- Email: [email protected]
Impact of surface modification on the toxicity of zerovalent iron nanoparticles in aquatic and terrestrial organisms
Ecotoxicology and Environmental Safety 163 (2018) 436–443
Contents lists available at ScienceDirect
Ecotoxicology and Environmental Safety journal homepage: www.elsevier.com/locate/ecoenv
Impact of surface modification on the toxicity of zerovalent iron nanoparticles in aquatic and terrestrial organisms Hakwon Yoona,1, Monmi Pangginga,1, Min-Hee Jangb, Yu Sik Hwangb, Yoon-Seok Changa, a b
T ⁎
Division of Environmental Science and Engineering, Pohang University of Science and Technology (POSTECH), Pohang 37673, Republic of Korea Future Environmental Research Center, Korea Institute of Toxicology (KIT), Jinju 52834, Republic of Korea
A R T I C LE I N FO
A B S T R A C T
Keywords: Surface-modified nZVI Nanotoxicity Ecotoxicity Oxidative stress Physicochemical property
Nanoscale zerovalent iron (nZVI)-based materials are increasingly being applied in environmental remediation, thereby lead to their exposure to aquatic and terrestrial biota. However, little is known regarding the toxic effects of surface-modified nZVI on multiple species in the ecosystem. In this study, we systematically compared the toxicities of different forms of nZVIs, such as bare nZVI, carboxymethyl cellulose (CMC)-stabilized nZVI, tetrapolyphosphate (TPP)-coated nZVI and bismuth (Bi)-doped nZVI, on a range of aquatic and terrestrial organisms, including bacteria (Escherichia coli and Bacillus subtilis), plant (Arabidopsis thaliana), water flea (Daphnia magna) and earthworm (Eisenia fetida). The Bi- and CMC-nZVI induced adverse biological responses across all the test systems, except E. fetida, varying from cell death in E. coli and B. subtilis to inhibition of the physiological states in D. magna and A. thaliana. The particle characterization under exposure conditions indicated that the surface modification of nZVI played a significant role in their toxicities by changing their physicochemical properties. The underlying mechanisms by which nZVI induces toxicity might be a combination of oxidative stress and another mechanism such as cell membrane disruption, chlorosis and hypoxia. Overall, our findings could provide important implications for the development of environment-friendly nanomaterials and direct further ecotoxicological researches regarding interspecies exploration.
1. Introduction Nanoscale zerovalent iron (nZVI) is the only commercially available engineered nanoparticle that can be injected into the contaminated soil and groundwater on a large scale. Ample literature exists regarding the splendid application prospects of nZVI for the removal of various contaminants. These beneficial properties have led to a rapid increase of site remediation with nZVI in the USA and Europe (Mueller et al., 2012). The wide range of applications of nZVI in in situ remediation has led to extensive work on nZVI surface modification due to its low transportability and rapid oxidation. Various modification strategies, e.g., secondary metal deposition, sulfidation and polymer coating, have been reported in the literature to overcome the shortcomings of bare nZVI (Kim et al., 2011; Liu et al., 2014; Zhao et al., 2011). Therefore, onethird of the sites were applied the surface-modified nZVIs for the real contaminated soil and groundwater remediation (Karn et al., 2009). However, conformational changes in nZVI can affect particle mobility and reactivity, subsequently altering the fate of nZVI in biological
systems and potentially posing risk. Despite the increasing use of surface-modified nZVIs in the field and the intense interest in nanomaterial safety, only a few studies have addressed the ecotoxicological impact of surface-modified nZVIs. Several studies demonstrated that certain surface coatings with carboxymethyl cellulose (CMC) or humic acid alleviated nZVI toxicity toward bacteria by limiting particle adhesion to cells (Li et al., 2010). In contrast, bimetallic nZVI in oxygenic water enhanced the production of reactive oxygen species (ROS) and thus its cytotoxicity (Kim et al., 2014a). The toxic action of nanomaterials varies considerably depending on the test organisms in addition to their physicochemical properties and concentrations (Nel et al., 2009). Thus far, most of the studies have been focused on the toxicity of nZVI on each species in the ecosystem. In addition, compared to the amount of literature on bacteria, available data on the adverse biological impact on a range of aquatic and terrestrial organisms is limited. It was found that the toxic effects on medaka fish were noticeable at the nZVI concentrations of 50–100 mg L−1 (Chen et al., 2012), whereas higher order organisms, i.e., plant and earthworm, required relatively higher nZVI
⁎
Corresponding author. E-mail address: [email protected] (Y.-S. Chang). 1 These two authors contributed equally to this work. https://doi.org/10.1016/j.ecoenv.2018.07.099 Received 26 November 2017; Received in revised form 12 July 2018; Accepted 24 July 2018 0147-6513/ © 2018 Elsevier Inc. All rights reserved.
Ecotoxicology and Environmental Safety 163 (2018) 436–443
H. Yoon et al.
bacterial toxicity was examined in terms of cell viability, intracellular ROS and cellular damage. Cell viability was assessed using Fluorescein Diacetate (FDA). The bacterial suspensions were treated with various concentrations of nZVIs (1–500 mg L−1) in distilled water and incubated for 24 h in a shaking incubator at 150 rpm. Control and nZVIstreated bacteria suspensions were washed, and then FDA was added to the suspensions to a final concentration of 10 µM. The samples were incubated for 30 min and the fluorescent signal was measured at excitation wavelength of 485 nm and emission wavelength of 530 nm using Fluorescence spectrometer (Fluoromax 4, Horiba). The cell viability was calculated by comparing between the control group and nZVIs-treated bacteria after 24 h of incubation. The equation for the cell viability (%) for Fig. 1:
concentrations for induction of observable toxicities. According to ElTemsah and Joner, the median effective concentrations estimated from the avoidance data of earthworm species were > 500 mg nZVI kg−1 (El-Temsah and Joner, 2012a). The phytotoxicity of nZVI was assessed in Lepidium sativum and Sinapis alba, with no significant inhibition on seed germination at approximately 300 mg L−1 (Libralato et al., 2016). Extrapolating effects of toxicants from a single test species to the ecosystem as a whole is an essential part of environmental risk assessment. However, the limited number of test species can’t take into account the interactions between the toxicants. nZVI can be transferred from injection sites to soil and groundwater streams via waterborne transportation (Karn et al., 2009; Mueller et al., 2012), their toxic effects should be considered at multiple levels of organization along a potential exposure route to provide a more realistic risk assessment. In the present study, the toxicities of surface-modified nZVIs in various aquatic and terrestrial organisms (bacteria, water flea, plant and earthworm) were systematically compared with bare nZVI through various ecotoxicity tests. The test species and endpoints were chosen under the worst assumption that nZVI has spread to aquatic and terrestrial environments at high concentrations. Three different types of surface-modified nZVIs [CMC-stabilized nZVI (CMC-nZVI), tetrapolyphosphate-coated nZVI (TPP-nZVI) and bismuth-iron bimetallic nanoparticle (Bi-nZVI)] were employed because of their effectiveness and attentions, as reported in previous studies (Bokare et al., 2010; Gong et al., 2015; Kim et al., 2015a). Additionally, we aimed to determine how the toxicity relates to the test species and the physicochemical properties of each surface-modified nZVIs.
[FDA]sample ⎞ Cell viability inhibition (%) = ⎛1− × 100 ⎝ [FDA]control ⎠ ⎜
⎟
The intracellular ROS in nZVIs-treated cells were measured using 2’, 7’-dichlorofluorescein diacetate (DCF-DA). The experimental method was same as the FDA method, but different incubation time (10 min) and dye concentration (2 µM) were applied. The fluorescent signal was measured at an excitation wavelength of 488 nm. To check the cellular damage by nZVIs, Fourier transform infrared spectroscopy (FTIR) analysis has been used. The bacterial suspensions were treated with 100 mg L−1 of nZVIs in distilled water. After 24 h of treatment, the suspensions were centrifuged to collect the pellets. The bacterial pellets were dried overnight at 60 °C in the vacuum oven and analyzed using FTIR (iS50, Thermo) in 800–3000 cm−1 range. For TEM imaging, the nZVI treated and control cells were fixed in formaldehyde and glutaraldehyde, dehydrated in ethanol, and embedded into EMBed 812 resin and propylene oxide. After sample was sectioned by an ultratome, analysis of bacterial sections was carried out using the TEM (H-7600, Hitachi).
2. Materials and methods 2.1. Preparation of nZVIs Four different nanomaterials were used in this study. Bare nZVI (Nanofer STAR) was commercially obtained from NANOIRON (Czech Republic). CMC-, TPP- and Bi-nZVIs were prepared using the same procedure as described in previous works (Bokare et al., 2010; Gong et al., 2015; Kim et al., 2015a). CMC- and Bi-nZVI were produced through the reduction of a Fe(II)–CMC mixture ([CMC] / [Fe(II)] = 0.2) and Bi-Fe mixture ([Bi(III)] / [Fe(II)] = 0.07) with NaBH4, respectively. TPP-nZVI was prepared by mixing an aliquot of nZVI with TPP in a pH-adjusted solution.
2.3.2. Plant The seeds of A. thaliana ecotype Columbia were sterilized using 70% ethanol with 20% Clorox and washed five times with distilled water. Seeds were germinated in hydroponics or soil for 30 days in a plant growth chamber (DS-330DHL, Daewon Sci., Korea) at 22–24 °C and 60% humidity with a 16:8 h (light: dark) photoperiod. In the hydroponic culture, the nZVI slurry was mixed with 1/2 Murashige and Skoog medium (MS; Duchefa Biochemie, Netherlands) at pH 5.8 (adjusted with 0.1 N KOH, Sigma-Aldrich, USA). In the soil culture, the slurry of nZVI was added to the test soil (purchased from Hungnong Co., Korea) to make the concentration 500 mg kg−1. The toxicity in A. thaliana was assessed based on the following endpoints: seed germination (%), root lengths (mm), relative biomass (%), H2O2 imaging, and iron accumulation. After 30 days of growth, weights of the harvested plants were quantified. The weight measurements were conducted on the shoots except roots. Average weights of three groups consisting of ten individual plants were measured. For H2O2 detection, segments of roots and leaves were incubated in 25 µM DCF-DA for 30 min in the dark and imaged using a confocal microscope (FV 1000, Olympus) in 485–530 nm wavelength. To determine the iron accumulation in tissues, freeze-dried plant samples were dissolved in 60% HNO3 at 120 °C overnight. After diluting the sample, the element contents were measured using an ICP-OES.
