, it has been shown that their exposure to low endosomal pH levels results in a degradation of the QDots themselves as a result of acid etching12,28 and subsequent release of highly toxic Cd2+ ions in

Figure 2. Fluorescence intensity levels of QDotInP and QDotZnSe exposed to buffer systems of different pH values of 7.4 (cytoplasmic); 5.5 (late endosomes); or 4.5 (lysosomes) for up to 5 days. Fluorescence intensity levels were measured every day. Data are presented as values relative to the fluorescence intensity of QDots exposed to pH 7.4 and measured immediately. Data are expressed as the mean ± SEM (n = 3).

their immediate environment, increasing nanoparticle cytotoxicity.28,34 Acute and Long-Term Toxicity. The cytotoxic effects of both types of QDots were determined using the Alamar Blue assay for QDot concentrations ranging from 0 to 100 nM. Figure 3A (Supporting Information, Figure S5) shows significant cytotoxicity starting from 60 nM for QDotsZnSe and 80 nM for QDotsInP. The difference in toxicity between the two QDot types likely reflects the differences in cellular nanoparticle levels, which was also slightly higher for the QDotsZnSe. However, the differences in cellular uptake are only minimal, suggesting that other factors intrinsic to the different types of QDots played an additional role in their cytotoxicity, such as the type of heavy metal used in the chemical composition of the QDots. As QDot degradation results in the release of heavy metal ions which can be highly toxic, such as Cd2+, the intracellular presence of QDots can result in a slow but gradually increasing damage of the cells.28 To test this, cells were first labeled with various concentrations of either type of QDot, after which they were kept in culture under nonproliferative conditions as detailed elsewhere.12 The halt in cell proliferation is essential to allow one to accurately determine the effect of QDot degradation on particle cytotoxicity without the rapid and asymmetric dilution of cellular particle numbers which accompanies cell division.35 Furthermore, under these conditions, the cells more accurately mimic the in vivo situation, where most cells are also not actively dividing. Figure 3B reveals no increase in cytotoxicity up to at least 7 days after initial cell labeling. Together with the high stability of the particle fluorescence in time, these data indicate a high chemical stability of the QDots, where environmentally induced degradation appears to be minimal. This is in stark contrast to common Cd2+-containing QDots that have been found to degrade under endosomal pH conditions.12 This may be explained by differences in the crystal structure of the QDots where some Cd2+-free QDots have been described to be very robust against environmental influences such as oxidation. Oxidative Stress and Secondary Effects. The use of nanomaterials, such as zinc-containing NPs, is known to be closely linked to their induction of oxidative stress, which, when it persists over longer time periods or reaches higher levels, can result in cell death.36−39 Figure 3C (Supporting Information, Figure S5) shows a clear induction of reactive oxygen species 1053

dx.doi.org/10.1021/tx5000975 | Chem. Res. Toxicol. 2014, 27, 1050−1059

Chemical Research in Toxicology

Article

Figure 3. (A) Viability of HUVEC as a function of QDot concentration (ranging from 20 to 100 nM) for both QDotInP (light gray) and QDotZnSe (dark gray) after 24 h of incubation. (B) Viability of nonproliferating HUVEC cells as a function of both QDot concentration (40, 60, and 80 nM) and time (1−7 days post-QDot incubation). (C) ROS levels of HUVEC cells exposed to varying concentrations (10−60 nM) of QDotInP (light gray) or QDotZnSe (dark gray). Data are represented as the mean ± SEM (n = 4) and expressed as relative to untreated control cells. The degree of significance between samples and controls is indicated when appropriate (*, p < 0.05; **, p < 0.01; ***, p < 0.001). Statistical significance between the QDotInP and QDotZnSe samples is indicated by #.

The difference between QDotsZnSe and QDotsInP is somewhat strange as both types of QDots have the same shell layer in terms of composition (ZnS). Additionally, they are coated with the same organic coating and have a similar hydrodynamic size and ζ-potential. Yet, similar to cellular uptake, the differences in the core structure of the QDots appear to affect their final cellular effects. Thus far, the exact cause of this remains rather unclear. Given the high stability of the particles against degradation, it is rather unlikely that many ions from the particle core will be free in the environment to affect cell viability. Even if very low levels of degradation would occur in time, some Zn2+ ions will be released from the shell layer in both types of QDots. For the QDotsZnSe, more Zn2+ ions can be released from the core, elevating the total level of cellular zinc, an essential component in cell homeostasis, above toxic thresholds. For the QDotsInP, In3+ will be released, which in itself may not reach the levels that are above the toxic threshold. The total level of Zn2+ ions will then likely also be lower for the latter QDots, where the contribution of In3+ and Zn2+ separately on inducing oxidative stress may be less outspoken than higher levels of Zn2+ alone. To evaluate this in more detail, the effect of In3+ and Zn2+ ions on cell viability (Supporting Information, Figure S9) and oxidative stress (Supporting Information, Figure S10) has been investigated. These data clearly reveal that neither of the ions results in any significant cytotoxicity at concentrations up to 12 μM (which exceeds the level that can be released from the QDots), as observed for Cd2+-containing QDots.12,28 The level of cellular In3+ or Zn2+ ions were not assessed in the current study because for Zn2+ ions, which are naturally present, determining the level of QDot-derived Zn2+ in a cellular environment by ICP-MS is very challenging. The lack of any toxic effects for these ions under the conditions used are in line with other reports, where,

