Toxicity and uptake of TRI- and dibutyltin in Daphnia magna in the absence and presence of nano-charcoal

TOXICITY AND UPTAKE OF TRI- AND DIBUTYLTIN IN DAPHNIA MAGNAIN THE ABSENCE AND PRESENCE OF NANO-CHARCOAL

LIPING FANG,* OLE K. BORGGAARD, PETER E. HOLM, HANS CHRISTIAN BRUUN HANSEN, and NINA CEDERGREEN

Department of Basic Sciences and Environment, University of Copenhagen, Copenhagen, Denmark

(Submitted 24 March 2011; Returned for Revision 17 May 2011; Accepted 26 July 2011)

Abstract—Butyltins (BTs), such as tributyltin (TBT) and dibutyltin (DBT), are toxic to aquatic organisms, but the presence of the strongadsorbent, black carbon (BC), can markedly influence BT toxicity and uptake in organisms. In the present study, the acute toxicity anduptake of TBT and DBT in the crustacean, Daphnia magna, were investigated with and without addition of nano-charcoal at differentpHs and water hardnesses. The results showed that the toxicity of TBT and DBT increased by lowering the pH from 8 to 6. This reflects arelatively higher toxicity of cationic BT species than of the neutral species. At pH 6, by enhancing the water hardness of the media from0.6 to 2.5mM, the toxicity of TBT and DBT consistently decreased due to competitive binding of bivalent cations (Mg2þ, Ca2þ) to bioticligands of D. magna. Furthermore, the toxicity of TBT to D. magna significantly decreased in the presence of nano-charcoal comparedwith experiments without nano-charcoal at pH 6 and 8, while no significant decrease in toxicity of DBT was observed in the presence ofnano-charcoal. This can be attributed to the insignificant decrease of free DBT concentration in the presence of nano-charcoal comparedwith that for TBT. Conversely, it was observed that more TBT and DBT were taken up in D. magna in the presence of nano-charcoaldue to the uptake of TBT or DBT associated with nano-charcoal by Daphnia in gut systems, as seen by light microscopy. Thisindicated that only free nonadsorbed BTs were toxic to D. magna, at least during short periods of exposure. Environ. Toxicol. Chem.2011;30:2553–2561. # 2011 SETAC

Keywords—Black carbon Organotin Nanoparticles Bioavailability Adsorption

INTRODUCTION

Tributyltin (TBT) and dibutyltin (DBT) are two importantbutyltin compounds (BTs) which have been widely applied inindustries and agriculture since the 1970 s. Some uses of TBThave been for wood preservatives, antifouling paints in ship-yards, and biocides, and DBT has been used as a stabilizeradditive in polyvinyl chloride (PVC) [1,2]. Tributyltin is foundto have high chronic and acute toxicity to aquatic organisms atlevels of nanogram tin (Sn) per liter, which causes gastropodimposex, mussel larvae mortality, and oyster malformation[3–5]. Dibutyltin shows less toxicity to aquatic organisms incomparison with TBT; for example, the 24-h half-maximaleffective concentration (EC50) value of 900mg/L for DBT toDaphnia magna is approximately 70 times higher than that of13mg/L for TBT [6]. However, DBT exhibits much strongerimmunologic toxicity to mammals than TBT [7]. Tributyltin hasbeen phased out from use in ship paints in the 2000 s, as seriousadverse effects on the environment were observed [8].However, TBT still remains in the environment with a half-life up to decades [9], although TBT can be debutylated to DBT,and further to monobutyltin [2]. Dibutyltin continues to beintroduced into sediments and water by leaching from PVCmaterials and degradation of TBT [2].

A number of investigations on the toxicity of TBT todifferent organisms, such as amphipod crustaceans [10], fishlarvae [11], and fungi [12] have been performed. Most ofthese studies have been conducted in pure water without theaddition of sediments and organic matter naturally present in theenvironment [6,11,12]. Both TBT and DBT sorb strongly to

different natural adsorbents such as minerals, organic matter(OM), and black carbon (BC), with partitioning coefficients(Kd) up to 105 L/kg [13–15]. Studies have shown that theadsorption can reduce the bioavailability and toxicity of pollu-tants to aquatic organisms [16–18], whereas the opposite effecthas also been observed in the presence of adsorbents such asnano-C60 and OM [17,19]. These findings suggest that the effectof adsorbents to toxicity of organic compounds not onlydepends on the properties of the chemicals, but also on thecharacteristics of the adsorbents such as particle size and sur-face functionalities [17,19]. However, very little has been doneto study the effect of adsorbents on TBT and DBT toxicity, apartfrom some toxicity tests showing decreased toxicity of TBTtowardD. magna after addition of humic acids [20,21]. No workhas been done to investigate the effect of BCs on the toxicity ofTBT and DBT to organisms.

Recent studies on toxicity of industrial carbon nanoparticlesto various organisms have been conducted, and both toxic andnontoxic effects have been reported [17,22,23]. Although verycommon in the environment, the toxicity of BCs such as nano-BC, which may be taken up by and become harmful toorganisms, still remains unknown. In addition, nano-BC mayincrease the uptake of toxic TBT and DBT, because adsorptionby small BC particles may facilitate the uptake by the organ-isms. Moreover, studies have demonstrated that the pH andwater hardness (Ca2þ andMg2þ) can influence the adsorption ofTBT and DBT by BC [2,13,15], which may also affect theirbioaccumulation and toxicity.

In an attempt to remedy this lack of knowledge, the classicorganism model D. magna, which is a good representative offilter-feeding zooplanktons, was used for the toxicity and uptaketests in the present study, including toxicity test of nano-charcoal; toxicity tests of TBT and DBT with/without charcoalat pH 6 and pH 8; toxicity tests of TBT and DBT in the absenceand presence of charcoal in media with three different water

Environmental Toxicology and Chemistry, Vol. 30, No. 11, pp. 2553–2561, 2011# 2011 SETAC

Printed in the USADOI: 10.1002/etc.649

All Supplemental Data may be found in the online version of this article.* To whom correspondence may be addressed

([email protected]).Published online 19 August 2011 in Wiley Online Library

(wileyonlinelibrary.com).

