Aquatic Toxicology 161 (2015) 154–169 Contents lists available at ScienceDirect Aquatic Toxicology journal homepage: www.elsevier.com/locate/aquatox Combined toxicity of two crystalline phases (anatase and rutile) of Titania nanoparticles towards freshwater microalgae: Chlorella sp V. Iswarya a , M. Bhuvaneshwari a , Sruthi Ann Alex a , Siddharth Iyer a , Gouri Chaudhuri a , Prathna Thanjavur Chandrasekaran b , Gopalkrishna M. Bhalerao c , Sujoy Chakravarty c , Ashok M. Raichur b , N. Chandrasekaran a , Amitava Mukherjee a,∗ a Centre for Nanobiotechnology, VIT University, Vellore, India Department of Materials Engineering, Indian Institute of Science, Bangalore, India c UGC-DAE CSR, Kalpakkam Node, Kokilamedu, India b a r t i c l e i n f o Article history: Received 13 December 2014 Received in revised form 9 February 2015 Accepted 11 February 2015 Available online 13 February 2015 Keywords: Binary mixture Bio uptake Crystalline phases Titania NPs Toxicity a b s t r a c t In view of the increasing usage of anatase and rutile crystalline phases of titania NPs in the consumer products, their entry into the aquatic environment may pose a serious risk to the ecosystem. In the present study, the possible toxic impact of anatase and rutile nanoparticles (individually and in binary mixture) was investigated using freshwater microalgae, Chlorella sp. at low exposure concentrations (0.25, 0.5 and 1 mg/L) in freshwater medium under UV irradiation. Reduction of cell viability as well as a reduction in chlorophyll content were observed due to the presence of NPs. An antagonistic effect was noted at certain concentrations of binary mixture such as (0.25, 0.25), (0.25, 0.5), and (0.5, 0.5) mg/L, and an additive effect for the other combinations, (0.25, 1), (0.5, 0.25), (0.5, 1), (1, 0.25), (1, 0.5), and (1, 1) mg/L. The hydrodynamic size analyses in the test medium revealed that rutile NPs were more stable in lake water than the anatase and binary mixtures [at 6 h, the sizes of anatase (1 mg/L), rutile NPs (1 mg/L), and binary mixture (1, 1 mg/L) were 948.83 ± 35.01 nm, 555.74 ± 19.93 nm, and 1620.24 ± 237.87 nm, respectively]. The generation of oxidative stress was found to be strongly dependent on the crystallinity of the nanoparticles. The transmission electron microscopic images revealed damages in the nucleus and cell membrane of algal cells due to the interaction of anatase NPs, whereas rutile NPs were found to cause chloroplast and internal organelle damages. Mis-shaped chloroplasts, lack of nucleus, and starchpyrenoid complex were noted in binary-treated cells. The findings from the current study may facilitate the environmental risk assessment of titania NPs in an aquatic ecosystem. © 2015 Elsevier B.V. All rights reserved. 1. Introduction Titanium dioxide (TiO2 ) nanoparticles were extensively used in various industrial applications and consumer products such as water treatment, medicine, cosmetics, and engineering (ICIS Chemical Report, 2010). Excessive usage of TiO2 nanoparticles have led to their exposure to the aquatic environment and their consequent hazards to the ecosystem (Furman et al., 2013). There are mainly three crystalline phases of TiO2 viz., anatase (tetragonal), rutile (tetragonal), and brookite (orthorhombic) (Cho et al., 2013). Among these, rutile is the most common and natural form of TiO2 , and it is an integral part of heavy minerals. It is employed in ∗ Corresponding author. Centre for Nanobiotechnology, VIT University Vellore, 632014 Tamil Nadu. Tel.: +91 416 2202620; fax: +91 416 2243092. E-mail address: amit.mookerjea@gmail.com (A. Mukherjee). http://dx.doi.org/10.1016/j.aquatox.2015.02.006 0166-445X/© 2015 Elsevier B.V. All rights reserved. optical elements due to its highest refractive indices and also used as a construct for refractory ceramics, pigments, etc. (Winkler, 2003; Yu et al., 2013). Anatase is extensively used in organic photovoltaics as an electron collecting layer (Small et al., 2012). Anatase is also applied as a catalytic support for the production of nanotubes and nanoribbons (Gregory et al., 2008). Both rutile and anatase phases are being extensively used in sunscreens (due to their high-energy-absorbing property), paints, plastics, paper, foods, electronics, and other applications (Ferguson et al., 2005; Mueller and Nowack 2008; Wang et al., 2006; Winkler, 2003). Since there is scarcity of brookite in nature, this form does not have significant economic importance (Allen et al., 2009). The released nanomaterials from different industries, consumer products may inevitably end up in the water bodies. They may potentially exert adverse impacts on the aquatic ecosystem due to their unique physical and chemical characteristics such as high reactivity and the photoactivity (Cardinale et al., 2012). Kaegi et al. V. Iswarya et al. / Aquatic Toxicology 161 (2015) 154–169 (2008) mentioned about the direct release of TiO2 NPs of about 16 ␮g/L into the surface water from aged paints. The predicted environmental concentration (PEC) of TiO2 NPs in surface water has been stated to be less than 1 ␮g/mL (Gottschalk et al., 2009). Hence, it becomes necessary to evaluate the toxicity of TiO2 NPs at its environmentally relevant concentrations. In our previous studies (Dalai et al., 2013; Pakrashi et al., 2013), we have evaluated the toxicity of TiO2 and Al2 O3 NPs under environmentally relevant low-exposure concentrations, i.e., 0.05, 0.5, and 1 mg/L, towards freshwater algae. Microalgae are of great importance for the maintenance of aquatic ecosystem. It can be used as a model for the studies of aquatic risk assessment of the nanomaterials (Aruoja et al., 2009). Lubick (2008) and Navarro et al. (2008) reported that the interaction of nanoparticles with algae influenced the aquatic toxicity of nanomaterials. Recently, Ji et al. (2011) noted that ZnO and TiO2 (anatase) nanoparticles caused severe damage to freshwater green algae. Cardinale et al. (2012) evaluated the toxicity of TiO2 nanoparticles (Degussa, 82% anatase/18% rutile) on three algal species viz., Chlorella vulgaris, Scenedesmus quadricauda, and Chlamydomonas moewusii. They observed that the gross primary production of these algae were reduced, and the reduction rate varied depending on the species type. Dalai et al. (2013) reported the photoinduced toxicity of TiO2 anatase NPs on Scenedesmus obliquus at low-exposure doses (≤1 ␮g/mL). They observed reduced cell viability, increased reactive oxygen species (ROS) generation, and membrane damage. Miller et al. (2012) observed the increased toxicity of TiO2 NPs on marine algae under UV-A irradiation than non-irradiation condition due to their increased photocatalytic activity. A number of prior reports strongly indicated that titanium dioxide nanoparticles caused severe toxicity towards freshwater microalgae species. However, information on their fate, behavior, and mechanism of uptake (pathways) based on crystallinity, shape, and other properties of materials is still lacking. Ji et al. (2011) stated that TiO2 NP toxicity varied with respect to the crystalline structure of TiO2 NPs. The two allotropic forms of TiO2 NPs viz., anatase and rutile have different surface properties and reactivity. Prior studies have demonstrated that anatase phases are more cytotoxic than those of rutile phases (Hirakawa et al., 2004). Braydich-Stolle et al. (2009) have observed that rutile TiO2 NPs were capable of initiating apoptosis through the formation of ROS, whereas pure anatase TiO2 NPs caused cell necrosis and membrane leakage in cells. Most of the previous toxicity studies on microalgae dealt with the anatase phase and P25 form of TiO2 (Chen et al., 2012; Clement et al., 2013; Dalai et al., 2013; Lee and An, 2013; Wang et al., 2008) and only a handful studies are available with the rutile phase (Ji et al., 2011). Therefore, it is pertinent to study the other crystalline phase, i.e., rutile TiO2 NPs, which is also commonly employed in commercial products/applications (Winkler, 2003; Yu et al., 2013). As UV-C radiation is a shorter and higher energized radiation than UV-A and UV-B, the photocatalytic action of TiO2 NPs was reported to be enhanced significantly under UV-C irradiation than other UV radiations (Termtanun, 2013). Due to its greater photolytic activity and energy, UV-C irradiation is being widely used for water disinfection and in the photodegradation studies, along with TiO2 NPs (Bushnaq et al., 2004). Since, UV-C radiation gets absorbed by the earth’s atmosphere, not much attention was given in evaluating its effects on the environment and the transformations caused by it (Holzinger and Lütz, 2006; Basti et al., 2009). McGivney (2007) studied the combined effect of UV-C, vacuum UV, and TiO2 on freshwater algae, Pseudokirchneriella subcapitata, and marine algae, Tetraselmis suecica, in a ballast water treatment system. They noticed that UV-C/TiO2 exerted a higher mortality compared to UV irradiation