(ROS), reaching significant levels at 30 nM for QDotsZnSe and 50 nM for QDotsInP. The higher level of oxidative stress for the QDotsZnSe suggests the important role of ROS in QDot toxicity given their higher cytotoxicity. The link between oxidative stress and cytotoxicity is quite unclear, as all cell types have different degrees of antioxidative capacity.40 Oxidative stress can affect cell viability through several possible mechanisms, such as the induction of mitochondrial damage, alterations in calcium fluxes, or DNA damage. Therefore, the effect of both types of QDots on secondary oxidative stress mechanisms was investigated. Figure 4A (Supporting Information, Figure S6) shows no significant increase in the number of double strand breaks for either type of QDot as revealed by phosphorylated γH2Ax staining, a common marker for DNA double strand breaks.41 Furthermore, cellular calcium levels were slightly but not significantly increased for both types of QDots at higher particle concentrations (Figure 4B, Supporting Information, Figure S7). Changes in calcium waves are important signaling mediators in cell homeostasis, where alterations in cytoplasmic calcium waves can have drastic effects on many important cellular processes such as cell division or cell viability.42 Calcium waves, however, have only a short lifetime and die out rather fast so they can easily be overlooked in the case of minor alterations.43 Therefore, the effect of the QDots on alterations in mitochondrial membrane potential (Δψm) was investigated as a more robust marker. Figure 4C (Supporting Information, Figure S8) shows a clear and significant loss of Δψm for both types of QDots, starting from 30 nM for QDotsZnSe and 60 nM for QDotsInP. These data are closely matched with the differences in oxidative stress levels induced by both types of QDots as well as their cytotoxicity and therefore suggest the importance of oxidative stress in the cytotoxic profile of both types of QDots. 1054

dx.doi.org/10.1021/tx5000975 | Chem. Res. Toxicol. 2014, 27, 1050−1059

Chemical Research in Toxicology

Article

selected at which no cytotoxicity was observed, being 60 nM for QDotsZnSe and 80 nM for QDotsInP. In the present study, the three cell types used offer optimal conditions for assessing important functional parameters. The widely spread HUVEC cells are ideally suited to investigate the effect of the QDots on cell morphology and cytoskeletal architecture.30 For QDots that are ideally suited for bioimaging applications in live cells, it is essential that the intracellular presence of the particles does not affect cell structure. Some particles, including QDots, have been found to disturb cell morphology and the cell cytoskeleton, and some studies have shown direct effects on actin fibers.12,47 The HUVEC cells were therefore stained for Factin and α-tubulin, after which the spreading of the cells was analyzed by confocal microscopy and image analysis, as performed previously.43 Figure 5A shows representative images of the cells which display a clear spreading and nice cytoskeletal architecture for untreated control cells and cells labeled with low concentrations of QDots. Cell area distribution was then analyzed (Figure 5B), which shows a clear decrease in cell spreading at 40 nM for QDotsZnSe and 50 nM for QDotsInP. These effects may in part be due to the induction of ROS, but when cells were treated with 5 mM N-acetylcysteine (NAC), a free radical scavenger that reduced cellular ROS levels to near control values (data not shown), the decrease in cell spreading could only be partially overcome (Figure 5C). The results demonstrate that oxidative stress plays a major role in the cytotoxic effects of the QDots but that for both types of particles, other mechanisms also played an important role, such as cytoskeletal rearrangements and reductions in cell spreading. For C17.2 cells, cell functionality can be evaluated by determining the efficiency of cell differentiation, where the neural progenitor cells can be differentiated into full neurons. Figure 5D shows a clear and significant loss in differentiation efficiency of the C17.2 cells when treated with 50 nM QDotsZnSe, whereas QDotsInP did not appear to have any significant effect. Treating the cells with 5 mM NAC could again only partially overcome the observed effects, suggesting that cell differentiation was affected through other non-ROS associated mechanisms. The differentiation of the C17.2 cells is a long process that is accompanied by high levels of cell death, which can harden any analysis of minor effects.26 The PC12 cells present a more elegant approach for investigating nanoparticle effects on cell functionality as they can reversibly induce rapid neurite outgrowth that can easily be quantitated.48 This process is also accompanied by only low levels of cell death and therefore allows picking up any effects with higher sensitivity than the C17.2 cells. Figure 5E shows significant reductions in neurite outgrowth efficiency at 10 nM QDotsZnSe and 20 nM QDotsInP. Taken together, these data show that at subcytotoxic concentrations, both types of QDots still affect cell homeostasis, by impeding cell functionality and inducing oxidative stress. To evaluate the functionality of the particles themselves, further tests are performed at the concentrations of the particles at which no adverse effects were observed: 10 nM for QDotsZnSe and 20 nM for QDotsInP. When compared to different types of Cd2+-containing core−shell QDots studied under identical conditions, these concentrations are approximately 10-fold higher (ranging from 0.5 to 2 nM).12,28 This high level of particle concentration that can be tolerated by the cells is likely caused by a combination of several factors, in part the lower intrinsic cytotoxicity of both types of Cd2+-free QDots, which is in line with other studies reporting lower