2.2. Particle characterization The particle size, morphology and elemental analysis of the products were analyzed using high-resolution transmission electron microscopy (HR-TEM, JEM-2010, Jeol Ltd) coupled with electron energy loss spectroscopy (EELS). Hydrodynamic diameter and ζ-potential measurements of nZVIs dispersed in deionized (DI) water or media were performed on a zetasizer (Nano ZS90, Malvern Instruments). Surface area analysis was conducted using the Brunauer-Emmett-Teller (BET) method with a micropore physisorption analyzer (ASAP-2020 M, Micrometrics). The dissolved iron concentration of each sample was measured with an inductively coupled plasma-optical emission spectrometer (ICP-OES, Optima 7000 DV, PerkinElmer). 2.3. Test organism and toxicity assessment
2.3.3. Water flea D. magna was cultured in M4 medium prepared according to the Organization for Economic Co-operation and Development (OECD) test guideline (pH; 7.8 ± 0.1, hardness; 250 ± 25 mg CaCO3 L−1) at 20 ± 1 °C with a 16:8 h (light:dark) photoperiod. The green algae, Pseudokirchneriella subcapitata and a mixture (yeast and CEROPHYLL®) were provided as food on a daily basis. The acute toxicity tests for D. magna were conducted in accordance with the OECD Test Guidelines
The model organisms used in the present study were two bacterial strains (Gram-positive Bacillus subtilis and Gram-negative Escherichia coli), a water flea (Daphnia magna), a plant (Arabidopsis thaliana) and an earthworm (Eisenia fetida). 2.3.1. Bacteria E. coli and B. subtilis (from the Korean Agricultural Culture Collection) were grown aerobically at 37 °C and 30 °C, respectively. The 437
Ecotoxicology and Environmental Safety 163 (2018) 436–443
H. Yoon et al.
media could promote aggregation of the particles (Tiraferri et al., 2008). The point of zero charge (PZC) of the nZVI samples varied from approximately pH 4–9.5. The iron contents of Nanofer, CMC-nZVI, TPPnZVI and Bi-nZVI were 79.3%, 24.3%, 13.7% and 75.3% (w/w), respectively. To investigate the iron dissolution from the nanoparticle in aqueous solutions, the production of total dissolved iron was monitored in water (Fig. S1 (B)) Nanofer, CMC-nZVI and Bi-nZVI showed insignificant amounts of dissolved iron for 48 h, which was consistent with the previous finding that the Fe2+ produced by nZVI corrosion remained insoluble as iron oxides (Kanel et al., 2005). However, TPP-nZVI produced approximately 3 mg L−1 (approximately 20% of TPP-nZVI) of dissolved iron after 48 h because of the iron-chelating character of TPP (Kim et al., 2015a).
202 (OECD, 1984). Exposure media was fully aerated M4 media without ethylenediaminetetraacetic acid (EDTA). Toxicity tests comprise of five concentrations of test solution and one control with four replicates. For each exposure, five randomly selected neonates (≤ 24 h old) were placed in a glass beaker containing 40 mL of test solutions. Immobilization of the test species was assessed after 48 h of exposure. Neonates not able to swim within 15 s after gentle agitation were considered to be immobilized. The median effective concentration (EC50) values for D. magna as well as their associated 95% confidence intervals (95% CI) were calculated accordingly. Temporal changes of pH, oxidation-reduction potential (ORP), dissolved oxygen (DO) and concentration of iron species in the modified nZVIs-containing M4 solutions were monitored following Chen et al. (2012). Daphnids (n = 6) were placed in nZVI suspensions concentrated at 1 or 5 mg L−1 to determine an individual oxidative stress response using confocal fluorescent microscopy in 485–530 nm wavelength. The samples were incubated for 4 h at 50 µM DCF-DA concentration. Relative fluorescence intensity was recorded and analyzed using ImageJ. An equivalent number of daphnids as negative controls were also included without nZVI.
3.2. Toxicity to bacteria The antibacterial effects of selected nZVIs, represented by the decrease of bacterial viability, were investigated using Fluorescein diacetate (FDA) fluorescence staining at different nZVI concentrations (1–500 mg L−1). As observed in Fig. 1, the nZVI treatment significantly decreased the viabilities of the two bacterial species in a dose-response manner. Of these, Bi-nZVI exhibited the highest toxicity in the range of all concentrations. This Bi-nZVI toxicity might be attributed to the presence of bismuth on the nZVI surface. Similar findings have shown that zerovalent bismuth NPs cause strong antimicrobial effects by the inhibition of biofilm formation (Hernandez-Delgadillo et al., 2012). On the other hand, the growth inhibition by CMC-nZVI rapidly increased depending on the nanoparticle concentrations. The main toxicological factor of CMC-nZVI toxicity would be the high dispersibility and inhibition of aggregation, thereby allowing favorable contact with bacterial cells (Chen et al., 2011). Previous studies have reported that oxidative stress was the major toxicological mechanism after exposure to nZVI (Sacca et al., 2014). 2′,7′–dichlorofluorescein diacetate (DCF-DA) staining was used to measure the intracellular ROS level as an indicator of the oxidative stress. Strong fluorescent signals were observed in E. coli treated with all nZVIs (Fig. 2), indicating that those NPs could significantly induce intracellular ROS. In case of B. subtilis, only Bi-nZVI significantly elevated the ROS level. The high ROS level detection could be explained by the nZVI promoting free radical generation via the Fenton reaction inside the cell (Ševců et al., 2011). However, the ROS level does not fully explain why nZVI tended to have negative effects on bacterial viability. Kim et al. reported that bimetallic Fe-Pd and Fe-Pt induced high ROS levels in E. coli, whereas their antibacterial activities were not significant (Kim et al., 2014a). It appears that bacterial cells possess sufficient antioxidant systems to defense oxidative stress from NPs. Therefore, an additional toxic mechanism might exist beside the oxidative stress. In the present study, B. subtilis was found to be more tolerant to the tested nZVIs than E. coli. These differences toward E. coli (Gram-negative) and B. subtilis (Gram-positive) are likely to be originated from the different cell wall structures and compositions between the two bacterial species. Furthermore, the lower ROS level in B. subtilis could be due to the presence of nitric oxides that boost the catalase activity involved in H2O2 scavenging. To date, the bacterial nitric oxide synthases have been known to exist in Gram-positive bacteria only (Gusarov and Nudler, 2005). The current hypothesized toxic mechanisms also include structural change and disruption of cell membranes (Jiang et al., 2010). To provide insight into the mechanism of interaction between NPs and cell membranes, the bacterial cells were analyzed with Fourier transform infrared spectroscopy-diffuse reflectance (FTIR-DRIFT) to characterize the change in biomolecule composition. IR spectra of the bacterial cells exposed at 100 mg L−1 nZVI are shown in Fig. S2. The fatty acid, protein and peptide and carbohydrate regions were characterized by the peaks in the wavelengths of 100–2800 cm−1, 1800–1500 cm−1 and
2.3.4. Earthworm E. fetida were purchased from Carolina Biological Supply (Burlington, USA). The earthworms were acclimated in plastic boxes containing moist bedding and food (Magic Worm Food, USA). The acute toxicity tests for E. fetida were conducted in accordance with the OECD Test Guidelines 207, (OECD, 2004). Standardized OECD soil substrate was used as exposure medium which consists of industrial fine sand, kaolin clay, and sphagnum peat in a 7:2:1 ratio, respectively and were mixed with nZVI. Four replicates were carried out and each group consisted of ten worms. For each exposure, adult earthworms (at least 2 months old with individual weights of 350–450 mg) were selected. 500 g (dry weight) of test medium was placed into each crystallizing dishes and earthworms were placed on the test medium surface. Treated earthworms were maintained at a controlled temperature of 20 ± 2 °C with 80–85% humidity under 400–800 lx of constant light. Mortality, abnormal behavior, and weight change of earthworms were assessed after 14 days of exposure. The wet mass of each exposure group (ten earthworms) was weighed and expressed as average weight of earthworms. To quantify the body burden of total iron in earthworms, whole bodies of each exposure group were placed in a digestion vessel with 8 mL of 60% HNO3 and 2 mL of H2O2. After complete digestion (heating at 120 °C for 4 h), the solution was analyzed by ICPOES. 2.4. Statistical analysis Each treatment was conducted in at least triplicate. Data were presented as the mean ± standard deviation (SD) and analyzed by oneway analysis of variance (ANOVA) with the Tukey HSD test using IBM SPSS statistics 20. Probability values (P) < 0.05 were considered statistically significant. The 48 h EC50 for the D. magna acute toxicity test were estimated by the Spearman-Karber method using CETIS version 1.8.7.15 (Tidepool Scientific Software, USA). 3. Results and discussion 3.1. Characteristics of nZVI materials The physicochemical properties of each surface-modified nZVIs are summarized in Table 1. Although the primary sizes of four nZVIs were determined to be approximately 40–116 nm by TEM (Fig. S1 (A)), the particles in DI water rapidly aggregated with an average hydrodynamic diameter range of approximately 300–770 nm. All nanoparticles suspended in M4 media had significantly larger hydrodynamic diameters compared to those in water since the high salt concentrations in M4 438
Ecotoxicology and Environmental Safety 163 (2018) 436–443
H. Yoon et al.
Table 1 Physicochemical characterization of selected nanoscale zerovalent iron (nZVI). Results are shown as the means ± SD (n = 3). Types of nZVI
Coating
Surface Area (BET) (m2/g)
Primary size (TEM) (nm)a
Hydrodynamic diameter (nm)b
Zeta potential (mV)b
Point of zero charge (PZC)c
Iron contents (%)
Nanofer
–
14.3
59.8 ± 1.3
79.3 ± 0.5
Carboxy methyl cellulose Tetrapolyphosphate Bismuth
19.5
39.8 ± 1.7
<4
24.3 ± 8.9
11.3
72.2 ± 2.7
<4
13.7 ± 0.2
11.0
116.7 ± 5.4
− 5.8 ± 0.6 (DW) − 7.4 ± 0.4 (M4) − 32.1 ± 1.8 (DW) − 9.9 ± 0.1 (M4) − 62.6 ± 3.4 (DW) − 21.8 ± 0.4 (M4) 31.4 ± 0.9 (DW) − 2.8 ± 0.4 (M4)
5.5
CMC-nZVI
775.4 ± 20.2 (DW) 4365.0 ± 245.7 (M4) 443.8 ± 34.4 (DW) 3383.7 ± 619.2 (M4) 701.2 ± 29.1 (DW) 2466 ± 105.1 (M4) 313.5 ± 20.7 (DW) 4155.3 ± 513.2 (M4)
9.5
75.3 ± 15.4
TPP-nZVI Bi-nZVI
a b c
The values of each NPs size represent the mean of the 10 replicates. Hydrodynamic diameter and zeta potential were measured with 10 mg L−1 nZVIs in deionized water or M4 media. Point of zero charge was measured with 10 mg L−1 nZVIs in 10 mM NaCl.
Fig. 2. ROS levels in two species of bacterial cells after 24 h exposure of 100 mg L−1 nZVI. Results are shown as the means ± SD (n = 3). Different letters represent significant differences among the different groups (one-way ANOVA with Tukey HSD post-hoc test).
penetration into the cell membrane, thereby inducing high toxicity. The rough surface of Bi-nZVI promoted adhesion (Fig. S1 (A)), leading to easy translation into the cell (Nel et al., 2009). To compare actual images of bacterial interactions with nZVIs, the exposed bacteria were visualized by TEM (Fig. S3). Bare-nZVI, CMC-nZVI and Bi-nZVI attached to the cells caused serious damages to the membrane integrity, eventually leading to death. In the protein region, there were changes caused by binding of NPs or ions released from NPs (Eckhardt et al., 2013; Faghihzadeh et al., 2016). For TPP-nZVI, only the peaks in the protein region were shifted. Moreover, in the TEM images, only TPPnZVI was aggregated out of the cell membrane without any penetration. Previous studies reported a positive correlation between the dissolved Fe ions and the nZVI toxicity to bacteria (Auffan et al., 2008). TPP-nZVI showed a higher dissolution than other nZVIs in DI water (Fig. S1 (B)). The Fe ions released from TPP-nZVI might interact with amino acids, thereby inducing the alteration of proteins or enzymes. Finally, in the carbohydrate region, ROS could cause the alternation of C-O-C and C-O vibrations in the sugar rings of polysaccharides depicted in Fig. 2 (Eckhardt et al., 2013; Faghihzadeh et al., 2016). In case of B. subtilis, however, only Bi-nZVI caused an obvious change to the spectrum, which agreed well with the results of the cell viability and ROS level.
