Environmental Toxicology and Chemistry, Vol. 33, No. 3, pp. 632–640, 2014 # 2013 SETAC Printed in the USA
SALINITY INFLUENCES ON THE UPTAKE OF SILVER NANOPARTICLES AND SILVER NITRATE BY MARINE MEDAKA (ORYZIAS MELASTIGMA) JIAN WANG and WEN-XIONG WANG* Division of Life Science, State Key Laboratory of Marine Pollution, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong Special Administrative Region, China (Submitted 21 June 2013; Returned for Revision 3 September 2013; Accepted 14 November 2013) Abstract: With increasing use of silver nanoparticles (AgNPs), concerns about their potential deleterious effects on aquatic ecosystems have increased. Most previous studies have focused on the toxicity of AgNPs while their bioavailability has been seldom investigated. The present study examined the effects of salinity on the aggregation kinetics as well as the bioavailability of commercial 80-nm citrate-coated AgNPs (c-AgNPs) in the presence or absence of a nonionic surfactant (Tween 20) to marine medaka (Oryzias melastigma). In addition, the uptake of soluble Ag was quantified for comparison and for deducting the uptake of soluble Ag during AgNP exposure by applying a biokinetic model. The authors found that the addition of Tween 20 immediately slowed down the process of aggregation of AgNPs, and an elevated amount of Tween 20 (20 mM) kept AgNPs well dispersed, even in the 30-psu salinity medium. Uptake rate constants (ku) of AgNPs were less than half the soluble Ag at low salinities (1 psu and 5 psu), while limited bioavailability of c-AgNPs was observed at high salinities (15 psu and 30 psu). However, the Tween 20–stabilized AgNPs (t-AgNPs) were accumulated by medaka at comparable rates as the soluble Ag, indicating the importance of dispersion for bioavailability of AgNPs in a highly ionic environment. The present study provided the first insight of the bioavailability of AgNPs to fish in a high-ionic environment. More studies are needed to gain a full understanding of bioavailability of AgNPs in marine environments. Environ Toxicol Chem 2014;33:632–640. # 2013 SETAC Keywords: Silver nanoparticles
Uptake kinetic
Bioavailability
Aggregation
water, sediment, and particles) with continuous dissolution throughout the whole process [11]. Therefore, both dissolved Ag and the AgNPs themselves need to be considered in evaluating the potential impacts of AgNPs. Soluble Ag has been reported to be a toxic metal that inhibits several osmoregulation-related enzymes of fish [3]. Previous study has shown that AgNPs at a size range of 5 nm to 46 nm can be transported into the zebrafish embryo and distributed into the heart, yolk, and blood of the embryo, causing toxic effects such as delayed hatching, mortality, and malformation [12]. Waterborne exposure to AgNPs also caused an elevated silver burden in the gills of zebrafish [13], and enzymes of juvenile rainbow trouts such as Naþ, Kþ-adenosine triphosphate-ase (ATPase) were suppressed after 3 h of exposure to 10 mg/L AgNPs [14]. Biomarkers such as metallothionein, glutathione-S-transferase, or the p53 gene responded after chronic exposure to AgNPs in medaka [15]. In these earlier studies, the observable toxic concentrations of AgNPs were magnitudes higher than the estimated environmental concentrations [16]. The bioavailability of AgNPs at sublethal concentrations has seldom been studied. In addition, these studies mainly focused on freshwater fish, while the impact of AgNPs on marine fish remains unknown. Although AgNPs tend to form aggregates in highionic environments [17], it is now possible to keep AgNPs suspended in high-ionic medium with the application of nonionic surfactant Tween 80 [18]. Therefore, the present study investigated the stabilization of the nonionic surfactant Tween 20 on the aggregation kinetics and uptake of the c-AgNPs at different salinities. A euryhaline species of marine medaka, Oryzias melastigma, was used as the model organism in the present study. A newly developed radiolabeling method was applied to study the bioavailability of AgNPs at sublethal concentrations. In addition, because dissolution of AgNPs is unavoidable in the saline medium,
INTRODUCTION
