Article pubs.acs.org/est
TiO2 Nanoparticles Act As a Carrier of Cd Bioaccumulation in the Ciliate Tetrahymena thermophila Wei-Wan Yang,† Ying Wang,† Bin Huang, Ning-Xin Wang, Zhong-Bo Wei, Jun Luo, Ai-Jun Miao,* and Liu-Yan Yang State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, Nanjing, Jiangsu Province 210046, China S Supporting Information *
ABSTRACT: When nanoparticles can enter a unicellular organism directly, how may they affect the bioaccumulation and toxicity of other pollutants already present in the environment? To answer this question, we conducted experiments with a protozoan Tetrahymena thermophila. The well-dispersed polyacrylate-coated TiO2 nanoparticles (PAATiO2−NPs) were used as a representative nanomaterial, and Cd as a conventional pollutant. We found that PAA-TiO2− NPs could get into Tetrahymena cells directly. Such internalization was first induced by low concentrations of Cd, but later suppressed when Cd concentrations were higher than 1 μg/L. Considering its significant adsorption on PAA-TiO2−NPs, Cd could be taken up by T. thermophila in the form of free ion or metal-nanoparticle complexes. The latter route accounted for 46.3% of Cd internalization. During the 5 h depuration period, 4.34−22.1% of Cd was excreted out, which was independent of the concentrations of intracellular Cd and PAA-TiO2−NPs. On the other hand, both free and intracellular Cd concentrations only partly predicted its toxicity at different levels of PAA-TiO2− NPs. This may have resulted from PAA-TiO2−NPs’ synergistic effects and the distinct subcellular distribution of Cd taken up via the two routes above. Overall, we should pay attention to the carrier effects of nanoparticles when assessing their environmental risks.
■
INTRODUCTION With the rapid advancement of nanotechnology, a substantial amount of nanoparticles will inevitably find their way into the aquatic environment. Recent years have thus witnessed the intensive exploration of nanoparticles’ adverse effects on aquatic organisms at different trophic levels.1−3 Besides nanoparticles, various pollutants are widely present in the environment. Hence, it is important to reveal how nanoparticles may influence the behavior, effects, and fate of these conventional pollutants. Nanoparticle effects on the bioavailability and toxicity of metals, metalloids, and organic pollutants to a variety of aquatic animals (e.g., zooplankton, fish, and benthic organisms) have been examined.4−8 In most cases, nanoparticles can be ingested by these multicellular organisms and thus facilitate the bioaccumulation of the pollutants adsorbed on the nanoparticle surface. Once inside the intestines, these pollutants may partly get dissociated, move to other tissues, and thereby their toxicity is exacerbated. By contrast, unicellular organisms have a completely different story. In our previous studies,9,10 titanium dioxide nanoparticles (TiO2−NPs) with different coatings could adsorb Cd from the experimental medium and diminish its free ion concentration. These TiO2−NPs cannot enter the cells of the green alga Chlamydomonas reinhardtii. Hence, Cd accumulation and © 2014 American Chemical Society
toxicity in this alga are reduced. Nevertheless, Cd toxicity could still be well predicted by the conventional free ion activity model (FIAM).11 Similar results were reported by Hartmann et al.12 for one of the TiO2−NPs they tested. However, algal growth inhibition is greater than what can be explained by the concentration of dissolved Cd for the other TiO2−NPs. Besides nanoparticles’ physicochemical features, their effects on metal bioavailability also appear to be cell-specific. For instance, carboxyl-CdSe/ZnS quantum dots elevate Cu and Pb accumulation in a wall-less strain of C. reinhardtii.13 But its wall-containing counterpart displays an inverse pattern. This phenomenon is explained by the observations that quantum dots are attached to the surface of the wall-less strain, but not to the walled one. Then the authors concluded that nanoparticlecell interactions play a critical role in metal bioaccumulation. Following this logic, further studies are warranted to demonstrate how metal bioavailability and toxicity may vary if the nanoparticles can enter the cells directly. Received: Revised: Accepted: Published: 7568
February 10, 2014 May 21, 2014 June 9, 2014 June 9, 2014 dx.doi.org/10.1021/es500694t | Environ. Sci. Technol. 2014, 48, 7568−7575
Environmental Science & Technology
Article
showed that this concentration of EDTA has negligible effects on T. thermophila. Then the intracellular concentrations of Cd ([Cd]intra) and PAA-TiO2−NPs ([TiO2−NPs]intra) were quantified by graphite furnace atomic absorption spectrophotometry (Thermo Fisher Scientific Inc., Waltham, MA).9 Whenever applicable, [Cd]med and [TiO2−NPs]med were also measured before the inoculation of T. thermophila. Further, 10 mL aliquot from each replicate was ultrafiltered through a 10 kDa (kD) PALL Nanosep centrifugal device at the beginning and the end of each toxicity test. The amount of Cd retained by the membrane represents its adsorption on PAA-TiO2−NPs. According to the Cd concentration in the 0.99) with [Cd]med at each level of [TiO2−NPs]med. Despite the association of Cd with PAATiO2−NPs, there was still a linear correlation (p < 0.05, r2 = 0.99) between [Cd]intra and [Cd2+]F (Figure 1b). Further, [Cd]intra was higher at higher [TiO2−NPs]med, as was opposite to FIAM. In this model, metal bioaccumulation is considered to be determined by its free ion concentration only. Therefore, PAA-TiO2−NPs might serve as a carrier of Cd bioaccumulation besides their ability to reduce [Cd2+]F. This hypothesis was later supported by the noteworthy internalization of PAATiO2−NPs into the cells (Figure 2g, h). After 24 h exposure to 1.3 (13.2) mg-Ti/L PAA-TiO2−NPs, [TiO2−NPs]intra ranged from 5.1 (79.5) to 7.5 (101.5) pg-Ti/cell in different Cd concentration treatments (Figure 1c). On the other hand, both PAA-TiO2−NPs and Cd suppressed the growth of T. thermophila (Figure 1d). Its μ was 0.8, 0.5, and 0.1 d−1, respectively, in the control treatments containing 0, 1.3, and 13.2 mg-Ti/L PAA-TiO2−NPs. Afterward, μ descended consecutively to 0.1, 0, and 0.02 d−1 when [Cd]med reached its
Figure 1. Linear correlations (a) between the free ([Cd2+]F) and total Cd concentration ([Cd]med) in the medium, and (b) between the intracellular Cd concentration ([Cd]intra) and [Cd2+]F in the three toxicity tests with the addition of 0, 1.3, and 13.2 mg-Ti/L PAA-TiO2− NPs, respectively; (c) intracellular concentration of PAA-TiO2−NPs ([TiO2−NPs]intra) and (d) the cell-specific growth rate μ of Treatment A−H in the toxicity tests above. [Cd]med was 0, 200, 500, 800, 1000, 1200, 1500, and 2000 μg/L for Treatment A−H. Data are mean ± standard deviation (n = 2).
Figure 2. (a, d) The differential interference contrast (DIC) and (b, e) fluorescence (excitation 488 nm, emission 519 nm) images of T. thermophila pre-exposed to 1 mg/L Cd with (d, e) or without (a, b) the addition of 13.2 mg-Ti/L PAA-TiO2−NPs for 2 h as obtained by confocal laser scanning microscopy; (c, f) combined images of (a) and (b) or (d) and (e); (g) brightfield microscope image and (h) subcellular distribution of Ti for the 2 μm thick cell slice of T. thermophila pre-exposed to 1 mg/L Cd and 13.2 mg-Ti/L PAA-TiO2− NPs for 2 h. The image of (h) was achieved through STXM and Ti signal was highlighted by yellow color.