alone. Most of the algal toxicity studies on TiO2 NPs till date were carried out under different irradiation conditions such as visible light and UV light especially, UV A and UV B (Lee and An, 2013; Ji et al., 2011). To the best of our knowledge, there are limited reports highlighting the photocatalytic effects of UV-C on the toxicity of TiO2 NPs. Vileno et al. (2007) studied the effect of TiO2 NPs on the stiffness of human skin fibroblasts in the presence of UV-A and UV-C. Hence, it is crucial to evaluate the risk of TiO2 NPs in the presence of UV-C on Chlorella sp. In the natural ecosystem, various toxicants are expected to be present in the mixed form rather than as individuals. The toxicity assessment of a single toxicant alone does not adequately reflect the actual impact in the aquatic environment. The mode of action may vary for individual toxicants in the mixture; they may mask the effect of each other. It is indispensable to study the effect of their mixture in addition to the individual toxicants to adequately assess the environmental toxicity of different forms of toxicants (Jak et al., 1996). Zou et al. (2014) studied the toxicity of silver (Ag) NPs in the presence of TiO2 NPs on a ciliated protozoan, Tetrahymena pyriformis. Increased ecotoxicity was noted due to the coexistence of TiO2 NPs and Ag NPs. This elucidates that the level of toxicity increases in the presence of a mixture of nanoparticles rather than the individual forms (Utgikar et al., 2004) and provides an understanding of the complex interaction between different substances. The present investigation is the first of its nature to evaluate the combined toxicity of anatase and rutile NPs towards freshwater microalgae in a freshwater matrix. It may be hypothesized that there are inherent differences in the toxic effects of the two different crystal phases of titania NPs (anatase and rutile). Their binary combination would be more toxic than the respective individual phases. The aim of the present study was to elucidate the toxic effects of the two crystalline phases of titania nanoparticles i.e., anatase and rutile, as well as their binary mixture towards freshwater algae, Chlorella sp. at environmentally relevant low concentration levels (0.25, 0.5 and 1 mg/L) in the lake water matrix under UV-C irradiation. 2. Materials and methods 2.1. Chemicals Dry titanium(IV) dioxide (TiO2 ) nanopowder (anatase, <25 nm, CAS No: 1317-70-0, 99.7% trace metal basis; and rutile, <100 nm (∼10 nm Diam. × 40 nm L), CAS No: 1317-80-2, 99.5% trace metals basis) were purchased from Sigma–Aldrich, Missouri, USA, and their supplier information was summarized in Table 1. BG-11 broth was purchased from Himedia Labs Pvt., Ltd. (Mumbai, India). N, N-dimethylformamide was procured from SD fine chemicals Table 1 Information about the physicochemical parameters of two different types of TiO2 NPs has been represented in the table as per the supplier. Assay Form Particle size CAS No Surface area Density Bulk density 155 Anatase NPs Rutile NPs 99.7% trace metal basis Nanopowder <25 nm 1317–70–0 Spec. surface area 45–55 m2 /g 3.9 g/mL at 25 ◦ C 0.04–0.06 g/mL 99.5% trace metals basis Nanopowder <100 nm (∼10 nm Diam. × 40 nm L) 1317–80–2 Spec. surface area 130–190 m2 /g 4.17 g/mL at 25 ◦ C (lit.) 0.06–0.10 g/mL 156 V. Iswarya et al. / Aquatic Toxicology 161 (2015) 154–169 (Mumbai, India). 2′ ,7′ -dichlorofluorescein diacetate (DCFH-DA), and propidium iodide were obtained from Sigma–Aldrich (St. Louis, MO, USA). All the chemicals used in the study were of analytical grade. 2.2. Stock preparation of TiO2 nanoparticles A stock solution of TiO2 NPs (anatase and rutile) at a concentration of 100 mg/L was prepared in Milli-Q water. TiO2 NP suspension in Milli-Q water was sonicated for about 10 min in an ultrasonicator (130 W, 20 kHz, Sonics, USA) and further used for the characterization and toxicity studies. A stock suspension of TiO2 NPs was freshly prepared every time prior to the experiment. Homogeneity of the stock solution was ensured through the DLS analysis of the upper, middle, and bottom layers of the stock suspension. of algal species. Illumination was provided by white fluorescent lights (TL-D Super 80 Linear fluorescent tube, Philips, India) with an intensity of 3000 Lux. After 15 days, the algal species were examined under an optical microscope (Zeiss Axiostar Microscope, USA) to confirm the presence of green algae. Then, the pure algal cultures were obtained by the streak plate method. The obtained pure algal cultures were maintained in BG-11 broth in a day/night rhythm of 16 h/8 h under white fluorescent light at 23 ◦ C. The dominant algal species were isolated as the most occurring algal species and identified as Chlorella sp. through morphology identification (Fig. S1, Supplementary information). Chlorella sp., a single-celled, spherical green alga was further used for the toxicity assessment of NPs as per OECD guidelines (Organisation for Economic Cooperation and Development, 2011). 2.5. Characterization of lake water matrix used for the study 2.3. Characterization of nanoparticles (anatase and rutile) 2.3.1. Primary characterization of nanoparticles Surface morphology, primary particle size, and shape of TiO2 NPs were analyzed using transmission electron microscopy (Field Emission TEM, Libra Model 200, Zeiss, Germany) and scanning electron microscopy (SEM, Model S400, HITACHI, Japan). The hydrodynamic size of titania NPs (0.25, 0.5 and 1 mg/L) was analyzed in Milli-Q water using a Dynamic Light Scattering analyzer (90 Plus Particle Size Analyzer, Brookhaven Instruments Corp., USA) in order to determine their size in the aqueous solution. The specific surface areas of anatase and rutile nanoparticles were determined using Brunauer-Emmett-Teller (BET) method (Micrometrics, Tristar II 3020, USA). The UV-Visible absorption and diffuse reflectance spectra (DRS) of TiO2 nanoparticles (anatase and rutile) were evaluated to confirm the photoactivity of TiO2 NPs. X-Ray diffraction pattern (XRD) was analyzed to confirm the difference in the crystalline pattern of anatase and rutile nanoparticles. The surface functional groups of TiO2 NPs were analyzed for both anatase and rutile phases with the help of Fourier Transform Infra Red (FT-IR) Spectroscopy (IR Affinity 1, Shimadzu, Kyoto, Japan). 2.3.2. Solubility analysis The solubility of TiO2 nanoparticles was analyzed to evaluate the effect of dissolution in lake water (as detailed in our previous study by Dalai et al., 2012). TiO2 NPs (1 mg/L) dispersed in sterile filtered lake water was incubated for 72 h at room temperature under UV light condition (UV-C, Philips, 15 W, Wavelength <280 nm). After incubation, the dispersion was subjected to centrifugation at 12,000 rpm for 20 min, followed by filtration through a 0.1-␮m membrane filter and 3-kDa filter. The hydrodynamic size analysis was performed to ensure the complete removal of NPs in the filtrate. The concentration of Ti4+ ions retained in the filtrate was measured at a wavelength of 334.94 nm using inductively coupled plasma-optical emission spectrometry (ICP-OES, PerkinElmer Optima 5300 DV, USA) with a detection limit of 0.003 mg/L. The concentration of dissolved Ti4+ ions in the NP dispersion at 72 h was found to be below the detection limit of the instrument (ICPOES) for both anatase and rutile NPs. Thus, toxicity due to Ti4+ ion dissolution was negligible. 2.4. Isolation and identification of freshwater algal species The freshwater samples collected from different sites of VIT Lake, VIT University, Vellore, Tamilnadu, India was used for the isolation of freshwater green algal species. A standard isolation protocol was followed for the isolation of microalgae as described by Sadiq et al. (2011). The freshwater sample collected from VIT Lake was inoculated into sterile BG-11 broth and cultured in a growth chamber (I.L.E Co., India) at 23 ◦ C, for about 15 days for the growth Freshwater collected from the VIT Lake was used as an experimental matrix without any nutrient supplements in order to mimic the environmental conditions for the toxicity studies. The physicochemical parameters of the lake water were analyzed and found to be, pH: 7.76 ± 0.16, temperature: 26 ± 1.2 ◦ C, conductance: 2.145 ± 0.085 mS/cm, total dissolved solids: 775 ± 50 mg/L, total carbon: 26.795 ± 0.185 mg/L, and total organic carbon (TOC): 13.89 ± 0.72 mg/L. The lake water was also found to contain trace amounts of some metal ions (ICP-OES) such as Cu2+ , Zn2+ , Mn2+ , Cr6+ , and Al3+ . It was also found to contain some other inorganic ions (as detailed in our previous study by Pakrashi et al., 2011). Freshwater collected from the VIT Lake was immediately filtered, sterilized, and stored for further toxicity studies. The lake water was first coarse filtered through a blotting paper and then by Whatman No. 1 (pore size: 11 ␮m). It was further sterilized for about 15 min, at 121 ◦ C, 15 psi and filtered again with Whatmann No. 1 to remove the cell debris. This sterile lake water was further used for all the experimental studies. 