Silver is used extensively in various sectors such as jewelry, coins, and conductive elements, with an annual global consumption of 28 million kg. [1]. The recent development of nanotechnology extends the applications of silver (e.g., cosmetics, optics, and textiles) in the form of silver nanoparticles (AgNPs). Extensive use of AgNPs in these products will result in increased inputs of AgNPs in the environment. Among the potential discharge routes, release of commercial nanoparticles into the aquatic system has attracted the maximum concern [2]. As a toxic rare metal, soluble silver and its impact on aquatic systems have long been studied [3], but our understanding of the bioavailability of AgNPs is still limited. Compared with Agþ, the behavior of AgNPs is much more complex. Once AgNPs are introduced into the aquatic systems, various transformations of AgNPs take place [4]; processes such as aggregation and dissolution have been addressed in many earlier studies [5]. Particles of different sizes and coatings also behave differently. Silver nanoparticles of smaller sizes tend to aggregate faster [6], and coatings such as sodium dodecyl sulfate and Tween 80 can effectively elevate the critical coagulation concentrations of AgNPs [7]. Based on these aggregation processes, several models have been developed to assess the particle–particle and particle–surface interaction under laboratory conditions [8]. However, in aquatic systems, the presence of natural organic matter (NOM) as well as various environmental factors (e.g., temperature, pH, and dissolved oxygen) can complicate the interaction [9] and affect the dissolution process [10]. The AgNPs distribute into different compartments (e.g., * Address correspondence to [email protected]. Published online 28 November 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/etc.2471 632
Salinity influences on silver nanoparticle uptake by marine medaka
the uptake kinetics of AgNO3 were also studied for comparison, and the uptake rate constant was calculated to deduct the soluble Ag uptake by applying a biokinetic model. MATERIALS AND METHODS
Chemicals and organisms
Silver nitrate (ACS reagent, >99%) and sodium citrate tribasic dihydrate (Purum p.a., >99%) were purchased from Sigma Aldrich. Commercially available 80-nm c-AgNPs were purchased from Ted Pella. Radioactive 110mAgNO3 (POLATOM) were purchased in the form of an aqueous solution with 0.1 M HNO3, and an equal volume of 0.1 M NaOH was added for neutralization before use. Instant Sea Salts were purchased from Instant Ocean, and the test media were prepared by diluting the filtered (pore size 0.22 mm, Milipore) concentrated stock to a specific salinity (1 psu, 5 psu, 15 psu, and 30 psu). Marine medaka (Oryzias melastigma) were obtained from City University of Hong Kong and reared in sand-filtered natural flowing seawater from Clear Water Bay (Kowloon, Hong Kong). The fish were kept in a natural light cycle at room temperature (22–25 8C). Dry food powder (TROUW) was fed to the fish twice a day, and the tank was cleaned regularly to remove the algae and feces on the bottom. After 3 mo of growth, the adult fish were randomly transferred into 20-L glass aquaria with 15 L of 1-psu, 5-psu, 15-psu, or 30-psu salinity medium for acclimation at a density of 2 individuals/L. The acclimation period lasted for 3 wk before the initiation of uptake experiments. Within the acclimation period, the fish were kept at 25 8C with gentle aeration and fed twice a day with dry food powder. Two-thirds of the acclimation medium was renewed every 3 d to maintain constant salinity. AgNP characterization
Transmission electronic microscope (TEM; JEOL 100CXII) was used for morphology and size observation. For the TEM sample preparation, the diluted AgNP stock was ultrasonicated for 2 min to disturb any aggregates formed during dilution. The dispersion obtained was dripped onto a carbon film–covered copper grid with a piece of filter paper below. After being dried in the desiccator, the sample was placed into the specimen holder and vacuumed for 30 s before observation. The TEM pictures were obtained at an acceleration voltage of 100 kV. The mean size of AgNPs was determined by measuring 200 randomly selected particles. Dynamic light scattering (Brookhaven Instruments) was used to study the distribution and zeta potential of AgNPs in the tested medium. Particle sizes were measured 3 times under irradiation with a 35-mW red solid-state laser (660-nm wave length) using a time interval of 15 s. The zeta potential of the particles was measured twice by a palladium electrode, with 40 circles each. All measurements were conducted at 22 8C. Atomic absorption spectroscopy (Perkin-Elmer furnace) was used for silver concentration analysis. Dispersions containing AgNPs were digested in 65% HNO3 at 80 8C for 10 h, and solutions without AgNPs were mixed with 1 mL 2% HNO3 before analysis. Samples were diluted to certain ranges (between 0.5 mg/L and 1.2 mg/L) before analysis. Radioactive AgNP preparation and stability
