TiO2 nanoparticles in seawater: Aggregation and interactions with the green alga Dunaliella tertiolecta

Ecotoxicology and Environmental Safety 148 (2018) 184–193

Contents lists available at ScienceDirect

Ecotoxicology and Environmental Safety journal homepage: www.elsevier.com/locate/ecoenv

TiO2 nanoparticles in seawater: Aggregation and interactions with the green alga Dunaliella tertiolecta

MARK

⁎

Elisabetta Morellia, , Edi Gabellieria, Alessandra Bonominia,b, Danika Tognottia, Giacomo Grassib, Ilaria Corsib a b

Institute of Biophysics, National Research Council, 56124 Pisa, Italy Department of Physical, Earth and Environmental Sciences, University of Siena, 53100 Siena, Italy

A R T I C L E I N F O

A B S T R A C T

Keywords: Titanium dioxide nanoparticles Marine microalgae Exopolymeric substances Toxicological effects Oxidative stress

Titanium dioxide nanoparticles (TiO2 NPs) have been widely employed in industrial applications, thus rising concern about their impact in the aquatic environment. In this study we investigated the chemical behaviour of TiO2 NPs in the culture medium and its effect on the green alga Dunaliella tertiolecta, in terms of growth inhibition, oxidative stress, ROS (Reactive Oxygen Species) accumulation and chlorophyll content. In addition, the influence of exopolymeric substances (EPS) excreted by the microalgae on the stability of NPs has been evaluated. The physicochemical characterization showed a high propensity of TiO2 NPs to form micrometric-sized aggregates within 30 min, large enough to partially settle to the bottom of the test vessel. Indeed, an increasing amount of TiO2 particles settled out with time, but the presence of EPS seemed to mitigate this behaviour in the first 6 h of exposure where the main effects in D. tertiolecta were observed. TiO2 NPs did not inhibit the 72-h growth rate of D. tertiolecta, nor affected the cellular chlorophyll concentration in the range 0.01–10 mg L−1. The time-course of ROS production showed an initial transient increase of ROS in TiO2 NP-exposed algae compared to the control, concomitant with an enhancement of catalase activity. Interestingly, intracellular ROS was a small fraction of total ROS, the highest amount being extracellular. The occurrence of cell-mediated chemical transformations of TiO2 NPs in the external medium, related to the presence of EPS, has been evaluated. Our results showed that carbohydrates were the major component of EPS, whereas proteins of medium molecular weight (20–80 kDa) were preferentially bound to TiO2 NPs, likely influencing their biological fate.

1. Introduction Titanium dioxide nanoparticles (TiO2 NPs) are extensively used in a wide range of industrial applications and consumer products such as water treatments, photocatalysis, solar cells, self-cleaning paints, cosmetics and sunscreens (Handy et al., 2008; Mu and Sprando, 2010; Botta et al., 2011; Piccinno et al., 2012). Many applications are based on the ability of these metal-oxide NPs to absorb UV-light and to exhibit photocatalytic properties (Wold, 1993; EPA, 2009). The wide spread use of TiO2 NPs has raised concerns about their impact on the aquatic environment and human health (Klaine et al., 2008). Moreover, the use of these NPs in new marine nanotechnologies, such as pollution remediation systems, has increased the need to understand their fate and sustainability for the marine environment (Corsi et al., 2014). Actually, no data on the amount of TiO2 NPs in marine waters are available, but the predicted environmental concentrations in surface waters are in the order of a few µg L−1 (Minetto et al., 2014). Despite the importance of

⁎

the marine environment, most of papers concerning the effects of TiO2 NPs on aquatic biota have been carried out on freshwater rather than on marine organisms (Minetto et al., 2014). Since coastal waters and sediments are the final sink for TiO2 NPs, a great variety of marine organisms, like crustaceans, molluscs and algae, may be exposed to them (Matranga and Corsi, 2012). Among these, microalgae represent the major primary producers in marine ecosystems, and are of great importance for the maintenance of the aquatic ecosystem. In this perspective, marine microalgae, which are widespread in coastal waters and are particularly sensitive to pollutants, can be used as a model for the studies of aquatic risk assessment of nanomaterials (Amiard-Triquet et al., 2015). Several studies have shown that TiO2 NPs can cause adverse effects to aquatic organisms, such as fishes (Paterson et al., 2011), invertebrates (Canesi and Corsi, 2016; Della Torre et al., 2015; Zhu et al., 2011; Galloway et al., 2010), and bacteria (Blaise et al., 2008; Cherchi and Gu, 2010). Only a few recent studies have addressed the interactions of these NPs with marine phytoplankton. A marked

Correspondence to: Institute of Biophysics, Area della Ricerca del CNR, via Moruzzi, 1, 56124 Pisa, Italy. E-mail address: [email protected] (E. Morelli).

http://dx.doi.org/10.1016/j.ecoenv.2017.10.024 Received 20 February 2017; Received in revised form 2 October 2017; Accepted 9 October 2017 0147-6513/ © 2017 Elsevier Inc. All rights reserved.

Ecotoxicology and Environmental Safety 148 (2018) 184–193

E. Morelli et al.

suspensions in the culture media both alone and conditioned with EPS, in order to assess their stability in seawater during the exposure experiment. In addition, we focused on the chemical characterization of EPS and their interaction with NPs, in order to improve our understanding on their role in TiO2 NPs toxicity.

variability of the effects is apparent from the literature, depending on the species studied and the experimental conditions, such as NPs size and type, ionic strength of the culture medium and UV irradiation (Castro-Bugallo et al., 2014; Xia et al., 2015; Sendra et al., 2017a). Indeed, Miller et al. (2012) reported that TiO2 NPs inhibit the growth rates of Thalassiosira pseudonana, Skeletonema marinoi, Dunaliella tertiolecta and Isochrysis galbana only under UV irradiation. Usually, the effect of TiO2 NPs in algae is described by parameters reflecting the physiological state of the cells, such as growth rate, photosynthetic pigment content, antioxidant response, assimilation rate of NPs (Chen et al., 2012; Xia et al., 2015; Li et al., 2015; Manzo et al., 2015; Wang et al., 2016; Schiavo et al., 2016). Direct toxic effects, such as membrane and cellular damages, have been described in Phaeodactylum tricornutum (Wang et al., 2016), in Nitzschia closterium (Xia et al., 2015) and in the freshwater green alga Chlamydomonas reinhardtii (Chen et al., 2012) in response to high levels of TiO2 NPs exposure. Indirect effects of NPs can be caused by the aggregation of NPs on cell surface which could reduce light availability for photosynthesis (shading effect), and/or the uptake of nutrients through the cell surface (Aruoja et al., 2009; Chen et al., 2012; Li et al., 2015). Another well described mechanism of TiO2 NPs toxicity is the production of reactive oxygen species (ROS), because of their photocatalytic properties and interaction with organisms or biomolecules present in the environment (Von Moss and Slaveykova, 2014). Studies aiming to address the toxicity of NPs cannot be separated from the knowledge of the chemical behaviour of these NPs in a complex medium, such as seawater (Handy et al., 2008). In the aquatic environment, NPs generally tend to aggregate as a function of size, surface charge, pH of the medium, as well as of the presence of dissolved organic matter (Klaine et al., 2008; Keller et al., 2010). In particular, in seawater, the high ionic strength can effectively shield the repulsive forces among charged NPs, thereby favouring the aggregation processes. In particular, TiO2 NPs aggregate very quickly in seawater, within a few hours (Keller et al., 2010), with consequent fast sedimentation rates. The aggregation process, common to a large variety of NPs, reduces the exposure of organisms living in the water column and increases the risk for benthic and filter-feeding organisms (Klaine et al., 2008; Morelli et al., 2013; Baker et al., 2014; Zhou et al., 2016). Moreover, recently it has been shown that heteroaggregation of TiO2 NPs with microalgae can play an important role in the toxicity mechanisms of these nanoparticles, especially at high ionic strength (Sendra et al., 2017b). A few studies have shown that TiO2 NPs toxicity is related not only to their physicochemical properties (i.e. surface charge, size, shape, photocatalytic activity), but also to the environmental conditions (i.e. pH, ionic strength, natural organic matter). One of the main components of natural organic matter is constituted EPS excreted by bacteria and phytoplankton. EPS are ubiquitous in the marine environment. They are mainly composed by polysaccharides and proteins, with a variable composition depending on species and environmental conditions (Verdugo et al., 2004). EPS are amphiphilic biopolymers which can assemble to form marine gels, but can also play a key role in the interaction between NPs and marine organisms. Recent papers have shown that EPS can interact with NPs, thus affecting NPs aggregation and /or degradation processes (Quigg et al., 2013). The interaction of EPS derived from phytoplankton with NPs in the marine environment has been described for quantum dots (Zhang et al., 2012), copper-based NPs (Adeleye et al., 2014) and silver engineered NPs (Miao et al., 2009) but, to our knowledge, no study has been performed with TiO2 NPs. Moreover, no information is available about the chemical characterization of biomolecules, such as proteins, which could adsorb on NPs, and consequently affect their biological fate. The aim of the present study was to investigate the effect of TiO2 NPs on the green alga Dunaliella tertiolecta, in terms of growth rate, chlorophyll content, oxidative stress and ROS accumulation. We used Dynamic Light Scattering (DLS) to monitor the behaviour of TiO2 NPs