2.6. Binary mixture toxicity study (UV light conditions) Algal cells in the exponential phase were harvested from the cultures, which were grown in media with a day and night rhythm of about 16 h:8 h in visible light (Philips, 18 W, intensity of 3000 Lux), by centrifugation at 7000 rpm, 0 ◦ C for 10 min. Then, an algal cell suspension with an initial cell population of 5 × 105 cells/mL was prepared in sterile lake water and further used for the toxicity studies. The algal cell suspension (5 mL) prepared in sterile lake water was interacted with an appropriate concentration of anatase and rutile NPs (NPs concentration per cell: 0.05 ␮g NPs per 105 cells for 0.25 mg/L, 0.1 ␮g NPs per 105 cells for 0.5 mg/L, 0.2 ␮g NPs per 105 cells for 1 mg/L). Toxicity studies were carried out with continuous UV irradiation (UV-C, Philips, 15 W, Wavelength <280 nm) for about 72 h, at 23 ◦ C, under static condition, i.e., without any mechanical shaking. Preliminary experiments on UV-C revealed that UV-C had a negligible effect on the growth of algal cells (93.22% growth in comparison with visible light). Hence, all the toxicity experiments were continued under UV-C irradiation. Individual toxicity of anatase and rutile NPs was evaluated over a range of concentrations (0.25, 0.5, 1, 2, 4, 8, 16, and 32 mg/L) with the help of a cell enumeration method. Their effective concentration values such as EC10, EC50, and EC90 were determined using the EPA Probit Analysis Program, Version 1.5. The concentration of the TiO2 NPs (anatase, rutile) used for the binary toxicity experiments was tabulated in Table 2. In the table, (0.25, 0.5) represents that 0.25 mg/L of anatase (A) NPs were interacted with 0.5 mg/L of rutile (R) NPs. Similarly, the toxicity experiments were also conducted for anatase and rutile nanoparticles individually at various concentrations, i.e., 0.25, 0.5, and 1 mg/L. V. Iswarya et al. / Aquatic Toxicology 161 (2015) 154–169 Table 2 Matrix represents the concentration of TiO2 NPs (A, R) used for the binary mixture toxicity study. Anatase (A) NPs (mg/L) Rutile (R) NPs (mg/L) 0.25 0.5 1 0.25 (0.25, 0.25) (0.5, 0.25) (1, 0.25) 0.5 (0.25, 0.5) (0.5, 0.5) (1,0.5) 1 (0.25, 1) (0.5, 1) (1,1) 2.6.1. Cell viability assessment 2.6.1.1. Cell enumeration. The algal growth inhibition was evaluated by the cell enumeration method in order to determine the effect of TiO2 NPs on the algal growth after 72 h under UV-C irradiation. Aliquots of algal cell suspension (TiO2 NPs treated and untreated cells) were loaded into the Neubauer chamber. The number of intact cells, i.e., without any distortion in the size and shape of algal cells was counted. The percentage growth inhibition of the NP-treated cells was calculated with respect to untreated cells. 2.6.1.2. Chlorophyll estimation. Chlorophyll is the primary pigment that plays a significant role in the photosynthesis of algae. Hence, it is measured to analyze the direct impact of nanoparticles on algal cells. The chlorophyll pigments were extracted with N, N, dimethylformamide (DMF) and quantified according to the protocol (Suzuki and Ishimaru 1990). After 72 h interaction, the algal cells were centrifuged at 7000 rpm for 10 min at 4 ◦ C. To the pellet containing only the algal cells, one mL of N, N, dimethylformamide was added and incubated for about 30 min at 4 ◦ C in dark condition. After a 30-min incubation, the dissolved pellet was centrifuged again at 7000 rpm for 10 min (4 ◦ C). The supernatant containing chlorophyll extract was subjected to chlorophyll analysis using a UV–vis spectrophotometer (Model U2910, HITACHI, Japan) at the wavelengths, 649 and 665 nm. Then, the reduction in the chlorophyll yield was calculated after its normalization with the chlorophyll yield of untreated cells. The type of nanoparticle interactions in the binary mixture such as synergism, antagonism, or addition was evaluated by the Abott’s statistical model (Teisseire et al., 1999). Abott’s modelling is widely used to estimate the effect of toxicants in the presence of a natural mortality source (Bliss, 1939). It is also a best way to compare the observed inhibitions with the expected growth inhibitions, in the case of binary mixtures (Chesworth et al., 2004). From the cell enumeration and chlorophyll results, the ratio of inhibition (RI ) was calculated for the assessment of binary mixture toxicity using the Abott’s formula. For this binary toxicity approach, the expected toxicity (Cexp ) of the binary mixture was computed using Eq. (1) from the inhibitions caused by the individual nanoparticles (anatase and rutile NPs). The expected toxicity is the percentage growth inhibition predicted from the growth inhibition observed for the individual NPs (anatase and rutile NPs) using the endpoints such as % growth inhibition (from cell enumeration) and reduction in chlorophyll yield (%). C exp = A + B − (AB/100) (1) where, A and B are the inhibitions caused by the individual anatase and rutile NPs, respectively, which was observed from the endpoints such as, % growth inhibition (from cell enumeration) and reduction in the chlorophyll yield (%). Then, the expected toxicity was compared with the observed toxicity using Eq. (2), and RI values were calculated. The ratio of inhibition (RI ) is the ratio of observed toxicity to expected toxicity. RI = observedtoxicity/expectedtoxicity(C exp ) (2) where, the observed toxicity is the percentage growth inhibition observed for binary mixtures after interaction with algae for 72 h 157 using the end points such as % growth inhibition (from cell enumeration) and reduction in chlorophyll yield (%). The interactive effects of the binary mixture were evaluated by comparing the RI with 1. If RI < 1, the toxic action is said to be antagonistic, if RI > 1, it is synergism, and if RI = 1, it is additive. The mean RI values calculated from the triplicates of the treatment should be greater or lower than standard deviation (SD) from 1, i.e. 1 ± SD, such that the interactive effect was assumed to be statistically different from additivity. Statistical differences were analyzed with two-way ANOVA at p < 0.01 using a graph pad prism, Version 5. 2.6.2. Stability of TiO2 nanoparticles in lake water matrix The hydrodynamic size of the nanoparticles was evaluated in the sterile filtered lake water under UV-C irradiation to determine the stability of nanoparticles individually and in a binary mixture over a period (0, 6, and 72 h). The morphological variations between the nanoparticles in a binary mixture was evaluated in an abiotic system. After interaction with anatase and rutile nanoparticles (equal ratio) devoid of algal cells under UV light condition for about 72 h, the NP suspension was subjected to TEM and SEM analysis. 2.6.3. Uptake/internalization of Ti (ICP-OES) Bioavailability of titania in the algal cells was analyzed to predict their uptake by the algae. TiO2 NP-treated algal samples were centrifuged at 7000 rpm for 10 min. The pellet containing only algal cells were further washed with PBS(1X) to remove the loosely bound NPs by centrifuging it once again at 7000 rpm for 10 min. Then, the pellet obtained was stored to evaluate the intracellular metal content (as detailed in our previous study by Dalai et al., 2013). Then, the samples were acid digested using concentrated HNO3 and subjected to ICP-OES analysis at a wavelength of 334.94 nm for Ti analysis. Relative Ti uptake was calculated by normalizing the Ti uptake obtained with the initial total metal (Ti) concentrations available in their respective concentrations for individual NPs, as well as binary combinations. The initial total metal concentration is the Ti content present in their respective concentrations of TiO2 NPs. 2.6.4. Microscopic studies 2.6.4.1. Scanning electron microscopy. Scanning electron microscopy helps to reveal the changes that occurred in the algal cell surface due to the nanoparticles. Aliquots of untreated and NP-treated algal cells (Anatase (1 mg/L), rutile (1 mg/L), and binary mixture (1, 1 mg/L)) were coated on a thin glass piece and air dried (Dalai et al., 2013). Then, it was subjected to gold sputtering and analyzed under a scanning electron microscope. 2.6.4.2. Confocal laser scanning microscopy. Confocal laser scanning microscopy (CLSM) helps to reveal the three-dimensional structure of the organisms (Pawley, 2006). Nanoparticle-treated (anatase, rutile and (1, 1) mg/L) and untreated algal cells were stained with 500 ␮L of propidium iodide for about 10 min. Stained algal cells were washed with 2X saline-sodium citrate (SSC) buffer thrice (as detailed in our previous study by Pakrashi et al., 2013). The stained algal cells were observed using confocal laser scanning microscopy (Zeiss LSM 510 META Confocal system, Germany) by employing the emission filter BP, 565–615 nm, and the excitation filter, LP 543 nm. 2.6.4.3. Transmission electron microscopy. The ultrastructural changes in algal cells due to the interaction with nanoparticles (anatase and rutile (1 mg/L) and binary mixture (1, 1) mg/L) were determined with the help of TEM (Dalai et al., 2014). Ultrathin sections of algal cells (untreated and treated cells) were prepared with a microtome and placed on the copper grids. Then, the copper 158 V. Iswarya et al. / Aquatic Toxicology 161 (2015) 154–169 Fig. 1. Transmission electron microscopy images of (A) anatase and (B) rutile nanoparticles. grids were observed under a transmission electron microscope (Philips CM12, Netherlands). 