A radiolabeling method newly developed by Hildebrand and Franke [19] was adopted with slight modifications in the present
Environ Toxicol Chem 33, 2014
633
study. In detail, the aqueous AgNP dispersion (25 mL) was centrifuged at 2000 g for 30 min. After centrifugation, the supernatant was carefully discarded. Neutralized 110mAgNO3 stock containing 7.6 mCi 110mAgþ (400 mL) was introduced by flushing the centrifuge tube wall to bring all the AgNPs to the bottom; these were then mixed by shaking gently 10 times. The mixture was thereafter heated in an oil bath at 50 8C for 3 h. The radiolabeled AgNPs were dispersed in 25 mL MiniQ water (18 MV, Thermo) by ultrasonication for 10 min in an ice-water bath. Afterwards, they were centrifuged at 2000 g for 30 min to pellet the AgNPs. The wash efficiency was checked by adding 100 mL stock into 900 mL ultrapure water and centrifugation at 10 000 g for 10 min. Supernatant (0.5 mL) was dripped out, and the remaining 0.5 mL mixture was measured for radioactivity using a Wallac 1480 NaI (T1) gamma counter (Turku). Finally, the pellet was redispersed in 25 mL trisodium citrate (1 mM) solution and stabilized in the dark overnight. The radioactivity of the stock was measured before the experiment. The concentration of the radiolabeled AgNPs was calculated as C f ¼ ðC t C d Þ Rf =R0
ð1Þ
where Cf (mg/L) is the final concentration of AgNPs based on silver mass, Ct (mg/L) is the total concentration of added AgNPs stock, Cd (mg/L) is the silver concentration of the discarded supernatant before radiolabeling (all the concentrations were transferred into a volume of 25 mL), Rf is the final radioactivity of the AgNPs dispersion, and R0 is the initial radioactivity of the AgNPs before washing. An accompanying experiment was also conducted using stable Agþ instead of 110mAgþ under the same conditions, for TEM observation of AgNPs morphology and size. The binding of radiolabeled AgNPs was checked before and after the uptake experiments by centrifuging a 1.5-mL water sample at 10 000 g for 10 min. A 0.5-mL supernatant was dripped out, and the remaining 1-mL mixture was assayed for radioactivity. The binding efficiency was calculated as the remaining percentage of radioactivity in the nanoparticles. In all 4 salinities, the average binding efficiencies were above 90%, suggesting an efficient radiolabeling method. Aggregation kinetics of AgNPs
Aggregation kinetics of AgNPs were examined at 4 different salinities in artificial seawater (1 psu, 5 psu, 15 psu, and 30 psu) in the absence and presence of Tween 20 (2 mM and 20 mM). In detail, the citrate-capped AgNP dispersion was ultrasonicated in an ice-water bath for 5 min to disrupt any possible aggregates before the experiment. The aggregation experiment was conducted in 50-mL polypropylene conical tubes (Becton Dickinson), and 30-mL solutions of artificial water with different salinities were prepared by diluting the concentrated stock to the desired salinity, using deionized water. Specific volumes of AgNPs stock were added to achieve a silver concentration of 1 mg/L. Before each measurement the tube was briefly shaken to mix the dispersion well, and 2-mL samples were taken out for hydrodynamic diameter and zeta-potential measurements at time points of 0 h, 1.5 h, 3 h, 4.5 h, and 6 h. All the aggregation experiments were conducted at room temperature of 22 8C. The main ion concentrations for the 30-psu medium were Naþ 423.5 mM, Kþ 18.6 mM, Mg2þ 65.3 mM, Ca2þ 8.9 mM, and Cl– 473.7 mM based on the measurements conducted by Tsueng et al. [20]. The pH values for the salinities studied (1 psu, 5 psu, 15 psu, and 30 psu) were 8.41, 9.02, 9.46, and 9.45, respectively.