2. Materials and methods 2.1. Preparation and behavioural characterization of TiO2 NPs Nanosized titanium dioxide (TiO2 NPs) Aeroxide® P25 was supplied by Evonik Degussa (Germany). According to the manufacturer, TiO2 NPs have an average primary size of 25 nm and are composed of anatase: rutile 80%:20% crystalline phase (purity > 99.5%). Brunauer, Emmet and Teller (BET) specific surface area is 50 ± 15 m2 g−1. The pristine nanopowder was employed for dispersions as received, with no surface modifications. Stock dispersions were prepared suspending TiO2 NPs powder in ultrapure Milli-Q water (Millipore, Bedford, MA, USA) up to a concentration of 5000 mg L−1 by tip sonication (45 min at 100 w, 50% cycle off) in an ice-cool bath. Initially, NPs dispersed in Milli-Q water were imaged through transmission electron microscopy (TEM) by deposition of a 10 µL drop on formvar/carbon-coated copper grids and overnight drying, and micrographs were acquired at 200 keV. NPs stocks were stored in dark at room temperature up to 24 h and diluted in f/2 algal medium (salinity 38‰, pH 8), in order to achieve 10 mg L−1 working dispersions, which were bath-sonicated for 15 min prior to behavioural analysis. In order to assess the role of green algae exopolymeric substances (EPS) on TiO2 NPs behaviour, such working dispersions were prepared in f/2 medium conditioned by culturing D. tertiolecta for 96 h (final cell density, 1·106 cells mL−1), which was then removed by centrifugation and 1.2 µm filtration, prior to test TiO2 NPs dispersion. The concentrations of carbohydrates and proteins in this algae-conditioned culture medium were 0.93 and 0.45 mg L−1, respectively. The aggregation of TiO2 NPs in standard f/2 and algaeconditioned f/2 medium was assessed via Dynamic Light Scattering (DLS) on a Zetasiser Nano ZS90 (Malvern Instruments Ltd., Malvern, UK), operating with 90° backscattering angle and a 639 nm wavelength laser. Hydrodynamic diameter (Z-average) growth and Polydispersity Index (PdI) were monitored every 10 min for 2 h. Each measurement consisted of three independent runs, all made of 11 sub-runs. The dispersion's electrostatic stability in both f/2 algal media was assessed measuring the NPs’ surface charge (ζ-potential) and reported as an average of three measurements. In addition, the dynamic sedimentation process was followed spectrophotometrically over 72 h from dispersion, using a well-established procedure (Keller et al., 2010; Della Torre et al., 2015), on a Perkin-Elmer Lambda 650 UV–Vis spectrophotometer. First, a calibration curve was obtained by measuring absorbance at 269 nm of a range of TiO2 dispersions of known concentration. The lowering of suspended TiO2 concentration due to settling was estimated through absorbance measurements at 0, 1, 2, 4, 6, 24, 48 and 72 h after dispersion. For each time point, absorbance was recorded in triplicate, from volumes withdrawn in the upper layer of dispersions (< 1 cm from the surface), which were left stand unmoved in 50 mL Falcon tubes, at RT, for the whole experimental duration in order to avoid perturbation and resuspension. The sedimentation profile was then achieved plotting normalized concentration values (i.e. C/ C0, where C0 is the initial concentration at 0 h and C refers to specific time points) vs. time, expressed as mean ± SD. 2.2. Exposure experiments The unicellular green alga Dunaliella tertiolecta was obtained from the Culture Collection of Algae and Protozoa, Dunstaffnage Marine Laboratory, U.K.. Stock cultures were maintained in axenic conditions, in a growth chamber at 21 ± 1 °C and fluorescent daylight (100 µmol photons m−2 s−1) in a 16:8 light-dark cycle photoperiod. Culture 185

Ecotoxicology and Environmental Safety 148 (2018) 184–193

E. Morelli et al.

520 nm emission wavelength (λexcitation = 485 nm), using a Fluoromax-4 Spectrofluorometer (Horiba JobinYvon, Milano, Italy). Emission intensity was used to estimate the level of total ROS in the culture. Thereafter, the sample was centrifuged (10000 g, 20 °C, 15 min), the pellet containing algae was re-suspended in a fresh culture medium (4 mL) and the emission was measured again in order to obtain the intracellular ROS estimation. Extracellular ROS level was calculated by subtracting intracellular ROS from the total ROS amount. Fluorescence values were always corrected for the fluorescence of the respective culture medium, without cells. The ROS level in the cultures was expressed as the ratio of emission intensity at 520 nm to cell density (cell number mL−1) in the cuvette.

medium was natural seawater enriched with f/2 medium, modified as to obtain a f/10 medium for trace metal concentrations. Seawater (salinity = 38‰) was collected offshore in the Tyrrhenian sea (Italy), filtered through 0.2 µm membrane filter (Sartorius) and stored in the dark at +4 °C. Stock cultures were maintained in exponential growth by inoculating cells weekly into a new sterilized medium. In order to perform exposure experiments, algae from a stock culture were inoculated in a new medium at a cell density of 1·105 cells mL−1. This culture was immediately mixed with an equal volume of culture medium containing an appropriate amount of nanoparticles (prepared as described above), so as to obtain cultures at an initial cell density of 5·104 cells mL−1 and TiO2 NPs concentrations ranging from 0.01 to 10 mg L−1. Control cultures were prepared by adding an equivalent volume of f/2 in order to obtain the same cellular density as TiO2 NPs cultures. Algal growth assays were carried out in 2.5 mL multi-well plates, using the experimental conditions described for the stock cultures. Cultures were shaken twice per day with a pipette tip. Cell density was measured at the third day of growth by recording the absorbance at 680 nm (Jasco V-550 UV/Vis spectrophotometer, Lecco, Italy). A standard equation was established between optical density and cell number per mL using a haemocytometer, under a light microscope (Zeiss, Oberkochen, Germany). Finally, the specific growth rate (µ) was calculated by using the equation µ = ln(N3-N0) / (t3-t0), where N3 and N0 are the cell number mL−1 at the third day (t3) and initial day (t0) of the exponential growth phase. Exposure experiments, designed to follow the kinetics of ROS generation, antioxidant enzymatic activity and pigment concentration, were carried out in 900 mL D. tertiolecta cultures. For this purpose, a pre-culture of exponentially growing algae at the 4th day of growth, at approx. 0.7 − 1·106 cells mL−1, was mixed with a suspension of TiO2 NPs as to obtain an initial cell density of 5·105 cells mL−1 and a TiO2 concentration of 10 mg L−1. Cultures were continuously stirred with a magnetic stirrer bar during this time and grown for 72 h. At time intervals (1, 3, 5, 7, 24, 48, 72 h), 10 and 100 mL aliquots were collected for pigment and enzymatic activity measurements, respectively and centrifuged (10000 g, 20 °C, 15 min). The pellets were stored at −70 °C. At 1, 6, 24, 48 and 72 h, aliquots of 4 mL were collected and immediately subjected to ROS measurement. Cell density was always determined and all parameters were normalized per cell. Control cultures (no TiO2 NPs added) were always carried out. In order to evaluate the shading effects of nanoparticle suspensions, a vessel containing the same amount of TiO2 NPs than that of the test cultures was superimposed on a set of control cultures. All experiments were carried out in triplicate.

