Higher silver bioavailability after nanoparticle dietary exposure in marine amphipods

Materials, reagents, and equipment

The reagents used were HNO₃ with 65% purity (Merck; Sigma-Aldrich); 1000 mg L⁻¹ Ag monoelementar standard solution (Quemis), AgNO₃ with ≥99% purity (Sigma-Aldrich), and other reagents generally used in chemical analysis. Glassware, plastic bottles, and other materials used during the collection and analysis of samples were decontaminated with 10% (v/v) HNO₃ for 24 h. All solutions, dilutions, and washes were performed with ultrapure water (Millipore; 18 MΩ cm resistivity).

An Analyst 600 graphite furnace atomic absorption spectrometer (GF-AAS; PerkinElmer) was used for Ag determinations in the hemolymph, in the saline water, and in the food used in the dietary exposure experiments. A 7700× inductively coupled plasma mass spectrometer (ICP-MS; Agilent Technologies) was used for Ag determinations in the hemolymph and in the saline water from AgNO₃ exposure via water. Other equipment used included a magnetic shaker (Fisaton, model 753), a pH meter (Thermo Scientific, model Orion Star A211), a conductivity/salinity meter (Thermo Scientific, model Orion Star A212), an oxygen meter (YSI, model 55), an incubator with photoperiod (Marconi, model MA403, and Eletrolab, model EL 202/4), and an analytical scale (Shimadzu, model AUW220D).

Exposure to AgNO₃ via water

Adult organisms were individually exposed via water to nonlethal concentrations (0, 5, 10, 25, 50, and 100 μg L⁻¹ of Ag from AgNO₃) for 14, 24, 48, 72, and 96 h without feeding or aeration. Twelve replicates were used with a 1:1 sex ratio for each treatment. The hemolymph was collected according to Vannuci-Silva et al. (2018), and Ag concentrations were determined using an ICP-MS. Silver concentrations in the exposure solutions were determined at the end of experiments, and a comparison between the final and the nominal concentrations was done. The saline water sample was acidified with HNO₃ 2% for Ag preservation. After 1:10 dilution, the sample was directly introduced into the ICP-MS using a High Matrix Introduction System because it supports a high content of dissolved solids (up to 5%; Vannuci-Silva et al. 2018). The limit of detection (LOD) and the limit of quantification (LOQ) for the hemolymph were 0.13 and 0.44 µg L⁻¹, respectively. For the saline water, the LOD was 1.2 µg L⁻¹ and the LOQ was 4.1 µg L⁻¹, as described by Vannuci-Silva et al. (2018).

Exposure to AgNP and AgCl via food

Feeding exposure experiments were conducted with 2 types of food, one containing AgNP and another with Ag salt (AgCl). A control group was fed with the same basal diet used in the contaminated food (∼40% protein and 6% lipid). The contaminants were incorporated into the basal diet by adding the unmodified powdered form within the feed pellets as described by Merrifield et al. (2013). The AgNPs were obtained from Sigma-Aldrich (nano <100 nm), with a mean particle diameter of 58.6 ±18.6 nm (mean ± standard deviation [SD], n = 64), and were from the same batch as reported in Bradford et al. (2009) and Merrifield et al. (2013). Silver as AgCl (Sigma-Aldrich, with 99% purity) was also added to the basal diet to provide a feeding treatment with an Ag salt form. The metal concentration in both types of food was evaluated by GF-AAS. The determinations indicated that Ag concentrations were 155 and 195 mg kg⁻¹ in the food preparations for AgNP and AgCl, respectively.

Adult animals were individually allocated in plastic containers with 100 mL of reconstituted saline water (30 ± 2 salinity). They were fed with control food or with food containing either AgNP or AgCl pellets. After 1 h of feeding, organisms were rinsed with ultrapure water and placed in new plastic containers and clean artificial saline water to ensure that the exposure was only via food. The times of exposure were 7, 14, and 28 d. The tests were carried out at the same conditions of salinity, temperature, and photoperiod as used in the husbandry (30, 24 ± 2 °C, and 12:12-h light: dark, respectively) but without aeration. Twelve replicates were used for each treatment, with a 1:1 sex ratio; and the experiment was performed twice. In the first experiment, the organisms were fed daily. Because it was observed that not all organisms ate every day, a second experiment was carried out with the same conditions but with food provided on alternate days.

Silver determination in hemolymph was carried out by GF-AAS as described by Vannuci-Silva et al. (2018). The measurements were performed using a wavelength of 328.1 nm, and samples and chemical modifier (Pd 5 µg/Mg[NO₃]₂ 3 µg) volumes injected into the graphite tube were 20 and 5 µL, respectively. Silver concentrations in saline water of the exposure assays were determined in 17 samples at the beginning and at the end of 1 h of feeding, to ensure that no Ag was present in the water. The saline water sample was acidified with HNO₃ 2% for Ag preservation. Samples were directly introduced into the GF-AAS after 1:10 dilution. The LOD and the LOQ for Ag in the hemolymph were 0.11 and 0.37 ng mg⁻¹, respectively. The LOD and LOQ for Ag in saline water were 0.16 and 0.54 μg L⁻¹, respectively (Vannuci-Silva et al. 2018).