634
Environ Toxicol Chem 33, 2014
J. Wang and W.-X. Wang
Uptake kinetics of AgNO3 and AgNPs in medaka
To compare the AgNP uptake and estimate the contribution of dissolved 110mAg to the overall AgNPs uptake, 110mAgNO3 was used to quantify the soluble Ag uptake. The uptake experiment was conducted in 500-mL plastic beakers with 200 mL of test medium. The medium was prepared by diluting filtered concentrated artificial seawater to a specific salinity (1 psu, 5 psu, 15 psu, and 30 psu) and aerated for several hours to achieve a saturated O2 concentration. For the AgNO3 uptake experiment, 0.37 mCi of 110mAgNO3 and varied volumes of stable AgNO3 (stock concentration 81 mg/mL) were added into a 200-mL medium to achieve a final Ag concentration of 8.12 mg/L and 47.6 mg/L, respectively. Finally, these test media were equilibrated in the dark overnight. The influx rate (IRAgNO3, ng/g/h) was calculated as the slope of the linear regression between the newly accumulated Ag in fish and the exposure time. The uptake rate constant (kuAgNO3, L/kg/d) was calculated as shown below kuAgNO3 ¼ IRAgNO3 24=C water1
ð2Þ
where Cwater1 (mg/L) is the exposure AgNO3 concentration. In the AgNPs uptake experiment, AgNPs were added immediately before the introduction of fish. Radiolabeled AgNPs (800 mL) stock plus 90 mL stable AgNPs stock (commercial c-AgNPs) were added to achieve a final concentration of 87.0 mg/L Ag. Four replicates were set up for each treatment. The concentration (87.0 mg/L) was selected based on the preliminary experiment, to give a reliable radioactivity count. Well-selected medaka with normal development with an average body weight of 0.16 0.02 g were not fed for 1 d preceding the uptake experiment and were maintained in fresh medium to evacuate the gut contents. One individual fish was introduced into each replicated beaker for the AgNO3 treatment while 2 individuals were introduced into each beaker for the AgNPs treatments because of the low radioactivity count (e.g.,
SALINITY INFLUENCES ON THE UPTAKE OF SILVER NANOPARTICLES AND SILVER NITRATE BY MARINE MEDAKA (ORYZIAS MELASTIGMA) JIAN WANG and WEN-XIONG WANG* Division of Life Science, State Key Laboratory of Marine Pollution, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong Special Administrative Region, China (Submitted 21 June 2013; Returned for Revision 3 September 2013; Accepted 14 November 2013) Abstract: With increasing use of silver nanoparticles (AgNPs), concerns about their potential deleterious effects on aquatic ecosystems have increased. Most previous studies have focused on the toxicity of AgNPs while their bioavailability has been seldom investigated. The present study examined the effects of salinity on the aggregation kinetics as well as the bioavailability of commercial 80-nm citrate-coated AgNPs (c-AgNPs) in the presence or absence of a nonionic surfactant (Tween 20) to marine medaka (Oryzias melastigma). In addition, the uptake of soluble Ag was quantified for comparison and for deducting the uptake of soluble Ag during AgNP exposure by applying a biokinetic model. The authors found that the addition of Tween 20 immediately slowed down the process of aggregation of AgNPs, and an elevated amount of Tween 20 (20 mM) kept AgNPs well dispersed, even in the 30-psu salinity medium. Uptake rate constants (ku) of AgNPs were less than half the soluble Ag at low salinities (1 psu and 5 psu), while limited bioavailability of c-AgNPs was observed at high salinities (15 psu and 30 psu). However, the Tween 20–stabilized AgNPs (t-AgNPs) were accumulated by medaka at comparable rates as the soluble Ag, indicating the importance of dispersion for bioavailability of AgNPs in a highly ionic environment. The present study provided the first insight of the bioavailability of AgNPs to fish in a high-ionic environment. More studies are needed to gain a full understanding of bioavailability of AgNPs in marine environments. Environ Toxicol Chem 2014;33:632–640. # 2013 SETAC Keywords: Silver nanoparticles
Uptake kinetic
Bioavailability
Aggregation
