Article pubs.acs.org/est
Accumulation and Embryotoxicity of Polystyrene Nanoparticles at Early Stage of Development of Sea Urchin Embryos Paracentrotus lividus C. Della Torre,† E. Bergami,† A. Salvati,‡,∥ C. Faleri,§ P. Cirino,⊥ K. A. Dawson,‡ and I. Corsi*,† †
Department of Physical, Earth and Environmental Sciences, University of Siena, 53100 Siena, Italy Centre for BioNano Interactions, School of Chemistry and Chemical Biology, University College Dublin, Dublin 4, Ireland § Department of Life Sciences, University of Siena, 53100 Siena, Italy ⊥ Anton Dohrn Zoological Station, 80121 Naples, Italy ‡
S Supporting Information *
ABSTRACT: Nanoplastic debris, resulted from runoff and weathering breakdown of macro- and microplastics, represents an emerging concern for marine ecosystems. The aim of the present study was to investigate disposition and toxicity of polystyrene nanoparticles (NPs) in early development of sea urchin embryos (Paracentrotus lividus). NPs with two different surface charges where chosen, carboxylated (PS-COOH) and amine (PS-NH2) polystyrene, the latter being a less common variant, known to induce cell death in several in vitro cell systems. NPs stability in natural seawater (NSW) was measured while disposition and embryotoxicity were monitored within 48 h of postfertilization (hpf). Modulation of genes involved in cellular stress response (cas8, 14-3-3ε, p-38 MAPK, Abcb1, Abcc5) was investigated. PSCOOH forms microaggregates (PDI > 0.4) in NSW, whereas PS-NH2 results are better dispersed (89 ± 2 nm) initially, though they also aggregated partially with time. Their respectively anionic and cationic nature was confirmed by ζ-potential measurements. No embryotoxicity was observed for PS-COOH up to 50 μg mL−1 whereas PS-NH2 caused severe developmental defects (EC50 3.85 μg mL−1 24 hpf and EC50 2.61 μg mL−1 48 hpf). PS-COOH accumulated inside embryo’s digestive tract while PS-NH2 were more dispersed. Abcb1 gene resulted up-regulated at 48 hpf by PS-COOH whereas PS-NH2 induced cas8 gene at 24 hpf, suggesting an apoptotic pathway. In line with the results obtained with the same PS NPs in several human cell lines, also in sea urchin embryos, differences in surface charges and aggregation in seawater strongly affect their embryotoxicity. biota.15,16 Being small in size, they are more likely to be ingested and accumulate in marine organisms and consequently are biomagnified along trophic webs.2,17 Nanoscale materials end up a significant agglomeration in seawater, but dispersion in the water column might also occur due to counterbalance of several parameters as pH, salts, natural organic matter (NOM), natural colloids as well as nanoparticle’s size and surface chemistry (i.e., charges).18 Marine aggregates of nanosized polystyrene (100 nm) (PS) facilitate ingestion in suspension feeding bivalves, which translocate them from the gut to the circulatory system and particles are retained for more than 1 month.19 Food chain transfer of PS from algae, through zooplankton to fish, has been reported as well as affecting behavior and fat metabolism of the top consumers of such food chains, fish.20 PS beads have been reported to be ingested by zooplankton and their uptake varies
1. INTRODUCTION Microplastics have been acknowledged by the international scientific community as an emerging worldwide threat for the marine ecosystem based on their huge accumulation in five convergence zones named gyres and their impact on marine wildlife from entanglement and ingestion at various trophic levels.1−8 Microplastics encounter a continual fragmentation mainly due to physical, mechanical and chemical attack from wind, waves and UV degradation and hydrolysis.9 Microsized plastics and also nanoscale ones will thus increase consistently with time10,11 and their persistence and abundance in seawater will enhance their global distribution, even reaching remote areas (i.e., around Antarctica).12 Micro- and nanoscale debris are also directly released as granules, pellets and powders used in various industrial applications including nanomedicine (drug delivery) and in personal care products (cosmetics, detergents, food and cleaning products).13,14 Although microplastics have been quite well studied, the fate and impact of nanoscale plastics in the marine environment is almost unknown and this is raising concern due their increasing abundance in the water column and food webs, which could imply toxicity to marine © XXXX American Chemical Society
Received: May 26, 2014 Revised: September 24, 2014 Accepted: September 26, 2014
A
dx.doi.org/10.1021/es502569w | Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Environmental Science & Technology
Article
