Oxidative stress effects of titanium dioxide nanoparticle aggregates in zebrafish embryos

Science of the Total Environment 470–471 (2014) 379–389

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Oxidative stress effects of titanium dioxide nanoparticle aggregates in zebrafish embryos Melissa Faria a,⁎, José M. Navas b, Amadeu M.V.M. Soares a, Carlos Barata c a b c

CESAM, Departamento de Biologia, Universidade de Aveiro, 3810-193 Aveiro, Portugal INIA, Dpto. Medio Ambiente, Laboratorio de Ecotoxicología, Ctra de la Coruña km 7.5, E-28040 Madrid, Spain Department of Environmental Chemistry, IDAEA-CSIC, Jordi Girona 18, 08034 Barcelona, Spain

H I G H L I G H T S • At 1 mg ml− 1 TiO2 aggregates impaired embryo growth. • All TiO2 aggregates generated oxidative stress in the absence of solar simulated radiation. • The nanomaterial P25 was the most phototoxic material across all studied particles.

a r t i c l e

i n f o

Article history: Received 29 July 2013 Received in revised form 17 September 2013 Accepted 17 September 2013 Available online 18 October 2013 Editor: Damia Barcelo Keywords: Zebrafish Titanium dioxide nanoparticles Phototoxicity Oxidative stress Antioxidant markers

a b s t r a c t There is limited data on the sub-lethal oxidative stress effects of titanium dioxide nanoparticle aggregates (NMTiO2) and its modulation by simulated solar radiation (SSR) to aquatic organisms. This study aimed to examine sublethal oxidative stress effects of aqueous exposure to three different types of NM-TiO2 differing in their coating or crystal structure but of similar primary size (20 nm) plus a micron-sized bulk material to zebrafish embryos without and with SSR. Oxidative stress responses of known model prooxidant (tert-Butyl hydroperoxide) and photoprooxidant (fluoranthene) compounds were also studied. Results evidenced a low bio-availability of NM-TiO2 to embryos with detrimental effects on growth at 1 mg ml−1. Phototoxicity increased moderately, by 3 and 1.5 fold, under co-exposures to fluoranthene (100 μg l−1) and to the NM-TiO2 P25 (1 mg ml−1), respectively, being unchanged in the other TiO2 aggregates. In vitro exposures under SSR confirmed that the NM-TiO2 P25 had the highest potential to generate reactive oxygen species (ROS). Antioxidant enzyme activities of superoxide dismutase increased shortly after exposure to the studied materials, whereas the levels of glutathione tend to be altered after longer exposures. All compounds were able to produce oxidative stress enhancing the senescence-associated β galactosidase pigment (SA-β-gal). Under SSR radiation the NM-TiO2 P25 affected antioxidant and oxidative stress responses as the phototoxic compound fluoranthene. These results indicated that despite the low bio-availability of NM-TiO2 to zebrafish embryos, P25 was phototoxic due to the production of reactive oxygen species. Nevertheless, overall our results indicated that fish development may not be at high risk in the face of NM-TiO2, even when combined with prooxidant conditions. © 2013 Elsevier B.V. All rights reserved.

1. Introduction The study of the ecotoxicity of manufactured nanomaterials to aquatic organisms has a relative recent history when compared with that of other toxic substances (Braun et al., 2008). Manufactured nano-scale titanium dioxide particles (NM-TiO2) are used in a broad range of consumer products including sunscreens, cosmetics, paints and other coatings, and embedded in glass and other surfaces. NMTiO2 are also applied in wastewater treatments (Chen et al., 2004) and

⁎ Corresponding author. Tel.: +351 234 370350; fax: +351 234 372 587. E-mail addresses: [email protected] (M. Faria), [email protected] (J.M. Navas), [email protected] (A.M.V.M. Soares), [email protected] (C. Barata). 0048-9697/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.scitotenv.2013.09.055

environmental remediation (Aitken et al., 2006). Release of these NMTiO2 to aquatic systems through bathing, sewage effluent discharges and other industrial applications has increased the chances of aquatic organisms and humans to be exposed to them (Klaine et al., 2008). The biological toxicity of NM-TiO2 is closely related to many physicochemical characteristics such as size, phase composition, surface modification, and radical formation. Size is a well-known important factor determining NM-TiO2 toxicity to aquatic organisms since penetration is eased with decreasing particle size while bioavailability is increased, so that more particles can be deposited inside the cell (Kim et al., 2010; Lovern and Klaper, 2006; Metzler et al., 2011). The phase composition of nanoscale titanium dioxide particles is also known to affect toxicity. Several research groups have reported that NM-TiO2 could cause oxidative stress and that they can be photoactivated by UV

380

M. Faria et al. / Science of the Total Environment 470–471 (2014) 379–389

generating the hydroxyl radical (•OH), a potent reactive oxygen species (ROS) (Colvin, 2003; Klaine et al., 2008; Xiong et al., 2011). In particular nanoscale TiO2 particles with a phase composition of anatase have been reported to be 100 times more toxic to human fibroblast or lung epithelial cells than an equivalent sample composed mostly by rutile (Sayes et al., 2006). There is, however, disagreement as to the exact nature of the ROS that are involved in cell death. Possible ROS that can damage cells and tissues are hydroxyl radicals (•OH), superoxide radical anions 1 (O− 2 ), hydrogen peroxide (H2O2) and singlet oxygen ( O2) (Cai et al., 1992; Chen et al., 2004; Konaka et al., 2001; Reeves et al., 2008; Uchino et al., 2002). In relation to this there are few studies that have compared the mechanisms of toxicity of phototoxic NM-TiO2 with those from non-phototoxic NM-TiO2 and most interestingly with other prooxidant substances (Oberdörster, 2004; Xiong et al., 2011). Oxidative stress is a common pathway of toxicity of many pollutants that may affect organisms through several mechanisms. It may be directly induced by oxidizing (e.g. hydrogen peroxide, H2O2) or photooxidizing (e.g. fluoranthene) agents that react with oxygen producing ROS (Bar-Ilan et al., 2012; Fu et al., 2012; Miller et al., 2012). Organism responses against ROS usually involve the increase of antioxidant enzyme activities such as superoxide dismutase enzyme (SOD) that convert the superoxide anion radical O− 2 to hydrogen peroxide (H2O2), catalase or glutathione peroxidase that further reduces H2O2 to water (Halliwell and Gutteridge, 1999). Oxidative stress may also be produced through the induction of cytochrome P4501A (Cyp1A), as is the case of co-planar polycyclic aromatic hydrocarbons (PAHs) (Wassenberg and Di Giulio, 2004). Xenobiotics may also disrupt the cellular antioxidant system by the inhibition of the synthesis of antioxidant molecules such as glutathione (GSH) or by the impairment of anti-oxidant enzyme activities that function to maintain oxidative balance (Anderson, 1998; Livingstone, 2001). Recently several studies reported that NM-TiO2 with a crystal structure of anatase were toxic to embryos and adult zebrafish and that toxicity was a photo-dependent due to the production of ROS which enhanced oxidative stress and oxidative damage (Bar-Ilan et al., 2013, 2012; Ramsden et al., 2013; Xiong et al., 2011). In the present study we use the embryonic zebrafish model to elucidate the mechanisms by which NM-TiO2 differing in their phase composition induced oxidative stress toxicity. This model organism has been used extensively for drug discovery and chemical or nanotoxicity screenings and to identify mechanisms of toxic actions (Usenko et al., 2007). Here, we studied oxidative stress responses caused by different NM-TiO2 with and without light using a set of known oxidative stress markers: superoxide dismutase (SOD) activity, levels of glutathione (GSH) and of senescenceassociated β galactosidase pigment (SA-β-gal). Altered patterns of SOD activities and GSH levels in zebrafish embryos exposed to prooxidant agents have been reported for TBBPA and PAHs (Hu et al., 2009; Timme-Laragy et al., 2009) and increased levels of SA-β-gal are known to be closely related with increased levels of oxidative tissue damage in zebrafish embryos (Kishi et al., 2008). These markers were first validated with known oxidizing (tert-Butyl hydroperoxide; BHP (Kishi et al., 2008)) and photo-oxidizing (fluoranthene (Matson et al., 2008)) agents and then determined in embryos exposed to distinct NM-TiO2. One of the key features of oxidative stress markers is their transient response (Dinu et al., 2010; Parihar et al., 1997), which makes it necessary to include a time-course experimental design.

