Uptake and Transfer of 13C-Fullerenols from Scenedesmus obliquus

Oct 18, 2018 - Fullerenol, a water-soluble polyhydroxylated fullerene nanomaterial, enters aquatic organisms and ecosystems through different ingestio...
0 downloads 0 Views 1MB Size
Download PDF
Article Cite This: Environ. Sci. Technol. 2018, 52, 12133−12141

pubs.acs.org/est

Uptake and Transfer of 13C‑Fullerenols from Scenedesmus obliquus to Daphnia magna in an Aquatic Environment Chenglong Wang,† Xue-Ling Chang,*,† Qiuyue Shi,‡ and Xian Zhang*,‡ †

Key Lab for Biomedical Effects of Nanomaterials and Nanosafety, Institute of High Energy Physics, Chinese Academy of Sciences, Beijing 100049, China ‡ Key Lab of Urban Environment and Health, Institute of Urban Environment, Chinese Academy of Sciences, Xiamen 361021, China

Downloaded via UNIV OF LOUISIANA AT LAFAYETTE on November 6, 2018 at 20:17:52 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: Fullerenol, a water-soluble polyhydroxylated fullerene nanomaterial, enters aquatic organisms and ecosystems through different ingestion exposures and may pose environmental risks. The study of their uptake routes and transfer in aquatic systems is scarce. Herein, we quantitatively investigated the aquatic uptake and transfer of 13C-fullerenols from Scenedesmus obliquus to Daphnia magna using 13C-skeletonlabeling techniques. The bioaccumulation and depuration of fullerenol in Daphnia magna increased with exposure doses and time, reaching steady state within 16 h in aqueous and feeding-affected aqueous routes. The capacity of Daphnia magna to ingest fullerenol via the aqueous route was much higher than that via the dietary route. From the aqueous to feeding-affected aqueous, the kinetic analysis demonstrated the bioaccumulation factors decreases, which revealed that algae suppressed Daphnia magna uptake of fullerenols. The aqueous route was the primary fullerenols ingestion pathway for Daphnia magna. Kinetic analysis of the accumulation and transfer in Daphnia magna via the dietary route indicated low transfer efficiency of fullerenol along the Scenedesmus obliquus-Daphnia magna food chain. Using stable isotope labeling techniques, these quantitative data revealed that carbon nanomaterials underwent complex aquatic accumulation and transfer from primary producers to secondary consumers and algae inhibited their transfer in food chains.

■

INTRODUCTION Fullerenol nanomaterial is a water-soluble polyhydroxylated fullerene derivative that has been widely developed into many commercial applications because of its prominent biocompatibility and biofunctionality.1−4 For instance, fullerene and its derivatives were found to suppress tumor tissue without any damage to normal skin because of its ability to release singlet oxygen under visible light.5 Moreover, the metal polyhydroxylated derivative of fullerene (Gd@C82(OH)22) exhibited novel immunomodulatory effects during anticancer therapy both in vivo and in vitro.2,3,6 Progress in nanotechnology and the increasing use of nanomaterials have been accompanied by biosafety concerns because of the inevitable release of these materials to the environment, which may result in the accumulation of nanomaterials in the ecosystem, negative effects on organisms, and eventually effects on public health via transfer through the food chain.7−11 Oberdorster et al. found significant lipid peroxidation of uncoated nC60 in the brains and gills of largemouth bass.12 Specifically, they found that 21 days of exposure to 2.5 and 5 ppm nC60 resulted in a significant delay in molting and significantly reduced production of offspring by Daphnia.13 Consequently, it is important to investigate the environmental fate and toxicity or bioeffects of nanomaterials on ecosystems.14−17 However, the bioaccumulation and transfer processes of these materials among aquatic organisms of different trophic levels are still relatively unknown. © 2018 American Chemical Society

Quantitative exposure is a fundamental question in the assessment of ecological and human health risks of nanoparticles for aquatic organisms. To understand and predict the environmental ecological effects of nanoparticles, one must first understand the exposure route under natural conditions. Similar to conventional metal environmental pollutants,18 aquatic organisms can be exposed to nanoparticles directly from aqueous surroundings (aqueous route) and indirectly via their prey (dietary route), which is simultaneously exposed to the nanoparticles. The former exposure route can be further divided into two pathways, the simple aqueous route (only nanoparticles are ingested) and the feeding-affected route (nanoparticles are ingested simultaneously with food). For example, zinc from isotopically modified 67ZnO nanoparticles was efficiently assimilated by freshwater snails when ingested with food, and the dietary exposure to high concentrations of ZnO nanoparticles was found to exert adverse responses in the snails.19 However, comparison of the bioaccumulation and toxicity in Lymnaea stagnalis after waterborne and diet borne exposure revealed that diet borne exposure to 65CuO nanoparticles was more likely to cause adverse effects than waterborne exposure.20 When comparing CuO NP indirect Received: Revised: Accepted: Published: 12133