2553

hardnesses; and uptake of TBT and DBT in D. magna with/without charcoal at pH 6 and pH 8. The results were assessed interms of adsorption isotherms and EC50 values.

MATERIALS AND METHODS

Reagents

Tributyltin and dibutyltin were purchased as liquid tribu-tyltin chloride ([C4H9]3SnCl; density 1.2 g/ml; 96% purity), andsolid dibutyltin dichloride ([C4H9]2SnCl2; 96% purity) fromSigma Aldrich. Tripropyltin chloride (TPrT; [C3H7]3SnCl;density 1.27 g/ml; 98%), used as an internal standard, waspurchased from Merck. The ethylating agent, sodium tetrae-thylborate (NaBEt4; 97%), was purchased from Sigma Aldrich.The buffer, 3-[N-morpholino]propanesulphonic acid (MOPS),was purchased from Sigma Aldrich. Concentrated HNO3 (69%)were obtained from J.T. Baker, and HPLC-grade methanol andhexane were purchased from Sigma Aldrich. All chemicalswere pro analysis or better quality, and Milli-Q water (Milli-pore) (<10mS/m) was used for preparing aqueous workingsolutions throughout the adsorption experiments.

Stock solutions containing 1,000mg/L of TBT or 2,065mg/Lof DBT were prepared by dissolving 250ml TBT or 554mgDBT in 100ml methanol, respectively. The stock solutions ofthe internal standard (IS) TPrT (1,000mg/L) was prepared bydissolving 200ml TPrT in 100mlmethanol. A 2% (w/v) NaBEt4solution was prepared by dissolving 1 g NaBEt4 in 50mlmethanol under N2 flushing to avoid degradation of NaBEt4.All stock solutions were stored in the dark at �208C. Theconcentrations of TBT and DBT are given as mg Sn/L through-out the article.

Test media

The International Organization for Standardization (ISO)medium was prepared according to ISO guideline 6341 [24].Then the ISOmedium was buffered by adding 750mgMOPS in1 L ISO media to obtain good pH buffering in the range of pH 6to 8 [25]. Preliminary tests showed no toxicity of MOPS(750mg/L) to D. magna after exposure for 48 h, in agreementwith the finding of no-observed-effect concentration of 750mg/Lby Schamphelaere et al. [25]. In addition, no significant differ-ence was observed between the toxicity of TBT or DBT to D.magna with/without MOPS buffer in ISO media (p< 0.05)(data not shown). Also found not to complex with chargedorganotins is 3-[N-morpholino]propanesulphonic acid [26].The pH of MOPS-buffered ISO media was finally adjustedto pH 6 or 8 using 1M of NaOH or HCl. The modified ISOmedia with three different water hardnesses (equivalent asCaCO3) were prepared by adding different amounts of CaCl2and MgSO4 to obtain water hardnesses of 0.6, 2.5, and 5mM.The chemical characteristics of the six test media (three waterhardnesses and two pH values) are listed in Table 1.

Charcoal suspensions and characterization

Wheat charcoal was produced as described by Fang et al.[15]. Due to a high content of hydrophilic functional groups onthe surfaces of the particles (1.5mmol/g; [15]), stock solutionsof charcoal suspensions were prepared directly by adding2,000mg of charcoal into 1 L of ISO media. Then suspensionswere ultrasonicated using a Branson Sonifier 150 equipped witha microtip (Branson Ultrasonics) for 10min at 20 watts todisperse the charcoal particles, and then standing for one weekbefore further use. The validation of the concentration ofcharcoal stock solutions was performed by evaporating

100ml of the prepared BC suspensions to dryness at 1008Cand measuring the residual solid weight [15], and less than 5%loss was found compared with that added initially. The workingsuspensions of BC suspension for toxicity and uptake tests wereprepared by diluting the homogenous stock suspension afterhand shaking [15]. The working charcoal suspensions wereallowed to stand in the dark for one week. Finally, the super-natant was decanted and used for the tests. The concentration ofBC in the supernatant of the ISO media was measured using anultraviolet/visible 1800 spectrophotometer (Shimadzu) at awavelength of 336 nm. Linear standard curve was obtainedby measuring the absorbance of the concentration of 0 to50mg/L BC standards prepared from the BC stock suspensionin ISO media. Furthermore, no significant precipitation inworking suspension was observed due to aggregation after a48-h period. The maximum difference of charcoal concentra-tion before and after toxicity tests in Daphnia-free samples wasless than 10%. The nano-charcoal was examined by bothscanning and transmission electron microscopy (SEM andTEM). For analysis on a Quanta 200 SEM (FEI), samples werecollected as solid particles by drying 10ml charcoal suspensionin an oven at 1008C overnight. Then the particles were preparedon 12-mm carbon adhesive tabs and sputter-coated with 5 nm ofgold using a Polaron SC 7640 (Quorum Tech). Alongside, adroplet of suspension was added on a palladium grid, andobserved using a Philips CM100 transmission electron micro-scope.

Adsorption experiment of TBT and DBT

Adsorption isotherms for TBT and DBT bonding to charcoalat pH 6 and pH 8 were determined by adding known amounts ofBT stock solutions to 25ml (TBT) or 7ml (DBT) of 16mg/Lcharcoal suspensions with a hardness of 2.5mM in amber glasstubes (30ml or 10ml capacity), respectively. To cover therelevant BT concentrations of the acute toxicity tests, the initialconcentrations for TBT and DBT were made to range from 0 to50mg/L, and from 0 to 1,500mg/L, respectively. The suspen-sions were equilibrated by horizontal shaking (50 strokes perminute) for 24 h in the dark at room temperature (22� 18C)[15]. In addition, the adsorption of the highest initial concen-trations of 50mg/L for TBT, and 1,500mg/L for DBT, wereexamined at the hardnesses of 0.6 and 5.0mM, respectively.The final concentrations of TBT and DBT, in the aqueous phaseafter adsorption experiments, were analyzed using gas chro-matography mass spectrometry (GC-MS) after derivatizationand extraction into organic solvent, as described below.