were obtained by natural mating and maintained in a dark incubator in fish culture water at 28.5 °C. Animal stages were recorded as hours postfertilization (hpf) or days postfertilization (dpf). 2.2. Model prooxidants Fluoranthene 98% purity was purchased from ACROS-ORGANICS (Thermo Fisher Scientific, CAS — 206-44-0) and tert-Butyl hydroperoxide (BHP) was purchased from Aldrich (Sigma, CAS — 75-91-2). Aqueous solution of 10 and 100 μg l− 1 fluoranthene was prepared in fish water using DMSO (b1 ml l− 1) as carrier and BHP 500 μM was prepared by directly diluting the supplied stock solution in fish medium. 2.3. TiO2 nanomaterial Two of the used NM-TiO2 were supplied by the repository of the European Commission (EC) Joint Research Centre at Ispra, Italy. In order to facilitate the collection of data and the comparison with other studies performed with the same NM-TiO2, the codes used in this work are those of the repository. NM-103 and NM-104 are TiO2 of rutile modification, with a TiO2 total content of 89%, and a primary crystal size of 20 nm. According to the received information NM-103 exhibits hydrophobic properties while NM-104 is hydrophilic (Table 1). Both nanomaterials are used as UV screening agents in sunscreen products and NM-104 has received an organic treatment with glycerin. The other NM used in this work was P-25 (Evonik Degussa, Germany). It is a TiO2 of anatase–rutile modification (78% anatase, 14% rutile and 8% amorphous phase), according to Ohtani et al. (2010). The manufacturer reports 99.5% purity and a primary size of 21 nm (Sigma-Aldrich, http://www.sigmaaldrich.com/) (Table 1). Finally, a micro-sized TiO2 (Tiona AT-1 (Cristal Global, MD)) was used as reference and identified from here on as micro-TiO2. The manufacturer reports that the crystal form is anatase, with 98.5% TiO2 content and with untreated surface (Table 1). Primary particle size determined by transmission electron microscope (TEM) confirmed the sizes reported by manufacturers (i.e. approx. 20 nm for NM-103, NM-104, and P-25 and approx. of 200 nm for micro-TiO2) (Campos et al., 2013). The initial stock solutions of particle suspension of 1 mg ml−1 were dispersed with the aid of an ultrasonic bath during 30 min in fish medium and left for 24 h under agitation in darkness prior to the preparation of test solutions. Test solutions were prepared from stocks still under agitation and further dispersed in an ultrasonic bath 15 min prior to exposure. 2.4. Physicochemical characterization of TiO2 materials The aggregate size of nanoparticles in the water column was determined with dynamic light scattering (DLS) using a Zeta-Sizer Nano ZS (Malvern Instruments Ltd., Malvern, UK). Experimental water without particles was used as a control. Before measuring the samples the instrument temperature was set at 25 °C. Three independent measurements were taken with 4 readings per measurement, each reading consisting of six runs of 10 s duration. The mean hydrodynamic size was calculated from the histogram of size distribution frequency for each nanoparticle. The polydispersity index (Pdi) that informs about the width of the size distribution was also included.

2. Material and methods 2.1. Culture conditions Adult zebra fish (Danio rerio) wild type were maintained in reverseosmosis purified water containing 90 μg/ml Instant Ocean (Aquarium Systems, Sarrebourg, France), 0.58 mM CaSO4, and 2H2O, at 28 ± 1 °C. Photoperiod was set to a 12 h:12 h light: dark cycle and fish fed on flake dry food complemented with live Artemia. Zebrafish embryos

Table 1 Physical characteristics of TiO2 particles. TiO2 material

Particle size (nm)

Prime crystal form

Surface treatment

NM103 NM104 P25 MicroTiO2

20 20 21 200

Rutile Rutile Anatase Anatase

Hydrophobic Hydrophilic Untreated Untreated

M. Faria et al. / Science of the Total Environment 470–471 (2014) 379–389

The relative potency of the tested TiO2 materials to produce ROS was tested in vitro using simulated solar radiation (SSR, SUNTEST CPS +, ATLAS material testing technology, Germany) comparing the concentration-fluorescence inhibition curves of aqueous suspensions of TiO2 materials using fluorescein as a fluorescent probe. This fluorescent dye is degraded under the presence of ROS (Ou et al., 2001). Different concentrations of waterborne TiO2 materials were dispersed in glass containers in the presence of 20 nM of fluorescein and fluorescent background was measured at (ex.485 nm, em.520 nm). Subsequently the mixture was exposed to SSR of 180 J m−2 (19.5 J m−2 of UV) and following exposure, fluorescence was once more monitored. Treatments of fluorescein and TiO2 materials non-exposed to SSR were used as controls. Fluorescein alone exposed to the same SSR radiation dose showed no degradation. 2.5. Experimental design Experiments were conducted as follows. Within 2 h after spawning either 10 or 100 embryos, respectively, were transferred to 6-well plates containing 5 ml of test medium (two replicates per treatment) or 100 ml glass beakers containing 50 ml of test medium. In all experiments organism concentration was kept to 2 embryos ml−1 of test medium. Larvae were not fed throughout the exposure periods. Exposures were held in a dark incubator at 28.5 °C. Control groups received only fish culture medium. Hatching status and survival were registered daily according to the OECD 212, “Test Guideline 212”., 1998 standards along with any obvious morphological abnormalities. 2.6. Microscopy All individuals were observed and photographed under a Nikon SMZ1500 (NIKON Instruments INC. New York, USA) dissecting microscope fitted with Nikon Digital Sight DSRi1 camera and NIS Elements AR software (version 3.0) (NIKON Instruments INC. New York, USA) and saved as high resolution (3840 pixels × 3005 pixels) tagged image file format (TIFF). All images were taken on the left side of each fish. The total body length (μm) was obtained by using a measuring tool from the NIS Elements software where a line was drawn from the anterior-most part of the snout to posterior-most point of the tail. 2.7. Toxicity assays Within 2 h after spawning, 10 embryos were transferred to 6-well plates containing 5 ml of test medium (two replicates per treatment and TiO2 material). Embryos were exposed to 0.01, 0.1 and 1 mg ml−1 of each substance for 8 days in a dark incubator at 28.5 °C without food. Control groups received only embryo medium. Hatching status and survival were registered daily according to the OECD TG 212 standards along with any obvious morphological abnormalities. Body length and morphological abnormalities were monitored in fixed 8 dpf larvae as follows: larvae were fixed in 4% paraformaldehyde (PFA) overnight at 4 °C, followed by several washes in phosphatebuffered saline (PBS: 2.68 mM KCl, 7.81 mM Na2HPO4·2H2O, phosphate, 1.47 mM KH2PO4, 0.14 M NaCl) and gradually transferred to 90% glycerol. Larvae were examined to observe the phenotype and images were obtained to determine the total body length. 2.8. Titanium uptake assays Groups of embryos were exposed to the highest concentration tested (1 mg ml−1) for 8 days as described above. Following exposures about 25 fish embryos were collected, rinsed with clean culture water and freeze dried for TiO2 determination. Another 25 embryos were transferred to clean, TiO2 free culture water for 24 h to allow the cleansing of TiO2 from the digestive tract and then treated as above for

