Accumulation and Distribution of Multiwalled Carbon Nanotubes in

Sep 26, 2014 - Institute for Environmental Research (Biology V), RWTH Aachen ... RWTH Aachen University, Lukasstrasse 1, 52056 Aachen, Germany...
0 downloads 0 Views 1MB Size
Subscriber access provided by ECU Libraries

Article

Accumulation and distribution of multiwalled carbon nanotubes in zebrafish (Danio rerio) Hanna Maja Maes, Felix Stibany, Sebastian Giefers, Benjamin Daniels, Björn Deutschmann, Werner Baumgartner, and Andreas Schaeffer Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/es503006v • Publication Date (Web): 26 Sep 2014 Downloaded from http://pubs.acs.org on September 30, 2014

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23

Environmental Science & Technology

TOC art 230x133mm (150 x 150 DPI)

ACS Paragon Plus Environment

Environmental Science & Technology

Page 2 of 23

Accumulation and distribution of multiwalled carbon nanotubes in zebrafish (Danio rerio)

1 2 3 4 5

Hanna M. Maes*1, Felix Stibany1, Sebastian Giefers1, Benjamin Daniels1, Björn Deutschmann1, Werner Baumgartner2, and Andreas Schäffer1

6 7 8 9

1

10 11 12

Institute for Environmental Research (Biology V), RWTH Aachen University, Worringerweg 1, 52074 Aachen, Germany 2 Department for Cellular Neurobionics (Biology II), RWTH Aachen University, Lukasstrasse 1, 52056 Aachen, Germany *Corresponding author:

mail: [email protected] phone: +49-241-80-26698 fax: +49-241-80-22182

13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 1 ACS Paragon Plus Environment

Page 3 of 23

31

Environmental Science & Technology

TOC Art

32 33 34 35 36

Abstract

37

No data on the bioaccumulation and distribution of multiwalled carbon nanotubes (MWCNT) in

38

aquatic vertebrates is available till now. We quantified uptake and elimination of dispersed

39

radiolabelled MWCNT (14C-MWCNT; 1 mg/L) by zebrafish (Danio rerio) over time. The influences of

40

the feeding regime and presence of dissolved organic carbon (DOC) on accumulation of the

41

nanomaterial were determined. The partitioning of radioactivity to different organs and tissues was

42

measured in all experiments. A bioaccumulation factor of 16 L/kg fish wet weight was derived.

43

MWCNT quickly associated with the fish and steady state was reached within one day. After transfer

44

to clear medium, MWCNT were quickly released to the water phase, but on average 5 mg

45

MWCNT/kg fish dry weight remained associated with the fish. The nanomaterial mainly accumulated

46

in the gut of all fish. Feeding led to lower internal concentrations due to facilitated elimination via

47

the digestive tract. In presence of DOC, a tenfold less was taken up by the fish after 48 h of exposure

48

compared to without DOC. Quick adhesion to and detachment from superficial tissues were

49

observed. Remarkably, little fractions of the internalized radioactivity were detected in the blood and

50

muscle tissue of exposed fish. The part accumulated in these fish compartments remained constant

51

during the elimination phase. Hence, biomagnification of MWCNT in the food chain is possible and

52

should be subject of further research.

53 2 ACS Paragon Plus Environment

Environmental Science & Technology

Page 4 of 23

54

Introduction

55

Carbon nanotubes (CNT) are considered promising materials in nanotechnology1. Investment in

56

nanotechnology research and development as well as nanoparticle production volumes are growing

57

rapidly worldwide2, 3. Based on the assumption that this will lead to increasing CNT release in the

58

environment, a lot of research has recently been dedicated to investigating the possible (eco)toxicity

59

of this nanomaterial4. However, little information is available on the impact of CNT and other

60

carbonaceous nanoparticles on aquatic vertebrates. Some studies found that these nanomaterials

61

did not affect exposed organisms5, but it was also reported that single walled CNT (SWCNT) exerted

62

respiratory toxicity to rainbow trout (Oncorhynchus mykiss)6, that SWCNT and fullerenol altered the

63

antioxidant capacity of adult zebrafish (Danio rerio)7, that multiwalled CNT (MWCNT) and fullerene

64

caused toxicity in zebrafish embryos8, 9, and that double walled CNT (DWCNT) physically blocked

65

epithelial tissue of larvae of the African clawed frog (Xenopus laevis)10. As can be extracted from a

