- Email: [email protected]
Disruption of zebrafish (Danio rerio) reproduction upon chronic exposure to TiO2 nanoparticles
Chemosphere 83 (2011) 461–467
Contents lists available at ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
Disruption of zebrafish (Danio rerio) reproduction upon chronic exposure to TiO2 nanoparticles Jiangxin Wang a, Xiaoshan Zhu b,c, Xuezhi Zhang b, Zheng Zhao d, Huan Liu d, Rajani George e, Jeanne Wilson-Rawls e, Yung Chang a,⇑, Yongsheng Chen b,f,⇑ a
Center for Infectious Diseases and Vaccinology, The Biodesign Institute, Arizona State University, Tempe, AZ, USA The School of Sustainable Engineering and the Built Environment, Arizona State University, Tempe, AZ, USA Graduate School at Shenzhen, Tsinghua University, Shenzhen, PR China d School of Computing, Informatics, and Decision Systems Engineering, Ira A. Fulton School of Engineering, Arizona State University, Tempe, USA e Genomics, Evolution and Bioinformatics, School of Life Sciences, Arizona State University, Tempe, AZ, USA f School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, GA, USA b c
a r t i c l e
i n f o
Article history: Received 10 October 2010 Received in revised form 8 December 2010 Accepted 13 December 2010 Available online 15 January 2011 Keywords: Acute toxicity Bioaccumulation Chronic toxicity Microarry Reproduction Zebrafish TiO2 Nanoparticles
a b s t r a c t As common engineered nanomaterials, TiO2 nanoparticles (nTiO2) are usually perceived as non-toxic, and have already been widely used in many products and applications. Such a perception might have been shaped by some short-term studies that revealed no/low toxicity of nTiO2 to cells and eco-relevant organisms. However, given the ultimate release of nTiO2 into the aquatic environment, which can act as a sink for engineered nanoparticles, their long-term impact on the environment and human health is still a concern and deserves more research efforts. Here, for the first time, we demonstrate that chronic exposure of zebrafish to 0.1 mg L 1 nTiO2, can significantly impair zebrafish reproduction. For instance, there was a 29.5% reduction in the cumulative number of zebrafish eggs after 13 weeks of nTiO2 exposure. Thus, we provided timely information on indicating a serious risk of reproductive impairment of environments contaminated with low levels of nTiO2 on aquatic organisms, leading to alterations in population dynamics and aquatic ecosystem balance, and thus warrants a careful scrutiny on toxicity assessment of nTiO2, especially their long-term impact. Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction Thus far, more than 1000 products with nanomaterials are already on the market (Nanovip, 2010). Revenues from nanotechnology-based products are expected to grow worldwide to 3.1 trillion US$ by 2015 (Industrial week; Jonathan, 2008). The large-scale production and utilization of engineered nanomaterials (ENMs) have raised concerns over the release of ENMs to the environment and its potential impact to human and environmental health. Nanoscale titanium dioxide (nTiO2) is widely used in many commercial products, including sunscreens, cosmetics, paints, surface coatings, and food additives (Fisher and Egerton, 2001; Kaida et al., 2004). It is also used as a photocatalyst for environmental decontamination of air, soil, and water (Esterkin et al., 2005; Choi et al., 2006). Such wide application of nTiO2 might have been influenced under the assumption that nTiO2 is somewhat safe ⇑ Corresponding authors. Address: School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, GA, USA. Tel.: +1 404 894 3089; fax: +1 404 894 2278 (Y. Chen), Tel.: +1 480 965 8672; fax: +1 480 727 0599 (Y. Chang). E-mail addresses: [email protected] (Y. Chang), [email protected]. edu (Y. Chen). 0045-6535/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2010.12.069
as several acute and short-term ecotoxicity studies showed no/low toxicity of nTiO2 exerted to testing organisms, ranging from bacteria, algae, Daphnia to fish (Heinlaan et al., 2008; Hund-Rinke and Simon, 2008; Griffitt et al., 2008, 2009; Velzeboer et al., 2008; Zhu et al., 2008, 2009a; Wiench et al., 2009). For instance, Heinlaan et al. (2008) found that neither nTiO2 nor bulk TiO2, were toxic to bacteria or crustaceans. Both Wiench et al. (2009) and Zhu et al. (2009a) reported EC50 > 100 mg L 1 in Daphnia over 48 h exposure to nano-scale and non-nano-scale TiO2. Furthermore, zebrafish embryos or adult fish displayed no apparent abnormality 2 days after being exposed to 10 mg L 1 or 500 mg L 1 TiO2, respectively (Griffitt et al., 2008; Zhu et al., 2008). However, some recent studies, including ours, revealed that an extended exposure up to 21 days did cause some mortality, reproductive impairment and tissue damage in both Daphnia and fish (Federici et al., 2007; Wiench et al., 2009; Zhu et al., 2010a). These studies indicate that chronic exposure may be critical in assessing potential toxicity of nTiO2. Given the wide applications of nTiO2 and its potential for release into the aquatic environment, exposure of aquatic organisms to nTiO2 would be long-term. Thus, the adverse impact of chronic exposure to nTiO2 is worth to be carefully examined, especially at the dose range that is low and considered non-toxic.
462
J. Wang et al. / Chemosphere 83 (2011) 461–467
To further address this issue, we conducted a chronic toxicity study with prolonged zebrafish exposure (91 days) to nTiO2 to systemically evaluate potential impact of such exposure on the zebrafish reproductive system, including egg production; embryo survival; ovary histology, nTiO2 accumulation in ovaries; and global gene expression in ovarian tissues by microarray analyses. We found that zebrafish reproduction could be impaired by nTiO2 exposure at as low as 0.1 mg L 1, which is an environmentally relevant dose range. Our findings indicate that long-term impact of engineered nanomaterials on the aquatic ecosystem should be taken into consideration in advancing nanotechnology.
2. Materials and methods 2.1. Nanoparticle preparation and characterization All chemicals used, including uncoated and powdered nanoscale anatase TiO2 (nTiO2), were obtained from Sigma Aldrich (99+% purity). All samples were prepared in Mili-Q water purified using a Milli-Q plus water purification system (Millipore, Bedford, MA). A stock solution of 10.0 g L 1 nTiO2 was prepared by dispersing the nanoparticles in DI water (resistivity: 18.2 MX cm 1 at 25 °C) with sonication for 10 min (50 W L 1 at 40 kHz). In order to get working concentrations of 0.1 and 1.0 mg L 1 nTiO2 in the fish tanks, each tank was dosed with 10.0 or 100 ll of the 10.0 g L 1 stock solution, respectively. The test solution was changed and re-dosed every 24 h with new aliquot of nTiO2 to maintain the exposure at relatively consistent levels of nTiO2. The actual concentration of nTiO2 in the fish solution over 24 h was tested using inductively coupled plasma mass spectrometry (ICP-MS). The size distribution of nTiO2 in the solution was determined using a dynamic light scattering device (DLS, Brookhaven Instrument Corporation, Holtsville, NY, USA). In this case, three replicate water samples were obtained from three different tanks (sample num-
ber = 3 3) in each group at each time point post-nTiO2 exposure. Specifically, 1 mL solutions from the middle area of different tanks were collected for immediate DLS analysis, which was repeated several times over the 24 h of nTiO2 exposure. The size distribution (median size) of nTiO2 in the test solution is shown in the insert of the Fig. 1A. The particle sizes of nTiO2 in the test solutions were also verified by scanning electron microscope (SEM) with energy dispersive X-ray (EDX) analysis. For SEM test, 1 mL solutions from three different tanks in each group were collected at different time points post-nTiO2 exposure, air dried on pre-cleaned glass slides for standard SEM observation. The representative images are presented in Fig. 1B. For nTiO2 concentration analysis in culture media, 10 mL nTiO2 suspensions at different time points were transferred to triangular flasks and evaporated to dryness. nTiO2 was decomposed into Ti4+ ion by heating with 2 mL of a sulphuric acid–ammonium sulphate solution (400 g ammonium sulphate in 700 mL concentrated sulphuric acid, boiled) until the pellets were completely dissolved. After cooling down, the above solutions were transferred to 10 mL volumetric flasks. Released Ti4+ was determined using ICPMS (Element2, Thermo Fisher, USA). Using this procedure, concentrations of nTiO2 in fish medium were calculated from the corresponding Ti4+ concentrations measured, which were slightly deviated from the initial concentration obtained by a dilution of the nTiO2 stock solution. For the assessment of nTiO2 bioaccumulation in the fish ovary, freshly obtained tissues were completely digested in 5 mL of sulphuric acid–ammonium sulphate solutions to decompose TiO2 into Ti4+ using a microwave digestion procedure. Briefly, samples were digested by a microwave digestion system (Mars 5, CEM, USA) with a four-stage digestion protocol (5 min at 150 °C, 5 min at 190 °C, 10 min at 230 °C and 10 min at 240 °C). After digestion, samples were evaporated to dryness. Then nTiO2 was converted to Ti4+ using sulphuric acid-ammonium sulphate, and then Ti4+ was analyzed by ICP-MS. In this study, nTiO2 recovery in biological samples ranged from 80% to 90%.