2.6.5. Oxidative stress assay-ROS Reactive oxygen species (ROS) generation is one of the most important mechanisms of cell death. It also acts as a bioindicator of the stress caused by some external factors on biological organisms (Mittler, 2002; Sevcu et al., 2011). DCFH-DA, a non-fluorescent cell membrane-permeable dye reacts with the ROS produced intracellularly and becomes fluorescent, and this can be further quantified by a standard protocol as described by Wang and Joseph (1999) with minor modifications. 5 mL of treated algal culture was incubated for about 30 min at room temperature under dark after the addition of 5 ␮L of DCFH-DA (100 ␮M). The fluorescence intensity of DCFH-DA was analyzed at excitation and emission wavelengths of 530 and 485 nm using a fluorescence spectrophotometer (Model G9800A, Cary Eclipse fluorescence spectrophotometer, Agilent technologies, USA). 2.7. Statistical analysis All the experiments were conducted in triplicates, results were represented as mean ± SE. Significant differences between the individual NPs and the binary mixture were calculated using a graph pad prism, Version 5. Two-way ANOVA (p < 0.01) was performed for the stability study of TiO2 NPs, cell viability assessment, and chlorophyll assays. One-way ANOVA (Tukey multiple comparison test, p < 0.05) was used for oxidative stress assay and internalization studies. 3. Results and discussion 3.1. Primary characterization of nanoparticles Literature reports on TiO2 NPs revealed that TiO2 nanoparticles were more reactive than its bulk form. They have enhanced physical, chemical, and electrical properties due to their size and large surface area per given mass (Karakoti et al., 2006). Nevertheless, the composition and phase of the material were still considered as a determining factor in the toxicological studies (Gojova et al., 2007; Sayes et al., 2006). Thus, the preliminary characterizations of as received nanoparticles were carried out. Anatase NPs were found to have cubical- and spherical-shaped particles, in the size range of about 2–8 nm (TEM, Fig. 1A). Aggregates of nanoparticles in a size range of 60–100 nm were also observed. Rutile NPs were found to be rod-shaped (Fig. 1B) and in a bunch of bundles to form spherical structures with a roughened surface as observed in SEM micrograph (Fig. S2, Supplementary information). The size of rod-shaped rutile NPs ranged from 20–100 nm in length and 2–14 nm in breadth. Their colloidal size was evaluated at 0 h in Milli-Q water using dynamic light scattering analysis. The effective diameters of the anatase NPs in MilliQ-water were found to be 445.48 ± 7.3 nm, 409.44 ± 4.87 nm, and 407.63 ± 5.57 nm for 0.25, 0.5, and 1 mg/L, respectively. Whereas, the effective diameters of the rutile NPs in Milli-Q water were noted to be 206.99 ± 6.99 nm, 199.41 ± 0.49 nm, and 205.02 ± 4.32 nm for 0.25, 0.5, and 1 mg/L, respectively. From the DLS analysis, it can be confirmed that the size of the nanoparticles did not vary significantly with increase in concentration (p > 0.05), for both anatase and rutile nanoparticles in Milli-Q water at 0 h. The surface areas of anatase and rutile NPs were found to be 94.90 and 90.46 m2 /g, respectively. The X-ray diffraction pattern of both crystalline phases of TiO2 NPs showed dominant characteristic crystalline peaks at (1 0 1) and (1 1 1) for anatase and rutile NPs, respectively (Fig. S3, Supplementary information). The FT-IR spectra of both anatase and rutile NPs showed a strong, broad peak at 3406.29 cm−1 due to O H stretching (Fig. S4, Supplementary information). It also showed the O H and C H bending at 1629.85 and 1402.25 cm−1 for anatase and 1631.78 and 1435.04 cm−1 for rutile NPs, respectively. Ti O O stretching vibration peaks were noted at 428.20 and 460.99 cm−1 for anatase and rutile NPs, respectively. The absorption spectra of TiO2 NPs revealed the characteristic absorption peaks at 337 and 296 nm for anatase and rutile NPs, respectively, in the UV region (Fig. S5, Supplementary information). The DRS spectra revealed absorption peak edges at 390 (UV region) and 407 nm (visible region) corresponding to the band gap energy (Eg ) of about 3.18 and 3.05 eV for anatase and rutile NPs, respectively (Fig. S6, Supplementary information). Marcone et al. (2012) and Reyes-Coronado et al. (2008) reported similar band gap energy values for anatase and rutile NPs. The higher band gap energy of anatase NPs clearly represents its higher photocatalytic activity than that of the rutile NPs (Scanlon et al., 2013). TiO2 NPs (anatase and rutile) exhibit differential photocatalytic action on different irradiation conditions, which might play a significant role in the cytotoxicity of NPs (Marcone et al., 2012). The preliminary characterizations of TiO2 NPs revealed that the properties of anatase and rutile NPs varied with their hydrodynamic size, shape, and photocatalytic activity. Their primary sizes and BET surface areas were found to be almost similar. 159 V. Iswarya et al. / Aquatic Toxicology 161 (2015) 154–169 Table 3 The effective concentration (EC) of anatase and rutile NPs for the growth inhibition of green algae, Chlorella sp. tested over a range of concentrations of NPs (0, 0.25, 0.5, 1, 2, 4, 8, 16, and 32 mg/L) under UV irradiation for 72 h. The median effective concentration (EC50 ) of rutile NPs was found to be 6.255 which is double than the EC50 of anatase NPs (3.362 mg/L). EC10 EC50 EC90 Anatase NPs (mg/L) Rutile NPs (mg/L) 0.053 3.362 213.177 0.067 6.255 585.961 3.2. Toxicity of TiO2 NPs as individual and binary toxicants 3.2.1. Cell viability assessment EC10 , EC50 , and EC90 values of anatase and rutile NPs as individual toxicants towards green algae, Chlorella sp. were represented in Table 3. The median effective concentrations (EC50 ) were about 3.36 and 6.25 mg/L for anatase and rutile NPs, respectively. A few recent studies on Chlorella sp. reported an EC50 value of about 16.12 ␮g/mL (Sadiq et al., 2011) and 4.9 ␮g/mL (Lin et al., 2012) for anatase nanoparticles with a particle size of <25 nm and 5–10 nm, respectively. Similarly, Sadiq et al. (2011) reported an EC50 value of 35.50 mg/L for micron-sized titania on Chlorella sp. The Ti4+ LC50 value was noted to be around 1404 ␮g Ti/L in the freshwater amphipod, Hyalella azteca (Malhi, 2012). To the best of our knowledge, EC50 and LC50 values for rutile NPs and dissolved Ti4+ ion on microalgae have not been reported so far. Individual toxicity of both the NPs was analyzed for low exposure concentrations (0.25, 0.5, and 1 mg/L), and the results were presented in Table 4. The algal growth rates were found to be reduced by about 18.41 ± 1.98%, 27.24 ± 1.53%, and 38.59 ± 1.28% for 0.25, 0.5, and 1 mg/L anatase NPs, respectively. Rutile NPs also showed a similar concentrationdependent growth inhibition of about 17.91 ± 0.38%, 24.21 ± 3.54%, and 29.76 ± 5.90% for 0.25, 0.5, and 1 mg/L, respectively. Though, a concentration-dependent growth inhibition was observed for both anatase and rutile NPs individually, the differences between them were statistically insignificant (p > 0.05) at low exposure concentrations (≤1 mg/L). Braydich-Stolle et al. (2009) reported a similar insignificant cytotoxic effect on HEL-30 mouse keratinocyte cell lines between the anatase and rutile TiO2 NPs at low concentrations (5 to 50 mg/L). Similarly, the algal growth inhibition was studied for the binary mixtures (Table 5). As the concentration of rutile NPs was increased at a fixed concentration of anatase NPs and vice versa, a concentration-dependent inhibition was observed. In the presence of 0.25 mg/L of anatase NPs, the algal growth inhibition was observed to be 15.89 ± 1.12%, 23.83 ± 0.5%, and 35.43 ± 1.77% at increasing concentrations of rutile NPs, 0.25, 0.5, and 1 mg/L, respectively. Similarly, concentration-dependent increments in the growth inhibition of about 29.89 ± 2.33%, 32.92 ± 0.67%, and 38.71 ± 1.22% were noted for 0.25, 0.5, and 1 mg/L rutile NPs, respectively, on addition of 0.5 mg/L anatase NPs to the respective mixture. Algal growth was found to be significantly reduced by about 39.6 ± 2.1%, 45.14 ± 0.76%, and 50.69 ± 2.56%, for 0.25, 0.5 and 1 mg/L rutile NPs, respectively, with 1 mg/L anatase NPs. A maximum growth inhibition of about 50.69 ± 2.56% was noted for (1, 1) mg/L, indicating an additive