Regulation of sodium and calcium in Daphnia magna exposed to silver nanoparticles

Authors


Abstract

The toxicity of manufactured silver nanoparticles (AgNPs) has been widely studied, but the influence of AgNPs on the major ions (such as sodium [Na] and calcium [Ca]) regulations are unknown. In the present study, a freshwater cladoceran Daphnia magna was exposed to commercial AgNPs coated with polyvinylpyrrolidone. After 48 h, the Na body content was significantly reduced by AgNO3 exposure, but the Ca body content was significantly increased under AgNO3 and AgNP exposures, respectively. No effect was observed on the body concentrations of Na and Ca at 50 to 500 µg/L AgNPs with 1-µM cysteine addition. Exposure of AgNO3 and AgNPs inhibited the Na influx and elevated the Na efflux. In contrast, their exposure increased the Ca influx, but did not affect the Ca efflux. The results of the present study demonstrated the significant influences of AgNO3 and AgNPs (without cysteine) on Na and Ca regulations. Such effect of AgNPs on Na and Ca regulation disappeared after cysteine addition, indicating that the soluble Ag released from AgNPs played a major role in the ionoregulatory dysfunction. Environ. Toxicol. Chem. 2013;32:913–919. © 2013 SETAC

INTRODUCTION

Nanoparticles have unique physicochemical properties due to their small sizes (1–100 nm), and are now widely applied in the industries of electronics, catalysis, and medicine 1, 2. Among the different types of nanoparticles, silver nanoparticles (AgNPs) have a wide spectrum of antibacterial property and are used in a large number of consumer and personal care products, such as socks, detergents, and medical bandages 3, 4. The AgNPs have now become one of the fast-growing manufactured products because of the high demand for products that contain AgNPs. Thus, it is inevitable that AgNPs will enter into the environment during their production and consumption 5–7. After aggregation and sedimentation, AgNPs remaining in the water may be toxic to organisms. The AgNP toxicity has already been investigated in algae 8, 9, bacteria 10, 11, cladocerans 12, 13, and fish 14, 15. Physiological and biochemical endpoints related to toxicity have been addressed in previous studies 16, 17. However, the ion flux that requires radioisotopes for accurate quantification has not yet been investigated in aquatic organisms.

Sodium (Na) and calcium (Ca) are macroelements necessary for the metabolism of organisms. Normal regulation of Na and Ca is important for freshwater crustaceans such as Daphnia magna. However, it was found that soluble silver could inhibit the activity of Na-K adenosine triphosphatase (Na-K-ATPase) located in the absolateral membrane of epithelium, decrease the Na influx, and finally result in dysfunction of ion regulation 18. Whether AgNPs have similar effects on the Na flux and the underlying mechanism is still not known. In addition, daphnids require a large amount of Ca during their life cycle 19. Calcium regulation may play an important role not only in molting but also in other capacities such as nerve function. It was found that nanoparticles could affect the Ca content in cell lines and cause an abrupt increase of both extracelluar and intracellular Ca influx 19. Whether AgNPs can affect the Ca homeostasis and regulation in daphnids is still unknown.

To understand the influence of AgNP exposure on Na and Ca regulation, we initially exposed daphnids to commercial AgNPs coated with polyvinylpyrrolidone (PVP). Then, Na and Ca body contents and their flux (influx and efflux) in daphnids were quantified using radioisotope methodology. To explore the underlying mechanisms of AgNPs, effects, we conducted AgNP exposure at low (2 µg/L, without cysteine) and high (200 µg/L, with 1 µM cysteine) concentrations, and made comparisons with AgNO3.

MATERIALS AND METHODS

Organisms, medium, and nanoparticles

Daphnia magna, the freshwater cladoceran, was raised in creek water collected from the campus of Hong Kong University of Science and Technology. Water was filtered through a glass-fiber membrane (GF/C, pore size 1.2 µm; Maistone) to remove the particles before use. The density of daphnids was kept at 1 individual/10 ml, and water was refreshed every 2 d. The freshwater green alga, Chlamydomonas reinhardtii, was cultured in the artificial Woods Hole Chu-10 medium (containing 36.8 mg/L CaCl2 × 2 H2O, 37 mg/L MgSO4 × 7 H2O, 12.6 mg/L NaHCO3, 11.4 mg/L K2HPO4 × 3 H2O, 85 mg/L NaNO3, 21.1 mg/L Na2SiO3 × 9 H2O, micronutrients and vitamin solution 20) with bubbled air and harvested at the exponential phase of growth. Then algae were centrifuged and resuspended in filtered creek water, and fed at a food concentration of 5 × 104 (daphnids ≤ 3 d) or 1 × 105 cells/ml (daphnids > 3 d). Both daphnids and algae were maintained in a walk-in chamber at 23.5°C with 16:8 light:dark cycle.