according to species, life-stage and size of PS.17 PSs are one of the most largely used plastics worldwide, used in food and industrial packaging, disposable cutlery, compact disc cases, building insulation, medical products and toys.21−24 This versatile and nonbiodegradable polymer is the forth most plastic found in the oceans as micro- and nanodebris and accounts for 24% of the macroplastics found in estuarine habitat.25 Spherules of PS or fragments have been found already 30 years ago in stomach of fish from U.K.26 and U.S. marine coastal waters, accounting for over 20% and 60% of the total fish species collected.27 They reach the oceans as other plastics, but some direct release has been recently documented from aging and biological corrosion (boring crustaceans) of PS floats used under floating docks in harbors, aquaculture facilities and marine coastal areas.28−30 Hydrodynamic and morphological characteristics of marine coastal areas could influence the dispersion of nanodebris derived from various wastes and consequently, coastal systems may be particularly affected by polymeric NPs as polystyrenes.31−33 PS NPs refers to particles with nanoscale dimensions having a PS core and variable functional groups that determine the effective surface charge of the particle.34 The most common PS NPs are plain polystyrenes, typically with sulfonated residues (due to the polymerization reaction) on their surface. Common surface modifications of this material, introduced to enable further functionalization or in order to obtain particles with different surface charge, include those displaying carboxylated groups (COOH) or unsaturated amine (NH2) as anionic and cationic NPs, respectively.35 Positively charged PS NPs (and negatively charged, to a lesser extent) have been shown to induce oxidative stress, mitochondrial damage and cellular toxicity in a phagocytic cell line.36 Some authors37,38 report that PS-NH2 NPs induce cytotoxicity, whereas PS-COOH NPs are reported as nontoxic. In particular, it has been observed that these NPs accumulate in the lysosomes, where they induce lysosomal damage, leading to release of lysosomal content in the cytosol, followed by apoptosis.39,40 Furthermore, the same NPs at sublethal doses have been found to induce cell cycle arrest.41 These data overall suggest that the toxic potential of PS NPs strongly varies depending on NP properties (e.g., size and surface chemistry) and its interaction with the surrounding medium, specific cell types, and also the mechanism used by NPs to enter the cells and reach target molecules.39,42,43 In most cases, when NPs are exposed to cells in realistic conditions, such as in the presence of biological fluids, rather than simple buffer, NP uptake might be different.44−47 Although such mechanisms have been quite well described in human cell models, how PS NPs can interact with their surroundings as, for instance, with marine waters and enter the cells of marine organisms, is still largely unknown. PS NPs can be thus considered as good model for studying both environmental fate of nanoplastics, in terms of interactions with the surrounding media, and toxicity for marine organisms focusing on specific pathways of cellular uptake.48−52 In particular, as extensive literature on PS-COOH and PS-NH2 NPs is already available and the mechanisms by which the PSNH2 induce cell death have been already described in details,36,39,41,43 we have chosen these 2 NPs as a good model to translate what observed in human cell models toward a more complex organism and different environment, such as sea urchin embryos and seawater in general.
Marine invertebrates are among the primary biological targets of NPs, being exposed both to polymeric NPs in suspension, as planktonic larvae, and to the fraction in sediments, as adult organism.15,31,53 Mediterranean Sea urchin Paracentrotus lividus (Lamarck, 1816) is a key species of coastal ecosystems, diffused in the Mediterranean Sea.54 This organism has been largely employed in ecotoxicological studies for evaluating many environmental stressors.55−59 More recently, P. lividus has been proposed15 as alternative model for testing the toxicity of engineered NPs in seawater. Based on possible aggregation of suspended NPs in seawater, bottom grazers species as sea urchins are expected to be exposed to high concentrations in their natural environment. The aim of the present study was to investigate disposition and toxicity of fluorescent carboxylated (PS-COOH) and amine (PS-NH2) polystyrene NPs in sea urchin embryos of P. lividus during early stages of development in order to determine whether these materials have similar effects in full embryos and marine environment as what observed in common human and other mammalian cell lines.