June 8, 2018 September 26, 2018 October 18, 2018 October 18, 2018 DOI: 10.1021/acs.est.8b03121 Environ. Sci. Technol. 2018, 52, 12133−12141

Article

Environmental Science & Technology

according to the method described by Li et al.48 The obtained C-fullerenols were then characterized by time-of-flight mass spectrometry (TOF-MS; UltrafleXtreme MALDI-TOF/TOF, Bruker, Karlsruhe, Germany), Fourier transform infrared spectroscopy (FT-IR; Spotlight 200i, PerkinElmer, Waltham, USA), and X-ray photoelectron spectroscopy (XPS; ESCALAB 250Xi, Thermo, Waltham, USA). The 13C-isotope rate of fullerenols was determined by isotope ratio mass spectrometry (IRMS; Delta V Advantage, Thermo, Bremen, Germany). The morphology of 13C-fullerenols dispersed in water was characterized by transmission electron microscopy (TEM; Tecnai F30, Philips-FEI, Netherlands). Preparation and Analysis of 13C-Fullerenols Solution. Stock solution was produced by dissolving 5 mg 13C-fullerenols in artifact freshwater (KCl, 1.2 mg/L; CaCl2·2H2O, 58.6 mg/ L; NaHCO3, 13.0 mg/L; and MgSO4·2H2O, 24.5 mg/L). Next, the stock solution was diluted with artifact freshwater to 1.0 and 0.1 mg/mL. Stock solutions of 1.0 mg/L and 0.1 mg/L 13 C-fullerenols were subsequently ultrasonicated for 5 min before measuring the hydrodynamic diameter of 13C-fullerenols using a Zetasizer Nano ZS system (ZEN 3600, Malvern, UK). Test Organisms. Scenedesmus obliquus and D. magna were purchased from the State Key Laboratory of Environmental Aquatic Chemistry, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences. Scenedesmus obliquus was incubated in 2 L conical flasks with 1.5 L of BG11 artifact freshwater and maintained in a manual climatic box at 25 ± 1 °C under a 16:8 h light/dark photoperiod. The conical flasks were shaken three times a day. Daphnia magna were maintained in 2 L beakers with 1.5 L of artifact freshwater at 20 ± 1 °C under a 16:8 h light/dark photoperiod. The D. magna were fed S. obliquus daily, and 80% of the artifact freshwater was refreshed every other day. Preparation of 13C-Fullerenols Incubated S. obliquus. 13 C-Fullerenols ingested by algal cells were prepared as a food source for dietary exposure experiments. Next, 50 mL of algal suspension in the exponential growth phase containing 0.4 mL of 13C-fullerenols stock solution of 1.0 or 0.1 mg/mL was transferred to a 500 mL conical flask with 350 mL of artifact freshwater and cultivated in a manual climatic box at 25 ± 1 °C under a 16:8 h light/dark photoperiod for 9 days. During cultivation, the conical flask was shaken three times a day. Three replicates were prepared for each concentration, and a control group was set. Next, 20 mL of algal suspension was collected at 1, 2, 3, 5, 7, and 9 d to monitor the growth of algae and quantify 13C-fullerenols. The growth of algae was indicated by the optical absorbance of algae suspension which related with the algae density using UV−vis spectrophotometry (UV3600, Shimadzu, Japan). To quantify fullerenols in algae, algal samples were then centrifuged in 50 mL centrifuge tubes at 6000 rpm for 3 min. Following centrifugation, the supernatant was removed, and the remaining algae were resuspended in ultrapure water. The resuspension and centrifugation process was then performed three more times, after which the algal samples were preserved at −20 °C. The concentrations of 13C-fullerenols in algae sampled at each time were subsequently determined by IRMS, while another set of algae was sampled for TEM analysis at 9 d. The remaining exposed algae were centrifuged and resuspended three times, after which they were used as food for dietary exposure