The medium used in the present study was the International Organization for Standardization (ISO) test water, which was recommended in Daphnia sp. Acute Immobilization Test (2004) by the U.S. Environmental Protection Agency (U.S. EPA), with modified Ca concentration (20 mg/L) because Ca had significant effect on the size distribution of AgNPs in the medium 21. It contained four constituents, including 73.5 mg/L CaCl2 × 2 H2O, 123.3 mg/L MgSO4 × 7 H2O, 63.8 mg/L NaHCO3, and 5.8 mg/L KCl, and was freshly prepared before each experiment. The pH value was adjusted from 8 to 8.2 with NaOH (for Ca influx) or KOH (for Na influx).

The suspension of AgNPs coated with PVP was purchased from NanoSys GmbH. To ensure the particle size in the nanoscale, the suspension was initially passed through a 100-nm membrane (Millipore). The filtrate was kept as AgNP stock. The shape and original size of AgNPs were observed by transmission electron microscopy (TEM; JEOL 2010F) with an acceleration voltage of 200 kV. The TEM sample was prepared by dropping ethanol-diluted AgNP suspension onto the copper grid covered with carbon membrane. After being dried in the desiccator, the sample was placed on the specimen holder and inserted into the TEM for observation. The size distribution and zeta potential, which reflected the stability of AgNPs in medium, were determined by dynamic light scattering (DLS; Brookhaven Instruments). Total Ag concentration in AgNP stock was determined by atomic absorption spectrometry (AAS; Perkin-Elmer) after digestion in 65% HNO3 at 80°C. Soluble Ag released from AgNPs in the stock was collected by ultracentrifugation through a 3-kD membrane (pore size around 1 nm; Millipore) at 160 g for 20 min. The Ag concentration in the filtrate was also analyzed by AAS. Therefore, AgNP concentration in the stock was calculated as the difference between the total Ag concentration and the soluble Ag concentration.

Na and Ca body burdens in daphnids exposed to Ag

To determine the influence of AgNO3 or AgNP exposure on Na and Ca body burdens, daphnids were exposed to AgNO3 and AgNPs for 48 h without food addition. The Na and Ca concentrations in the medium were 18 and 20 mg/L, respectively. The Ag concentrations were as follows: 0 (control); 1, 2, and 4 µg/L of AgNPs; 0.1, 0.2, and 0.4 µg/L of AgNO3; and 50, 200, and 500 µg/L of AgNPs with 1-µM cysteine. Cysteine was used to combine with soluble Ag released from AgNPs; thus, it reduced the soluble Ag bioavailability in daphnids 22. There were three replicates in each treatment and each replicate contained 25 individual daphnids. To reduce the variation of AgNP concentration, exposure media were refreshed every 24 h. After the 48-h exposure, the daphnids in each treatment were collected by the mesh and washed in the MilliQ for 1 min. Then they were filtered onto 14-µm polycarbonate membrane and rinsed another three times. After washing, daphnids were dried at 80°C and digested in 65% HNO3 for 24 h. The Ag concentration in the digestion was measured by AAS. The Na and Ca concentrations were analyzed by inductively coupled plasma atomic emission spectroscopy (Perkin-Elmer).

Na influx and efflux under AgNP and AgNO3 exposures

The influences of AgNP exposure on Na influx were investigated in two experiments. In the first experiment, Na influx was quantified at one fixed Na concentration (18-mg/L Na, spiked with 12-µCi/L 22Na; Perkin Elmer). The Ag concentrations were set as follows: control; 0.1, 0.2, and 0.4 µg/L AgNO3; 1, 2, and 4 µg/L AgNPs (without cysteine); and 50, 200, and 500 µg/L AgNPs (with 1-µM cysteine addition). In the second experiment, we quantified the Na influx at various Na concentrations (0.5, 2, 10, 50, and 200 mg/L) under the exposure of control (no Ag), 0.3 µg/L AgNO3, 2 µg/L AgNPs, and 200 µg/L AgNPs (with 1-µM cysteine), respectively. The 22Na concentrations spiked were 8 µCi/L for 0.5 mg/L Na, 10 µCi/L for 2 and 10 mg/L Na, 15 µCi/L for 50 mg/L Na, and 25 µCi/L for 200 mg/L Na, respectively.