2. MATERIALS AND METHODS 2.1. Nanoparticles Physico-Chemical Characterization. 40 nm green carboxylated polystyrene (PS-COOH) NPs (505 nm excitation, 515 nm emission) were purchased from Invitrogen. Unlabeled and blue fluorescently labeled (358 nm excitation, 410 nm emission) 50 nm amino modified polystyrene (PS-NH2) NPs were purchased from Bangs Laboratories and Sigma, respectively. The unlabeled variant was used after dilution of the stock to 10 mg mL−1 in Milli-Q water. The same NPs have been widely investigated within the framework of the FP7 Research infrastructure QualityNano. Thanks to their easy detection, fluorescently labeled PS beads have been recommended as priority test material to be developed and used for ecotoxicological studies.60 Primary particle diameter of PS NPs was determined by transmission electron microscopy (Philips Morgagni 268D electronics, at 80 kV and equipped with a MegaView II CCd camera. Size (Z-average and polydispersity index, PDI) and ζpotential (mV) were determined by dynamic light scattering (Malvern instruments), using Zetasizer Nano Series software, version 7.02 (Particular Sciences, U.K.). Measurements have been performed in triplicate, each containing 11 runs of 10 s for determining Z-average, 20 runs for the ζ-potential. NSW without PS NPs added was analyzed as the control. Timedependent variations in size were investigated at 6, 24 and 48 h after suspension in NSW in order to determine particle stability at the same exposure times used to perform sea urchin embryotoxicity test. PS NPs suspension were prepared in NSW (0.45 μm filtered, T = 18 °C, salinity 38‰, pH 8.3, conductivity 6 S/m) and quickly vortexed prior to use but not sonicated (see the Supporting Information for further details). Concentration up to 50 μg mL−1 was shown to induce apoptosis in 1321N1 human cells treated for 24 h with PS-NH2 NPs.43 2.2. Sea Urchin Maintenance, Gametes Collection and PS NPs Exposure. Adults of Mediterranean Sea urchin P. lividus, were kindly provided by the Anton Dohrn Zoological Station of Naples (Italy) and transported within few hours in an insulated box to the Marine Aquarium facility at the Department of Physical, Earth and Environmental Sciences of the University of Siena. Sea urchins were maintained in 140 L flow-through circulating aquarium at 18 °C, constant (12:12 B
dx.doi.org/10.1021/es502569w | Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Environmental Science & Technology
Article
Figure 1. Primary and secondary characterization of PS NPs using TEM imaging and DLS, respectively. TEM images of PS-COOH (A) and PS-NH2 (C) suspended in distilled water. Scale bar: 200 nm. Size distribution graphs for 40 nm PS-COOH (B) and 50 nm PS-NH2 (D) in NSW, as determined by Contin analysis of the data obtained by DLS and edited with GraphPad Prism5. Particle size distribution at 50 μg mL−1 is given by intensity and for each graph, 3 independent replicates have been reported. The x-axis minimum is set at 10 nm and a logarithmic scale is used for the y-axis. ζ-potential values (mV) obtained by DLS analysis are shown shown for PS-COOH (B) and PS-NH2 (D) as average ± s.d.
(Calbiochem) and they were left overnight in the dark at room temperature to allow the mounting medium to harden before imaging. Images were analyzed using the software FV10-ASW 4.0a Viewer. 2.4. Embryotoxicity. The embryotoxicity was performed in 24-well plates. About 30 min after fertilization, embryos were exposed to different nominal concentrations of PS-COOH (2.5, 5, 10, 25 and 50 μg mL−1) and PS-NH2 (1, 2.5, 3, 5, 10 and 50 μg mL−1) NPs suspension in NSW and then incubated at 18 °C in the dark. At specific exposure times of 6−24−48 h post fertilization (hpf) corresponding to blastula, gastrula and pluteus stages respectively, embryos were fixed in 70% ethanol and observed using an inverted microscope (INV-3, BEL photonics). As validity of the test, controls (only NSW) must show at least >85% of normal developed embryos. Copper sulfate (0−400 ng mL−1) was tested as a reference toxicant.61 2.5. Gene Expression by Real Time q-PCR. Fertilized eggs were exposed to 25 μg mL−1 PS-COOH and 3 μg mL−1 PS-NH2 NPs respectively and embryos collected at 24 and 48 hpf and then centrifuged at 2000g for 8 min. RNA extraction was performed following standardized procedures (see the Supporting Information for further details). Specific primers for stress response genes (cas8, 14-3-3ε, p-38 MAPK) and the housekeeping Z12-1 were already published,62,63 while primers for Abcb1 and Abcc5 were designed IDTDNA www.idtdna.com. Primer sequences used for q-PCR are listed and shown in Table S1 in the Supporting Information. q-PCR was performed using standardized protocols (see the Supporting Information for further details). Data were analyzed by the ΔΔCt method.64 Fold changes were considered significant