2.5. Enzyme assays The harvested algae were placed in 1.5 mL of extraction buffer containing 50 mM sodium phosphate buffer (pH 7.0) and 1 mM EDTA (ethylenediaminetetraacetic acid, disodium salt) and immediately disrupted by sonication (Sonopuls Ultrasonic Homogenizer, Bandelin) for 3 min with a repeating duty cycle of 0.3 s, in an ice bath. The cell homogenate was centrifuged at 20000 g for 30 min at 4 °C and the supernatant was used to measure enzymatic activities. The activity of superoxide dismutase (SOD) was assayed by measuring its ability to inhibit the photoreduction of nitro blue tetrazolium (NBT), according to the method of Beyer and Fridovich (1987). In this assay one unit (U) of SOD is defined as the amount required to inhibit the photoreduction of NBT by 50%. The activity of catalase (CAT) was measured according to Aebi (1974), by monitoring the decomposition of H2O2, through the decrease in absorbance at 240 nm (ε = 0.04 mM−1 cm−1). One unit of activity (U) was defined as the amount of enzyme that can transform 1 μmole of H2O2 per minute. All enzymatic activities were calculated per mg of protein. The protein content in the cell extract was determined according to Bradford (1976) using bovine serum albumin as standard. Enzymatic assays were carried out at 25 °C using the JASCO V-550 UV/Vis Spectrophotometer. 2.6. EPS isolation and characterization In order to characterize the chemical composition of EPS, a culture of D. tertiolecta exposed to 10 mg L−1 TiO2 NPs for 24 h and a control culture (cell density 1.1 ± 0.4·106 cells mL−1) were used. The cultures were centrifuged (10000 g, 20 °C, 15 min) and the supernatant was filtered through membrane filters (Sartorius 1.2 µm) to remove cells. Six successive aliquots of 12 mL of the filtrate (Vol total = 72 mL) were centrifuged (5000 g, 60 min, 20 °C) by using a centrifugal filter device with a 3 kDa cut-off (Amicon Ultra-15 mL, Millipore, USA), and reduced to a final volume of about 300–400 µL. Thus, in the retentate, the organic components were concentrated about 200 times. The retentate was analysed for the protein content by the Pierce micro BCA protein assay kit (Pierce, USA) and for total carbohydrate content using the anthrone method with glucose as a standard (Brooks et al., 1986). A further experiment was designed in order to isolate nanoparticlesEPS complexes. A 300 mL culture of D. tertiolecta at the end of the logarithmic growth phase (cell density about 1 × 106 cells mL−1) was centrifuged and the supernatant filtered (Sartorius 1.2 µm) to remove cells. The filtrate was spiked with a suspension of TiO2 NPs pre-sonicated in the culture medium, in order to obtain a final concentration of 10 mg L−1. After 24 h of incubation, nanoparticles-EPS complexes were recovered by centrifugal isolation according to the procedure of Monopoli et al. (2013), with some modifications. Briefly, the suspension was centrifuged (14000 g, 30 min, 20 °C) and the supernatant (SN) was concentrated by using a centrifugal filter device with a 3 kDa cutoff, as described above. The pellet was washed with 1.5 mL of PBS buffer pH 7, vortexed and centrifuged again (17000 g, 30 min, 20 °C) to recover the nanoparticles-EPS complexes. This washing procedure was repeated three times and the supernatants (W1, W2 and W3) were also

2.3. Pigment quantification Algae stored at −70 °C were thawed in 1 mL of 95% ethanol, vortexed vigorously and kept overnight at room temperature. The extracts were then centrifuged (8000 g, 4 °C, 15 min) to remove cellular debris and the supernatants used for spectrophotometric measurements (Jasco V-550 UV/Vis spectrophotometer, Lecco, Italy). The concentrations of chlorophyll a, b and carotenoids were determined measuring the optical density at 664, 648.5 and 470 nm, according to the protocol of Lichtenthaler and Buschmann (2001). 2.4. Determination of ROS The production of ROS was measured by following the cellular conversion of the non-fluorescent 2′,7′- dichlorofluoresceindiacetate (DCF-DA) to the highly fluorescent compound dichlorofluorescein (DCF) as described by Wang and Joseph (1999a, 1999b). The samples (4 mL) were spiked with 40 µL of 1 mM DCF-DA, as to obtain 10 µM concentration in the culture, and left to react for 1 h at room temperature, with stirring. The formation of DCF was determined at 186

Ecotoxicology and Environmental Safety 148 (2018) 184–193

E. Morelli et al.

collected, in order to test the success of the protocol. The final pellet, containing TiO2 NPs with strongly bound EPS (named HC, hard corona), was re-suspended with 200 µL of PBS. Samples SN, W1, W2, W3 and HC were analysed for protein content and for total carbohydrate content as described above. 2.7. SDS-Polyacrylamide gel electrophoresis Protein content of samples obtained from the separation procedure of nanoparticle-EPS complexes was analysed by SDS-PAGE. Electrophoresis was performed on a 4% stacking gel and a 12% resolving gel in a mini-electrophoresis apparatus (BioRad, USA), according to the Laemmli method (1970). Protein bands were stained with silver stain (Merril et al., 1981). Molecular mass was estimated by comparison with SDS-Broad Range standard proteins run in the same gel. 2.8. Data analysis One-Way ANOVA test and post hoc analysis of variance (Tuckey test) were conducted using Origin 9.0 Software (Original Corporation, MA, USA). Significant differences were accepted for p < 0.05. Mean values are expressed with standard deviation. 3. Results and discussion 3.1. Behaviour of TiO2 NPs in culture media The typical irregular/spheroidal shape of TiO2 NPs was confirmed by TEM analysis (Fig. SM1) and primary particle size was found to be 24 ± 7 nm, as reported in our previous work for the same NPs batch (Della Torre et al., 2015), in keeping with the data declared by the manufacturer. The dispersions proved highly unstable with an average ζ-potential value of – 5.65 ± 0.44 and – 8.59 ± 0.53 mV, in the f/2 and algae-conditioned f/2 media, respectively, as NPs are generally considered charge-stabilized with values > +30 or < −30 mV. Mediuminduced electric double layer compression due to surface charge shielding is well documented for high ionic strength electrolytes solutions (Keller et al., 2010; Brunelli et al., 2013). In our case, the standard marine algal culture medium f/2 has a high salts content (salinity = 38%) and an abundance of monovalent and divalent counter ions (Ca2+, Mg 2+) that can partly saturate the NPs surface charge, which at the present value of pH (i.e. 8) was negative (TiO2 NPs isoelectric point ~ 6.2 – 6.5) (Keller et al., 2010; Kosmulski, 2009; Shih et al., 2012). Therefore, NPs dispersed in f/2 media colliding due to casual movements had a high attachment efficiency and were highly prone to form aggregates. This trend was indeed confirmed by DLS analysis (Fig. 1A). The hydrodynamic diameter in f/2 showed a progressive growth, from 731.4 ± 122 nm, immediately after dispersion and sonication, up to ~ 1.65 µm at 2 h. TiO2 NPs showed an aggregative behaviour with the same fashion in algae-conditioned f/2 (Fig. 1A), ranging from 816.3 ± 32.58 nm to 1.62 µm during the examined time lapse. In both cases, the PdI values increased with time, from 0.23 in f/2 and 0.27 in the algae-conditioned f/2 medium, to > 0.5 after 2 h, not allowing further analysis after that time-point due to unreliable instrumental measurements. Overall, TiO2 NPs dispersion in the present media was highly prone to form aggregates that reached micrometric sizes within 30 min, also showing a high degree of heterogeneity in size distribution. Under these conditions, NPs can easily form big clusters able to progressively settle in test vessels under the influence of gravity, lowering the amount actually suspended in the sample. Indeed, it is of primary importance to determine, during exposure experiments, the suspended fraction of NPs actually capable to come in contact with and exert an effect on the exposed organisms. Therefore in the present study, TiO2 NPs sedimentation was determined indirectly via UV–Vis spectrophotometry, which has already been proven a suitable technique for assessing TiO2