Fig. 1. Viability of (A) Escherichia coli and (B) Bacillus subtilis after exposure to tested nZVIs in deionized (DI) water for 24 h. Results are shown as the means ± SD (n = 3).
1200–900 cm−1, respectively. Overall, the tendency of the spectra alternation was relative to the order of toxicity. For example, a remarkable difference was found in the bacteria treated by Bi-nZVI, which had a high inhibition effect on growth rate. In detail, the bands of E. coli cells exposed to bare nZVI, CMC-nZVI and Bi-nZVI showed almost complete disappearance of spectra in the fatty acid region, possibly due to the strong binding of NPs with –CH groups of the membrane, which changed the fluidity of cell wall lipids (Faghihzadeh et al., 2016). In addition, there is high probability of interaction between positively charged Bi-nZVI and negatively charged bacteria membrane via electrostatic attraction. The shape and small hydrodynamic size of nZVI also contributed to
3.3. Toxicity to plant To investigate the influence of the tested nZVIs on plant growth, the physiological attributes such as seed germination, root length and biomass of a model plant, A. thaliana, were measured. As illustrated in Fig. S4 (A), the tested nZVIs did not have deleterious influences on the 439
Ecotoxicology and Environmental Safety 163 (2018) 436–443
H. Yoon et al.
treated plant. It was expected that the greater amount of H2O2 accumulated by Bi-nZVI should adversely affect plant growth via oxidative stress. The chlorosis is caused by iron deficiency in plants. Therefore, iron accumulation in plant tissues was analyzed to identify the bioavailability of nZVI. The iron contents in shoots increased significantly by all tested nZVIs (Fig. S4 (C)). In particular, the Bi-nZVI treated plants absorbed an excessive amount of iron compared to the control, but yellowing of the leaves and chlorosis were also observed. Similar symptoms were studied by Wang et al. (2016), where the higher concentration of nZVI induced iron deficiency in rice because the transport of active iron from the root to the shoot was blocked (Wang et al., 2016). As a result, chlorosis cannot be explained solely by the total concentration of iron, because excessive iron enters the plant but cannot be confirmed in what form it is stored inside. Furthermore, the results indicated that the phenomena observed in the present study did not originate from the dissolved ions but from the NPs themselves. Zhao et al. (2012) found that the additives released from the surface of the NPs played an important role in ion bioavailability to plants. Therefore, further study of the uptake, translocation and transformation of the tested nZVIs into plants is highly needed.
germination percentages of plant. However, the various nZVI types exhibited different effects on the growth of plants. In a previous study, El-Temsah et al., examined the effects of different types and concentrations of nZVI on the germination of various pasture plants such as flax, barley and ryegrass. They found that although there was no negative effect on the germination percentage up to 250 mg L−1, no seeds were able to germinate at 1000–2000 mg L−1 of poly(acrylic acid) coated nZVI (El-Temsah and Joner, 2012b). In addition, the germination percentages of rape and cabbage were also found to significantly decrease by 10% and 40%, respectively at 1500 mg kg−1 of CMC-nZVI (Wang et al., 2014). Those findings suggest that the phytotoxic responses resulting from nZVI are more likely to depend on not only the type and concentration of nZVI but also the plant species (Du et al., 2016). The present study also obtained a similar result of root elongation and biomass depend on the type of nZVI. The root length of plants treated with CMC-nZVI and Bi-nZVI decreased by approximately 40% and 69% compared to the control plant (non-treated), respectively (Fig. S4 (B)). For biomass, CMC and Bi-nZVI similarly inhibited the growth of plant by at least 40% and Bi-nZVI, in particular, might cause severe damages to the plant in a hydroponic system (Fig. 3). Bismuth NP was reported to interact with the sulfhydryl groups (SH-) present in plant membrane proteins, thereby causing a DNA damage (Liman, 2013). In addition to Bi-nZVI, CMC-nZVI also showed an inhibitory effect on the plant root growth (Wang et al., 2014). On the other hand, bare nZVI was observed to be capable of increasing root length and significantly promoting plant growth. In the case of bare nZVI, the results agreed fairly well with our previous studies demonstrating that bare nZVI triggered root elongation and showed an increase of biomass (i.e., leaf area) by activating stomatal opening (Kim et al., 2014b, 2015b). H2O2 has essential roles in plant metabolism such as root development and photosynthesis (Slesak et al., 2007). However, the concurrent abiotic stress related to the overproduction of H2O2 is a concern, as to be potentially damaging to plant growth (Cheeseman, 2007). To utilize H2O2 as a signaling molecule, the non-toxic levels must be maintained in a delicate balancing act between H2O2 production and scavenging (Hossain et al., 2015). To elucidate the toxicity mechanism caused by the tested nZVIs on plants, imaging of H2O2 was performed using DCFDA on the roots and leaves. Fig. 4 illustrates that all plants, except the TPP-nZVI treated plant, showed higher fluorescent intensities in the roots compared to the control. The trend of the results shown in the roots was approximately similar to the bacterial results (Fig. 2). However, the results obtained from the leaves were of interest and a significantly higher level of fluorescence was observed in the Bi-nZVI
3.4. Toxicity to water flea The concentration immobilizing 50% of neonates after 48 h of nZVI exposure was expressed as the effective concentration (EC50). Bi-nZVI (EC50 = 42.5 mg L−1) and CMC-nZVI (EC50 = 45.6 mg L−1) appeared to be the most toxic to water flea, whereas no immobilization occurred when exposed to bare and TPP-nZVI (EC50 > 100 mg L−1). In oxygenated water, nZVI reacts with O2 to generate Fe2+ and oxidizing intermediates (i.e., H2O2, •OH) (Harada et al., 2016), which may enter the cells and induce the oxidative stress. In the existing literature, oxidative stress has been regarded as an important factor in toxicity to daphnids (Dalai et al., 2013). To confirm that oxidative stress induced by nZVI in Daphina magna, the formation of ROS was measured by the DCF-DA method that was previously used in the bacterial and plant ROS determination. As shown in Fig. 5, similar to that of the bacteria and plant results, all daphnids, except for the TPP-nZVI treated group, had significant fluorescent intensities compared to that of the control even at a lower concentration than the no observed effect concentration (NOEC). Similar to the bacterial results, these results do not fully explain why nZVI tend to produce negative effects on daphnid immobilization. The temporal fluctuations in hydrochemical variables of the test medium (i.e., pH, oxidation-reduction potential (ORP) and dissolved oxygen (DO)) indicated that the exposure to nZVI would affect the NP bioavailability and toxicity in an aquatic environment. As shown in Fig. S5, the pH values of all nZVI solutions slightly increased for 4 h and they were stabilized at pH 8.0–8.2. The addition of Bi- and CMC-nZVI led to decreases of ORP and Bi-nZVI gave more negative ORP than CMC-nZVI for 48 h. Conversely, the initial ORP values did not change significantly with bare and TPP-nZVI. The pattern of DO changes in four nZVI suspensions was similar to what we observed in the ORP measurement. The DO depletion at the initial phase was most pronounced in the Bi-nZVI suspension, reaching almost zero. The observed higher toxicities of Bi- and CMC-nZVI were apparently related to the hypoxia. Our data were in agreement with the previous studies on lethal and critical O2 concentrations, down to 3 mg L−1, for aerobic metabolism in D. magna. The DO depletion was the stressor to daphnids because anthropogenic impacts of Bi- and CMC-nZVI induced the hypoxia in aquatic systems, resulting in the change of physiological processes (Kobayashi and Hoshi, 1984). Although hypoxia appeared to the main causative factor, the contribution of other factors cannot be overlooked. Several studies have demonstrated that the dissolved form of NPs is a main cause of toxicity to aquatic species (Griffitt et al., 2008). To determine the relative
Fig. 3. Relative growth rate of nZVI-treated Arabidopsis thaliana seedlings after 30 days in a hydroponic and soil system. Results are shown as the means ± SD (n = 3). Different letters represent significant differences among the different groups (one-way ANOVA with Tukey HSD post-hoc test). 440
Ecotoxicology and Environmental Safety 163 (2018) 436–443
H. Yoon et al.
Fig. 4. H2O2 imaging in Arabidopsis thaliana root and leaf treated with 100 mg L−1 nZVI.
significant difference in the body weight of E. fetida was observed between the nZVI-treated soils and the control (Fig. 6), indicating that all the nZVIs did not exhibit acute toxicity to earthworm. In addition, the iron concentration in tissue was measured after 14 d exposure to nZVIs. As depicted in Fig. 6(B), the tissue iron concentration varied slightly depending on the nZVI type, ranging from 80 to 150 mg kg−1. The relatively low uptake of total iron even at high exposure concentrations might be attributed to the sorption of Fe ions to soil or the aggregation of NPs to themselves. As previously noted in literature, the higher concentrations of iron NPs did in several cases influence the measurement principles behind the earthworm toxicity tests because of their aggregation, agglomeration and ultimately sedimentation of nZVI (Hjorth et al., 2017). However, despite the limitations and challenges in the terrestrial ecotoxicity test, these attempts can still provide useful information for developing a new catalyst for environmental remediation. Fig. 5. Level of ROS in Daphnia magna after 24 h exposure of 1 and 5 mg L−1 nZVI. Results are shown as the means ± SD (n = 3). Different letters represent significant differences among the different groups (one-way ANOVA with Tukey HSD post-hoc test).
4. Conclusions In the present work, we have tested the potential toxicity of surfacemodified nZVIs on the various terrestrial and aquatic species (bacteria, plant, water flea and earthworm). Unlike previous studies, we focused on the toxic effects of surface-modified nZVIs on multiple species in the ecosystem. Among the tested nZVIs, Bi- and CMC-nZVI caused strong toxic effects on all of the tested organisms, except for earthworm. It is worth noting that the oxidative stress via the Fenton reaction appeared to mainly contribute to the observed toxicity endpoints. In addition, membrane disruption, chlorosis and hypoxia could be additional toxic factors, depending on the test species. The commonalities in the toxicity of tested nZVIs on the diverse trophic groups suggested that the interspecies comparison would be critical for the risk assessment of engineered nanomaterials. In addition, the results emphasized the importance of surface modifiers in the toxic effects of nZVI, which deserved further evaluations given that a variety of surface-modified nZVIs are currently available at the on-site remediation. Finally, this
contribution of Fe ions to the observed toxic effects, the release profiles of dissolved iron from nZVI at 100 mg L−1 were investigated in M4 media. It should be noted that the dissolution behavior of nZVI can vary depending on the particle composition and exposure conditions (Adeleye et al., 2013). In the present study, the dissolved iron concentrations for each nZVI ranged from 0.2 to 3 mg L−1 (data not shown), which was far below the 48 h EC50 (59.7 mg L−1) value of the Fe2+ solution (Baumann et al., 2014). Thus, the overall toxicity of nZVI to Daphnia magna was not related to the direct effects of free iron, but the combination of hypoxia and oxidative stress. 3.5. Toxicity to earthworm Despite the high concentration of nZVI (1000 mg kg−1) applied, no 441