Hemolymph collection

Hemolymph was obtained as described by Vannuci-Silva et al. (2018). The fluid was collected using a thin glass needle that was manually made from a capillary glass. Using tweezers, the animals were immobilized and placed with the dorsal segments clearly visible on a decontaminated glass plate. The needle was inserted into the first or second dorsal segment, and the hemolymph was extracted with the capillary needle. The amount of hemolymph collected was placed into a 2-mL Eppendorf tube containing 0.5 or 1 mL of HNO₃ 0.05%. Each capillary glass was weighed before and after the hemolymph collection. We collected 0.87 ± 0.42 mg of hemolymph per animal (n = 574). Three pooled samples of 4 organisms (2 females and 2 males) were tested per treatment.

Uptake rates

Silver concentrations were measured in the hemolymph of animals exposed via either water or diet. The uptake kinetics were modeled as described by Reinardy et al. (2011), with a single-component first-order kinetic model

(1)

where C_t and C_ss represent activity concentration at time t (days) and at steady state, respectively, and k_e represents the biological uptake rate constant (per day). If there was no indication of reaching a steady state during the time of exposure (nonsignificant fit to the model), a simple linear regression model was applied:

(2)

where k_u is the slope of regression.

Expression of results and statistics

The results are reported as the mean ± SD and expressed in milligrams per kilogram for Ag concentration in the food, micrograms per liter for Ag concentration in the water, and nanograms per milligram for Ag concentration in the hemolymph.

Silver concentration data sets were tested for normality (Shapiro–Wilk test) and homogeneity of variance (Levene test). When data were acceptable, analyses of variance were performed to assess the relation between Ag hemolymph concentration and the independent factors: Ag concentration in the water/food and exposure time. A Tukey test was posteriorly applied. The IBM SPSS statistics program (Ver 24) was used for all statistical analyses, and p < 0.05 was considered significant.

Parameters and statistics of uptake models were estimated by iterative adjustments by nonlinear exponential rise to maximum and linear functions in SigmaPlot 12.5 (Systat Software). Where the significance of the model was satisfied (p < 0.05), the model was applied to the data.

Exposure to AgNO₃ via water

The Ag determinations in the saline water, at the end of the experiments, were 72 to 87% of the Ag concentrations added at the beginning of the experiment. Silver concentration in the hemolymph increased with exposure time and Ag concentration in the water (Figure 1); differences were significant between exposure times (p < 0.001), concentrations (p < 0.001), and their interaction (p < 0.001). However, no significant difference between 50 and 100 μg L⁻¹ treatments were found (p = 0.088).

We calculated uptake rates for each Ag water concentration tested. The general equation that fitted all data was f = y⁰ + a × [1 – exp(–b × x)]. The a and b coefficients for each concentration, the slope correlation (R²), and the p value for the regressions are described in Table 1.

The angular coefficient (a) in the equations increased proportionally with the increase of Ag concentration in the water, up to 50 μg L⁻¹. Above 50 μg L⁻¹, the uptake rate reached a plateau, suggesting some mechanism of regulation when Ag concentrations in water reach this amount. The proposed equation can be used to estimate the Ag concentration in the hemolymph of organisms exposed to Ag via water.

Exposure to AgNP and AgCl via food

Data from animals fed daily and those fed on alternate days were combined and analyzed together because they were not statistically different (p = 0.177). Silver was not released from food to the water media during the experiments (concentrations in the water were lower than the LOQ); therefore, the only source of Ag was the diet.

There was no difference in the Ag concentration in the hemolymph among the organisms in the control group at different exposure times (p = 0.963). Silver concentrations in the hemolymph for AgCl food–exposed organisms showed no difference between 7 and 14 d of exposure (p = 0.784), and they were 1.3 ± 0.2 and 1.6 ± 0.3 ng mg⁻¹, respectively. However, for 28 d of exposure, an Ag increase was observed, reaching 3.7 ± 1.0 ng/mg hemolymph (p > 0.001 when compared to 7 and 14 d). Silver concentrations in the hemolymph for AgNP food exposure during 7 and 14 d were 3.3 ± 2.0 and 5.1 ± 0.2 ng mg⁻¹, respectively; and there was no significant difference between these values (p = 0.271), probably because of the high SD observed in the measurements for 7 d–exposure organisms. Nevertheless, at 28 d, Ag concentration in the hemolymph reached 8.4 ± 0.7 ng mg⁻¹, and it was significantly higher than the 7 (p = 0.001) and the 14 (p = 0.027) d exposures.

Silver concentrations in the hemolymph of organisms exposed to AgNPs were higher than the ones exposed to AgCl along time (Figure 2), although the Ag concentration in food containing AgCl was slightly higher than that in the AgNP-containing food, highlighting the higher bioavailability of Ag from AgNPs. No difference was observed between the 2 food treatments on day 7 of exposure (p = 0.07), possibly because of the higher SD observed for AgNP. The hemolymph of organisms fed with AgNP food showed significant higher Ag concentration than of those fed with AgCl on 14 (p = 0.001) and 28 (p = 0.001) d of exposure.

Silver uptake from AgNP food is 2.8 times higher than that from AgCl (Figure 2). More studies are required to verify the mechanisms involved and the risks associated with AgNP and AgCl exposures.