Fig. 1. (A) Kinetics of aggregation of TiO2 NPs (10 mg L−1) monitored for 2 h by DLS. (B) Sedimentation profile of 10 mg L−1 TiO2 NPs. C and Co represent the concentration of TiO2 NPs at the specific time point and at t=0, respectively, measured by absorbance at 269 nm. Marine algal medium f/2 alone (•) or conditioned with EPS (■). Data are mean values ± standard deviation of three replicates. Pairwise comparison of normalized TiO2 NP concentrations in f/2 and f/2+EPS was conducted by one-way ANOVA followed by Tukey's post hoc test, and significant differences detected at p < 0.05 were indicated by asterisks.

NPs concentration (Ramirez-Garcia et al., 2011). The calibration line of the instrument showed good linearity (R2= 0.9994) at λ = 269 nm, in the concentration range employed (Fig. SM2). The sedimentation profiles were performed both in f/2 and in algae-conditioned f/2 medium (Fig. 1B), in triplicate. The former showed a fairly constant sedimentation rate up to 6 h, with approximately 58% of suspended TiO2 settling. A slower rate of sedimentation took place up to 24 h, at the end of which little (~ 8%) of the initial TiO2 concentration remained suspended, which remained nearly unchanged at 48 h and 72 h. On the contrary, TiO2 NPs showed a significantly milder (p < 0.05) sedimentation regime when dispersed in algae-conditioned f/2. In this case, even if the sedimentation trend was similar, the suspended NPs concentration remained well above the former values, in the time range from 1 to 24 h. Nonetheless, during the first hour after dispersion, TiO2 sedimentation was still dramatic, due to very large NPs clusters settling rapidly. After that time point, however, it was possible to observe a longer residence time of NPs in the water column. This allowed us to hypothesize a modest EPS-driven stabilization effect of smaller TiO2 aggregates. Soluble biomolecules, like those abundantly present in biological fluids such as proteins, are capable of adsorbing onto suspended NPs surface, providing in some cases single particle steric stabilization (Monopoli et al., 2013). In our case, the EPS molecules may have caused a mild stabilization of small TiO2 aggregates, possibly slowing, to a certain extent, the formation of bigger, rapidly-settling clusters. This would explain the slackening of settling speed of TiO2 suspensions in algae-conditioned f/2 compared to f/2 alone. NPs dispersion state and behaviour in standard media can consistently affect 187

Ecotoxicology and Environmental Safety 148 (2018) 184–193

E. Morelli et al.

after 3 days of exposure, although short-term negative effects cannot be excluded. Our findings are in strong agreement with Miller et al. (2010) who reported no effect of TiO2 NPs (from 0.01 to 1 mg L−1) on the growth of four species of marine phytoplankton. Wang et al. (2016) also reported a low toxicity of TiO2 NPs to P. tricornutum which exhibited a significant growth inhibition at nanoparticle concentrations higher than 20 mg L−1 and a 120-h EC50 of 167 mg L−1. Growth inhibition was reported in the marine diatom N. closterium exposed to TiO2 NPs, but only at very high concentrations, such as a 96-h EC50 of 89–119 mg L−1 (Xia et al., 2015). Furthermore our data are roughly in agreement with Manzo et al. (2015) who reported a NOEC of 7.5 mg L−1 and EC50 of 25 mg L−1 for D. tertiolecta exposed to similar TiO2 NPs. On theother hand, Li et al. (2015) reported 72-h EC50 values of 10.7 and 7.4 mg L−1 of TiO2 NPs for Karenia brevis and Skeletonema costatum, respectively. These contrasting results suggest that TiO2 NPs toxicity can depend not only on the algal species tested, but also on the type of NPs tested (different size or proportion of anatase/rutile) or the different experimental conditions which include illumination, salinity of culture medium, natural vs artificial seawater, initial cell density. Moreover, toxic effects of TiO2 NPs were found in freshwater algae at lower concentrations. In fact, measurable growth inhibition was found for Chlamydomonas reinhardtii at just 0.1 mg L−1 TiO2 NPs (Chen et al., 2012) and an EC50 of only 5.8 mg L−1for Pseudokirchneriella subcapitata (Aruoja et al., 2009). These results can be explained with a lower aggregation process occurring in fresh water compared to sea water which allow the presence of small aggregates or even dispersed NPs (Handy et al., 2008). Keller et al. (2010) report that the size of aggregates of TiO2 NPs is about 300 nm in freshwater and 1000–2000 nm in seawater. Overall, our results evidenced that TiO2 NPs had no significant effect in the growth of D. tertiolecta cultures, at least in the range of concentration from 0.01 to 10 mg L−1. Biochemical markers have been analysed to assess if sub-lethal effects, such as the activation of defence mechanisms, appear in response to exposure to these levels of NPs.

Fig. 2. Growth rate of cultures of D. tertiolecta exposed for 72 h to increasing concentrations of TiO2 NPs. Growth rate of cultures with shading effect at 1 and 10 mg L−1 NP concentrations are also shown. Error bars represent standard deviation (n = 6). No significant difference was detected with respect to the control cultures, at p < 0.05 level according to the one-way ANOVA and Tukey's comparison.

exposure levels and potential toxic effects towards organisms selected for ecotoxicological assays (Brunelli et al., 2016). Therefore such parameters have to be thoroughly characterized throughout the duration of the experiment. In our case, it is highly unlikely that D. tertiolecta cells were exposed to the actual nanosized form of TiO2, possibly excluding a toxic effect relative to a single NP-cell interaction pattern. Moreover, the instability of NPs in f/2 media allowed us to speculate that the level of exposure decreased with time as TiO2 was removed from the exposure media by settling. The highest exposure to NPs took place within the first 6 h, while from 24 h on, the low NPs concentration in the test vessel may have affected bioavailability to algae and cellular toxic responses. In agreement, Li et al. (2017) suggested that the toxic effect of a metal-oxide particle (ZnO) in two marine microalgae was mainly related to the aggregation and sedimentation processes.

3.3. Chlorophyll Chlorophyll content in algae can be used as a measurement of physiological health status (Metzler et al., 2012). In order to evaluate whether TiO2 NPs affected photosynthetic pigment production, we monitored the cellular levels of chlorophyll a, b and carotenoids in D. tertiolecta during a 72 h exposure experiment at 10 mg L−1 TiO2 NPs. The time-course of pigment concentration (Fig. 3) showed time-dependent fluctuations for the three treatments (control, shading and TiO2 NPs), with values slightly higher during the first 24 h for TiO2exposed cultures, although these values were not significantly different from the control or the shading cultures. These findings suggest that TiO2 NPs exposure did not affect the ability of the alga to synthesize photosynthetic pigments either for the direct interaction with NPs, or for the indirect shading effect. The ratios chlorophyll a / chlorophyll b and (chlorophyll a + chlorophyll b) / carotenoids were not affected either by TiO2 NPs or shading (data not shown), and exhibited mean values of 2.2 ± 0.2 and 3.4 ± 0.3, respectively. The finding is in agreement with the absence of effect on growth and hence with the lack of toxicity, at least under our experimental conditions. Indeed, other authors reported that growth inhibition in algae, such as D. salina, K. brevis, S. costatum, P. subcapitata exposed to other NPs goes along with a decrease of the chlorophyll content (Shirazi et al., 2015; Li et al., 2015; Metzler et al., 2012).