if they were ±2-fold change from the reference point (0.5 < not significant 95% fertilization success were used for embryotoxicity and the other experimental tests. Fertilization was carried out adding diluted sperm (1:1000 in NSW) to the egg suspension. The obtained zygotes were washed twice and diluted to the final concentration of 250 embryos mL−1 (embryotoxicity test) and 600 embryos mL−1 (gene expression and disposition studies) in NSW. 2.3. Disposition Study. PS-COOH and PS-NH2 NPs disposition studies were investigated using 40 mL sterile flasks and embryos exposed to 25 μg mL−1 PS-COOH and 3 μg mL−1 PS-NH2 based on results of embryotoxicity. Embryos were observed at 4−6, 24 and 48 hpf. Each experiment has been performed three times. PS NPs tracking was performed under optical fluorescent microscope AXIO IMAGER Z1 using an Apotome system (Zeiss) (filter FITC 470/525 for PS-COOH; filter DAPI 365/ 445 for the labeled PS-NH2). Images were taken with AxioCam MRm camera at 20−40× using AxioVision software. For more accurate observation, a confocal laser scanning microscope, with the aid of the Olympus FluoView 1000 software, has been used at 20−40×, (filter FITC 488 nm for PS-COOH). Nuclei were labeled using DAPI (filter DAPI 405 nm). Specific slides were prepared adding a drop of Mowiol C
dx.doi.org/10.1021/es502569w | Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Environmental Science & Technology
Article
Figure 2. (A) Optical fluorescence microscopy showing disposition of PS-NH2 (3 μg mL−1) (blue) and of PS-COOH (25 μg mL−1) (green) in P. lividus embryos at 4−6, 24 and 48 hpf. For PS-NH2 control (NSW) is shown. (B) Details of confocal images only for PS-COOH (25 μg mL−1) (green) showing embryos nuclei stained with DAPI (blue). Scale bar: 20 μm.
The ζ-potential in NSW (Figure 1) indicated a negative surface charge (−7 mV) for PS-COOH NPs and a positive one for PS-NH2 NPs (+13 mV) as expected from the chemistry of the NPs and in agreement with previous finding for PBS.38 No time-dependent variations in ζ-potential were observed for both PS NPs (data not shown), though ζ-potential is also affected by particle aggregation. Lower absolute values of ζ-potentials observed for both PS NPs in NSW (as compared to the values in Milli-Q water, which were −51.8 mV for PS-COOH and +42.8 mV for PSNH2), suggest a screening effect of surface charges due to the higher salt content, as discussed above, and also by proteins or other compounds in the surroundings42 as, for instance, the natural organic matter (NOM) present in seawater. The average amount of NOM in NSW is about 3 mg L−1 and it is reported to affect dispersion of other NPs due to the formation of complexes.66−72 In general, it is important to note that this concentration is roughly 1000 lower the mass of biomolecules in typical in vitro cell culture conditions (where typically 10% fetal calf serum is added, corresponding roughly to 4 mg mL−1). The above considerations drive to the conclusion that medium choice and related parameters should be always taken into account for NPs characterization in complex natural media as for instance seawater. Furthermore, it is strongly recommended to perform ecotoxicological studies using natural matrixes, since a combination of parameters such as pH, ionic strength, salt concentrations and the presence of other biomolecules, similarly to what observed for proteins forming a corona on the NP in human blood, are fundamental for studying not only NP behavior but more important interactions with cells and potential toxicity.40 The detailed secondary characterization performed in the present study clearly showed that NSW sea urchin embryos are exposed mostly to nanoscale aggregates of PS-NH2 (1000 nm). 3.2. Disposition in Sea Urchin Developing Embryos. For the purpose of understanding the behavior of PS NPs in seawater and their disposition in sea urchin developing embryos, fluorescent PS NPs were used. At the morula/blastula stage (4−6 hpf), PS-NH2 showed a diffuse fluorescence (4−6 hpf) (Figure 2A); however, due to the low fluorescence intensity of these NPs, the diffuse pattern
2.6. Statistical Analysis. All statistical analysis were performed using Graphpad Prism5. Analysis of variance (ANOVA) was performed to compare the various treatments, and p < 0.05 was taken as the significant cutoff. Results of embryotoxicity are mean of at least three independent experiments. When possible, EC50 values were calculated by fitting the percentage of normal embryos to a classical sigmoidal dose−response model according to the equation y = b + (a − b) / 1 + 10 (Log EC50−x) where y is response, b response minimum, a response maximum, x the logarithm of inhibitor concentration and EC50 the concentration of inhibitor giving 50% of maximum effect. Each experiment has been performed three to five times.