381

titanium determination. Extraction of titanium was conducted following Querol et al. (2001) (for details see Supplementary material) Because biological standard reference materials containing TiO2 were not available, laboratory controls were prepared by using samples of fish embryos fortified with 0.1 mg of TiO2. Recoveries varied between 80 and 98% across NM-TiO2. Titanium levels were determined by inductively quadrupole plasma mass spectrometry (ICP-MS, XSeriesII ICPMS, Thermo Electron Corporation, USA). Measurements were performed for each TiO2 material per triplicate. 2.9. Time dependent oxidative stress Embryos were placed in clean fish culture medium for controls and medium containing 500 μM of BHP, which according to Kishi et al. (2008) should cause oxidative stress in D. rerio embryos. The TiO2 particles were tested at 1 mg ml−1. Exposures were conducted in 100 ml glass beakers and the medium and respective control changed every other day. Embryos were exposed to BHP and TiO2 material for 1, 2 and 8 days. It is important to highlight that all organisms were sampled at 8 dpf. SOD and GSH were measured across the three exposure periods using 10 pools with 10 to 15 larvae that were collected and immediately frozen in liquid nitrogen and stored at −80 °C until biochemical determination. SA-β-gal levels were only determined in 8 day exposed larvae. 2.10. Phototoxicity and oxidative stress Embryos and larvae were exposed to 10 and 100 μg l−1 of fluoranthene and 1 mg ml−1 of TiO2 particles for 6 days in 100 ml glass beakers and the medium changed every day. On day 6, embryos were well rinsed and transferred in groups of 5 larvae to 6 well plates containing 4 ml of clean fish medium water. In total there were 22 replicates (total 110 embryos) for each treatment. Embryos were exposed to acute simulated solar radiation (SSR) doses ranging from 15 to 450 kJ m−2 (equivalent to radiation doses of 1.6–49 kJ m−2 of UV; SUNTEST CPS+, ATLAS material testing technology, Germany). The selected range corresponds to half of the daily doses of UV in Western Bengal where zebrafish is a native species (Behrendt et al., 2010). As negative controls, one replicate per treatment was covered with aluminium and exposed to 450 kJ m−2 (49 kJ m−2 of UV). Cumulative mortality observations were carried out 24 h after exposures. Photo-oxidative stress responses were limited to fluoranthene (100 μg l−1) and TiO2 material (1 mg ml−1). Embryos were exposed to the selected compounds for 6 days and then co-exposed or not to sublethal levels of SSR radiation set to 105 kJ m−2 (11.4 kJ m−2 UV) for fluoranthene and 180 kJ m−2 (19.5 kJ m−2 of UV) for TiO2 material. SSR exposures were conducted following previously described procedures. Six and 24 h following exposures oxidative stress responses (SOD, GSH; SA-β-gal) were determined in SSR co-exposed and nonSSR co-exposed embryos. 2.11. Biochemical assays Biomarkers and SA-β-gal assessment were determined according to Faria et al. (2009) and Kishi et al. (2008), respectively. SA-β-gal staining images are depicted in Fig. 1. Detailed procedures of these methods are given in the Supplementary material section. 2.12. Chemical analysis and oxidative stress responses Accumulated levels of titanium, body length measurements or biomarker responses were compared using one way ANOVA followed by post hoc Dunnett's or Tukey's tests (P b 0.05). Mortality responses across SSR radiation doses were fitted to the Hill model of Eq. 1.

382

M. Faria et al. / Science of the Total Environment 470–471 (2014) 379–389

A

B

C

Fig. 1. Images of 8 dpf zebrafish embryos showing SA-β-gal activity after being exposed for 8 days to 500 μM BHP (A), 100 μg l−1 fluoranthene (B) and 1 mg ml−1 of NM-TiO2 and microTiO2 (C). Right panel images show the neural tube.

Mortality ¼

100 1 þ ðLD50 =xÞp

ð1Þ

where p = slope; LD = lethal dose; and x = dose (kJ). The Hill model was also used to determine the relative potency of the tested TiO2 materials to produce ROS under SSR in vitro. In that case fluorescein fluorescence inhibition–TiO2 concentration curves were fitted to Eq. 2 to estimate median inhibitory concentration effects.

Fuorescence inhibition ¼

100 1 þ ðIC50 =xÞp

ð2Þ

where p = slope; IC = inhibition concentration; and x = concentration (mg ml−1). Analyses were conducted using the IBM SPSS 19.0 software. 3. Results 3.1. Particle characterization and uptake of NM-TiO2 DLS measurements conducted at 100 μg ml−1 are reported in Table 2 and indicate aggregate sizes greater than 1 μm. Polydispersity indexes were also higher than 0.5, which denoted the existence of polydispersity. The micro-TiO2 particle suspensions had polydispersity indexes of 1 preventing a reliable determination of aggregate sizes.

M. Faria et al. / Science of the Total Environment 470–471 (2014) 379–389

383

Table 2 NM-TiO2 characterization by dynamic light scattering (DLS) and ROS production. Results are depicted as mean ± SD. The size of micron-sized particle suspensions could not be characterized (nc) due to the high polydispersity. N

NM103 NM104 P25 MicroTiO2

Hydrodynamic size (nm)

12 12 12 12

Polydispersity ind

Fluorescein inhibition

Mean

SD

Mean

SD

IC50 (mg ml−1)

SD

1320.43 1361.33 1243.16 nc

340.02 70.50 249.23 nc

0.57 0.54 0.55 nc

0.16 0.08 0.04 nc

0.46 0.73 0.02 1.39

0.02 0.03 b0.01 0.11

The ability of TiO2 particle aggregates to produce ROS was evaluated in vitro upon SSR assessing the loss of fluorescence of the dye fluorescein. Fig. 2A showed substantial differences between NM-TiO2 P25 and the rest of particles inhibiting fluorescein fluorescence due to the production of ROS. According to the median inhibition concentrations of fluorescein exposed to SSR and TiO2 aggregates of Table 2 (IC50), NM-TiO2 P25 had the greatest potential to produce ROS and micro-TiO2 the lowest. Despite the increase of particle size caused by aggregation, we found evidence that NM-TiO2 and micro-TiO2 were uptaken by the embryos. Titanium levels in embryos measured by ICP-MS were negligible in

control animals (Mean ± SEM, 0.08 ± 0.01 ng Ti/embryo, N = 6), showing a large increase (75. 8 ± 2.9 ng Ti/embryo) in those exposed to NM-TiO2 that decreased to 28.6 ± 2.5 ng Ti/embryo following transfer to clean medium water for 1 day (Fig. 2B). Accumulated levels of titanium after 8 days of exposure changed significantly (P b 0.05) across particle suspensions, accumulation and depuration periods (F9,20 = 115.9, Fig. 2B). Measured levels of titanium in embryos were two- to three-fold higher than in those that were allowed to clean for 1 day. NM-103 accumulated the most, followed in decreasing order by NM-104, P25 and micro-TiO2. 3.2. Toxicity