66

recent review, there is a distinct lack on data on bioaccumulation of CNT11. This is consistent with the

67

limited number of analytical methods available to differentiate carbon-based nanoparticles from

68

organic matrices, and thus, to detect these materials and measure their concentration in biological

69

samples12-15. Using mainly microscopic techniques, it was observed that CNTs clogged the filter

70

apparatus and covered the carapace of daphnids16, precipitated on the gills of rainbow trout6, or

71

filled the gastrointestinal tract of several aquatic invertebrates16-18 after exposure of the organisms

72

via aqueous dispersions of CNT. From research published up to now, it seems that nanotubes do not

73

pass the gut epithelium to accumulate in exposed aquatic organisms4, 15, 19. Benthic invertebrates and

74

earthworms were shown to incorporate CNT from sediment, food, and soil matrices, but the

75

nanomaterials did not appear to pass the gut epithelium either20-22. Since most described effects

76

could not be linked to internal distribution of nanoparticles and/or their agglomerates, the

77

underlying mechanisms of possible CNT toxicity remain subject of many current investigations. In a

78

recent study, SWCNT were tracked in fish using near infrared fluorescence after single gavage

79

exposure15. 3 ACS Paragon Plus Environment

Page 5 of 23

Environmental Science & Technology

80

We studied the accumulation and distribution of MWCNT for the first time in a vertebrate after

81

waterborne exposure. Adult zebrafish were exposed to radiolabeled MWCNT (14C-MWCNT) for up to

82

one month and uptake of radioactivity by the fish was quantified over time. Afterwards, exposed fish

83

were transferred to clear water to examine elimination of MWCNT. The influences of the feeding

84

regime and the presence of dissolved organic carbon (DOC) on MWCNT accumulation in zebrafish

85

were additionally determined. In all experiments, fish were dissected to assess the fraction of

86

MWCNT partitioned to different fish compartments.

87 88

Materials and Methods

89

Preparation of 14C-MWCNT test suspensions for exposure of zebrafish

90

One batch of

91

synthesized (as described in the SI) and used to perform all experiments. 500 µg of

92

agglomerates were weighed on a microbalance (0.0001 mg readability; Mettler Toledo UMX2,

93

Mettler Toledo GmbH, Germany) and added to a 1000 mL test beaker, filled with 500 mL of test

94

medium, which contained 294.0 mg/L CaCl2, 123.3 mg/L MgSO4, 63.0 mg/L NaHCO3, and 5.3 mg/L

95

KCl. The beaker was placed in an ice bath and the MWCNT material was homogenized by means of

96

ultrasonication with a microtip (Sonoplus HD 2070, Bandelin, Germany) set to a 0.2 seconds pulse at

97

70 W – 0.8 seconds pause regime for 5 minutes. The procedure was repeated two times. This

98

method resulted in the presence of small agglomerates and single nanotubes of 0.2 to 1.0 µm tube

99

length (both referred to as MWCNT) in the medium, as was visualized by means of transmission

100

electron microscopy (TEM; SI Figure S3). In case DOC was supplemented to the medium, a stock

101

solution was prepared by mixing 2.5 g Aldrich humic acid (Sigma Aldrich Chemie GmbH, Germany)

102

and 0.4 g NaOH (Carl Roth GmbH, Germany) in 1 L of distilled water. After stirring overnight, the

103

suspension was centrifuged for 15 minutes at 350 g and then filtered through three filters of

104

different pore sizes (1.2, 0.8, and 0.45 µm) to remove suspended particles. The pH was set to 7 with

105

HCl before the total organic carbon (TOC) content was determined using a TOC-Analyzer (Shimadzu

14

C-MWCNT with a specific radioactivity of 1.3 ± 0.1 MBq/mg (SI Table S5) was 14

C-MWCNT

4 ACS Paragon Plus Environment

Environmental Science & Technology

Page 6 of 23

106

Europe GmbH, Duisburg, Germany). The volume to add to 500 mL medium to obtain 8 mg DOC/L was

107

calculated and amounted to about 6 mL. In this case, the MWCNT agglomerates were sonicated in

108

about 494 mL and 6 mL DOC stock solution was added afterwards. The 14C-MWCNT concentration (1

109

mg/L) and homogeneity of the test dispersion were verified by measuring the amount of radioactivity

110

in replicate samples directly after sonication. Four samples of 1 mL were mixed with 5 mL of

111

scintillation cocktail (Insta-Gel PlusTM, Perkin Elmer, Germany) and subjected to liquid scintillation

112

counting (LSC; LS 5000 TD, Beckmann Instruments GmbH, Germany; detection limit: 1 Bq

113

corresponding to 0.8 ng MWCNT). Using a radioactive standard (SPEC-CHEC-14C, Perkin Elmer,

114

Germany), it was verified that the presence of DOC did not influence LSC measurements in the

115

applied amounts. After placing the test beakers in a heated water bath (25 ± 1°C), one adult female

116

fish was added to the MWCNT suspension.

