Potential Impact of Low-Concentration Silver Nanoparticles onPredator−Prey Interactions between Predatory Dragonfly Nymphsand Daphnia magna as a PreyLok R. Pokhrel† and Brajesh Dubey*,‡

†Department of Environmental Health, College of Public Health, East Tennessee State University, Johnson City, Tennessee37614-1700, United States‡Environmental Engineering, School of Engineering, University of Guelph, 50 Stone Road East, Guelph, Ontario, Canada

*S Supporting Information

ABSTRACT: This study investigated the potential impacts of low-concentration citrate-coated silver nanoparticles (citrate−nAg; 2 μg L−1 astotal Ag) on the interactions of Daphnia magna Straus (as a prey) withthe predatory dragonfly (Anax junius: Odonata) nymph using thebehavioral, survival, and reproductive end points. Four different toxicitybioassays were evaluated: (i) horizontal migration; (ii) vertical migration;(iii) 48 h survival; and (iv) 21 day reproduction; using four differenttreatment combinations: (i) Daphnia + citrate−nAg; (ii) Daphnia +predator; (iii) Daphnia + citrate−nAg + predator; and (iv) Daphnia only(control). Daphnia avoided the predators using the horizontal and verticalmovements, indicating that Daphnia might have perceived a significantrisk of predation. However, with citrate−nAg + predator treatment,Daphnia response did not differ from control in the vertical migrationtest, suggesting that Daphnia were unable to detect the presence ofpredator with citrate−nAg treatment and this may have potential implication on daphnids population structure owing topredation risk. The 48 h survival test showed a significant mortality of Daphnia individuals in the presence of predators, with orwithout citrate−nAg, in the test environment. Average reproduction of daphnids increased by 185% with low-concentrationcitrate−nAg treatment alone but was severely compromised in the presence of predators (decreased by 91.3%). Daphniareproduction was slightly enhanced by approximately 128% with citrate−nAg + predator treatment. Potential mechanisms ofthese differential effects of low-concentration citrate−nAg, with or without predators, are discussed. Because silver dissolutionwas minimal, the observed toxicity could not be explained by dissolved Ag alone. These findings offer novel insights into howexposure to low-concentration silver nanoparticles could influence predator−prey interactions in the fresh water systems.

■ INTRODUCTIONWith better understanding of the physicochemical properties ofengineered nanomaterials (ENM) in recent years, it has beenpossible to tune nanomaterials with desired novel function-alities.1,2 Hence, these novel nanomaterials have becomeintegral components of hundreds of consumer productsavailable in the market, today.3 Increasing applications ofnanoproducts could lead to environmental contamination dueto leaching of chemicals, possibly in ionic- and/or nanoform, asthese products age and weather during their life cycle.4,5

Increasing number of both in vitro and in vivo studies suggestthat ENMs could potentially pose risk to the living systems.6−8

Among a wide variety of ENMs, silver nanoparticles (AgNP)have found applications in more than 300 products asdocumented by the Project on Emerging Nanotechnologies(PEN) inventory.3 Historically, Ag (as ions or compounds) isknown for its antimicrobial properties;6,9,10 but with theprogress in nanotechnology, AgNPs incorporated into productsseem to offer the same end result at relatively low

concentrations, thus minimizing11 or potentially replacing theuse of bulk silver in material applications. On the other hand,limited understanding of nanotoxicology has hindered riskassessment of AgNPs in aquatic environments. An absence ofstandard analytical technique and/or protocol to quantifyENMs in the test matrix or in vivo has been the main hurdle tounderstand whether the response(s) displayed by an organismis primarily due to NPs or ions released from NPs or is thecombined effects of both.6,12 Furthermore, as nanotoxicityliterature generally accounts for exceedingly higher concen-trations,13,14 it offers less information about the potential risksthat nanomaterials might have on natural aqueous systems.Acute toxicity studies of ENMs using Daphnia species as test

organisms are abundant,15−17 but literature accounting for

Received: November 13, 2011Revised: June 9, 2012Accepted: June 14, 2012Published: June 14, 2012

Article

pubs.acs.org/est

© 2012 American Chemical Society 7755 dx.doi.org/10.1021/es204055c | Environ. Sci. Technol. 2012, 46, 7755−7762

chronic toxicity on Daphnia is rather scarce.18,19 Moreimportantly, studies investigating long-term exposure ofdaphnids (as a prey) to low-concentration NPs, with specialfocus on potential interactions among the model NPs,predators (dragonfly nymphs), and prey, could enable betterunderstanding of the potential toxicity of NPs to daphnids inthe aquatic environments. Toxicological studies of conventionaltoxicants at sublethal doses suggest that changes in the behaviorand/or physiology of zooplanktons, such as daphnids, couldelicit long-term impacts in the population structure andecosystem dynamics.20,21 More specifically, altered behavioralresponses of daphnids on exposure to a pesticide have beenshown to promote predation risk.21 For example, daphnidsexposed to a carbaryl pesticide exhibited abnormal swimmingpatterns and were preferentially preyed upon by predatorybluegills (Lepomis macrochirus) compared to the unexposeddaphnids that exhibited normal swimming behavior.21 Liter-ature suggests that the behavioral end points, such as horizontaland vertical migrations, can serve as sensitive indicators for low-concentration xenobiotic exposure.20−22