Ten individual daphnids were exposed in 100 ml medium for 3 to 4 h, with three replicates for each treatment. During exposure, all the daphnids were collected and determined for Na radioactivity by a Wallac 1480 NaI (T1) gamma counter after 1 min of washing. Then, they were quickly returned to the original beakers. At the beginning and end of the experiments, 1 ml of medium was sampled to monitor the changes of radioactivity in the medium and to calculate the specific activity. The influx rate (I, mg/g/h) was the linear regression between the newly accumulated Na and the exposure time.

To investigate whether Na efflux was affected by AgNP or AgNO3 exposure, daphnids were initially radiolabeled at 18 mg/L Na (with spikes of 12-µCi/L 22Na) for 24 h. During radiolabeling, food was added at a density of 1 × 105 cells/ml to keep the daphnids healthy. No Ag was spiked into the exposure medium during radiolabeling to avoid an interaction between Ag and food. After 24-h exposure, the radioactivity of daphnids was analyzed. Subsequently, daphnids were transferred into the medium containing 0, 0.3 µg/L AgNO3, 2 µg/L AgNPs, and 200 µg/L AgNPs (with 1 µM cysteine) without 22Na for 12 h of depuration. No food was added during the depuration period. At 1, 2, 4, 6, 8, and 12 h, daphnids were collected and washed for 1 min, then measured for their radioactivity. After that, they were returned back to the refreshed medium. The density of daphnids in the depuration was 1 individual/5 ml. There were 20 individuals in each replicate and three replicates in one treatment. The Na efflux rate constant (ke, 1/d) was calculated by linear regression between the natural logarithm of the percentage of Na retention in daphnids and the depuration time.

Ca influx and efflux under AgNPs and AgNO3 exposures

The Ca influx was quantified at a fixed Ag concentration with various Ca concentrations. Daphnids were initially preexposed to 0 (control), 2 µg/L AgNPs, 200 µg/L AgNPs (with 1 µM cysteine), and 0.3 µg/L AgNO3 in the medium (Ca 20 mg/L) for 48 h. After that, 25 daphnids were transferred into the beaker containing 100 ml medium with 0.5, 2, 10, 50, and 200 mg/L of Ca. The 45Ca concentrations spiked in the medium were 5 µCi/L at 0.5, 2, and 10 mg/L Ca; 10 µCi/L at 50 mg/L Ca; and 20 µCi/L at 200 mg/L Ca. In each treatment, AgNPs or AgNO3 concentrations were the same as those used during the preexposure period. At 1, 2, 3, and 4 h, six to seven individual daphnids were randomly collected with mesh and washed in deionized water for 1 min. Then they were transferred into 7-ml scintillation tubes containing 0.5 ml of 20% HNO3. After digestion in 80°C water bath overnight, samples were added with a 5-ml cocktail (Perkin Elmer; the scintillation solution) and shaken to form the homogenous mixture. The 45Ca radioactivity in the samples was determined by LS6500 multipurpose scintillation counter (Beckman). Before and after Ca exposure, 1 ml of medium was sampled and mixed with a 5-ml cocktail to determine the radioactivity of medium. The Ca influx rate was calculated by the same method as that of Na mentioned above.

To quantify the Ca efflux, daphnids were initially radiolabeled at 0 (control), 2 µg/L AgNPs, 0.3 µg/L AgNO3, and 200 µg/L AgNPs with 20 mg/L Ca and 50 µCi/L 45Ca for 48 h without food addition. Then, the daphnids (≈55 individuals/beaker) were transferred into the medium with the same Ag and Ca concentrations, but without 45Ca. Ten individuals were collected at the beginning and end of depuration. They were washed in deionized water and then digested in 0.5 ml 20% HNO3 at 80°C (water bath) for 12 h. After adding a 5-ml cocktail, the radioactivity of samples was determined. At 6, 12, 18, 24, 36, and 48 h, the radioactivities of 20-ml medium, molts, and neonates (collected and digested with the same method mentioned before) were sampled for determination. Then, the radioactivity in daphnids at each time point was calculated by the mass balance method. The Ca efflux rate constant (ke, 1/d) was the slope of linear regression between the natural logarithm of the percentage of Ca retained in daphnids and the time.