maximum. Although the adverse effects of Cd or TiO2−NPs on various aquatic organisms have been widely investigated, their toxicity to the ciliate T. thermophila was scarcely studied. The median lethal concentration of Cd is 0.2 mg/L for T. thermophila SB196927 and 0.5 mg/L for Tetrahymena sp. RT228 based on live/dead cell viability assays. Both values were 7570
dx.doi.org/10.1021/es500694t | Environ. Sci. Technol. 2014, 48, 7568−7575
Environmental Science & Technology
Article
lower than the median growth inhibition concentration (EC50 = 1.2 mg/L) of Cd we observed herein. Such discrepancy might result from the different toxicity end points, ciliate species/ strain, and Cd speciation of these studies. In the only study about bare TiO 2 −NP toxicity to the same clone of Tetrahymena,15 significant growth inhibition was observed with a 20-h EC50 of over 60 mg-Ti/L. In this situation, PAA-TiO2−NPs with their EC50 between 1.3 and 13.2 mg-Ti/ L were more toxic than bare TiO2−NPs. The toxicity dissimilarity between these two TiO2−NPs might be explained by the difference in their dispersibility and surface coating and further by the discrepancy in the toxicity medium. To further reveal the underlying mechanisms of Cd toxicity, the relative change of μ at different Cd levels, as compared to its counterpart in the respective control treatment with the same [TiO2−NPs]med, was calculated. It was then plotted against [Cd2+]F and [Cd]intra (SI, Figure S2). In the former case, Cd was less noxious when no PAA-TiO2−NPs were applied in the toxicity media (SI, Table S2). By contrast, when the Cd dose was expressed as [Cd]intra (SI, Figure S2b), its toxicity discrepancy at various concentrations of PAA-TiO2− NPs was reduced. There was only 4.8-fold difference between the highest and lowest [Cd]intra-EC50 in contrast to 12.2-fold for [Cd2+]F-EC50. Therefore, the distinct Cd toxicity at various [TiO2−NPs]med could only be partly accounted for by its different bioaccumulation. This could be explained by two possibilities. First, PAA-TiO2−NPs might have some synergistic effects on Cd toxicity, as both pollutants could produce ROS intracellularly and cause oxidative stresses to T. thermophila.15,28 The synergistic effects between TiO2−NPs and Cd have already been reported by Hartmann et al.12 They found that Cd toxicity is alleviated in the presence of TiO2−NPs, but is still higher than what is expected by [Cd2+]F. Second, as PAATiO2−NPs were mainly concentrated in the endosomes of T. thermophila (Figure 2h), the subcellular distribution of Cd taken up in the form of metal-nanoparticle complexes might be different from what was absorbed as free Cd2+. So was its toxicity.29 The second possibility was supported by our observation that Cd concentration in endosomes was much higher when it entered the cells mainly in the form of metalnanoparticle complexes instead of free ion (Figure 2a−f). [Cd]intra was comparable in both situations according to the uptake experiments below. Uptake of PAA-TiO2−NPs and Cd. During the 2 h uptake period, both [TiO2−NPs]med and [Cd]med remained constant. A substantial amount of Cd was also adsorbed on PAA-TiO2− NPs, similar to the toxicity experiment above. Then the adsorption data in both the toxicity and uptake experiments were combined and plotted in Figure 3a. The amount of Cd adsorbed on PAA-TiO2−NPs ([Cd]ads‑NP) in both experiments was the same at the same [Cd2+]F. All data points could be fitted to a single Freundlich isotherm ([Cd]ads‑NP = 2.14 × [Cd2+]F0.84, r2 = 0.99). As for the uptake kinetics of PAA-TiO2− NPs, [TiO2−NPs]intra increased linearly with exposure time (SI, Figure S3). Our preliminary experiment with heat-killed cells manifests that PAA-TiO2−NP attachment to the surface of T. thermophila is completed within a few minutes, similar to quantum dots.30 Consequently, the y-intercept of the linear regression between [TiO2−NPs]intra and exposure time represents the amount of PAA-TiO2−NPs adsorbed on the cell surface. It was close to zero, suggesting that [TiO2− NPs]intra was mostly accounted for by their intracellular fraction. Since PAA-TiO2−NPs and T. thermophila were both
Figure 3. (a) Linear correlation between the amount of Cd adsorbed on PAA-TiO2−NPs ([Cd]ads‑NP) and [Cd2+]F in the uptake and toxicity experiments, respectively. [Cd]med in both experiments is listed in SI, Table S1. [TiO2−NPs]med was 0.4, 1.3, 4.0, and 13.2 mg-Ti/L (0.4, 1.3, 4.0, and 13.2) in the uptake experiment and was 1.3 and 13.2 mg-Ti/L (1.3-T and 13.2-T) in the toxicity experiment. (b) Linear correlations between PAA-TiO2−NP uptake rate and its ambient concentration [TiO2−NPs]med in the uptake experiment with different [Cd]med (0, 0.1, 1, 10, and 30 μg/L). Arrows indicate the increase (up arrow) or decrease (down arrow) of PAA-TiO2−NP uptake rate with [Cd]med. (c) Linear correlation between the cell-surface-adsorbed Cd ([Cd]ads‑cell) and [Cd2+]F, and (d) the hyperbolic correlations between the rate of Cd uptake via the first pathway (VCd‑first) and [Cd2+]F in the uptake experiment with the addition of 0, 0.4, 1.3, 4.0, and 13.2 mgTi/L PAA-TiO2−NPs. Solid line in (d) is the maximum flux of free Cd2+ to the cell surface (J*Cd) at different [Cd2+]F, as calculated from eq 3. Data are mean ± standard deviation (n = 2).
negatively charged, and electrostatic repulsion was dominant, the nanoparticles thus had a low tendency to attach to the cells. Moreover, PAA-TiO2−NP uptake rate, as the slope of each regression line, increased proportionally with [TiO2−NPs]med (Figure 3b). No saturation was reached even when [TiO2− NPs]med was as high as 13.2 mg-Ti/L. In view of the fact that PAA-TiO2−NPs had an average particle size of only 37.6 nm, pinocytosis was their major route to enter the ciliate cells.31 If each aggregate was internalized separately, then approximately 0.1−8.3% of the cell membrane was invaginated every minute to take up PAA-TiO2−NPs at the rate of 0.2−17.4 pg-Ti/cell/ h. The invagination rate we estimated herein was much higher than that of the mixotrophic alga Ochromonas danica (0.015− 0.067%).30 But it was comparable to that of the ciliate Euplotes, which could produce food vacuoles equivalent to 50−150% of its total cell surface area within 5−10 min.32 Further, the clearance rate of PAA-TiO2−NPs (i.e., the slope of the linear correlation between their uptake rate and [TiO2−NPs]med) was in the range of 0.4 × 10−6−1.4 × 10−6 mL/cell/h. The uptake of different-sized polystyrene microspheres (1.1−6.2 μm) by Tetrahymena pyriformis was previously investigated.33 Their clearance rate (5.3 × 10−5−2.1 × 10−4 mL/cell/h) depends on the particle size and is approximately one or 2 orders of magnitude higher than what was achieved herein. Such discrepancy might result from the different uptake routes for nano (pinocytosis, PAA-TiO2−NPs) and micrometer-sized (phagocytosis, polystyrene microsphere) particles. On the other hand, PAA-TiO2−NP uptake rate went up with [Cd]med in each of the four nanoparticle concentration treatments first and declined afterward when [Cd]med was 7571
dx.doi.org/10.1021/es500694t | Environ. Sci. Technol. 2014, 48, 7568−7575
Environmental Science & Technology
Article
Table 1. Values of VmaxCd‑first (fg/cell/h) and Km (μg/L) as Obtained by the Michaelis-Menton Modeling of the Hyperbolic Correlations Between VCd‑first and [Cd2+]F in the Uptake Experiment with the Addition of 0, 0.4, 1.3, 4.0, and 13.2 mg-Ti/L PAA-TiO2−NPsa PAA-TiO2−NP concentration (mg-Ti/L) VmaxCd‑first Km a
0
0.4
1.3
4.0
13.2
709.7 ± 51.1 467.3 ± 75.8
42.2 ± 3.4 29.1 ± 3.5
28.7 ± 2.5 1.8 ± 0.5
15.1 ± 0.4 1.1 ± 0.09
9.2 ± 0.3 0.11 ± 0.02
Data are mean ± standard deviation (n = 2).
higher than 1 μg/L (Figure 3b). The inducive effects of Cd at low concentrations might result from their ability to partly substitute intracellular Ca, which plays an important role in endocytosis.34 On the contrary, when its concentration was high enough, Cd could impair the acidification of endosomes by reducing the activity or the amount of their ATPases.35,36 Accordingly, the binding capacity of the receptors on the cell surface diminished, and the endocytosis rate was lowered. Diffusion from the bulk solution to the cell surface is a critical step in the bioaccumulation of various pollutants.37 Regarding PAA-TiO2−NPs as rigid spheres, their diffusion coefficient (DNP, 1.5 × 10−11 m2/s) could then be calculated from the Stokes−Einstein equation below,
D NP =
kBT 6πηsRH
adsorbed on the cell surface and this process was suppressed by PAA-TiO2−NPs. Nevertheless, the concentration of cellsurface-adsorbed Cd ([Cd]ads‑cell) was comparable at similar [Cd2+]F regardless of [TiO2−NPs]med (Figure 3c). Further considering the negligible attachment of PAA-TiO2−NPs on T. thermophila, the decreased Cd adsorption was thus mainly caused by the lowered [Cd2+]F in the presence of PAA-TiO2− NPs. As the slope of the linear regression between [Cd]cell and exposure time, Cd uptake rate (VCd‑T) went up hyperbolically with [Cd2+]F at each level of PAA-TiO2−NPs (SI, Figure S5a). Moreover, its value was significantly (p < 0.05) enhanced with the increase in [TiO2−NPs]med. This phenomenon disagreed with the concept of FIAM and further supports our hypothesis that PAA-TiO 2 −NPs might serve as a carrier of Cd bioaccumulation. Since PAA-TiO2−NPs can enter the cells of T. thermophila directly, Cd might be taken up in three ways. Namely, it can be internalized in the form of free Cd2+ directly from the bulk solution (first route) or in the form of metalnanoparticle complexes (second route). Alternatively, although not all constants for Cd complexation with the TiO2 core and the PAA shell are well-known, the Cd-PAA-TiO2−NP complex is speculated to be labile (i.e., lability ≫1).40 Hence, Cd2+ dissociated from Cd-PAA-TiO2−NPs during the diffusion of these complexes from the bulk medium to the cell surface would also have a significant contribution to Cd uptake (third route, dynamic process).41 Based on PAA-TiO2−NP uptake rate and their Cd adsorption, we then calculated the rate of Cd uptake via the second route (VCd‑second, fg/cell/h). It was linearly correlated to VCd‑T (SI, Figure S5b). Accordingly, 46.3% of Cd uptake was accomplished through this pathway in different metal and nanoparticle concentration treatments. To further investigate the potential contribution of the third pathway, the maximum flux of free Cd2+ to the cell surface (J*Cd) was calculated (solid line in Figure 3d) as follows,38,42
(1)
where kB (m2 kg/s2/K) means the Boltzmann constant, T (K) represents the absolute temperature, ηs (kg/s/m) is the solvent viscosity, and RH (m) indicates the hydrodynamic radius of the particle as determined by DLS. Then the maximum diffusion of PAA-TiO2−NPs to the cell surface (J*NP, g-Ti/cell/s) was obtained as follows,37 * = D NP × S × ([TiO2 − NPs]med − [TiO2 − NPs]p ) JNP ×
⎛1 1⎞ ⎜ + ⎟ ⎝R δ⎠
(2)
where S (m2 cell−1) is the average cell surface area, [TiO2− NPs]p (mg-Ti/L) represents PAA-TiO2−NP concentration in the cell periphery, R (1.5 × 10−5 m) and δ (2.0 × 10−5 m) signify the radius and the diffusion layer thickness of T. thermophila, respectively.38 Assuming that nanoparticle uptake by the cells was rather quick (i.e., [TiO2−NPs]p = 0 mg-Ti/L), then J*NP was in the range of 7.5 × 10−16 − 2.5 × 10−14 g-Ti/ cell/s (i.e., 2.7−90.5 pg-Ti/cell/h). It was only 2.9−15.1 times higher than the actual uptake rate (0.2−17.4 pg-Ti/cell/h) of PAA-TiO2−NPs. Further, the value of J*NP derived from eq 2 was even overestimated considering the electrostatic repulsion between the negatively charged cells and PAA-TiO2−NPs. Under this condition, the difference between J*NP and the actual uptake rate of PAA-TiO2−NPs would be further lowered. Therefore, diffusion from the bulk solution to the cell surface determined the rate of PAA-TiO 2 −NP uptake by T. thermophila.39 Similarly, quantum dot uptake by O. danica was also found to be diffusion-limited in our previous study.30 We further proposed that nanoparticle uptake tends to be restricted by their diffusion to the cell surface unless their particle size is smaller than 10 nm based on the equations above. Similar to PAA-TiO2−NPs, [Cd]cell also increased linearly with exposure time in the uptake experiment (SI, Figure S4). However, a substantial amount of Cd (the y-intercept) was