117

Test setup

118

Adult female zebrafish (Danio rerio held as described in the SI; 453 ± 103 mg wet weight) were

119

individually exposed to 1 mg

120

was shown that this rather high concentration was necessary to be able to detect radioactivity in fish

121

compartments, but did not cause adverse effects on the growth and condition of zebrafish (data not

122

shown). Females were selected for this bioaccumulation study as they are bigger than male fish of

123

the same age. For comparability reasons, only one sex was tested. At least five fish were subjected to

124

each treatment, i.e. in separate test beakers. Since preliminary experiments had shown that data of

125

MWCNT accumulation involved large scatter, more replicates were provided for important sampling

126

points. Moreover, data of different experiments involving the same treatment and exposure duration

127

were pooled, when no significant differences in the results were observed. In the uptake experiment,

128

uptake of MWCNT by the fish was quantified over time. Sampling of the water phase and the fish

129

was performed after 3 h, 1 day, 2 days, 1 week, 2 weeks, 3 weeks, and 4 weeks, as described below.

130

In the elimination experiment, fish were exposed to 14C-MWCNT for two weeks, and then rinsed and

131

transferred to clear water to quantify elimination of MWCNT. Samples were taken after exposure,

14

C-MWCNT/L in 500 mL test medium. In preliminary experiments, it

5 ACS Paragon Plus Environment

Page 7 of 23

Environmental Science & Technology

132

and after an elimination period of 4, 24, 48, and 168 h. In both tests, fish were fed every second day

133

before the MWCNT suspension or clear medium was completely renewed (water renewal every 48

134

h). They were supplied with 2% of their wet weight in Artemia nauplii that were consumed within ten

135

minutes. In the feeding experiment, the influence of the feeding regime on MWCNT accumulation in

136

the fish was determined. Fish were either not fed or fed once a day with 1% of their body weight in

137

Artemia nauplii during a 48 h exposure period, or additionally fed every second day with 2% of their

138

body weight in Artemia nauplii after complete water renewal during a 168 h exposure period. In the

139

DOC experiment, it was examined whether DOC has an influence on accumulation of MWCNT in

140

zebrafish. Fish were exposed to 1 mg MWCNT/L in the presence or absence of 8 mg DOC/L. The fish

141

were not fed and the water was not renewed during this experiment. Sampling was performed after

142

3, 24, and 48 h.

143

Sampling

144

At each sampling point, fish were caught with a steel grid and rinsed to remove MWCNT containing

145

water from the skin. Their length (L) and weight (W) were measured to determine the condition

146

factor (CF: W/L³) that was compared with the CF of five control fish held under the same conditions

147

in clear water. Afterwards, they were anaesthetized in a saturated benzocaine (Sigma Aldrich,

148

Germany) solution in tap water. Depending on the aim of the experiment, only the gut of the fish was

149

removed, or they were fully dissected in order to study the distribution of CNT in zebrafish (see SI).

150

Except for the blood, all tissues were separately placed on small pre-weighed pieces of aluminium

151

foil, dried in an oven at 60 °C, and weighed on a microbalance. All samples were transferred to stable

152

glass test tubes, covered with tissue solubilizer (BTS-450, Beckman Coulter, USA), and incubated for

153

24 h at 37°C. Afterwards, a 35% hydrogen peroxide solution was added to bleach the samples. They

154

were transferred to 20 mL scintillation vials using methanol to rinse the glass tubes. The vials were

155

filled with scintillation cocktail, and subjected to LSC after storage overnight at 4°C in the dark. The

156

amount of radioactivity in samples of unexposed fish, which were bleached and solubilized in the

157

same way as those of exposed fish, was below the limit of detection. Adding a known amount of 14C6 ACS Paragon Plus Environment

Environmental Science & Technology

Page 8 of 23

158

MWCNT to blank tissue samples, a recovery higher than 95% could be derived (data not shown). To

159

make a balance of the total radioactivity present in the system (SI Figure S7), the MWCNT

160

concentration in the medium was again measured. Thereto, the remaining water in the test beakers

161

was sonicated as described above, in order to again homogenize the test dispersion, now containing

162

settled agglomerates of MWCNT. The radioactivity in four 1 mL water samples per test beaker was

163

measured by means of LSC.