Ecotoxicology and Environmental Safety 163 (2018) 436–443
H. Yoon et al.
101, 6354–6360. Cheeseman, J.M., 2007. Hydrogen peroxide and plant stress: a challenging relationship. Plant Stress 1, 4–15. Chen, J., Xiu, Z., Lowry, G.V., Alvarez, P.J., 2011. Effect of natural organic matter on toxicity and reactivity of nano-scale zero-valent iron. Water Res. 45, 1995–2001. Chen, P.-J., Tan, S.-W., Wu, W.-L., 2012. Stabilization or oxidation of nanoscale zerovalent iron at environmentally relevant exposure changes bioavailability and toxicity in medaka fish. Environ. Sci. Technol. 46, 8431–8439. Dalai, S., Pakrashi, S., Chandrasekaran, N., Mukherjee, A., 2013. Acute toxicity of TiO2 nanoparticles to Ceriodaphnia dubia under visible light and dark conditions in a freshwater system. PLoS One 8, e62970. Du, W., Tan, W., Peralta-Videa, J.R., Gardea-Torresdey, J.L., Ji, R., Yin, Y., Guo, H., 2016. Interaction of metal oxide nanoparticles with higher terrestrial plants: physiological and biochemical aspects. Plant Physiol. Biochem. 110, 210–225. Eckhardt, S., Brunetto, P.S., Gagnon, J., Priebe, M., Giese, B., Fromm, K.M., 2013. Nanobio silver: its interactions with peptides and bacteria, and its uses in medicine. Chem. Rev. 113, 4708–4754. El-Temsah, Y.S., Joner, E.J., 2012a. Ecotoxicological effects on earthworms of fresh and aged nano-sized zero-valent iron (nZVI) in soil. Chemosphere 89, 76–82. El-Temsah, Y.S., Joner, E.J., 2012b. Impact of Fe and Ag nanoparticles on seed germination and differences in bioavailability during exposure in aqueous suspension and soil. Environ. Toxicol. 27, 42–49. Faghihzadeh, F., Anaya, N.M., Schifman, L.A., Oyanedel-Craver, V., 2016. Fourier transform infrared spectroscopy to assess molecular-level changes in microorganisms exposed to nanoparticles. Nanotechnol. Environ. Eng. 1, 1–16. Gong, J., Lee, C.-S., Chang, Y.-Y., Chang, Y.-S., 2015. Novel self-assembled bimetallic structure of Bi/Fe 0: the oxidative and reductive degradation of hexahydro-1, 3, 5trinitro-1, 3, 5-triazine (RDX). J. Hazard. Mater. 286, 107–117. Griffitt, R.J., Luo, J., Gao, J., Bonzongo, J.C., Barber, D.S., 2008. Effects of particle composition and species on toxicity of metallic nanomaterials in aquatic organisms. Environ. Toxicol. Chem. 27, 1972–1978. Gusarov, I., Nudler, E., 2005. NO-mediated cytoprotection: instant adaptation to oxidative stress in bacteria. Proc. Natl. Acad. Sci. USA 102, 13855–13860. Harada, T., Yatagai, T., Kawase, Y., 2016. Hydroxyl radical generation linked with iron dissolution and dissolved oxygen consumption in zero-valent iron wastewater treatment process. Chem. Eng. J. 303, 611–620. Hernandez-Delgadillo, R., Velasco-Arias, D., Diaz, D., Arevalo-Niño, K., Garza-Enriquez, M., De la Garza-Ramos, M.A., Cabral-Romero, C., 2012. Zerovalent bismuth nanoparticles inhibit Streptococcus mutans growth and formation of biofilm. Int. J. Nanomed. 7, e2113. Hjorth, R., Coutris, C., Nguyen, N.H., Sevcu, A., Gallego-Urrea, J.A., Baun, A., Joner, E.J., 2017. Ecotoxicity testing and environmental risk assessment of iron nanomaterials for sub-surface remediation–Recommendations from the FP7 project NanoRem. Chemosphere 182, 525–531. Hossain, M.A., Bhattacharjee, S., Armin, S.-M., Qian, P., Xin, W., Li, H.-Y., Burritt, D.J., Fujita, M., Tran, L.-S.P., 2015. Hydrogen peroxide priming modulates abiotic oxidative stress tolerance: insights from ROS detoxification and scavenging. Front. Plant Sci. 6, 420. Jiang, W., Yang, K., Vachet, R.W., Xing, B., 2010. Interaction between oxide nanoparticles and biomolecules of the bacterial cell envelope as examined by infrared spectroscopy. Langmuir 26, 18071–18077. Kanel, S.R., Manning, B., Charlet, L., Choi, H., 2005. Removal of arsenic (III) from groundwater by nanoscale zero-valent iron. Environ. Sci. Technol. 39, 1291–1298. Karn, B., Kuiken, T., Otto, M., 2009. Nanotechnology and in situ remediation: a review of the benefits and potential risks. Environ. Health Persp. 117, 1813–1831. Kim, E.-J., Kim, J.-H., Azad, A.-M., Chang, Y.-S., 2011. Facile synthesis and characterization of Fe/FeS nanoparticles for environmental applications. ACS Appl. Mater. Interfaces 3, 1457–1462. Kim, E.-J., Le Thanh, T., Chang, Y.-S., 2014a. Comparative toxicity of bimetallic Fe nanoparticles toward Escherichia coli: mechanism and environmental implications. Environ. Sci.: Nano. 1, 233. Kim, H.-H., Lee, H., Kim, H.-E., Seo, J., Hong, S.W., Lee, J.-Y., Lee, C., 2015a. Polyphosphate-enhanced production of reactive oxidants by nanoparticulate zerovalent iron and ferrous ion in the presence of oxygen: yield and nature of oxidants. Water Res. 86, 66–73. Kim, J.H., Lee, Y., Kim, E.J., Gu, S., Sohn, E.J., Seo, Y.S., An, H.J., Chang, Y.S., 2014b. Exposure of iron nanoparticles to Arabidopsis thaliana enhances root elongation by triggering cell wall loosening. Environ. Sci. Technol. 48, 3477–3485. Kim, J.H., Oh, Y., Yoon, H., Hwang, I., Chang, Y.S., 2015b. Iron nanoparticle-induced activation of plasma membrane H(+)-ATPase promotes stomatal opening in Arabidopsis thaliana. Environ. Sci. Technol. 49, 1113–1119. Kobayashi, M., Hoshi, T., 1984. Analysis of respiratory role of haemoglobin in Daphnia magna. Zool. Sci. 1, 523–532. Li, Z., Greden, K., Alvarez, P.J., Gregory, K.B., Lowry, G.V., 2010. Adsorbed polymer and NOM limits adhesion and toxicity of nano scale zerovalent iron to E. coli. Environ. Sci. Technol. 44, 3462–3467. Libralato, G., Devoti, A.C., Zanella, M., Sabbioni, E., Mičetić, I., Manodori, L., Pigozzo, A., Manenti, S., Groppi, F., Ghirardini, A.V., 2016. Phytotoxicity of ionic, micro-and nano-sized iron in three plant species. Ecotoxicol. Environ. Saf. 123, 81–88. Liman, R., 2013. Genotoxic effects of Bismuth (III) oxide nanoparticles by Allium and Comet assay. Chemosphere 93, 269–273. Liu, W.-J., Qian, T.-T., Jiang, H., 2014. Bimetallic Fe nanoparticles: recent advances in synthesis and application in catalytic elimination of environmental pollutants. Chem. Eng. J. 236, 448–463. Mueller, N.C., Braun, J., Bruns, J., Černík, M., Rissing, P., Rickerby, D., Nowack, B., 2012. Application of nanoscale zero valent iron (NZVI) for groundwater remediation in
Fig. 6. (A) Change in weight and (B) Iron accumulation in Eisenia fetida exposed to 1000 mg L−1 nZVI. Results are shown as the means ± SD (n = 3). Different letters represent significant differences among the different groups (one-way ANOVA with Tukey HSD post-hoc test).
research has explored the nZVI toxicity to individual organism levels, but the toxicity to community levels or synergistic impacts needs to be further investigated for eco-friendly remediation using nanomaterials. Acknowledgements This work was supported by ‘‘The GAIA Project” by the Korea Ministry of Environment (RE201402059) and the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No. NRF-2017R1A2B3012681). Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.ecoenv.2018.07.099. References Adeleye, A.S., Keller, A.A., Miller, R.J., Lenihan, H.S., 2013. Persistence of commercial nanoscaled zero-valent iron (nZVI) and by-products. J. Nanopart. Res. 15, 1–18. Auffan, M. l., Achouak, W., Rose, J. r., Roncato, M.-A., Chanéac, C., Waite, D.T., Masion, A., Woicik, J.C., Wiesner, M.R., Bottero, J.-Y., 2008. Relation between the redox state of iron-based nanoparticles and their cytotoxicity toward Escherichia coli. Environ. Sci. Technol. 42, 6730–6735. Baumann, J., Köser, J., Arndt, D., Filser, J., 2014. The coating makes the difference: acute effects of iron oxide nanoparticles on Daphnia magna. Sci. Total Environ. 484, 176–184. Bokare, V., Murugesan, K., Kim, Y.-M., Jeon, J.-R., Kim, E.-J., Chang, Y.S., 2010. Degradation of triclosan by an integrated nano-bio redox process. Bioresour. Technol.
442
Ecotoxicology and Environmental Safety 163 (2018) 436–443
H. Yoon et al.
Tiraferri, A., Chen, K.L., Sethi, R., Elimelech, M., 2008. Reduced aggregation and sedimentation of zero-valent iron nanoparticles in the presence of guar gum. J. Colloid Interface Sci. 324, 71–79. Wang, J., Fang, Z.Q., Cheng, W., Yan, X.M., Tsang, P.E., Zhao, D.Y., 2016. Higher concentrations of nanoscale zero-valent iron (nZVI) in soil induced rice chlorosis due to inhibited active iron transportation. Environ. Pollut. 210, 338–345. Wang, Y., Fang, Z., Kang, Y., Tsang, E.P., 2014. Immobilization and phytotoxicity of chromium in contaminated soil remediated by CMC-stabilized nZVI. J. Hazard. Mater. 275, 230–237. Zhao, L., Peralta-Videa, J.R., Varela-Ramirez, A., Castillo-Michel, H., Li, C., Zhang, J., Aguilera, R.J., Keller, A.A., Gardea-Torresdey, J.L., 2012. Effect of surface coating and organic matter on the uptake of CeO 2 NPs by corn plants grown in soil: insight into the uptake mechanism. J. Hazard. Mater. 225, 131–138. Zhao, X., Lv, L., Pan, B., Zhang, W., Zhang, S., Zhang, Q., 2011. Polymer-supported nanocomposites for environmental application: a review. Chem. Eng. J. 236, 448–463 (170, 381-394).
Europe. Environ. Sci. Pollut. Res. Int. 19, 550–558. Nel, A.E., Mädler, L., Velegol, D., Xia, T., Hoek, E.M., Somasundaran, P., Klaessig, F., Castranova, V., Thompson, M., 2009. Understanding biophysicochemical interactions at the nano–bio interface. Nat. Mater. 8, 543–557. OECD, 1984. OECD Guidelines for the Testing of Chemicals, Section 2: Effects on Biotic Systems 207: Earthworm, Acute Toxicity Tests. OECD Publishing, pp. 1–9. OECD, 2004. OECD Guidelines for the Testing of Chemicals Test No. 202: Daphnia sp. Acute Immobilisation Test. OECD Publishing. Sacca, M.L., Fajardo, C., Martinez-Gomariz, M., Costa, G., Nande, M., Martin, M., 2014. Molecular stress responses to nano-sized zero-valent iron (nZVI) particles in the soil bacterium Pseudomonas stutzeri. PLoS One 9, e89677. Ševců, A., El-Temsah, Y.S., Joner, E.J., Černík, M., 2011. Oxidative stress induced in microorganisms by zero-valent iron nanoparticles. Microbes Environ. 26, 271–281. Slesak, I., Libik, M., Karpinska, B., Karpinski, S., Miszalski, Z., 2007. The role of hydrogen peroxide in regulation of plant metabolism and cellular signalling in response to environmental stresses. Acta Biochim. Pol. 54, 39.