* = DCd × S × [Cd2 +]F × JCd
⎛1 1⎞ ⎜ + ⎟ ⎝R δ⎠
(3)
Where DCd (7 × 10−10 m2/s) is the diffusion coefficient of Cd2+.40 The value of J*Cd was much higher than the difference between VCd‑T and VCd‑second (i.e., the rate of total Cd uptake via the first and the third routes). This phenomenon indicates that the turnover rate of Cd transport sites was slow and the potentially labile Cd-PAA-TiO2−NP complexes were not required even though they could theoretically contribute to Cd uptake. Namely, the third pathway had a negligible contribution to the overall bioaccumulation of Cd. Under this condition, the difference between VCd‑T and VCd‑second is defined as VCd‑first (fg/cell/h, uptake via the first route), which should have conformed to FIAM in theory. However, VCd‑first at similar [Cd2+]F differed significantly (p < 0.05) from each other at the five levels of PAA-TiO2−NPs (Figure 3d). The biphasic 7572
dx.doi.org/10.1021/es500694t | Environ. Sci. Technol. 2014, 48, 7568−7575
Environmental Science & Technology
Article
correlation between VCd‑first and [Cd2+]F was then simulated by the Michaelis−Menten equation for each [TiO2−NPs]med as follows, VCd ‐ first =
max 2+ VCd ‐ first × [Cd ]F
K m + [Cd2 +]F
(4)
where VmaxCd‑first (fg/cell/h) represents the maximum uptake rate of Cd via the first pathway, Km (μg/L) is the Michaelis− Menten constant and indicates the binding affinity between Cd and its transporter on the cell surface. Values of both parameters thus obtained are listed in Table 1, and they both declined with the enhancement in [TiO2−NPs]med. It implies that the amount of Cd transporters on the cell surface decreased, but their metal binding affinity was reinforced in the presence of PAA-TiO2−NPs. According to our previous calculation, approximately 0.10−8.27% of the plasma membrane was invaginated by T. thermophila every minute. Moreover, the invagination rate was higher, and the cell membrane was refreshed more quickly at higher [TiO2− NPs]med. The rate of plasma membrane recycling, as depended on [TiO2−NPs]med, might thus play a critical role in metal association onto the cell surface and in its internalization. Nevertheless, more evidence is required to illuminate the underlying mechanisms about nanoparticle impacts on metal uptake via the first route. Efflux of PAA-TiO2−NPs and Cd. T. thermophila grew significantly (p < 0.05) in most treatments of the efflux experiment. Such growth dilution effects on [TiO2−NPs]intra and [Cd]intra were corrected from the calculation of the net efflux. While it seems that [TiO2−NPs]intra diminished with time in two of the nine treatments, this trend was insignificant (p > 0.05) (SI, Figure S6). The proportion of PAA-TiO2−NPs retained in cells fluctuated around 100% during the 5 h depuration period. Based on the few studies about the kinetics of nanoparticle expulsion out of cells (mostly human cells),43−46 nanoparticles can be expelled out quickly with the half-life ranging from a few seconds to several hours. Despite their fast depuration, there is also a substantial amount of internalized nanoparticles, which can not be removed, and their intracellular concentrations level off at the end.46 Moreover, no significant exocytosis of quantum dots was observed for a murine “macrophage-like” cell (J774.A1) within a 2 h period.47 Likewise, the expulsion of quantum dots by a freshwater mixotrophic alga O. danica was also slow, and their intracellular concentration decreased by no more than 10% after 4 h.30 The murine cell, the mixotrophic alga, and the ciliate used in the present study are all able to ingest particles through phagocytosis, as was different from the cells used in other nanoparticle efflux studies. They might thus exhibit a completely different pattern of nanoparticle exocytosis, which needs to be proved in the future. In contrast to PAA-TiO2−NPs, a substantial amount of intracellular Cd (4.3−22.1%) was eliminated by T. thermophila (Figure 4). As mentioned before, Cd was taken up in the form of either free ion directly from the bulk solution or metalnanoparticle complexes. The latter route had a contribution of 46.3% to total Cd uptake (SI, Figure S5b). Additionally, the subcellular distribution of Cd accumulated through these two pathways was different (Figure 2a−f). However, the overall elimination of Cd was independent of [Cd]intra and [TiO2− NPs]intra. Hence, Cd taken up via the second pathway was possibly eliminated in the same rate as the first one. Since there
Figure 4. Proportion of Cd retained in the cells of T. thermophila precultured in the media containing (a) 0.1 and (b) 10 μg/L [Cd]med with the addition of 0, 0.4, 1.3, and 4.0 mg-Ti/L PAA-TiO2−NPs during the 5 h depuration period. Data are mean ± standard deviation (n = 2).
was no significant exocytosis of PAA-TiO2−NPs, the nanoparticle-associated Cd inside the cells cannot be excreted out. Therefore, it could be speculated that part of the nanoparticleadsorbed Cd would get dissociated once inside the acidic vacuoles containing various metal-binding ligands.48 Further, Cd dissociated from the nanoparticle surface was liberated out of the cells even more rapidly than (or at least no lower than) its counterpart accumulated via the first pathway. Based on the findings above and on the results of our previous studies,9,10 we propose that nanoparticles have at least two different kinds of effects on metal bioavailability and toxicity. On one hand, they could adsorb a substantial amount of metal ions on the surface and decrease its free ion concentration in the environment. If the nanoparticles cannot enter the cells as what was observed in Yang et al.,9,10 then its bioaccumulation and toxicity are reduced in agreement with FIAM. On the other hand, when the nanoparticles could get into the cells directly with a significant adsorption of metal ions, they serve as a carrier of metal uptake. In this situation, both metal bioaccumulation and its toxicity would be changed abruptly and could not be well predicted by FIAM. In the natural environment, the free metal ion concentration is determined by a number of factors, and a small variation in one factor is hard to alter this concentration dramatically. Therefore, the carrier effects from nanoparticles may be more important under this condition. What’s more, the dual effects of nanoparticles should not be limited to metal ions only and could be extrapolated to other conventional pollutants already existing in the environment. These facts cannot be ignored when we assess the environmental risks of nanoparticles.
■
ASSOCIATED CONTENT
S Supporting Information *
Detailed information how the subcellular distribution of PAATiO2−NPs and Cd was measured, [Cd]med and [TiO2− NPs]med in different experiments, the [Cd2+]F and [Cd]intrabased EC50s, the physicochemical properties of PAA-TiO2− NPs, dose−response curves of Cd, the uptake kinetics of PAATiO2−NPs and Cd, correlations between VCd‑T and [Cd2+]F as well as between VCd‑T and VCd‑second, and PAA-TiO2−NP efflux are included as Supporting Information. This material is available free of charge via the Internet at http://pubs.acs.org. 7573
dx.doi.org/10.1021/es500694t | Environ. Sci. Technol. 2014, 48, 7568−7575
Environmental Science & Technology
■
Article
(14) Mortimer, M.; Kasemets, K.; Kahru, A. Toxicity of ZnO and CuO nanoparticles to ciliated protozoa Tetrahymena thermophila. Toxicology 2010, 269 (2−3), 182−189. (15) Ud-Daula, A.; Pfister, G.; Schramm, K. W. Method for toxicity test of titanium dioxide nanoparticles in ciliate protozoan Tetrahymena. J. Environ. Sci. Health A 2013, 48 (11), 1343−1348. (16) Larue, C.; Veronesi, G.; Flank, A. M.; Surble, S.; Herlin-Boime, N.; Carriere, M. Comparative uptake and impact of TiO2 nanoparticles in wheat and rapeseed. J. Toxicol. Environ. Health A 2012, 75 (13−15), 722−734. (17) Gorovsky, M. A.; Yao, M. C.; Keevert, J. B.; Pleger, G. L. Isolation of micro- and macronuclei of Tetrahymena pyriformis. Method. Cell Biol. 1975, 9 (0), 311−327. (18) Dryl, S. Antigenic transformation in Paramecium aurelia after homologous antiserum treatment during autogamy and conjugation. J. Protozool. 1959, 6 (3), 25−25. (19) Miao, A. J.; Zhang, X. Y.; Luo, Z.; Chen, C. S.; Chin, W. C.; Santschi, P. H.; Quigg, A. Zinc oxide engineered nanoparticles dissolution and toxicity to marine phytoplankton. Environ. Toxicol. Chem. 2010, 29 (12), 2814−2822. (20) Miao, A. J.; Wang, W. X. Relationships between cell-specific growth rate and uptake rate of cadmium and zinc by a coastal diatom. Mar. Ecol.: Prog. Ser. 2004, 275, 103−113. (21) Wang, N. X.; Li, Y.; Deng, X. H.; Miao, A. J.; Ji, R.; Yang, L. Y. Toxicity and bioaccumulation kinetics of arsenate in two freshwater green algae under different phosphate regimes. Water Res. 2013, 47 (7), 2497−2506. (22) Hassler, C. S.; Slaveykova, V. I.; Wilkinson, K. J. Discriminating between intra- and extracellular metals using chemical extractions. Limnol. Oceanogr. Methods 2004, 2, 237−247. (23) Miao, A. J.; Wang, W. X. Fulfilling iron requirements of a coastal diatom under different temperatures and irradiances. Limnol. Oceanogr. 2006, 51 (2), 925−935. (24) Liufu, S. C.; Mao, H. N.; Li, Y. P. Adsorption of poly(acrylic acid) onto the surface of titanium dioxide and the colloidal stability of aqueous suspension. J. Colloid Interface Sci. 2005, 281 (1), 155−163. (25) Navarro, E.; Baun, A.; Behra, R.; Hartmann, N. B.; Filser, J.; Miao, A. J.; Quigg, A.; Santschi, P. H.; Sigg, L. Environmental behavior and ecotoxicity of engineered nanoparticles to algae, plants, and fungi. Ecotoxicology 2008, 17 (5), 372−386. (26) Kirwan, L. J.; Fawell, P. D.; van Bronswijk, W. An in situ FTIRATR study of polyacrylate adsorbed onto hematite at high pH and high ionic strength. Langmuir 2004, 20 (10), 4093−4100. (27) Gallego, A.; Martin-Gonzalez, A.; Ortega, R.; Gutierrez, J. C. Flow cytometry assessment of cytotoxicity and reactive oxygen species generation by single and binary mixtures of cadmium, zinc and copper on populations of the ciliated protozoan Tetrahymena thermophila. Chemosphere 2007, 68 (4), 647−661. (28) Rico, D.; Martin-Gonzalez, A.; Diaz, S.; de Lucas, P.; Gutierrez, J. C. Heavy metals generate reactive oxygen species in terrestrial and aquatic ciliated protozoa. Comp. Biochem. Phys. C 2009, 149 (1), 90− 96. (29) Miao, A. J.; Wang, W. X. Predicting copper toxicity with its intracellular or subcellular concentration and the thiol synthesis in a marine diatom. Environ. Sci. Technol. 2007, 41 (5), 1777−1782. (30) Wang, Y.; Miao, A. J.; Luo, J.; Wei, Z. B.; Zhu, J. J.; Yang, L. Y. Bioaccumulation of CdTe quantum dots in a freshwater alga Ochromonas danica: A kinetics study. Environ. Sci. Technol. 2013, 47 (18), 10601−10610. (31) Iversen, T. G.; Skotland, T.; Sandvig, K. Endocytosis and intracellular transport of nanoparticles: Present knowledge and need for future studies. Nano Today 2011, 6 (2), 176−185. (32) Kloetzel, J. A. Feeding in ciliated protozoa 0.1. pharyngeal disks in euplotesSource of membrane for food vacuole formation. J. Cell Sci. 1974, 15 (2), 379−401. (33) Lavin, D. P.; Hatzis, C.; Srienc, F.; Fredrickson, A. G. Size effects on the uptake of particles by populations of Tetrahymena pyriformis cells. J. Protozool. 1990, 37 (3), 157−163.
