Matthew S. Hull

Fractionating Nanosilver: Importance for Determining Toxicity to Aquatic Test Organisms

This investigation applied novel techniques for characterizing and fractionating nanosilver particles and aggregates and relating these measurements to toxicological endpoints. The acute toxicity of eight nanosilver suspensions of varying primary particle sizes (10-80 nm) and coatings (citrate, polyvinylpyrrolidone, EDTA, proprietary) was assessed using three aquatic test organisms (Daphnia magna, Pimephales promelas, Pseudokirchneriella subcapitata). When 48-h lethal median concentrations (LC50) were expressed as total silver, both D. magna and P. promelas were significantly more sensitive to ionic silver (Ag(+)) as AgNO(3) (mean LC50 = 1.2 and 6.3 μg/L, respectively) relative to a wide range in LC50 values determined for the nanosilver suspensions (2 -126 μg/L). However, when LC50 values for nanosilver suspensions were expressed as fractionated nanosilver (Ag(+) and/or <4 nm particles), determined by ultracentrifugation of particles and confirmed field-flow-fractograms, the LC50 values (0.3-5.6 μg/L) were comparable to the values obtained for ionic Ag(+) as AgNO(3). These results suggest that dissolved Ag(+) plays a critical role in acute toxicity and underscores the importance of characterizing dissolved fractions in nanometal suspensions.

Factors influencing the partitioning and toxicity of nanotubes in the aquatic environment

Carbon nanotubes (NTs) may be among the most useful engineered nanomaterials for structural applications but could be difficult to study in ecotoxicological evaluations using existing tools relative to nanomaterials with a lower aspect ratio. Whereas the hydrophobicity and van der Waals interactions of NTs may suggest aggregation and sedimentation in aquatic systems, consideration regarding how engineered surface modifications influence their environmental fate and toxicology is needed. Surface modifications (e.g., functional groups and coatings) are intended to create conditions to make NTs dispersible in aqueous suspension, as required for some applications. In the present study, column stability and settling experiments indicated that raw, multiwalled NTs (MWNTs) settled more rapidly than carbon black and activated carbon particles, suggesting sediment as the ultimate repository. The presence of functional groups, however, slowed the settling of MWNTs (increasing order of stability: hydroxyl > carboxyl > raw), especially in combination with natural organic matter (NOM). Stabilized MWNTs in high concentrations of NOM provided relevance for water transport and toxicity studies. Aqueous exposures to raw MWNTs decreased Ceriodaphnia dubia viability, but such effects were not observed during exposure to functionalized MWNTs (> 80 mg/L). Sediment exposures of the amphipods Leptocheirus plumulosus and Hyalella azteca to different sizes of sediment-borne carbon particles at high concentration indicated mortality increased as particle size decreased, although raw MWNTs induced lower mortality (median lethal concentration [LC50], 50 to >264 g/kg) than carbon black (LC50, 18-40 g/kg) and activated carbon (LC50, 12-29 g/kg). Our findings stress that it may be inappropriate to classify all NTs into one category in terms of their environmental regulation.

Influence of nanotube preparation in Aquatic Bioassays

Knowledge gaps in nanomaterial fate and toxicity currently limit the ability of risk assessments to characterize the environmental implications of nanomaterials. This problem is further complicated by the lack of standardized characterization and preparation methodologies for researchers to gain the needed information to assist risk assessors. In the present study, data were generated to determine if multiwalled nanotube (MWNT) fate and toxicity are altered by engineered surface modifications or by different dispersal methods. While dissolved organic matter was a good dispersing agent of MWNTs in water, the humic acid fraction was a more effective dispersant than the fulvic acid fraction. When stabilized in organic matter, the functional group attached to the MWNT controlled its toxicity. Underivatized MWNTs induced relatively moderate toxicity to Ceriodaphnia dubia after 96 h (25 +/- 19% survival at 26 mg/L), while hydrophilic groups (hydroxyl, carboxyl) reduced this toxicity (93 +/- 12% survival at 48 mg/L). However, other functional groups (alkyl, amine) increased toxicity (0 +/- 0% survival at <15 mg/L). In dispersal method studies, sonication of MWNTs increased fragmentation relative to magnetic stirring. The sonication treatment of MWNTs also slightly reduced the mortality of C. dubia in the water column but increased toxicity in the sediment to Leptocheirus plumulosus and Hyalella azteca. Findings in the present study indicate that nanotubes engineered for specific applications need to be managed independently and that laboratory methods to disperse and test nanotubes in bioassays need to be standardized to obtain repeatable results for comparison of materials.