164

Data interpretation

165

In order to calculate the bioaccumulation factor (BAF), a one-compartment model was fitted to the

166

uptake and elimination data (according to Connell, 199823):

 =  ∗

∗ 1 −  ∗  +   ∗   ∗



167

where Cb and Cw are the concentrations of MWCNT detected in the fish and the water, t represents

168

time, and k1 and k2 are the uptake and elimination rate constants. Since the amount of MWCNT in

169

the water body did not change a lot due to uptake by the fish (SI Figure S7) and since the water was

170

renewed every second day, Cw can be regarded as invariable. It amounted to 1 and 0 mg/L during the

171

uptake and depuration phase, respectively. At steady state Cb/Cw equals k1/k2, giving the BAF.

172

Differences between treatments were tested after checking the data for outliers by means of

173

Dixon/Grubbs test. Then, the data were checked for normality and variance homogeneity with

174

Shapiro-Wilk’s test and Levene’s test, respectively. Statistical differences were identified by means of

175

pairwise comparison using Student’s t-test in case of homoscedasticity and by Welch’s t-test in case

176

of heteroscedasticity. For the evaluation of statistically significant fluctuations over time, Dunnett’s t-

177

test was applied when variances were homogeneous and Welch’s t-test with Bonferroni adjustment

178

when variances were heterogeneous. A significance level (α) of 0.05 was chosen for all tests.

179 180

Results

181

There was no significant difference in the condition factor (CF) of exposed and control fish at all time

182

points (SI Table S1). In the uptake experiment, MWCNT quickly associated with zebrafish over time, 7 ACS Paragon Plus Environment

Page 9 of 23

Environmental Science & Technology

183

but the variability between replicates was extremely high (with coefficients of variation of 70 to

184

150%). Steady state was already reached within 24 h (Fig. 1, A, Y1). Maximal concentrations were

185

measured after two weeks of exposure, where on average 6% (25 µg) of the total MWCNT amount

186

present in the water was taken up by the fish (Table 3). The mean MWCNT level (± standard

187

deviation) in all complete fish of the 24 h up to 4 weeks treatments amounted to 73 (± 93) ng

188

MWCNT/mg fish dry weight (= 16 ng MWCNT/mg fish wet weight) (Fig. 1, A, Y1), and since steady

189

state was reached, a bioaccumulation factor of 73 on dry weight (16 on wet weight) could be

190

calculated (BAF = Cb/Cw = (73 mg/kg)/(1 mg/L)). In the elimination experiment, MWCNT were quickly

191

rejected to the water phase, but after 24 h, no further elimination of the material could be observed

192

(Fig. 1, B), i.e. the fish MWCNT concentrations after 24, 48, and 168 h of presence in clear water were

193

not significantly different, amounting to 5 ng MWCNT/mg fish dry weight on average (SI Table S4).

194

The data were modelled over 48 h, the time between the beginning of the depuration phase and the

195

first water renewal. For complete fish, an elimination rate constant (k2) of 0.14 h-1 was calculated

196

(Fig. 1, B). With this value, an uptake rate constant (k1) of 9.64 L*kg-1*h-1 was derived. Dividing k1 by

197

k2, a BAF of 70 is obtained, verifying the above calculated BAF using Cb and Cw at steady state.

198

MWCNT mainly accumulated in the digestive tract of the fish (Table 1, 2, 3; SI Table S2, S3, S4), which

199

is not necessarily presenting a bioavailable fraction. Hence, it was removed and measured separately

200

from the total remaining fish body. The one-compartment model did not fit to the data of fish from

201

which the gut was removed. In the uptake phase, steady state was directly reached, i.e. no

202

differences in the fish MWCNT content were observed over time (Fig. 1, A, Y2). Similarly, elimination

203

was quicker than predicted by the model (Fig. 1, C). After 4 h, most of the MWCNT were no longer

204

associated with the gut excised fish and the internal MWCNT content remained constant up to the

205

last sampling point of 168 h, amounting to 0.24 ng/mg fish dry weight on average.