A

NM103 NM104 P25 microTiO2

1.2

% inhibition of fluorescence

1.0 0.8 0.6 0.4 0.2 0.0 -0.2 0.0

0.2

0.4

0.6

0.8

1.0

1.2

mg/ml

B

100

Accumulation Depuration

a

a 80

ng Ti embryo-1

a 60

40

b bc

cd d

20

d

0 NM103

NM104

P25

microTiO2

Particle type Fig. 2. Fluorescein fluorescence inhibition due to ROS production under SSR radiation (A) and accumulated levels of Ti in 8 dpf embryos after 8 d exposures (Accumulation— dark bars) and 8 d exposures × 1 d clearance (Depuration—grey bars) across the studied TiO2 particles at 1 mg ml−1. In graph A inhibition patterns have been fitted to Eq. 2 and in graph B different letters mean significant (P b 0.05) differences between treatments following ANOVA and Tukey's test.

Survival of developing zebrafish embryos in controls and groups exposed to the studied contaminants was not significantly (P b 0.05) different and varied between 92 and 98% (data not shown). Nano-material aggregates of TiO2 were marginally toxic to 8 dpf zebrafish embryos, affecting significantly (P b 0.05; one way ANOVA tests) their body length at 1 mg ml−1 (Fig. 3A). No other morphological effects were observed in 8 dpf exposed embryos. Sublethal doses of BHP (500 μM) also decreased significantly (P b 0.05; F1,48 = 4.13; Fig. 3B) the body size of 8 dpf embryos. Fluoranthene alone at 10 or 100 μg l−1 did not affect either the survival or the length (F2,54 = 0.31; Fig. 3C) of 6 dpf embryos. Phototoxic effects of the studied compounds are represented as UV radiation dose-mortality response curves of 6 dpf embryos exposed to fluoranthene and TiO2 material Fig 4. Fluoranthene at 100 μg l−1 had the highest phototoxic effects (lowest LD50, Table 3), followed by P-25 at 1 mg ml−1, fluoranthene at 10 μg l−1 and NM-103 at 1 mg ml−1. The remaining TiO2 materials were not phototoxic to 6 dpf embryos since their LD50 were comparable to those of control embryos (Fig 4, Table 3). 3.3. Oxidative stress markers Time course-exposure experiments conducted in 8 dpf embryos exposed to BHP and TiO2 particle suspensions under no SSR radiation are depicted in Fig. 5. SOD activity in embryos exposed to sublethal doses (500 μM) of the pro-oxidant model, BHP, increased significantly (F3,34 = 9.56) shortly after (1 d) exposure decreasing to control values in non-exposed embryos after 2 days (Fig. 5A). Levels of GSH, however, decreased significantly (P b 0.05; F3,34 = 28.45) and monotonically with exposure time (Fig. 5B). Levels of SA-β-gal were over expressed significantly (P b 0.05; F1,48 = 86.54) in embryos exposed during the entire period (8 d, Fig 5C). Exposure to particle suspension followed a different pattern. SOD activity increased significantly (P b 0.05, F12, 86 = 7.9) in 8 dpf embryos exposed to NM-104, P25 and micro-TiO2 at 8, 2–8 and 1 d of exposure, respectively (Fig. 5D). Levels of GSH were only significantly different (P b 0.05; F12, 82 = 3.9) than control treatments in embryos exposed to micro-TiO2 during 8 d (Fig. 4E). SAβ-gal levels in 8 dpf embryos exposed to NM-TiO2 and micro-TiO2 during 8 d were higher than those of control treatments (P b 0.05; F4,108 =5.8; Fig. 5F). Effects of SSR radiation on the studied markers across controls, fluoranthene and TiO2 aggregates were determined using differential

384

M. Faria et al. / Science of the Total Environment 470–471 (2014) 379–389

A

0.01 mg ml-1

C

1 mg ml-1

0.1mg ml-1

Body length (µm)

3800

3600

*

*

*

3400

3200

3000 NM103

B

*

3200

3800

3600

Body length (µm)

3400

microTiO2

P25

C

3800

3600

Body length (µm)

NM104

3400

3200

3000

3000 C

C

500

10

100

-1 Fluoranthene (µg l )

BHP (µM)

Fig. 3. Body length measurements of 8 and 6 dpf zebrafish embryos exposed to 1 mg ml−1 of the studied particle suspensions (A), 500 μM of BHP (B) and 10 and 100 μg l−1 fluoranthene (C). Error bars are SEM, N = 30. Asterisks indicate significant (P b 0.05) differences from the respective control treatments. In graph A, NM-103 and NM-104 were tested together and NM-P25 and micro-TiO2 separately.

Control microTiO2

NM103

NM104

Fu 10

Fu 100

P25

1.2

Cumulative mortality after 24 h

1.0

0.8

0.6

responses (ΔR = Rnon-irradiated − Rirradiated embryos). Embryos of 6 dpf coexposed to the phototoxic compound fluoranthene (100 μg l− 1) and to a sublethal SSR doses had differentially increased SOD activities at 6 h (t12 = 3.7) but not after 24 h (t10 = 1.5) relative to those of controls (Fig. 6A). Differential levels of GSH were higher in embryos co-exposed and fluoranthene after 6 h (t11 = 2.7) and 24 h (t10 = 2.8) of SSR exposure than in controls (Fig. 6B). Differential SA-β-gal levels of co-exposed fluoranthene and SSR embryos were significantly (P b 0.05) higher than those of controls only after 24 h (t28 = 3.3, Fig. 6C). Six hours after being exposed to SSR treatments, differentially SOD activity measured in 6 dpf embryos co-exposed to nanoparticle aggregates increased significantly (P b 0.05) relative to those of controls only for P25 (F4,31 = 31.0, Fig. 7A). Differential levels of GSH and

0.4

0.2

Table 3 Regression parameters according to Eq. 1 of fitted models to mortality responses following co-exposures to SSR (expressed as UV radiation) with the studied particle aggregates and fluoranthene. All regressions were significant (P b 0.001). N, sample size, r2, r square. All TiO2 particle suspensions were at doses 1 mg ml−1.

0.0

5

10

50

UV exposure (KJ m-2) Fig. 4. Simulated solar radiation (SSR) doses–cumulative mortality responses (expressed as UV radiation in log scale) of zebrafish embryos co-exposed to fluoranthene and particle suspensions. Each point represents a single replicate. Responses have been fitted to the Hill regression equation (Eq. 1). Dotted line represents control.