2.2. Zebrafish exposure experiment
Fig. 1. nTiO2 characterization in solution within 24 h. (A) The actual concentration of nTiO2 in the fish solution over 24 h. The insert indicates the size distribution (median size) of nTiO2 in the test solution during the same time period. (B) Representative scanning electron microscope (SEM) images for nTiO2 aggregates in the test solution at 24 h. Scale bar: 1 lm. The inserts indicate the sizes of one aggregate with a scale bar of 200 nm. The nTiO2 particles were confirmed with energy dispersive X-ray (EDX) spectroscopy (data not shown).
Zebrafish were maintained in a fish room at The Biodesign Institute of Arizona State University following regulations. Adult zebrafish (more than 3 months old; 0.33 ± 0.09 g) were used for this study. Prior to nTiO2 treatment, fish were acclimated to experimental conditions (28 ± 0.5 °C; light:dark/14:10 h; daily water change), including daily manipulation and weekly breeding for at least 4 weeks. Throughout the experiment, including the breeding period, fish were kept in fish tanks or breeding tanks containing 1 L standard culture medium (containing 64.75 mg L 1 NaHCO3, 5.75 mg L 1 KCl, 123.25 mg L 1 MgSO47H2O, and 294 mg L 1 CaCl22H2O and prepared according to ISO standard 7346-3:1996 and OECD Guideline 202), and the health status of the fish was monitored and recorded daily. All fish were fed daily with TetraMin dry flakes (Tetra, Melle, Germany) containing no nTiO2. As illustrated in Figure S1, after the acclimation period, reproductively active fish (12 females in two tanks and 12 males in two tanks) were exposed to one of the following treatments for 13 weeks using a semi-static exposure regime (water change every 24 h with a new aliquot of nTiO2): control (culture medium only), 0.1, or 1.0 mg L 1 nTiO2. Each week, one male and one female from the same group were set up in a breeding tank (one pair of fish in one breeding tank; twelve pairs for one treatment group); spawning was triggered when the light was turned on in the morning and completed within 30 min. Numbers of spawning fish and eggs from individual breeding tanks were then recorded. The water qualities
J. Wang et al. / Chemosphere 83 (2011) 461–467
such as pH, temperature, hardness and dissolved oxygen concentration were monitored throughout the experiments.
463
data sources to comprehensively evaluate the biological relevance of the genes, which is further explained in the Supporting Information.
2.3. Histological examination Histological examination was performed following protocols described previously (Wilson-Rawls et al., 1999). Briefly, freshly isolated ovary tissues were fixed in newly made 4% paraformaldehyde in PBS overnight at 4 °C. The tissue was embedded in paraffin after a stepwise dehydration in ethanol and xylene, and then 5 lm transverse sections were prepared, deparaffinized in xylene, and stained with hematoxylin and eosin. The sections were collected for further analyses from the tissue 1 mm below the ovary surface. Eight to ten sections of each ovary were examined, and more than 100 follicles from 1 to 3 sections of each ovary (total five ovaries for each group) were counted and assigned to their appropriate stages based on the size and morphology as defined by Selman et al. (1993). The distribution of stages I, II, III and IV in each ovary was calculated as the percentage of total follicles counted, and the averages of different stages of follicles among five individual ovaries of each experimental group are presented. 2.4. RNA extraction and qRT-PCR Fish tissues were removed for total RNA extraction and subsequent reverse transcription, as detailed in the Supporting Information. To monitor the expression of selected genes, we applied realtime quantitative PCR (qRT-PCR) to analyze individually prepared ovary and liver samples using SYBR green procedure (Zhu et al., 2009b) on an ABI-7000 system with primers listed in Table S1 of the Supporting Information. 2.5. Zebrafish microarray analyses and bioinformatics Global gene expressions were analyzed with the Agilent 4 44 k zebrafish microarray (Santa Clara, CA) using a two-color reference design. Quality and concentration of RNA samples were evaluated using a 2100 Bioanalyzer (Agilent, USA) and Nanodrop ND-1000 (Thermo Fisher Scientific, USA). In this study, RNA samples with RNA integrity number higher than 9.0 (10 being intact and 1 being totally degraded) were chosen for microarray and qRT-PCR analyses. The control RNA was constructed by pooling equal amounts of RNA from four individual control ovary samples. For treated groups, both pooled (equal amounts of RNA, 250 ng each, from four individual treated ovary samples) and two individual ovary RNA samples were submitted to the microarray analysis. cRNA synthesis and reverse transcription were conducted using a two-color Amplification Kit and Gene Expression Hybridization Kit (Agilent, USA), according to the manufacturer’s protocols. The labeled samples were hybridized to the arrays for 17 h at 65 °C, washed, and immediately scanned using a G2505 B Microarray Scanner (Agilent, USA) and raw expression data were extracted using Feature Extraction 9.5 (Agilent, USA). To identify genes differentially expressed between exposed and control zebrafish, p-values of Student’s 1-sample t-test and mean fold changes between the exposed and control samples were calculated. A threshold based on both p-value and fold change (p < 0.05 and fold change > 2) was used to select genes that responded to the exposure, either up-regulated or down-regulated by nTiO2 treatments. The genes with consistent expression patterns under 0.1 and 1.0 mg L 1 nTiO2 treatments were selected, and the Hierarchical clustering analysis was employed to search for common gene expression patterns. To extract multiple lines of information that arise in the context with simultaneous observation of more than 40,000 probes in the microarray, our integrative framework (Zhao et al., 2010) was applied to fuse different types of knowledge and
2.6. Data analysis The reproduction experiment was conducted with 12 pairs of adult fish in each group. Values for the average number of egg per spawned fish, histological analysis of folliculogenesis, nTiO2 accumulation in ovary tissues, and qRT-PCR analyses were evaluated in at least five replicates for statistical evaluation of the differences between the treatment and control group. Data are represented as the mean ± SD. Differences among means were determined using one-way analysis of variance (ANOVA) for all parameters. Results were considered to be statically significant if p < 0.05. 3. Results and discussion 3.1. Nanoparticle characterization nTiO2 in our tests were found to aggregate in culture medium, consistent with our previous observation of metal oxide nanoparticles (Zhu et al., 2009b). The quantitative analyses were used to determine the level and kinetics of the nTiO2 aggregation process. To assess the stability of nTiO2 in fish water, we examined the concentration and particle size of nTiO2 during the 24 h of exposure time. At 0.1 mg L 1, the concentration of nTiO2 remained relatively constant over 24 h. However, at 1.0 mg L 1, the concentration dropped to 0.6 mg L 1 over the first hour and then remained unchanged for the rest of time (Fig. 1A). The drop in concentration might have been caused by aggregation of the nanoparticles, which is concentration dependent, as reported in our previous finding (Zhu et al., 2009b). Using DLS, the median values of particle size were detected with mean sizes of 240–280 nm (0.1 mg L 1) and 259–360 nm (1.0 mg L 1) at 0, 1, 3, 6, 12 and 24 h. Observed by SEM, these nTiO2 aggregates had variable sizes from a few hundred nanometers to several microns in diameter (Fig. 1B), which is in line with our previous reports (Zhu et al., 2009b). 3.2. Impairment of zebrafish reproduction and embryo survival by chronic exposure to nTiO2 We set up a long-term nTiO2 exposure to examine its effect on zebrafish reproduction. Specifically, zebrafish were exposed to 0.1 mg L 1 or 1.0 mg L 1 of nTiO2 daily for 13 weeks. Given the possible fluctuation of nTiO2 over time in both concentrations and particle sizes, we transferred zebrafish every 24 h to a tank containing a new aliquot of nTiO2 at appropriate concentrations (i.e., 0.1 mg L 1 or 1.0 mg L 1) to ensure a relatively consistent nTiO2 exposure for these fish. The zebrafish reproduction was evaluated weekly by assessing the number of spawned females, egg production, and embryo survival rates. Daily monitoring of the mortality and general health of the fish revealed no apparent abnormalities. However, the total cumulative number of eggs and the number of eggs per spawned female was gradually reduced beginning at week 9 in groups treated with nTiO2 although the number of spawning females was comparable between control and treated groups (Fig. 2A and B). The number of spawned females each week for the control group (open circles) was 9, 9, 10, 10, 11, 12, 11, and 10 from week 6 to week 13, respectively, whereas the numbers for the 0.1 mg L 1 group (cross) were 10, 10, 8, 11, 10, 5, 4, and 9 during the same weeks. Significant reduction in embryo survival was observed in the nTiO2-treated group at 8, 10, 11, 12 and 13 weeks (p < 0.05).