AUTHOR INFORMATION
Corresponding Author
*Phone: +86 25 89680255; fax: +86 25 89680569; e-mail: [email protected]. Author Contributions †
W.W.Y. and Y.W. contributed equally to this work.
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We thank three anonymous reviewers as well as Dr. Qiaoguo Tan and Peter H. Santschi for their constructive suggestions on this paper. The ciliate is courtesy of Dr. Wei Miao. The financial support offered by the National Natural Science Foundation of China (41271486, 41001338, and 21237001) and the Natural Science Foundation of Jiangsu Province (BK2010371) to A. J. Miao have made this work possible.
■
REFERENCES
(1) Moore, M. N. Do nanoparticles present ecotoxicological risks for the health of the aquatic environment? Environ. Int. 2006, 32 (8), 967−976. (2) Fabrega, J.; Luoma, S. N.; Tyler, C. R.; Galloway, T. S.; Lead, J. R. Silver nanoparticles: Behaviour and effects in the aquatic environment. Environ. Int. 2011, 37 (2), 517−531. (3) Hou, W. C.; Westerhoff, P.; Posner, J. D. Biological accumulation of engineered nanomaterials: A review of current knowledge. Environ. Sci.: Processes Impacts 2013, 15 (1), 103−122. (4) Sun, H.; Zhang, X.; Niu, Q.; Chen, Y.; Crittenden, J. C. Enhanced accumulation of arsenate in carp in the presence of titanium dioxide nanoparticles. Water, Air, Soil Pollut. 2007, 178 (1−4), 245−254. (5) Baun, A.; Sorensen, S. N.; Rasmussen, R. F.; Hartmann, N. B.; Koch, C. B. Toxicity and bioaccumulation of xenobiotic organic compounds in the presence of aqueous suspensions of aggregates of nano-C-60. Aquat. Toxicol. 2008, 86 (3), 379−387. (6) Fan, W.; Cui, M.; Liu, H.; Wang, C.; Shi, Z.; Tan, C.; Yang, X. Nano-TiO2 enhances the toxicity of copper in natural water to Daphnia magna. Environ. Pollut. 2011, 159 (3), 729−734. (7) Zhang, W.; Miao, Y.; Lin, K.; Chen, L.; Dong, Q.; Huang, C. Toxic effects of copper ion in zebrafish in the joint presence of CdTe QDs. Environ. Pollut. 2013, 176, 158−164. (8) Tan, C.; Fan, W. H.; Wang, W. X. Role of titanium dioxide nanoparticles in the elevated uptake and retention of cadmium and zinc in Daphnia magna. Environ. Sci. Technol. 2012, 46 (1), 469−476. (9) Miao, A. J.; Wang, N. X.; Yang, L. Y.; Wang, W. X. Accumulation kinetics of arsenic in Daphnia magna under different phosphorus and food density regimes. Environ. Toxicol. Chem. 2012, 31 (6), 1283− 1291. (10) Yang, W. W.; Li, Y.; Miao, A. J.; Yang, L. Y. Cd2+ toxicity as affected by bare TiO2 nanoparticles and their bulk counterpart. Ecotoxicol. Environ. Saf. 2012, 85, 44−51. (11) Campbell, P. G. C., Interactions between trace metals and aquatic organisms: A critique of the free-ion activity model. In Metal Speciation and Bioavailability in Aquatic Systems; Tessier, A., Turner, D. R., Eds.; John Wiley & Sons: Chichester, 1995; pp 45−102. (12) Hartmann, N. B.; Von der Kammer, F.; Hofmann, T.; Baalousha, M.; Ottofuelling, S.; Baun, A. Algal testing of titanium dioxide nanoparticles-Testing considerations, inhibitory effects and modification of cadmium bioavailability. Toxicology 2010, 269 (2−3), 190−197. (13) Worms, I. A. M.; Boltzman, J.; Garcia, M.; Slaveykova, V. I. Cellwall-dependent effect of carboxyl-CdSe/ZnS quantum dots on lead and copper availability to green microalgae. Environ. Pollut. 2012, 167, 27−33. 7574
dx.doi.org/10.1021/es500694t | Environ. Sci. Technol. 2014, 48, 7568−7575
Environmental Science & Technology
Article
(34) Zefirov, A. L.; Abdrakhmanov, M. M.; Mukhamedyarov, M. A.; Grigoryev, P. N. The role of extracellular calcium in exo- and endocytosis of synaptic vesicles at the frog motor nerve terminals. Neuroscience 2006, 143 (4), 905−910. (35) Choi, J. S.; Kim, K. R.; Ahn, D. W.; Park, Y. S. Cadmium inhibits albumin endocytosis in opossum kidney epithelial cells. Toxicol. Appl. Pharmacol. 1999, 161 (2), 146−152. (36) Herak-Kramberger, C. M.; Brown, D.; Sabolic, I. Cadmium inhibits vacuolar H+-ATPase and endocytosis in rat kidney cortex. Kidney Int. 1998, 53 (6), 1713−1726. (37) Buffle, J.; Wilkinson, K. J.; van Leeuwen, H. P. Chemodynamics and bioavailability in natural waters. Environ. Sci. Technol. 2009, 43 (19), 7170−7174. (38) Van Leeuwen, H. P.; Koster, W. Physicochemical Kinetics and Transport at Biointerfaces; John Wiley & Sons Ltd.: West Sussex, 2004. (39) Hudson, R. J. M.; Morel, F. M. M. Iron transport in marine phytoplankton - kinetics of cellular and medium coordination reactions. Limnol. Oceanogr. 1990, 35 (5), 1002−1020. (40) Van Leeuwen, H. P.; Town, R. M.; Buffle, J.; Cleven, R. F. M. J.; Davison, W.; Puy, J.; van Riemsdijk, W. H.; Sigg, L. Dynamic speciation analysis and bioavailability of metals in aquatic systems. Environ. Sci. Technol. 2005, 39 (22), 8545−8556. (41) Van Leeuwen, H. P. Metal speciation dynamics and bioavailability: Inert and labile complexes. Environ. Sci. Technol. 1999, 33 (21), 3743−3748. (42) Kola, H.; Wilkinson, K. J. Cadmium uptake by a green alga can be predicted by equilibrium modelling. Environ. Sci. Technol. 2005, 39 (9), 3040−3047. (43) Jiang, X.; Rocker, C.; Hafner, M.; Brandholt, S.; Dorlich, R. M.; Nienhaus, G. U. Endo- and exocytosis of zwitterionic quantum dot nanoparticles by live HeLa cells. ACS Nano 2010, 4 (11), 6787−6797. (44) Chu, Z.; Huang, Y.; Tao, Q.; Li, Q. Cellular uptake, evolution, and excretion of silica nanoparticles in human cells. Nanoscale 2011, 3 (8), 3291−3299. (45) Chithrani, B. D.; Chan, W. C. W. Elucidating the mechanism of cellular uptake and removal of protein-coated gold nanoparticles of different sizes and shapes. Nano Lett. 2007, 7 (6), 1542−1550. (46) Ohta, S.; Inasawa, S.; Yamaguchi, Y. Real time observation and kinetic modeling of the cellular uptake and removal of silicon quantum dots. Biomaterials 2012, 33 (18), 4639−4645. (47) Clift, M. J. D.; Rothen-Rutishauser, B.; Brown, D. M.; Duffin, R.; Donaldson, K.; Proudfoot, L.; Guy, K.; Stone, V. The impact of different nanoparticle surface chemistry and size on uptake and toxicity in a murine macrophage cell line. Toxicol. Appl. Pharmacol. 2008, 232 (3), 418−427. (48) Nilsson, J. R. On food vacuoles in Tetrahymena pyriformis GL. J. Protozool. 1977, 24 (4), 502−507.