BT pretreatments and analyses

For water samples, after 24 h of equilibration, 10ml of TBTspiked or 5ml of DBT spiked to charcoal suspensions werefiltered through a mixed cellulose ester syringe filter (MCE;

Table 1. Characteristics of the synthetic test media used in toxicity testsa

Testmedium

Major ion (mM)

Hardness(mM as CaCO3)pH Ca2þ Mg2þ Naþ Kþ

ISO-1 6 0.5 0.1 0.8 0.1 0.6ISO-1 8 0.5 0.1 1.7 0.1ISO-2 6 2 0.5 0.8 0.1 2.5ISO-2 8 2 0.5 1.7 0.1ISO-3 6 4 1 0.8 0.1 5ISO-3 8 4 1 1.7 0.1

a ISO¼ International Organization for Standardization.

2554 Environ. Toxicol. Chem. 30, 2011 L. Fang et al.

0.22-mm pore size, 25-mm diameter, acrylic housing, AdvantecMFC) into a 5- or 10-ml volumetric flask containing 100-mlTrPT internal standard. To avoid adsorption of TBT or DBT tothe glass walls of the flask, 30ml of HNO3 (69%) was added intoeach volumetric flask. The Daphnia-containing samples werehomogenized using a Bransonic 1510 ultrasonicator (BransonUltrasonics) for 2min after adding 0.5ml 20% CH3COONa/CH3COOH buffer (pH 5) and 100ml TrPT internal standard.The process of sonication showed negligible effect on thestability of the BT compounds, because no change in signalfor BT was observed in BT stock solution on GC-MS, using thescanning mode before and after ultrasonication. The derivati-zation and extraction of TBT and DBT were performed simul-taneously in 20-ml clear glass centrifuge tubes with a glass lidfor water samples, or directly in 2-ml microcentrifuge tubes forDaphnia samples. Subsequently, the pH was adjusted to 5 with2ml 20% (w/v) acetic acid/sodium acetate buffer for watersamples, while no adjustment was needed for the Daphniasamples. The ethylation and extraction were carried out simul-taneously by adding 500ml (water samples) or 1ml (Daphniasamples) of 2% (w/v) NaBEt4 stock solution and 1ml of hexaneinto the microcentrifuge tubes. The mixture was vigorouslyshaken for 1min using a vortex shaker. After phase separation,the hexane phase was transferred to a 5ml clear glass tube. Thisprocedure was repeated three times and the hexane phasescombined. Finally, the hexane extract was transferred to asample vial after being concentrated under nitrogen flow toapproximately 0.5ml. This solution was injected into theGC-MS.

Analyses of TBT and DBT were performed on an Agilent7890 series gas chromatograph (GC) coupled with a 5975BMSD mass spectrometer (MS) (Agilent Technologies). The GCsystem was equipped with a Phenomenex ZB-5 capillary col-umn (60m� 0.25mm i.d., 0.25mm film thickness). Injectionwas performed in splitless mode using a 7683B Agilent auto-mated sample injector. Helium was employed as the carrier gaswith a constant flow of 1.1ml/min. The oven temperature wasprogrammed as follows: 508C for 2min, from 50 to 2508C at158C/min, and held at 2508C for 5min. The injector temper-ature was held at 2508C. The electron energy and temperature ofthe electron ionization (EI) ion source were 70 eV and 2308C,and the temperature of the transfer line was 2508C. The quadru-pole was kept at 1508C. Selected ion monitoring (SIM) modewas chosen for BT analyses. The target ions (m/z) for TBT,DBT, and MBT were 291, 263, and 235, respectively.

Losses of BTs to glass walls and filters during adsorptionexperiments and filtration were estimated by performing experi-ment without adding nano-charcoal. Approximately 5 and 10%of BT losses to glass walls and syringe filters were observed.Losses of BTs during the acute toxicity tests were found to be 8to 10%, which was estimated by measuring BT concentration intest media both with and without adding D. magna after 48 h.The concentrations of TBT and DBT were corrected to includethe losses. In preliminary tests, by extracting the total BT inaqueous solution and Daphnia samples simultaneously afterexposing Daphnia to BT in the presence of charcoal for 48 h,above 97% recovery for BT indicated that the effect of charcoalon the recovery of TBT or DBT was negligible.

D. magna immobilization tests

Daphnia magna were cultured in Elendt’s M7 media at20� 18C with a 16:8 h light:dark photoperiod [27]. Daphniamagna were fed daily with a culture of green algae (Pseudo-kirchneriella subcapitata). Acute 48-h immobilization tests for

TBT or DBT with/without charcoal in the six types of media(Table 1) at 20� 18C were performed using juvenile D. magna(<24-h old) at growth conditions according to Organisation forEconomic Co-operation and Development (OECD) Test Guide-line 202 [28], while acute tests for charcoal were only per-formed in ISO-2 media at pH 8. TBT or DBT test media wereprepared by adding known amounts of stock solutions into100ml ISO media to achieve a nominal concentration rangingfrom 1 to 12mg/L for TBT, and from 60 to 1,500mg/L for DBT.For charcoal-associated tests, TBT or DBT was added tocharcoal ISO suspensions and shaken for 24 h until equilibrium,prior to the acute tests. Because there was no significant acutetoxicity of charcoal to D. magna observed at charcoal concen-trations up to 50mg/L, the concentration of charcoal was kept at12 to 16mg/L in the toxicity test of TBT and DBT. The toxicityof TBT and DBT with addition of nano-charcoal was performedwith three different water hardnesses (ISO-1, ISO-2, and ISO-3)at pH 6, and only in one medium (ISO-2) at pH 8. At least five tosix test concentrations in four replicates, and a control with sixreplicates, were included in each test. In addition to samplecontrol, six replicates of solvent control were also performed foreach test. Each replicate comprised five juveniles in 25-ml testsolution in a 100-ml glass beaker. The number of immobilejuveniles in each beaker was recorded after 24 h and 48 h,respectively. Each toxicity test was repeated two times. Daph-nia magna exposed to charcoal were photographed using aSMZ800 digital light microscope (Nikon) to assess the locationof the charcoal on/in the animals.