3.2. Effect of TiO2 nanoparticles on D. tertiolecta growth TiO2 NPs had no effect on the growth of D. tertiolecta cultures during 3-day exposure experiments in a range of TiO2 concentration 0.01–10 mg L−1 (ANOVA, P > 0.05), as shown in Fig. 2. Due to the documented strong aggregation of TiO2 NPs and the consequent reduction of light transmission through the culture medium, a decrease in the photosynthetic ability of algae exposed to TiO2 NPs could be expected. In this work we evaluated whether the effect of shading was implied in the possible mechanisms of TiO2 NPs toxicity, for each biological endpoint. We verified that shading at two selected concentrations (1 and 10 mg L−1) did not inhibit algal growth (Fig. 2), thus confirming that TiO2 NPs did not cause any inhibition to algal growth rate, either directly or indirectly by a shading effect. These results are in agreement with the behavioural data showing strong aggregation and then settling processes occurring in culture medium soon after addition of TiO2 NPs (Fig. 1A and B). Indeed, algal cells were scarcely in contact with particles of nanometer size, but mainly with aggregates of NPs originating within minutes from TiO2 NPs addition to the culture medium. Large aggregates entrapping algal cells were observed in cultures exposed 5 h to TiO2 NPs, as shown in Fig. SM3. Other authors (Manzo et al., 2015) report that the toxic effects of TiO2 NPs on D. tertiolecta are due to large aggregates, mainly occurring during the first hours of exposure. Moreover, the observed slight reduction in growth rate is recovered at the end of the exposure test. Our results showed that the contact of large aggregates did not affect growth rate of microalgae

3.4. Oxidative stress Due to the recognized photocatalytic activity of TiO2 NPs, an enhanced ROS production was expected. Therefore we examined the activity of the main antioxidant enzymes (SOD and CAT) in concomitance with ROS generation in cultures of D. tertiolecta exposed to 10 mg L−1 188

Ecotoxicology and Environmental Safety 148 (2018) 184–193

E. Morelli et al.

Fig. 4. Time course of the activity of SOD (A) and CAT (B) in cultures of D. tertiolecta exposed to 10 mg L−1 TiO2 NPs and in cultures with shading effect, during 72 h exposure. Enzymatic activity is expressed as a percentage of that of the control culture. Data are mean values ± standard deviation of three independent experiments. Asterisks mean significant differences with the control cultures detected at p < 0.05 level according to the one-way ANOVA and Tukey's comparison.

inhibition at longer exposure times. On the contrary, CAT activity increased significantly just at 1–3 h of exposure, in TiO2-exposed cultures with respect to the control ones, yielding an increase close to 100%, without any concomitant effect of shading. However, starting from the fifth hour, it decreased to values similar to the control or slightly lower (not significant), at 48 and 72 h of exposure. Our results indicate an early oxidative stress response within the first 6 h of exposure, as evidenced by the rapid increase of CAT at the beginning of the exposure. This biological effect is in agreement with the behaviour of TiO2 NPs in the culture medium (Fig. 1B), showing the highest propensity of NPs to remain in suspension, and likely to interact with algae, during the first hours. The finding that CAT, but not SOD, activity increased suggests an increased production of hydrogen peroxide, rather than of the superoxide anion. This result is in agreement with other authors (Li et al., 2015), who report a dose-dependent increase of H2O2 but not of O2·- in K. brevis and S. costatum exposed to TiO2 NPs. The same authors show a different trend of SOD and CAT activity in the two species of marine microalgae, suggesting a species-specific response. In agreement with our data, Xia et al. (2015) report values of SOD activity significantly lower than the control in cultures of N. closterium exposed to 5 mgL−1 TiO2 NPs for 6, 12, 24 and 48 h, as well as an early and pronounced increase of CAT activity, followed by a decrease at longer exposure times. The time course of total, intracellular and extracellular production

Fig. 3. Time course of chlorophyll a (A), chlorophyll b (B) and carotenoids (C) in cultures of D. tertiolecta exposed to 10 mg L−1 TiO2 NPs, control cultures and in cultures with shading effect, during 72 h exposure. Data are mean values ± standard deviation of three independent experiments. No significant difference was detected with respect to the control cultures, at p < 0.05 level according to the one-way ANOVA and Tukey's comparison.

TiO2 NPs. Fig. 4 shows the time course of SOD and CAT activities measured in the cultures directly exposed to TiO2 NPs and in the shading cultures, both reported as a percentage of the control culture. Original enzyme activity data are shown in Fig. SM4. During the first 5 h of exposure, the SOD activity in TiO2-exposed cells was not significantly affected compared to control culture. Thereafter, it followed a decreasing trend, exhibiting values significantly lower than the control at 7, 48 and 72 h of exposure. The shading culture did not show any alteration in SOD activity compared to control culture, indicating that shading was not the cause of SOD 189

Ecotoxicology and Environmental Safety 148 (2018) 184–193

E. Morelli et al.

trend is in agreement with the behaviour of TiO2 NPs in the culture medium (Fig. 1A and B). As shown above, TiO2 NPs rapidly aggregated in the culture medium, and settled to the bottom of the test vessel, thus it is expected that the main effects may occur during the first 24 h, and not later, at 48 and 72 h exposure. It should be underlined that these measurements of ROS have been corrected for the contribution of ROS in the culture medium (without cells) both in the presence and in the absence of TiO2NPs, thus the ROS levels account for the presence of the cells, including their extracellular products, such as EPS. Indeed, chemical redox reactions with biomolecules in the surrounding medium have been proposed as a mechanism of ROS generation (Von Moss and Slaveykova, 2014). Our previous finding that the presence of EPS in the culture medium possibly contributes to stabilize the NP suspension (Fig. 1B) supports the hypothesis that the organic components of cellular exudates can come into contact with NPs. It can be supposed that EPS can play an important role in the effect of NPs, since they can interact with TiO2 NPs modulating possible toxic effects towards cells. In the following experiments, we characterized the chemical composition of EPS and their interaction with TiO2 NPs. 3.5. Exopolymeric Substances (EPS) The chemical composition of EPS extracted from D. tertiolecta cultures has been evaluated by measuring the protein and carbohydrate concentrations. The data related to a culture exposed 24 h to TiO2 NPs (10 mg L−1) and a control culture are shown in Table 1. The results show that both the carbohydrate and the protein concentrations were unchanged in response to TiO2 NPs, suggesting that NPs didn’t affect either the amount or the composition of EPS in terms of total carbohydrates and protein content, at least under our experimental conditions. The carbohydrates are the major component of EPS, since the protein/carbohydrate ratio is about 0.3. This result is in agreement with other authors who reported similar ratios for EPS extracted from the marine microalgae I. galbana (Adeleye et al., 2014) and P. tricornutum (Chen et al., 2011). The question if some algal species can modulate the production of EPS in response to NP exposure, as a general defence mechanism against toxic compounds, still remains open (Quigg et al., 2013). Our data do not seem to support such a hypothesis in D. tertiolecta in terms of changes in chemical composition of EPS upon TiO2 NPs exposure, while on the contrary an involvement of EPS in the stability and consequently, bioavailability of NPs can be hypothesised. Indeed, we found that the presence of EPS in the culture medium mitigated the process of settling of TiO2 NPs, as reported in Fig. 1B, suggesting that the organic substances might interact, for instance adhering to the NPs, and thus contribute to stabilize the suspension. In order to test this hypothesis, we carried out an additional experiment, in which TiO2 NPs suspensions were incubated 24 h with the algal medium containing EPS. Then, the EPS-NPs complexes were subjected to a separation procedure, aimed to isolate the organic molecules bound to the NPs (Monopoli et al., 2013). Table 1 shows the concentrations of