464
J. Wang et al. / Chemosphere 83 (2011) 461–467
Fig. 2. Disruption of zebrafish reproduction upon long-term exposure to nTiO2 and the embryo survival rate of the F1 generation from the control and 0.1 mg L 1 nTiO2treated groups 6–13 weeks post-nTiO2 treatment. (A) A representative experiment shows the reduction in the cumulative number of zebrafish eggs after long-term nTiO2 exposure. (B) The inhibitory effect of nTiO2 treatment is evident from the reduction in the average numbers of eggs produced per spawning female (p < 0.05, n = 12). (C) The data points indicate the survival rate, at 48 hpf, of embryos derived from individual spawned females.
As a result, even at a concentration as low as 0.1 mg L 1, there was a 29.5% reduction in the cumulative number of zebrafish eggs (the eggs were collected from all spawned fish of each experimental group and the total egg numbers were tabulated for each group) over 13 weeks of nTiO2 exposure, as compared to the control group (Fig. 2A). The effect of nTiO2 exposure on embryo survival was also monitored. The embryos spawned from the females exposed to 0.1 mg L 1 nTiO2 for 8 weeks or longer showed an increase in mortality at 48 h post-fertilization (hpf), as compared to the control group. A significant increase in embryo mortality was observed in groups with 8, 10, 11, 12, 13 weeks of exposure (Fig. 2C). Our findings reveal that zebrafish reproduction and embryo survival could be negatively affected by a prolonged exposure to nTiO2 at concentrations as low as 0.1 mg L 1, close to the ENM level disseminated into the environment (Zhang et al., 2008; Kiser et al., 2009). 3.3. Defect in folliculogenesis The reduced egg production may reflect a defect in folliculogenesis. To test this hypothesis, the ovaries from the control and nTiO2-treated fish were sectioned and stained for histology examination to characterize the developmental stages (i.e., I–IV) of the follicles therein according to Selman’s classification (Selman et al., 1993). As compared to the control group, the distribution of follicular developmental stages was skewed by the nTiO2
treatment toward the immature stage, as evident by an increase in stage I follicles and some reduction in other stages, especially, stage IV follicles (Fig. 3A and B). This indicates a developmental block at the transition from stage I to stage II and possibly at follicular maturation to stage IV as well (Selman et al., 1993). This block may be attributable to a direct effect of nTiO2 on developing follicles or to an indirect disruption of the normal pituitary-gonad endocrine system. 3.4. nTiO2 accumulation in ovaries To determine whether the ovary is a direct target of nTiO2, we examined nTiO2 accumulation in fish ovaries using ICP-MS. Fish ovaries were found to accumulate nTiO2 at concentrations of approximately 2.5 mg kg 1 and 7.2 mg kg 1 under 0.1 mg L 1 and 1.0 mg L 1 doses, respectively (Fig. 3C). This finding suggests that nTiO2 can enter the ovary, presumably via blood circulation, and may directly act on this organ to disrupt oocyte development. 3.5. Effect of nTiO2 on gene expression Given the apparent disruption in oocyte maturation in the females exposed to nTiO2, genes involved in oocyte development might have been altered by nTiO2 treatment. To test this scenario, we performed expression microarray analyses of ovarian tissues from both control and nTiO2-treated groups to identify genes
J. Wang et al. / Chemosphere 83 (2011) 461–467
465
Fig. 3. Histological changes and nTiO2 accumulation in zebrafish ovary afterlong-term nTiO2 exposure. (A) Representative sections of ovaries taken from control (left), 0.1 mg L 1 (middle) and 1.0 mg L 1 nTiO2-treated (right) zebrafish females. Scale bar, 200 lm. (B) The average percentage of follicles present at different developmental stages (p < 0.05, n = 5). (C) nTiO2 accumulation in individual ovaries from females in the control, 0.1 mg L 1, or 1.0 mg L 1 nTiO2-treated groups is presented as mg kg 1 (p < 0.05, n = 4).
associated with the disrupted folliculogenesis in the nTiO2-treated groups. The array analyses revealed a significant alteration caused by long-term nTiO2 exposures, even at a low concentration as low as 0.1 mg L 1 (Fig. 4). For instance, total 2383 and 1043 genes were found to be down-regulated in 0.1 and 1.0 mg L 1 treated groups, respectively, while 2069 and 471 genes were up-regulated genes detected in 0.1 and 1.0 mg L 1 exposed fish, respectively. Among these differentially expressed genes, 405 up-regulated and 174 down-regulated genes were shared in both 0.1 and 1.0 mg L 1 treated groups. Table S2 lists a subset of the genes altered by nTiO2 at both 0.1 mg L 1 and 1.0 mg L 1. Pathway and gene function analysis showed that some of these genes are involved in proteolysis, oxidative stress regulation, metabolism, insulin signaling, and apoptosis, as well as oocyte maturation, suggesting multiple and complicated modes of action for nTiO2-mediated disruption of reproduction. However, these differentially expressed genes have little overlap with those derived from microarray analyses of other toxicants
Fig. 4. Venn diagram analyses of genes identified by microarray analysis. Number of genes significantly differentially expressed in the ovary following zebrafish exposure to nTiO2 at different doses. (A) Down-regulated genes. (B) Up-regulated genes.
and chemicals, such as 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) (Heiden et al., 2008) and estradiol (E2) dependent treatments (Levi et al., 2009), or with the gene sets differentially expressed in gills after short-term exposure to different ENMs (Griffitt et al., 2009). Thus, although nTiO2, TCDD and E2 all cause impaired oocyte maturation, the mode of action mediated by nTiO2 likely differs from those by hormone disrupting compounds (Heiden et al., 2008; Levi et al., 2009). Our finding of a significant differential gene expression profile even at as low as 0.1 mg L 1 of nTiO2 reveals that the maturation and functionality of zebrafish ovary could be affected by chronic exposure to a very low dose of nTiO2 (Table S2, 0.1 mg L 1). Thus, we focused on this experimental group, i.e., treated with 0.1 mg L 1 of nTiO2, for further characterization. Three genes that control oocyte maturation, cyp11a, tgfb1 and egf, were found significantly decreased upon chronic exposure to nTiO2 (Table S2). The reduction of these genes was confirmed by qRT-PCR (Fig. 5A, Supporting information, and Table S1). These genes are expressed primarily in stage I/II follicles and promote stage I/II growth and maturation (Kohli et al., 2003; Wang and Ge, 2004; Ge, 2005; Clelland et al., 2006). Despite an elevated number of morphologically stage I follicles (Fig. 3B), the reduction in these genes implies that nTiO2 may inhibit the growth and maturation of stage I follicles (Kohli et al., 2003), presumably by altering the expression of several regulators that are critical to this stage of folliculogenesis. The growth factors coded by tgfb1 and egf have been implicated as paracrine factors that promote growth and differentiation of stage I follicles. cyp11a encodes an enzyme that converts cholesterol into testosterone, the rate limiting control step of E2 synthesis in the follicle, which in turn controls the expression of vitellogenin in the liver to promote follicular
466
J. Wang et al. / Chemosphere 83 (2011) 461–467
Fig. 5. Expression analysis by qRT-PCR on selected genes. (A) Expression of genes associated with oocyte development (cyp11a1, tgfb1 and egf) in ovaries from the control and 0.1 mg L 1 nTiO2 treated groups ( represents p < 0.05, n = 8). (B) vtg1 expression in livers from the control and 0.1 mg L 1 nTiO2 treated groups ( represents p < 0.05, n = 6). Actin was used as a housekeeping gene. The data in both (A) and (B) were obtained from fish samples in two independent experiments.