Horizontal migration plays a crucial role by allowingdaphnids to migrate from pelagic (open water) to shallowwaters and thus avoid fish predation by retreating against thevegetation.23,24 On the other hand, vertical migration enablesdaphnids to acquire planktonic food via filter feeding,25 which isessential for their survival and reproduction.26 Recently, a fewstudies that explored behavioral and physiological responses ofdaphnids on exposure to carbon-based NPs (i.e., fullerenes)27,28

or nTiO227 showed potential effects on their survival,

reproduction, and food web dynamics in the aquaticecosystems.27,28 Brausch et al.28 further explored potentialinteractions of the predator + fullerene NPs on the behaviors ofD. magna using water conditioned with predatory bluegill(Lepomis macrochirus) and used as a predatory cue.However, an application of visual predators, such as the live

dragonfly nymphs or fish, in the test environment has not yetbeen realized in the nanotoxicology research. Physical presenceof the predators in the laboratory experiment should provideboth the visual29 and chemical cues to Daphnia,28 which couldincur stress via predation risk as in the natural environ-ment.27,28,30 This critical knowledge gap is addressed here byinvestigating the potential impacts of low-concentration citrate-coated silver nanoparticles (citrate−nAg; 2 μg L−1 as total Ag)on the interactions of D. magna (as a prey) with the predatorydragonfly (Anax junius: Odonata) nymphs. We evaluated thespatial migratory patterns, survival, and reproduction in D.magna using four different toxicity bioassays: (i) test forhorizontal migration; (ii) test for vertical migration; (iii) 48 hsurvival test; and (iv) 21 day reproduction test; andcomplementary treatments using the ensuing combinations:(i) Daphnia + citrate−nAg; (ii) Daphnia + predator; (iii)Daphnia + citrate−nAg + predator; and (iv) Daphnia only(control).

■ MATERIALS AND METHODSNanomaterial Preparation and Dosimetry. Citrate-

coated silver nanoparticles (citrate−nAg) were synthesizedand stabilized in the laboratory following a previously describedprocedure.31 By using the protocol described in our previouswork,6 citrate−nAg was cleaned using a Kros Flo TangentialFlow Filtration (TFF) System equipped with polysulfone (PS)10 kD hollow fiber membranes (P/N: X31S-300-02P, surfacearea = 145 cm2; Spectrum Laboratories, CA). Because lower

electrical conductance (EC) has been associated with lowerimpurity fraction in similar NP suspensions, our citrate−nAgsuspension (as retentate) with EC of 5 μS cm−1 indicated nearabsence of ions and/or impurities in the purified suspension(19 mg L−1 of total Ag) (Supporting Information, TableS1).31,32

A low-concentration of 2 μg citrate−nAg L−1 (as total Ag)was used in this study for all four toxicity bioassays tested. Thisconcentration value was adapted from the 48 h acute toxicitydata collected previously in our laboratory for the sameDaphnia species and the laboratory conditions.33 Additionally, arecent study by Kennedy et al. that reported median lethalconcentration (LC50 for 48 h test) of 0.3 to 2.2 μg Ag L−1 (asdissolved Ag) for D. magna34 and the revised USEPA nationalrecommended water quality criteria of 1.9 and 3.2 μg Ag L−1

for salt and fresh water systems, respectively, also providedbases for this study.35

Nanoparticle Characterization. Purified citrate−nAgsuspension was characterized using dynamic light scattering(DLS), UV−Vis spectrophotometry, and transmission electronmicroscopy (TEM). Average hydrodynamic diameter (HDD,volume weighted), electrophoretic mobility, and zeta (ζ)potential of citrate−nAg samples were estimated with a PSSNicomp 380ZLS particle sizer/zeta potential unit (ParticleSizing Systems, CA, USA) using the DLS method. The unit wascalibrated at 23 °C with Duke 500 (491 nm) NIST 3490Astandard (PSS Nicomp, FL, USA) prior to the measurements.The stability of NPs in moderately hard water was evaluated fora period of 21 days, measuring average HDD, ζ potential, andelectrophoretic mobility. The surface plasmon resonance (SPR)spectra were recorded using a HACH DR5000 UV−Visspectrophotometer (HACH Company, CO, USA). By using aPhilips EM 420 transmission electron microscope, citrate−nAgsuspension was imaged in the bright-field mode at 120 kV tovisualize NP morphology. An aliquot of sample was droppedonto a carbon-coated copper Formvar grid (Ted Pella, cat. no.01813-F) and was air-dried before acquiring TEM images.Using ImageJ 1.44 particle size distribution (PSD) of thesample was estimated from a representative TEM imagery.36

Graphite furnace- or flame-atomic absorption spectroscopy(AAS, Varian 220Z/220FS) was used to quantify total Agconcentrations in the nanosuspension digested using HNO3(trace element grade) following the standard USEPA method3050B.37

Daphnia magna Toxicity Bioassays. A year oldestablished culture of Daphnia magna Straus was used for thisstudy. The stock culture of D. magna was originally procuredfrom Aquatic Biosystems, Inc., Fort Collins, CO. Four differenttoxicity bioassays were conducted: (i) test for horizontalmigration; (ii) test for vertical migration; (iii) 48 h survival test;and (iv) 21 day reproduction test. The aim of these four-pronged tests was to understand the potential impacts of low-concentration (2 μg L−1 as total Ag) citrate−nAg on theinteractions of D. magna (as a prey) with the predatorydragonfly (A. junius) nymphs, with focus on the spatialmigratory patterns, survival, and reproduction in D. magna,but not that of predatory nymphs as they were tied with threadrestricting their locomotion (detail presented in the followingsections; schematics shown in the Supporting Information,Figures S1 and S2). All tests were run in triplicates in 9.4 L(30.7 L × 15.5 W × 21 H cm3) aquaria, each containing 5 L ofmoderately hard reconstituted water (MHRW), 2 × 103 algal(Selenastrum capricornutum) cells mL−1 test media, and 1 mL

Environmental Science & Technology Article

dx.doi.org/10.1021/es204055c | Environ. Sci. Technol. 2012, 46, 7755−77627756

trout food L−1 test media, unless noted otherwise. Thefollowing four different treatment combinations were used forall bioassays: Daphnia + citrate−nAg; Daphnia + predator;Daphnia + citrate−nAg + predator; and Daphnia only(control).Test for Horizontal Migration. A total of 30, 7−8 days old