Figure 4. Secondary ROS effects of Cd2+-free QDots. (A) The level of phosphorylated γ-H2Ax, (B) the level of cellular calcium, or (C) mitochondrial viability in HUVEC cells as a function of QDot concentration (ranging from 10 to 60 nM) for both QDotInP (light gray) and QDotZnSe (dark gray) after 24 h of incubation. Data are expressed as the mean ± SEM (n = 3). The degree of significance between samples and controls is indicated when appropriate (*, p < 0.05; **, p < 0.01; ***, p < 0.001). Statistical significance between the QDotInP and QDotZnSe samples is indicated by #.

for example, for freshwater shrimp, the LC50 of In3+ ions was found to be 60 μM compared to 0.5 μM for Cd2+.44 Interesting, however, is the induction of oxidative stress for both ions (Supporting Information, Figure S10). In3+ ions, which are known ROS inducers,45 show a clear concentrationdependent induction of ROS that reaches saturation levels at higher In3+ concentrations and then slowly drop to lower levels. Zn2+ ions, however, initially do not affect cellular ROS levels up to a concentration of 10 μM but then result in a sharp increase at higher level. The lack of any significant ROS inductions by low levels of Zn2+ ions is in line with their status as potent antioxidants.46 The higher toxicity and higher induction of ROS by QDotZnSe is in apparent contrast to these findings. However, it has also been shown that Zn2+ ions can aggravate the cellular stress induced by other stimuli.46 In the present study, the QDots themselves may induce oxidative stress, where low levels of free Zn2+ may further increase cellular ROS levels, specifically in the case of QDotZnSe. Effects of QDots on Cell Functionality. Nanomaterials can affect cells in many different ways, requiring a large number of parameters to be analyzed to get a clear overview of potentially negative effects that these particles may exert. For functionally active cells, such as HUVEC or C17.2 cells that are actively being explored in cell transplantation studies,31 assessing cell functionality is an important aspect. It is important to note that dead cells will logically not have any more functionality, and therefore, only those conditions are 1055

dx.doi.org/10.1021/tx5000975 | Chem. Res. Toxicol. 2014, 27, 1050−1059

Chemical Research in Toxicology

Article

Figure 5. (A) Representative fluorescence images of HUVEC cells either untreated (top row) or exposed to 20 nM QDotZnSe (second row), 50 nM QDotZnSe (third row), 20 nM QDotInP (fourth row), or 50 nM QDotInP (bottom row) for 24 h. Cells have been stained for F-actin (red, left column) and α-tubulin (green, middle column). The right column shows a merged image of both the green and red channels. Scale bars: 150 μm. (B) The average cell area of HUVEC cells as a function of QDot concentration (ranging from 0 to 60 nM) for both QDotInP (light gray) and QDotZnSe (dark gray) after 24 h incubation. (C) The average cell area of HUVEC cells as a function of QDot concentration (ranging from 0 to 60 nM) upon 24 h of incubation with either QDotInP (light gray) or QDotZnSe (dark gray) in the presence of 5 mM NAC, a free radical scavenger. (D) The level of C17.2 cells that differentiated into full neurons upon 1 week of exposure to neuronal induction media as a function of QDot concentration (ranging from 0 to 60 nM) for both QDotInP (light gray) and QDotZnSe (dark gray) after 24 h of incubation. (E) The level of PC12 neurite outgrowth upon 2 days of exposure to nerve growth factor as a function of QDot concentration (ranging from 0 to 60 nM) for both QDotInP (light gray) and QDotZnSe (dark gray) after 24 h of incubation. (B−E) Data are expressed as the mean ± SEM (n = 4). The degree of significance between samples and controls is indicated when appropriate (*, p < 0.05; **, p < 0.01; ***, p < 0.001). Statistical significance between the QDotInP and QDotZnSe samples is indicated by #.