Treatments

LD50 (kJ/m2)

SE

p

r2

N

Control Fluoranthene 10 μg l−1 Fluoranthene 100 μg l−1 NM103 NM104 P25 MicroTiO2

31.2429 25.8205 10.8499 28.3835 32.7402 21.5556 32.1251

1.4065 0.4395 0.2841 1.1774 1.7763 1.3727 2.1779

5.8576 8.9901 6.415 4.7335 3.978 4.2892 5.6629

0.7681 0.9774 0.9772 0.8402 0.7481 0.7174 0.6008

34 20 20 26 27 40 25

M. Faria et al. / Science of the Total Environment 470–471 (2014) 379–389

C

BHP

A

2d

8d

D

*

4

1d

385

6

*

*

*

*

SOD (U mg prot-1)

5 3 4 2

3 2

1 1 0

0 C

B

1

2

8

800

GSH(nmol mg prot-1)

NM103

E

NM104

P25

microTiO2

1000

*

*

600

800

*

600

400 400 200

200

0

0 C

C

1

2

8

NM103

F

4

NM104

P25

microTiO2

4

SGAL (pixel µm-1)

* 3

3

2

2

1

1

*

*

*

*

NM103

NM104

P25

microTiO2

0

0 C

8

Exposure days

C

Particle suspensions

Fig. 5. Superoxide dismutase (SOD) activities (A, D), levels of reduced glutathione (GSH) (B, E) and SA-β-gal (C, F) of 8 dpf zebrafish embryos exposed to 500 μM of BHP (left graph panel) or TiO2 particle aggregates (1 mg ml−1, right graph panel) for 1, 2 and 8 days. Errors bars are SE, N = 10, 30. Asterisks indicate significant (P b 0.05) differences from the respective control treatments after ANOVA and Dunnet's tests.

of SA-β-gal in 6 dpf embryos were similar to those of SOD activity however with significant effects (P b 0.05) only after 24 h of exposure (F4,30 = 10.8, for SOD, Fig. 7B; F4,95 = 3.7, for SA-β-gal, Fig. 7C).

other descriptions of NM-TiO2 behaviour in similar aqueous media (Bar-Ilan et al., 2012; Xiong et al., 2011). Therefore, the culture medium leads to the aggregation of nanoparticles.

4. Discussion

4.2. ROS production and accumulation

4.1. Particle characterization

ROS production by nanoparticle aggregates confirmed previous studies indicating that the NM-TiO2 P25 having a phase composition of anatase was much more phototoxic than the other nano-particle aggregates composed of rutile (Bar-Ilan et al., 2012; Sayes et al., 2006; Xiong et al., 2011). Measured accumulated levels of whole body titanium of nano-particle suspensions measured in 8 dpf embryos (approx. 80 ng/embryo) were similar to those reported by Bar-Ilan et al. (2012) (approx. 50 ng/embryo) but decreased by half in embryos transferred to clean medium for 24 h. These results may suggest that only a fraction of measured titanium was internalised. In a dry weight basis our measured titanium concentrations of 80 ng/embryo were equivalent to 1 mg g dw−1 (data not shown). Ramsden et al. (2013) reported that the accumulated levels of titanium in whole adult zebrafish

The particle size distribution determined by laser scattering analysis showed that NM-TiO2 were consistently agglomerated in test suspensions. With the approach used in this work, determination of frequency size distribution is done by measuring the intensity of the signal generated by particles of different sizes. The strong signal produced by particles of a given size (normally big particles, aggregates or agglomerates) attenuates this of other sizes independent of their relative abundance, leading to underestimation of the number of particles of low sizes. Therefore, the results presented are indicative of the presence of particles of distinct sizes, although big aggregates appear to predominate. The observed aggregation effects of NM-TiO2 are in agreement with

386

M. Faria et al. / Science of the Total Environment 470–471 (2014) 379–389

A

6h

24 h

3

3

Differential SOD

* 2

2

1

1

0

0

-1

-1 0

B

0

100

250

250

Differential GSH

* 200

200

150

150

100

100

50

50

0

Differential SGAL

*

0

0

C

100

0

100

0.6

0.6

0.5

0.5

0.4

0.4

0.3

0.3

0.2

0.2

0.1

0.1

0.0

100

*

0.0 0

100

0

100

Fluoranthene (µg l-1) Fig. 6. Differential superoxide dismutase (SOD, U mg protein−1) activities (A), levels of reduced glutathione (GSH, nmol mg protein−1) (B) and of SA-β-gal (pixels μm−1) (C) of 6 dpf zebrafish embryos exposed to 100 μg l−1 of fluoranthene (grey bars), 6 and 24 h after being co-exposed to SSR treatments. Error bars are SE, N = 10, 30. Asterisks indicate significant (P b 0.05) differences from the respective control treatments after ANOVA and Dunnet's tests.

(7–12 μg Ti g dw−1) exposed to TiO2 NM, at concentrations three orders of magnitude lower than in this present study (1 mg ml−1) returned to control levels shortly after being cultured in clean medium. In contrast Chen et al. (2011)) showed low accumulation of titanium (5–100 ng g dw− 1) in internal organs of zebrafish exposed to 1–7 mg l− 1 of TiO2 NM during 6 months. Our results and those of Ramsden et al. (2013) and Chen et al. (2011) indicate that titanium nanoparticle had a low potential to be accumulated in fish. 4.3. Acute toxicity Reported toxicity for the NM-TiO2 was low except at the highest tested concentration of 1 mg ml−1 that impaired growth in 8 dpf

embryos. Previous studies have also found that toxicity of the NMTiO2 P25 was marginal without illumination (Bar-Ilan et al., 2013, 2012). Toxicities of BHP (500 μM) and fluoranthene (≤100 μg l−1) are known to be non-lethal to zebrafish embryos without UV radiation (Kishi et al., 2008; Matson et al., 2008). Embryotoxic effects on the growth of developing zebrafish embryos are often associated with other malformations (Raldúa et al., 2008; Shi et al., 2008). Nevertheless, for oxidizing agents such as H2O2 and BHP administered at sublethal doses Kishi et al. (2008) reported no apparent phenotypic effects other than altered oxidative stress markers in 6 dpf embryos. In the present study, however, we observed that BHP at 500 μM marginally impaired the growth of 8 dpf embryos. Prolonged exposure to NMTiO2 under UV radiation or to aryl hydrocarbon receptor agonists such

M. Faria et al. / Science of the Total Environment 470–471 (2014) 379–389

6h 1.5

phototoxic in in vivo exposures. Similar results were obtained by BarIlan et al. (2013, 2012)) who also found that NM-TiO2 P25 were toxic to 8 dpf embryos under UV radiation and that toxicity was related to ROS production.