Uptake studies with D. magna

Quantification of bioaccumulation of TBT or DBT with/without charcoal was performed in ISO-2 media during anuptake period of 48 h at pH 6 and pH 8. On the basis ofEC50 values observed from acute toxicity tests in the presentstudy, a nominal added concentration of 3.6mg/L and 310mg/Lwas used for TBT and DBT in uptake experiments, respectively.Seven beakers of 200ml TBT- or DBT-spiked ISO media with/without charcoal were prepared prior to the addition of 20 to 30D. magna that were from 6 to 8 days old. The D. magna werekept under growth conditions without feeding for up to 48 h.One beaker was successively removed after 0, 1.5, 3, 5, 7, 24,and 48 h, and all living D. magna in each beaker were trans-ferred to another clean beaker. After rinsing three times withclean ISO media, the Daphnia were dried on filter paper.Finally, all Daphnia from each beaker were weighed and storedin a 2-ml polypropylene microcentrifuge tube (VWR Interna-tional) at �208C. The uptake experiment was performed intriplicate. The concentration of TBT or DBT in the Daphniawas analyzed using GC-MS as described above.

Modeling and statistical analysis

For TBT and DBT adsorption isotherms, adsorption data ofTBT and DBT were fitted by a linear adsorption isothermequation (Eqn. 1) and by the Langmuir equation (Eqn. 2),respectively.

Cs ¼ Kd � Cw (1)

Cs ¼ Cs;maxKLCw

1þ KLCw

(2)

where Cs (mg/kg) and Cw (mg/L) denote the concentrations ofTBT or DBT on solid phase nano-charcoal and in the aqueoussolution, respectively; Kd (L/kg) denotes the partitioningcoefficient of TBT between solid phase (i.e., nano-charcoal)

Nano-charcoal lowers toxicity of TBT and DBT to D. magna Environ. Toxicol. Chem. 30, 2011 2555

and aqueous solution; Cs, max (mg/L) and KL (L/kg) denotes themaximum adsorption capacity by charcoal and the Langmuiradsorption coefficient, respectively.

The fraction of immobile D. magna as a function ofcorrected TBT or DBT solution concentration was describedby using a sigmoidal dose–response model (Eqn. 3)

Y ¼ 1

ð1þ X=EC50Þb(3)

where Y denotes the response (proportion of immobility); EC50is the effect concentration leading to immobility of 50% of testorganisms, X is the concentration of TBT or DBT, and b denotesthe Hill slope.

Statistical analysis was performed using the R program [29].The software package drc (www.R-project.org) was used for theanalysis of dose–response data [30].

RESULTS AND DISCUSSION

Images of nano-charcoals

Particle sizes, morphology, and aggregation in the charcoalsuspension were assessed by using SEM and TEM. Represen-tative images are shown in Figure 1. Scanning electron micro-scopy revealed that the particle sizes of the dried charcoalsample ranged from approximately 50 nm to several micro-meters (Fig. 1a), and thus showed a broad size distribution.Notably, the morphology of smaller particles appeared to bedifferent from that the larger ones, with the smaller particlesbeing more rounded in comparison with the more angular largerparticles. Also, a regular dotted pattern was observed on thesurface of larger particles (Fig. 1a). This may indicate that thelarger particles were aggregates of smaller ones. The TEM(Fig. 1b) imaging further supported the SEM observations; theimage showed two single particles of 40 to 50 nm with morerounded edges (in circle of Fig. 1b). Although it is difficult toidentify single particles from their aggregates, the images stillhelp to explain better the sizes of single particles on the basis oftheir morphologies, which also have been successfully appliedin the study of C60 particles [17]. Hence, we termed the charcoalsample as nano-charcoal in the present study.

Adsorption of TBT and DBT to nano-charcoal

Because the adsorption of TBT to nano-charcoal at pH 6and pH 8 was taken at the lowest part of the isotherms, whereTBT solution concentration ranged from 0 to 50mg/L, theadsorption isotherms can be described with a linear equation(Eqn. 1). For the adsorption of DBT, due to its relatively lowtoxicity, the adsorption isotherms were prepared at considerablyhigher DBT aqueous concentrations up to 1,500mg/L, andhence the adsorption isotherms are curved and can be fittedby the Langmuir equation (Eqn. 2).

The adsorption isotherms and the parameters for adsorptionof TBT and DBT are presented in Figure 2. The figure showsthat the adsorption of TBT or DBT is pH-dependent, exhibitingstronger adsorption at pH 6 than at pH 8. This is in agreementwith our previous study and can be explained by the adsorptionthrough electrostatic bonding of cationic species and hydro-phobic adsorption of neutral species [15]. The log-partitioningcoefficient, log Kd for TBT was 5.0� 0.3 and 4.3� 0.1 at pH 6and pH 8, respectively, while the log Kd for DBT ranged from4.5� 0.2 to 5.5� 0.1 at pH 6, and from 3.9� 0.1 to 4.4� 0.3at pH 8, respectively, due to the nonlinear adsorption isothermobserved (Fig. 2). In the toxicity tests, it was found that

approximately 75% of the total TBT mass was adsorbed bythe nano-charcoal at pH 6, in comparison with approximately30% at pH 8. The adsorbed DBT mass ranged from 12 to 50%at pH 6, and from only 2 to 9% at pH 8, respectively. Theconsiderably lower adsorption for DBT is due to the weakeradsorption of DBT compared with TBT, and the much higheradsorbate:adsorbent ratios of up to 0.1 for DBT compared withthat of 0.003 for TBT.

