The purpose of this study was to assess the biological response of mice to inhaled nanosilver and to determine the amount of silver retained in lungs and other organs. We observed nanosilver agglomerates in the inhalation chamber with a GM mobility diameter of 79 nm (Table 1) and SEM-EDS analysis showed the presence of Ag aerosols in the exposure chamber (Figure 4). ICP-OES analysis of nanosilver incubated in the simulated biological fluids revealed negligible formation of Ag ions (Figure 3). Moreover, most of the nanoparticles, even PVP coated particles, aggregated and settled out in these simulated biological fluids at the relatively high concentrations investigated after 24 hours and the XRD analysis of the precipitates revealed the presence of metallic Ag similar to Figure 2A.
Since there are limited data available on the concentrations of nanosilver in consumer products or in the air in occupational settings where materials with nanosilver are being produced, we established exposure concentrations in the whole-body chamber that were similar to our previous studies of toxicity assessment of titanium dioxide, copper, and iron. This facilitated comparison of toxicologic responses with our previous studies with different nanomaterials [34, 41, 42]. The median amount of silver measured in lungs of nanosilver-exposed mice was 31 and 10 μg/g lung (d.w.) for 0 wk and 3 wks animals, respectively, which corresponds to 100 μg/L and 32 μg/L Ag concentrations measured by ICP-OES in the lungs. Moreover, low Ag concentrations in the particle-free supernatants of BAL fluid (13.9 ± 0.9 μg/L and 1.7 ± 0.2 μg/L in animals necropsied at 0 wk and 3 wks post exposure, respectively) were detected by ICP-MS. Higher concentration of Ag measured in BAL fluid from animals necropsied at 0 wk agrees with the higher amount of Ag measured in the 0 wk lungs. Thus, these data suggest that there is some evidence for Ag dissolution in the lungs. It has been shown that Ag nanoparticles might fractionate (or dissolve) in aquatic environments at very low (< 100 μg/L) concentrations since nanoparticles might be less aggregated than in higher concentrated samples [43, 44].
The amount of Ag detected in the 0 wk group corresponds to the lung burden accumulated by a 70 kg person exposed to a very high concentration of 1.0 mg/m3 for 16.6 hours. Hence, this represents a substantial dose. To calculate the exposure time of the human, we assumed exposure concentration of 1.0 mg/m3, frequency of breathing 15 breaths/min, tidal volume of the lungs 600 mL/breath, pulmonary deposition fraction for 80 nm particles of 0.4 (based on human model by Cassee et al. ) and delivered Ag dose to human 3574 μg. The amount of nanosilver measured from the pulmonary region of the exposed mice was below the estimated levels based on the minute ventilation, deposition fraction and gravimetric measurements of chamber outflow. This could be due to lower breathing zone concentrations for the mice than expected from the chamber outflow measurements, an overestimation of the deposition coefficient for the pulmonary region, clearance by mucociliary escalator to the trachea or gastrointestinal tract, translocation of nanosilver to other organs that we did not measure, or poor sensitivity of the ICP-OES assay methods. Work by Asgharian and colleagues has shown that particles below 10 nm are deposited in the nasopharynx and tracheal regions with little penetration into the pulmonary region . Deposition occurs primarily through axial transport by diffusion and convective mixing. Model predictions show that for nanoparticles in the 80 nm range, the deposition fraction for the pulmonary region is about 0.4 in humans (for mouth breathing) and 0.15 in rodents which are obligate nose breathers .
We observed increased total cell numbers in Ag-exposed animals (mainly macrophages) and a slight inflammatory response (increased number of neutrophils in BAL fluid) in 0 wk mice (Figures 5). Recruitment of macrophages is a normal clearance mechanism following particle exposure [45, 46]. Neutrophilia decreased after 3 weeks without exposure, however total number of cells remained elevated. Interestingly, comparison of these results to other metal nanoparticle exposures studied by our group shows that nanosilver is most similar to iron nanoparticles, which are coated with an iron oxide layer. In the previous study, iron nanoparticles were found to be significantly less toxic than copper nanoparticles . Copper nanoparticles induced a 25 times increase in the number of BAL cells after sub-acute exposure as well as an increase in most cytokines.
The levels of total protein and LDH activity in BAL fluid did not change significantly with exposure to nanosilver (Figure 5B). This is another indication of low nanosilver toxicity at the exposure concentration used in this study. Of 7 cytokines tested only two, IL-12(p40) and KC, had slightly elevated concentrations at 0 wk and most were below detection even using a very sensitive assay (Table 2). IL-12(p40), which exhibited the most elevated concentration, functions primarily to induce naïve T-cells to differentiate to Th0 cells and to enhance NK cell activity. It is produced by dendritic cells and macrophages . This is consistent with the observed doubling in BAL macrophages but is very low in comparison to our previous study of Cu nanoparticles when we detected a mean level of IL-12(p40) of 1210 pg/ml (Table 2). Keratinocyte-derived cytokine (KC) that is involved in chemotaxis and cell activiation of neutrophils, was also slightly elevated. However, this increase was 50 times lower than in Cu-exposed mice. Furthermore, other chemokines/cytokines such as MCP-1, MIP-1α, TNF-α, RANTES, GM-CSF were much more elevated in case of Cu nanoparticle exposure .
We observed agglomerated nanosilver in macrophages that were recovered from BAL fluid and also in lung tissue (Figure 6). Dark-field microscopy revealed lower particle load at 3 wks post exposure than immediately post exposure, showing some particle clearance or possible translocation from the airways to the interstitium. Lung tissues were without remarkable pathological changes. In a 90 day rat study, lung inflammation and an increase in granulomatous lesions as well as alterations in lung functions were observed . In another inhalation study in rats, minimal, toxicologically insignificant effects of nanosilver on the nasal respiratory mucosa were indicated .
Here we used the same experimental mouse model and exposure system as in our previous studies on the toxicity of different nanomaterials [34, 41, 42]. Of all nanomaterials studied in the size range from ca. 5 to 20 nm (TiO2, Fe, Cu, Ag), copper nanoparticles, which had a copper oxide coating, induced much higher total cell and neutrophil recruitment in BAL fluid, elevated total protein, activity of LDH and levels of inflammatory cytokines in BAL fluid than nanosilver . This increased inflammatory response in copper-exposed mice was associated with the nanoparticle size and increased ion concentration produced from the dissolving nanoparticles in vivo. Mice exposed sub-acutely to 2-5 nm TiO2 nanoparticles showed a moderate inflammatory response at 0, 1, and 2 weeks post-exposure that resolved after 3 weeks post-exposure. Iron nanoparticles induced a minimal inflammatory response similar to nanosilver. The toxicity of Ag nanoparticles in this murine inhalation model was similar to Fe nanoparticles. Both types do not dissolve in artificial interstitial fluid which could be associated with the negligible toxic response observed for metal and metal oxide nanomaterials. Other studies also support a role for solubility of nanomaterials in their toxicity [51, 52].