8 ACS Paragon Plus Environment

Environmental Science & Technology

Page 10 of 23

A

206 B

C

207 208 209 210 211 212 213

Fig. 1 Concentration of multiwalled carbon nanotubes (MWCNT) in zebrafish on dry weight (dw) after exposure to 1 mg MWCNT/L via the water phase in function of time. The upper graph (A) shows uptake by the fish over time. Filled rounds represent the concentrations in complete fish (Y-axis on the left: Y1), non-filled rounds those in fish from which the gut was removed (Y-axis on the right: Y2). The graphs below (B, C) present elimination of MWCNT from two weeks exposed zebrafish (time = 0 h) that were transferred to clear water. Again filled and non-filled rounds represent complete and gut excised fish, respectively. A one compartment model is plotted to all data (R²: determination coefficient).

214 215

Feeding of the fish led to significantly facilitated elimination of MWCNT via the digestive tract, as was

216

shown in the feeding experiment (Fig. 2). Within an exposure period of 48 h without water renewal,

217

unfed fish had an MWCNT body burden that was on average 45 times higher than fish that were daily

218

supplied with uncontaminated Artemia nauplii (Fig. 2, A). After 168 h, fish that were fed once a day

219

accumulated less MWCNT than fish that were fed once every second day and much less than fish

220

that were not fed at all (Fig. 2, B). No significant differences in the MWCNT content of fish from

221

which the gut was removed were observed (Fig. 2, D). The differences originated from the fraction in

222

the gut (Fig. 2, C). In the DOC experiment, it was shown that the presence of DOC in the medium also

223

resulted in the accumulation of lower amounts of MWCNT by the fish (Fig. 3). After 48 h, the total

224

concentrations in fish exposed to MWCNT in presence and absence of DOC amounted to 9 and 91 9 ACS Paragon Plus Environment

Page 11 of 23

Environmental Science & Technology

225

ng/mg dry weight, respectively (Fig. 3; SI Table S2 and Table S3). Compared to unfed fish exposed

226

without DOC, MWCNT got quicker associated with the gills and to larger relative amounts when DOC

227

was present in the medium (Table 1 vs. Table 2). The sum of MWCNT present in the skin, filet, and

228

blood (= rest fish), and the gills was relatively larger with (11%) than without (2%) DOC after 48 h of

229

exposure. Furthermore, relatively less was present in the gut with (89%) than without (96%) DOC

230

(Table 1 vs. Table 2).

231 A

B

232 D

C

233 234 235 236 237 238 239

Fig. 2 Concentration of multiwalled carbon nanotubes (MWCNT) in zebrafish (gut) after exposure to 1 mg MWCNT/L via the water phase for 48 or 168 h, during which food was provided in different regimes (no food, one time per two days, and once per day) and the MWCNT dispersion was either renewed one time every 48 h or not at all. The values are expressed on dry weight (dw). The rounds, the asterisks, the vertical bars, and the error bars represent the data points for separately exposed fish, outliers, the mean, and the standard deviation on the mean of remaining replicates, respectively. Significant differences between treatments are indicated by different letters.

10 ACS Paragon Plus Environment

Environmental Science & Technology

Page 12 of 23

240 241 242 243 244

Fig. 3 Concentration of multiwalled carbon nanotubes (MWCNT) in zebrafish after exposure to 1 mg CNT/L via the water phase in function of time and in absence (-) or presence (+) of dissolved organic carbon (DOC, 8 mg/L). The mean of five replicates is presented by the vertical bars to which the error bars represent the standard deviation, all expressed on fish dry weight (dw). Significant differences between DOC treatments at one time point are indicated by the asterisks.

245 246

As was mentioned above, MWCNT mainly accumulated in the gut of all fish (Table 1, 2, 3; SI Table S2,

247

S3, S4). Large relative amounts of radioactivity were also detected in gills, skin and muscle samples of

248

briefly exposed fish (3 h). Afterwards, only low percentages of the total amount were associated to

249

these fish compartments. No distribution to the liver, the gonads, and the brain was observed.