dietary and direct waterborne exposure in D. magna, Harper et al. found that D. magna regulate internal Cu differently depending on uptake routes of CuO NP exposure.21 These reports demonstrated the dose-dependent bioaccumulation pathways and route-regulated toxic effects of exogenous nanoparticles to aquatic creatures. The bioaccumulation and transfer processes of fullerenol nanoparticles among aquatic organisms of different trophic levels (from Scenedesmus obliquus to D. magna) were found to be crucial for evaluation of their environmental biosafety and risk. Before investigating the exposure route of fullerene, it is necessary to establish an effective quantification method for complex environmental or biological systems.22 Because of the high background level of carbon in the environment and organisms, it is extremely important to develop a method for high-precision quantification of ecosystem carbon nanomaterials.23,24 Among the existing analytical methods, isotope labeling methods, including radioactive and stable isotope labeling, comprise a very powerful tool to trace exogenous carbon elements and quantify carbon nanomaterials in organisms.25−27 Radioactive labeling has been widely applied in the evaluation of bioaccumulation of carbon nanomaterials in vivo; however, there are significant drawbacks associated with this method, such as the strict conditions required for use of radioactivity, the generation of radioactive wastes, and the shorter half-lives of some radioactive elements. To overcome these drawbacks, stable isotope labeling has been successfully applied to quantify and trace metal/metal oxides and carbon nanoparticles in vivo.19,28−33 To investigate stable 13C isotope labeled carbon nanomaterials, Liu et al. synthesized 13C labeled carbon nanotubes and studied the bioaccumulation and biodistribution of single-walled carbon nanotubes in mice using isotope ratio mass spectrometry (IRMS). 34 We developed 13C skeleton-labeled fullerene, graphene oxide, and their derivatives35−38 and successfully applied the 13C labeling to investigate their nanobio-effects and structure−activity relationships in environmental animals/plants and aquatic plankton.38−41 Moreover, we quantified the bioaccumulation/ depuration of 13C-labeled fullerenol nanoparticles in D. magna and their subgeneration transmission effect at low concentration by IRMS, which had competitive sensitivity to liquid chromatography−tandem mass spectrometry (LC−MS/MS) coupled ultraviolet/visible (UV/vis) spectroscopy.42−45 Stable isotopic tracing has shown great potential for use in investigations of environmental contamination and nutrition studies of nanoparticles to address questions regarding the uptake of elements that naturally have multiple isotopes. Scenedesmus obliquus is a primary producer,46,47 while D. magna are filter feeders in the freshwater food web. Herein, we investigated the accumulation of fullerenols from S. obliquus to D. magna via the aqueous route (direct uptake from water), dietary route (uptake via the food chain), and feeding-affected aqueous route. The specific goals of this study were to (a) compare the contribution of the aqueous route and dietary route, (b) evaluate the transferability of fullerenols via food chains, and (c) provide basic concentrations for subsequent exposure experiments using higher level aquatic organisms (such as fish) in ecological food chains.

13

■

MATERIALS AND METHODS Materials. The method for synthetic production of 13Clabeled fullerenes (13C60) has been described in our previous reports.35,36,39 Additionally, 13C-fullerenols were prepared 12134

DOI: 10.1021/acs.est.8b03121 Environ. Sci. Technol. 2018, 52, 12133−12141

Article

Environmental Science & Technology

Quantification of 13C-Fullerenols. The quantification of C-fullerenols was conducted as described in our previous studies.40−42 The carbon isotope ratios (13C/12C) of 13Cfullerenols, algae, and D. magna sampled in each experiment were determined by IRMS. Moreover, the carbon elemental contents were measured according to the peak areas obtained from mass spectrometry. Labeled urea was analyzed after every 10 samples to check the accuracy and precision of isotope ratios. According to the carbon elemental content, the 13C/12C ratio was converted to 13C concentration expressed as ω13C. The concentrations of 13C-fullerenols in algae and D. magna were then calculated according to the ω13C values of 13Cfullerenols, algae, and D. magna using eq 1 and expressed as milligrams per gram of dry weight (mg/g) of D. magna.

experiments, while algae in the control group were collected for feeding-affected aqueous exposure experiments. Aqueous Exposure Experiments. Aqueous exposure experiments, including a 24 h uptake period followed by a 24 h depuration period, were conducted in suspensions of 1.0 mg/L and 0.1 mg/L 13C-fullerenols. Here, the low and high fullerenol concentrations were selected to evaluate the concentration effect at nontoxic doses, where the experimental doses were selected under the guidance of other reports of bioeffect-experiments of fullerenols and other carbon nanomaterials on S. obliquus and D. magna.42,49,50 Briefly, 13Cfullerenols in artificial freshwater were freshly prepared and sonicated for 60 s to obtain full dispersions, after which 6-dayold D. magna were transferred to fresh culture media. After 1 day without feeding, 100 organisms were transferred to each beaker containing 500 mL of artificial freshwater with 0.1 mg/ L or 1.0 mg/L 13C-fullerenols. Three replicates in each treatment were used, and a control group was set. Daphnia magna were sampled at 0, 1, 4, 8, 12, 16, 20, and 24 h by sacrificing four mobile D. magna from each beaker at each sampling time. After sampling, the D. magna were rinsed three times using ultrapure water to remove fullerenols attached to their carapaces and then stored at −20 °C for 13C-isotope analysis. Another four D. magna were sampled at 24 h for TEM observations. At the end of the 24 h uptake period, all mobile D. magna in each beaker were transferred to clean beakers with fresh artifact freshwater for the depuration studies. In the depuration period, four D. magna from each breaker were collected at 4, 8, 12, 16, 20, and 24 h after transfer to clean freshwater. In addition, remaining D. magna were sampled daily (at 48 and 72 h) for a preliminary prolonged study of depuration. Sample preparation was the same as for the uptake experiments. At the end of the experiments, D. magna samples were freeze-dried (Boyikang, Beijing, China), weighed (Mettler Toledo microbalance, Zurich, Switzerland), and then transferred into tin capsules for carbon isotope analysis to determine the concentrations of 13C-fullerenols. Dietary Exposure Experiments. Dietary experiments included a 24 h uptake period and a 24 h depuration period. Briefly, 6-day-old D. magna were transferred to fresh culture media. After 1 day without feeding, 100 D. magna were added to each beaker containing 500 mL of artificial freshwater. There were three replicates of each treatment as well as a control group. After uptake for 0, 8, and 16 h, D. magna were fed algae that had been incubated in 1.0 and 0.1 mg/L 13Cfullerenols suspension at a concentration of 1.0 × 108 cells/L to ensure there was sufficient food available. The D. magna in control beakers were fed clean algae at the same concentration. The depuration period of dietary exposure experiments was the same as that in the aqueous exposure experiments. The sampling times and sample preparation in both the uptake and depuration period and the determination of 13C-fullerenols concentration were the same as for the aqueous exposure experiments. Another four D. magna were also sacrificed after 24 h of exposure for TEM observation. Feeding-Affected Aqueous Exposure Experiments. The influence of feeding on aqueous uptake was evaluated based on the aqueous exposure experiments. Briefly, the experimental setup and sample preparation were the same as for the preceding aqueous exposure experiments, except that the D. magna were fed clean algae at a concentration of 1.0 × 108 cells/L after 0, 8, and 16 h of uptake according to the dietary exposure experiments.