D. magna were released into the center of each aquarium usinga sterile pipet and were allowed to acclimate for 2 h in the testmedium (with or without nanoparticles and/or with or withoutnymphs) before the test was initiated. Live dragonfly nymphswere collected from the nearby Brush Creek (vicinity of EastTennessee State University, Johnson City, TN, USA) and werecleaned several times with water prior to their placement in theaquaria. A 2 ft long wide-spectrum fluorescent tube was used asa light source. Two dragonfly nymphs (body length 2−3 cm)tied to a thread were introduced into the water at one side of anaquarium in a way such that they could swim to a distance ofnot more than 5 cm from that region of the aquarium. Hence,the area beyond 5 cm from that corner of an aquarium wherethe nymphs were located was defined as an “open area”(Supporting Information, Figure S1). Daphnids present in theopen area were counted at the end of every 30 min for fourconsecutive time periods with an hour interval between thereadings as a response to the presence (or absence) of thepredatory nymphs. Whether this response was altered (if so, towhat extent) in the presence of low-concentration citrate−nAgwas of special interest. This simple microcosm experiment maysimulate interaction between prey and predator in the aqueousenvironment contaminated with AgNPs.Test for Vertical Migration. Test for vertical migration

was conducted as described previously for D. magna by Brauschet al.,28 with ensuing modifications. The experimental setup wasmanually designed, which was composed of a 25 mL plasticpipet glued to one end at 90° onto the center of a Petri dish,while the top-end was cut wide open to facilitate efficientcleaning and transfer of test solution, including two 7−8 daysold D. magna, into the tube. After erasing the milliliter scale thatwas on the outer surface of the pipet with acetone, the 250 mmlong tube was divided into eight 30 mm sections and a 10 mmsection at the top end. This setup was placed inside a 1 Lbeaker, containing MHRW with 10 unexposed daphnids tomimic the natural population.28 The beaker was wrapped with asheet of black plastic from all sides including the bottom(Supporting Information, Figure S2), and the daphnids wereacclimated for an hour in dark in the respective testsuspensions. Soon the highest vertical position of Daphniawas recorded under light for a period of 5 min, with 30 minintervals between each of the four repeated measures. A 2 ftlong wide-spectrum fluorescent tube provided light for this test.One dragonfly nymph that was placed at the topmost part ofthe tube represented a predator and was tied to a thread,restricting its movement within the topmost 3 cm of the tube.Animals were fed during the entire acclimation and test periodsas described earlier for horizontal migration. Tests were run infive replicates.48 h Survival Test. The 48 h acute toxicity test was

conducted following the standard USEPA guidelines,38 withsome modifications as noted. Thirty <24 h old D. magnaneonates were randomly introduced into each test aquarium,consisting of 5 L of MHRW and were acclimated in it for 2 hbefore being exposed to different treatments. Daphnia wereexposed to a low-concentration citrate−nAg (2 μg L−1 as totalAg) with (or without) two dragonfly nymphs, which were

placed in the respective aquaria as described for horizontalmigration (Supporting Information, Figure S1). Animals wereunfed during the test period. A 16:8 h light/dark photoperiodwas maintained using 2 ft long wide-spectrum fluorescent tubes.At the end of 48 h, the total live daphnids in each chamber werecounted.

21 Day Reproduction Test. The 21 day chronic toxicitytest was conducted following the standard USEPA guidelines,38

with modifications as described for the survival test. Twenty<24 h old D. magna neonates were randomly assigned to eachtest aquarium. Following 2 h acclimation in MHRW, daphnidswere exposed to a low-concentration citrate−nAg (2 μg L−1 astotal Ag), with (or without) two dragonfly nymphs, which wereplaced in the respective aquaria (Supporting Information,Figure S1). Daphnids were fed every other day. After everythree days one-half of the test medium was replaced withfreshly prepared medium to maintain an acceptable waterquality per the USEPA guidelines. On the 21st day, total livedaphnids in each test chamber were counted to evaluate theirreproductive potential.

Dissolution of Citrate−nAg. The dissolved fraction ofcitrate−nAg in the test media (MHRW) was analyticallydetermined using GF-AAS as follows. MHRW (50 mL)containing citrate−nAg at concentrations of 2 or 100 μg L−1

(as total silver), with or without food, was maintained under thesame test environment (16:8 light/dark cycle, 20 ± 1 °C) in a50 mL plastic centrifuge tube (Fisher Scientific, cat. no. 06-443-18) for three different time periods (5 h, 48 h, and 21 day) toanalyze the rate of dissolved Ag released from citrate−nAg. Acitrate−nAg concentration of 2 μg L−1 represented an exposureconcentration applied for all the test bioassays, while the higherconcentration of 100 μg L−1 citrate−nAg was used to comparethe rate of Ag released to the one at lower concentration. Eachsample was centrifuged at 4000 rpm (∼3150g) for 30 min at 20°C (Thermo Electron, IEC Centra CL3 series centrifuge), afterwhich 5 mL of supernatant was pipetted. Each supernatantsample was then digested using HNO3 (trace metal grade)following the USEPA method 3050B37 and analyzed using GF-AAS to determine Ag+ released (as total silver) during thecourse of the toxicity tests. Each set of analysis was run induplicates and values averaged.

Analysis of Water Quality Parameters. Water qualityparameters (i.e., pH, temperature, dissolved oxygen (DO),conductivity, hardness, alkalinity, and ammonia−nitrogen(NH3−N2)) that were assessed in the test media after everythree days until the completion of the experiments wereacceptable per the USEPA guidelines for D. magna culture(Supporting Information, Table S2).38 Total hardness, totalalkalinity, and NH3−N2 were measured using standardcolorimetric HACH methods as described by the manufacturer(HACH Company, Loveland, CO). Water conductivity, pH,DO, and temperature were determined using Hanna Instru-ment HI9828 multiparameter meter (Transcat Inc., Rochester,NY).