vitellogenesis and maturation, according to the hypothalamuspituitary-gonad (HPG) regulatory circuit in teleosts (Ings and Van der Kraak, 2006; Clelland and Peng, 2009). On the other hand, several up-regulated genes are associated with protein degradation or reactive oxygen species (ROS) production (Table S2), which may reflect stress responses of ovaries that are dealing with nTiO2 particles, e.g., trying to remove foreign entities. Based on the analyses of the differentially expressed genes, we argue that nTiO2 may affect the ovary directly, which is consistent with our observation of nTiO2 accumulation in this tissue (Fig. 3C). nTiO2 may cause oxidative stress to the developing follicles and interfere with the expression of key regulatory genes, ultimately impairing the growth and function of stage I/II follicles. If so, the aborted follicle development, as well as the possible inhibition of E2 synthesis as a result of cyp11a down-regulation, could impinge on vitellogenin synthesis in the liver, which in turn alters the maturation of vitellogenic follicles (i.e., stages III and IV) (Clelland and Peng, 2009). To test this scenario, we analyzed the expression of vtg1, one of the vitellogenin genes, in the livers of zebrafish treated with nTiO2 and control fish. As shown in Fig. 5B, vtg1 expression was found significantly reduced in the livers of nTiO2-treated fish, consistent with the abnormality noted in the folliculogenesis of these fish (Fig. 3B). Taken together, our data, which are derived from zebrafish egg production, ovary histology, nTiO2 accumulation and altered gene expression, reveals that long-term exposure to low concentrations of nTiO2 is toxic to the zebrafish reproductive system. Thus, the ultimate consequence of chronic exposure to nTiO2 nanoparticles is similar to the one caused by many reproductive toxicants, such as endocrine disrupting chemicals (EDCs). However, unlike EDCs, which usually perturb fish reproduction by interfering with hormone regulation, nTiO2 may act on primary follicles directly and/ or subsequently inhibit interfere with vitellogenin synthesis. Regardless of the mode of action, our current findings along with our previous study on Daphnia (Zhu et al., 2010a) and recent demonstration of nTiO2 transfer from Daphnia to zebrafish in a simplified freshwater food chain (Zhu et al., 2010b) indicate that chronic exposure to low doses of nTiO2 could alter the reproduction of certain aquatic organisms and ultimately cause perturbation in the population dynamic of these organisms in aquatic environments. Our studies highlight the importance of long-term studies in assessing the negative impact of nTiO2 at concentrations relevant or close to the levels of nTiO2 released into the environment, which will provide critical information for risk assessment and hazard management of environmental contamination by nTiO2 and other ENMs.
Acknowledgements The authors would like to thank T. Fu (Biodesign Inst. at Arizona State University) for her critical technical assistance on fish maintenance, treatments and data collection; S. Bingham and J. Hock (DNA Lab., Arizona State University) for their assistance on microarray experiments and the data extraction process, and P. Prapaopong (Chemistry Dept., Arizona State University) for assistance on fish tissue HNO3 microwave digestion and ICP-MS experiments. This study was supported by the U.S. Environmental Protection Agency Science to Achieve Results Program (Grant # RD831713) (to Y. Chen and Y. Chang) and National Science Foundation Grant (0812551) (to H. Liu). Article contents are sole responsibility of the authors and do not represent official views of the sponsors. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.chemosphere.2010.12.069. References Choi, H., Stathatos, E., Dionysiou, D.D., 2006. Sol–gel preparation of mesoporous photocatalytic TiO2 films and TiO2/Al2O3 composite membranes for environmental applications. Appl. Catal. B – Environ. 63, 60–67. Clelland, E., Kohli, G., Campbell, R.K., Sharma, S., Shimasaki, S., Peng, C., 2006. Bone morphogenetic protein-15 in the zebrafish ovary: complementary deoxyribonucleic acid cloning, genomic organization, tissue distribution, and role in oocyte maturation. Endocrinology 147, 201–209. Clelland, E., Peng, C., 2009. Endocrine/paracrine control of zebrafish ovarian development. Mol. Cell. Endocrinol. 312, 42–52. Esterkin, C.R., Negro, A.C., Alfano, O.M., Cassano, A.E., 2005. Air pollution remediation in a fixed bed photocatalytic reactor coated with TiO2. AIChE. J. 51, 2298–2310. Federici, G., Shaw, B.J., Handy, R.D., 2007. Toxicity of titanium dioxide nanoparticles to rainbow trout (Oncorhynchus mykiss): gill injury, oxidative stress, and other physiological effects. Aquat. Toxicol. 84, 415–430. Fisher, J., Egerton, T., 2001. Titanium Compounds, Inorganic. Kirk-Othmer Encyclopedia of Chemical Technology. John Wiley & Sons, New York. Ge, W., 2005. Intrafollicular paracrine communication in the zebrafish ovary: the state of the art of an emerging model for the study of vertebrate folliculogenesis. Mol. Cell. Endocrinol. 237, 1–10. Griffitt, R.J., Hyndman, K., Denslow, N.D., Barber, D.S., 2009. Comparison of molecular and histological changes in zebrafish exposed to metallic nanoparticles. Toxicol. Sci. 107, 404–415. Griffitt, R.J., Luo, J., Gao, J., Bonzongo, J.C., Barber, D.S., 2008. Effects of particle composition and species on toxicity of metallic nanomaterials in aquatic organisms. Environ. Toxicol. Chem. 27, 1972–1978. Heiden, T.C., Struble, C.A., Rise, M.L., Hessner, M.J., Hutz, R.J., Carvan, M.J., 2008. Molecular targets of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) within the zebrafish ovary: insights into TCDD-induced endocrine disruption and reproductive toxicity. Reprod. Toxicol. 25, 47–57.
J. Wang et al. / Chemosphere 83 (2011) 461–467 Heinlaan, M., Ivask, A., Blinova, I., Dubourguier, H.C., Kahru, A., 2008. Toxicity of nanosized and bulk ZnO, CuO and TiO2 to bacteria Vibrio fischeri and crustaceans Daphnia magna and Thamnocephalus platyurus. Chemosphere 71, 1308–1316. Hund-Rinke, K., Simon, M., 2008. Ecotoxic effect of photocatalytic active nanoparticles (TiO2) on algae and daphnids. Environ. Sci. Pollut. Res. 13, 225– 232. Ings, J.S., Van Der Kraak, G.J., 2006. Characterization of the mRNA expression of StAR and steroidogenic enzymes in zebrafish ovarian follicles. Mol. Reprod. Dev. 73, 943–954. Jonathan, K., 2008. Nanotechnology Boom Expected by 2015. Industrial week (July 22). Kaida, T., Kobayashi, K., Adachi, M., Suzuki, F., 2004. Optical characteristics of titanium oxide interference film and the film laminated with oxides and their applications for cosmetics. J. Cosmet. Sci. 55, 219–220. Kiser, M.A., Westerhoff, P., Benn, T., Wang, Y., Pérez-Rivera, J., Hristovski, K., 2009. Titanium nanomaterial removal and release from wastewater treatment plants. Environ. Sci. Technol. 43, 6757–6763. Kohli, G., Hu, S.Q., Clelland, E., Di Muccio, T., Rothenstein, J., Peng, C., 2003. Cloning of transforming growth factor-beta 1 (TGF-beta 1) and its type II receptor from zebrafish ovary and role of TGF-beta 1 in oocyte maturation. Endocrinology 144, 1931–1941. Levi, L., Pekarski, I., Gutman, E., Fortina, P., Hyslop, T., Biran, J., Levavi-Sivan, B., Lubzens, E., 2009. Revealing genes associated with vitellogenesis in the liver of the zebrafish (Danio rerio) by transcriptome profiling. BMC Genom. 10, 141. Selman, K., Wallace, R.A., Sarka, A., Qi, X.P., 1993. Stages of oocyte development in the zebrafish, Brachydanio rerio. J. Morphol. 218, 203–224. Velzeboer, I., Hendriks, A.J., Ragas, A.M., van de Meent, D., 2008. Aquatic ecotoxicity tests of some nanomaterials. Environ. Toxicol. Chem. 27, 1942–1947. Wang, Y., Ge, W., 2004. Cloning of epidermal growth factor (EGF) and EGF receptor from the zebrafish ovary: evidence for EGF as a potential paracrine factor from
467
the oocyte to regulate activin/follistatin system in the follicle cells. Biol. Reprod. 71, 749–760. Who’s Regulating Nanotechnology? Nanovip, April 23, 2010.. Wiench, K., Wohlleben, W., Hisgen, V., Radke, K., Salinas, E., Zok, S., Landsiedel, R., 2009. Acute and chronic effects of nano- and non-nano-scale TiO2 and ZnO particles on mobility and reproduction of the freshwater invertebrate Daphnia magna. Chemosphere 76, 1356–1365. Wilson-Rawls, J., Hurt, C.R., Parsons, S.M., Rawls, A., 1999. Differential regulation of epaxial and hypaxial muscle development by paraxis. Development 126, 5217. Zhang, Y., Chen, Y.S., Westerhoff, P., Hristovski, K., Crittenden, J.C., 2008. Stability of commercial metal oxide nanoparticles in water. Water Res. 42, 2204–2212. Zhao, Z., Wang, J.X., Sharma, S., Agarwal, N., Liu, H., Chang, Y., 2010. An integrative approach to identifying biologically relevant genes. In: Proceedings of SIAM International Conference on Data Mining (SDM). Zhu, X.S., Zhu, L., Duan, Z., Qi, R., Li, Y., Lang, Y.J., 2008. Comparative toxicity of several metal oxide nano-particle aqueous suspensions to zebrafish (Danio rerio) early developmental stage. Environ. Sci. Health A 43, 278–284. Zhu, X.S., Zhu, L., Li, Y., Chen, Y.S., Tian, S., 2009a. Acute toxicities of six manufactured nanomaterial suspensions to Daphnia magna. J. Nanoparticle Res. 11, 67–75. Zhu, X.S., Wang, J.X., Zhang, X.Z., Chang, Y., Chen, Y.S., 2009b. The impact of ZnO nanoparticle aggregates on the embryonic development of zebrafish (Danio rerio). Nanotechnology 20, 195103. Zhu, X.S., Chang, Y., Chen, Y.S., 2010a. Toxicity and bioaccumulation of TiO2 nanoparticle aggregates in Daphnia magna. Chemosphere 78, 209–215. Zhu, X.S., Wang, J.X., Zhang, X.Z., Chang, Y., Chen, Y.S., 2010b. Trophic transfer of TiO2 nanoparticles from daphnia to zebrafish in a simplified freshwater food chain. Chemosphere 79, 928–933.