13

ωfullerenols =

ω13C(sample) − ω13C(control) ω13C(fullerenols)

(1)

Data Analysis. Rates for uptake (ku) and elimination (ke) were modeled using the first-order kinetic model given in eq 2, where Ct is the concentration in the organism at time t, and Cw is the water phase concentration.51 The bioaccumulation factors based on the kinetic method (BAFFit) were calculated from the kinetic parameters using eq 3. The biomagnification factor at time n (BMFn) was calculated for dietary exposure experiments using eq 4, where CF is the concentration of 13Cfullerenols in food (S. obliquus), and Con is the concentration of 13C-fullerenols in D. magna obtained from 13C-fullerenols containing S. obliquus at time n.52 Ct =

■

Cwku (1 − e−ket ) ke

(2)

BAFFit =

ku ke

(3)

BMFn =

Con CF

(4)

RESULTS AND DISCUSSION Characterization of 13C-Fullerenols. As obtained 13Cfullerenols were characterized by TOF-MS. The spectrum peaked at 725 m/z and had a Poisson distribution of 720 to 730 m/z (Figure 1a), which resulted from the incorporation of 13 C atoms in fullerene cages and indicated that there were 5− 10 13C atoms in each 13C-labeled fullerene molecule.35,36,39 Meanwhile, the Poisson distribution of fullerene cages confirmed that the process of synthesizing 13C-fullerenols did not destruct the cage structure of 13C−C60 and consequently would not result in the loss of 13C atmos. This result was consistent with our previous study on the structural properties of fullerenols.37,40−42,53 The FT-IR spectrum of fullerenols (Figure 1b) had a broad peak at 3408 cm−1 for νO−H, an intense peak at 1598 cm−1 for νCC, a characteristic peak of δsC−O−H bending at 1401 cm−1, and a weak vibration absorption peak of νC−O at 1084 cm−1, which indicated an abundance of oxygen containing groups. Curve-fitting analysis of the C 1s XPS spectrum revealed it was divided into three components (Figure 1c) as nonoxidized carbons, C−OH, and C−O−. The atomic ratio of carbon:oxygen was about 10:2.7 based on XPS, which was equal to 16 hydroxyl groups per fullerene cage [C60(OH)16].54 The mass ratio of 13C atoms to 12135

DOI: 10.1021/acs.est.8b03121 Environ. Sci. Technol. 2018, 52, 12133−12141

Article

Environmental Science & Technology

Figure 2. Changes in 13C-fullerenols concentration in S. obliquus incubated with 13C-fullerenols. (The low concentration group is indicated by the right vertical axis, and the high concentration group is indicated by the left vertical axis.)