Statistical Analyses. Data were tested for normality usingthe Kolmogorov−Smirnov (K−S) test, and as data satisfiednormality (p > 0.1 in all cases), no transformation was required.The General Linear Model (GLM) was used to assess thepotential effect of time and its interaction with treatments onthe response of daphnids in horizontal and vertical migrations.One way analysis of variance (ANOVA) followed by Tukey’sposthoc test for multiple comparisons tested for significantdifference between the treatment means and with that of

Environmental Science & Technology Article

dx.doi.org/10.1021/es204055c | Environ. Sci. Technol. 2012, 46, 7755−77627757

control at the p ≤ 0.05 level. Statistical analyses were performedusing SPSS version 18.0.39

■ RESULTS AND DISCUSSIONCharacteristics and Dissolution of Citrate−nAg. The

HDD of purified citrate−nAg was 11.6 (SD = 1.1) nm, and theaverage TEM diameter was 55.9 (SD = 14.6) nm (SupportingInformation, Figure S3 and Table S2). Its SPR peak wasobserved at 425 nm with an absorbance of 2.38 au (SupportingInformation, Figure S4 and Table S3). Data showed that theNPs were stable in the test matrix (MHRW, without food) overa period of 21 days, although slight decrement in electro-phoretic mobility occurred over time (Supporting Information,Table S4).Our dissolution experiment suggests that there was (null to)

negligible amount of dissolved silver in the supernatantanalyzed as the concentrations were below the GF-AASmethod detection limit (MDL = 0.54 μg/L as total Ag;Supporting Information, Table S5). UV−Vis spectra of the 48 hand 21 day supernatants showed very low absorbance,indicative of an absence of AgNPs and perhaps Ag ions inthe supernatants (Supporting Information, Figure S4).40 Anestimation of dissolved Ag concentrations using dissolution rate(%) from pertinent literature of similar (with respect to TEMsize, coating, and pH) AgNPs showed estimated concentrationsto be minimal and well below the MDL of GF-AAS(Supporting Information, Table S5, and references thereof).Hence, minimal dissolution of citrate−nAg under ourexperimental conditions suggests that dissolved silver may notbe the main predictor of the observed toxicity in this study. Thetotal Ag concentrations in the respective test chambers weremaintained at approximately 2 μg L‑1 during the test period of21 days (Supporting Information, Table S6).Impact on Horizontal Migration. Levene’s test showed

that the variances were not significantly different between thetreatments including the control (F = 1.001; p = 0.402).ANOVA showed that the mean count of Daphnia in the “openarea” was significantly different between the groups (Table 1).

As expected, daphnids showed negative taxis to the presence ofpredatory nymphs, with or without citrate−nAg, in the testenvironment (Tukey’s posthoc, p < 0.01 in both cases; Table1). However, with low-concentration citrate−nAg treatmentalone, daphnids’ response was not different to that observedwhen predators were present in the test chamber (p > 0.5).This suggests that either the chemical cue(s) due to citrate−nAg in the test chamber was indistinguishable to that of

predators or this may rather indicate a general response of the

daphnids to the stressors, such as chemicals or predators, in the

short test period (5 h).

Figure 2A illustrates average variability among treatments onDaphnia counts in open area over the 5 h time period. TheGLM showed no significant main effect of time (GLM: F =0.265, p > 0.5), and no significant interaction between time andtreatment on Daphnia migration toward open area (GLM: F =1.405, p > 0.1). This suggests that the response of Daphnia todifferent treatments was not influenced by time in ourexperiment. On average, with low-concentration citrate−nAg+ predator treatment, Daphnia response was highest among thetreatments (Figure 1A). However, Tukey’s posthoc test showedthat this combined effect was not significantly different withpredator only (p > 0.5) or citrate−nAg only treatment (p >0.1). These results are consistent with the literature onpredator−prey interactions.41

Impact on Vertical Migration. Daphnia displayednegative taxis to the presence of live predator and remainedfurthest away from the predator, which was present on thetopmost part of the experimental setup, compared to othertreatments (Figure 1B). Prey individuals descending downwardor residing at the bottom of the experimental setup whenpredatory nymph and light were present on the surface water is

Table 1. ANOVA Followed by Tukey’s Posthoc Test forDaphnia magna Horizontal Migration to Open Area underDifferent Treatments

sum of squares mean square F p

between groups 88.917 29.639 6.688 0.001Tukey’s Posthoc for Multiple Comparisons

comparison betweenmean

differencestderror p

control citrate−nAg −2.0 0.859 0.107nymph −2.833 0.859 0.010citrate−nAg + nymph −3.666 0.859 0.001

citrate−nAg nymph −0.833 0.859 0.767citrate−nAg + nymph −1.666 0.859 0.227

nymph citrate−nAg + nymph −0.833 0.859 0.767

Figure 1. Horizontal (A) and vertical (B) migration patterns inDaphnia magna upon exposure to citrate−nAg (NP; 2 μg L−1 as totalAg), or predatory dragonfly nymphs (NY), or a combination of both(NP + NY). C represents control (without NP or NY). Error barsrepresent ±1 standard error of the means. For (A), ** denotes p =0.01, and *** denotes p = 0.001 compared to control; for (B), *indicates p < 0.05.