Contents lists available at ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
Disruption of zebrafish (Danio rerio) reproduction upon chronic exposure to TiO2 nanoparticles Jiangxin Wang a, Xiaoshan Zhu b,c, Xuezhi Zhang b, Zheng Zhao d, Huan Liu d, Rajani George e, Jeanne Wilson-Rawls e, Yung Chang a,⇑, Yongsheng Chen b,f,⇑ a
Center for Infectious Diseases and Vaccinology, The Biodesign Institute, Arizona State University, Tempe, AZ, USA The School of Sustainable Engineering and the Built Environment, Arizona State University, Tempe, AZ, USA Graduate School at Shenzhen, Tsinghua University, Shenzhen, PR China d School of Computing, Informatics, and Decision Systems Engineering, Ira A. Fulton School of Engineering, Arizona State University, Tempe, USA e Genomics, Evolution and Bioinformatics, School of Life Sciences, Arizona State University, Tempe, AZ, USA f School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, GA, USA b c
a r t i c l e
i n f o
Article history: Received 10 October 2010 Received in revised form 8 December 2010 Accepted 13 December 2010 Available online 15 January 2011 Keywords: Acute toxicity Bioaccumulation Chronic toxicity Microarry Reproduction Zebrafish TiO2 Nanoparticles
a b s t r a c t As common engineered nanomaterials, TiO2 nanoparticles (nTiO2) are usually perceived as non-toxic, and have already been widely used in many products and applications. Such a perception might have been shaped by some short-term studies that revealed no/low toxicity of nTiO2 to cells and eco-relevant organisms. However, given the ultimate release of nTiO2 into the aquatic environment, which can act as a sink for engineered nanoparticles, their long-term impact on the environment and human health is still a concern and deserves more research efforts. Here, for the first time, we demonstrate that chronic exposure of zebrafish to 0.1 mg L 1 nTiO2, can significantly impair zebrafish reproduction. For instance, there was a 29.5% reduction in the cumulative number of zebrafish eggs after 13 weeks of nTiO2 exposure. Thus, we provided timely information on indicating a serious risk of reproductive impairment of environments contaminated with low levels of nTiO2 on aquatic organisms, leading to alterations in population dynamics and aquatic ecosystem balance, and thus warrants a careful scrutiny on toxicity assessment of nTiO2, especially their long-term impact. Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction Thus far, more than 1000 products with nanomaterials are already on the market (Nanovip, 2010). Revenues from nanotechnology-based products are expected to grow worldwide to 3.1 trillion US$ by 2015 (Industrial week; Jonathan, 2008). The large-scale production and utilization of engineered nanomaterials (ENMs) have raised concerns over the release of ENMs to the environment and its potential impact to human and environmental health. Nanoscale titanium dioxide (nTiO2) is widely used in many commercial products, including sunscreens, cosmetics, paints, surface coatings, and food additives (Fisher and Egerton, 2001; Kaida et al., 2004). It is also used as a photocatalyst for environmental decontamination of air, soil, and water (Esterkin et al., 2005; Choi et al., 2006). Such wide application of nTiO2 might have been influenced under the assumption that nTiO2 is somewhat safe ⇑ Corresponding authors. Address: School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, GA, USA. Tel.: +1 404 894 3089; fax: +1 404 894 2278 (Y. Chen), Tel.: +1 480 965 8672; fax: +1 480 727 0599 (Y. Chang). E-mail addresses: [email protected] (Y. Chang), [email protected]. edu (Y. Chen). 0045-6535/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2010.12.069
as several acute and short-term ecotoxicity studies showed no/low toxicity of nTiO2 exerted to testing organisms, ranging from bacteria, algae, Daphnia to fish (Heinlaan et al., 2008; Hund-Rinke and Simon, 2008; Griffitt et al., 2008, 2009; Velzeboer et al., 2008; Zhu et al., 2008, 2009a; Wiench et al., 2009). For instance, Heinlaan et al. (2008) found that neither nTiO2 nor bulk TiO2, were toxic to bacteria or crustaceans. Both Wiench et al. (2009) and Zhu et al. (2009a) reported EC50 > 100 mg L 1 in Daphnia over 48 h exposure to nano-scale and non-nano-scale TiO2. Furthermore, zebrafish embryos or adult fish displayed no apparent abnormality 2 days after being exposed to 10 mg L 1 or 500 mg L 1 TiO2, respectively (Griffitt et al., 2008; Zhu et al., 2008). However, some recent studies, including ours, revealed that an extended exposure up to 21 days did cause some mortality, reproductive impairment and tissue damage in both Daphnia and fish (Federici et al., 2007; Wiench et al., 2009; Zhu et al., 2010a). These studies indicate that chronic exposure may be critical in assessing potential toxicity of nTiO2. Given the wide applications of nTiO2 and its potential for release into the aquatic environment, exposure of aquatic organisms to nTiO2 would be long-term. Thus, the adverse impact of chronic exposure to nTiO2 is worth to be carefully examined, especially at the dose range that is low and considered non-toxic.
462
J. Wang et al. / Chemosphere 83 (2011) 461–467
To further address this issue, we conducted a chronic toxicity study with prolonged zebrafish exposure (91 days) to nTiO2 to systemically evaluate potential impact of such exposure on the zebrafish reproductive system, including egg production; embryo survival; ovary histology, nTiO2 accumulation in ovaries; and global gene expression in ovarian tissues by microarray analyses. We found that zebrafish reproduction could be impaired by nTiO2 exposure at as low as 0.1 mg L 1, which is an environmentally relevant dose range. Our findings indicate that long-term impact of engineered nanomaterials on the aquatic ecosystem should be taken into consideration in advancing nanotechnology.
2. Materials and methods 2.1. Nanoparticle preparation and characterization All chemicals used, including uncoated and powdered nanoscale anatase TiO2 (nTiO2), were obtained from Sigma Aldrich (99+% purity). All samples were prepared in Mili-Q water purified using a Milli-Q plus water purification system (Millipore, Bedford, MA). A stock solution of 10.0 g L 1 nTiO2 was prepared by dispersing the nanoparticles in DI water (resistivity: 18.2 MX cm 1 at 25 °C) with sonication for 10 min (50 W L 1 at 40 kHz). In order to get working concentrations of 0.1 and 1.0 mg L 1 nTiO2 in the fish tanks, each tank was dosed with 10.0 or 100 ll of the 10.0 g L 1 stock solution, respectively. The test solution was changed and re-dosed every 24 h with new aliquot of nTiO2 to maintain the exposure at relatively consistent levels of nTiO2. The actual concentration of nTiO2 in the fish solution over 24 h was tested using inductively coupled plasma mass spectrometry (ICP-MS). The size distribution of nTiO2 in the solution was determined using a dynamic light scattering device (DLS, Brookhaven Instrument Corporation, Holtsville, NY, USA). In this case, three replicate water samples were obtained from three different tanks (sample num-
ber = 3 3) in each group at each time point post-nTiO2 exposure. Specifically, 1 mL solutions from the middle area of different tanks were collected for immediate DLS analysis, which was repeated several times over the 24 h of nTiO2 exposure. The size distribution (median size) of nTiO2 in the test solution is shown in the insert of the Fig. 1A. The particle sizes of nTiO2 in the test solutions were also verified by scanning electron microscope (SEM) with energy dispersive X-ray (EDX) analysis. For SEM test, 1 mL solutions from three different tanks in each group were collected at different time points post-nTiO2 exposure, air dried on pre-cleaned glass slides for standard SEM observation. The representative images are presented in Fig. 1B. For nTiO2 concentration analysis in culture media, 10 mL nTiO2 suspensions at different time points were transferred to triangular flasks and evaporated to dryness. nTiO2 was decomposed into Ti4+ ion by heating with 2 mL of a sulphuric acid–ammonium sulphate solution (400 g ammonium sulphate in 700 mL concentrated sulphuric acid, boiled) until the pellets were completely dissolved. After cooling down, the above solutions were transferred to 10 mL volumetric flasks. Released Ti4+ was determined using ICPMS (Element2, Thermo Fisher, USA). Using this procedure, concentrations of nTiO2 in fish medium were calculated from the corresponding Ti4+ concentrations measured, which were slightly deviated from the initial concentration obtained by a dilution of the nTiO2 stock solution. For the assessment of nTiO2 bioaccumulation in the fish ovary, freshly obtained tissues were completely digested in 5 mL of sulphuric acid–ammonium sulphate solutions to decompose TiO2 into Ti4+ using a microwave digestion procedure. Briefly, samples were digested by a microwave digestion system (Mars 5, CEM, USA) with a four-stage digestion protocol (5 min at 150 °C, 5 min at 190 °C, 10 min at 230 °C and 10 min at 240 °C). After digestion, samples were evaporated to dryness. Then nTiO2 was converted to Ti4+ using sulphuric acid-ammonium sulphate, and then Ti4+ was analyzed by ICP-MS. In this study, nTiO2 recovery in biological samples ranged from 80% to 90%.