Fig. 5. Time course of ROS levels in cultures of D. tertiolecta exposed to 10 mg L−1 TiO2 NPs and in control cultures, during 72 h exposure. Total (A), intracellular (B) and extracellular (C) ROS levels are expressed as DCF-fluorescence intensity at 520 nm (F520) normalized to cell density (N, cell number mL−1). Data are mean values ± standard deviation of three independent experiments. Asterisks mean significant differences with the control cultures detected at p < 0.05 level according to the one-way ANOVA and Tukey's comparison.

of ROS is shown in Fig. 5A–C. The histograms show a transient increase in total ROS concentration starting from 1 h of exposure, reaching a maximum value after 24 h and a subsequent decrease. Intracellular ROS was a small fraction of total ROS (range 5–20%), showing that the higher amount of ROS (80–95%) was extracellular, independently from the presence of NPs. The time course of intracellular ROS exhibited a four-fold increase in TiO2-treated algae with respect to the control culture after just one h of exposure. Thereafter, intracellular ROS in TiO2-treated algae slowly decreased, maintaining a value 3 times higher than that of the control culture at 24 h, and reached values similar to the control culture at 48 and 72 h. Extracellular ROS increased sharply during the first 24 h in cultures exposed to TiO2 NPs, reaching values 6–7 times higher than that of the control culture after 6 and 24 h of exposure. However, at 48 and 72 h of exposure, extracellular ROS dramatically decreased reaching values slightly, but significantly, higher than those of the control culture. This

Table 1 Chemical analysis of EPS extracted from cultures of D. tertiolecta either not-exposed (Control) or exposed for 24 h to 10 mg L−1 TiO2 NPs (TiO2-exposed), and of the fractions SN, W1, W2, W3 and HC obtained from the centrifugal separation procedure of the EPSnanoparticles complexes. Mean values ± standard deviation (n=3) are reported. SN, supernatant; W1, W2 and W3, three successive washings of the pellet; HC, hard corona (EPS strongly bound to TiO2 NPs).

190

Sample

Carbohydrate (mg L−1)

Protein (mg L−1)

Protein/carb Ratio

Control TiO2-exposed SN W1 W2 W3 HC

1.13 ± 0.40 1.06 ± 0.37 0.740 ± 0.142 0.010 ± 0.002 0.010 ± 0.005 0.001 ± 0.002 0.078 ± 0.008

0.34 ± 0.16 0.32 ± 0.21 0.261 ± 0.030 0.014 ± 0.008 0.012 ± 0.003 0.008 ± 0.008 0.110 ± 0.010

0.30 0.30 0.35 1.47 1.20 n. d. 1.41

Ecotoxicology and Environmental Safety 148 (2018) 184–193

E. Morelli et al.

Fig. 6. A) Typical SDS-PAGE of silver-stained proteins present in the supernatant after the first centrifugation of the EPS-TiO2 NPs suspension (SN, supernatant) and in the pellet, after 3 repeated washing steps (HC, hard corona). B) Densitometric analysis of the two lanes: SN and HC. Culture medium containing EPS was incubated with 10 mg L−1 TiO2 NPs for 24 h.

TiO2 NPs in the marine environment. Our results indicate that TiO2 NPs aggregate quickly in seawater and tend to settle to the bottom of the test vessel, but the presence of biogenic organic matter released by microalgae seemed to attenuate this process, favouring the suspension stability in the first hours of exposure. We found no evidence that TiO2 NPs inhibited the growth of D. tertiolecta or affected the pigment cellular content up to a concentration of 10 mg L−1 during a 72 h exposure time. This low toxicity of TiO2 NPs towards the marine green alga D. tertiolecta went along with the fast sedimentation and removal of large aggregates. Furthermore, intracellular ROS increased at the beginning of the exposure to TiO2 NPs, concomitant with a transient enhancement of CAT activity, which scavenged potentially toxic oxygen radicals. The subsequent decrease of intracellular ROS coincided with the restoration of CAT activity. The high production of extracellular ROS induced by TiO2 NPs may be related to the interaction EPS-NPs. A first attempt to concentrate and characterize EPS- NPs complexes highlighted the propensity of NPs to bind proteins rather than carbohydrates, suggesting their important role in the stability and biological fate of NPs in the marine environment. Overall, our results indicated that exposure time and test conditions can have a significant role in toxicity response of marine algae to TiO2 NPs. As recently underlined, there is an urgent need to properly set experimental conditions in running bioassays for nanomaterials risk assessment, due to peculiar properties of the materials and their interactions with exposure media (Petersen et al., 2015). Indeed, the use of standardized test conditions should be further adapted to better mimic realistic environmental exposure scenarios in which marine species will come in contact with nanomaterials in their natural environment. Further studies on the interaction between TiO2 NPs and marine organisms at different degree of complexity will be necessary to better characterize the ecotoxicity of NPs. These results will help the understanding of the mechanisms of NP toxicity in the marine environment and the design of safe nanomaterials.

carbohydrates and proteins in the fractions obtained from the separation procedure. In the supernatant obtained from the first centrifugation (SN) we found the highest amount of carbohydrate (about 88%) and proteins (about 65%), compared with the other fractions. Furthermore, the protein/carbohydrate ratio was of 0.35, confirming the results obtained for the whole cellular exudates (Table 1). These results indicate that the highest amount of EPS remained in suspension unattached or attached to suspended NPs, if any. Very low amounts of carbohydrates and proteins were detected in the washes 1–3 (W1, W2, W3), with a decreasing trend from W1 to W3, confirming the suitability of the separation protocol. In the final pellet, containing the EPS strongly bound to NPs, we found about 9% of total carbohydrates and 27% of proteins, with a protein/carbohydrate ratio of 1.4. This ratio was higher than the value 0.35 calculated for the SN sample. A comparable increase in the protein/carbohydrate ratio was also observed in W1 and W2 samples. This finding indicates that proteins were preferentially adsorbed to NPs with respect to carbohydrates. The protein content of SN and HC samples was characterized by SDS-Page, and a representative gel is reported in Fig. 6A. A comparison between SN and HC lanes shows that some protein bands evident in HC were undetectable in the SN sample. In fact, qualitative analysis of the gel densitometric profile (Fig. 6 B) of HC sample showed an increase in the content of proteins in the molecular weight range from about 80–20 kDa with respect to SN, suggesting that a selective attachment of some proteins to NP occurred. These proteins strongly bound to TiO2 NPs constitute a protein corona and might be involved in NP bioavailability and defence mechanisms as previously described (Monopoli et al., 2013). Studies regarding the formation of a NP protein corona have been reported in biological fluids of the marine bivalve Mytilus galloprovincialis (Canesi et al., 2016), of the terrestrial earthworm Eisenia fetida (Hayashi et al., 2013) and in human plasma (Monopoli et al., 2011), and some proteins have been identified. To our knowledge, this is the first report regarding the electrophoretic characterization of proteins involved in the formation of a NP protein corona in marine phytoplanktonic exudates as EPS. Additional research will be devoted to identify these specific proteins which could play an important role in stability of suspension of TiO2 NPs and hence driving their biological fate and toxicity in marine environment.

Acknowledgments This research was supported by the National Research Council and the University of Siena. The authors wish to thank Dr Sabina Lucia for the images by light microscopy and Alberto Pietrangeli for his technical assistance in microalgae culturing.