Toxicity of nano-charcoal to D. magna

The toxicity of nano-charcoal to D. magna in aqueoussolution was studied in ISO-2 medium at pH 8. No significantimmobilization was observed for juveniles of D. magnaexposed to the nano-charcoal with concentrations up to50mg/L (data not shown). Furthermore, light microscopyshowed that nano-charcoal was mainly located in the gut aftera 24-h exposure to the nano-charcoal suspension (Fig. 3). Thisindicated that the nano-charcoal particles or aggregates wereingested by D. magna during the exposure, which is expectedfor a filter feeder such as D. magna. This lack of toxicity toD. magna was in accordance with several previous studies onvarious types of BC, e.g., soot [16], C60 [17], and carbonnanotubes [31].

pH influence on the toxicity of TBT and DBT

The toxicity of TBT and DBT to D. magna was assessed atpH 6 and 8. The results showed an increase in toxicity from pH 8to pH 6, in which the EC50 value decreased from 3.4� 0.1 to2.7� 0.2mg/L for TBT, and from 484� 17 to 331� 15mg/Lfor DBT, respectively (p< 0.05; Table 2). Furthermore, it wasfound that TBT possessed considerably stronger toxicity thanDBT in accordance with the findings by Vighi and Calamari [6].Tributyltin and dibutyltin can undergo hydrolysis with an acid-ity constant, pKa of 6.3 for TBT, and pKa,1 of 3.0 and pKa,2 of 5.1for DBT [13] (Supplemental Data, Table S1). According tothese pKa values, TBT consists of approximately 70% of TBTþ

species and 30% of TBTOH species in the solution at pH 6,whereas the DBT distribution comprises approximately 45%cationic DBT(OH)þ and 55% neutral DBT(OH)2 species. At pH8, both TBT and DBT are present mainly as the neutral species(TBTOH and DBT[OH]2). The higher toxicity of TBT and DBTat pH 6 than at pH 8 therefore indicates that the cationic speciesTBTþ and DBT(OH)þ are more toxic to D. magna than theneutral forms. In contrast, Fent and Looser [32] reported thatTBT causes approximately three times higher mortality at pH 8than at pH 6, as observed in a single concentration test. Thediscrepancy between their results and the results of the presentstudy can be ascribed to the different experimental conditions,because phosphate buffer was used in their work which canform a strong complex with TBTþ [33], whereas noncomplex-ing MOPS was used in the present study. In agreement with ourresults, White and Tobin [12] found that the damaging effect ofTBTþ to yeast membranes (Candida maltosa) was stronger thanthat of TBTOH. This indicates that membrane damage may bean important mechanism for TBT toxicity to D. magna, whichmay also apply to DBT.

Water hardness effect on toxicity of TBT and DBT

The effect of water hardness on the toxicity of TBT and DBTwas determined at pH 6 and 8. The results showed that thetoxicity of BT toD. magna consistently increased from an EC50of 3.2� 0.1 to 1.6� 0.1mg/L for TBT, and from 309� 18 to191� 16mg/L for DBT, when water hardness decreased from5.0 to 0.6mM (Table 2; Fig. 4), respectively. Importantly, the


EC50 decreased for TBT and DBT when the water hardnessdecreased from 2.5 to 0.6mM, whereas no significant decreasewas observed when reducing the water hardness from 5.0 to2.5mM. Comparably, Heijerick et al. [34] reported thatD. magna EC50 (48 h) exposed to Zn2þ increased when the

water hardness (Ca2þ) was increased from 0 to 3mM, whereasthe EC50 was unaffected at higher hardness. A similar resultwas found in a study on the toxicity of Cu2þ to D. magna as afunction of water hardness (Ca2þ and Mg2þ) by De Scham-phelaere and Janssen [35]. At pH 6, a considerable fraction of

Fig. 1. (a) Scanning electronicmicroscopy images of nano-charcoal single particles (aa), and its aggregates (ab). The dotted pattern on surface (in red circle) of arepresentative aggregate may be single particles. (b) Transmission electronic microscopy image of nano-charcoal showing two single particles in red circle (ba).[Color figure can be seen in the online version of this article, available at wileyonlinelibrary.com]


the two BTs occur as cations (TBTþ and DBT[OH]þ) which canbind to the negatively charged sites (the biotic ligands) on livingorganisms [36], but these sites also attract other cations such asthe nontoxic Ca2þ and Mg2þ ions. Consequently, the toxic BTcations will compete with Ca2þ and Mg2þ ions, which mayexplain the decreased BT toxicity at increased water hardnessdue to increased competition from Ca2þ and Mg2þ ions at theactive sites. In addition, the result showed that the toxicity ofTBT at pH 8 was not substantially affected by variation inhardness. This supports the hypothesis that TBT and DBT enterorganisms through diffusion at pH 8 [12], because the two BTsexclusively exist as the neutral species at that pH (SupplementalData, Table S1). Studies have found that some organisms suchas Paracentrotus lividus have an EC50 value of 300 ng/L whenexposed to TBT, even in artificial seawater with a very highconcentration of cations at approximately pH 8 [37]. This alsoindicates that TBT can still be extremely toxic to organismsat pH 8, although the hardness is very high; this agrees with theresults of the present study.

Effect of nano-charcoal on toxicity of TBT and DBT

The D. magna EC50 (48 h) values for TBT and DBT, withand without addition of nano-charcoal, are given in Table 2,and selected dose–response curves for TBT are presented inFigure 4 (all dose–response curves for both TBT and DBT areshown in Supplemental Data, Fig. S1).

At pH 6, the addition of 12.0mg/L nano-charcoal remark-ably decreased the toxicity of TBT to D. magna by shifting theEC50 value from the range of 1.6 to 3.2mg/L without nano-charcoal, to the range of 6.7 to 9.3mg/L after addition of nano-charcoal, depending on the water hardness. The decreasingtoxicity was also found for TBT after adding 10.0mg/L ofnano-charcoal to the medium with 2.5mM hardness at pH 8(Table 2; Fig. 4). For DBT, the decreased toxicity to D. magnaonly occurred at the hardness of 0.6mM medium at pH 6 afterthe addition of 16.0mg/L nano-charcoal. For the other hard-nesses, there was no significant difference in toxicity of DBTbetween systems with or without nano-charcoal (Table 2; Sup-plemental Data, Fig. S1). Conversely, the toxicity of DBT in the

Fig. 3. Light microscopic graphs of Daphnia magna exposed to nano-charcoal–free aqueous solution (a), and to the nano-charcoal in solution for 24 h (b). Theconcentration of nano-charcoal was 20mg/L. [Color figure can be seen in the online version of this article, available at wileyonlinelibrary.com]

Fig. 2. Adsorption isotherms of tributyltin (TBT) and dibutyltin (DBT) tonano-charcoal in media with hardness of 2.5mM at pH 6 (*) and pH 8 (~).The scatter points within the circle represent the adsorption data at thesame initial concentrations in media with the hardnesses of 0.6mM (*)and 5mM ( ) for TBT and DBT, respectively. For the full isotherm, theinitial concentrations of TBT and DBT ranged from 8 to 50mg Sn/L, and 60to 1,500mg Sn/L, respectively. The concentration of nano-charcoal was16mg/L. Vertical bars represent standard deviation (n¼ 3).


presence of nano-charcoal (12.0mg/L) at pH 8 slightlyincreased, but not significantly.