2.2. Zebrafish exposure experiment
Fig. 1. nTiO2 characterization in solution within 24 h. (A) The actual concentration of nTiO2 in the fish solution over 24 h. The insert indicates the size distribution (median size) of nTiO2 in the test solution during the same time period. (B) Representative scanning electron microscope (SEM) images for nTiO2 aggregates in the test solution at 24 h. Scale bar: 1 lm. The inserts indicate the sizes of one aggregate with a scale bar of 200 nm. The nTiO2 particles were confirmed with energy dispersive X-ray (EDX) spectroscopy (data not shown).
Zebrafish were maintained in a fish room at The Biodesign Institute of Arizona State University following regulations. Adult zebrafish (more than 3 months old; 0.33 ± 0.09 g) were used for this study. Prior to nTiO2 treatment, fish were acclimated to experimental conditions (28 ± 0.5 °C; light:dark/14:10 h; daily water change), including daily manipulation and weekly breeding for at least 4 weeks. Throughout the experiment, including the breeding period, fish were kept in fish tanks or breeding tanks containing 1 L standard culture medium (containing 64.75 mg L 1 NaHCO3, 5.75 mg L 1 KCl, 123.25 mg L 1 MgSO47H2O, and 294 mg L 1 CaCl22H2O and prepared according to ISO standard 7346-3:1996 and OECD Guideline 202), and the health status of the fish was monitored and recorded daily. All fish were fed daily with TetraMin dry flakes (Tetra, Melle, Germany) containing no nTiO2. As illustrated in Figure S1, after the acclimation period, reproductively active fish (12 females in two tanks and 12 males in two tanks) were exposed to one of the following treatments for 13 weeks using a semi-static exposure regime (water change every 24 h with a new aliquot of nTiO2): control (culture medium only), 0.1, or 1.0 mg L 1 nTiO2. Each week, one male and one female from the same group were set up in a breeding tank (one pair of fish in one breeding tank; twelve pairs for one treatment group); spawning was triggered when the light was turned on in the morning and completed within 30 min. Numbers of spawning fish and eggs from individual breeding tanks were then recorded. The water qualities
J. Wang et al. / Chemosphere 83 (2011) 461–467
such as pH, temperature, hardness and dissolved oxygen concentration were monitored throughout the experiments.
463
data sources to comprehensively evaluate the biological relevance of the genes, which is further explained in the Supporting Information.
2.3. Histological examination Histological examination was performed following protocols described previously (Wilson-Rawls et al., 1999). Briefly, freshly isolated ovary tissues were fixed in newly made 4% paraformaldehyde in PBS overnight at 4 °C. The tissue was embedded in paraffin after a stepwise dehydration in ethanol and xylene, and then 5 lm transverse sections were prepared, deparaffinized in xylene, and stained with hematoxylin and eosin. The sections were collected for further analyses from the tissue 1 mm below the ovary surface. Eight to ten sections of each ovary were examined, and more than 100 follicles from 1 to 3 sections of each ovary (total five ovaries for each group) were counted and assigned to their appropriate stages based on the size and morphology as defined by Selman et al. (1993). The distribution of stages I, II, III and IV in each ovary was calculated as the percentage of total follicles counted, and the averages of different stages of follicles among five individual ovaries of each experimental group are presented. 2.4. RNA extraction and qRT-PCR Fish tissues were removed for total RNA extraction and subsequent reverse transcription, as detailed in the Supporting Information. To monitor the expression of selected genes, we applied realtime quantitative PCR (qRT-PCR) to analyze individually prepared ovary and liver samples using SYBR green procedure (Zhu et al., 2009b) on an ABI-7000 system with primers listed in Table S1 of the Supporting Information. 2.5. Zebrafish microarray analyses and bioinformatics Global gene expressions were analyzed with the Agilent 4 44 k zebrafish microarray (Santa Clara, CA) using a two-color reference design. Quality and concentration of RNA samples were evaluated using a 2100 Bioanalyzer (Agilent, USA) and Nanodrop ND-1000 (Thermo Fisher Scientific, USA). In this study, RNA samples with RNA integrity number higher than 9.0 (10 being intact and 1 being totally degraded) were chosen for microarray and qRT-PCR analyses. The control RNA was constructed by pooling equal amounts of RNA from four individual control ovary samples. For treated groups, both pooled (equal amounts of RNA, 250 ng each, from four individual treated ovary samples) and two individual ovary RNA samples were submitted to the microarray analysis. cRNA synthesis and reverse transcription were conducted using a two-color Amplification Kit and Gene Expression Hybridization Kit (Agilent, USA), according to the manufacturer’s protocols. The labeled samples were hybridized to the arrays for 17 h at 65 °C, washed, and immediately scanned using a G2505 B Microarray Scanner (Agilent, USA) and raw expression data were extracted using Feature Extraction 9.5 (Agilent, USA). To identify genes differentially expressed between exposed and control zebrafish, p-values of Student’s 1-sample t-test and mean fold changes between the exposed and control samples were calculated. A threshold based on both p-value and fold change (p < 0.05 and fold change > 2) was used to select genes that responded to the exposure, either up-regulated or down-regulated by nTiO2 treatments. The genes with consistent expression patterns under 0.1 and 1.0 mg L 1 nTiO2 treatments were selected, and the Hierarchical clustering analysis was employed to search for common gene expression patterns. To extract multiple lines of information that arise in the context with simultaneous observation of more than 40,000 probes in the microarray, our integrative framework (Zhao et al., 2010) was applied to fuse different types of knowledge and
2.6. Data analysis The reproduction experiment was conducted with 12 pairs of adult fish in each group. Values for the average number of egg per spawned fish, histological analysis of folliculogenesis, nTiO2 accumulation in ovary tissues, and qRT-PCR analyses were evaluated in at least five replicates for statistical evaluation of the differences between the treatment and control group. Data are represented as the mean ± SD. Differences among means were determined using one-way analysis of variance (ANOVA) for all parameters. Results were considered to be statically significant if p < 0.05. 3. Results and discussion 3.1. Nanoparticle characterization nTiO2 in our tests were found to aggregate in culture medium, consistent with our previous observation of metal oxide nanoparticles (Zhu et al., 2009b). The quantitative analyses were used to determine the level and kinetics of the nTiO2 aggregation process. To assess the stability of nTiO2 in fish water, we examined the concentration and particle size of nTiO2 during the 24 h of exposure time. At 0.1 mg L 1, the concentration of nTiO2 remained relatively constant over 24 h. However, at 1.0 mg L 1, the concentration dropped to 0.6 mg L 1 over the first hour and then remained unchanged for the rest of time (Fig. 1A). The drop in concentration might have been caused by aggregation of the nanoparticles, which is concentration dependent, as reported in our previous finding (Zhu et al., 2009b). Using DLS, the median values of particle size were detected with mean sizes of 240–280 nm (0.1 mg L 1) and 259–360 nm (1.0 mg L 1) at 0, 1, 3, 6, 12 and 24 h. Observed by SEM, these nTiO2 aggregates had variable sizes from a few hundred nanometers to several microns in diameter (Fig. 1B), which is in line with our previous reports (Zhu et al., 2009b). 3.2. Impairment of zebrafish reproduction and embryo survival by chronic exposure to nTiO2 We set up a long-term nTiO2 exposure to examine its effect on zebrafish reproduction. Specifically, zebrafish were exposed to 0.1 mg L 1 or 1.0 mg L 1 of nTiO2 daily for 13 weeks. Given the possible fluctuation of nTiO2 over time in both concentrations and particle sizes, we transferred zebrafish every 24 h to a tank containing a new aliquot of nTiO2 at appropriate concentrations (i.e., 0.1 mg L 1 or 1.0 mg L 1) to ensure a relatively consistent nTiO2 exposure for these fish. The zebrafish reproduction was evaluated weekly by assessing the number of spawned females, egg production, and embryo survival rates. Daily monitoring of the mortality and general health of the fish revealed no apparent abnormalities. However, the total cumulative number of eggs and the number of eggs per spawned female was gradually reduced beginning at week 9 in groups treated with nTiO2 although the number of spawning females was comparable between control and treated groups (Fig. 2A and B). The number of spawned females each week for the control group (open circles) was 9, 9, 10, 10, 11, 12, 11, and 10 from week 6 to week 13, respectively, whereas the numbers for the 0.1 mg L 1 group (cross) were 10, 10, 8, 11, 10, 5, 4, and 9 during the same weeks. Significant reduction in embryo survival was observed in the nTiO2-treated group at 8, 10, 11, 12 and 13 weeks (p < 0.05).