4. Conclusions This study contributes to the understanding of the ecotoxicity of 191

Ecotoxicology and Environmental Safety 148 (2018) 184–193

E. Morelli et al.

algae and the site of reactive oxygen species production. Aquat. Toxicol. 158, 1–13. Li, J., Schiavo, S., Rametta, G., Miglietta, M.L., La Ferrara, V., Wu, C., Manzo, S., 2017. Comparative toxicity of nano ZnO and bulk ZnO towards marine algae Tetraselmis suecica and Phaeodactylum tricornutum. Environ. Sci. Pollut. Res. 24, 6543–6553. Lichtenthaler, H.K., Buschmann, C., 2001. Chlorophylls and carotenoids: measurementand characterization by UV–VIS spectroscopy. Curr. Protoc. Food Anal. Chem (F4.3.1-F4.3.8). Manzo, S., Buono, S., Rametta, G., Miglietta, M., Schiavo, S., Di Francia, G., 2015. The diverse toxic effect of SiO2 and TiO2 nanoparticles toward the marine microalgae Dunaliella tertiolecta. Environ. Sci. Pollut. Res. 22, 15941–15951. Matranga, V., Corsi, I., 2012. Toxic effects of engineered nanoparticles in the marine environment: model organisms and molecular approaches. Mar. Environ. Res. 76, 32–40. Merril, D.M., Goldman, D., Sedman, S.A., Ebert, M.H., 1981. Ultrasensitive stain for proteins in polyacrylamide gels shows regional variation in cerebrospinal fluid proteins. Science 211 (4489), 1437–1438. Metzler, D.M., Erdem, A., Tseng, Y.H., Huang, C.P., 2012. Responses of algal cells to engineered nanoparticles measured as algal cell population, chlorophyll a, and lipid peroxidation: effect of particle size and type. J. Nanotechnol. 12 ID 237284. Miao, A.J., Schwehr, K.A., Xu, C., Zhang, S.J., Luo, Z., Quigg, A., Santschi, P.H., 2009. The algal toxicity of silver engineered nanoparticles and detoxification by exopolymeric substances. Environ. Pollut. 157 (11), 3034–3041. Miller, R.J., Lenihan, H.S., Muller, E.B., Tseng, N., Hanna, S.K., Keller, A.A., 2010. Impacts of metal oxide nanoparticles on marine phytoplankton. Environ. Sci. Technol. 44, 7329–7334. Miller, R.J., Bennett, S., Keller, A.A., Pease, S., Lenihan, H.S., 2012. TiO2 nanoparticles are phototoxic to marine phytoplankton. PLoS One 7 (1), e30321. Minetto, D., Libralato, G., Ghirardini, A.V., 2014. Ecotoxicity of engineered TiO2 nanoparticles to saltwater organisms: an overview. Environ. Int. 66, 18–27. Monopoli, M.P., Pitek, A.S., Lynch, I., Dawson, K.A., 2013. Formation and characterization of the nanoparticle-protein corona. Methods Mol. Biol. 1025, 137–155. Monopoli, M.P., Walczyk, D., Campbell, A., Elia, G., Lynch, I., Baldelli Bombelli, F., Dawson, K.A., 2011. Physical-chemical aspects of protein corona: relevance to in vitro and in vivo biological impacts of nanoparticles. J. Am. Chem. Soc. 133, 2525–2534. Morelli, E., Salvadori, E., Bizzarri, R., Cioni, P., Gabellieri, E., 2013. Interaction of CdSe/ ZnS quantum dots with the marine diatom Phaeodactylum tricornutum and the green alga Dunaliella tertiolecta: a biophysical approach. Biophys. Chem. 182, 4–10. Mu, L., Sprando, R.L., 2010. Application of nanotechnology in cosmetics. Pharm. Res. 27 (8), 1746–1749. Paterson, G., Ataria, J.M., Hoque, M.E., Burns, D.C., Metcalfe, C.D., 2011. The toxicity of titanium dioxide nanopowder to early life stages of the Japanese medaka (Oryzias latipes). Chemosphere 82, 1002–1009. Petersen, E.J., Diamond, S.A., Kennedy, A.J., Goss, G.G., Ho, K., Lead, J., Hanna, S.K., Hartmann, N.B., Hund-Rinke, K., Mader, B., Manier, N., Pandard, P., Salinas, E.R., Sayre, P., 2015. Adapting OECD Aquatic Toxicity Tests for Use with Manufactured Nanomaterials: key Issues and Consensus Recommendations. Environ. Sci. Technol. 49, 9532–9547. Piccinno, F., Gottschalk, F., Seeger, S., Nowack, B., 2012. Industrial production quantities and uses of ten engineered nanomaterials in Europe and the world. J. Nanopart. Res. 14, 1–11. Quigg, A., Chin, W., Chen, C., Zhang, S., Jiang, Y., Miao, A., Schwern, K., Xu, C., Santschi, P.H., 2013. Direct and indirect toxic effects of engineered nanoparticles on algae: role of natural organic matter. ACS Sustain. Chem. Eng. 1, 686–702. Ramirez-Garcia, S., Chen, L., Morris, M.A., Dawson, K.A., 2011. A new methodology for studying nanoparticle interactions in biological systems: dispersing titania in biocompatible media using chemical stabilisers. Nanoscale 3, 4617. Schiavo, S., Oliviero, M., Miglietta, M., Rametta, G., Manzo, S., 2016. Genotoxic and cytotoxic effects of ZnO nanoparticles for Dunaliella tertiolecta and comparison with SiO2 and TiO2 effects at population growth inhibition levels. Sci. Total Environ. 550, 619–627. Sendra, M., Sánchez-Quiles, D., Blasco, J., Moreno-Garrido, I., Lubián, L.M., Pérez-García, S., Tovar-Sánchez, A., 2017a. Effects of TiO2 nanoparticles and sunscreens on coastal marine microalgae: ultraviolet radiation is key variable for toxicity assessment. Environ. Int. 98, 62–68. Sendra, M., Yeste, M.P., Gatica, J.M., Moreno-Garrido, I., Blasco, J., 2017b. Homoagglomeration and heteroagglomeration of TiO2, in nanoparticle and bulk form, onto freshwater and marine microalgae. Sci. Total Environ. 592, 403–411. Shih, Y.-H., Liu, W.-S., Su, Y.F., 2012. Aggregation of stabilized TiO2 nanoparticle suspensions in the presence of inorganic ions. Environ. Toxicol. Chem. 31 (8), 1693–1698. Shirazi, A., Shariati, M., Keshavarz, Ak, Ramezanpour, Z., 2015. Toxic effect of aluminium oxide nanoparticles on green micro-algae Dunaliella salina. Int. J. Environ. Res. 9 (2), 585–594. Verdugo, P., Alldredge, A.L., Azam, F., Kirchman, D.L., Passow, U., Santschi, P.H., 2004. The oceanic gel phase: a bridge in the DOM-POM continuum. Mar. Chem. 92, 67–85. Von Moss, N., Slaveykova, V.I., 2014. Oxidative stress induced by inorganic nanoparticles in bacteria and aquatic microalgae – state of the art and knowledge gaps. Nanotoxicology 8, 605–630. Wang, H., Joseph, J.A., 1999a. Quantifying cellular oxidative stress by dichlorofluorescein biomineralization as an inspirational source of new approaches to silica production. J. Biotechnol. 70, 39–51. Wang, H., Joseph, J.A., 1999b. Quantifying cellular oxidative stress by dichlorofluorescein biomineralization as an inspirational source of new approaches to silica production. J. Biotechnol. 70, 39–51. Wang, Y., Zhu, X., Lao, Y., Lv, X., Tao, Y., Huang, B., Wang, J., Zhou, J., Cai, Z., 2016. TiO2 nanoparticles in the marine environment: physical effects responsible for the

Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.ecoenv.2017.10.024. References Adeleye, A.S., Conway, J.R., Perez, T., Rutten, P., Keller, A.A., 2014. Influence of extracellular polymeric substances on the long-term fate, dissolution, and speciation of copper-based nanoparticles. Environ. Sci. Technol. 48, 12561–12568. Aebi, H., 1974. Catalase. In: Bergmeyer, H.U. (Ed.), Methods of Enzymatic Analysis 2. Academic Press, NY, pp. 673–684. Amiard-Triquet, C., Amiard, J.-C., Mouneyrac, C., 2015. Aquatic Ecotoxicology. Advancing tolls for dealing with emerging risks. Academic Press, Elsevier. Aruoja, V., Dubourguier, H.C., Kasemets, K., Kahru, A., 2009. Toxicity of nanoparticles of CuO, ZnO and TiO2 to microalgae Pseudokirchneriella subcapitata. Sci. Total Environ. 407 (4), 1461–1468. Baker, T.J., Tyler, C.R., Galloway, T.S., 2014. Impacts of metal and metal oxide nanoparticles on marine organisms. Environ. Pollut. 186, 257–271. Beyer Jr, W.F., Fridovich, I., 1987. Assaying for superoxide dismutase activity: some large consequences of minor changes in conditions. Anal. Biochem. 161, 559–566. Blaise, C., Gagnè, F., Ferard, J.F., Eullaffroy, P., 2008. Ecotoxicity of selected nano-materials to aquatic organisms. Environ. Toxicol. 23, 591–598. Botta, C., Labille, J., Auffan, M., Borschneck, D., Miche, H., Cabié, M., Masion, A., Rose, J., Bottero, J.Y., 2011. TiO2-based nanoparticles released in water from commercialized sunscreens in a life-cycle perspective: structures and quantities. Environ. Pollut. 159 (6), 1543–1550. Bradford, M.M., 1976. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254. Brooks, J.R., Griffin, V.K., Kattan, M.W., 1986. A modified method for total carbohydrate analysis of glucose syrups, maltodextrins, and other starch hydrolysis products. Cereal Chem. 63 (5), 465–466. Brunelli, A., Pojana, G., Callegaro, S., Marcomini, A., 2013. Agglomeration and sedimentation of titanium dioxide nanoparticles (n-TiO2) in synthetic and real waters. J. Nanopart. Res. 15, 1684. Brunelli, A., Zabeo, A., Semenzin, E., Hirstozov, D., Marcomini, A., 2016. Extrapolated long-term stability of titanium dioxide and multi-walled carbon nanotubes in artificial freshwater. J. Nanopart. Res. 18, 113. Canesi, L., Corsi, I., 2016. Effects of nanomaterials on marine invertebrates. Sci. Total Environ. 565, 933–940. Castro-Bugallo, A., Gonzalez-Fernandez, A., Guisande, C., Barreiro, A., 2014. Comparative responses to metal oxide nanoparticles in marine phytoplankton. Arch. Environ. Contam. Toxicol. 67, 483–493. Chen, C.S., Anaya, J.M., Zhang, S., Spurgin, J., Chuang, C.Y., Xu, C., Miao, A.J., Chen, E.Y.T., Schwehr, K.A., Jiang, Y., Quigg, A., Santschi, P.H., Chin, W.C., 2011. Effects of engineered nanoparticles on the assembly of exopolymeric substances from phytoplankton. PLoS One 6 (7), e21865. Chen, L., Zhou, L., Liu, Y., Deng, S., Wu, H., Wang, G., 2012. Toxicological effects of nanometer titanium dioxide (nano-TiO2) on Chlamydomonas reinhardtii. Ecotoxicol. Environ. Saf. 84, 155–162. Cherchi, C., Gu, A.Z., 2010. Impact of titanium dioxide nanomaterials on nitrogen fixation rate and intracellular nitron storage in Anabaena variabilis. Environ. Sci. Technol. 44, 8302–8307. Corsi, I., Cherr, G.N., Lenihan, H.S., Labille, J., Hassellov, M., Canesi, L., Pondero, F., Frenzilli, G., Hristozov, D., Puntes, V., Della Torre, C., Pinsino, A., Libralato, G., Marcomini, A., Sabbioni, E., Matranga, V., 2014. Common strategies and technologies for the ecosafety assessment and design of nanomaterials entering the marine environment. ACS Nano 8, 9694–9709. Della Torre, C., Balbi, T., Grassi, G., Frenzilli, G., Bernardeschi, M., Smerilli, A., Guidi, P., Canesi, L., Nigro, M., Monaci, F., Scarcelli, V., Rocco, L., Focardi, F., Monopoli, M., Corsi, I., 2015. Titanium dioxide nanoparticles modulate the toxicological response to cadmium in the gills of Mytilus galloprovincialis. J. Hazard. Mat. 297, 92–100. EPA, 2009. External Review Draft-Nanomaterial Case Studies: Nanoscale Titanium Dioxide in Water Treatment and in Topical Sunscreen. National Center for Environmental Assessment, Office of Research and Development. US Environmental Protection Agency, pp. 222. Galloway, T., Lewis, C., Dolciotti, I., Johnston, B.D., Moger, J., Regoli, F., 2010. Sublethal toxicity of nano-titanium dioxide and carbon nanotubes in a sediment dwelling marine polychaete. Environ. Pollut. 158, 1748–1755. Handy, R.D., Owen, R., Valsami-Jones, E., 2008. The ecotoxicology of nanoparticles and nanomaterials: current status, knowledge gaps, challenges and future needs. Ecotoxicology 17, 315–325. Keller, A.A., Wang, H., Zhou, D., Lenihan, H.S., Cherr, G., Cardinale, B.J., Miller, R., Ji, Z., 2010. Stability and aggregation of metal oxide nanoparticles in natural aqueous matrices. Environ. Sci. Technol. 44 (6), 1962–1967. Klaine, S.J., Alvarez, P.J.J., Batley, G.E., Fernandes, T.F., Handy, R.D., Lyon, D.Y., Mahendra, S., McLaughlin, M.J., Lead, J.R., 2008. Nanomaterials in the environment: behavior, fate, bioavailability, and effects. Environ. Toxicol. Chem. 27, 1825–1851. Kosmulski, M., 2009. pH-dependent surface charging and point of zero charge. IV. Update and new approach. J. Colloid Interface Sci. 337, 439–448. Laemmli, U.K., 1970. Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature 227, 680–685. Li, F., Liang, Z., Zheng, X., Zhao, W., Wu, M., Wang, Z., 2015. Toxicity of nano-TiO2 on

192

Ecotoxicology and Environmental Safety 148 (2018) 184–193

E. Morelli et al.

environments: importance of extracellular polymeric substances. Environ. Sci. Technol. 46, 8764–8772. Zhou, C., Vitiello, V., Pellegrini, D., Wu, C., Morelli, E., Buttino, I., 2016. Toxicological effects of CdSe/ZnS quantum dots on marine planktonic organisms. Ecotoxicol. Environ. Saf. 123, 26–31. Zhu, X., Zhou, J., Cai, Z., 2011. The toxicity and oxidative stress of TiO2 nanoparticles in marine abalone (Haliotis diversicolor supertexta). Mar. Pollut. Bull. 63, 334–338.

toxicity on algae Phaeodactylum tricornutum. Sci. Total Environ. 565, 818–826. Wold, A., 1993. Photocatalytic Properties of TiO2. Chem. Mater. 5, 280–283. Xia, B., Chen, B., Sun, X., Qu, K., Ma, F., Du, M., 2015. Interaction of TiO2 nanoparticles with the marine microalga Nitzschia closterium: growth inhibition, oxidative stress and internalization. Sci. Total Environ. 508, 525–533. Zhang, S., Jiang, Y., Chen, C., Spurgin, J., Schwehr, K.A., Quigg, A., Chin, W., Santschi, P.H., 2012. Aggregation, dissolution, and stability of quantum dots in marine

193