24 h

*

Differential SOD

2 1.0

4.4. Oxidative stress

1 0.5 0 0.0 -1 -0.5 70

Differential GSH

60

*

60 50

40

40 20

30 20

0 10

N

M

Ti ro ic m

4 10

M

P2 5

3 M

10 N

N

O 2

0.0 Ti

0.0

ro

0.1

ic

0.1

m

0.2

4

0.2

P2 5

0.3

3

0.3

10

0.4

M

0.4

*

N

0.5

C

0.5

10

0

O 2

-20

C

Differential SGAL

387

TiO2 particle suspensions Fig. 7. Differential superoxide dismutase (SOD, U mg protein−1) activities (A), levels of reduced glutathione (GSH, nmol mg protein−1) (B) and of SA-β-gal (pixels μm−1) (C) of 6 dpf zebrafish embryos exposed to 1 mg ml−1 of TiO2 particle aggregates (grey bars) 6 and 24 h after being co-exposed to SSR treatments. Error bars are SE, N = 10, 30. Asterisks indicate significant (P b 0.05) differences from the respective control treatments after ANOVA and Dunnet's tests.

as co-planar PAHs can cause pericardial effusion and craniofacial malformations and death (Bar-Ilan et al., 2013, 2012; Timme-Laragy et al., 2009). In this present study UV exposures and effect assessment using SSR were conducted in advanced developmental stages of zebrafish embryos (6 dpf) and were relatively short (≤24 h). This means that our experimental design of UV exposure rules out the detection of embryonic developmental effects. Nevertheless, toxicity results under SSR agree with previous studies. As expected fluoranthene at 100 μg l−1 was the most phototoxic compound. Under UV radiation, fluoranthene becomes toxic to fish and amphibian embryos at doses N50 μg l−1 (Diamond et al., 2006; Hatch and Burton, 1998. The NMTiO2 P25 were also phototoxic whereas the remaining nano and microsized TiO2 aggregates showed similar phototoxic responses as those of unexposed embryos. Phototoxic behaviour of NM-TiO2 agrees with the results reported in vitro, just showing that the NM-TiO2 P25 had the greatest potential to produce ROS under SSR and hence it was

In zebrafish embryos several antioxidant enzyme activities (i.e. SOD, catalase, glutathione peroxidase among others) and markers of oxidative tissue damage (i.e. lipid peroxidation, DNA adducts) have been described and used in toxicological evaluations of chemical, nanoparticle and several environmental factors (Bar-Ilan et al., 2012; Cazenave, 2006; Charron et al., 2000; Kishi et al., 2008; Usenko et al., 2008; Wu et al., 2011; Zhao et al., 2013). We found that embryos had high SOD activities and elevated levels of GSH compared to those of other antioxidant markers. We also found that SA-β-gal was a reliable and robust marker of oxidative tissue damage. Under no UV radiation (no SSR exposure) TiO2 particle suspensions produced oxidative tissue damage in zebrafish embryos similar to the pro-oxidant BHP measured as increase levels of SA-β-gal (Kishi et al., 2008). SA-β-gal is considered a robust marker of oxidative tissue damage that has been widely used in senescence studies conducted in zebrafish and other vertebrates (Kishi et al., 2008). SA-β-gal is in fact the well characterized lysosomal β galactosidase enzyme (Lee et al., 2006), that increases in response to cellular senescence or oxidative stress (Kishi et al., 2008). Increase of SA-β-gal is related to the accumulation of oxidized proteins and lipid-by-products in lysosomes. Oxidative tissue damage levels obtained in embryos exposed to BHP are in agreement with previous results (Kishi et al., 2008) Oxidative damage caused by TiO2 suspensions is indicative of the toxic effect of the chemical TiO2 rather than to nano-scale particles alone. This apparently contradicts the findings of Xiong et al. (2011) who reported that NM-TiO2 but not bulk material of TiO2 caused oxidative damage in different zebrafish organs. Nevertheless, the previous study used a different type of bulk TiO2 material, determined other oxidative damage markers and worked in adult fish. Indeed a closer look to Xiong et al. (2011) evidenced enhanced but not significant levels of lipid peroxidation and protein carboxyl in gills and liver of adults exposed to bulk material of TiO2. The two studied antioxidant markers were also affected differently by BHP and TiO2 suspensions under no illumination conditions. SOD activity, which is involved in the first reduction of the superoxide ion radical (O2−) to hydrogen peroxide, increased shortly after exposure (≤1 d) to the pro-oxidant compound BHP, acting as a first antioxidant defensive system. At latter exposure periods (N 2 days) levels of GSH in zebrafish embryos were depleted by BHP. Conversely, SOD activity induction varied across TiO2 particle suspensions: it was unaffected by NM-103, increased at 1 d in embryos exposed to micro-TiO2 or later in embryos exposed to NM-104 and P25. GSH levels were unaffected in embryos exposed to all but micro-TiO2 particle suspensions that showed enhanced levels at 8 d. In zebrafish embryos, enhanced activities of SOD have been reported under exposure to tetrabromobisphenol A and cypermethrin (Hu et al., 2009; Shi et al., 2008). Gu et al. (2010) also describe that SOD activity increased in zebrafish embryos shortly (6 h) after being exposed to fenvalerate showing a subsequent inhibition in longer exposures (24 h). Zebrafish embryos contain high levels of GSH, thus effects on this thiol are observed under prolonged and/or severe pro-oxidant conditions such as the one observed under BHP exposure. Indeed TimmeLaragy et al. (2009) showed that GSH levels in zebrafish embryos were reduced only after severe exposures for example by co-exposure to Cyp1A1 inducers such as β-naphthoflavone and to the GSH synthesis inhibitor L-buthionine (S,R)-sulfoximine. Thus observed SOD and GSH responses under the prooxidant BHP model confirm the findings of other authors indicating that SOD activity increases shortly after exposure responding to the production of superoxide radical anions (O2−)

388

M. Faria et al. / Science of the Total Environment 470–471 (2014) 379–389

and GSH levels are depleted later under prolonged/severe pro-oxidant conditions. TiO2 aggregate suspensions increased SOD activity later in time and did not affect GSH levels, the exception being micro-TiO2, which may indicate that antioxidant responses observed in zebrafish embryos could be associated to a combination of mechanisms. Zebrafish embryos sampled in the oxidative stress study were 8 days old, which means that they are able to eat particles. NM-TiO2 and micro-TiO2 ingested by zebrafish may be deposited to different organs than the gut (Chen et al., 2011) and may generated ROS in the absence of light (Long et al., 2006). ROS can also be generated by NM-TiO2 that are directly in contact with the skin or the gills (Xiong et al., 2011). In both cases NM-TiO2 primarily produced hydroxyl radicals (•OH) that can react with oxygen generating superoxide anions that are the substrate of SOD. This means that retarded effects of SOD induction may be related to the period needed for particle aggregates to be ingested and reach embryo organs and/or for hydroxyl radicals to produce superfluous superoxide anions. Data on micro-TiO2, however, did not agree with the previous arguments since micro-TiO2 is less prooxidant than NM-TiO2 (Xiong et al., 2011). In that case the alternative hypothesis suggested by Ramsden et al. (2013)) that accumulation of TiO2 on external surfaces (i.e. skin) or gills caused respiratory distress that leads to the observed changes in GSH and SOD could also explain our results. Indeed there is reported evidence that local accumulation of nanoparticles in the lung tissue caused inflammation and oxidative stress (Xia et al., 2006). Measured antioxidant responses in embryos co-exposed to fluoranthene or TiO2 particle aggregates with SSR radiation followed a similar pattern, despite that effects were more evident for fluoranthene. Coexposure of fluoranthene or NM-TiO2 P25 with SSR increased SOD activity at 6 h. The observed transient responses of SOD were similar to those observed for BHP and support the view that this enzyme functions as the first antioxidant response to ROS generated by these compounds under SSR. GSH levels increased at 6 and 24 h or only at 24 h in embryos co-exposed to fluoranthene or NM-TiO2 P25 with SSR, respectively. Induction of glutathione has been reported in zebrafish adults exposed to NM-TiO2 (Ramsden et al., 2013) and in Japanese flounder larvae exposed to mercury (Huang et al., 2010). Oxidative stress damage measured as levels of SA-β-gal was enhanced only in those embryos co-exposed to SSR and fluoranthene or the NMTiO2 P25. There were marked differences to generate ROS in vitro under SSR between NM-TiO2 P25 and the rest of the particle aggregates and consequently toxicological and anti- and prooxidant responses in vivo were most evident for the former particle. Interestingly micro-TiO2, that has the same phase composition as NM-TiO2 P25 (anatase) was not phototoxic to zebrafish embryos. These results are consistent with the view that aggregates of nano-sized TiO2 are more reactive than those of microsized ones (Ramsden et al., 2013; Xiong et al., 2011) and that those having a phase composition of anatase are phototoxic due to the production of ROS (Bar-Ilan et al., 2012). 5. Conclusions Selection of several oxidative stress markers and the inclusion of a temporal analysis are important for studying the oxidative stress related toxicity of nanoparticles since many of these markers are interconnected and the induction/depletion of many of them is transient. The inclusion of prooxidant positive controls combined with temporal analysis of the selected oxidative stress markers allowed us to study oxidative stress mediated effects of NM-TiO2 aggregates in zebrafish embryos. Under no SSR, observed oxidative tissue damage (SA-β-gal) did not vary across the tested TiO2 and temporal patterns of the studied antioxidant markers (SOD and GSH) behave differently than those of the prooxidant substance BHP. This means that other mechanisms than ROS production per se can be involved in the