443
Contents lists available at ScienceDirect
Ecotoxicology and Environmental Safety journal homepage: www.elsevier.com/locate/ecoenv
Impact of surface modification on the toxicity of zerovalent iron nanoparticles in aquatic and terrestrial organisms Hakwon Yoona,1, Monmi Pangginga,1, Min-Hee Jangb, Yu Sik Hwangb, Yoon-Seok Changa, a b
T ⁎
Division of Environmental Science and Engineering, Pohang University of Science and Technology (POSTECH), Pohang 37673, Republic of Korea Future Environmental Research Center, Korea Institute of Toxicology (KIT), Jinju 52834, Republic of Korea
A R T I C LE I N FO
A B S T R A C T
Keywords: Surface-modified nZVI Nanotoxicity Ecotoxicity Oxidative stress Physicochemical property
Nanoscale zerovalent iron (nZVI)-based materials are increasingly being applied in environmental remediation, thereby lead to their exposure to aquatic and terrestrial biota. However, little is known regarding the toxic effects of surface-modified nZVI on multiple species in the ecosystem. In this study, we systematically compared the toxicities of different forms of nZVIs, such as bare nZVI, carboxymethyl cellulose (CMC)-stabilized nZVI, tetrapolyphosphate (TPP)-coated nZVI and bismuth (Bi)-doped nZVI, on a range of aquatic and terrestrial organisms, including bacteria (Escherichia coli and Bacillus subtilis), plant (Arabidopsis thaliana), water flea (Daphnia magna) and earthworm (Eisenia fetida). The Bi- and CMC-nZVI induced adverse biological responses across all the test systems, except E. fetida, varying from cell death in E. coli and B. subtilis to inhibition of the physiological states in D. magna and A. thaliana. The particle characterization under exposure conditions indicated that the surface modification of nZVI played a significant role in their toxicities by changing their physicochemical properties. The underlying mechanisms by which nZVI induces toxicity might be a combination of oxidative stress and another mechanism such as cell membrane disruption, chlorosis and hypoxia. Overall, our findings could provide important implications for the development of environment-friendly nanomaterials and direct further ecotoxicological researches regarding interspecies exploration.
1. Introduction Nanoscale zerovalent iron (nZVI) is the only commercially available engineered nanoparticle that can be injected into the contaminated soil and groundwater on a large scale. Ample literature exists regarding the splendid application prospects of nZVI for the removal of various contaminants. These beneficial properties have led to a rapid increase of site remediation with nZVI in the USA and Europe (Mueller et al., 2012). The wide range of applications of nZVI in in situ remediation has led to extensive work on nZVI surface modification due to its low transportability and rapid oxidation. Various modification strategies, e.g., secondary metal deposition, sulfidation and polymer coating, have been reported in the literature to overcome the shortcomings of bare nZVI (Kim et al., 2011; Liu et al., 2014; Zhao et al., 2011). Therefore, onethird of the sites were applied the surface-modified nZVIs for the real contaminated soil and groundwater remediation (Karn et al., 2009). However, conformational changes in nZVI can affect particle mobility and reactivity, subsequently altering the fate of nZVI in biological
systems and potentially posing risk. Despite the increasing use of surface-modified nZVIs in the field and the intense interest in nanomaterial safety, only a few studies have addressed the ecotoxicological impact of surface-modified nZVIs. Several studies demonstrated that certain surface coatings with carboxymethyl cellulose (CMC) or humic acid alleviated nZVI toxicity toward bacteria by limiting particle adhesion to cells (Li et al., 2010). In contrast, bimetallic nZVI in oxygenic water enhanced the production of reactive oxygen species (ROS) and thus its cytotoxicity (Kim et al., 2014a). The toxic action of nanomaterials varies considerably depending on the test organisms in addition to their physicochemical properties and concentrations (Nel et al., 2009). Thus far, most of the studies have been focused on the toxicity of nZVI on each species in the ecosystem. In addition, compared to the amount of literature on bacteria, available data on the adverse biological impact on a range of aquatic and terrestrial organisms is limited. It was found that the toxic effects on medaka fish were noticeable at the nZVI concentrations of 50–100 mg L−1 (Chen et al., 2012), whereas higher order organisms, i.e., plant and earthworm, required relatively higher nZVI
⁎
Corresponding author. E-mail address: [email protected] (Y.-S. Chang). 1 These two authors contributed equally to this work. https://doi.org/10.1016/j.ecoenv.2018.07.099 Received 26 November 2017; Received in revised form 12 July 2018; Accepted 24 July 2018 0147-6513/ © 2018 Elsevier Inc. All rights reserved.
Ecotoxicology and Environmental Safety 163 (2018) 436–443
H. Yoon et al.
bacterial toxicity was examined in terms of cell viability, intracellular ROS and cellular damage. Cell viability was assessed using Fluorescein Diacetate (FDA). The bacterial suspensions were treated with various concentrations of nZVIs (1–500 mg L−1) in distilled water and incubated for 24 h in a shaking incubator at 150 rpm. Control and nZVIstreated bacteria suspensions were washed, and then FDA was added to the suspensions to a final concentration of 10 µM. The samples were incubated for 30 min and the fluorescent signal was measured at excitation wavelength of 485 nm and emission wavelength of 530 nm using Fluorescence spectrometer (Fluoromax 4, Horiba). The cell viability was calculated by comparing between the control group and nZVIs-treated bacteria after 24 h of incubation. The equation for the cell viability (%) for Fig. 1:
concentrations for induction of observable toxicities. According to ElTemsah and Joner, the median effective concentrations estimated from the avoidance data of earthworm species were > 500 mg nZVI kg−1 (El-Temsah and Joner, 2012a). The phytotoxicity of nZVI was assessed in Lepidium sativum and Sinapis alba, with no significant inhibition on seed germination at approximately 300 mg L−1 (Libralato et al., 2016). Extrapolating effects of toxicants from a single test species to the ecosystem as a whole is an essential part of environmental risk assessment. However, the limited number of test species can’t take into account the interactions between the toxicants. nZVI can be transferred from injection sites to soil and groundwater streams via waterborne transportation (Karn et al., 2009; Mueller et al., 2012), their toxic effects should be considered at multiple levels of organization along a potential exposure route to provide a more realistic risk assessment. In the present study, the toxicities of surface-modified nZVIs in various aquatic and terrestrial organisms (bacteria, water flea, plant and earthworm) were systematically compared with bare nZVI through various ecotoxicity tests. The test species and endpoints were chosen under the worst assumption that nZVI has spread to aquatic and terrestrial environments at high concentrations. Three different types of surface-modified nZVIs [CMC-stabilized nZVI (CMC-nZVI), tetrapolyphosphate-coated nZVI (TPP-nZVI) and bismuth-iron bimetallic nanoparticle (Bi-nZVI)] were employed because of their effectiveness and attentions, as reported in previous studies (Bokare et al., 2010; Gong et al., 2015; Kim et al., 2015a). Additionally, we aimed to determine how the toxicity relates to the test species and the physicochemical properties of each surface-modified nZVIs.
[FDA]sample ⎞ Cell viability inhibition (%) = ⎛1− × 100 ⎝ [FDA]control ⎠ ⎜
⎟
The intracellular ROS in nZVIs-treated cells were measured using 2’, 7’-dichlorofluorescein diacetate (DCF-DA). The experimental method was same as the FDA method, but different incubation time (10 min) and dye concentration (2 µM) were applied. The fluorescent signal was measured at an excitation wavelength of 488 nm. To check the cellular damage by nZVIs, Fourier transform infrared spectroscopy (FTIR) analysis has been used. The bacterial suspensions were treated with 100 mg L−1 of nZVIs in distilled water. After 24 h of treatment, the suspensions were centrifuged to collect the pellets. The bacterial pellets were dried overnight at 60 °C in the vacuum oven and analyzed using FTIR (iS50, Thermo) in 800–3000 cm−1 range. For TEM imaging, the nZVI treated and control cells were fixed in formaldehyde and glutaraldehyde, dehydrated in ethanol, and embedded into EMBed 812 resin and propylene oxide. After sample was sectioned by an ultratome, analysis of bacterial sections was carried out using the TEM (H-7600, Hitachi).
2. Materials and methods 2.1. Preparation of nZVIs Four different nanomaterials were used in this study. Bare nZVI (Nanofer STAR) was commercially obtained from NANOIRON (Czech Republic). CMC-, TPP- and Bi-nZVIs were prepared using the same procedure as described in previous works (Bokare et al., 2010; Gong et al., 2015; Kim et al., 2015a). CMC- and Bi-nZVI were produced through the reduction of a Fe(II)–CMC mixture ([CMC] / [Fe(II)] = 0.2) and Bi-Fe mixture ([Bi(III)] / [Fe(II)] = 0.07) with NaBH4, respectively. TPP-nZVI was prepared by mixing an aliquot of nZVI with TPP in a pH-adjusted solution.
2.3.2. Plant The seeds of A. thaliana ecotype Columbia were sterilized using 70% ethanol with 20% Clorox and washed five times with distilled water. Seeds were germinated in hydroponics or soil for 30 days in a plant growth chamber (DS-330DHL, Daewon Sci., Korea) at 22–24 °C and 60% humidity with a 16:8 h (light: dark) photoperiod. In the hydroponic culture, the nZVI slurry was mixed with 1/2 Murashige and Skoog medium (MS; Duchefa Biochemie, Netherlands) at pH 5.8 (adjusted with 0.1 N KOH, Sigma-Aldrich, USA). In the soil culture, the slurry of nZVI was added to the test soil (purchased from Hungnong Co., Korea) to make the concentration 500 mg kg−1. The toxicity in A. thaliana was assessed based on the following endpoints: seed germination (%), root lengths (mm), relative biomass (%), H2O2 imaging, and iron accumulation. After 30 days of growth, weights of the harvested plants were quantified. The weight measurements were conducted on the shoots except roots. Average weights of three groups consisting of ten individual plants were measured. For H2O2 detection, segments of roots and leaves were incubated in 25 µM DCF-DA for 30 min in the dark and imaged using a confocal microscope (FV 1000, Olympus) in 485–530 nm wavelength. To determine the iron accumulation in tissues, freeze-dried plant samples were dissolved in 60% HNO3 at 120 °C overnight. After diluting the sample, the element contents were measured using an ICP-OES.