250

However, low amounts of radioactivity were detected in blood of fish exposed for more than one

251

week. The average total amount of radioactivity in blood samples of unfed fish (exposed for only one

252

week) was higher than in fish fed every second day (exposed for two weeks), corresponding to 7.6

253

and 2.5 ng 14C-MWCNT, respectively (Table 2 vs. Table 3). During the elimination phase, the absolute

254

blood MWCNT content did not significantly change, leading to an increase of the relative amount of

255

radioactivity in the blood current over time. On the other hand, the absolute amount of MWCNT

256

associated with the rest of the fish, i.e. the skin and filet, drastically decreased (from ca. 100 ng to 20

257

ng) within the first 4 h of depuration. Afterwards, the relative amount of CNT in the rest fish

258

increased, but the absolute amount corresponded to about 20 to 30 ng 14C-MWCNT at all elimination

259

sampling points (Table 3). The MWCNT concentration in the skin and filet indeed decreased within

260

the first four hours of elimination and remained more or less constant thereafter (SI Table S4). It is

261

suggested that quick detachment from the skin occurred, and that the remaining radioactivity

262

concerned nanomaterial that reached muscle tissue. Quick dissociation of MWCNT from external fish 11 ACS Paragon Plus Environment

Page 13 of 23

Environmental Science & Technology

263

tissue was demonstrated for the gills, to which no more MWCNT were attached after 168 h of

264

depuration (Table 3). Similarly, quick adhesion to the fish surface was measured in the uptake phase:

265

on average 16 and 26% of the total amount of MWCNT associated with the fish were detected in gill

266

tissue and skin within 3 h (Table 2). It seems therefore that connection of MWCNT to the fish exterior

267

is reversible. During the depuration period, the part accumulated in internal fish compartments

268

remained constant (as shown for blood and assumed for the filet), and the main MWCNT fraction

269

was purged from the gut and detached from the gills and skin (as shown for the gills and assumed for

270

the skin).

271 272 273 274 275 276 277

Table 1 Distribution of multiwalled carbon nanotubes (MWCNT) in different compartments of zebrafish, which were 14 exposed to 1 mg C-MWCNT/L via the water phase in the presence of 8 mg dissolved organic carbon/L. The fraction in each fish compartment is expressed as the percentage of the total amount of radioactivity detected in the fish. The sum represents the total amount, and is expressed as the corresponding amount of MWCNT in µg. The rest is the relative amount of radioactivity measured in the residual fish, from which gut and gills were removed. Values that are struck out represent outliers (M: mean; SD: standard deviation).

exposure time (h) 3

elimination time (h) 0

M SD 24

0

M SD 48

0

M SD

gut (%) 8.2 68.6 11.4 5.5 6.4 7.9 2.6 90.4 93.2 99.0 97.3 12.9 95.0 3.9 92.2 83.8 95.4 84.4 89.4 89.0 5.0

gills (%) 43.8 4.8 12.4 39.4 40.0 28.1 18.1 3.0 0.2 0.3 0.5 28.6 1.0 1.3 7.1 9.8 1.3 3.5 8.7 6.1 3.6

rest = fish-gut-gills (%) 48.0 26.6 76.2 55.1 53.5 51.9 17.7 6.7 6.6 0.7 2.2 58.5 4.0 3.1 0.8 6.4 3.3 12.1 1.9 4.9 4.5

sum (µg) 0.035 0.122 0.107 0.039 0.035 0.068 0.043 0.054 0.198 0.709 0.197 0.020 0.235 0.277 0.368 0.048 1.602 0.655 0.662 0.667 0.580

278 279 280 281 12 ACS Paragon Plus Environment

Environmental Science & Technology

282 283 284 285 286

Page 14 of 23

Table 2 Distribution of multiwalled carbon nanotubes (MWCNT) in different compartments of unfed zebrafish, which 14 were exposed to 1 mg C-MWCNT/L via the water phase. No water renewal was performed. The fraction in each fish compartment is expressed as the percentage of the total amount of radioactivity detected in the fish. The sum represents the total amount, and is expressed as the corresponding amount of MWCNT in µg. Values that are struck out represent outliers (M: mean; SD: standard deviation; DL: detection limit).

exposure time (h) 3

elimination time (h) 0

M SD 24

0

M SD 48

0

M SD 168

0

M SD

gut (%) 1.2 6.3 2.7 6.8 29.3 4.3 2.7 31.0 99.0 92.3 99.5 75.0 91.3 11.3 99.1 91.2 94.9 97.5 98.8 96.3 3.3 99.7 99.8 99.9 98.6 99.9 99.6 0.6

gills (%) 7.9 32.9 24.6 8.0 6.0 15.9 12.1 35.0 0.4 2.5 0.2 24.2 12.4 16.2 0.5 0.0 4.2 1.8 0.4 1.4 1.7 0.11 0.03 0.02 0.36 0.03 0.03 0.00

blood (%)