7575
dx.doi.org/10.1021/es500694t | Environ. Sci. Technol. 2014, 48, 7568−7575
TiO2 Nanoparticles Act As a Carrier of Cd Bioaccumulation in the Ciliate Tetrahymena thermophila Wei-Wan Yang,† Ying Wang,† Bin Huang, Ning-Xin Wang, Zhong-Bo Wei, Jun Luo, Ai-Jun Miao,* and Liu-Yan Yang State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, Nanjing, Jiangsu Province 210046, China S Supporting Information *
ABSTRACT: When nanoparticles can enter a unicellular organism directly, how may they affect the bioaccumulation and toxicity of other pollutants already present in the environment? To answer this question, we conducted experiments with a protozoan Tetrahymena thermophila. The well-dispersed polyacrylate-coated TiO2 nanoparticles (PAATiO2−NPs) were used as a representative nanomaterial, and Cd as a conventional pollutant. We found that PAA-TiO2− NPs could get into Tetrahymena cells directly. Such internalization was first induced by low concentrations of Cd, but later suppressed when Cd concentrations were higher than 1 μg/L. Considering its significant adsorption on PAA-TiO2−NPs, Cd could be taken up by T. thermophila in the form of free ion or metal-nanoparticle complexes. The latter route accounted for 46.3% of Cd internalization. During the 5 h depuration period, 4.34−22.1% of Cd was excreted out, which was independent of the concentrations of intracellular Cd and PAA-TiO2−NPs. On the other hand, both free and intracellular Cd concentrations only partly predicted its toxicity at different levels of PAA-TiO2− NPs. This may have resulted from PAA-TiO2−NPs’ synergistic effects and the distinct subcellular distribution of Cd taken up via the two routes above. Overall, we should pay attention to the carrier effects of nanoparticles when assessing their environmental risks.
■
INTRODUCTION With the rapid advancement of nanotechnology, a substantial amount of nanoparticles will inevitably find their way into the aquatic environment. Recent years have thus witnessed the intensive exploration of nanoparticles’ adverse effects on aquatic organisms at different trophic levels.1−3 Besides nanoparticles, various pollutants are widely present in the environment. Hence, it is important to reveal how nanoparticles may influence the behavior, effects, and fate of these conventional pollutants. Nanoparticle effects on the bioavailability and toxicity of metals, metalloids, and organic pollutants to a variety of aquatic animals (e.g., zooplankton, fish, and benthic organisms) have been examined.4−8 In most cases, nanoparticles can be ingested by these multicellular organisms and thus facilitate the bioaccumulation of the pollutants adsorbed on the nanoparticle surface. Once inside the intestines, these pollutants may partly get dissociated, move to other tissues, and thereby their toxicity is exacerbated. By contrast, unicellular organisms have a completely different story. In our previous studies,9,10 titanium dioxide nanoparticles (TiO2−NPs) with different coatings could adsorb Cd from the experimental medium and diminish its free ion concentration. These TiO2−NPs cannot enter the cells of the green alga Chlamydomonas reinhardtii. Hence, Cd accumulation and © 2014 American Chemical Society
toxicity in this alga are reduced. Nevertheless, Cd toxicity could still be well predicted by the conventional free ion activity model (FIAM).11 Similar results were reported by Hartmann et al.12 for one of the TiO2−NPs they tested. However, algal growth inhibition is greater than what can be explained by the concentration of dissolved Cd for the other TiO2−NPs. Besides nanoparticles’ physicochemical features, their effects on metal bioavailability also appear to be cell-specific. For instance, carboxyl-CdSe/ZnS quantum dots elevate Cu and Pb accumulation in a wall-less strain of C. reinhardtii.13 But its wall-containing counterpart displays an inverse pattern. This phenomenon is explained by the observations that quantum dots are attached to the surface of the wall-less strain, but not to the walled one. Then the authors concluded that nanoparticlecell interactions play a critical role in metal bioaccumulation. Following this logic, further studies are warranted to demonstrate how metal bioavailability and toxicity may vary if the nanoparticles can enter the cells directly. Received: Revised: Accepted: Published: 7568
February 10, 2014 May 21, 2014 June 9, 2014 June 9, 2014 dx.doi.org/10.1021/es500694t | Environ. Sci. Technol. 2014, 48, 7568−7575
Environmental Science & Technology
Article
showed that this concentration of EDTA has negligible effects on T. thermophila. Then the intracellular concentrations of Cd ([Cd]intra) and PAA-TiO2−NPs ([TiO2−NPs]intra) were quantified by graphite furnace atomic absorption spectrophotometry (Thermo Fisher Scientific Inc., Waltham, MA).9 Whenever applicable, [Cd]med and [TiO2−NPs]med were also measured before the inoculation of T. thermophila. Further, 10 mL aliquot from each replicate was ultrafiltered through a 10 kDa (kD) PALL Nanosep centrifugal device at the beginning and the end of each toxicity test. The amount of Cd retained by the membrane represents its adsorption on PAA-TiO2−NPs. According to the Cd concentration in the 0.99) with [Cd]med at each level of [TiO2−NPs]med. Despite the association of Cd with PAATiO2−NPs, there was still a linear correlation (p < 0.05, r2 = 0.99) between [Cd]intra and [Cd2+]F (Figure 1b). Further, [Cd]intra was higher at higher [TiO2−NPs]med, as was opposite to FIAM. In this model, metal bioaccumulation is considered to be determined by its free ion concentration only. Therefore, PAA-TiO2−NPs might serve as a carrier of Cd bioaccumulation besides their ability to reduce [Cd2+]F. This hypothesis was later supported by the noteworthy internalization of PAATiO2−NPs into the cells (Figure 2g, h). After 24 h exposure to 1.3 (13.2) mg-Ti/L PAA-TiO2−NPs, [TiO2−NPs]intra ranged from 5.1 (79.5) to 7.5 (101.5) pg-Ti/cell in different Cd concentration treatments (Figure 1c). On the other hand, both PAA-TiO2−NPs and Cd suppressed the growth of T. thermophila (Figure 1d). Its μ was 0.8, 0.5, and 0.1 d−1, respectively, in the control treatments containing 0, 1.3, and 13.2 mg-Ti/L PAA-TiO2−NPs. Afterward, μ descended consecutively to 0.1, 0, and 0.02 d−1 when [Cd]med reached its
Figure 1. Linear correlations (a) between the free ([Cd2+]F) and total Cd concentration ([Cd]med) in the medium, and (b) between the intracellular Cd concentration ([Cd]intra) and [Cd2+]F in the three toxicity tests with the addition of 0, 1.3, and 13.2 mg-Ti/L PAA-TiO2− NPs, respectively; (c) intracellular concentration of PAA-TiO2−NPs ([TiO2−NPs]intra) and (d) the cell-specific growth rate μ of Treatment A−H in the toxicity tests above. [Cd]med was 0, 200, 500, 800, 1000, 1200, 1500, and 2000 μg/L for Treatment A−H. Data are mean ± standard deviation (n = 2).
Figure 2. (a, d) The differential interference contrast (DIC) and (b, e) fluorescence (excitation 488 nm, emission 519 nm) images of T. thermophila pre-exposed to 1 mg/L Cd with (d, e) or without (a, b) the addition of 13.2 mg-Ti/L PAA-TiO2−NPs for 2 h as obtained by confocal laser scanning microscopy; (c, f) combined images of (a) and (b) or (d) and (e); (g) brightfield microscope image and (h) subcellular distribution of Ti for the 2 μm thick cell slice of T. thermophila pre-exposed to 1 mg/L Cd and 13.2 mg-Ti/L PAA-TiO2− NPs for 2 h. The image of (h) was achieved through STXM and Ti signal was highlighted by yellow color.