464
J. Wang et al. / Chemosphere 83 (2011) 461–467
Fig. 2. Disruption of zebrafish reproduction upon long-term exposure to nTiO2 and the embryo survival rate of the F1 generation from the control and 0.1 mg L 1 nTiO2treated groups 6–13 weeks post-nTiO2 treatment. (A) A representative experiment shows the reduction in the cumulative number of zebrafish eggs after long-term nTiO2 exposure. (B) The inhibitory effect of nTiO2 treatment is evident from the reduction in the average numbers of eggs produced per spawning female (p < 0.05, n = 12). (C) The data points indicate the survival rate, at 48 hpf, of embryos derived from individual spawned females.
As a result, even at a concentration as low as 0.1 mg L 1, there was a 29.5% reduction in the cumulative number of zebrafish eggs (the eggs were collected from all spawned fish of each experimental group and the total egg numbers were tabulated for each group) over 13 weeks of nTiO2 exposure, as compared to the control group (Fig. 2A). The effect of nTiO2 exposure on embryo survival was also monitored. The embryos spawned from the females exposed to 0.1 mg L 1 nTiO2 for 8 weeks or longer showed an increase in mortality at 48 h post-fertilization (hpf), as compared to the control group. A significant increase in embryo mortality was observed in groups with 8, 10, 11, 12, 13 weeks of exposure (Fig. 2C). Our findings reveal that zebrafish reproduction and embryo survival could be negatively affected by a prolonged exposure to nTiO2 at concentrations as low as 0.1 mg L 1, close to the ENM level disseminated into the environment (Zhang et al., 2008; Kiser et al., 2009). 3.3. Defect in folliculogenesis The reduced egg production may reflect a defect in folliculogenesis. To test this hypothesis, the ovaries from the control and nTiO2-treated fish were sectioned and stained for histology examination to characterize the developmental stages (i.e., I–IV) of the follicles therein according to Selman’s classification (Selman et al., 1993). As compared to the control group, the distribution of follicular developmental stages was skewed by the nTiO2
treatment toward the immature stage, as evident by an increase in stage I follicles and some reduction in other stages, especially, stage IV follicles (Fig. 3A and B). This indicates a developmental block at the transition from stage I to stage II and possibly at follicular maturation to stage IV as well (Selman et al., 1993). This block may be attributable to a direct effect of nTiO2 on developing follicles or to an indirect disruption of the normal pituitary-gonad endocrine system. 3.4. nTiO2 accumulation in ovaries To determine whether the ovary is a direct target of nTiO2, we examined nTiO2 accumulation in fish ovaries using ICP-MS. Fish ovaries were found to accumulate nTiO2 at concentrations of approximately 2.5 mg kg 1 and 7.2 mg kg 1 under 0.1 mg L 1 and 1.0 mg L 1 doses, respectively (Fig. 3C). This finding suggests that nTiO2 can enter the ovary, presumably via blood circulation, and may directly act on this organ to disrupt oocyte development. 3.5. Effect of nTiO2 on gene expression Given the apparent disruption in oocyte maturation in the females exposed to nTiO2, genes involved in oocyte development might have been altered by nTiO2 treatment. To test this scenario, we performed expression microarray analyses of ovarian tissues from both control and nTiO2-treated groups to identify genes
J. Wang et al. / Chemosphere 83 (2011) 461–467
465
Fig. 3. Histological changes and nTiO2 accumulation in zebrafish ovary afterlong-term nTiO2 exposure. (A) Representative sections of ovaries taken from control (left), 0.1 mg L 1 (middle) and 1.0 mg L 1 nTiO2-treated (right) zebrafish females. Scale bar, 200 lm. (B) The average percentage of follicles present at different developmental stages (p < 0.05, n = 5). (C) nTiO2 accumulation in individual ovaries from females in the control, 0.1 mg L 1, or 1.0 mg L 1 nTiO2-treated groups is presented as mg kg 1 (p < 0.05, n = 4).
associated with the disrupted folliculogenesis in the nTiO2-treated groups. The array analyses revealed a significant alteration caused by long-term nTiO2 exposures, even at a low concentration as low as 0.1 mg L 1 (Fig. 4). For instance, total 2383 and 1043 genes were found to be down-regulated in 0.1 and 1.0 mg L 1 treated groups, respectively, while 2069 and 471 genes were up-regulated genes detected in 0.1 and 1.0 mg L 1 exposed fish, respectively. Among these differentially expressed genes, 405 up-regulated and 174 down-regulated genes were shared in both 0.1 and 1.0 mg L 1 treated groups. Table S2 lists a subset of the genes altered by nTiO2 at both 0.1 mg L 1 and 1.0 mg L 1. Pathway and gene function analysis showed that some of these genes are involved in proteolysis, oxidative stress regulation, metabolism, insulin signaling, and apoptosis, as well as oocyte maturation, suggesting multiple and complicated modes of action for nTiO2-mediated disruption of reproduction. However, these differentially expressed genes have little overlap with those derived from microarray analyses of other toxicants
Fig. 4. Venn diagram analyses of genes identified by microarray analysis. Number of genes significantly differentially expressed in the ovary following zebrafish exposure to nTiO2 at different doses. (A) Down-regulated genes. (B) Up-regulated genes.
and chemicals, such as 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) (Heiden et al., 2008) and estradiol (E2) dependent treatments (Levi et al., 2009), or with the gene sets differentially expressed in gills after short-term exposure to different ENMs (Griffitt et al., 2009). Thus, although nTiO2, TCDD and E2 all cause impaired oocyte maturation, the mode of action mediated by nTiO2 likely differs from those by hormone disrupting compounds (Heiden et al., 2008; Levi et al., 2009). Our finding of a significant differential gene expression profile even at as low as 0.1 mg L 1 of nTiO2 reveals that the maturation and functionality of zebrafish ovary could be affected by chronic exposure to a very low dose of nTiO2 (Table S2, 0.1 mg L 1). Thus, we focused on this experimental group, i.e., treated with 0.1 mg L 1 of nTiO2, for further characterization. Three genes that control oocyte maturation, cyp11a, tgfb1 and egf, were found significantly decreased upon chronic exposure to nTiO2 (Table S2). The reduction of these genes was confirmed by qRT-PCR (Fig. 5A, Supporting information, and Table S1). These genes are expressed primarily in stage I/II follicles and promote stage I/II growth and maturation (Kohli et al., 2003; Wang and Ge, 2004; Ge, 2005; Clelland et al., 2006). Despite an elevated number of morphologically stage I follicles (Fig. 3B), the reduction in these genes implies that nTiO2 may inhibit the growth and maturation of stage I follicles (Kohli et al., 2003), presumably by altering the expression of several regulators that are critical to this stage of folliculogenesis. The growth factors coded by tgfb1 and egf have been implicated as paracrine factors that promote growth and differentiation of stage I follicles. cyp11a encodes an enzyme that converts cholesterol into testosterone, the rate limiting control step of E2 synthesis in the follicle, which in turn controls the expression of vitellogenin in the liver to promote follicular
466
J. Wang et al. / Chemosphere 83 (2011) 461–467
Fig. 5. Expression analysis by qRT-PCR on selected genes. (A) Expression of genes associated with oocyte development (cyp11a1, tgfb1 and egf) in ovaries from the control and 0.1 mg L 1 nTiO2 treated groups ( represents p < 0.05, n = 8). (B) vtg1 expression in livers from the control and 0.1 mg L 1 nTiO2 treated groups ( represents p < 0.05, n = 6). Actin was used as a housekeeping gene. The data in both (A) and (B) were obtained from fish samples in two independent experiments.