Data were expressed as mean ± standard deviation. The statistical analysis including t-test and one-way analysis of variance (ANOVA) was specified wherever necessary.

RESULTS

Characteristics of AgNPs

The PVP-coated AgNPs had spherical shape with sizes of approximately 10 to 20 nm under TEM observation (Fig. 1). The effective diameter of AgNPs in the medium measured by DLS was 58.5 ± 1.4 nm. It increased slightly after 24 h of incubation (61.8 ± 1.6 nm) or with a 1-µM cysteine addition (62.7 ± 2.7 nm). However, the zeta potential was kept relatively constant during 24 h (–27.1 ± 4.8 mV at t = 0, and –26.2 ± 5.4 mV after 24 h) or with cysteine addition (–24.8 ± 5.7 mV). Generally, AgNPs were relatively well dispersed in the medium. AgNP concentration in the stock was 737.2 mg/L and the concentration of soluble Ag released from AgNPs in the stock was 26.7 mg/L (nearly 3.6% of total Ag). After transfer into the medium, AgNPs (1–4 µg/L) released 0.096 to 0.48 µg/L soluble Ag during 24 h, which accounted for 10.5 to 19.2% of the total AgNPs (Fig. 2). More than 80% AgNO3 remained in the soluble form in medium. After 24-h exposure, soluble Ag concentration released from AgNPs was 1.4-, 1.2-, and 1.5-fold higher than that of AgNO3 (i.e., 1 µg/L AgNPs vs 0.1 µg/L AgNO3, 2 µg/L AgNPs vs 0.2 µg/L AgNO3, and 4 µg/L AgNPs vs 0.4 µg/L AgNO3, respectively). Cysteine stimulated soluble Ag release. After the addition of 1 µM cysteine for 24 h, the concentrations of soluble Ag released from 50, 200, and 500 µg/L AgNPs were 22.8, 26.5, and 31.6 µg/L, which accounted for 49.6, 14.8, and 7.6% of the actual AgNP concentrations, respectively. Because Ag+ had a strong affinity for cysteine with a stability constant of silver–cysteine complex AgCysH (log K) as high as 11.9 23, Ag+ immediately combined with cysteine after its release from AgNPs and resulted in the reduction of nanoparticle concentration. Thus, to avoid the decrease of AgNP concentration in the 48-h exposure, the medium was refreshed every 24 h in the bioaccumulation experiment. Most of the AgNPs were still in the particulate form, except in the 50-µg/L AgNP treatment in which approximately 50% was in a dissolved phase after 24 h.

Figure 1.

Commercial silver nanoparticles (AgNPs) coated with polyvinylpyrrolidone observed by transmission electron microscopy.

Figure 2.

Concentration of soluble silver in the medium at 0, 3, and 24 h of suspension. Silver nanoparticles (AgNPs)-L: AgNPs at low concentrations. AgNPs-H: AgNPs at high concentrations with cysteine. Data are mean ± standard deviation (n = 3).

Na and Ca bioaccumulation under silver exposure

Under AgNO3 and AgNP exposure, Ag body content in daphnids increased with Ag concentration in the medium (Fig. 3A and B). The bioaccumulation of AgNPs at 1 µg/L (5.0 ± 1.1 µg/g dry wt) was comparable to that of 0.4 µg/L AgNO3 (6.7 ± 0.4 µg/g dry wt, p > 0.05, t-test). At the exposure of 500 µg/L AgNPs with 1 µM cysteine, the Ag concentration in daphnids reached 900 µg/g dry weight. Such a large bioaccumulation of AgNPs had no significant influence on Na and Ca body contents (p < 0.05, t-test; Fig. 3D and F). However, AgNO3 (0.2 and 0.4 µg/L) significantly decreased the Na body burden (Fig. 3C) but increased the Ca body content (t-test, p < 0.05; Fig. 3E). Similarly, AgNPs at 2 µg/L also elevated the Ca bioaccumulation (p < 0.05, t-test) but had no significant influence on the Na body content.

Figure 3.