Environmental Science & Technology Article

dx.doi.org/10.1021/es204055c | Environ. Sci. Technol. 2012, 46, 7755−77627758

in good agreement with the visual predator avoidancehypothesis.42 ANOVA followed by Tukey’s posthoc testshowed that citrate−nAg treatment did not significantlyinfluence daphnids’ response compared to control (p > 0.5;Supporting Information, Table S7). However, with citrate−nAg+ predator treatment, it appeared that the NPs somehowreduced or perhaps neutralized the chemical/visual stimuli fromthe predator, allowing daphnids to react in the way similar towhen there was no predator in the test environment, andmigrated vertically upward in the water column reaching closestthe predator that was confined to the top (Figure 1B). In oneinstance, a daphnid that almost reached the water surface wasinstantly preyed upon by the predatory nymph. Tukey’sposthoc test revealed a significant difference between predatoronly and citrate−nAg + predator treatments (p = 0.01). Thesedata suggest the possibilities that (i) the sensory system ofdaphnids might have been compromised on exposure tocitrate−nAg, and/or (ii) the nymph was unable to release anychemical cue into the test environment as citrate−nAg mayhave also impacted it, which might have led daphnids tomistaken for the absence of predator when it was actuallypresent.Average variability in vertical migration of Daphnia among

treatments over a time period of approximately 2 h is presentedin Figure 2B. No significant main effect of time (F = 0.272, p >0.5) and no significant interaction between time and treatments(F = 1.425, p > 0.1) on vertical migration in Daphnia wereshown by the GLM. This suggests that the response of Daphniato different treatments was not influenced by time during ourexperimental period of ∼2 h.Impact on 48 h Survival. Results of the 48 h survival

bioassay showed that, in the presence of predatory nymphs,with or without citrate−nAg, Daphnia survival was significantlycompromised (Figure 3; p < 0.05 in both cases). Furthermore,this result also suggests that the combined effect of citrate−nAg+ predators was not different to that when only predators werepresent (p = 1.0). With low-concentration citrate−nAgtreatment alone, average survival of Daphnia decreased by50% in the 48 h test period (Figure 3); however, this was notsignificantly different from control (p > 0.1; SupportingInformation, Table S8).

Impact on 21 Day Reproduction. Average reproductionin Daphnia was significantly reduced by 91.3% in the presenceof predators (without citrate−nAg) compared to control(Figure 4). On the other hand, reproduction was significantlyenhanced, on average, by 185% with low-concentration citrate−nAg treatment alone (Figure 4; Table 2). Whether hormesiswas triggered at this low concentration remained to be tested,which would require multiple doses of citrate−nAg treatments.Reproduction was also higher with citrate−nAg + predatortreatment (Figure 4), which increased by an average ofapproximately 128% of control. This might be due to lowerthreat perceived by Daphnia of the predators as the lattermovement was restricted and/or the potential interference withthe chemical cue released by predators due to citrate−nAg,thereby reducing the predation risk on the prey individuals.Predation risk is known to be energetically costly,43 and giventhat the predation risk was lowered with citrate−nAg, daphnids’investment of energy into foraging might have promoted itsreproduction.44,45 Potential adaptation of Daphnia to the fixedpredators in a 21 day period might also explain in part theobserved increase in reproduction. In nature, this may not likely

Figure 2. Variation in horizontal (A) and vertical (B) migrations in Daphnia magna over time upon exposure to citrate−nAg (2 μg L−1 as total Ag),or predatory dragonfly nymphs (NY), or a combination of both (NPNY). Daphnids count in the open area represented horizontal migrationresponse. Data are presented as mean ±1 standard error of the means. For horizontal migration, each data point represents means of triplicate runs;for vertical migration, each data point represents means of five replicates. NP, 2 μg L−1 citrate−nAg; NPNY, citrate−nAg + predatory dragonflynymphs combined; Control indicates D. magna only (no NPs, no nymphs).

Figure 3. Variation in Daphnia magna survival as shown by 48 hsurvival test upon exposure to citrate−nAg (NP; 2 μg L−1 as total Ag),or predatory dragonfly nymphs (NY), or a combination of both (NP +NY). Y-axis represents average percent survival of the triplicate runs,each run consisting of 30 daphnids. Error bars represent ±1 standarderror of the means. * denotes p < 0.05 compared to control (C).

Environmental Science & Technology Article

dx.doi.org/10.1021/es204055c | Environ. Sci. Technol. 2012, 46, 7755−77627759

happen as predators are not fixed to a place. Given thatDaphnia perceive low predation risk with low-concentrationcitrate−nAg in aquatic systems, they would be more vulnerableto predation as the predators such as dragonfly nymphs wouldbe actively seeking for prey (unlike in our test environmentwhere predators’ locomotion was restricted with a thread), andthis may potentially modify the behavior, survival, andreproduction of D. magna.25,27,28,46

Differential effects were observed with low-concentrationcitrate−nAg treatment on the interactions of D. magna with thepredatory dragonfly nymphs according to the types of testconducted. In the horizontal migration test, the presence ofcitrate−nAg together with the predators showed the greatestnegative taxis as significantly more daphnids avoided the zoneoccupied by the predators and remained in the open area(Figure 1A). However, in the vertical migration test citrate−nAg + predator treatment modified the daphnids’ behavior asthey moved relatively closer to the predator than with othertreatments, including control (Figure 1B). In the 48 h survivaltest, daphnids’ response to citrate−nAg + predator treatmentwas indistinguishable to that when only predators were present;in both cases, daphnids survival was significantly reduced bymore than 75% (Figure 3).