maximum. Although the adverse effects of Cd or TiO2−NPs on various aquatic organisms have been widely investigated, their toxicity to the ciliate T. thermophila was scarcely studied. The median lethal concentration of Cd is 0.2 mg/L for T. thermophila SB196927 and 0.5 mg/L for Tetrahymena sp. RT228 based on live/dead cell viability assays. Both values were 7570
dx.doi.org/10.1021/es500694t | Environ. Sci. Technol. 2014, 48, 7568−7575
Environmental Science & Technology
Article
lower than the median growth inhibition concentration (EC50 = 1.2 mg/L) of Cd we observed herein. Such discrepancy might result from the different toxicity end points, ciliate species/ strain, and Cd speciation of these studies. In the only study about bare TiO 2 −NP toxicity to the same clone of Tetrahymena,15 significant growth inhibition was observed with a 20-h EC50 of over 60 mg-Ti/L. In this situation, PAA-TiO2−NPs with their EC50 between 1.3 and 13.2 mg-Ti/ L were more toxic than bare TiO2−NPs. The toxicity dissimilarity between these two TiO2−NPs might be explained by the difference in their dispersibility and surface coating and further by the discrepancy in the toxicity medium. To further reveal the underlying mechanisms of Cd toxicity, the relative change of μ at different Cd levels, as compared to its counterpart in the respective control treatment with the same [TiO2−NPs]med, was calculated. It was then plotted against [Cd2+]F and [Cd]intra (SI, Figure S2). In the former case, Cd was less noxious when no PAA-TiO2−NPs were applied in the toxicity media (SI, Table S2). By contrast, when the Cd dose was expressed as [Cd]intra (SI, Figure S2b), its toxicity discrepancy at various concentrations of PAA-TiO2− NPs was reduced. There was only 4.8-fold difference between the highest and lowest [Cd]intra-EC50 in contrast to 12.2-fold for [Cd2+]F-EC50. Therefore, the distinct Cd toxicity at various [TiO2−NPs]med could only be partly accounted for by its different bioaccumulation. This could be explained by two possibilities. First, PAA-TiO2−NPs might have some synergistic effects on Cd toxicity, as both pollutants could produce ROS intracellularly and cause oxidative stresses to T. thermophila.15,28 The synergistic effects between TiO2−NPs and Cd have already been reported by Hartmann et al.12 They found that Cd toxicity is alleviated in the presence of TiO2−NPs, but is still higher than what is expected by [Cd2+]F. Second, as PAATiO2−NPs were mainly concentrated in the endosomes of T. thermophila (Figure 2h), the subcellular distribution of Cd taken up in the form of metal-nanoparticle complexes might be different from what was absorbed as free Cd2+. So was its toxicity.29 The second possibility was supported by our observation that Cd concentration in endosomes was much higher when it entered the cells mainly in the form of metalnanoparticle complexes instead of free ion (Figure 2a−f). [Cd]intra was comparable in both situations according to the uptake experiments below. Uptake of PAA-TiO2−NPs and Cd. During the 2 h uptake period, both [TiO2−NPs]med and [Cd]med remained constant. A substantial amount of Cd was also adsorbed on PAA-TiO2− NPs, similar to the toxicity experiment above. Then the adsorption data in both the toxicity and uptake experiments were combined and plotted in Figure 3a. The amount of Cd adsorbed on PAA-TiO2−NPs ([Cd]ads‑NP) in both experiments was the same at the same [Cd2+]F. All data points could be fitted to a single Freundlich isotherm ([Cd]ads‑NP = 2.14 × [Cd2+]F0.84, r2 = 0.99). As for the uptake kinetics of PAA-TiO2− NPs, [TiO2−NPs]intra increased linearly with exposure time (SI, Figure S3). Our preliminary experiment with heat-killed cells manifests that PAA-TiO2−NP attachment to the surface of T. thermophila is completed within a few minutes, similar to quantum dots.30 Consequently, the y-intercept of the linear regression between [TiO2−NPs]intra and exposure time represents the amount of PAA-TiO2−NPs adsorbed on the cell surface. It was close to zero, suggesting that [TiO2− NPs]intra was mostly accounted for by their intracellular fraction. Since PAA-TiO2−NPs and T. thermophila were both
Figure 3. (a) Linear correlation between the amount of Cd adsorbed on PAA-TiO2−NPs ([Cd]ads‑NP) and [Cd2+]F in the uptake and toxicity experiments, respectively. [Cd]med in both experiments is listed in SI, Table S1. [TiO2−NPs]med was 0.4, 1.3, 4.0, and 13.2 mg-Ti/L (0.4, 1.3, 4.0, and 13.2) in the uptake experiment and was 1.3 and 13.2 mg-Ti/L (1.3-T and 13.2-T) in the toxicity experiment. (b) Linear correlations between PAA-TiO2−NP uptake rate and its ambient concentration [TiO2−NPs]med in the uptake experiment with different [Cd]med (0, 0.1, 1, 10, and 30 μg/L). Arrows indicate the increase (up arrow) or decrease (down arrow) of PAA-TiO2−NP uptake rate with [Cd]med. (c) Linear correlation between the cell-surface-adsorbed Cd ([Cd]ads‑cell) and [Cd2+]F, and (d) the hyperbolic correlations between the rate of Cd uptake via the first pathway (VCd‑first) and [Cd2+]F in the uptake experiment with the addition of 0, 0.4, 1.3, 4.0, and 13.2 mgTi/L PAA-TiO2−NPs. Solid line in (d) is the maximum flux of free Cd2+ to the cell surface (J*Cd) at different [Cd2+]F, as calculated from eq 3. Data are mean ± standard deviation (n = 2).
negatively charged, and electrostatic repulsion was dominant, the nanoparticles thus had a low tendency to attach to the cells. Moreover, PAA-TiO2−NP uptake rate, as the slope of each regression line, increased proportionally with [TiO2−NPs]med (Figure 3b). No saturation was reached even when [TiO2− NPs]med was as high as 13.2 mg-Ti/L. In view of the fact that PAA-TiO2−NPs had an average particle size of only 37.6 nm, pinocytosis was their major route to enter the ciliate cells.31 If each aggregate was internalized separately, then approximately 0.1−8.3% of the cell membrane was invaginated every minute to take up PAA-TiO2−NPs at the rate of 0.2−17.4 pg-Ti/cell/ h. The invagination rate we estimated herein was much higher than that of the mixotrophic alga Ochromonas danica (0.015− 0.067%).30 But it was comparable to that of the ciliate Euplotes, which could produce food vacuoles equivalent to 50−150% of its total cell surface area within 5−10 min.32 Further, the clearance rate of PAA-TiO2−NPs (i.e., the slope of the linear correlation between their uptake rate and [TiO2−NPs]med) was in the range of 0.4 × 10−6−1.4 × 10−6 mL/cell/h. The uptake of different-sized polystyrene microspheres (1.1−6.2 μm) by Tetrahymena pyriformis was previously investigated.33 Their clearance rate (5.3 × 10−5−2.1 × 10−4 mL/cell/h) depends on the particle size and is approximately one or 2 orders of magnitude higher than what was achieved herein. Such discrepancy might result from the different uptake routes for nano (pinocytosis, PAA-TiO2−NPs) and micrometer-sized (phagocytosis, polystyrene microsphere) particles. On the other hand, PAA-TiO2−NP uptake rate went up with [Cd]med in each of the four nanoparticle concentration treatments first and declined afterward when [Cd]med was 7571
dx.doi.org/10.1021/es500694t | Environ. Sci. Technol. 2014, 48, 7568−7575
Environmental Science & Technology
Article
Table 1. Values of VmaxCd‑first (fg/cell/h) and Km (μg/L) as Obtained by the Michaelis-Menton Modeling of the Hyperbolic Correlations Between VCd‑first and [Cd2+]F in the Uptake Experiment with the Addition of 0, 0.4, 1.3, 4.0, and 13.2 mg-Ti/L PAA-TiO2−NPsa PAA-TiO2−NP concentration (mg-Ti/L) VmaxCd‑first Km a
0
0.4
1.3
4.0
13.2
709.7 ± 51.1 467.3 ± 75.8
42.2 ± 3.4 29.1 ± 3.5
28.7 ± 2.5 1.8 ± 0.5
15.1 ± 0.4 1.1 ± 0.09
9.2 ± 0.3 0.11 ± 0.02
Data are mean ± standard deviation (n = 2).
higher than 1 μg/L (Figure 3b). The inducive effects of Cd at low concentrations might result from their ability to partly substitute intracellular Ca, which plays an important role in endocytosis.34 On the contrary, when its concentration was high enough, Cd could impair the acidification of endosomes by reducing the activity or the amount of their ATPases.35,36 Accordingly, the binding capacity of the receptors on the cell surface diminished, and the endocytosis rate was lowered. Diffusion from the bulk solution to the cell surface is a critical step in the bioaccumulation of various pollutants.37 Regarding PAA-TiO2−NPs as rigid spheres, their diffusion coefficient (DNP, 1.5 × 10−11 m2/s) could then be calculated from the Stokes−Einstein equation below,
D NP =
kBT 6πηsRH
adsorbed on the cell surface and this process was suppressed by PAA-TiO2−NPs. Nevertheless, the concentration of cellsurface-adsorbed Cd ([Cd]ads‑cell) was comparable at similar [Cd2+]F regardless of [TiO2−NPs]med (Figure 3c). Further considering the negligible attachment of PAA-TiO2−NPs on T. thermophila, the decreased Cd adsorption was thus mainly caused by the lowered [Cd2+]F in the presence of PAA-TiO2− NPs. As the slope of the linear regression between [Cd]cell and exposure time, Cd uptake rate (VCd‑T) went up hyperbolically with [Cd2+]F at each level of PAA-TiO2−NPs (SI, Figure S5a). Moreover, its value was significantly (p < 0.05) enhanced with the increase in [TiO2−NPs]med. This phenomenon disagreed with the concept of FIAM and further supports our hypothesis that PAA-TiO 2 −NPs might serve as a carrier of Cd bioaccumulation. Since PAA-TiO2−NPs can enter the cells of T. thermophila directly, Cd might be taken up in three ways. Namely, it can be internalized in the form of free Cd2+ directly from the bulk solution (first route) or in the form of metalnanoparticle complexes (second route). Alternatively, although not all constants for Cd complexation with the TiO2 core and the PAA shell are well-known, the Cd-PAA-TiO2−NP complex is speculated to be labile (i.e., lability ≫1).40 Hence, Cd2+ dissociated from Cd-PAA-TiO2−NPs during the diffusion of these complexes from the bulk medium to the cell surface would also have a significant contribution to Cd uptake (third route, dynamic process).41 Based on PAA-TiO2−NP uptake rate and their Cd adsorption, we then calculated the rate of Cd uptake via the second route (VCd‑second, fg/cell/h). It was linearly correlated to VCd‑T (SI, Figure S5b). Accordingly, 46.3% of Cd uptake was accomplished through this pathway in different metal and nanoparticle concentration treatments. To further investigate the potential contribution of the third pathway, the maximum flux of free Cd2+ to the cell surface (J*Cd) was calculated (solid line in Figure 3d) as follows,38,42
(1)
where kB (m2 kg/s2/K) means the Boltzmann constant, T (K) represents the absolute temperature, ηs (kg/s/m) is the solvent viscosity, and RH (m) indicates the hydrodynamic radius of the particle as determined by DLS. Then the maximum diffusion of PAA-TiO2−NPs to the cell surface (J*NP, g-Ti/cell/s) was obtained as follows,37 * = D NP × S × ([TiO2 − NPs]med − [TiO2 − NPs]p ) JNP ×
⎛1 1⎞ ⎜ + ⎟ ⎝R δ⎠
(2)
where S (m2 cell−1) is the average cell surface area, [TiO2− NPs]p (mg-Ti/L) represents PAA-TiO2−NP concentration in the cell periphery, R (1.5 × 10−5 m) and δ (2.0 × 10−5 m) signify the radius and the diffusion layer thickness of T. thermophila, respectively.38 Assuming that nanoparticle uptake by the cells was rather quick (i.e., [TiO2−NPs]p = 0 mg-Ti/L), then J*NP was in the range of 7.5 × 10−16 − 2.5 × 10−14 g-Ti/ cell/s (i.e., 2.7−90.5 pg-Ti/cell/h). It was only 2.9−15.1 times higher than the actual uptake rate (0.2−17.4 pg-Ti/cell/h) of PAA-TiO2−NPs. Further, the value of J*NP derived from eq 2 was even overestimated considering the electrostatic repulsion between the negatively charged cells and PAA-TiO2−NPs. Under this condition, the difference between J*NP and the actual uptake rate of PAA-TiO2−NPs would be further lowered. Therefore, diffusion from the bulk solution to the cell surface determined the rate of PAA-TiO 2 −NP uptake by T. thermophila.39 Similarly, quantum dot uptake by O. danica was also found to be diffusion-limited in our previous study.30 We further proposed that nanoparticle uptake tends to be restricted by their diffusion to the cell surface unless their particle size is smaller than 10 nm based on the equations above. Similar to PAA-TiO2−NPs, [Cd]cell also increased linearly with exposure time in the uptake experiment (SI, Figure S4). However, a substantial amount of Cd (the y-intercept) was