Bao et al. [20] found a twofold reduction of the toxicity ofTBT by adding approximately 40mgC/L of peat or soil humic

acids, which can be compared with twofold to sixfold reductionby adding 10 to 12mg/L nano-charcoal in the present study. Thereduced toxicity of TBT is most likely due to the adsorption ofTBT to the nano-charcoal, resulting in reduced TBT aqueous

Table 2. The 48-h half maximal effective concentration (EC50) values of tributyltin and dibutyltin with Daphnia magna in the different test media with andwithout addition of nano-charcoala,b

Compound Test media pH Hardness Concentration of charcoal (mg/L) EC50 value (mg/L) b (Hill slope)

Tributyltin ISO-1 6 0.6 0 1.6� 0.1 �2.6� 0.312.0 8.9� 0.4 �5.1� 0.7

ISO-2 6 2.5 0 2.7� 0.2 �4.0� 0.512.0 9.3� 0.3 �5.3� 0.7

ISO-3 6 5 0.0 3.2� 0.1 �6.1� 0.912.0 6.7� 0.2 �4.9� 0.7

ISO-1 8 0.6 0 3.3� 0.2 �8.6� 0.8ISO-2 2.5 0 3.4� 0.1 �9.2� 1.7

10.0 7.0� 0.2 �12.7� 2.0Dibutyltin ISO-1 6 0.6 0 191� 16 �3.5� 0.4

16.0 285� 17 �3.7� 0.5ISO-2 6 2.5 0 331� 15 �4.6� 0.6

16.0 344� 21 �3.2� 0.4ISO-3 6 5 0 309� 18 �4.1� 0.7

16.0 327� 22 �3.3� 0.4ISO-2 8 2.5 0 484� 17 �3.6� 0.5

12.0 454� 21 �5.7� 0.7

a EC50 and the slope parameters is given� SE.b ISO¼ International Organization for Standardization.

Fig. 4. The dose–response curves of tributyltin with (——) and without (–––) addition of nano-charcoal (12mg/L), in the media with a hardness of 0.6mM(a), 2.5mM (b), and 5.0mM (c) at pH 6, and 2.5mM at pH 8 (d), respectively. The scatter points* and* represent two replicates for each toxicity test withaddition of nano-charcoal (12mg/L); and the scatter points~ and~ represent two replicates for each toxicity test in nano-charcoal–free solution. The verticalbars represent standard deviation (n¼ 4).


concentration. Studies suggest that the nonassociated TBTspecies are responsible for the toxicity when an adsorbent ispresent in aqueous solution [16,17]. According to the calcu-lation for free TBT by means of the adsorption isotherms, theeffective EC50 value, i.e., EC50 for free TBT in the ISO-2medium is estimated to be approximately 2.3mg/L at pH 6, and4.9mg/L at pH 8, respectively, which is comparable to thoseobtained from the tests without addition of nano-charcoal(Table 2). Similarly, the effective EC50 value for DBT basedon calculated free DBT in the ISO-2 medium is approximately300mg/L at pH 6, and 440mg/L at pH 8, respectively, whichis close to the EC50 values obtained in toxicity tests withoutnano-charcoal (Table 2). These results strongly suggest thatbioavailability, and hence acute toxicity of BT to D. magna,depends only on the concentration of free BT species insolution, even though the nano-charcoal–adsorbed BT can beingested by D. magna (Fig. 3). A similar dependency was alsoseen in a study on the toxicity of diuron in the presence of soot,although the uptake of soot in algae was not determined [16].Baun et al. [17] reported a contradictive result, where thepresence of C60 increased the toxicity of phenanthrene nearly10 times, which was interpreted as the delivery of the complexof phenanthrene-C60 directly into cell membranes. In addition,the present study found that an increase of the water hardnessfrom 2.5 to 5.0mM slightly but significantly increased thetoxicity of TBT in the presence of nano-charcoal, where theEC50 value decreased from 9.3 to 6.7mg/L. This was not seenfor DBT (Table 2). The discrepancy of effects of water hardness(Ca2þ and Mg2þ) on TBT and DBT toxicity is probably due tocompetition between toxic BT cations and nontoxic Ca2þ/Mg2þ

ions at the negatively charged surface sites of both the charcoaland the Daphnia.

Effect of nano-charcoal on TBT and DBT uptake

Tributyltin and dibutyltin uptake in D. magna was studiedwith and without the addition of nano-charcoal at pH 6 and pH 8(Fig. 5). Accumulated TBT and DBT in the Daphnia tended toreach an equilibrium (plateau) within approximately 20 h,which resembles the uptake of TBT by algae (Pavlova lutheri)[38]. As shown in Figure 3 and adsorption of the BTs bycharcoal (Fig. 2), the increased BTs uptake in the presenceof nano-charcoal, especially at pH 6, is undoubtedly due toingestion of charcoal-adsorbed BTs. This is in line with pre-vious findings that the content of TBT in Hyallella azteca ismuch higher when exposed to sediment-adsorbed TBT than toTBT in the absence of sediment [39]. However, studies foundthat addition of humic acid reduced the uptake of TBT inD. magna and larval midge (Chironomus riparius) [21,32],probably because humic acid-associated TBT cannot be accu-mulated and is not bioavailable to organisms [19,21]. Addi-tionally, negligible amounts of metabolites, DBT and MBTfrom TBT, or MBT from DBT were found in the body ofD. magna after 48 h uptake tests, respectively. This indicatesvery low metabolism of TBT and DBT in the organism, inaccordance with the observations by Fent and Looser [32].