Concentrations of Ag, Na, and Ca in daphnids after 48-h exposure to 0.1 to 0.4 µg/L AgNO3, 1 to 4 µg/L AgNPs (without cysteine) (A,C,E), and 50 to 500 µg/L AgNPs with 1-µM cysteine addition (B,D,F). The asterisk represents significant difference between treatment and control (t-test, p < 0.05). Data are mean ± standard deviation (n = 3).

Na influx and efflux

At fixed Ag concentrations, the Na influx was initially increased and then gradually leveled off at a high Na concentration (Fig. 4A). Silver exposure decreased the Na influx, especially at the high Na concentration of 200 mg/L (compared with the control, p < 0.05, t-test). Because influx measurements were conducted with different batches of daphnids, the control (Na influx at the 50 mg/L Na without Ag addition) was repeated at each time of Ag exposure to avoid the deviation among batches. The 2 µg/L AgNPs significantly decreased the Na influx by 7.5% (p < 0.05, t-test; Fig. 4B). The exposure of 0.3 µg/L AgNO3 and 200 µg/L AgNPs with 1 µM cysteine did not influence the Na influx at 50 mg/L Na.

Figure 4.

Na influx rate in daphnids under the exposure of control (no Ag), 0.3 µg/L AgNO3, 2 µg/L silver nanoparticles (AgNPs; without cysteine), 200 µg/L AgNPs (with 1 µM cysteine). (A) Na concentrations in the medium ranged from 0.5 to 200 mg/L; (B) at 50 mg/L Na. The asterisk indicates the significant difference between treatment and the control (t-test, p < 0.05). Data are mean ± standard deviation (n = 3).

In the second experiment, the Na influx was quantified at one fixed Na concentration (18 mg/L Na). The Na influx significantly decreased at 4 µg/L AgNP without cysteine addition (p < 0.05, one-way ANOVA; Fig. 5). Although AgNO3 exposure did not significantly affect Na influx compared with the control, Na influx was indeed reduced along AgNO3 concentrations (p < 0.05, one-way ANOVA). High AgNP concentrations (200 µg/L) with 1 µM cysteine addition did not influence Na influx.

Figure 5.

Na influx rate in daphnids at 18 mg/L Na under the exposure of AgNO3 (left panel), low silver nanoparticles (AgNPs; without cysteine; middle panel), and high AgNPs (with 1 µM cysteine; right panel). Different letters indicate the difference between treatments (p < 0.05). Data are mean ± standard deviation (n = 3).

In the Na efflux experiment, the percentage of Na incorporated in daphnids decreased dramatically during 12 h of depuration, indicating a fast Na metabolism in daphnids (Fig. 6A). Na depuration in daphnids fit into a one-compartment model. The AgNO3 exposure led to the lowest percentage of Na retaining in daphnids, followed by the 2 µg/L AgNP exposure. The retention of Na at 200 µg/L AgNPs with 1 µM cysteine was comparable to that of the control. The ke of Na increased significantly with the exposure of AgNO3 and 2 µg/L AgNPs (t-test, p < 0.05; Fig. 6B), indicating that soluble Ag accelerated the Na efflux in daphnids.

Figure 6.

(A) Percentage of Na retained in daphnids during 12-h depuration; (B) Na efflux rate constant (ke) at the exposure of control (no Ag), 0.3 µg/L AgNO3, 2 µg/L AgNPs (without cysteine), and 200 µg/L silver nanoparticles (AgNPs; with 1 µM cysteine). The asterisk indicates significant difference between treatment and control (t-test, p < 0.05). Data are mean ± standard deviation (n = 3).

Ca influx and efflux

Similar to Na, the Ca influx rate was increased (from 0.5 to 10 mg/L Ca) and then progressively reached the plateau at a high Ca concentration (10–200 mg/L; Fig. 7). Both AgNO3 and 2-µg/L AgNP exposures significantly elevated the Ca influx at 10 to 200 mg/L Ca (compared with the control at each concentration; t-test, p < 0.05). At 200 µg/L AgNPs with 1 µM cysteine, the Ca influx was comparable to that of the control at each Ca concentration (t-test, p > 0.05). For Ca efflux, the percentage of Ca retained in daphnids decreased continuously within 48 h of depuration, and more than 80% of Ca remained in the body at the end of depuration (Fig. 8A). The AgNO3 and AgNP exposure did not affect the Ca efflux (compared with control; t-test, p > 0.05). Approximately 5.2 to 6.5% and 2.1 to 4.5% of Ca were distributed in the excretion and molting compartments, respectively (Fig. 8B).