3. RESULTS AND DISCUSSION The present study investigated PS NPs impact on early life developmental stages of P. lividus sea urchin embryo both in terms of disposition and embryotoxicity. 3.1. Behavior in Sea Water. TEM images confirmed nominal size of PS-COOH and unlabeled PS-NH2 as 40 and 50 nm, respectively (Figure 1). DLS analysis of NSW without PS NPs showed the presence of a background of molecules of broad size range (mean Zaverage = 380 ± 101 nm standard deviation). Such aggregates become irrelevant once the PS NPs are added to NSW. Zaverage values determined by cumulant analysis of the data obtained by DLS showed formation of large PS-COOH aggregates (1764 ± 409 nm s.d.) in NSW, whereas unlabeled PS-NH2 resulted far less aggregated with a Z-average of 89 nm (±2 nm s.d.) and a PDI of 0.32. Size distributions by Intensity obtained by contin analysis also confirmed this outcome (Figure 1), though peaks at larger sizes were visible also for PSNH2, suggesting that also this sample may be partially aggregated. Both PS NPs showed aggregation increasing with time (0, 6, 24 and 48 h) in NSW: for PS-NH2 PDI, from 0.32 to 0.4; for PS-COOH, remains >0.4. Overall, the lower stability of these NPs in NSW compared to what observed in in vitro cell culture media is probably due to the higher salinity of NSW in respect to PBS and cell culture conditions. The higher salt concentration in fact can screen the particle surface charges leading to the observed aggregation, unless the particles are stabilized by other factors, such as for instance absorption of biomolecules on their surface. D
dx.doi.org/10.1021/es502569w | Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Environmental Science & Technology
Article
was difficult to distinguish from the autofluorescence of embryos in the spectrum of PS-NH2 observed in the same conditions (358 nm excitation, 410 nm emission) (Figure 2A, showing controls). At the gastrula stage (24 hpf), PS-NH2 seems to be more clearly distributed on the outer surface membrane. At 48 hpf, any clear internalization of PS-NH2 in embryos was observed but several larval (pluteus) malformations were evident, showing a clear sign of toxicity (Figure 2A). Concerning PS-COOH, at the morula/blastula stage (4−6 hpf), they were found to be equally distributed on the external surface of the embryo (Figure 2A,B). Later, in the gastrula stage (24 hpf), PS-COOH were present in the rudimentary gut (Figure 2A,B). The observed disposition might be related to the migration during the invagination into the blastocoel of primary mesenchyme cells derived from the vegetal pole. Such migration might be also enhanced by the ciliary beat of mesenchyme cells of sea urchin embryos. The gastrulation process continued with intestine elongation leading to an accumulation of PS NPs inside the digestive tract. At 48 hpf, the four-armed pluteus showed in fact PS-COOH clearly sequestered in the archenteron as large aggregates (Figure 2 A,B). Confocal images confirmed such disposition at different stages sea urchin embryo’s development (Figure 2B). A low accumulation of PS-COOH at 48 hpf was observed after hindering the ciliary beat of sea urchin embryos at 18 hpf (shown in Figure S1 in SI). Such findings might suggest that embryonic cilia might have an active role in carrying PS-COOH inside the sea urchin gastrula (after 24 hpf) and that PS-COOH might be not internalized into the single embryonic cells being large aggregates. Cilia might also have a role in breaking down PS aggregates as already suggested in gills and labial palps of marine mussels.19 The accumulation of PS-COOH inside the digestive tract of sea urchin embryos suggests a peculiar mechanism for these PS NPs, as already described for PS microbeads in a marine copepod.73 Translocation through the digestive gland has also been reported for PS micro- and nanospheres beads in bivalves.19,25 The observed accumulation might enhance trophic food web transfer by predation (biomagnification) as already recently reported for micro PS.74 Similar behavior has been described for other NPs as quantum dots and n-TiO2.75,76 The transfer of nanoplastics through the marine food web raises serious concern about the exposure of organisms at higher trophic levels, including humans. Our findings suggest that the different aggregation of the two tested PS NPs in NSW (89 nm for PS-NH2 and PDI > 0.4 for PS-COOH) and, more importantly, the different surface charge might affect their cellular uptake and disposition and consequent toxicity during embryo development. 3.3. Embryotoxicity. According to the classification given by ref 59, the impact of PS NPs on sea urchin embryo development was assessed in terms of embryonic defects during 48 hpf (Figure 3). As reference toxicant for sea urchin embryotoxicity,61 CuSO4 was used and an EC50 of 129.5 ng mL−1 was calculated at 48 hpf. PS-COOH NPs did not show any relevant effect on embryo development in the range of tested concentrations (2.5−50 μg mL−1) except for slight toxic malformations in