In accordance with the results of previous studies of effectsof humic acids and sediments on TBT toxicity, our investigationhas shown increased ingestion but reduced toxicity of TBT andDBT to D. magna in the presence of nano-charcoal. Strongadsorption of the BTs to the charcoal causing reduced bioavail-ability, and hence toxicity, can explain this behavior. Althoughthe increased bioaccumulation of nano-charcoal–bound TBT orDBT does not lead to an increase in acute toxicity, long-term

investigations of chronic toxicity of nano-charcoal–associatedTBT and DBT might lead to a different result.

CONCLUSIONS

The results of the present study demonstrate that pH andwater hardness can significantly influence the toxicity of TBTand DBT to D. magna. The toxicity of TBT and DBT notablyincreased when pH decreased from 8 to 6. Enhancing thehardness of the media, from 0.6 to 2.5mM at pH 6 consistentlydecreased the toxicity of TBT and DBT, whereas no furtherdecrease in the toxicity at a further increase of hardness wasseen. This is due to competitive binding of TBT/DBT andbivalent cations (e.g., Ca2þ, Mg2þ) to biotic ligands ofD. magna. Exposure to nano-charcoal presented no acutetoxicity. The uptake of TBT and DBT in D. magna wasremarkably enhanced with addition of nano-charcoal, whichis attributed to uptake of nano-charcoal–associated TBT andDBT. In spite of the higher uptake, the toxicity of TBT toD. magna prominently decreased in the presence of nano-charcoal compared with absence of nano-charcoal at pH 6and 8. No significant decrease in toxicity of DBT was observedin the presence of 12 to 16mg/L nano-charcoal, due to theinsignificant decrease of free DBT concentration compared withthat for TBT in the presence of the same concentration of BC.

Fig. 5. The uptake of tributyltin (TBT) and dibutyltin (DBT) by Daphniamagna without the nano-charcoal at pH 6 (*) and pH 8 (*), and withadditionofnano-charcoal at pH6 (~) andpH8 (~).The initial concentrationofTBTandDBTwas3.6mg/Land310mg/L, respectively.Theconcentrationof nano-charcoal was 12.0 to 14.0mg/L. The error bars represent standarddeviation (n¼ 3).


These findings demonstrate that only the free BT species arebiologically active, which is consistent with the experimentswhere nano-charcoal was added. Long-term investigations ofchronic toxicity of nano-charcoal-associated TBT and DBTneed to be answered in future work.

SUPPLEMENTAL DATA

Table S1. The hydrolytic reactions in aqueous solution oftributyltin (TBT) and dibutyltin (DBT).

Fig. S1.Dose–response curves of dibutyltin with (——) andwithout (–––) addition of nano-charcoal, in ISO-1media in pH 6(a), ISO-2 media at pH 6 (b), ISO-3 media at pH 6 (c), and ISO-2 media at pH 8 (d). (299 KB DOC).

Acknowledgement—The authors thank Anders Baun, from the TechnicalUniversity of Denmark, for his valuable advice. Anja W. Stubbe, from theUniversity of Copenhagen, provided excellent assistance with the experi-ments. We are also grateful to Poul Hyttel, from the University ofCopenhagen, for assisting with the transmission electronic microscope.Christian Ritz, from the University of Copenhagen, gave great help with thestatistics for our article.

REFERENCES

1. Huggett RJ, Unger MA, Seligman PF. 1992. The marine biocidetributyltin.Assessingandmanaging the environmental risks.EnvironSciTechnol 26:232–237.

2. Hoch M. 2001. Organotin compounds in the environment—Anoverview. Appl Geochem 16:719–743.

3. Blaber SJM. 1970. The occurrence of a penis-like outgrowth behind theright tentacle in spent females ofNucella lapillus (L.).ProcMalacol Soc39:231–233.

4. Gibbs PE, Bryan GW. 1986. Reproductive failure in population of thedogwhelk, Nucella lapillus, caused by imposex induced by tributyltinfrom antifouling paints. J Mar Biol Assoc 66:767–777.

5. Matthiessen P, Gibbs PE. 1998. Critical appraisal of the evidence fortributyltin-mediated endocrine disruption in mollusks. Environ ToxicolChem 17:37–43.

6. Vighi M, Calamari D. 1985. QSAR for organotin compounds onDaphnia magna. Chemosphere 14:1925–1932.

7. FrouinH, LebeufM, Saint-LouisR,HammillM,Pelletier E, FournierM.2008.Toxic effect of tributyltin and itsmetabolitesonharbor seal (Phocavitulina) immune cells in vitro. Aquat Toxicol (Amst) 90:243–251.

8. TheCouncil of theEuropeanUnion. 2003.Regulation (EC)No782/2003of the European parliament and of the council of 14 April 2003 on theprohibition of organotin compounds on ships. Off J Eur Union L115:1–11.

9. Craiq PJ, ed. 2003. Organometallic Compounds in the Environment,2nd ed. John Wiley & Sons, Hoboken, NJ, USA.

10. Bartlett AJ, Borgmann U, Dixon DG, Batchelor SP, Maguire RJ. 2004.Accumulation of tributyltin inHyalella azteca as an indicator of chronictoxicity: Survival, growth, and reproduction. Environ Toxicol Chem23:2878–2888.

11. Adema-Hannes R, Shenker J. 2008. Acute lethal and teratogenic effectsof tributyltin chloride and copper chloride on mahi mahi (Coryphaenahippurus) eggs and larvae. Environ Toxicol Chem 27:2131–2135.

12. White JS, Tobin JM. 2004. Role of speciation in organotin toxicity to theyeast Candida maltosa. Environ Sci Technol 38:3877–3884.

13. Weidenhaupt A, Arnold C, Muller SR, Haderlein SB, SchwarzenbachRP.1997.Sorptionoforganotinbiocides tomineral surfaces.EnvironSciTechnol 31:2603–2609.