toxicity levels for Cd2+-free QDots.25 However, in order to accurately assess particle toxicity, the intrinsic toxicity of the particles must be considered. A direct comparison of the different particles is made impossible by differences in their size, surface coating, and the associated changes in cellular interaction such as alterations in cellular uptake levels.28 Furthermore, a comparison of particle toxicity based on concentrations is prone to big errors as accurate determination of nanoparticle numbers in stock suspensions is extremely tedious.14 Functionality of the QDots for Cell Imaging. On the basis of the determination of the safe concentrations of both types of QDots, their functionality can be assessed by evaluating how long they can be used for visualizing live, actively dividing cells, which would be one major advantage of the application of QDots. This then allows a much better comparison of different nanoparticles with the same function-

ality as it includes a large number of potential differences between particles such as differences in uptake levels, errors in determining stock concentrations, or QDot PLQY.28 In practice, cells are seeded with either type of QDot at noncytotoxic concentrations for 24 h (being 10 nM for QDotsZnSe and 20 nM for QDotsInP), after which cells are kept in culture. Upon every two cell divisions, cells are reseeded at their initial density, and part of the cell population is kept for microscopy analysis, as described in detail in the Supporting Information, Experimental Methodology. The reseeding of the cells is performed in order to avoid the need for low cell densities at initial time points (HUVEC and PC12 cell proliferation requires optimal cell densities) and to avoid cell cultures becoming overpopulated, which would impede cell division due to contact inhibition. Figure 6 shows that the QDotsZnSe can be used to visualize cells for approximately 4 cell cycles, whereas the QDotsInP allow cells to be tracked for 1056

dx.doi.org/10.1021/tx5000975 | Chem. Res. Toxicol. 2014, 27, 1050−1059

Chemical Research in Toxicology

Article

Figure 6. (A) Representative fluorescence images of HUVEC cells exposed to noncytotoxic concentrations of QDotInP (20 nM; top row) or QDotZnSe (10 nM; bottom row) for 24 h, after which the cells were kept in culture. The images shown are taken immediately after QDot exposure (D0) or after 2 (D2), 4 (D4), or 6 (D6) average cell doubling times. The images are merged images of the fluorescent QDots (QDotInP, red; QDotZnSe, blue) and phase contrast images of the cells. Scale bars: 150 μm. (B−D) The percentage of (B) C17.2, (C) HUVEC, or (D) PC12 cells containing QDotsInP (light gray) or QDotZnSe (dark gray) upon further culture of preincubated cells. The data are expressed as a function of the number of cell divisions the cells underwent during postincubation. Data are given as the mean ± SEM (n = 4). The horizontal line indicates 50% of QDot positive cells, which is seen as the cutoff value below which the ability of the QDots to visualize cells is referred to as insufficient.

Cd2+-containing QDots (i.e., so-called gradient alloy QDots, with identical surface coating to the QDots evaluated here) that were found to have a PLQY of up to 90%, resulted in a similar duration of cell tracking at nontoxic conditions (up to 6 cell doublings). Therefore, the Cd2+-free QDots examined in the current study appear to be similar to more advanced Cd2+containing QDots. In the future, further improvements in surface chemistry and PLQY of the Cd2+-free QDots may boost their functionality to higher levels, especially for the QDotsInP. At this stage, however, more research needs to be performed for these particles to be optimized and to reach their full potential in replacing current Cd2+-containing QDots.

approximately 6 cell cycles. For various types of core−shell Cd2+-containing QDots, cells could be tracked for approximately 2 to 4 cell cycles, indicating that the Cd2+-free QDots are approximately 1.5-fold better than some Cd2+-containing QDots previously tested.12,28 The difference in particle functionality is therefore far less than expected based on the concentrations of the particles only. This is likely caused by the relatively low brightness of the Cd2+-free QDots compared to that of some Cd2+-containing QDots. Additionally, this may be due to a lower concentration of QDots in the stock suspension than originally determined as well as the short organic ligands, which may result in lower colloidal stability of the particles in cell medium and reduced cellular uptake.49 Overall, the Cd2+free QDots appear to be well suited alternatives for Cd2+containing QDots for cell bioimaging purposes, where the QDotsZnSe appear to be slightly less suited than the QDotsInP, likely because of a higher intrinsic toxicity and lower brightness of the QDotsZnSe. This is further supported upon determining the cellular fluorescence intensity of the QDot-labeled cells, where QDotZnSe-labeled cells have a much lower fluorescence intensity than the QDotInP-labeled cells (Supporting Information, Figure S11). At this stage, however, both types of QDots appear to be similar to common Cd2+-containing QDots. This is further supported by a recent study, where more advanced