observed oxidative stress responses. Under illumination (SSR) the oxidative stress markers behaved similarly across the prooxidant phototoxic compound fluoranthene and the phototoxic NM-TIO2 P25. These results support the argument that under illumination the production of ROS is causing oxidative stress in embryos exposed to TiO2 aggregates. Conflict of interest The authors declare that there is no conflict of interest. Acknowledgements This project was funded by the Spanish projects CTM2011-30471C02-01 and AEG 07-060 and the Portuguese Foundation for Science and Technology (FCT) fellowship SFRH/BPD/78342/2011 funded by the Program POPH-QREN through the Portuguese Ministry of Education and Science and the European Social Fund. This work was done in the framework of the OECD Working Party on Manufactured Nanomaterials project Safety Testing of a Representative Set of Manufactured Nanomaterials. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2013.09.055. References Aitken RJ, Chaudhry MQ, Boxall ABA, Hull M. Manufacture and use of nanomaterials: current status in the UK and global trends. Occup Med-Oxf 2006;56:300–6. Anderson ME. Glutathione: an overview of biosynthesis and modulation. Chem Biol Interact 1998;111–112:1–14. Bar-Ilan O, Louis KM, Yang SP, Pedersen JA, Hamers RJ, Peterson RE, et al. Titanium dioxide nanoparticles produce phototoxicity in the developing zebrafish. Nanotoxicology 2012;6:670–9. Bar-Ilan O, Chuang CC, Schwahn DJ, Yang S, Joshi S, Pedersen JA, et al. TiO2 nanoparticle exposure and illumination during zebrafish development: mortality at parts per billion concentrations. Environ Sci Tech 2013;47:4726–33. Behrendt L, Jönsson ME, Goldstone JV, Stegeman JJ. Induction of cytochrome P450 1 genes and stress response genes in developing zebrafish exposed to ultraviolet radiation. Aquat Toxicol 2010;98:74–82. Braun A, Hartmann NB, Grieger K, Kusk KO. Ecotoxicity of engineered nanoparticles to aquatic invertebrates: a brief review and recommendations for future toxicity testing. Ecotoxicology 2008;17:387–95. Cai R, Kubota Y, Shuin T, Sakai H, Hashimoto K, Fujishima A. Induction of cytotoxicity by photoexcited TiO2 particles. Cancer Res 1992;52:2346–8. Campos B, Rivetti C, Rosenkranz P, Navas JM, Barata C. Effects of nanoparticles of TiO2 on food depletion and life-history responses of Daphnia magna. Aquat Toxicol 2013;130–131:174–83. Cazenave J, De Los Ángeles Bistoni M, Zwirnmann E, Wunderlin DA. Attenuating effects of natural organic matter on microcystin toxicity in zebra fish (Danio rerio) embryos — benefits and costs of microcystin detoxication. Environ Toxicol 2006;21:22–32. Charron RA, Fenwick JC, Lean DRS, Moon TW. Ultraviolet-B radiation effects on antioxidant status and survival in the zebrafish, Brachydanio rerio. Photochem Photobiol 2000;72:327–33. Chen C, Zhao W, Lei P, Zhao J, Serpone N. Photosensitized degradation of dyes in polyoxometalate solutions versus TiO2 dispersions under visible-light irradiation: mechanistic implications. Chem Eur J 2004;10:1956–65. Chen J, Dong X, Xin Y, Zhao M. Effects of titanium dioxide nano-particles on growth and some histological parameters of zebrafish (Danio rerio) after a long-term exposure. Aquat Toxicol 2011;101:493–9. Colvin VL. The potential environmental impact of engineered nanomaterials. Nat Biotechnol 2003;21:1166–70. Diamond SA, Mount DR, Mattson VR, Heinis LJ, Highland TL, Adams AD, et al. Photoactivated polycyclic aromatic hydrocarbon toxicity in medaka (Oryzias latipes) embryos: relevance to environmental risk in contaminated sites. Environ Toxicol Chem 2006;25:3015–23. Dinu D, Marinescu D, Munteanu MC, Staicu AC, Costache M, Dinischiotu A. Modulatory effects of deltamethrin on antioxidant defense mechanisms and lipid peroxidation in Carassius auratus gibelio liver and intestine. Arch Environ Contam Toxicol 2010;58:757–64. Faria M, Carrasco L, Diez S, Riva MC, Bayona JM, Barata C. Multi-biomarker responses in the freshwater mussel Dreissena polymorpha exposed to polychlorobiphenyls and metals. Comp Biochem Physiol C 2009;149:281–8.