2.2. Particle characterization The particle size, morphology and elemental analysis of the products were analyzed using high-resolution transmission electron microscopy (HR-TEM, JEM-2010, Jeol Ltd) coupled with electron energy loss spectroscopy (EELS). Hydrodynamic diameter and ζ-potential measurements of nZVIs dispersed in deionized (DI) water or media were performed on a zetasizer (Nano ZS90, Malvern Instruments). Surface area analysis was conducted using the Brunauer-Emmett-Teller (BET) method with a micropore physisorption analyzer (ASAP-2020 M, Micrometrics). The dissolved iron concentration of each sample was measured with an inductively coupled plasma-optical emission spectrometer (ICP-OES, Optima 7000 DV, PerkinElmer). 2.3. Test organism and toxicity assessment
2.3.3. Water flea D. magna was cultured in M4 medium prepared according to the Organization for Economic Co-operation and Development (OECD) test guideline (pH; 7.8 ± 0.1, hardness; 250 ± 25 mg CaCO3 L−1) at 20 ± 1 °C with a 16:8 h (light:dark) photoperiod. The green algae, Pseudokirchneriella subcapitata and a mixture (yeast and CEROPHYLL®) were provided as food on a daily basis. The acute toxicity tests for D. magna were conducted in accordance with the OECD Test Guidelines
The model organisms used in the present study were two bacterial strains (Gram-positive Bacillus subtilis and Gram-negative Escherichia coli), a water flea (Daphnia magna), a plant (Arabidopsis thaliana) and an earthworm (Eisenia fetida). 2.3.1. Bacteria E. coli and B. subtilis (from the Korean Agricultural Culture Collection) were grown aerobically at 37 °C and 30 °C, respectively. The 437
Ecotoxicology and Environmental Safety 163 (2018) 436–443
H. Yoon et al.
media could promote aggregation of the particles (Tiraferri et al., 2008). The point of zero charge (PZC) of the nZVI samples varied from approximately pH 4–9.5. The iron contents of Nanofer, CMC-nZVI, TPPnZVI and Bi-nZVI were 79.3%, 24.3%, 13.7% and 75.3% (w/w), respectively. To investigate the iron dissolution from the nanoparticle in aqueous solutions, the production of total dissolved iron was monitored in water (Fig. S1 (B)) Nanofer, CMC-nZVI and Bi-nZVI showed insignificant amounts of dissolved iron for 48 h, which was consistent with the previous finding that the Fe2+ produced by nZVI corrosion remained insoluble as iron oxides (Kanel et al., 2005). However, TPP-nZVI produced approximately 3 mg L−1 (approximately 20% of TPP-nZVI) of dissolved iron after 48 h because of the iron-chelating character of TPP (Kim et al., 2015a).
202 (OECD, 1984). Exposure media was fully aerated M4 media without ethylenediaminetetraacetic acid (EDTA). Toxicity tests comprise of five concentrations of test solution and one control with four replicates. For each exposure, five randomly selected neonates (≤ 24 h old) were placed in a glass beaker containing 40 mL of test solutions. Immobilization of the test species was assessed after 48 h of exposure. Neonates not able to swim within 15 s after gentle agitation were considered to be immobilized. The median effective concentration (EC50) values for D. magna as well as their associated 95% confidence intervals (95% CI) were calculated accordingly. Temporal changes of pH, oxidation-reduction potential (ORP), dissolved oxygen (DO) and concentration of iron species in the modified nZVIs-containing M4 solutions were monitored following Chen et al. (2012). Daphnids (n = 6) were placed in nZVI suspensions concentrated at 1 or 5 mg L−1 to determine an individual oxidative stress response using confocal fluorescent microscopy in 485–530 nm wavelength. The samples were incubated for 4 h at 50 µM DCF-DA concentration. Relative fluorescence intensity was recorded and analyzed using ImageJ. An equivalent number of daphnids as negative controls were also included without nZVI.
3.2. Toxicity to bacteria The antibacterial effects of selected nZVIs, represented by the decrease of bacterial viability, were investigated using Fluorescein diacetate (FDA) fluorescence staining at different nZVI concentrations (1–500 mg L−1). As observed in Fig. 1, the nZVI treatment significantly decreased the viabilities of the two bacterial species in a dose-response manner. Of these, Bi-nZVI exhibited the highest toxicity in the range of all concentrations. This Bi-nZVI toxicity might be attributed to the presence of bismuth on the nZVI surface. Similar findings have shown that zerovalent bismuth NPs cause strong antimicrobial effects by the inhibition of biofilm formation (Hernandez-Delgadillo et al., 2012). On the other hand, the growth inhibition by CMC-nZVI rapidly increased depending on the nanoparticle concentrations. The main toxicological factor of CMC-nZVI toxicity would be the high dispersibility and inhibition of aggregation, thereby allowing favorable contact with bacterial cells (Chen et al., 2011). Previous studies have reported that oxidative stress was the major toxicological mechanism after exposure to nZVI (Sacca et al., 2014). 2′,7′–dichlorofluorescein diacetate (DCF-DA) staining was used to measure the intracellular ROS level as an indicator of the oxidative stress. Strong fluorescent signals were observed in E. coli treated with all nZVIs (Fig. 2), indicating that those NPs could significantly induce intracellular ROS. In case of B. subtilis, only Bi-nZVI significantly elevated the ROS level. The high ROS level detection could be explained by the nZVI promoting free radical generation via the Fenton reaction inside the cell (Ševců et al., 2011). However, the ROS level does not fully explain why nZVI tended to have negative effects on bacterial viability. Kim et al. reported that bimetallic Fe-Pd and Fe-Pt induced high ROS levels in E. coli, whereas their antibacterial activities were not significant (Kim et al., 2014a). It appears that bacterial cells possess sufficient antioxidant systems to defense oxidative stress from NPs. Therefore, an additional toxic mechanism might exist beside the oxidative stress. In the present study, B. subtilis was found to be more tolerant to the tested nZVIs than E. coli. These differences toward E. coli (Gram-negative) and B. subtilis (Gram-positive) are likely to be originated from the different cell wall structures and compositions between the two bacterial species. Furthermore, the lower ROS level in B. subtilis could be due to the presence of nitric oxides that boost the catalase activity involved in H2O2 scavenging. To date, the bacterial nitric oxide synthases have been known to exist in Gram-positive bacteria only (Gusarov and Nudler, 2005). The current hypothesized toxic mechanisms also include structural change and disruption of cell membranes (Jiang et al., 2010). To provide insight into the mechanism of interaction between NPs and cell membranes, the bacterial cells were analyzed with Fourier transform infrared spectroscopy-diffuse reflectance (FTIR-DRIFT) to characterize the change in biomolecule composition. IR spectra of the bacterial cells exposed at 100 mg L−1 nZVI are shown in Fig. S2. The fatty acid, protein and peptide and carbohydrate regions were characterized by the peaks in the wavelengths of 100–2800 cm−1, 1800–1500 cm−1 and
2.3.4. Earthworm E. fetida were purchased from Carolina Biological Supply (Burlington, USA). The earthworms were acclimated in plastic boxes containing moist bedding and food (Magic Worm Food, USA). The acute toxicity tests for E. fetida were conducted in accordance with the OECD Test Guidelines 207, (OECD, 2004). Standardized OECD soil substrate was used as exposure medium which consists of industrial fine sand, kaolin clay, and sphagnum peat in a 7:2:1 ratio, respectively and were mixed with nZVI. Four replicates were carried out and each group consisted of ten worms. For each exposure, adult earthworms (at least 2 months old with individual weights of 350–450 mg) were selected. 500 g (dry weight) of test medium was placed into each crystallizing dishes and earthworms were placed on the test medium surface. Treated earthworms were maintained at a controlled temperature of 20 ± 2 °C with 80–85% humidity under 400–800 lx of constant light. Mortality, abnormal behavior, and weight change of earthworms were assessed after 14 days of exposure. The wet mass of each exposure group (ten earthworms) was weighed and expressed as average weight of earthworms. To quantify the body burden of total iron in earthworms, whole bodies of each exposure group were placed in a digestion vessel with 8 mL of 60% HNO3 and 2 mL of H2O2. After complete digestion (heating at 120 °C for 4 h), the solution was analyzed by ICPOES. 2.4. Statistical analysis Each treatment was conducted in at least triplicate. Data were presented as the mean ± standard deviation (SD) and analyzed by oneway analysis of variance (ANOVA) with the Tukey HSD test using IBM SPSS statistics 20. Probability values (P) < 0.05 were considered statistically significant. The 48 h EC50 for the D. magna acute toxicity test were estimated by the Spearman-Karber method using CETIS version 1.8.7.15 (Tidepool Scientific Software, USA). 3. Results and discussion 3.1. Characteristics of nZVI materials The physicochemical properties of each surface-modified nZVIs are summarized in Table 1. Although the primary sizes of four nZVIs were determined to be approximately 40–116 nm by TEM (Fig. S1 (A)), the particles in DI water rapidly aggregated with an average hydrodynamic diameter range of approximately 300–770 nm. All nanoparticles suspended in M4 media had significantly larger hydrodynamic diameters compared to those in water since the high salt concentrations in M4 438
Ecotoxicology and Environmental Safety 163 (2018) 436–443
H. Yoon et al.
Table 1 Physicochemical characterization of selected nanoscale zerovalent iron (nZVI). Results are shown as the means ± SD (n = 3). Types of nZVI
Coating
Surface Area (BET) (m2/g)
Primary size (TEM) (nm)a
Hydrodynamic diameter (nm)b
Zeta potential (mV)b
Point of zero charge (PZC)c
Iron contents (%)
Nanofer
–
14.3
59.8 ± 1.3
79.3 ± 0.5
Carboxy methyl cellulose Tetrapolyphosphate Bismuth
19.5
39.8 ± 1.7
<4
24.3 ± 8.9
11.3
72.2 ± 2.7
<4
13.7 ± 0.2
11.0
116.7 ± 5.4
− 5.8 ± 0.6 (DW) − 7.4 ± 0.4 (M4) − 32.1 ± 1.8 (DW) − 9.9 ± 0.1 (M4) − 62.6 ± 3.4 (DW) − 21.8 ± 0.4 (M4) 31.4 ± 0.9 (DW) − 2.8 ± 0.4 (M4)
5.5
CMC-nZVI
775.4 ± 20.2 (DW) 4365.0 ± 245.7 (M4) 443.8 ± 34.4 (DW) 3383.7 ± 619.2 (M4) 701.2 ± 29.1 (DW) 2466 ± 105.1 (M4) 313.5 ± 20.7 (DW) 4155.3 ± 513.2 (M4)
9.5
75.3 ± 15.4
TPP-nZVI Bi-nZVI
a b c
The values of each NPs size represent the mean of the 10 replicates. Hydrodynamic diameter and zeta potential were measured with 10 mg L−1 nZVIs in deionized water or M4 media. Point of zero charge was measured with 10 mg L−1 nZVIs in 10 mM NaCl.
Fig. 2. ROS levels in two species of bacterial cells after 24 h exposure of 100 mg L−1 nZVI. Results are shown as the means ± SD (n = 3). Different letters represent significant differences among the different groups (one-way ANOVA with Tukey HSD post-hoc test).
penetration into the cell membrane, thereby inducing high toxicity. The rough surface of Bi-nZVI promoted adhesion (Fig. S1 (A)), leading to easy translation into the cell (Nel et al., 2009). To compare actual images of bacterial interactions with nZVIs, the exposed bacteria were visualized by TEM (Fig. S3). Bare-nZVI, CMC-nZVI and Bi-nZVI attached to the cells caused serious damages to the membrane integrity, eventually leading to death. In the protein region, there were changes caused by binding of NPs or ions released from NPs (Eckhardt et al., 2013; Faghihzadeh et al., 2016). For TPP-nZVI, only the peaks in the protein region were shifted. Moreover, in the TEM images, only TPPnZVI was aggregated out of the cell membrane without any penetration. Previous studies reported a positive correlation between the dissolved Fe ions and the nZVI toxicity to bacteria (Auffan et al., 2008). TPP-nZVI showed a higher dissolution than other nZVIs in DI water (Fig. S1 (B)). The Fe ions released from TPP-nZVI might interact with amino acids, thereby inducing the alteration of proteins or enzymes. Finally, in the carbohydrate region, ROS could cause the alternation of C-O-C and C-O vibrations in the sugar rings of polysaccharides depicted in Fig. 2 (Eckhardt et al., 2013; Faghihzadeh et al., 2016). In case of B. subtilis, however, only Bi-nZVI caused an obvious change to the spectrum, which agreed well with the results of the cell viability and ROS level.
Fig. 1. Viability of (A) Escherichia coli and (B) Bacillus subtilis after exposure to tested nZVIs in deionized (DI) water for 24 h. Results are shown as the means ± SD (n = 3).