■

CONCLUSIONS The present work shows that the cadmium-free QDots, despite having a high chemical stability and low intrinsic toxicity, appear to only be slightly better than commonly used Cd2+containing QDots in terms of cell visualization. This is primarily due to the relatively low PLQY and associated lower brightness of both types of cadmium-free QDots tested in this study. Although the QDotInP and QDotZnSe both possess a ZnS shell layer, are of similar size, and have the same surface chemistry, they result in slight differences in cellular uptake levels, but significant differences in toxic effects, which are most 1057

dx.doi.org/10.1021/tx5000975 | Chem. Res. Toxicol. 2014, 27, 1050−1059

Chemical Research in Toxicology

■

pronounced for QDotZnSe. These data indicate the importance of the chemical composition of the QDot core in their final interactions with biological entities such as cells or whole organisms. The use of the multiparametric methodology enables one to define nontoxic particle concentrations and sheds some insight into the mechanisms involved in the negative effects of the QDots. Additionally, using the same methods and cell types, and investigating particle toxicity under well-defined and controllable conditions allows one to compare the results obtained with results for other particles studied in the same way. This comparison reveals that cells can tolerate a higher level of cadmium-free QDots than cadmium-containing QDots, but when taking particle brightness and uptake levels into account, similar results are obtained, where cadmium-free QDots are only moderately more functional. This comparison shows the importance of particle functionality when assessing particle toxicity because comparing particle concentrations does not give an appropriate estimation of the true differences in particle toxicity, as many parameters such as errors in determining particle stock concentration, cellular uptake levels, etc. are not taken into account. The low intrinsic toxicity of the cadmium-free QDots offers a lot of potential for future applications, where more stable coatings, such as functionalized silica shells could be applied to optimize colloidal stability and particle brightness in aqueous solvents while maximizing cellular uptake.