* = DCd × S × [Cd2 +]F × JCd
⎛1 1⎞ ⎜ + ⎟ ⎝R δ⎠
(3)
Where DCd (7 × 10−10 m2/s) is the diffusion coefficient of Cd2+.40 The value of J*Cd was much higher than the difference between VCd‑T and VCd‑second (i.e., the rate of total Cd uptake via the first and the third routes). This phenomenon indicates that the turnover rate of Cd transport sites was slow and the potentially labile Cd-PAA-TiO2−NP complexes were not required even though they could theoretically contribute to Cd uptake. Namely, the third pathway had a negligible contribution to the overall bioaccumulation of Cd. Under this condition, the difference between VCd‑T and VCd‑second is defined as VCd‑first (fg/cell/h, uptake via the first route), which should have conformed to FIAM in theory. However, VCd‑first at similar [Cd2+]F differed significantly (p < 0.05) from each other at the five levels of PAA-TiO2−NPs (Figure 3d). The biphasic 7572
dx.doi.org/10.1021/es500694t | Environ. Sci. Technol. 2014, 48, 7568−7575
Environmental Science & Technology
Article
correlation between VCd‑first and [Cd2+]F was then simulated by the Michaelis−Menten equation for each [TiO2−NPs]med as follows, VCd ‐ first =
max 2+ VCd ‐ first × [Cd ]F
K m + [Cd2 +]F
(4)
where VmaxCd‑first (fg/cell/h) represents the maximum uptake rate of Cd via the first pathway, Km (μg/L) is the Michaelis− Menten constant and indicates the binding affinity between Cd and its transporter on the cell surface. Values of both parameters thus obtained are listed in Table 1, and they both declined with the enhancement in [TiO2−NPs]med. It implies that the amount of Cd transporters on the cell surface decreased, but their metal binding affinity was reinforced in the presence of PAA-TiO2−NPs. According to our previous calculation, approximately 0.10−8.27% of the plasma membrane was invaginated by T. thermophila every minute. Moreover, the invagination rate was higher, and the cell membrane was refreshed more quickly at higher [TiO2− NPs]med. The rate of plasma membrane recycling, as depended on [TiO2−NPs]med, might thus play a critical role in metal association onto the cell surface and in its internalization. Nevertheless, more evidence is required to illuminate the underlying mechanisms about nanoparticle impacts on metal uptake via the first route. Efflux of PAA-TiO2−NPs and Cd. T. thermophila grew significantly (p < 0.05) in most treatments of the efflux experiment. Such growth dilution effects on [TiO2−NPs]intra and [Cd]intra were corrected from the calculation of the net efflux. While it seems that [TiO2−NPs]intra diminished with time in two of the nine treatments, this trend was insignificant (p > 0.05) (SI, Figure S6). The proportion of PAA-TiO2−NPs retained in cells fluctuated around 100% during the 5 h depuration period. Based on the few studies about the kinetics of nanoparticle expulsion out of cells (mostly human cells),43−46 nanoparticles can be expelled out quickly with the half-life ranging from a few seconds to several hours. Despite their fast depuration, there is also a substantial amount of internalized nanoparticles, which can not be removed, and their intracellular concentrations level off at the end.46 Moreover, no significant exocytosis of quantum dots was observed for a murine “macrophage-like” cell (J774.A1) within a 2 h period.47 Likewise, the expulsion of quantum dots by a freshwater mixotrophic alga O. danica was also slow, and their intracellular concentration decreased by no more than 10% after 4 h.30 The murine cell, the mixotrophic alga, and the ciliate used in the present study are all able to ingest particles through phagocytosis, as was different from the cells used in other nanoparticle efflux studies. They might thus exhibit a completely different pattern of nanoparticle exocytosis, which needs to be proved in the future. In contrast to PAA-TiO2−NPs, a substantial amount of intracellular Cd (4.3−22.1%) was eliminated by T. thermophila (Figure 4). As mentioned before, Cd was taken up in the form of either free ion directly from the bulk solution or metalnanoparticle complexes. The latter route had a contribution of 46.3% to total Cd uptake (SI, Figure S5b). Additionally, the subcellular distribution of Cd accumulated through these two pathways was different (Figure 2a−f). However, the overall elimination of Cd was independent of [Cd]intra and [TiO2− NPs]intra. Hence, Cd taken up via the second pathway was possibly eliminated in the same rate as the first one. Since there
Figure 4. Proportion of Cd retained in the cells of T. thermophila precultured in the media containing (a) 0.1 and (b) 10 μg/L [Cd]med with the addition of 0, 0.4, 1.3, and 4.0 mg-Ti/L PAA-TiO2−NPs during the 5 h depuration period. Data are mean ± standard deviation (n = 2).
was no significant exocytosis of PAA-TiO2−NPs, the nanoparticle-associated Cd inside the cells cannot be excreted out. Therefore, it could be speculated that part of the nanoparticleadsorbed Cd would get dissociated once inside the acidic vacuoles containing various metal-binding ligands.48 Further, Cd dissociated from the nanoparticle surface was liberated out of the cells even more rapidly than (or at least no lower than) its counterpart accumulated via the first pathway. Based on the findings above and on the results of our previous studies,9,10 we propose that nanoparticles have at least two different kinds of effects on metal bioavailability and toxicity. On one hand, they could adsorb a substantial amount of metal ions on the surface and decrease its free ion concentration in the environment. If the nanoparticles cannot enter the cells as what was observed in Yang et al.,9,10 then its bioaccumulation and toxicity are reduced in agreement with FIAM. On the other hand, when the nanoparticles could get into the cells directly with a significant adsorption of metal ions, they serve as a carrier of metal uptake. In this situation, both metal bioaccumulation and its toxicity would be changed abruptly and could not be well predicted by FIAM. In the natural environment, the free metal ion concentration is determined by a number of factors, and a small variation in one factor is hard to alter this concentration dramatically. Therefore, the carrier effects from nanoparticles may be more important under this condition. What’s more, the dual effects of nanoparticles should not be limited to metal ions only and could be extrapolated to other conventional pollutants already existing in the environment. These facts cannot be ignored when we assess the environmental risks of nanoparticles.
■
ASSOCIATED CONTENT
S Supporting Information *
Detailed information how the subcellular distribution of PAATiO2−NPs and Cd was measured, [Cd]med and [TiO2− NPs]med in different experiments, the [Cd2+]F and [Cd]intrabased EC50s, the physicochemical properties of PAA-TiO2− NPs, dose−response curves of Cd, the uptake kinetics of PAATiO2−NPs and Cd, correlations between VCd‑T and [Cd2+]F as well as between VCd‑T and VCd‑second, and PAA-TiO2−NP efflux are included as Supporting Information. This material is available free of charge via the Internet at http://pubs.acs.org. 7573
dx.doi.org/10.1021/es500694t | Environ. Sci. Technol. 2014, 48, 7568−7575
Environmental Science & Technology
■
Article
(14) Mortimer, M.; Kasemets, K.; Kahru, A. Toxicity of ZnO and CuO nanoparticles to ciliated protozoa Tetrahymena thermophila. Toxicology 2010, 269 (2−3), 182−189. (15) Ud-Daula, A.; Pfister, G.; Schramm, K. W. Method for toxicity test of titanium dioxide nanoparticles in ciliate protozoan Tetrahymena. J. Environ. Sci. Health A 2013, 48 (11), 1343−1348. (16) Larue, C.; Veronesi, G.; Flank, A. M.; Surble, S.; Herlin-Boime, N.; Carriere, M. Comparative uptake and impact of TiO2 nanoparticles in wheat and rapeseed. J. Toxicol. Environ. Health A 2012, 75 (13−15), 722−734. (17) Gorovsky, M. A.; Yao, M. C.; Keevert, J. B.; Pleger, G. L. Isolation of micro- and macronuclei of Tetrahymena pyriformis. Method. Cell Biol. 1975, 9 (0), 311−327. (18) Dryl, S. Antigenic transformation in Paramecium aurelia after homologous antiserum treatment during autogamy and conjugation. J. Protozool. 1959, 6 (3), 25−25. (19) Miao, A. J.; Zhang, X. Y.; Luo, Z.; Chen, C. S.; Chin, W. C.; Santschi, P. H.; Quigg, A. Zinc oxide engineered nanoparticles dissolution and toxicity to marine phytoplankton. Environ. Toxicol. Chem. 2010, 29 (12), 2814−2822. (20) Miao, A. J.; Wang, W. X. Relationships between cell-specific growth rate and uptake rate of cadmium and zinc by a coastal diatom. Mar. Ecol.: Prog. Ser. 2004, 275, 103−113. (21) Wang, N. X.; Li, Y.; Deng, X. H.; Miao, A. J.; Ji, R.; Yang, L. Y. Toxicity and bioaccumulation kinetics of arsenate in two freshwater green algae under different phosphate regimes. Water Res. 2013, 47 (7), 2497−2506. (22) Hassler, C. S.; Slaveykova, V. I.; Wilkinson, K. J. Discriminating between intra- and extracellular metals using chemical extractions. Limnol. Oceanogr. Methods 2004, 2, 237−247. (23) Miao, A. J.; Wang, W. X. Fulfilling iron requirements of a coastal diatom under different temperatures and irradiances. Limnol. Oceanogr. 2006, 51 (2), 925−935. (24) Liufu, S. C.; Mao, H. N.; Li, Y. P. Adsorption of poly(acrylic acid) onto the surface of titanium dioxide and the colloidal stability of aqueous suspension. J. Colloid Interface Sci. 2005, 281 (1), 155−163. (25) Navarro, E.; Baun, A.; Behra, R.; Hartmann, N. B.; Filser, J.; Miao, A. J.; Quigg, A.; Santschi, P. H.; Sigg, L. Environmental behavior and ecotoxicity of engineered nanoparticles to algae, plants, and fungi. Ecotoxicology 2008, 17 (5), 372−386. (26) Kirwan, L. J.; Fawell, P. D.; van Bronswijk, W. An in situ FTIRATR study of polyacrylate adsorbed onto hematite at high pH and high ionic strength. Langmuir 2004, 20 (10), 4093−4100. (27) Gallego, A.; Martin-Gonzalez, A.; Ortega, R.; Gutierrez, J. C. Flow cytometry assessment of cytotoxicity and reactive oxygen species generation by single and binary mixtures of cadmium, zinc and copper on populations of the ciliated protozoan Tetrahymena thermophila. Chemosphere 2007, 68 (4), 647−661. (28) Rico, D.; Martin-Gonzalez, A.; Diaz, S.; de Lucas, P.; Gutierrez, J. C. Heavy metals generate reactive oxygen species in terrestrial and aquatic ciliated protozoa. Comp. Biochem. Phys. C 2009, 149 (1), 90− 96. (29) Miao, A. J.; Wang, W. X. Predicting copper toxicity with its intracellular or subcellular concentration and the thiol synthesis in a marine diatom. Environ. Sci. Technol. 2007, 41 (5), 1777−1782. (30) Wang, Y.; Miao, A. J.; Luo, J.; Wei, Z. B.; Zhu, J. J.; Yang, L. Y. Bioaccumulation of CdTe quantum dots in a freshwater alga Ochromonas danica: A kinetics study. Environ. Sci. Technol. 2013, 47 (18), 10601−10610. (31) Iversen, T. G.; Skotland, T.; Sandvig, K. Endocytosis and intracellular transport of nanoparticles: Present knowledge and need for future studies. Nano Today 2011, 6 (2), 176−185. (32) Kloetzel, J. A. Feeding in ciliated protozoa 0.1. pharyngeal disks in euplotesSource of membrane for food vacuole formation. J. Cell Sci. 1974, 15 (2), 379−401. (33) Lavin, D. P.; Hatzis, C.; Srienc, F.; Fredrickson, A. G. Size effects on the uptake of particles by populations of Tetrahymena pyriformis cells. J. Protozool. 1990, 37 (3), 157−163.