14. Huang JH, Matzner E. 2004. Adsorption and desorption of organotincompounds in organic and mineral soils. Eur J Soil Sci 55:693–698.

15. FangL,BorggaardOK,MarcussenH,HolmPE,HansenHCB.2010.ThepH-dependent adsorption of tributyltin to charcoals and soot. EnvironPollut 158:3642–3649.

16. KnauerK, SobekA,Bucheli TD. 2007. Reduced toxicity of diuron to thefreshwatergreenalgaPseudokirchneriella subcapitata in thepresenceofblack carbon. Aquat Toxicol (Amst) 83:143–148.

17. Baun A, Sørensen SN, Rasmussen RF, Hartmann NB, Koch CB. 2008.Toxicity and bioaccumulation of xenobiotic organic compound in thepresence of aqueous suspensions of aggregates of nano-C60. AquatToxicol (Amst) 86:379–387.

18. Ding Y, Harwood AD, Foslund HM, Lydy MJ. 2010. Distribution andtoxicity of sediment-associated pesticides in urban and agriculturalwaterways from Illinois, USA. Environ Toxicol Chem 29:149–157.

19. Haitzer M, Hoss S, Traunspurger W, Steinberg C. 1998. Effects ofdissolved organic matter (DOM) on the bioconcentration of organicchemicals in aquatic organisms—a review. Chemosphere 37:1335–1362.

20. BaoML,Dai SG, Pantani F. 1997. Effect of dissolved humicmaterial onthe toxicity of tributyltin chloride and triphenyltin chloride to Daphniamagna. Bull Environ Contam Toxicol 59:671–676.

21. Looser PW, Fent K, Berg M, Goudsmit GH, Schwarzenbach RP. 2000.Uptake and elimination of triogranotin compounds by larval midgeChironomus riparius in the absence and presence of Aldrich humic acid.Environ Sci Technol 34:5165–5171.

22. Jia G, Wang H, Yan L, Wang X, Pei R, Yan T, Zhao Y, Guo X. 2005.Cytotoxicity of carbon nanomaterials: Single-wall nanotube, multi-wallnanotube, and fullerene. Environ Sci Technol 39:1378–1383.

23. Cherukuri P, Gannon CJ, Leeuw TK, Schmidt HK, Smalley RE, CurleySA, Weisman RB. 2006. Mammalian pharmacokinetics of carbonnanotubes using intrinsic near-infrared fluorescence.ProcNatl Acad SciUSA 103:18882–18886.

24. International Organization for Standardization. 1996. Water qualitydetermination of the inhibition of the mobility ofDaphnia magna straus(Cladocera, Crustacea)—acute toxicity test. ISO 6341. Geneva,Switzerland.

25. Schamphelaere KAC, Heijerick DG, Janssen CR. 2004. Comparison ofthe effect of different pH buffering techniques on the toxicity of copperand zinc to Daphnia magna and Pseudokirchneriella subcapitata.Ecotoxicol 13:697–705.

26. Hoch M, Alonso-Azcarate J, Lischick M. 2002. Adsorption behavior oftoxic tributyltin to clay-rich sediments under various environmentalconditions. Environ Toxicol Chem 21:1390–1397.

27. Organisation for Economic Co-operation and Development. 1998.Daphnia magna reproduction test. OECD Guideline 211. Paris,France.

28. Organisation for Economic Co-operation and Development. 1993.Daphnia sp. acute immobilization test and reproduction test. OECDProcedure 202. Paris, France.

29. R Development Core Team. 2009. R: A Language and Environment forStatistical Computing. R Foundation for Statistical Computing, Vienna,Austria.

30. Ritz C. 2010. Toward a unified approach to dose-response modeling inecotoxicology. Environ Toxicol Chem 29:220–229.

31. Petersen EJ, Akkanen J, Kukkonen JVK, Weber WJ. 2009. Biologicaluptakeanddeuperationof carbonnanotubesbyDaphniamagna.EnvironSci Technol 43:2969–2975.

32. Fent K, Looser PW. 1995. Bioaccumulation and bioavailabilityof tributyltin chloride: Influence of pH and humic acids. Water Res29:1631–1637.

33. Arnold CG, Weidenhaupt A, David MM, Muller SR, Haderlein SB,Schwarzenbach RP. 1997. Aqueous speciation and 1-octano-waterportioning of tributyl- and triphenyltin: Effect of pH and ioncomposition. Environ Sci Technol 31:2595–2602.

34. Heijerick DG, De Schamphelaere KAC, Janssen CR. 2002. Predictingacute zinc toxicity for Daphnia magna as a function of key waterchemistry characteristics: Development and validation of a biotic ligandmodel. Environ Toxicol Chem 21:1309–1315.

35. De Schamphelaere KAC, Janssen CR. 2002. A biotic ligand modelpredicting acute copper toxicity for Daphnia magna: The effects ofcalcium, magnesium, sodium, potassium, and pH. Environ Sci Technol36:48–54.

36. Kinraide TB. 2006. Plasma membrane surface potential (cpm) as adeterminant of ion bioavailability: A critical analysis of new andpublished toxicological studies and a simplified method for thecomputation of plant cpm. Environ Toxicol Chem 25:3188–3198.

37. Bellas J, Beiras R, Marino-Balsa JC, Fernandez N. 2005. Toxicity oforganic compounds to marine invertebrate embryos and larvae: Acomparison between the sea urchin embryogenesis bioassay andalternative test species. Ecotoxicology 14:337–353.

38. St-Louis R, Pelletier E, Marsot P. 1997. A mechanistic approach totributyltin (TBT) sorption by marine microflagellated alga Pavlovalutheri. Appl Organomet Chem 11:543–550.

39. Bartlett AJ, Borgmann U, Dixon DG, Batchelor SP, Maguire RJ. 2004.Tributyltin uptake and depuration in Hyalella azteca: Implications forexperiment design. Environ Toxicol Chem 23:426–434.


Documents

Toxicity and uptake of TRI- and dibutyltin in Daphnia magna in the absence and presence of nano-charcoal