relative to their concentration disparity; consequently, the algae in the 1.0 mg/L group may have had more chances to interact with surrounding fullerenols particles, resulting in faster accumulation than in the 0.1 mg/L group. After the vertex, from 2 to 5 d, the concentrations declined remarkably in both the 0.1 mg/L and 1.0 mg/L group. In the later incubation process, the concentration continued to decline. The decrease in 13C-fullerenols concentration could be attributed to the rapid increase of algal cells in the exponential phase. When the exposed algae were collected for dietary exposure experiments after 9 days of incubation, the 13Cfullerenol concentrations of both the 0.1 mg/L and 1.0 mg/L group were found to be 1.07 and 8.23 mg 13C-fullerenols/g dw algae, respectively. After 9 days of incubation, the gap in 13Cfullerenols concentrations in algae between the two exposure groups was reduced, which may have contributed to the easier aggregation and sedimentation of fullerenol particles with a higher aqueous concentration. Under TEM (Figure S3a), the presence of a dark dot with a diameter of about 200 nm (circled in red line) in an algal cell was considered to be a 13Cfullerenol particle that had been absorbed. Uptake and Depuration of 13C-Fullerenols in D. magna. To investigate the contribution of the aqueous and dietary routes to the uptake of fullerenols for D. magna in the natural ecosystem, the uptake pathway was divided into an aqueous route, dietary route, and feeding-affected aqueous route (Figure 3). In aqueous exposure experiments, the 13C-fullerenols body burden of D. magna increased rapidly during the first 12 h

Figure 1. a) TOF-MS of 13C-fullerenols. b) FT-IR spectrum of 13Cfullerenols. c) XPS of 13C-fullerenols. d) TEM image of 13Cfullerenols in water suspension. 13

C-fullerenols (ω13C(f ullerenols)) was calculated as 4.57% according to IRMS, which was higher than the natural abundance (1.1%). The morphology of 13C-fullerenols particles characterized by TEM is shown in Figure 1d. The diameters of 13C-fullerenols particles suspended in water were about 200−300 nm. Dynamic light scattering (DLS) measurements (Figure S1) indicated the distribution of diameters of 13 C-fullerenols in 0.1 mg/L stock solution ranged from 255 to 615.1 nm, and the average diameter was 400.9 nm. Additionally, the distribution range of 1.0 mg/L stock solution was from 255 to 955.4 nm, and the average diameter was 480.4 nm. The result of DLS measurements was the hydrodynamic size of 13C-fullerenols which corresponded to the core and the swollen corona of the particles, where TEM only reflected the size of the core of the particles in a dried state where the corona with low electronic density was not visible. Consequently, the size of 13C-fullerenols measured by DLS was a little larger than that observed under TEM. These characterizations indicated 13C-labeled fullerenols were successfully prepared. Accumulation of 13C-Fullerenols in S. obliquus. To simulate the dietary route and compare it with the aqueous route, S. obliquus were incubated with 13C-fullerenols under the concentration conditions of the aqueous route (0.1 mg/L and 1.0 mg/L). According to the results of UV−vis spectrophotometry, the algae suspension had a maximum absorbance at around 685 nm. Figure S2 gives the absorbance of algae suspension at 685 nm of each group during the 9-day incubation. According to the significant difference analyses, the addition of fullerenols did not influence the growth of S. obliquus (p > 0.05). In the first day, a rapid increase in 13Cfullerenol concentration was observed in both the 0.1 mg/L and 1.0 mg/L group (Figure 2). After a slight increase, the concentration of 13C-fullerenols in the 0.1 mg/L group reached its maximum value of 1.95 mg 13C-fullerenols/g dry weight algae. The maximum concentration of 13C-fullerenols in S. obliquus exposed to 1.0 mg/L 13C-fullerenols was also attained at 2 d, reaching 26.7 mg 13C-fullerenols/g dry weight algae. The DLS analyses (Figure S1) revealed that the size distributions of fullerenols particles in the high and low concentration suspensions were approximately the same

Figure 3. Pathways of fullerenols uptake by D. magna. 12136

DOI: 10.1021/acs.est.8b03121 Environ. Sci. Technol. 2018, 52, 12133−12141

Article

Environmental Science & Technology

Figure 4. Body burden of 13C-fullerenols in D. magna by a) aqueous route, b) dietary route, and c) feeding-affected aqueous route. (The low concentration group is indicated by the right vertical axis, and the high concentration group is indicated by the left vertical axis.)