AUTHOR INFORMATION
Corresponding Author
*Phone: +86 25 89680255; fax: +86 25 89680569; e-mail: [email protected]. Author Contributions †
W.W.Y. and Y.W. contributed equally to this work.
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We thank three anonymous reviewers as well as Dr. Qiaoguo Tan and Peter H. Santschi for their constructive suggestions on this paper. The ciliate is courtesy of Dr. Wei Miao. The financial support offered by the National Natural Science Foundation of China (41271486, 41001338, and 21237001) and the Natural Science Foundation of Jiangsu Province (BK2010371) to A. J. Miao have made this work possible.
■
REFERENCES
(1) Moore, M. N. Do nanoparticles present ecotoxicological risks for the health of the aquatic environment? Environ. Int. 2006, 32 (8), 967−976. (2) Fabrega, J.; Luoma, S. N.; Tyler, C. R.; Galloway, T. S.; Lead, J. R. Silver nanoparticles: Behaviour and effects in the aquatic environment. Environ. Int. 2011, 37 (2), 517−531. (3) Hou, W. C.; Westerhoff, P.; Posner, J. D. Biological accumulation of engineered nanomaterials: A review of current knowledge. Environ. Sci.: Processes Impacts 2013, 15 (1), 103−122. (4) Sun, H.; Zhang, X.; Niu, Q.; Chen, Y.; Crittenden, J. C. Enhanced accumulation of arsenate in carp in the presence of titanium dioxide nanoparticles. Water, Air, Soil Pollut. 2007, 178 (1−4), 245−254. (5) Baun, A.; Sorensen, S. N.; Rasmussen, R. F.; Hartmann, N. B.; Koch, C. B. Toxicity and bioaccumulation of xenobiotic organic compounds in the presence of aqueous suspensions of aggregates of nano-C-60. Aquat. Toxicol. 2008, 86 (3), 379−387. (6) Fan, W.; Cui, M.; Liu, H.; Wang, C.; Shi, Z.; Tan, C.; Yang, X. Nano-TiO2 enhances the toxicity of copper in natural water to Daphnia magna. Environ. Pollut. 2011, 159 (3), 729−734. (7) Zhang, W.; Miao, Y.; Lin, K.; Chen, L.; Dong, Q.; Huang, C. Toxic effects of copper ion in zebrafish in the joint presence of CdTe QDs. Environ. Pollut. 2013, 176, 158−164. (8) Tan, C.; Fan, W. H.; Wang, W. X. Role of titanium dioxide nanoparticles in the elevated uptake and retention of cadmium and zinc in Daphnia magna. Environ. Sci. Technol. 2012, 46 (1), 469−476. (9) Miao, A. J.; Wang, N. X.; Yang, L. Y.; Wang, W. X. Accumulation kinetics of arsenic in Daphnia magna under different phosphorus and food density regimes. Environ. Toxicol. Chem. 2012, 31 (6), 1283− 1291. (10) Yang, W. W.; Li, Y.; Miao, A. J.; Yang, L. Y. Cd2+ toxicity as affected by bare TiO2 nanoparticles and their bulk counterpart. Ecotoxicol. Environ. Saf. 2012, 85, 44−51. (11) Campbell, P. G. C., Interactions between trace metals and aquatic organisms: A critique of the free-ion activity model. In Metal Speciation and Bioavailability in Aquatic Systems; Tessier, A., Turner, D. R., Eds.; John Wiley & Sons: Chichester, 1995; pp 45−102. (12) Hartmann, N. B.; Von der Kammer, F.; Hofmann, T.; Baalousha, M.; Ottofuelling, S.; Baun, A. Algal testing of titanium dioxide nanoparticles-Testing considerations, inhibitory effects and modification of cadmium bioavailability. Toxicology 2010, 269 (2−3), 190−197. (13) Worms, I. A. M.; Boltzman, J.; Garcia, M.; Slaveykova, V. I. Cellwall-dependent effect of carboxyl-CdSe/ZnS quantum dots on lead and copper availability to green microalgae. Environ. Pollut. 2012, 167, 27−33. 7574
dx.doi.org/10.1021/es500694t | Environ. Sci. Technol. 2014, 48, 7568−7575
Environmental Science & Technology
Article
(34) Zefirov, A. L.; Abdrakhmanov, M. M.; Mukhamedyarov, M. A.; Grigoryev, P. N. The role of extracellular calcium in exo- and endocytosis of synaptic vesicles at the frog motor nerve terminals. Neuroscience 2006, 143 (4), 905−910. (35) Choi, J. S.; Kim, K. R.; Ahn, D. W.; Park, Y. S. Cadmium inhibits albumin endocytosis in opossum kidney epithelial cells. Toxicol. Appl. Pharmacol. 1999, 161 (2), 146−152. (36) Herak-Kramberger, C. M.; Brown, D.; Sabolic, I. Cadmium inhibits vacuolar H+-ATPase and endocytosis in rat kidney cortex. Kidney Int. 1998, 53 (6), 1713−1726. (37) Buffle, J.; Wilkinson, K. J.; van Leeuwen, H. P. Chemodynamics and bioavailability in natural waters. Environ. Sci. Technol. 2009, 43 (19), 7170−7174. (38) Van Leeuwen, H. P.; Koster, W. Physicochemical Kinetics and Transport at Biointerfaces; John Wiley & Sons Ltd.: West Sussex, 2004. (39) Hudson, R. J. M.; Morel, F. M. M. Iron transport in marine phytoplankton - kinetics of cellular and medium coordination reactions. Limnol. Oceanogr. 1990, 35 (5), 1002−1020. (40) Van Leeuwen, H. P.; Town, R. M.; Buffle, J.; Cleven, R. F. M. J.; Davison, W.; Puy, J.; van Riemsdijk, W. H.; Sigg, L. Dynamic speciation analysis and bioavailability of metals in aquatic systems. Environ. Sci. Technol. 2005, 39 (22), 8545−8556. (41) Van Leeuwen, H. P. Metal speciation dynamics and bioavailability: Inert and labile complexes. Environ. Sci. Technol. 1999, 33 (21), 3743−3748. (42) Kola, H.; Wilkinson, K. J. Cadmium uptake by a green alga can be predicted by equilibrium modelling. Environ. Sci. Technol. 2005, 39 (9), 3040−3047. (43) Jiang, X.; Rocker, C.; Hafner, M.; Brandholt, S.; Dorlich, R. M.; Nienhaus, G. U. Endo- and exocytosis of zwitterionic quantum dot nanoparticles by live HeLa cells. ACS Nano 2010, 4 (11), 6787−6797. (44) Chu, Z.; Huang, Y.; Tao, Q.; Li, Q. Cellular uptake, evolution, and excretion of silica nanoparticles in human cells. Nanoscale 2011, 3 (8), 3291−3299. (45) Chithrani, B. D.; Chan, W. C. W. Elucidating the mechanism of cellular uptake and removal of protein-coated gold nanoparticles of different sizes and shapes. Nano Lett. 2007, 7 (6), 1542−1550. (46) Ohta, S.; Inasawa, S.; Yamaguchi, Y. Real time observation and kinetic modeling of the cellular uptake and removal of silicon quantum dots. Biomaterials 2012, 33 (18), 4639−4645. (47) Clift, M. J. D.; Rothen-Rutishauser, B.; Brown, D. M.; Duffin, R.; Donaldson, K.; Proudfoot, L.; Guy, K.; Stone, V. The impact of different nanoparticle surface chemistry and size on uptake and toxicity in a murine macrophage cell line. Toxicol. Appl. Pharmacol. 2008, 232 (3), 418−427. (48) Nilsson, J. R. On food vacuoles in Tetrahymena pyriformis GL. J. Protozool. 1977, 24 (4), 502−507.
7575
dx.doi.org/10.1021/es500694t | Environ. Sci. Technol. 2014, 48, 7568−7575