1200–900 cm−1, respectively. Overall, the tendency of the spectra alternation was relative to the order of toxicity. For example, a remarkable difference was found in the bacteria treated by Bi-nZVI, which had a high inhibition effect on growth rate. In detail, the bands of E. coli cells exposed to bare nZVI, CMC-nZVI and Bi-nZVI showed almost complete disappearance of spectra in the fatty acid region, possibly due to the strong binding of NPs with –CH groups of the membrane, which changed the fluidity of cell wall lipids (Faghihzadeh et al., 2016). In addition, there is high probability of interaction between positively charged Bi-nZVI and negatively charged bacteria membrane via electrostatic attraction. The shape and small hydrodynamic size of nZVI also contributed to
3.3. Toxicity to plant To investigate the influence of the tested nZVIs on plant growth, the physiological attributes such as seed germination, root length and biomass of a model plant, A. thaliana, were measured. As illustrated in Fig. S4 (A), the tested nZVIs did not have deleterious influences on the 439
Ecotoxicology and Environmental Safety 163 (2018) 436–443
H. Yoon et al.
treated plant. It was expected that the greater amount of H2O2 accumulated by Bi-nZVI should adversely affect plant growth via oxidative stress. The chlorosis is caused by iron deficiency in plants. Therefore, iron accumulation in plant tissues was analyzed to identify the bioavailability of nZVI. The iron contents in shoots increased significantly by all tested nZVIs (Fig. S4 (C)). In particular, the Bi-nZVI treated plants absorbed an excessive amount of iron compared to the control, but yellowing of the leaves and chlorosis were also observed. Similar symptoms were studied by Wang et al. (2016), where the higher concentration of nZVI induced iron deficiency in rice because the transport of active iron from the root to the shoot was blocked (Wang et al., 2016). As a result, chlorosis cannot be explained solely by the total concentration of iron, because excessive iron enters the plant but cannot be confirmed in what form it is stored inside. Furthermore, the results indicated that the phenomena observed in the present study did not originate from the dissolved ions but from the NPs themselves. Zhao et al. (2012) found that the additives released from the surface of the NPs played an important role in ion bioavailability to plants. Therefore, further study of the uptake, translocation and transformation of the tested nZVIs into plants is highly needed.
germination percentages of plant. However, the various nZVI types exhibited different effects on the growth of plants. In a previous study, El-Temsah et al., examined the effects of different types and concentrations of nZVI on the germination of various pasture plants such as flax, barley and ryegrass. They found that although there was no negative effect on the germination percentage up to 250 mg L−1, no seeds were able to germinate at 1000–2000 mg L−1 of poly(acrylic acid) coated nZVI (El-Temsah and Joner, 2012b). In addition, the germination percentages of rape and cabbage were also found to significantly decrease by 10% and 40%, respectively at 1500 mg kg−1 of CMC-nZVI (Wang et al., 2014). Those findings suggest that the phytotoxic responses resulting from nZVI are more likely to depend on not only the type and concentration of nZVI but also the plant species (Du et al., 2016). The present study also obtained a similar result of root elongation and biomass depend on the type of nZVI. The root length of plants treated with CMC-nZVI and Bi-nZVI decreased by approximately 40% and 69% compared to the control plant (non-treated), respectively (Fig. S4 (B)). For biomass, CMC and Bi-nZVI similarly inhibited the growth of plant by at least 40% and Bi-nZVI, in particular, might cause severe damages to the plant in a hydroponic system (Fig. 3). Bismuth NP was reported to interact with the sulfhydryl groups (SH-) present in plant membrane proteins, thereby causing a DNA damage (Liman, 2013). In addition to Bi-nZVI, CMC-nZVI also showed an inhibitory effect on the plant root growth (Wang et al., 2014). On the other hand, bare nZVI was observed to be capable of increasing root length and significantly promoting plant growth. In the case of bare nZVI, the results agreed fairly well with our previous studies demonstrating that bare nZVI triggered root elongation and showed an increase of biomass (i.e., leaf area) by activating stomatal opening (Kim et al., 2014b, 2015b). H2O2 has essential roles in plant metabolism such as root development and photosynthesis (Slesak et al., 2007). However, the concurrent abiotic stress related to the overproduction of H2O2 is a concern, as to be potentially damaging to plant growth (Cheeseman, 2007). To utilize H2O2 as a signaling molecule, the non-toxic levels must be maintained in a delicate balancing act between H2O2 production and scavenging (Hossain et al., 2015). To elucidate the toxicity mechanism caused by the tested nZVIs on plants, imaging of H2O2 was performed using DCFDA on the roots and leaves. Fig. 4 illustrates that all plants, except the TPP-nZVI treated plant, showed higher fluorescent intensities in the roots compared to the control. The trend of the results shown in the roots was approximately similar to the bacterial results (Fig. 2). However, the results obtained from the leaves were of interest and a significantly higher level of fluorescence was observed in the Bi-nZVI
3.4. Toxicity to water flea The concentration immobilizing 50% of neonates after 48 h of nZVI exposure was expressed as the effective concentration (EC50). Bi-nZVI (EC50 = 42.5 mg L−1) and CMC-nZVI (EC50 = 45.6 mg L−1) appeared to be the most toxic to water flea, whereas no immobilization occurred when exposed to bare and TPP-nZVI (EC50 > 100 mg L−1). In oxygenated water, nZVI reacts with O2 to generate Fe2+ and oxidizing intermediates (i.e., H2O2, •OH) (Harada et al., 2016), which may enter the cells and induce the oxidative stress. In the existing literature, oxidative stress has been regarded as an important factor in toxicity to daphnids (Dalai et al., 2013). To confirm that oxidative stress induced by nZVI in Daphina magna, the formation of ROS was measured by the DCF-DA method that was previously used in the bacterial and plant ROS determination. As shown in Fig. 5, similar to that of the bacteria and plant results, all daphnids, except for the TPP-nZVI treated group, had significant fluorescent intensities compared to that of the control even at a lower concentration than the no observed effect concentration (NOEC). Similar to the bacterial results, these results do not fully explain why nZVI tend to produce negative effects on daphnid immobilization. The temporal fluctuations in hydrochemical variables of the test medium (i.e., pH, oxidation-reduction potential (ORP) and dissolved oxygen (DO)) indicated that the exposure to nZVI would affect the NP bioavailability and toxicity in an aquatic environment. As shown in Fig. S5, the pH values of all nZVI solutions slightly increased for 4 h and they were stabilized at pH 8.0–8.2. The addition of Bi- and CMC-nZVI led to decreases of ORP and Bi-nZVI gave more negative ORP than CMC-nZVI for 48 h. Conversely, the initial ORP values did not change significantly with bare and TPP-nZVI. The pattern of DO changes in four nZVI suspensions was similar to what we observed in the ORP measurement. The DO depletion at the initial phase was most pronounced in the Bi-nZVI suspension, reaching almost zero. The observed higher toxicities of Bi- and CMC-nZVI were apparently related to the hypoxia. Our data were in agreement with the previous studies on lethal and critical O2 concentrations, down to 3 mg L−1, for aerobic metabolism in D. magna. The DO depletion was the stressor to daphnids because anthropogenic impacts of Bi- and CMC-nZVI induced the hypoxia in aquatic systems, resulting in the change of physiological processes (Kobayashi and Hoshi, 1984). Although hypoxia appeared to the main causative factor, the contribution of other factors cannot be overlooked. Several studies have demonstrated that the dissolved form of NPs is a main cause of toxicity to aquatic species (Griffitt et al., 2008). To determine the relative
Fig. 3. Relative growth rate of nZVI-treated Arabidopsis thaliana seedlings after 30 days in a hydroponic and soil system. Results are shown as the means ± SD (n = 3). Different letters represent significant differences among the different groups (one-way ANOVA with Tukey HSD post-hoc test). 440
Ecotoxicology and Environmental Safety 163 (2018) 436–443
H. Yoon et al.
Fig. 4. H2O2 imaging in Arabidopsis thaliana root and leaf treated with 100 mg L−1 nZVI.
significant difference in the body weight of E. fetida was observed between the nZVI-treated soils and the control (Fig. 6), indicating that all the nZVIs did not exhibit acute toxicity to earthworm. In addition, the iron concentration in tissue was measured after 14 d exposure to nZVIs. As depicted in Fig. 6(B), the tissue iron concentration varied slightly depending on the nZVI type, ranging from 80 to 150 mg kg−1. The relatively low uptake of total iron even at high exposure concentrations might be attributed to the sorption of Fe ions to soil or the aggregation of NPs to themselves. As previously noted in literature, the higher concentrations of iron NPs did in several cases influence the measurement principles behind the earthworm toxicity tests because of their aggregation, agglomeration and ultimately sedimentation of nZVI (Hjorth et al., 2017). However, despite the limitations and challenges in the terrestrial ecotoxicity test, these attempts can still provide useful information for developing a new catalyst for environmental remediation. Fig. 5. Level of ROS in Daphnia magna after 24 h exposure of 1 and 5 mg L−1 nZVI. Results are shown as the means ± SD (n = 3). Different letters represent significant differences among the different groups (one-way ANOVA with Tukey HSD post-hoc test).
4. Conclusions In the present work, we have tested the potential toxicity of surfacemodified nZVIs on the various terrestrial and aquatic species (bacteria, plant, water flea and earthworm). Unlike previous studies, we focused on the toxic effects of surface-modified nZVIs on multiple species in the ecosystem. Among the tested nZVIs, Bi- and CMC-nZVI caused strong toxic effects on all of the tested organisms, except for earthworm. It is worth noting that the oxidative stress via the Fenton reaction appeared to mainly contribute to the observed toxicity endpoints. In addition, membrane disruption, chlorosis and hypoxia could be additional toxic factors, depending on the test species. The commonalities in the toxicity of tested nZVIs on the diverse trophic groups suggested that the interspecies comparison would be critical for the risk assessment of engineered nanomaterials. In addition, the results emphasized the importance of surface modifiers in the toxic effects of nZVI, which deserved further evaluations given that a variety of surface-modified nZVIs are currently available at the on-site remediation. Finally, this
contribution of Fe ions to the observed toxic effects, the release profiles of dissolved iron from nZVI at 100 mg L−1 were investigated in M4 media. It should be noted that the dissolution behavior of nZVI can vary depending on the particle composition and exposure conditions (Adeleye et al., 2013). In the present study, the dissolved iron concentrations for each nZVI ranged from 0.2 to 3 mg L−1 (data not shown), which was far below the 48 h EC50 (59.7 mg L−1) value of the Fe2+ solution (Baumann et al., 2014). Thus, the overall toxicity of nZVI to Daphnia magna was not related to the direct effects of free iron, but the combination of hypoxia and oxidative stress. 3.5. Toxicity to earthworm Despite the high concentration of nZVI (1000 mg kg−1) applied, no 441