revealed that the body burdens after 24 h of uptake were much lower than those observed in the aqueous exposure experiments. Even though excess exposed algae were added in the dietary exposure experiments (green algae could be found in the bottom of the beakers throughout the uptake period), the amount of fullerenols around D. magna was less than that observed in the aqueous exposure experiments. Consequently, the fullerenols accumulation via food uptake was lower than that obtained from the aqueous surroundings. In the depuration period, the body burdens also decreased during the first 4 h in both groups. The residual body burden after 24 h of depuration was 0.018 ± 0.001 μg/4 D. magna in the 0.1 mg/L group and 0.24 ± 0.04 μg 13C-fullerenols/4 D. magna in the 1.0 mg/L group. The rate between body burden after 24 h of uptake and residual body burden after 24 h of depuration was 85.71% and 61.54% in the 0.1 mg/L and 1.0 mg/L groups, respectively. In the feeding-affected aqueous exposure experiments, the presence of algae did not change the integral tendency of body burden when compared to the aqueous route (Figure 4c). However, the introduction of algae did cause the stable phase to arrive slightly earlier, at about 8 h, while the body burdens were smaller than those in aqueous exposure experiments. Specifically, the body burdens at 24 h were 0.23 ± 0.02 μg/4 D. magna and 2.81 ± 0.17 μg/4 D. magna in the 0.1 mg/L and 1.0 mg/L groups. Decreased body burdens were also observed in the uptake of gold nanoparticles with the presence of food.58 The presence of algae facilitated purging or clearing of the gut, resulting in less body burden of nanoparticles. The TEM image also confirmed the absorption of 13C-fullerenols by D. magna. After 24 h of depuration, the residual body burden was 0.06 ± 0.01 μg/4 D. magna in the 0.1 mg/L group and 0.64 ± 0.14 μg/4 D. magna in the 1.0 mg/L group. The rate between body burden after 24 h uptake and residual body burden after 24 h depuration was 26.07% and 22.78% in the 0.1 mg/L group and 1.0 mg/L group, respectively. The experimental results show that uptake of 13C-fullerenols by D. magna directly from water (aqueous route) was larger

(Figure 4a). As exposure concentration increased, the body burdens increased remarkably. For the remainder of the uptake period, the body burdens were relatively stable, fluctuated slightly, and declined distinctly in the following depuration period. After 24 h of uptake, the body burdens were 0.27 ± 0.03 μg/4 D. magna and 9.16 ± 0.91 μg/4 D. magna in the 0.1 mg/L and 1.0 mg/L groups, respectively. In the depuration period, the body burdens both plummeted during the first 4 h in the two groups. In our previous study, we found that 13Cfullerenols were deposited in high volumes in the gut.42 According to reports, the gut retention time of Daphnia varied from 2 to 55 min.55−57 This may explain why 13C-fullerenols body burden decreased promptly in the early depuration period. As the depuration process was occurring, the body burden increased slightly. This phenomenon was also observed in experiments investigating depuration of gold nanoparticles in D. magna conducted by Skjolding et al.58 Specifically, they found that the body burden of gold nanoparticles plummeted in the first hour and then fluctuated during the ongoing depuration. Considering the freshwater was not renewed throughout the depuration period, the excreted fullerenols could be recaptured by D. magna as the depuration process was occurring. After 24 h of depuration, the residual body burden was 0.14 ± 0.03 μg/4 D. magna in the 0.1 mg/L group, which was about 51.85% of the body burden after 24 h of exposure. In the 1.0 mg/L group, the residual body burden after 24 h of depuration was about 55.02% of the body burden after 24 h exposure, reaching 5.04 ± 0.63 μg/4 D. magna. Different from the aqueous exposure experiments, the results of dietary exposure experiments showed that the body burdens of 13C-fullerenols continued growing throughout the 24-h uptake period, reaching 0.021 ± 0.001 μg/4 D. magna in the 0.1 mg/L group and 0.39 ± 0.02 μg/4 D. magna in the 1.0 mg/ L group (Figure 4b). The group fed with algae that had been incubated with 1.0 mg/L 13C-fullerenols showed an obviously larger body burden than that fed with 0.1 mg/L 13C-fullerenols. The 13C-fullerenols assimilated by intestinal wall cells of D. magna were also observed under TEM (Figure S3c), which 12137

DOI: 10.1021/acs.est.8b03121 Environ. Sci. Technol. 2018, 52, 12133−12141

Article

Environmental Science & Technology

Table 1. Modeled Depuration Rates with Corresponding R2, the Burden Mass of 13C-Fullerenols at the End of a 24 h Uptake Period, and Residual Mass after a 24 h Depuration in Clean Freshwatera uptake route aqueous route aqueous route dietary route dietary route feeding-affected aqueous route feeding-affected aqueous route

exposure concn 0.1 1.0 0.1 1.0 0.1

mg/L mg/L mg/L mg/L mg/L

1.0 mg/L

ke (h−1)

R2

(0.04; (0.16; (0.00; (0.00; (0.05;

0.26) 0.57) 0.14) 0.03) 0.78)

0.79 0.88 0.75 0.95 0.75

0.56 (0.23; 0.89)

0.84

0.15 0.37 0.05 0.01 0.42

burden mass (μg/4 D. magna) 0.27 9.16 0.021 0.39 0.23

± ± ± ± ±

0.03 0.91 0.001 0.02 0.02

2.81 ± 0.17

a

residual mass (μg/4 D. magna) 0.14 5.04 0.018 0.24 0.06

± ± ± ± ±

0.03 0.63 0.001 0.04 0.01

0.64 ± 0.14

The values of uptake (ku) and elimination rate (ke) were derived from the following equation: Ct =

Cwku (1 ke

BAFFit (L g−1)

ku (L g−1 h−1)

11.77 ± 1.36 39.04 ± 3.55

1.77 ± 0.20 14.45 ± 1.31

0.14 ± 0.01

0.06 ± 0.001

17.33 ± 1.41

9.70 ± 0.79

− e−ket ). Cw refers to source

concentration. The values in the parentheses denote the 95% confidence interval with upper and lower boundary.