Figure 7.

Ca influx rate in daphnids at different Ca concentrations under the exposure of control (no Ag), 0.3 µg/L AgNO3, 2 µg/L silver nanoparticles (AgNPs; without cysteine), and 200 µg/L AgNPs (with 1 µM cysteine). Data are mean ± standard deviation (n = 3).

Figure 8.

(A) Percentage of Ca retained in daphnids during 48-h depuration; (B) relative percentage of Ca distributed among daphnids and efflux routes (excretion and molt). Data are mean ± standard deviation D (n = 3).

DISCUSSION

Na and Ca bioaccumulation under silver exposure

Silver nanoparticle bioaccumulation in daphnids reached 900 µg/g dry weight after 48-h exposure to 500 µg/L AgNPs, which was consistent with results from a previous study 24. Such a large bioaccumulation of AgNPs in daphnids was caused mainly by the ingestion of AgNPs 12, 25. The concentrations of soluble Ag released from 1, 2, and 4 µg/L AgNPs were approximately 1.2 to 1.5 times higher than those from 0.1, 0.2, and 0.4 µg/L AgNO3. However, Ag body concentrations in daphnids exposed to 1, 2, and 4 µg/L AgNPs were 2.6- to 10-fold greater than those of AgNO3. This indicated that nanoparticles still played a role in bioaccumulation of AgNPs even at the low AgNP concentration. The AgNO3 caused ionoregulatory disturbance in daphnids, which was indicated by a slight but significant decrease of Na body burden and increase of Ca body burden during the 48-h exposure. It has been found that a reduction of Na body content was caused by inhibition of Na-K-ATPase activity at the basolateral membrane of gill epithelium in daphnids 18.

Interestingly, the present study also found that the Ca body burden was raised under AgNO3 exposure. The protective effect of Ca on Ag toxicity was observed in some aquatic organisms 26. Wood et al. 27 found a slight increase of Ca in the plasma of rainbow trout during the first 2 d of 10-µg/L Ag exposure. However, little or no effect on Ca body burden was observed under chronic exposure. Although a similar effect on Ca content was found in the present study, it was unknown whether the increase of Ca content would persist under long-term exposure. Therefore, the ionoregulatory disturbance of Ca under AgNO3 exposure needs to be further confirmed during chronic exposure. Significant elevation of Ca content was also observed under the exposure of low AgNP concentration (2 µg/L). Nevertheless, we did not find any inhibition on the Na body burden, although AgNPs released a higher concentration of soluble Ag than AgNO3. This may be due to the speciation of soluble Ag. In AgNP suspension, Ag+ released from AgNPs may be associated with the free PVP (dissociated from the nanoparticles), leading to a decreased Ag+ toxicity. Thus, the effect of AgNPs and AgNO3 on Na and Ca body content may be better interpreted in terms of Ag+ concentration. At high concentrations of AgNPs, no effects on Na and Ca contents were observed when the soluble Ag was combined with cysteine. Although a large quantity of AgNPs was retained in the gut (whether the AgNPs were transported into the epithelia cells was unknown 12, 25), the daphnids were still in healthy condition with unaffected Na and Ca regulation. Therefore, soluble Ag in the AgNP suspension was most likely the cause of the ionoregulatory disturbance in the daphnids.

Na influx and efflux

When increasing the Na concentration from 0 to 200 mg/L, the Na influx quickly increased and then gradually leveled off. This pattern indicated that the Na influx was enzymatically involved and controlled by Na concentrations. Bianchini and Wood 18 also found a similar pattern of Na influx within 0 to 100 mg/L at 0.3-µg/L AgNO3 exposure. The decreased Na influx rate at 50 mg/L Na under 2-µg/L AgNP exposure indicated that the uptake of Na and Ag was competitive. Because the size of AgNPs was much larger than that of the Na channel, hence the soluble Ag released from AgNPs inhibited the Na influx instead of the nanoparticles themselves. This was also proven by the exposure of 200 µg/L AgNPs with 1 µM cysteine, which had no effect on Na influx.