■

REFERENCES

(1) Barkalina, N., Charalambous, C., Jones, C., and Coward, K. (2014) Nanotechnology in reproductive medicine: Emerging applications of nanomaterials. Nanomedicine, DOI: 10.1016/ j.nano.2014.01.001. (2) Ding, Y. P., Li, S. P., and Nie, G. J. (2013) Nanotechnological strategies for therapeutic targeting of tumor vasculature. Nanomedicine 8, 1209−1222. (3) Dvir, T., Timko, B. P., Kohane, D. S., and Langer, R. (2011) Nanotechnological strategies for engineering complex tissues. Nat. Nanotechnol. 6, 13−22. (4) Gu, Z. J., Yan, L., Tian, G., Li, S. J., Chai, Z. F., and Zhao, Y. L. (2013) Recent advances in design and fabrication of upconversion nanoparticles and their safe theranostic applications. Adv. Mater. 25, 3758−3779. (5) Ge, Z. S., and Liu, S. Y. (2013) Functional block copolymer assemblies responsive to tumor and intracellular microenvironments for site-specific drug delivery and enhanced imaging performance. Chem. Soc. Rev. 42, 7289−7325. (6) Wu, X., Wu, M., and Zhao, J. X. (2014) Recent development of silica nanoparticles as delivery vectors for cancer imaging and therapy. Nanomedicine 10, 297−312. (7) Walling, M. A., Novak, J. A., and Shepard, J. R. E. (2009) Quantum dots for live cell and in vivo imaging. Int. J. Mol. Sci. 10, 441−491. (8) Medintz, I. L., Uyeda, H. T., Goldman, E. R., and Mattoussi, H. (2005) Quantum dot bioconjugates for imaging, labelling and sensing. Nat. Mater. 4, 435−446. (9) Michalet, X., Pinaud, F. F., Bentolila, L. A., Tsay, J. M., Doose, S., Li, J. J., Sundaresan, G., Wu, A. M., Gambhir, S. S., and Weiss, S. (2005) Quantum dots for live cells, in vivo imaging, and diagnostics. Science 307, 538−544. (10) Dahan, M., Levi, S., Luccardini, C., Rostaing, P., Riveau, B., and Triller, A. (2003) Diffusion dynamics of glycine receptors revealed by single-quantum dot tracking. Science 302, 442−445. (11) Su, Y., Hu, M., Fan, C., He, Y., Li, Q., Li, W., Wang, L. H., Shen, P., and Huang, Q. (2010) The cytotoxicity of CdTe quantum dots and the relative contributions from released cadmium ions and nanoparticle properties. Biomaterials 31, 4829−34. (12) Soenen, S. J., Demeester, J., De Smedt, S. C., and Braeckmans, K. (2012) The cytotoxic effects of polymer-coated quantum dots and restrictions for live cell applications. Biomaterials 33, 4882−4888. (13) Rzigalinski, B. A., and Strobl, J. S. (2009) Cadmium-containing nanoparticles: Perspectives on pharmacology and toxicology of quantum dots. Toxicol. Appl. Pharmacol. 238, 280−288. (14) Rivera-Gil, P., De Aberasturi, D. J., Wulf, V., Pelaz, B., Del Pino, P., Zhao, Y. Y., De La Fuente, J. M., De Larramendi, I. R., Rojo, T., Liang, X. J., and Parak, W. J. (2013) The challenge to relate the physicochemical properties of colloidal nanoparticles to their cytotoxicity. Acc. Chem. Res. 46, 743−749. (15) Ye, L., Yong, K.-T., Liu, L., Roy, I., Hu, R., Zhu, J., Cai, H., Law, W.-C., Liu, J., Wang, K., Liu, J., Liu, Y., Hu, Y., Zhang, X., Swihart, M. T., and Prasad, P. N. (2012) A pilot study in non-human primates shows no adverse response to intravenous injection of quantum dots. Nat. Nanotechnol. 7, 453−458. (16) Sealy, C. (2012) Primates show mixed response to quantum dots. Nano Today 7, 223−224. (17) Soenen, S. J., Demeester, J., De Smedt, S. C., and Braeckmans, K. (2013) Turning a frown upside down: Exploiting nanoparticle toxicity for anticancer therapy. Nano Today 8, 121−125. (18) Tsoi, K. M., Dai, Q., Alman, B. A., and Chan, W. C. (2012) Are quantum dots toxic? Exploring the discrepancy between cell culture and animal studies. Acc. Chem. Res. 46, 662−671. (19) Joris, F., Manshian, B. B., Peynshaert, K., De Smedt, S. C., Braeckmans, K., and Soenen, S. J. (2013) Assessing nanoparticle toxicity in cell-based assays: influence of cell culture parameters and optimized models for bridging the in vitro-in vivo gap. Chem. Soc. Rev. 42, 8339−8359.

ASSOCIATED CONTENT

S Supporting Information *

Elemental analysis and transmission electron micrographs; dynamic light scattering plots of the nanoparticles used; quantum dot emission spectra; cellular uptake of the quantum dots in C17.2 and PC12 cells; effects of the quantum dots on viability, ROS, DNA damage, calcium levels, and mitochondrial health in C17.2 and PC12 cells; and full experimental methodology. This material is available free of charge via the Internet at http://pubs.acs.org.

■

Article

AUTHOR INFORMATION

Corresponding Authors

*(S.J.S.) E-mail: [email protected]. *(S.C.D.) E-mail: [email protected]. Funding

S.J.S. is a postdoctoral fellow from FWO-Vlaanderen. We acknowledge FWO-Vlaanderen for financial support. Notes

The authors declare no competing financial interest.

■

ABBREVIATIONS DLS, dynamic light scattering; DMEM, Dulbecco’s modified Eagle’s medium; EDX, energy dispersive X-ray spectroscopy; HUVEC, human umbilical vein endothelial cell; GFP, green fluorescent protein; MPA, 3-mercaptopropionic acid; NAC, Nacetylcysteine; NIR, near-infrared; NMs, nanomaterials; NPs, nanoparticles; PBS, phosphate buffered saline; PLQY, photoluminescence quantum yield; QDots, quantum dots; QDotZnSe, ZnSe/ZnS quantum dots; QDotInP, InP/ZnS quantum dots; ROS, reactive oxygen species; SEM, standard error to the mean; TEM, transmission electron microscopy 1058