water, sediment, and particles) with continuous dissolution throughout the whole process [11]. Therefore, both dissolved Ag and the AgNPs themselves need to be considered in evaluating the potential impacts of AgNPs. Soluble Ag has been reported to be a toxic metal that inhibits several osmoregulation-related enzymes of fish [3]. Previous study has shown that AgNPs at a size range of 5 nm to 46 nm can be transported into the zebrafish embryo and distributed into the heart, yolk, and blood of the embryo, causing toxic effects such as delayed hatching, mortality, and malformation [12]. Waterborne exposure to AgNPs also caused an elevated silver burden in the gills of zebrafish [13], and enzymes of juvenile rainbow trouts such as Naþ, Kþ-adenosine triphosphate-ase (ATPase) were suppressed after 3 h of exposure to 10 mg/L AgNPs [14]. Biomarkers such as metallothionein, glutathione-S-transferase, or the p53 gene responded after chronic exposure to AgNPs in medaka [15]. In these earlier studies, the observable toxic concentrations of AgNPs were magnitudes higher than the estimated environmental concentrations [16]. The bioavailability of AgNPs at sublethal concentrations has seldom been studied. In addition, these studies mainly focused on freshwater fish, while the impact of AgNPs on marine fish remains unknown. Although AgNPs tend to form aggregates in highionic environments [17], it is now possible to keep AgNPs suspended in high-ionic medium with the application of nonionic surfactant Tween 80 [18]. Therefore, the present study investigated the stabilization of the nonionic surfactant Tween 20 on the aggregation kinetics and uptake of the c-AgNPs at different salinities. A euryhaline species of marine medaka, Oryzias melastigma, was used as the model organism in the present study. A newly developed radiolabeling method was applied to study the bioavailability of AgNPs at sublethal concentrations. In addition, because dissolution of AgNPs is unavoidable in the saline medium,
INTRODUCTION
Silver is used extensively in various sectors such as jewelry, coins, and conductive elements, with an annual global consumption of 28 million kg. [1]. The recent development of nanotechnology extends the applications of silver (e.g., cosmetics, optics, and textiles) in the form of silver nanoparticles (AgNPs). Extensive use of AgNPs in these products will result in increased inputs of AgNPs in the environment. Among the potential discharge routes, release of commercial nanoparticles into the aquatic system has attracted the maximum concern [2]. As a toxic rare metal, soluble silver and its impact on aquatic systems have long been studied [3], but our understanding of the bioavailability of AgNPs is still limited. Compared with Agþ, the behavior of AgNPs is much more complex. Once AgNPs are introduced into the aquatic systems, various transformations of AgNPs take place [4]; processes such as aggregation and dissolution have been addressed in many earlier studies [5]. Particles of different sizes and coatings also behave differently. Silver nanoparticles of smaller sizes tend to aggregate faster [6], and coatings such as sodium dodecyl sulfate and Tween 80 can effectively elevate the critical coagulation concentrations of AgNPs [7]. Based on these aggregation processes, several models have been developed to assess the particle–particle and particle–surface interaction under laboratory conditions [8]. However, in aquatic systems, the presence of natural organic matter (NOM) as well as various environmental factors (e.g., temperature, pH, and dissolved oxygen) can complicate the interaction [9] and affect the dissolution process [10]. The AgNPs distribute into different compartments (e.g., * Address correspondence to [email protected]. Published online 28 November 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/etc.2471 632
Salinity influences on silver nanoparticle uptake by marine medaka
the uptake kinetics of AgNO3 were also studied for comparison, and the uptake rate constant was calculated to deduct the soluble Ag uptake by applying a biokinetic model. MATERIALS AND METHODS
Chemicals and organisms