In the second experiment, Na influx gradually decreased with increasing AgNO3 concentrations. Such inhibition was also found at low AgNP concentrations without cysteine addition. However, after cysteine addition, this effect disappeared at 50 to 500 µg/L AgNPs. These patterns were similar to that of Na influx (50 mg/L Na) under AgNO3 and AgNP exposures, again indicating that AgNPs inhibited the Na influx through soluble Ag release. AgNO3 could significantly inhibit the activity of Na-K-ATPase 18. Decreased activity of Na-K-ATPase may block Na+ transportation into the extracellular fluid and finally depress Na influx into the epithelial cells of gills 28. In the present study, the decrease of Na influx induced by AgNPs and AgNO3 was rather small (14.6% at 4 µg/L AgNPs compared with the control, and 11.9% at 0.4 µg/L AgNO3 compared with 0.1 µg/L AgNO3). This may be due to the short exposure time (3 h) because the inhibition of Na influx was related to the exposure time. In rainbow trout, AgNO3 exposure initially decreased the Na influx by nearly 40% (0- to 4-h exposure) and finally by 99% (46- to 50-h exposure), which was much more obvious than that of daphnids 27. It appeared that such an effect on Na influx could only be observed under waterborne Ag exposure. Dietborne Ag had no influence on Na influx and Na-K-ATPase activity of crayfish 29, indicating that the inhibition of Na influx may be related to Ag exposure pathways. Whether this phenomenon can be observed in daphnids still needs to be tested.

Compared with other metals, Na ke was more than 10 times higher than those of Ag, Cd, Cu, and Zn in daphnids 22, 30, 31, and even higher than that of Ca, which was highly regulated in the crustaceans 19. The AgNO3 and 2-µg/L AgNP exposures significantly increased the Na ke by 15.7 and 7.8% compared with the control. Nevertheless, 200 µg/L AgNPs with 1 µM cysteine had no influence on Na efflux. Increased Na efflux was also observed in crayfish at 8.4 µg/L AgNO3 within the first 24-h exposure, but it gradually recovered after 3 d 32. Bury 33 also found increased Na efflux in rainbow trout exposed to 8.3 nM AgNO3 for 24 h. Increased efflux with inhibited influx of Na could result in the suppressed Na body content in daphnids. However, in the present study, the highest reduction of Na body burden under AgNO3 and AgNP exposures was 11.0 to 13.6% after 2 d of exposure. Wood et al. 27 also found that Na influx was inhibited dramatically in 24 h, but Na content in the plasma of rainbow trout was not significantly reduced. This indicated that some other mechanisms on Na compensation may be activated.

Ca influx and efflux

It was interesting that the Ca influx rate could be stimulated under AgNO3 exposure, especially at high Ca concentration. Such an elevation of Ca influx was also found at 200 mg/L Ca under the 2-µg/L AgNP exposure (p < 0.05, t-test). This may result in an increased Ca body content because Ca efflux was not affected by Ag exposure. The elevation of Ca content may be due to the toxicity of soluble Ag in daphnids because intracellular Ca homeostasis may be in dysfunction under stress. Extracellular Ca influx could be stimulated, which would have resulted in an increased Ca influx and body content 34. The Ca efflux was rather slow in the present study. The Ca ke (0.11–0.14/d) was approximately 7 to 16 times lower than that under normal condition 19. After 48-h depuration, more than 80% Ca still remained in the daphnids. Such slow Ca metabolism may be due to the starvation because food was not supplied during exposure and depuration periods. This resulted in a slow Ca turnover in daphnids.

In summary, the Na influx in daphnids was inhibited under the exposure of 2 µg/L AgNPs. However, it was not affected by 200 µg/L AgNPs with 1 µM cysteine, indicating that soluble Ag played an important role in the reduction of Na influx. In addition, exposure of 0.3 µg/L AgNO3 and 2 µg/L AgNPs increased the Na efflux, which resulted in a decreased Na body content after 48 h of AgNO3 exposure. In contrast, AgNO3 and AgNPs (2 µg/L) increased the Ca influx, but did not influence the Ca efflux. This caused an increased Ca body content in the daphnids. Overall, the present study found that AgNPs could induce ionoregulatory dysfunction, which was likely caused by soluble Ag released from AgNPs.

Acknowledgements

We thank the anonymous reviewers for their comments on this work. This study was supported by General Research Funds from the Hong Kong Research Grants Council (663011) and Research Project Competition grant (RPC10.SC10) to W.-X. Wang.

Ancillary