effect of the binary mixture. The enhanced toxic action of the binary mixture may be due to their increased photocatalytic activity under UV irradiation. The algal growth inhibition of the binary mixtures was compared in terms of total concentration of TiO2 NPs (A, R) in a mixture with its equivalent concentration of individual TiO2 NPs. The binary combination, (0.25, 0.25) mg/L, showed an increase in the algal growth with its equivalent concentration of TiO2 NPs, i.e., 0.5 mg/L of individual anatase or rutile NPs. Similarly, a significant reduction in growth inhibition, i.e., lesser toxicity was observed for the mixture (0.5, 0.5) mg/L, in comparison with its equivalent concentration, 1 mg/L of anatase NPs. In contrast, higher toxicity was noted in comparison with 1 mg/L of rutile NPs and here an antagonistic effect of binary mixtures could be observed. The additive and antagonistic effects of the binary mixtures were further confirmed by Abott’s statistical modeling (Table 5). The ratio of inhibition (RI ) of binary mixtures was calculated from the observed algal growth inhibition with the expected growth inhibition (Cexp ). RI values were observed within the range of 0.48 to 0.90 for all the binary mixtures. At (1, 1) mg/L, the ratio of inhibition was calculated as 0.90 ± 0.08. However, this RI value was not statistically significant (p > 0.05) from 1, indicating the additive type of action. A similar additive effect was observed in different concentrations of binary mixtures, (0.25, 1), (0.5, 0.25), (0.5, 1), (1, .25), and (1, 0.5) mg/L. For all these combinations, RI values were found to be below 1 and were statistically insignificant indicating the additive effect. On the other hand, antagonistic effect was noted for the binary mixtures, (0.25, 0.25), (0.25, 0.5), and (0.5, 0.5) mg/L. 3.2.2. Chlorophyll estimation Since chlorophyll is a primary photosynthetic pigment necessary for the algal cell function, it was used as an indicator for toxicity assessment. Reduction in the chlorophyll yield was compared between the anatase and rutile NPs individually (Table 4) as well as for the binary mixtures (Table 5). Both the NPs showed a concentration-dependent decrease in the chlorophyll content with respect to untreated cells. At 1 mg/L, anatase and rutile NP-treated cells showed a reduction in the chlorophyll yield by 31.64 ± 6.22% and 54.26 ± 5.43%, respectively. Significant differences in the reduction of chlorophyll content (p < 0.05) were noticed between the anatase and rutile NPs. Hartmann et al. (2010) reported that entrapment of particles by the cells (direct shading effect) results in decreased light availability, which in turn leads to disturbance in the energy transduction processes. As a result, oxidative stress occurs and creates an impact on the algal growth and chlorophyll content of the algal cells. Thus, rutile NPs showed a higher decrement in the chlorophyll content than anatase NPs. It was found to be contradictory to the growth inhibition (%) results. A significant decrease in the chlorophyll yield was observed for binary mixtures with respect to untreated cells, similar to cytotoxicity results (p < 0.05). The maximum reduction in the chlorophyll yield was observed for the binary mixture (1, 1) mg/L of about 76.06 ± 4.45%. At a fixed concentration of anatase NPs (0.25, 0.5, or 1 mg/L) and varied concentrations of rutile NPs (0.25, 0.5 and 1 mg/L), no concentration-dependent effect was observed for the binary combinations, except at the combination of (0.5, 0.25), (0.5, Table 4 Individual toxicity (%) of anatase and rutile NPs towards the green algae Chlorella sp. after 72 h was assessed using cell enumeration and chlorophyll yield. * denotes a statistical significance with respect to untreated cells at p < 0.05, n = 3. Anatase NPs Rutile NPs Concentration of TiO2 NPs Growth inhibition (%) Reduction in chlorophyll yield (%) Growth inhibition (%) Reduction in chlorophyll yield (%) 0.25 mg/L 0.5 mg/L 1 mg/L 18.41 ± 1.98* 27.24 ± 1.53* 38.59 ± 1.28* 1.30 ± 0.001 17.15 ± 3.92 31.64 ± 6.22* 17.91 ± 0.38* 24.21 ± 3.54* 29.76 ± 5.9* 48.01 ± 1.94* 52.30 ± 4.51* 54.26 ± 5.43* 160 V. Iswarya et al. / Aquatic Toxicology 161 (2015) 154–169 Table 5 Binary mixture toxicity of anatase and rutile NPs towards a freshwater algae, Chlorella sp. and are represented as observed toxicity. The ratio of inhibition (RI ) for the binary mixture toxicity representing the antagonistic/synergistic model calculated using Abott’s formula. Concentration of A, R (mg/L) Observed toxicitya Expected toxicity (Cexp )b Ratio of inhibition (RI )c Statistically significant (p ≤ 0.05) Type of binary action (A) Cell enumeration (0.25, 0.25) (0.25, 0.5) (0.25, 1) (0.5, 0.25) (0.5, 0.5) (0.5, 1) (1, 0.25) (1, 0.5) (1, 1) 15.89 23.83 35.43 29.89 32.91 38.71 39.6 45.14 50.64 ± ± ± ± ± ± ± ± ± 1.12 0.5 0.7 2.33 0.67 1.22 2.1 0.76 2.96 33.03 38.5 42.76 40.26 44.96 48.87 49.58 53.49 56.75 ± ± ± ± ± ± ± ± ± 1.4 4.26 4.59 1.51 1.5 4.59 1.15 1.96 4.34 0.48 0.64 0.85 0.75 0.73 0.8 0.8 0.84 0.9 ± ± ± ± ± ± ± ± ± 0.04 0.06 0.09 0.08 0.01 0.07 0.05 0.02 0.08 Yes Yes No No Yes No No No No Antagonistic Antagonistic Additive Additive Antagonistic Additive Additive Additive Additive (B) Chlorophyll yield (0.25, 0.25) (0.25, 0.5) (0.25, 1) (0.5, 0.25) (0.5, 0.5) (0.5, 1) (1, 0.25) (1, 0.5) (1, 1) 6.88 22.75 12.85 8.15 55.89 66.79 53.92 51.90 76.06 ± ± ± ± ± ± ± ± ± 6.8 5.16 2.95 8.99 4.34 5.54 4.92 1.77 4.45 48.69 52.92 54.86 56.93 60.48 62.11 64.46 67.39 68.73 ± ± ± ± ± ± ± ± ± 2.04 4.63 5.6 3.19 4.32 3.42 1.79 0.98 3.65 0.14 0.43 0.23 0.14 0.92 1.07 0.84 0.77 1.10 ± ± ± ± ± ± ± ± ± 0.14 0.10 0.05 0.16 0.07 0.09 0.08 0.03 0.06 Yes Yes Yes Yes No No No No No Antagonistic Antagonistic Antagonistic Antagonistic Additive Additive Additive Additive Additive a Observed toxicity – percentage growth inhibition observed after interaction with algae for 72 h using the end points–% growth inhibition (cell enumeration) and reduction in chlorophyll yield (%). b Expected toxicity – percentage growth inhibition predicted from the inhibition observed for the individual toxicants (anatase and rutile NPs). c Ratio of inhibition – ratio of observed toxicity to expected toxicity. 0.5), and (0.5, 1) mg/L. When the concentration of anatase NPs was varied (0.25, 0.5 and 1 mg/L) at a fixed concentration of rutile NPs (either 0.25 or 1 mg/L), a concentration-dependent effect was observed for the binary combinations. On the other hand, in the presence of 0.5 mg/L of rutile NPs, the chlorophyll yield was reduced by 22.75 ± 5.16%, 55.89 ± 4.34%, and 51.90 ± 1.77% at increasing concentrations of anatase NPs, 0.25, 0.5, and 1 mg/L, respectively. The reduction in the chlorophyll yield of the binary mixtures was compared with its equivalent concentration of individual TiO2 NPs. It was observed that the reduction in the chlorophyll yield was lesser for the mixture (0.25, 0.25) in comparison with 0.5 mg/L of individual anatase or rutile NPs. In contrast, the highest reduction in the chlorophyll yield was observed for the mixture (0.5, 0.5) mg/L in comparison with 1 mg/L of individual anatase NPs. In addition, no difference was found as compared to 1 mg/L of individual rutile NPs. This proved the antagonistic effect of the binary mixtures towards Chlorella sp. Similar to cytotoxicity results, the ratio of inhibition was calculated based on the reduction in the chlorophyll yield (Table 5) and found to be in the range of 0.14-1.1. Additive effect was obtained for most of the binary combinations, (0.5, 0.5), (0.5, 1), (1, 0.25), (1, 0.5), and (1, 1) mg/L. Antagonistic effect was noted for the binary combinations, (0.25, 0.25), (0.25, 0.5), (0.25, 1), and (0.5, 0.25) mg/L. Fargasova (2001) investigated the interaction of various metals with freshwater algae, Scenedesmus quadricauda. He stated that the metal-metal interactions reduced the unfavorable effects of metals Fig. 2. Stability of TiO2 nanoparticles (anatase and rutile) in sterile and filtered lake water as individual and binary mixture. Rutile NPs were more stable than anatase NPs. Particle size of the NPs in the binary mixture were increased owing to the aggregation of NPs after 6 h. Asterisk (*) represents that the size of the NPs at 6 h were statistically significant with respect to particle size at 0 h (p < 0.01). The symbol, # represents that the size of the NPs at 72 h were statistically significant with respect to particle size at 6 h (p < 0.01). V. Iswarya et al. / Aquatic Toxicology 161 (2015) 154–169 Fig. 3. TEM micrograph of the binary mixture (A, R) showing the aggregation of NPs under UV irradiation. Rutile NPs were surrounded by the anatase NPs and created an interaction between them owing to the aggregation of NPs. on the algal growth and its photosynthetic activity, which resulted in antagonism. Thus, the anatase–rutile interaction may have influenced the mode of binary action (antagonism, additive) based on its concentration in the mixture apart from the shading effect and stress induced by nanoparticles. 