dx.doi.org/10.1021/tx5000975 | Chem. Res. Toxicol. 2014, 27, 1050−1059

Chemical Research in Toxicology

Article

(20) Mandal, G., Darragh, M., Wang, Y. A., and Heyes, C. D. (2013) Cadmium-free quantum dots as time-gated bioimaging probes in highly-autofluorescent human breast cancer cells. Chem. Commun. 49, 624−626. (21) Moret, S., Becue, A., and Champod, C. (2013) Cadmium-free quantum dots in aqueous solution: Potential for fingermark detection, synthesis and an application to the detection of fingermarks in blood on non-porous surfaces. Forensic Sci. Int. 224, 101−110. (22) Xie, R. G., Chen, K., Chen, X. Y., and Peng, X. G. (2008) InAs/ InP/ZnSe core/shell/shell quantum dots as near-infrared emitters: Bright, narrow-band, non-cadmium containing, and biocompatible. Nano Res. 2008 (1), 457−464. (23) Pons, T., Pic, E., Lequeux, N., Cassette, E., Bezdetnaya, L., Guillemin, F., Marchal, F., and Dubertret, B. (2010) Cadmium-free CuInS2/ZnS quantum dots for sentinel lymph node imaging with reduced toxicity. ACS Nano 4, 2531−2538. (24) Yong, K. T., Roy, I., Hu, R., Ding, H., Cai, H. X., Zhu, J., Zhang, X. H., Bergey, E. J., and Prasad, P. N. (2010) Synthesis of ternary CuInS2/ZnS quantum dot bioconjugates and their applications for targeted cancer bioimaging. Integr. Biol. 2, 121−129. (25) Brunetti, V., Chibli, H., Fiammengo, R., Galeone, A., Malvindi, M. A., Vecchio, G., Cingolani, R., Nadeau, J. L., and Pompa, P. P. (2013) InP/ZnS as a safer alternative to CdSe/ZnS core/shell quantum dots: in vitro and in vivo toxicity assessment. Nanoscale 5, 307−317. (26) Soenen, S. J., Rivera-Gil, P., Montenegro, J. M., Parak, W. J., De Smedt, S. C., and Braeckmans, K. (2011) Cellular toxicity of inorganic nanoparticles: Common aspects and guidelines for improved nanotoxicity evaluation. Nano Today 6, 446−465. (27) Soenen, S. J., De Backer, L., Manshian, B., Doak, S., Raemdonck, K., Demeester, J., Braeckmans, K., and De Smedt, S. (2014) Unraveling the effects of siRNA carrier systems on cell physiology: a multiparametric approach demonstrated on dextran nanogels. Nanomedicine (London, U. K.) 9, 61−76. (28) Soenen, S. J., Montenegro, J. M., Abdelmonem, A. M., Manshian, B. B., Doak, S. H., Parak, W. J., De Smedt, S. C., and Braeckmans, K. (2014) The effect of nanoparticle degradation on poly(methacrylic acid)-coated quantum dot toxicity: the importance of particle functionality assessment in toxicology. Acta Biomater. 10, 732− 741. (29) Soenen, S. J., Manshian, B., Doak, S. H., De Smedt, S. C., and Braeckmans, K. (2013) Fluorescent non-porous silica nanoparticles for long-term cell monitoring: Cytotoxicity and particle functionality. Acta Biomater. 9, 9183−9193. (30) Soenen, S. J., Manshian, B., Montenegro, J. M., Amin, F., Meermann, B., Thiron, T., Cornelissen, M., Vanhaecke, F., Doak, S., Parak, W. J., De Smedt, S., and Braeckmans, K. (2012) Cytotoxic effects of gold nanoparticles: A multiparametric study. ACS Nano 6, 5767−5783. (31) Berman, S. M. C., Kshitiz, Wang, C. J., Orukari, I., Levchenko, A., Bulte, J. W. M., and Walczak, P. (2013) Cell motility of neural stem cells is reduced after SPIO-labeling, which is mitigated after exocytosis. Magn. Reson. Med. 69, 255−262. (32) Ruan, G., Agrawal, A., Marcus, A. I., and Nie, S. (2007) Imaging and tracking of tat peptide-conjugated quantum dots in living cells: new insights into nanoparticle uptake, intracellular transport, and vesicle shedding. J. Am. Chem. Soc. 129, 14759−14766. (33) Laurent, S., Burtea, C., Thirifays, C., Hafeli, U. O., and Mahmoudi, M. (2012) Crucial ignored parameters on nanotoxicology: The importance of toxicity assay modifications and “cell vision”. PLoS One 7, 306−314. (34) Chen, N., He, Y., Su, Y. Y., Li, X. M., Huang, Q., Wang, H. F., Zhang, X. Z., Tai, R. Z., and Fan, C. H. (2012) The cytotoxicity of cadmium-based quantum dots. Biomaterials 33, 1238−1244. (35) Summers, H. D., Rees, P., Holton, M. D., Brown, M. R., Chappell, S. C., Smith, P. J., and Errington, R. J. (2011) Statistical analysis of nanoparticle dosing in a dynamic cellular system. Nat. Nanotechnol. 6, 170−174.