Moving beyond Mass: The Unmet Need to Consider Dose Metrics in Environmental Nanotoxicology Studies

knowledge of the particle size, shape, and crystallinity, which were reported in 60%, 48%, and 20% of the manuscripts, respectively. Only 8% of the studies reported that characterization was performed in the experimental media used to measure a toxicological response and none reported the fraction of dissolved ions present. Focused efforts to relate dosimetry to biological effects are required to better understand the ecotoxicological profiles of engineered nanomaterials and to allow risk assessors to more reliably relate exposure to effect. We challenge the research community to better address the dosimetry issue in future toxicological studies with engineered nanomaterials, and to report characterization information sufficient to convert mass

Effect of silver nanoparticles and antibiotics on antibiotic resistance genes in anaerobic digestion.

Water resource recovery facilities have been described as creating breeding ground conditions for the selection, transfer, and dissemination of antibiotic resistance genes (ARGs) among various bacteria. The objective of this study was to determine the effect of direct addition of antibiotic and silver nanoparticles (Ag NPs, or nanosilver) on the occurrence of ARGs in thermophilic anaerobic digesters. Test thermophilic digesters were amended with environmentally-relevant concentrations of Ag NP (0.01, 0.1, and 1.0 mg-Ag/L; corresponding to approximately 0.7, 7.0, and 70 mg-Ag/kg total solids) and sulfamethoxazole (SMX) that span susceptible to resistant classifications (1, 5, and 50 mg/L) as potential selection pressures for ARGs. Tetracycline (tet(O), tet(W)) and sulfonamide (sulI, sulII) ARGs and the integrase enzyme gene (intI1) associated with Class 1 integrons were measured in raw sludge, test thermophilic digesters, a control thermophilic digester, and a control mesophilic digester. There was no apparent effect of Ag NPs on thermophilic anaerobic digester performance. The maximum SMX addition (50 mg/L) resulted in accumulation of volatile fatty acids and low pH, alkalinity, and volatile solids reduction. There was no significant difference between ARG gene copy numbers (absolute or normalized to 16S rRNA genes) in amended thermophilic digesters and the control thermophilic digester. Antibiotic resistance gene copy numbers in digested sludge ranged from 10(3) to 10(6) copies per microL (approximately 8 x10(1) to 8 x 10(4) copies per microg) of sludge as result of a 1-log reduction of ARGs (2-log reduction for intI1). Quantities of the five ARGs in raw sludge ranged from 10(4) to 10(8) copies per microL (approximately 4 x 10(2) to 4 x 10(6) per microg) of sludge. Test and control thermophilic digesters (53 degrees C, 12-day solids retention time [SRT]) consistently reduced but did not eliminate levels of all analyzed genes. The mesophilic digester (37 degrees C, 20-day SRT) also reduced levels of sulI, sulII, and intI1 genes, but levels of tet(O) and tet(W) were the same or higher than in raw sludge. Antibiotic resistance gene reductions remained constant despite the application of selection pressures, which suggests that digester operating conditions are a strong governing factor of the bacterial community composition and thus the prevalence of ARGs.

Uptake and retention of metallic nanoparticles in the Mediterranean mussel (Mytilus galloprovincialis)

We measured the uptake, distribution and elimination of two types of metallic nanoparticles (MetNPs) by the aquatic mussel Mytilus galloprovincialis in static seawater column exposures. Test MetNPs included polyethylene glycol (PEG) functionalized Fe3O4 nanoparticles (PEG-FeOxNP) and PEG-functionalized cadmium-selenide quantum dots (PEG-Qdot). Exposure water was sampled at various times to assess MetNP clearance, and mussels were serially euthanized to assess uptake of MetNPs into the hemolymph, digestive gland, and remaining carcass. Results indicated that >90% of both types of MetNPs were taken up by mussels within 8h of initial exposure. Nearly the entire retained dose of FeOxNPs and PEG-Qdots was deposited in the digestive gland. Our results provide important insights on the uptake and elimination kinetics of MetNPs in filter-feeding marine bivalves, and will be useful for subsequent development of toxicokinetic models to predict the kinetics of these processes.

Gaining a Critical Mass: A Dose Metric Conversion Case Study Using Silver Nanoparticles

Mass concentration is the standard convention to express exposure in ecotoxicology for dissolved substances. However, nanotoxicology has challenged the suitability of the mass concentration dose metric. Alternative metrics often discussed in the literature include particle number, surface area, and ion release (kinetics, equilibrium). It is unlikely that any single metric is universally applicable to all types of nanoparticles. However, determining the optimal metric for a specific type of nanoparticle requires novel studies to generate supportive data and employ methods to compensate for current analytical capability gaps. This investigation generated acute toxicity data for two standard species (Ceriodaphnia dubia, Pimephales promelas) exposed to five sizes (10, 20, 30, 60, 100 nm) of monodispersed citrate- and polyvinylpyrrolidone-coated silver nanoparticles. Particles were sized by various techniques to populate available models for expressing the particle number, surface area, and dissolved fraction. Results indicate that the acute toxicity of the tested silver nanoparticles is best expressed by ion release, and is relatable to total exposed surface area. Particle number was not relatable to the observed acute silver nanoparticle effects.