3.55 to 17.33 ± 1.41Lg−1 in 1.0 mg/L exposure groups. The modeled uptake rates for the feeding-affected aqueous route were lower than those for the aqueous route, indicating that the presence of algae had a negative influence on the uptake of fullerenol nanoparticles by D. magna. Moreover, S. obliquus increased the elimination rate from 0.15 to 0.42 h−1 in the 0.1 mg/L group and from 0.37 to 0.56 h−1 in the 1.0 mg/L group. The elimination of fullerenols may increase with the metabolism of algal cells ingested by D. magna, which was confirmed by residual rates of body burdens after 24 h of depuration in the aqueous and feeding-affected aqueous groups. These values declined from 51.85% to 26.07% in the 0.1 mg/L groups and from 55.02% to 22.78% in the 1.0 mg/L groups. Based on the experimental data and the kinetic analyses above, fullerenols could be ingested directly from the aqueous environment in large volumes. Specifically, algae consumed fullerenols from the aqueous environment, which reduced the amount surrounding D. magna and therefore decreased their direct uptake from the aqueous environment. On the other hand, because D. magna are filter feeders, the algae competed with D. magna for ingestion of fullerenol particles. Even though fullerenols that accumulated in algae could be absorbed via predation of algae by D. magna, the amount was too small to overcome the negative effects of its presence in the aqueous environment. Thus, D. magna primarily took up 13C-fullerenols via the aqueous route. Upon evaluation of the uptake of unmodified graphene by freshwater snails, the association with algae enhanced its ingestion, resulting in absorption from algae (dietary route) being the main pathway,61 with the algae acting as a dispersant/carrier of the unmodified graphene and enhancing the interaction with involved organisms. The effects of improved dispersibility were also demonstrated by the addition of natural organic matter (NOM), which improved the dispersibility of graphene and increased its accumulation in zebrafish gut.62 However, the presence of algae had no evident effects on the dispersion or uptake of Zn from ZnO nanoparticles by freshwater snails relative to the uptake of Zn from contaminated food.19 Consequently, when evaluating the risk of a certain artificial nanomaterial in ecosystems, the surface modification determines its interactions with environmental organisms and its fate in the ecosystem. For nanomaterials with poor dispersibility, food or other matter may act as a dispersant to enhance uptake by the target organism, thereby increasing the environmental exposure hazard. While for other nanomaterials, their ingestion by involved organisms may compete with the

than that from predation on S. obliquus (dietary route) that had been incubated with 13C-fullerenols at the same concentration as in the aqueous exposure experiments. The presence of algae in water negatively influenced body burden and accelerated the arrival of the stable phase. For the filter feeder D. magna, the ingestion of materials dispersed in water (including algal cells and other particles) mainly occurred via body absorption and feeding.59 The presence of S. obliquus may compete with fullerenol nanoparticles, thereby reducing ingestion of the nanoparticles by D. magna. Moreover, the lower residues after 24 h of depuration may have occurred because selective assimilation of algae reduced the retention time of fullerenols in the D. magna gut. Similarly, D. magna have been found to selectively absorb lipids from lipid coated carbon nanotubes.60 The role of algae in our experiments was completely different from that observed in experiments conducted by Mao and co-workers.61 Specifically, they found that unmodified graphene could associate with algal cells rather than forming agglomerates. The association with algal cells facilitated the ingestion of graphene along with a freshwater snail’s predation on algae, thereby changing the fate of graphene in the ecosystem and promoting its bioaccumulation in involved organisms. Routes of 13C-Fullerenols Uptake by D. magna. To evaluate the contributions of aqueous and dietary routes, the body burdens in each experiment during the uptake period were fitted using first-order kinetics. As shown in Figure 4 and Table 1, the maximum accumulation of 13C-fullerenols in D. magna occurred at about 16 h in both the 0.1 mg/L and 1.0 mg/L group in the aqueous exposure experiments and showed steady state balance within 24 h only in the aqueous and feeding-affected aqueous routes. Additionally, after 24 h exposure, the body burdens were 0.27 ± 0.03 μg/4 D. magna and 9.16 ± 0.91 μg/4 D. magna in the 0.1 mg/L and 1.0 mg/L groups, respectively. However, in the dietary exposure experiments, the body burdens both continued to grow in the 0.1 mg/L and 1.0 mg/L groups, at 24 h experiment exposure, only reaching body burdens of 0.021 ± 0.001 μg/4 D. magna and 0.39 ± 0.02 μg/4 D. magna, respectively. These results and trend showed that many more fullerenol particles were ingested by the aqueous route than the dietary route during 24 h exposure. The influence of feeding should also be evaluated with the simultaneous presence of algae in an actual aquatic ecosystem. As shown in Table 1, from the aqueous to feeding-affected aqueous, the kinetic analyses revealed that the bioaccumulation factors (BAFs) decreased from 11.77 ± 1.36 to 0.14 ± 0.01 Lg1− in 0.1 mg/L exposure groups and 39.04 ± 12138