Ecotoxicology and Environmental Safety 163 (2018) 436–443
H. Yoon et al.
101, 6354–6360. Cheeseman, J.M., 2007. Hydrogen peroxide and plant stress: a challenging relationship. Plant Stress 1, 4–15. Chen, J., Xiu, Z., Lowry, G.V., Alvarez, P.J., 2011. Effect of natural organic matter on toxicity and reactivity of nano-scale zero-valent iron. Water Res. 45, 1995–2001. Chen, P.-J., Tan, S.-W., Wu, W.-L., 2012. Stabilization or oxidation of nanoscale zerovalent iron at environmentally relevant exposure changes bioavailability and toxicity in medaka fish. Environ. Sci. Technol. 46, 8431–8439. Dalai, S., Pakrashi, S., Chandrasekaran, N., Mukherjee, A., 2013. Acute toxicity of TiO2 nanoparticles to Ceriodaphnia dubia under visible light and dark conditions in a freshwater system. PLoS One 8, e62970. Du, W., Tan, W., Peralta-Videa, J.R., Gardea-Torresdey, J.L., Ji, R., Yin, Y., Guo, H., 2016. Interaction of metal oxide nanoparticles with higher terrestrial plants: physiological and biochemical aspects. Plant Physiol. Biochem. 110, 210–225. Eckhardt, S., Brunetto, P.S., Gagnon, J., Priebe, M., Giese, B., Fromm, K.M., 2013. Nanobio silver: its interactions with peptides and bacteria, and its uses in medicine. Chem. Rev. 113, 4708–4754. El-Temsah, Y.S., Joner, E.J., 2012a. Ecotoxicological effects on earthworms of fresh and aged nano-sized zero-valent iron (nZVI) in soil. Chemosphere 89, 76–82. El-Temsah, Y.S., Joner, E.J., 2012b. Impact of Fe and Ag nanoparticles on seed germination and differences in bioavailability during exposure in aqueous suspension and soil. Environ. Toxicol. 27, 42–49. Faghihzadeh, F., Anaya, N.M., Schifman, L.A., Oyanedel-Craver, V., 2016. Fourier transform infrared spectroscopy to assess molecular-level changes in microorganisms exposed to nanoparticles. Nanotechnol. Environ. Eng. 1, 1–16. Gong, J., Lee, C.-S., Chang, Y.-Y., Chang, Y.-S., 2015. Novel self-assembled bimetallic structure of Bi/Fe 0: the oxidative and reductive degradation of hexahydro-1, 3, 5trinitro-1, 3, 5-triazine (RDX). J. Hazard. Mater. 286, 107–117. Griffitt, R.J., Luo, J., Gao, J., Bonzongo, J.C., Barber, D.S., 2008. Effects of particle composition and species on toxicity of metallic nanomaterials in aquatic organisms. Environ. Toxicol. Chem. 27, 1972–1978. Gusarov, I., Nudler, E., 2005. NO-mediated cytoprotection: instant adaptation to oxidative stress in bacteria. Proc. Natl. Acad. Sci. USA 102, 13855–13860. Harada, T., Yatagai, T., Kawase, Y., 2016. Hydroxyl radical generation linked with iron dissolution and dissolved oxygen consumption in zero-valent iron wastewater treatment process. Chem. Eng. J. 303, 611–620. Hernandez-Delgadillo, R., Velasco-Arias, D., Diaz, D., Arevalo-Niño, K., Garza-Enriquez, M., De la Garza-Ramos, M.A., Cabral-Romero, C., 2012. Zerovalent bismuth nanoparticles inhibit Streptococcus mutans growth and formation of biofilm. Int. J. Nanomed. 7, e2113. Hjorth, R., Coutris, C., Nguyen, N.H., Sevcu, A., Gallego-Urrea, J.A., Baun, A., Joner, E.J., 2017. Ecotoxicity testing and environmental risk assessment of iron nanomaterials for sub-surface remediation–Recommendations from the FP7 project NanoRem. Chemosphere 182, 525–531. Hossain, M.A., Bhattacharjee, S., Armin, S.-M., Qian, P., Xin, W., Li, H.-Y., Burritt, D.J., Fujita, M., Tran, L.-S.P., 2015. Hydrogen peroxide priming modulates abiotic oxidative stress tolerance: insights from ROS detoxification and scavenging. Front. Plant Sci. 6, 420. Jiang, W., Yang, K., Vachet, R.W., Xing, B., 2010. Interaction between oxide nanoparticles and biomolecules of the bacterial cell envelope as examined by infrared spectroscopy. Langmuir 26, 18071–18077. Kanel, S.R., Manning, B., Charlet, L., Choi, H., 2005. Removal of arsenic (III) from groundwater by nanoscale zero-valent iron. Environ. Sci. Technol. 39, 1291–1298. Karn, B., Kuiken, T., Otto, M., 2009. Nanotechnology and in situ remediation: a review of the benefits and potential risks. Environ. Health Persp. 117, 1813–1831. Kim, E.-J., Kim, J.-H., Azad, A.-M., Chang, Y.-S., 2011. Facile synthesis and characterization of Fe/FeS nanoparticles for environmental applications. ACS Appl. Mater. Interfaces 3, 1457–1462. Kim, E.-J., Le Thanh, T., Chang, Y.-S., 2014a. Comparative toxicity of bimetallic Fe nanoparticles toward Escherichia coli: mechanism and environmental implications. Environ. Sci.: Nano. 1, 233. Kim, H.-H., Lee, H., Kim, H.-E., Seo, J., Hong, S.W., Lee, J.-Y., Lee, C., 2015a. Polyphosphate-enhanced production of reactive oxidants by nanoparticulate zerovalent iron and ferrous ion in the presence of oxygen: yield and nature of oxidants. Water Res. 86, 66–73. Kim, J.H., Lee, Y., Kim, E.J., Gu, S., Sohn, E.J., Seo, Y.S., An, H.J., Chang, Y.S., 2014b. Exposure of iron nanoparticles to Arabidopsis thaliana enhances root elongation by triggering cell wall loosening. Environ. Sci. Technol. 48, 3477–3485. Kim, J.H., Oh, Y., Yoon, H., Hwang, I., Chang, Y.S., 2015b. Iron nanoparticle-induced activation of plasma membrane H(+)-ATPase promotes stomatal opening in Arabidopsis thaliana. Environ. Sci. Technol. 49, 1113–1119. Kobayashi, M., Hoshi, T., 1984. Analysis of respiratory role of haemoglobin in Daphnia magna. Zool. Sci. 1, 523–532. Li, Z., Greden, K., Alvarez, P.J., Gregory, K.B., Lowry, G.V., 2010. Adsorbed polymer and NOM limits adhesion and toxicity of nano scale zerovalent iron to E. coli. Environ. Sci. Technol. 44, 3462–3467. Libralato, G., Devoti, A.C., Zanella, M., Sabbioni, E., Mičetić, I., Manodori, L., Pigozzo, A., Manenti, S., Groppi, F., Ghirardini, A.V., 2016. Phytotoxicity of ionic, micro-and nano-sized iron in three plant species. Ecotoxicol. Environ. Saf. 123, 81–88. Liman, R., 2013. Genotoxic effects of Bismuth (III) oxide nanoparticles by Allium and Comet assay. Chemosphere 93, 269–273. Liu, W.-J., Qian, T.-T., Jiang, H., 2014. Bimetallic Fe nanoparticles: recent advances in synthesis and application in catalytic elimination of environmental pollutants. Chem. Eng. J. 236, 448–463. Mueller, N.C., Braun, J., Bruns, J., Černík, M., Rissing, P., Rickerby, D., Nowack, B., 2012. Application of nanoscale zero valent iron (NZVI) for groundwater remediation in
Fig. 6. (A) Change in weight and (B) Iron accumulation in Eisenia fetida exposed to 1000 mg L−1 nZVI. Results are shown as the means ± SD (n = 3). Different letters represent significant differences among the different groups (one-way ANOVA with Tukey HSD post-hoc test).
research has explored the nZVI toxicity to individual organism levels, but the toxicity to community levels or synergistic impacts needs to be further investigated for eco-friendly remediation using nanomaterials. Acknowledgements This work was supported by ‘‘The GAIA Project” by the Korea Ministry of Environment (RE201402059) and the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No. NRF-2017R1A2B3012681). Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.ecoenv.2018.07.099. References Adeleye, A.S., Keller, A.A., Miller, R.J., Lenihan, H.S., 2013. Persistence of commercial nanoscaled zero-valent iron (nZVI) and by-products. J. Nanopart. Res. 15, 1–18. Auffan, M. l., Achouak, W., Rose, J. r., Roncato, M.-A., Chanéac, C., Waite, D.T., Masion, A., Woicik, J.C., Wiesner, M.R., Bottero, J.-Y., 2008. Relation between the redox state of iron-based nanoparticles and their cytotoxicity toward Escherichia coli. Environ. Sci. Technol. 42, 6730–6735. Baumann, J., Köser, J., Arndt, D., Filser, J., 2014. The coating makes the difference: acute effects of iron oxide nanoparticles on Daphnia magna. Sci. Total Environ. 484, 176–184. Bokare, V., Murugesan, K., Kim, Y.-M., Jeon, J.-R., Kim, E.-J., Chang, Y.S., 2010. Degradation of triclosan by an integrated nano-bio redox process. Bioresour. Technol.
442
Ecotoxicology and Environmental Safety 163 (2018) 436–443
H. Yoon et al.
Tiraferri, A., Chen, K.L., Sethi, R., Elimelech, M., 2008. Reduced aggregation and sedimentation of zero-valent iron nanoparticles in the presence of guar gum. J. Colloid Interface Sci. 324, 71–79. Wang, J., Fang, Z.Q., Cheng, W., Yan, X.M., Tsang, P.E., Zhao, D.Y., 2016. Higher concentrations of nanoscale zero-valent iron (nZVI) in soil induced rice chlorosis due to inhibited active iron transportation. Environ. Pollut. 210, 338–345. Wang, Y., Fang, Z., Kang, Y., Tsang, E.P., 2014. Immobilization and phytotoxicity of chromium in contaminated soil remediated by CMC-stabilized nZVI. J. Hazard. Mater. 275, 230–237. Zhao, L., Peralta-Videa, J.R., Varela-Ramirez, A., Castillo-Michel, H., Li, C., Zhang, J., Aguilera, R.J., Keller, A.A., Gardea-Torresdey, J.L., 2012. Effect of surface coating and organic matter on the uptake of CeO 2 NPs by corn plants grown in soil: insight into the uptake mechanism. J. Hazard. Mater. 225, 131–138. Zhao, X., Lv, L., Pan, B., Zhang, W., Zhang, S., Zhang, Q., 2011. Polymer-supported nanocomposites for environmental application: a review. Chem. Eng. J. 236, 448–463 (170, 381-394).
Europe. Environ. Sci. Pollut. Res. Int. 19, 550–558. Nel, A.E., Mädler, L., Velegol, D., Xia, T., Hoek, E.M., Somasundaran, P., Klaessig, F., Castranova, V., Thompson, M., 2009. Understanding biophysicochemical interactions at the nano–bio interface. Nat. Mater. 8, 543–557. OECD, 1984. OECD Guidelines for the Testing of Chemicals, Section 2: Effects on Biotic Systems 207: Earthworm, Acute Toxicity Tests. OECD Publishing, pp. 1–9. OECD, 2004. OECD Guidelines for the Testing of Chemicals Test No. 202: Daphnia sp. Acute Immobilisation Test. OECD Publishing. Sacca, M.L., Fajardo, C., Martinez-Gomariz, M., Costa, G., Nande, M., Martin, M., 2014. Molecular stress responses to nano-sized zero-valent iron (nZVI) particles in the soil bacterium Pseudomonas stutzeri. PLoS One 9, e89677. Ševců, A., El-Temsah, Y.S., Joner, E.J., Černík, M., 2011. Oxidative stress induced in microorganisms by zero-valent iron nanoparticles. Microbes Environ. 26, 271–281. Slesak, I., Libik, M., Karpinska, B., Karpinski, S., Miszalski, Z., 2007. The role of hydrogen peroxide in regulation of plant metabolism and cellular signalling in response to environmental stresses. Acta Biochim. Pol. 54, 39.
443