(36) Setyawati, M. I., Fang, W. R., Chia, S. L., and Leong, D. T. (2013) Nanotoxicology of common metal oxide based nanomaterials: their ROS-y and non-ROS-y consequences. Asia-Pac. J. Chem. Eng. 8, 205−217. (37) George, S., Xia, T., Rallo, R., Zhao, Y., Ji, Z., Lin, S., Wang, X., Zhang, H., France, B., Schoenfeld, D., Damoiseaux, R., Liu, R., Lin, S., Bradley, K. A., Cohen, Y., and Nel, A. E. (2011) Use of a highthroughput screening approach coupled with in vivo zebrafish embryo screening to develop hazard ranking for engineered nanomaterials. ACS Nano 5, 1805−1817. (38) Ng, K. W., Khoo, S. P., Heng, B. C., Setyawati, M. I., Tan, E. C., Zhao, X., Xiong, S., Fang, W., Leong, D. T., and Loo, J. S. (2011) The role of the tumor suppressor p53 pathway in the cellular DNA damage response to zinc oxide nanoparticles. Biomaterials 32, 8218−8225. (39) Setyawati, M. I., Tay, C. Y., and Leong, D. T. (2013) Effect of zinc oxide nanomaterials-induced oxidative stress on the p53 pathway. Biomaterials 34, 10133−10142. (40) Diaz, B., Sanchez-Espinel, C., Arruebo, M., Faro, J., de Miguel, E., Magadan, S., Yague, C., Fernandez-Pacheco, R., Ibarra, M. R., Santamaria, J., and Gonzalez-Fernandez, A. (2008) Assessing methods for blood cell cytotoxic responses to inorganic nanoparticles and nanoparticle aggregates. Small 4, 2025−2034. (41) Lamkowski, A., Forcheron, F., Agay, D., Ahmed, E. A., Drouet, M., Meineke, V., and Scherthan, H. (2014) DNA damage focus analysis in blood samples of minipigs reveals acute partial body irradiation. PLoS One 9, e87458. (42) Tang, M. L., Wang, M., Xing, T. R., Zeng, J., Wang, H. L., and Ruan, D. Y. (2008) Mechanisms of unmodified CdSe quantum dotinduced elevation of cytoplasmic calcium levels in primary cultures of rat hippocampal neurons. Biomaterials 29, 4383−4391. (43) Soenen, S. J., Himmelreich, U., Nuytten, N., and De Cuyper, M. (2011) Cytotoxic effects of iron oxide nanoparticles and implications for safety in cell labelling. Biomaterials 32, 195−205. (44) Yang, J. L. (2014) Comparative acute toxicity of gallium(III), antimony(III), indium(III), cadmium(II), and copper(II) on freshwater swamp shrimp (Macrobrachium nipponense). Biol. Res. 47, 13. (45) Lin, R. H., Yang, M. L., Li, Y. C., Chang, H. M., and Kuan, Y. H. (2013) Indium chloride-induced micronuclei via reactive oxygen species in chinese hamster lung fibroblast V79 cells. Environ. Toxicol. 28, 595−600. (46) Matsui, H., Oyama, T. M., Okano, Y., Hashimoto, E., Kawanai, T., and Oyama, Y. (2010) Low micromolar zinc exerts cytotoxic action under H2O2-induced oxidative stress: Excessive increase in intracellular Zn2+ concentration. Toxicology 276, 27−32. (47) Tay, C. Y., Cai, P. Q., Setyawati, M. I., Fang, W. R., Tan, L. P., Hong, C. H. L., Chen, X. D., and Leong, D. T. (2014) Nanoparticles strengthen intracellular tension and retard cellular migration. Nano Lett. 14, 83−88. (48) Pisanic, T. R., Blackwell, J. D., Shubayev, V. I., Finones, R. R., and Jin, S. (2007) Nanotoxicity of iron oxide nanoparticle internalization in growing neurons. Biomaterials 28, 2572−2581. (49) Kirchner, C., Liedl, T., Kudera, S., Pellegrino, T., Javier, A. M., Gaub, H. E., Stolzle, S., Fertig, N., and Parak, W. J. (2005) Cytotoxicity of colloidal CdSe and CdSe/ZnS nanoparticles. Nano Lett. 5, 331− 338.

1059

dx.doi.org/10.1021/tx5000975 | Chem. Res. Toxicol. 2014, 27, 1050−1059