Silver nitrate (ACS reagent, >99%) and sodium citrate tribasic dihydrate (Purum p.a., >99%) were purchased from Sigma Aldrich. Commercially available 80-nm c-AgNPs were purchased from Ted Pella. Radioactive 110mAgNO3 (POLATOM) were purchased in the form of an aqueous solution with 0.1 M HNO3, and an equal volume of 0.1 M NaOH was added for neutralization before use. Instant Sea Salts were purchased from Instant Ocean, and the test media were prepared by diluting the filtered (pore size 0.22 mm, Milipore) concentrated stock to a specific salinity (1 psu, 5 psu, 15 psu, and 30 psu). Marine medaka (Oryzias melastigma) were obtained from City University of Hong Kong and reared in sand-filtered natural flowing seawater from Clear Water Bay (Kowloon, Hong Kong). The fish were kept in a natural light cycle at room temperature (22–25 8C). Dry food powder (TROUW) was fed to the fish twice a day, and the tank was cleaned regularly to remove the algae and feces on the bottom. After 3 mo of growth, the adult fish were randomly transferred into 20-L glass aquaria with 15 L of 1-psu, 5-psu, 15-psu, or 30-psu salinity medium for acclimation at a density of 2 individuals/L. The acclimation period lasted for 3 wk before the initiation of uptake experiments. Within the acclimation period, the fish were kept at 25 8C with gentle aeration and fed twice a day with dry food powder. Two-thirds of the acclimation medium was renewed every 3 d to maintain constant salinity. AgNP characterization
Transmission electronic microscope (TEM; JEOL 100CXII) was used for morphology and size observation. For the TEM sample preparation, the diluted AgNP stock was ultrasonicated for 2 min to disturb any aggregates formed during dilution. The dispersion obtained was dripped onto a carbon film–covered copper grid with a piece of filter paper below. After being dried in the desiccator, the sample was placed into the specimen holder and vacuumed for 30 s before observation. The TEM pictures were obtained at an acceleration voltage of 100 kV. The mean size of AgNPs was determined by measuring 200 randomly selected particles. Dynamic light scattering (Brookhaven Instruments) was used to study the distribution and zeta potential of AgNPs in the tested medium. Particle sizes were measured 3 times under irradiation with a 35-mW red solid-state laser (660-nm wave length) using a time interval of 15 s. The zeta potential of the particles was measured twice by a palladium electrode, with 40 circles each. All measurements were conducted at 22 8C. Atomic absorption spectroscopy (Perkin-Elmer furnace) was used for silver concentration analysis. Dispersions containing AgNPs were digested in 65% HNO3 at 80 8C for 10 h, and solutions without AgNPs were mixed with 1 mL 2% HNO3 before analysis. Samples were diluted to certain ranges (between 0.5 mg/L and 1.2 mg/L) before analysis. Radioactive AgNP preparation and stability
A radiolabeling method newly developed by Hildebrand and Franke [19] was adopted with slight modifications in the present
Environ Toxicol Chem 33, 2014
633
study. In detail, the aqueous AgNP dispersion (25 mL) was centrifuged at 2000 g for 30 min. After centrifugation, the supernatant was carefully discarded. Neutralized 110mAgNO3 stock containing 7.6 mCi 110mAgþ (400 mL) was introduced by flushing the centrifuge tube wall to bring all the AgNPs to the bottom; these were then mixed by shaking gently 10 times. The mixture was thereafter heated in an oil bath at 50 8C for 3 h. The radiolabeled AgNPs were dispersed in 25 mL MiniQ water (18 MV, Thermo) by ultrasonication for 10 min in an ice-water bath. Afterwards, they were centrifuged at 2000 g for 30 min to pellet the AgNPs. The wash efficiency was checked by adding 100 mL stock into 900 mL ultrapure water and centrifugation at 10 000 g for 10 min. Supernatant (0.5 mL) was dripped out, and the remaining 0.5 mL mixture was measured for radioactivity using a Wallac 1480 NaI (T1) gamma counter (Turku). Finally, the pellet was redispersed in 25 mL trisodium citrate (1 mM) solution and stabilized in the dark overnight. The radioactivity of the stock was measured before the experiment. The concentration of the radiolabeled AgNPs was calculated as C f ¼ ðC t C d Þ Rf =R0
ð1Þ