vitellogenesis and maturation, according to the hypothalamuspituitary-gonad (HPG) regulatory circuit in teleosts (Ings and Van der Kraak, 2006; Clelland and Peng, 2009). On the other hand, several up-regulated genes are associated with protein degradation or reactive oxygen species (ROS) production (Table S2), which may reflect stress responses of ovaries that are dealing with nTiO2 particles, e.g., trying to remove foreign entities. Based on the analyses of the differentially expressed genes, we argue that nTiO2 may affect the ovary directly, which is consistent with our observation of nTiO2 accumulation in this tissue (Fig. 3C). nTiO2 may cause oxidative stress to the developing follicles and interfere with the expression of key regulatory genes, ultimately impairing the growth and function of stage I/II follicles. If so, the aborted follicle development, as well as the possible inhibition of E2 synthesis as a result of cyp11a down-regulation, could impinge on vitellogenin synthesis in the liver, which in turn alters the maturation of vitellogenic follicles (i.e., stages III and IV) (Clelland and Peng, 2009). To test this scenario, we analyzed the expression of vtg1, one of the vitellogenin genes, in the livers of zebrafish treated with nTiO2 and control fish. As shown in Fig. 5B, vtg1 expression was found significantly reduced in the livers of nTiO2-treated fish, consistent with the abnormality noted in the folliculogenesis of these fish (Fig. 3B). Taken together, our data, which are derived from zebrafish egg production, ovary histology, nTiO2 accumulation and altered gene expression, reveals that long-term exposure to low concentrations of nTiO2 is toxic to the zebrafish reproductive system. Thus, the ultimate consequence of chronic exposure to nTiO2 nanoparticles is similar to the one caused by many reproductive toxicants, such as endocrine disrupting chemicals (EDCs). However, unlike EDCs, which usually perturb fish reproduction by interfering with hormone regulation, nTiO2 may act on primary follicles directly and/ or subsequently inhibit interfere with vitellogenin synthesis. Regardless of the mode of action, our current findings along with our previous study on Daphnia (Zhu et al., 2010a) and recent demonstration of nTiO2 transfer from Daphnia to zebrafish in a simplified freshwater food chain (Zhu et al., 2010b) indicate that chronic exposure to low doses of nTiO2 could alter the reproduction of certain aquatic organisms and ultimately cause perturbation in the population dynamic of these organisms in aquatic environments. Our studies highlight the importance of long-term studies in assessing the negative impact of nTiO2 at concentrations relevant or close to the levels of nTiO2 released into the environment, which will provide critical information for risk assessment and hazard management of environmental contamination by nTiO2 and other ENMs.
Acknowledgements The authors would like to thank T. Fu (Biodesign Inst. at Arizona State University) for her critical technical assistance on fish maintenance, treatments and data collection; S. Bingham and J. Hock (DNA Lab., Arizona State University) for their assistance on microarray experiments and the data extraction process, and P. Prapaopong (Chemistry Dept., Arizona State University) for assistance on fish tissue HNO3 microwave digestion and ICP-MS experiments. This study was supported by the U.S. Environmental Protection Agency Science to Achieve Results Program (Grant # RD831713) (to Y. Chen and Y. Chang) and National Science Foundation Grant (0812551) (to H. Liu). Article contents are sole responsibility of the authors and do not represent official views of the sponsors. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.chemosphere.2010.12.069. References Choi, H., Stathatos, E., Dionysiou, D.D., 2006. Sol–gel preparation of mesoporous photocatalytic TiO2 films and TiO2/Al2O3 composite membranes for environmental applications. Appl. Catal. B – Environ. 63, 60–67. Clelland, E., Kohli, G., Campbell, R.K., Sharma, S., Shimasaki, S., Peng, C., 2006. Bone morphogenetic protein-15 in the zebrafish ovary: complementary deoxyribonucleic acid cloning, genomic organization, tissue distribution, and role in oocyte maturation. Endocrinology 147, 201–209. Clelland, E., Peng, C., 2009. Endocrine/paracrine control of zebrafish ovarian development. Mol. Cell. Endocrinol. 312, 42–52. Esterkin, C.R., Negro, A.C., Alfano, O.M., Cassano, A.E., 2005. Air pollution remediation in a fixed bed photocatalytic reactor coated with TiO2. AIChE. J. 51, 2298–2310. Federici, G., Shaw, B.J., Handy, R.D., 2007. Toxicity of titanium dioxide nanoparticles to rainbow trout (Oncorhynchus mykiss): gill injury, oxidative stress, and other physiological effects. Aquat. Toxicol. 84, 415–430. Fisher, J., Egerton, T., 2001. Titanium Compounds, Inorganic. Kirk-Othmer Encyclopedia of Chemical Technology. John Wiley & Sons, New York. Ge, W., 2005. Intrafollicular paracrine communication in the zebrafish ovary: the state of the art of an emerging model for the study of vertebrate folliculogenesis. Mol. Cell. Endocrinol. 237, 1–10. Griffitt, R.J., Hyndman, K., Denslow, N.D., Barber, D.S., 2009. Comparison of molecular and histological changes in zebrafish exposed to metallic nanoparticles. Toxicol. Sci. 107, 404–415. Griffitt, R.J., Luo, J., Gao, J., Bonzongo, J.C., Barber, D.S., 2008. Effects of particle composition and species on toxicity of metallic nanomaterials in aquatic organisms. Environ. Toxicol. Chem. 27, 1972–1978. Heiden, T.C., Struble, C.A., Rise, M.L., Hessner, M.J., Hutz, R.J., Carvan, M.J., 2008. Molecular targets of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) within the zebrafish ovary: insights into TCDD-induced endocrine disruption and reproductive toxicity. Reprod. Toxicol. 25, 47–57.
J. Wang et al. / Chemosphere 83 (2011) 461–467 Heinlaan, M., Ivask, A., Blinova, I., Dubourguier, H.C., Kahru, A., 2008. Toxicity of nanosized and bulk ZnO, CuO and TiO2 to bacteria Vibrio fischeri and crustaceans Daphnia magna and Thamnocephalus platyurus. Chemosphere 71, 1308–1316. Hund-Rinke, K., Simon, M., 2008. Ecotoxic effect of photocatalytic active nanoparticles (TiO2) on algae and daphnids. Environ. Sci. Pollut. Res. 13, 225– 232. Ings, J.S., Van Der Kraak, G.J., 2006. Characterization of the mRNA expression of StAR and steroidogenic enzymes in zebrafish ovarian follicles. Mol. Reprod. Dev. 73, 943–954. Jonathan, K., 2008. Nanotechnology Boom Expected by 2015. Industrial week (July 22). Kaida, T., Kobayashi, K., Adachi, M., Suzuki, F., 2004. Optical characteristics of titanium oxide interference film and the film laminated with oxides and their applications for cosmetics. J. Cosmet. Sci. 55, 219–220. Kiser, M.A., Westerhoff, P., Benn, T., Wang, Y., Pérez-Rivera, J., Hristovski, K., 2009. Titanium nanomaterial removal and release from wastewater treatment plants. Environ. Sci. Technol. 43, 6757–6763. Kohli, G., Hu, S.Q., Clelland, E., Di Muccio, T., Rothenstein, J., Peng, C., 2003. Cloning of transforming growth factor-beta 1 (TGF-beta 1) and its type II receptor from zebrafish ovary and role of TGF-beta 1 in oocyte maturation. Endocrinology 144, 1931–1941. Levi, L., Pekarski, I., Gutman, E., Fortina, P., Hyslop, T., Biran, J., Levavi-Sivan, B., Lubzens, E., 2009. Revealing genes associated with vitellogenesis in the liver of the zebrafish (Danio rerio) by transcriptome profiling. BMC Genom. 10, 141. Selman, K., Wallace, R.A., Sarka, A., Qi, X.P., 1993. Stages of oocyte development in the zebrafish, Brachydanio rerio. J. Morphol. 218, 203–224. Velzeboer, I., Hendriks, A.J., Ragas, A.M., van de Meent, D., 2008. Aquatic ecotoxicity tests of some nanomaterials. Environ. Toxicol. Chem. 27, 1942–1947. Wang, Y., Ge, W., 2004. Cloning of epidermal growth factor (EGF) and EGF receptor from the zebrafish ovary: evidence for EGF as a potential paracrine factor from
467
the oocyte to regulate activin/follistatin system in the follicle cells. Biol. Reprod. 71, 749–760. Who’s Regulating Nanotechnology? Nanovip, April 23, 2010.