M. Faria et al. / Science of the Total Environment 470–471 (2014) 379–389 Fu PP, Xia Q, Sun X, Yu H. Phototoxicity and environmental transformation of polycyclic aromatic hydrocarbons (PAHs)-light-induced reactive oxygen species, lipid peroxidation, and DNA damage. J Environ Sci Health, Part C Environ Carcinog Ecotoxicol Rev 2012;30:1–41. Gu A, Shi X, Yuan C, Ji G, Zhou Y, Long Y, et al. Exposure to fenvalerate causes brain impairment during zebrafish development. Toxicol Lett 2010;197:188–92. Halliwell B, Gutteridge JMC. Free radicals in biology and medicine. Oxford: Oxford University Press; 1999. Hatch AC, Burton Jr GA. Effects of photoinduced toxicity of fluoranthene on amphibian embryos and larvae. Environ Toxicol Chem 1998;17:1777–85. Hu J, Liang Y, Chen M, Wang X. Assessing the toxicity of TBBPA and HBCD by zebrafish embryo toxicity assay and biomarker analysis. Environ Toxicol 2009;24:334–42. Huang W, Cao L, Ye Z, Yin X, Dou S. Antioxidative responses and bioaccumulation in Japanese flounder larvae and juveniles under chronic mercury exposure. Comp Biochem Physiol C: Toxicol Pharmacol 2010;152:99–106. Kim KT, Klaine SJ, Cho J, Kim SH, Kim SD. Oxidative stress responses of Daphnia magna exposed to TiO2 nanoparticles according to size fraction. Sci. Total Environ 2010;408: 2268–72. Kishi S, Bayliss PE, Uchiyama J, Koshimizu E, Qi J, Nanjappa P, et al. The identification of zebrafish mutants showing alterations in senescence-associated biomarkers. PLoS Genet 2008;4. Klaine SJ, Alvarez PJJ, Batley GE, Fernandes TF, Handy RD, Lyon DY, et al. Nanomaterials in the environment: behavior, fate, bioavailability, and effects. Environ Toxicol Chem 2008;27:1825–51. Konaka R, Kasahara E, Dunlap WC, Yamamoto Y, Chien KC, Inoue M. Ultraviolet irradiation of titanium dioxide in aqueous dispersion generates singlet oxygen. Redox Rep 2001;6:319–25. Lee BY, Han JA, Im JS, Morrone A, Johung K, Goodwin EC, et al. Senescence-associated β-galactosidase is lysosomal β-galactosidase. Aging Cell 2006;5:187–95. Livingstone DR. Contaminant-stimulated reactive oxygen species production and oxidative damage in aquatic organisms. Mar Pollut Bull 2001;42:656–66. Long TC, Saleh N, Tilton RD, Lowry GV, Veronesi B. Titanium dioxide (P25) produces reactive oxygen species in immortalized brain microglia (BV2): implications for nanoparticle neurotoxicity. Environ Sci Tech 2006;40:4346–52. Lovern SB, Klaper R. Daphnia magna mortality when exposed to titanium dioxide and fullerene (C60) nanoparticles. Environ Toxicol Chem 2006;25:1132–7. Matson CW, Timme-Laragy AR, Di Giulio RT. Fluoranthene, but not benzo[a]pyrene, interacts with hypoxia resulting in pericardial effusion and lordosis in developing zebrafish. Chemosphere 2008;74:149–54. Metzler DM, Li M, Erdem A, Huang CP. Responses of algae to photocatalytic nano-TiO2 particles with an emphasis on the effect of particle size. Chem Eng J 2011;170:538–46. Miller RJ, Bennett S, Keller AA, Pease S, Lenihan HS. TiO2 nanoparticles are phototoxic to marine phytoplankton. PloS one 2012;7. Oberdörster E. Manufactured nanomaterials (fullerenes, C60) induce oxidative stress in the brain of juvenile largemouth bass. Environ Health Perspect 2004;112:1058–62. OECD 212, “Test Guideline 212”. Fish short-term toxicity test on embryos and sac-fry stages guideline for testing chemicels. OECD Organization for Economic Co-operation and Development; 1998C-15. Ohtani B, Prieto-Mahaney OO, Li D, Abe R. What is Degussa (Evonic) P25? Crystalline composition analysis, reconstruction from isolated pure particles and photocatalytic activity test. J Photochem Photobiol A Chem 2010;216:179–82.

389

Ou B, Hampsch-Woodill M, Prior RL. Development and validation of an improved oxygen radical absorbance capacity assay using fluorescein as the fluorescent probe. J Agric Food Chem 2001;49:4619–26. Parihar MS, Javeri T, Hemnani T, Dubey AK, Prakash P. Responses of superoxide dismutase, glutathione peroxidase and reduced glutathione antioxidant defenses in gills of the freshwater catfish (Heteropneustes fossilis) to short-term elevated temperature. J Therm Biol 1997;22:151–6. Querol X, Alastuey A, Rodriguez S, Plana F, Mantilla E, Ruiz CR. Monitoring of PM10 and PM2.5 around primary particulate anthropogenic emission sources. Atmos Environ 2001;35:845–58. Raldúa D, André M, Babin PJ. Clofibrate and gemfibrozil induce an embryonic malabsorption syndrome in zebrafish. Toxicol Appl Pharmacol 2008;228:301–14. Ramsden CS, Henry TB, Handy RD. Sub-lethal effects of titanium dioxide nanoparticles on the physiology and reproduction of zebrafish. Aquat Toxicol 2013;126:404–13. Reeves JF, Davies SJ, Dodd NJF, Jha AN. Hydroxyl radicals ({radical dot}OH) are associated with titanium dioxide (TiO2) nanoparticle-induced cytotoxicity and oxidative DNA damage in fish cells. Mutat Res Fundam Mol Mech Mutagen 2008;640:113–22. Sayes CM, Wahi R, Kurian PA, Liu Y, West JL, Ausman KD, et al. Correlating nanoscale titania structure with toxicity: a cytotoxicity and inflammatory response study with human dermal fibroblasts and human lung epithelial cells. Toxicol Sci 2006;92:174–85. Shi X, Du Y, Lam PKS, Wu RSS, Zhou B. Developmental toxicity and alteration of gene expression in zebrafish embryos exposed to PFOS. Toxicol Appl Pharmacol 2008;230: 23–32. Timme-Laragy AR, Van Tiem LA, Linney EA, Di Giulio RT. Antioxidant responses and NRF2 in synergistic developmental toxicity of PAHs in zebrafish. Toxicol Sci 2009;109: 217–27. Uchino T, Tokunaga H, Ando M, Utsumi H. Quantitative determination of OH radical generation and its cytotoxicity induced by TiO2-UVA treatment. Toxicol in Vitro 2002;16:629–35. Usenko CY, Harper SL, Tanguay RL. In vivo evaluation of carbon fullerene toxicity using embryonic zebrafish. Carbon 2007;45:1891–8. Usenko CY, Harper SL, Tanguay RL. Fullerene C60 exposure elicits an oxidative stress response in embryonic zebrafish. Toxicol Appl Pharmacol 2008;229:44–55. Wassenberg DM, Di Giulio RT. Synergistic embryotoxicity of polycyclic aromatic hydrocarbon aryl hydrocarbon receptor agonists with cytochrome P4501A inhibitors in Fundulus heteroclitus. Environ Health Perspect 2004;112:1658–64. Wu M, Xu H, Shen Y, Qiu W, Yang M. Oxidative stress in zebrafish embryos induced by short-term exposure to bisphenol A, nonylphenol, and their mixture. Environ Toxicol Chem 2011;30:2335–41. Xia T, Kovochich M, Brant J, Hotze M, Sempf J, Oberley T, et al. Comparison of the abilities of ambient and manufactured nanoparticles to induce cellular toxicity according to an oxidative stress paradigm. Nano Lett 2006;6:1794–807. Xiong D, Fang T, Yu L, Sima X, Zhu W. Effects of nano-scale TiO2, ZnO and their bulk counterparts on zebrafish: acute toxicity, oxidative stress and oxidative damage. Sci Total Environ 2011;409:1444–52. Zhao X, Wang S, Wu Y, You H, Lv L. Acute ZnO nanoparticles exposure induces developmental toxicity, oxidative stress and DNA damage in embryo-larval zebrafish. Aquat Toxicol 2013;136–137:49–59.