Phototaxis involves vertical migration of animals in responseto the light stimuli,47 and this migration is crucial for Daphniafor foraging or escaping predation from littoral predators suchas the dragonfly nymphs.25,48,49 Laboratory experimentsassessing vertical migration in Daphnia are found to resemblethe patterns observed in the field; hence, shorter verticaldistance experiments generally preferred in the laboratorysettings are sensitive enough to understand the potentialbehavioral changes in zooplanktons in the fresh watersystems.50,51 In the presence of the predatory nymph, Daphniasignificantly lowered their vertical distance (Figure 1B),supporting the predator avoidance tactics.21,42 Our observationof Daphnia showing relatively higher vertical movement towardpredator with low-concentration citrate−nAg treatment sug-gests that these daphnids would be vulnerable to predation(Figure 1B). This is consistent with the result of a previousstudy showing that Daphnia increased their vertical positionwith PCB153 (polychlorinated biphenyl congener) treatmentin the presence of fish kairomone.46 Changes in daphnids’migratory behaviors with citrate−nAg treatment, as our resultsindicate, could potentially affect their survival with associatedchanges in their ability to recognize the presence of predators innatural habitats.21

Because potential interactions of this planktonic prey withmultiple predators in the ecosystem might be complex, singlepredator−prey interaction models as described in this studyand one other28 should provide a basis to further ourunderstanding on nanoecotoxicity. How would low-concen-tration AgNPs affect the predator behavior and how changes (ifany) in the latter behavior would in turn affect the preyresponse to AgNPs remained to be tested. Unlike in this study,different experimental design(s) might be necessary to assessboth the predator and prey responses to low-concentrationAgNPs. As similar studies utilizing other nanomaterials at lowconcentrations should decrease the current uncertainty ofpotential risks to the environment, future studies should alsoaddress how environmental heterogeneity, such as thepresence/absence of vegetation stands (as refuge), would affectthe population and associated community structure when theaquatic systems are contaminated with different kinds of ENMs.Furthermore, better understanding of the fate of nanomaterialsin aqueous systems should enable researchers to studyinteractive effects of nanomaterials on the predator−preybehaviors and life history traits in the field environments.Daphnia species are important grazers of phytoplanktons,

while they are preferentially preyed upon by fish andinvertebrates such as Chaoborus, Bythotrephes, Leptodora orthe Odonata nymphs (e.g., dragonfly, damselfly, etc.).25 Thus,Daphnia is recognized as a keystone genus in the freshwaterecosystems, providing a major link in the energy flow betweenprimary producers and secondary consumers in these foodwebs.52 Besides occupying the trophic position of primaryconsumer,26 daphnids’ significance in maintaining water qualityvia filter feeding organic detritus from the water column hasalso been recognized to be crucial.21 Results of this studyindicate that the behavior, survival, and reproduction in D.magna could be altered upon exposure to low-concentrationcitrate−nAg, one of the most dominant ENMs3, in the presenceof its common natural predator, A. junius nymphs. Notably, ourstudy not only extended and supported the previous findings ofLovern et al.27 and Brausch et al.,28 our experimental designalso took the traditional ecotoxicity studies a step ahead byinvestigating the potential impact of low-concentration citrate−

Figure 4. Variation in Daphnia magna reproduction in a 21 day periodupon exposure to citrate−nAg (NP; 2 μg L−1 as total Ag), or predatorydragonfly nymphs (NY), or a combination of both (NP+NY). Y-axisrepresents average number of daphnids (both surviving adults andnewborn neonates) counted on the 21st day of the experiment. Errorbars represent ±1 standard error of the means. ** indicates significantdifference with control at p < 0.01; *** indicates significant differenceat p ≤ 0.001.

Table 2. ANOVA Followed by Tukey’s PosthocComparisons for Daphnia magna 21 Day Reproduction Testunder Different Treatments

sum of squares mean square F p

between groups 207394.917 69131.639 30.802 <0.0001Tukey’s Posthoc Test for Multiple Comparisons

comparison betweenmean

difference std error p

control citrate−nAg −175.333 38.681 0.008nymph 188.333 38.681 0.005citrate−nAg + nymph −58.666 38.681 0.472

citrate-nAg nymph 363.666 38.681 <0.0001citrate−nAg + nymph 116.666 38.681 0.065

nymph citrate−nAg + nymph −247.000 38.681 0.001

Environmental Science & Technology Article

dx.doi.org/10.1021/es204055c | Environ. Sci. Technol. 2012, 46, 7755−77627760

nAg on the predator−prey interactions between D. magna andA. junius nymphs, with focus on important behavioral and lifehistory traits in D. magna. Given the scenarios of ever-increasing ENM applications and subsequent disposition intothe environment, it is likely that low-concentration AgNPexposures could influence the life history and behavioral traitsin Daphnia, which may subsequently have implications in theecosystem dynamics and functioning.27,28

■ ASSOCIATED CONTENT

*S Supporting InformationExperimental setup for horizontal and vertical migration tests;synthesis procedure for citrate−nAg; characteristics of citrate−nAg; purification using TFF; stability of nanoparticles in a 21day period; Ag dissolution rate; UV−Vis spectra of citrate−nAg, blank, supernatant, and Ag ions; total Ag concentrations inthe test media; ANOVA and Tukey’s posthoc tests for verticalmigration and 48 h tests; and descriptive statistics for all tests.This material is available free of charge via the Internet athttp://pubs.acs.org.

■ AUTHOR INFORMATION

Corresponding Author*Phone: 001-519-824-4120 ext 52506; e-mail: bdubey@uoguelph.ca.

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis research was supported in part by the East TennesseeState University (ETSU) Research Development Council grantno. 82064 and the Research grant no. 83003 from the Office ofResearch and Sponsored Program (ETSU). The authorsgratefully thank Michael Coviello, TEM Analysis ServicesLab, Azle, TX, for his support in TEM characterization of NPs,and Drs. Kurt J. Maier (ETSU), Adrianna Wojtal-Frankiewicz(University of Lodz), and four anonymous reviewers for theirthoughtful comments.