where Cf (mg/L) is the final concentration of AgNPs based on silver mass, Ct (mg/L) is the total concentration of added AgNPs stock, Cd (mg/L) is the silver concentration of the discarded supernatant before radiolabeling (all the concentrations were transferred into a volume of 25 mL), Rf is the final radioactivity of the AgNPs dispersion, and R0 is the initial radioactivity of the AgNPs before washing. An accompanying experiment was also conducted using stable Agþ instead of 110mAgþ under the same conditions, for TEM observation of AgNPs morphology and size. The binding of radiolabeled AgNPs was checked before and after the uptake experiments by centrifuging a 1.5-mL water sample at 10 000 g for 10 min. A 0.5-mL supernatant was dripped out, and the remaining 1-mL mixture was assayed for radioactivity. The binding efficiency was calculated as the remaining percentage of radioactivity in the nanoparticles. In all 4 salinities, the average binding efficiencies were above 90%, suggesting an efficient radiolabeling method. Aggregation kinetics of AgNPs
Aggregation kinetics of AgNPs were examined at 4 different salinities in artificial seawater (1 psu, 5 psu, 15 psu, and 30 psu) in the absence and presence of Tween 20 (2 mM and 20 mM). In detail, the citrate-capped AgNP dispersion was ultrasonicated in an ice-water bath for 5 min to disrupt any possible aggregates before the experiment. The aggregation experiment was conducted in 50-mL polypropylene conical tubes (Becton Dickinson), and 30-mL solutions of artificial water with different salinities were prepared by diluting the concentrated stock to the desired salinity, using deionized water. Specific volumes of AgNPs stock were added to achieve a silver concentration of 1 mg/L. Before each measurement the tube was briefly shaken to mix the dispersion well, and 2-mL samples were taken out for hydrodynamic diameter and zeta-potential measurements at time points of 0 h, 1.5 h, 3 h, 4.5 h, and 6 h. All the aggregation experiments were conducted at room temperature of 22 8C. The main ion concentrations for the 30-psu medium were Naþ 423.5 mM, Kþ 18.6 mM, Mg2þ 65.3 mM, Ca2þ 8.9 mM, and Cl– 473.7 mM based on the measurements conducted by Tsueng et al. [20]. The pH values for the salinities studied (1 psu, 5 psu, 15 psu, and 30 psu) were 8.41, 9.02, 9.46, and 9.45, respectively.
634
Environ Toxicol Chem 33, 2014
J. Wang and W.-X. Wang
Uptake kinetics of AgNO3 and AgNPs in medaka
To compare the AgNP uptake and estimate the contribution of dissolved 110mAg to the overall AgNPs uptake, 110mAgNO3 was used to quantify the soluble Ag uptake. The uptake experiment was conducted in 500-mL plastic beakers with 200 mL of test medium. The medium was prepared by diluting filtered concentrated artificial seawater to a specific salinity (1 psu, 5 psu, 15 psu, and 30 psu) and aerated for several hours to achieve a saturated O2 concentration. For the AgNO3 uptake experiment, 0.37 mCi of 110mAgNO3 and varied volumes of stable AgNO3 (stock concentration 81 mg/mL) were added into a 200-mL medium to achieve a final Ag concentration of 8.12 mg/L and 47.6 mg/L, respectively. Finally, these test media were equilibrated in the dark overnight. The influx rate (IRAgNO3, ng/g/h) was calculated as the slope of the linear regression between the newly accumulated Ag in fish and the exposure time. The uptake rate constant (kuAgNO3, L/kg/d) was calculated as shown below kuAgNO3 ¼ IRAgNO3 24=C water1
ð2Þ
where Cwater1 (mg/L) is the exposure AgNO3 concentration. In the AgNPs uptake experiment, AgNPs were added immediately before the introduction of fish. Radiolabeled AgNPs (800 mL) stock plus 90 mL stable AgNPs stock (commercial c-AgNPs) were added to achieve a final concentration of 87.0 mg/L Ag. Four replicates were set up for each treatment. The concentration (87.0 mg/L) was selected based on the preliminary experiment, to give a reliable radioactivity count. Well-selected medaka with normal development with an average body weight of 0.16 0.02 g were not fed for 1 d preceding the uptake experiment and were maintained in fresh medium to evacuate the gut contents. One individual fish was introduced into each replicated beaker for the AgNO3 treatment while 2 individuals were introduced into each beaker for the AgNPs treatments because of the low radioactivity count (e.g.,