DOI: 10.1021/acs.est.8b03121 Environ. Sci. Technol. 2018, 52, 12133−12141

Article

Environmental Science & Technology Table 2. Transfer of 13C-Fullerenols through the S. obliquus-D. magna Food Chain concn in freshwater

concn in S. obliquus

concn in D. magna at 24 h

BMF at 24 h

fitted plateau in D. magna

fitted BMF

0.1 mg/L 1.0 mg/L

1.08 ± 0.10 mg/g 8.23 ± 0.02 mg/g

0.09 ± 0.001 mg/g 1.66 ± 0.15 mg/g

0.084 ± 0.001 0.20 ± 0.02

0.117 mg/g 15.753 mg/g

0.11 1.91

above highlight that the exposure concentration can directly influence the trophic biomagnification of nanomaterials. A high exposure concentration can induce the biomagnification of nanomaterial through the food chain, demonstrating that the concentration factor and exposure duration are very important when evaluating the biorisk posed by nanomaterials. During 24 h dietary exposure experiments, the bioaccumulation of fullerenols via the dietary route had not arrived at the plateau concentration, and we speculated the plateau concentrations based on first-order kinetic model analyses. As shown in Table 2, the plateau concentrations were 0.117 and 15.753 mg fullerenols/g dw D. magna for the 0.1 and 1.0 mg/L groups, and the half concentration times were 15.33 and 98.81 h, respectively. The reduced half concentration time of the low exposure group indicated that, under a low dietary exposure concentration, the stable accumulation amount will arrive early. These results further indicated that the dose−effect during dietary exposure should not be ignored. The fitted BMF value in the 0.1 mg/group was 1 (1.91), which implied that a relatively high exposure concentration and long-term exposure will increase the potential for transfer from S. obliquus to D. magna via the food chain. Kinetic analyses showed that the bioaccumulation of fullerenols may increase as the trophic level rises when the exposure concentration increases. Nanoparticles that bioaccumulate in organisms at one trophic level can transfer into organisms at the next trophic level in a way that is dependent on the delivery scenario, after which they can induce a biological response and generational transfer. For example, we previously showed that D. magna may transfer fullerenols to their offspring.42 Based on these findings, a longer and more comprehensive dietaryexposed experiment is warranted. It should be noted that, in the actual experiments, the concentration of nanomaterials directly affected not only the dose−effect of the toxicity study but also their uptake and transfer efficiency via the food chain, demonstrating a synergistic environmental effect.

ingestion of food and surrounding matter, and the presence of the food and matter may decrease the uptake of nanomaterials. For instance, the interaction between graphene oxide and soil constrained the transfer and dispersion of the nanomaterial in the cultivation medium, which reduced the environmental exposure hazard.63 Obviously, the feeding methods of D. magna and freshwater snails also influenced the bioaccumulations of the nanomaterials. Similarly, Sanchiś et al. investigated dietary exposure by feeding the filter feeder M. galloprovincialis algae that had been incubated with fullerene to simulate the potential uptake pathway of fullerene and investigate the ensuing metabolic responses of M. galloprovincialis.64 Taken together, these results indicate that when investigating the bioeffects of artificial nanomaterials on an environmental organism, it is important to pay close attention to its chemical and physical properties, which determine its fate in the environment, as well as the life habit of the target organism, which determines the bioaccumulation of nanomaterial by the organism. Transfer of 13C-Fullerenols through the S. obliquus-D. magna Food Chain. The biomagnification factor (BMF) is a crucial factor for estimation of the transfer potential of exogenous or toxic matter through food chains. In the dietary exposure experiments, S. obliquus and D. magna formed a twolevel food chain. At an aquatic 13C-fullerenols concentration of 0.1 mg/L, the concentrations of fullerenols in S. obliquus and D. magna were 1.08 ± 0.10 and 0.09 ± 0.001 mg/g, respectively (Table 2), while these values increased to 8.23 ± 0.02 mg/g and 1.66 ± 0.15 mg/g as the initial concentration increased to 1.0 mg/L. The BMF24 was 0.084 ± 0.001 in the low exposure group, while it was 0.20 ± 0.02 in the 1.0 mg/L group. This increase indicated that the transfer efficiency of fullerenols along the S. obliquus-D. magna food chain increased with the increasing exposure concentration. The levels of BMF24 in the 0.1 and 1.0 mg/L group were both