■ REFERENCES(1) Nakayama, Y.; Pauzauskie, P. J.; Radenovic, A.; Onorato, R. M.;Saykally, R. J.; Liphardt, J.; Yang, P. Tunable nanowire nonlinearoptical probe. Nature 2007, 447, 1098−1101.(2) Costanza, J.; El Badawy, A. M.; Tolaymat, T. M. Comment on“120 years of nanosilver history: implications for policy makers”.Environ. Sci. Technol. 2011, 45, 7591−7592.(3) Project on Emerging Nanotechnologies Inventory. http://www.nanotechproject.org/inventories/consumer/.(4) Benn, T. M.; Westerhoff, P. Nanoparticle silver released intowater from commercially available sock fabrics. Environ. Sci. Technol.2008, 42, 4133−4139.(5) Gottschalk, F.; Nowack, B. The release of engineerednanomaterials to the environment. J. Environ. Monit. 2011, 13,1145−1155.(6) Pokhrel, L. R.; Silva, T. U.; Dubey, B.; El Badawy, A. M.;Tolaymat, T. M.; Scheuerman, P. R. Rapid screening of aquatic toxicityof several metal-based nanoparticles using the MetPLATE bioassay.Sci. Total Environ. 2012, 426, 414−422.(7) Yamashita, K.; Yoshioka, Y.; Higashisaka, K.; Mimura, K.;Morishita, Y.; Nozaki, M.; Yoshida, T.; Ogura, T.; Nabeshi, H.;Nagano, K.; Abe, Y.; Kamada, H.; Monobe, Y.; Imazawa, T.; Aoshima,H.; Kawai, Y.; Mayumi, T.; Tsunoda, S.; Itoh, N.; Yoshikawa, T.;Yanagihara, I.; Saito, S.; Tsutsumi, Y. Silica and titanium dioxide

nanoparticles cause pregnancy complications in mice. Nat. Nano-technol. 2011, 6, 321−328.(8) Cho, E. C.; Zhang, Q.; Xia, Y. The effect of sedimentation anddiffusion on cellular uptake of gold nanoparticles. Nat. Nanotechnol.2011, 6, 385−389.(9) Paal, C. Uber colloidales Silber. Ber. Dtsch. Chem. Ges. 1902, 35,2224−2236.(10) Chen, X.; Schluesener, H. J. Nanosilver: a nanoproduct inmedical applications. Toxicol. Lett. 2008, 176, 1−12.(11) Ki, H. Y.; Kim, J. H.; Kwon, S. C.; Jeong, S. H. A study onmultifunctional wool textiles treated with nano-sized silver. J. Mater.Sci. 2007, 42, 8020−8024.(12) Navarro, E.; Piccapietra, F.; Wagner, B.; Marconi, F.; Kaegi, R.;Odzak, N.; Sigg, L.; Behra, R. Toxicity of silver nanoparticles toChlamydomonas reinhardtii. Environ. Sci. Technol. 2008, 42, 8959−64.(13) Aruoja, V.; Dubourguier, H. C.; Kasemets, K.; Kahru, A.Toxicity of nanoparticles of CuO, ZnO and TiO2 to microalgaePseudokirchneriella subcapitata. Sci. Total Environ. 2010, 407, 1461−1468.(14) Werlin, R.; Priester, J. H.; Mielke, R. E.; Kramer, S.; Jackson, S.;Stoimenov, P. K.; Stucky, G. D.; Cherr, G. N.; Orias, E.; Holden, P. A.Biomagnification of cadmium selenide quantum dots in a simpleexperimental microbial food chain. Nat. Nanotechnol. 2011, 6, 65−71.(15) Lee, S. W.; Kim, S. M.; Choi, J. Genotoxicity and ecotoxicityassays using the freshwater crustacean Daphnia magna and the larva ofthe aquatic midge Chironomus riparius to screen the ecological risks ofnanoparticle exposure. Environ. Toxicol. Pharmacol. 2009, 28, 86−91.(16) Kim, J.; Park, Y.; Yoon, T. H.; Yoon, C. S.; Choi, K.Phototoxicity of CdSe/ZnSe quantum dots with surface coatings of 3-mercaptopropionic acid or tri-n-octylphosphine oxide/gum Arabic inDaphnia magna under environmentally relevant UV-B light. Aquat.Toxicol. 2010, 97, 116−124.(17) García, A.; Espinosa, R.; Delgado, L.; Casals, E.; Gonzalez, E.;Puntes, V.; Barata, C.; Font, X.; Sanchez, A. Acute toxicity of ceriumoxide, titanium oxide and iron oxide nanoparticles using standardizedtests. Desalination 2011, 269, 136−141.(18) Zhu, H.; Chang, Y.; Chen, Y. Toxicity and bioaccumulation ofTiO2 nanoparticle aggregates in Daphnia magna. Chemosphere 2010,78, 209−215.(19) Li, M.; Czymmek, K. J.; Huang, C. P. Responses of Ceriodaphniadubia to TiO2 and Al2O3 nanoparticles: a dynamic nano-toxicityassessment of energy budget distribution. J. Hazard. Mater. 2011, 187,502−508.(20) Jones, M.; Folt, C.; Guarda, S. Characterizing individual,population and community effects of sublethal levels of aquatictoxicants - an experimental case-study using Daphnia. Freshwater Biol.1991, 26, 35−44.(21) Dodson, S. I.; Hanazato, T.; Gorski, P. R. Behavioral responsesof Daphnia pulex exposed to carbaryl and Chaoborus kairomone.Limnol. Oceanogr. 1995, 40, 700−709.(22) Yeardley, R. B., Jr.; Lazorchak, J. M.; Gast, L. C. The potential ofan earthworm avoidance test for evaluation of hazardous waste sites.Environ. Toxicol. Chem. 1996, 15, 1532−1537.(23) Lauridsen, T. L.; Buenk, I. Diel changes in the horizontaldistribution of zooplankton in two shallow eutrophic lakes. Arch.Hydrobiol. 1996, 137, 161−176.(24) Moss, B. Shallow lakes biomanipulation and eutrophication. Sci.Comm. Phosphates Eur. Newsl. 1998, 29, 45.(25) Wojtal, A.; Frankiewicz, P.; Andziak, M.; Zalewski, M. Theinfluence of invertebrate predators on Daphnia spatial distribution andsurvival in laboratory experiments: support for Daphnia horizontalmigration in shallow lakes. Int. Rev. Hydrobiol. 2007, 92, 23−32.(26) Gerritsen, J.; Porter, K. G. The role of surface chemistry in filterfeeding by zooplankton. Science 1982, 216, 1225−1227.(27) Lovern, S. B.; Strickler, J. R.; Klaper, R. Behavioral andphysiological changes in Daphnia magna when exposed to nanoparticlesuspensions (titanium dioxide, nano-C60, and C60HxC70Hx). Environ.Sci. Technol. 2007, 41, 4465−4470.

Environmental Science & Technology Article

dx.doi.org/10.1021/es204055c | Environ. Sci. Technol. 2012, 46, 7755−77627761

(28) Brausch, K.; Anderson, T. A.; Smith, P. N.; Maul, J. D. Theeffect of fullerenes and functionalized fullerenes on Daphnia magnaphototaxis and swimming behavior. Environ. Toxicol. Chem. 2011, 30,878−884.(29) Young, S.; Taylor, V. A.; Watts, E. Visual factors in Daphniafeeding. Limnol. Oceanogr. 1984, 29 (6), 1300−1308.(30) Bertolo, A.; Lacroix, F.; Lescher-Moutouea, F.; Sala, S. Effects ofphysical refuges on fish−plankton interactions. Freshwater Biol. 1999,41, 795−808.(31) El Badawy, A. M.; Luxton, T. P.; Silva, R. G.; Scheckel, K. G.;Suidan, M. T.; Tolaymat, T. M. Impact of environmental conditions(pH, ionic strength, and electrolyte type) on the surface charge andaggregation of silver nanoparticles suspensions. Environ. Sci. Technol.2010, 44, 1260−1266.(32) Kanel, S. R.; Al-Abed, S. R. Influence of pH on the transport ofnanoscale zinc oxide in saturated porous media. J. Nanopart. Res. 2011,13, 4035−4047.(33) Silva, T.-U. An evaluation of coating material dependent toxicityof silver nanoparticles. M.S. Thesis, East Tennessee State University,Johnson City, TN, 2011.(34) Kennedy, A. J.; Hull, M. S.; Bednar, A. J.; Goss, J. D.; Gunter, J.C.; Bouldin, J. L.; Vikesland, P. J.; Steevens, J. A. Fractionatingnanosilver: importance for determining toxicity to aquatic testorganisms. Environ. Sci. Technol. 2010, 44, 9571−9577.(35) National recommended water quality criteria; United StatesEnvironmental Protection Agency: Washington, D.C., 2009; http://water.epa.gov/scitech/swguidance/standards/current/upload/nrwqc-2009.pdf.(36) ImageJ Image Processing and Analysis in Java. http://rsb.info.nih.gov/ij/.(37) Acid digestion of sediments, sludges, and soils; Method 3050B;United States Environmental Protection Agency: Washington, D.C.;http://www.epa.gov/wastes/hazard/testmethods/sw846/pdfs/3050b.pdf.(38) Procedures for conducting Daphnia magna toxicity bioassay; EPA/600/8−87/011; United States Environmental Protection Agency:Washington, D.C., 1987.(39) SPSS Base 18.0 User’s Guide; SPSS Inc.: Chicago, IL.(40) Ma, R.; Levard, C.; Marinakos, S.; Cheng, Y.; Liu, J.; Michel, F.M.; Brown, G. E., Jr.; Lowry, G. V. Size-controlled dissolution oforganic-coated silver nanoparticles. Environ. Sci. Technol. 2012, 46 (2),752−759.(41) Relyea, R. A. How prey respond to combined predators: areview and an empirical test. Ecology 2003, 84, 1827−1839.(42) Zaret, T. M.; Suffern, J. S. Vertical migration in zooplankton as apredator avoidance mechanisms. Limnol. Oceanogr. 1976, 21, 804−813.(43) Alton, L. A.; White, C. R.; Wilson, R. S.; Franklin, C. E. Theenergetic cost of exposure to UV radiation for tadpoles is greater whenthey live with predators. Funct. Ecol. 2012, 26, 94−103.(44) Lynch, M. The life history consequences of resource depressionin Daphnia pulex. Ecology 1989, 70, 246−256.(45) Richman, S. The transformation of energy by Daphnia pulex.Ecol. Monogr. 1958, 28, 274−291.(46) Bernatowicz, P.; Pijanowska, J. Daphnia response to biotic stressis modified by PCBs. Ecotoxicol. Environ. Safety 2011, 74, 711−718.(47) Goodrich, M. S.; Lech, J. J. A behavioral screening assay forDaphnia magna: a method to assess the effects of xenobiotics onspecial orientation. Environ. Toxicol. Chem. 1990, 9, 21−30.(48) Ringelberg, J. The positively phototactic reaction of Daphniamagna Straus: A contribution to the understanding of diurnalmigration. Neth. J. Sea. Res. 1964, 2, 319−406.(49) De Meester, L. An analysis of the phototactic behavior ofDaphnia magna clones and their sexual descendants. Hydrobiology1991, 225, 217−227.(50) Harris, J. E. Physical factors involved in the vertical migration inzooplankton. Nature 1953, 320, 746−748.(51) Young, S.; Watt, P. Behavioral mechanisms controlling verticalmigration in Daphnia. Limnol. Oceanogr. 1993, 38, 70−79.

(52) Persson, J.; Brett, M. T.; Verde, T.; Ravet, J. L. Food quantityand quality regulation of trophic transfer between primary producersand a keystone grazer (Daphnia) in pelagic freshwater food webs.Oikos 2010, 116, 1152−1168.

Environmental Science & Technology Article

dx.doi.org/10.1021/es204055c | Environ. Sci. Technol. 2012, 46, 7755−77627762