3.3. Stability of anatase and rutile nanoparticles in the test system as individual and binary toxicants The colloidal stability of nanoparticles depends mainly on the test medium used for the study, which in turn has an impact on its reactivity and toxicity (Panessa-Warren et al., 2009). Therefore, the colloidal size of nanoparticles was evaluated in the sterile filtered lake water, in order to reveal the size effect of nanoparticles at 0, 6, and 72 h, especially in binary mixtures (Fig. 2). The effec- 161 tive diameter of anatase NPs (0.25 mg/L) in sterile filtered lake water was noted to be 524.49 ± 0.09 nm, 761.07 ± 93.48 nm, and 830.47 ± 4.54 nm at 0, 6, and 72 h, respectively. However, there were only insignificant variations in the size (p > 0.01) by comparing the time ranges from 0-6 h and 6–72 h. The effective diameter for 0.5 mg/L anatase NPs was found to be about 574.76 ± 30.02 nm, 838.19 ± 78.82 nm, 1164.205 ± 14.08 nm at 0, 6 and 72 h, respectively. Similar to the lower concentration (0.25 mg/L), the size differences for the time range 0–6 h and 6–72 h were found to be statistically insignificant (p > 0.01). The effective diameter of the anatase NPs, 1 mg/L was found to be significantly increased from 522.94 ± 18.54 (0 h) to 948.83 ± 35.01 and 1723.27 ± 51.18 nm at 6 and 72 h, respectively. The effective diameter of rutile NPs (0.25 mg/L) in sterile filtered lake water was found to be 311.95 ± 2.4, 485.15 ± 34.52 and 737.2 ± 16.76 nm at 0, 6, 72 h, respectively (Fig. 2). However, no significant size differences were observed for the time periods (0–6 h and 6–72 h). Similarly, the effective diameter (0.5 mg/L) was found to be increased from 333.10 ± 19.13 (0 h) to 508.85 ± 29.06 and 1228.79 ± 75.61 nm at 6 and 72 h, respectively. The effective diameter of the rutile NPs (1 mg/L) in sterile filtered lake water was found to be 301.34 ± 2.59, 555.74 ± 19.93 and 1745.84 ± 89.5 nm at 0, 6, 72 h, respectively. There was no significant increase in the particle size of rutile NPs until 6 h for 0.5 and 1 mg/L. However, both concentrations showed a significant increase (p < 0.01) in their size between 6 h and 72 h. This confirmed the aggregation of nanoparticles, which may be due to the interaction between the nanoparticles and the natural colloids (<200 nm) present in the lake water. Natural colloids of the lake water such as natural organic matter (NOM) derived from geochemical as well as microbial processes form complexes with the nanoparticles. These complexes in turn alter the colloidal stability of the nanoparticles in the aquatic environment (Ghosh et al., 2008; Hyung et al., 2007). The aggregation profile of the NPs revealed that rutile NPs were more stable in the sterile and filtered lake water up to 6 h than anatase NPs under UV-C irradiation. Thus, rutile NPs had a higher impact on the chlorophyll content of algal cells than anatase NPs owing to its stability. When rutile NPs are added to anatase NPs, the effective hydrodynamic size for the mixture in lake water was found to be less than the effective hydrodynamic size of individual NPs at 0 h (Fig. 2). A substantial increase in the effective diameter of the Fig. 4. Relative Ti uptake by the algal cells after interaction with anatase and rutile NPs, both individually and as a binary mixture for 72 h. Highest Ti internalization was observed at the binary mixture, (0.5, 0.5) mg/L. Asterisk (*) denotes that NP uptake were of statistically insignificant (p > 0.05) in comparison with individual and binary mixture. 162 V. Iswarya et al. / Aquatic Toxicology 161 (2015) 154–169 Fig. 5. Scanning electron microscopic images of Chlorella sp. (A) Untreated cells with uncompromised cell membrane; (B–D) Anatase NP-treated cells; (E and F) Rutile NP-treated cells and (G–I) binary mixture-treated cells (1, 1) mg/L. nanoparticles in a binary mixture was noticed over a period (0, 6, 72 h). At (1, 1) mg/L, the effective hydrodynamic size rapidly increased from 386.56 ± 8.7 nm (0 h) to 1620.24 ± 237.87 nm (6 h) and 3169.59 ± 223.38 nm (72 h). A significant increase in the effective diameter was noted for all the concentrations of the binary mixture used in the study from 0 to 6 h. After 6 h, a significant increase in the effective diameter was observed for the combinations, (0.5, 1), (1, 0.5), and (1, 1) mg/L. For the combinations, (0.25, 0.25) (0.25, 0.5), (0.25, 1), (0.5, 0.25), (0.5, 0.5), and (1, 0.25) mg/L, the increase in effective diameter from 6 to 72 h was found to be statistically insignificant (p > 0.01). From the hydrodynamic size analysis, it was noted that the nanoparticles tend to aggregate rapidly in the aqueous solution (lake water) when the anatase and rutile NPs coexist. The aggregation of nanoparticles was quite pos- sibly influenced by the interacting forces acting between the two different nanoparticles. In addition, the natural colloids available in the lake water exacerbated the aggregation of NPs. In addition, the pH, ionic composition, and ionic strength of the aqueous suspension may influence the aggregation of NPs in the lake water (Sharma, 2009; Keller et al., 2010). Under abiotic conditions, the surface interactions of nanoparticles in a binary mixture were observed using SEM and TEM analysis in order to reveal the type of interaction between them. The SEM micrographs (Fig. S7, Supplementary information) revealed that coexistence of anatase and rutile nanoparticles in Milli-Q water under UV light (after 72 h interaction) induced the aggregation of NPs (Fig. S7). The final size range of the aggregates reached 1–20 ␮m. Sun and Smirniotis (2003) reported that the agglomer- V. Iswarya et al. / Aquatic Toxicology 161 (2015) 154–169 163 Fig. 6. Confocal laser scanning micrographs of untreated and treated cells. (A) Untreated cell; (B–D) Anatase-treated cells (1 mg/L) with two to three nuclei (white arrows) and few are with scattered nucleus (yellow arrows); (E and F) Rutile-reated cells (1 mg/L) showed nuclear damage, smeared nucleus, and release of nuclear contents into the cytoplasm; G–J) Figure & Table captions 2 binary-treated cells (1, 1 mg/L) showing both multi-nuclei cells (white arrows) and smeared nucleus (yellow arrows). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). ation was one of the conditions in which the anatase and rutile titania nanoparticles can interact. They also stated that the interactions between anatase and rutile and also that its effect were dependent on their relative fermi levels and particle shape. TEM analysis also elucidated the similar aggregation of NPs in an abiotic system (Fig. 3) and substantiated the interparticle interactions. The spherically shaped, aggregated anatase NPs were noted to be surrounded by the rod-shaped rutile NPs. Even though both the NPs were hydrophilic (Creutzenberg, 2013), an interfacial interaction between the anatase and rutile NPs was observed and responsible for the additive toxicity of the binary mixture. Zou et al. (2014) reported that the Ag NP toxicity was enhanced in Tetrahymena pyriformis by the formation of activated TiO2 –Ag NPs complexes under continuous light condition. They also stated that the surface chemistry of Ag NPs was changed in the existence of TiO2 NPs under various illumination modes, which led to different toxicity effects. 3.4. Uptake/Internalization of Ti When two or more toxicants were present in the test system, their interaction with the algal cells might follow several pathways. Internalization or uptake of nanoparticles is one of the possible pathways apart from the cell membrane damage. Thus, the uptake of Ti by the algal cells was evaluated to understand the action of anatase and rutile NPs both individually as well as a binary mixture upon UV irradiation using ICP-OES analysis (Fig. 4). The highest Ti uptake of about 76% was observed at both 0.25 and 0.5 mg/L of rutile NP-treated cells. A significant decrease in the Ti uptake was noticed from 0.5 to 1 mg/L for both anatase and rutile NPs. Anatase NP-treated cells showed lesser Ti uptake (%) than rutile NPtreated cells. However, the differences in Ti uptake between anatase and rutile NPs were found to be statistically insignificant for 0.25, 0.5 and 1 mg/L with p > 0.05. Ekstrand-Hammarström et al. (2012) stated that the primary agglomeration of nanoparticles determines their availability and plays a significant role in the cellular uptake. Anatase NPs were found to be agglomerated in the sterile filtered lake water as observed from the DLS stability results, which in turn reduced their bioavailability for the cellular uptake. Thus, the highest Ti uptake was observed for rutile NPs owing to their higher bioavailability than the anatase NPs. NP adhesion on the algal cell surface facilitated the NP uptake that was further validated with the SEM micrograph. For a binary mixture of NPs (Fig. 4), the highest Ti uptake (%) was observed to be around 57% at (0.5, 0.5) mg/L. Based on the concentration of anatase and rutile NPs in a mixture, total Ti uptake was found to vary. A concentration-dependent increase in the relative 164 V. Iswarya et al. / Aquatic Toxicology 161 (2015) 154–169 Fig. 7. Transmission electron micrograph of untreated and anatase-treated cells (1 mg/L). (A and B) Untreated cells with the intact features: compact nucleus, cup-shaped chloroplast, and starch-pyrenoid complex; (C and D) Anatase-treated cells showing the destruction of starch grains in the starch-pyrenoid complex. (D and F) Nucleus were surrounded by the black colored particles (red-colored arrows); (E) NPs were accumulated in a vacuole space region that confirmed the uptake of NPs along with lipid globules formation (yellow arrows). (F) Nucleus damage along with mis-shaped chloroplast (inset: NP accumulation around the nucleus); (G) Lipid globules formation (yellow arrows); (H) Mis-shaped chloroplast. N-Nucleus, SP-Starch Pyrenoid complex, C-Chlorophyll, L-Lipid production, P-Pyrenoid. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). Ti uptake was observed for all the binary combinations, as the concentration of rutile NPs increased with a constant concentration of anatase NPs (0.25, 0.5 or 1 mg/L), except the combination, (0.5, 1) mg/L. As the concentration of anatase NPs was increased with a constant concentration of rutile NPs, the total Ti uptake was found to be reduced for all the binary mixtures except at (0.5, 0.5) mg/L. It might be due to the adsorption of rutile NPs over the anatase NPs (TEM, Fig. 3) in addition to the agglomeration of NPs, and this in turn resulted in less bioavailability of TiO2 NPs. Nur et al. (2014) stated that the agglomeration of nanoparticles leads to the sedimentation of nanoparticles and in turn results in less bioavailability of NPs in the suspension, thereby altering the dose-response relationship in the toxicity analysis. Differences in the NP uptake were found to be statistically insignificant (p > 0.05) when comparing individual with binary mixture. Thus, the crystalline behavior of anatase and rutile NPs does not have any significant effect on the Ti uptake by the algal cells. From these results, it can be concluded that Ti uptake in a binary mixture was found to be dependent on the bioavailability of NPs. In turn, it was also influenced by various factors such as, agglomeration of particles, concentrations of anatase and rutile NPs in the mixture, and interaction between nanoparticles. 3.5. Microscopic studies Microscopic examination was performed for the individual as well as binary mixture NPs at the highest toxic concentration, in order to visualize the internal damages in the cell by NPs. V. Iswarya et al. / Aquatic Toxicology 161 (2015) 154–169 165 Fig. 8. Transmission electron micrograph of rutile-treated cells (1 mg/L). (A) Complete destruction of the internal organelles; (B) Necrotized cell with more production of starch grains (blue arrows); (C and E) Undamaged starch-pyrenoid complex with destroyed chloroplasts and other organelles (yellow arrows); (D) Black putative body formation in the cell (orange-colored arrows); (E–G) Disrupted thylakoids, absence of nucleus, and lack of mitochondrial organelles. N-Nucleus, SP-Starch Pyrenoid complex, C-Chlorophyll. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). 3.5.1. Scanning electron microscopy The changes on the algal cell surface by the action of the nanoparticles were observed using SEM analysis (Fig. 5). The untreated cells were of round shape with a smooth surface showing an uncompromised cell membrane (Fig. 5A). The anatase NPtreated cells showed aggregation as well as attachment of cells (Fig. 5B) forming a network-like rough surface with altered morphology (Fig. 5C). The internalization of the NPs and the distorted cell membrane were clearly visible from the SEM micrograph (Fig. 5D). Rutile-treated cells also demonstrated similar aggregation of cells with distorted morphology and the release of exudates from the cells (Fig.5E and F). A differential distortion in the cell surface morphology was observed with anatase- and rutile-treated cells. Whereas, enhanced aggregation and grouping of cells with typically altered surface morphology (Fig. 5G–I) were noted for the binary mixture-treated cells (1, 1) mg/L. The exudates released by the algal cells helped in the aggregation and attachment of cells together, as a result of NP stress. Dalai et al. (2013) reported a similar agglomeration of cells on interaction with anatase NPs. 3.5.2. Confocal laser scanning microscopy Confocal laser scanning microscopic (CLSM) images (Fig. 6) further helped us to confirm the toxic behavior of anatase and rutile NPs both individually and in the binary mixture. The untreated cells (Fig. 6A) were not stained owing to the intact cell membrane. The treated cells showed red fluorescence due to the uptake of the nuclear-specific stain, propidium iodide, as a result of the compromised cell membrane. It was also observed that anatase and rutile-treated cells showed differential toxicity as compared to individual NPs. In anatase-treated cells, the nucleus was divided into two to three nuclei (Fig. 6B–D), and few are smeared (Fig. 6D) clearly indicating its nucleus (DNA) specific action on the algal cells. Similar observations were reported by Dalai et al. (2013) and Jin et al. (2011) for anatase-treated cells. The nuclear contents of the cell were released into the cytoplasm in the rutile-treated cells. This appeared as smeared or scattered nucleus (Fig. 6E and F, yellow arrows) as a result of the interaction with rutile NPs in the nucleus of the cell revealing their probable genotoxicity. Thus, both anatase and rutile NPs induced DNA damage as a result of NP entry into the cell, which further caused the cell membrane damage and NP uptake into the cell. Binary mixture-treated cells (1, 1 mg/L) revealed the combined action of anatase NPs (micronuclei formation) and rutile NPs (diffused nucleus) (Fig. 6G–J). This was found to have more toxic effects in comparison with the individual TiO2 NP (anatase and rutile)treated cells. Gurr et al. (2005) and Hund-Rinke and Simon (2006) noted lipid peroxidation and micronuclei formation on treatment with different phases of TiO2 NPs in various organisms, such as human bronchial epithelial cells, algae, and Daphnia sp. The aggregation of algal cells was also noted as a result of binary action of NPs. 3.5.3. Transmission electron microscopy Similarly, TEM images also confirmed the substantial damage due to the internalization of titania NPs into the algal cells. The untreated cells were compact and round shaped with its typical characteristic organelles, i.e., starch-pyrenoid complex centered along the nucleus and the well-packed cup-shaped chlorophyll (Fig. 7A and B). The anatase NP-treated cells displayed specific interactions of the NPs with the starch-pyrenoid complex and nucleus, which showed altered cell membrane. In the anatase-treated cells, starch grains were found to be destroyed in the starch-pyrenoid complex, leaving pyrenoid alone (Fig. 7C and D). The nucleus was found to be surrounded by black-colored particles (Fig. 7D and F, red pointed arrows). It might be due to the stress produced by the nanoparticles at the early stage necrosis of the cells. Lipid glob- 166 V. Iswarya et al. / Aquatic Toxicology 161 (2015) 154–169 Fig. 9. Transmission electron micrograph of binary mixture (A, R)-treated cells (1, 1mg/L). (A) Complete destruction of the cells; (B) Nanoparticle accumulation in the cell (red colored arrows); (C and E) Separation of starch grains (blue arrows) from starch-pyrenoid complex; (C, D, F) Disrupted thylakoids, absence of nucleus, and lack of mitochondrial organelles (yellow arrows); N-Nucleus, C-Chlorophyll. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). ules formation (Fig. 8E and G, yellow pointers) and mis-shaped chloroplast (Fig.7F and H) were also observed as a result of nanoparticle stress. Kang et al. (2014) reported that lipid production was enhanced in the Chlorella vulgaris by the TiO2 anatase NPs under UV-A irradiation through oxidative stress. In addition, NP internalization was very clearly observed in cells where the NPs were accumulated in a vacuole space region (Fig. 8E). In the rutile-treated cells, a distorted cell membrane with the clear changes in the intracellular organelles (Fig. 8) were noted. Rutile NPs were found to have interacted with the internal organelles, mainly mitochondria and chloroplasts. The cells were found to be necrotized with more production of starch grains (Fig. 8B). Cell membrane damage occurred with complete destruction of the internal organelles (Fig. 8A) leaving the starch-pyrenoid complex without any damage to it (Fig. 8C and D). In some cells, the black putative bodies were observed (Fig. 8D–G) along with disrupted thylakoids, absence of a nucleus, and lack of mitochondrial organelles. Therefore, rutile action was more specific to the chloroplasts, and other cellular organelles apart from the nucleus that varied its action from anatase NPs. Chloroplasts, together with mitochondria are the prime sites of reactive oxygen species production. Therefore, disruption of the chloroplasts and mitochondria may generate intermediate signals involved in programmed cell death and induces the apoptosis of cells (Apel and Hirt, 2004; Van Breusegem and Dat, 2006). The destabilized cell membrane, absence of nucleus and starchpyrenoid complex, complete destruction of mitochondria and other internal organelles, mis-shaped as well as distorted chlorophylls were noted in the binary NP-treated cells (Fig. 9A–F). Starch grains were found to be separated from the starch–pyrenoid complex (Fig. 9E) as of anatase-treated cells and putative oil bodies were also observed as a result of NP stress. Cell membrane injury and complete destruction of the cells indicated that the binary NPtreated cells underwent a necrotic process, which in turn induced cell death. Several previous reports suggested that ROS generation induces damage to DNA (Dick et al., 2003; Olmedo et al., 2005; Yeo and Kang 2006). However, the damage is not only to the DNA, but also to the proteins, lipids and other metabolites in cells (Hirakawa et al., 2004; Tucci et al., 2013) during the TiO2 photocatalysis. All these may lead to the release of proapoptotic factors and cause programmed cell death (Jin et al., 2011). 3.6. Oxidative stress assessment ROS generation is one of the most important mechanisms inducing cell death. A concentration-dependent ROS generation was observed for both anatase and rutile NP- treated cells (Fig. 10) and was statistically significant at p < 0.05 with respect to the untreated cells. Anatase-treated cells showed ROS generation of about 65.23 ± 0.22%, 77.53 ± 3.2%, and 101.54 ± 2.82% at increasing concentrations of 0.25, 0.5, and 1 mg/L, respectively. Similarly, rutile-treated cells showed an increase in ROS generation of about 27.01 ± 0.44%, 113 ± 3.72%, and 161.49 ± 4.71% for 0.25, 0.5, and 1 mg/L, respectively. Rutile-treated cells showed a higher ROS generation than anatase-treated cells. ROS generation was found to have a direct correlation with the decrement in the chlorophyll con- V. Iswarya et al. / Aquatic Toxicology 161 (2015) 154–169 167 Fig. 10. Oxidative stress assessment of individual toxicants and binary mixture with the help of ROS assay. * denotes a statistical significance with respect to untreated cells (p < 0.05). tent for both anatase and rutile NP-treated cells. Varela-Valencia et al. (2014) noted that the physicochemical characteristics of TiO2 NPs varied the CAT, GST, and SOD transcript levels in O. niloticus, which were treated with the NPs. They also reported that ROS generation was higher for the rutile phase than that of anatase phase. Hirakawa et al. (2004) and Braydich-Stolle et al. (2009) also described that anatase NPs induced cell necrosis through cell membrane damage and rutile NPs initiated apoptosis through ROS formation. Thus, the crystalline phase impacts the type of mechanism (cell necrosis or apoptosis) persuading the cell death. Similarly, ROS generated by the binary mixture of NPs was found to be significant with respect to the untreated cells for all the binary combinations. ROS generation was observed to be higher (156.05 ± 4.05%) at (0.5, 0.5) mg/L. While, the highest combination, (1, 1) mg/L showed a less ROS generation of about 77.02 ± 2.97%. Upon the addition of 0.25 mg/L rutile NPs to 0.25, 0.5, and 1 mg/L of anatase NPs, a concentration-dependent ROS generation was obtained and the levels were 56.02 ± 1.43%, 68.37 ± 9.79%, and 102.33 ± 5.16%, respectively. In contrast, no concentrationdependent ROS generation was observed for any other combination of binary mixtures, i.e., at the fixed concentration of rutile NPs (0.5 and 1 mg/L) with varying concentration of anatase NPs (0.25, 0.5, and 1 mg/L). ROS production was significantly lower at high concentration of anatase (1 mg/L) with increasing concentration of rutile (0.25, 0.5, and 1 mg/L). Braydich-Stolle et al. (2009) also indicated that the concentration and the level of rutile phase in the NP mixture play a major role in the ROS production. Nel et al. (2006) and Xia et al. (2006) illustrated that oxidative stress occurs when there is an imbalance between the oxidative pressure and antioxidant defense due to the ROS generation in the mitochondria. Thus, decreased ROS generation might be due to the antioxidant defense mechanisms induced by the cell in response to the NP stress. The decrease in the chlorophyll content, generation of reactive oxygen species, and cell membrane integrity was strongly influenced by the crystallinity of the nanoparticles. The surface interactions between the NPs and the algal cell wall might have facilitated the uptake of the NPs. The NP interaction with the algal cells and substantial morphological damages were confirmed through microscopic analysis (SEM, CLSM, and TEM). In summary, the current study explored the possible underlying mechanism(s) of the toxicity of two different TiO2 crystalline phases as well as their combined form towards freshwater microalgae paving the way for further in-depth studies along similar directions in nanoecotoxicology. Acknowledgements We acknowledge Life Science Research Board-Defense Research and Development Organization (LSRB-DRDO), Government of India for providing the grant throughout this research work. We would also like to thank Sophisticated Analytical Instrumentation Facility (SAIF), Department of Science & Technology (DST) at Indian Institute of Technology, Madras, for the Scanning electron microscopy (SEM) and ICP-OES analysis facility and Christian Medical College, Vellore, India, for the Transmission Electron Microscopy facilities used in our study. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.aquatox. 2015.02.006. References 4. Conclusion The present study clearly supports the hypothesis that the two crystalline phases of TiO2 NPs showed different toxic impacts towards the freshwater microalgae, Chlorella sp., and their combination demonstrated both antagonistic and additive toxicity effects, when compared to the individual phases in the freshwater system. The difference in reduction of cell viability caused by the two crystalline phases of titania NPs was statistically insignificant. Allen, N.S., Edge, M., Verran, J., Caballero, L., Abrusci, C., Stratton, J., Maltiby, J., Bygott, C., 2009. Photocatalytic surfaces: environmental benefits of Nanotitania. Open Mater. Sci. J. 3, 6–27. Apel, K., Hirt, H., 2004. Reactive oxygen species: metabolism oxidative stress and signal transduction. Annu. Rev. Plant Biol. 55, 373–399. 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. Basti, D., Bricknell, I., Beane, D., Bouchard, D., 2009. Recovery from a near-lethal exposure to ultraviolet-C radiation in a scleractinian coral. J. Inv. Path. 101, 43–48. 168 V. Iswarya et al. / Aquatic Toxicology 161 (2015) 154–169 Bliss, C.I., 1939. The toxicity of poisons applied jointly. Ann. Appl. Biol. 26, 585–615. Braydich-Stolle, L.K., Schaeublin, N.M., Murdock, R.C., Jiang, J., Biswas, P., John, J., Saber, M.H., 2009. Crystal structure mediates mode of cell death in TiO2 nanotoxicity. J. Nanopart. Res. 11, 1361–1374. Bushnaq, Z., Othma, M.Z., Roddick, F.A., 2004. Evaluation of Uva, Uvb and Uvc Photolysis for the Removal of Atrazine from Polluted Water. In: Nadebaun, P. (Ed.), Environ. 04 Proceedings. Australian Water Association, Sydney. Cardinale, B.J., Bier, R., Kwan, C., 2012. Effects of TiO2 nanoparticles on the growth and metabolism of three species of freshwater algae. J. Nanopart. Res. 14 (8), 1–8. 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. Chesworth, J.C., Donkin, M.E., Brown, M.T., 2004. The interactive effects of the antifouling herbicides Irgarol 1051 and Diuron on the seagrass Zostera marina (L.). Aquat. Toxicol. 66, 293–305. Cho, W.S., Kang, B.C., Lee, J.K., Jeong, J., Che, J.H., Seok, S.H., 2013. Comparative absorption, distribution, and excretion of titanium dioxide and zinc oxide nanoparticles after repeated oral administration. Part Fibre Toxicol. 10 (9). Clement, L., Hurel, C., Marmier, N., 2013. Toxicity of TiO2 nanoparticles to cladocerans, algae, rotifers and plants–effects of size and crystalline structure. Chemosphere 90 (3), 1083–1090. Creutzenberg, O., 2013. Toxic effects of various modifications of a nanoparticle following inhalation. (1. eds), 404. Project number: F 2246. Dalai, S., Pakrashi, S., Joyce Nirmala, M., Chaudhri, A., Chandrasekaran, N., Mandal, A.B., Mukherjee, A., 2013. Cytotoxicity of TiO2 nanoparticles and their detoxification in a freshwater system. Aquat. Toxicol. 138, 1–11. Dalai, S., Pakrashi, S., Kumar, R.S., Chandrasekaran, N., Mukherjee, A., 2012. A comparative cytotoxicity study of TiO2 nanoparticles under light and dark conditions at low exposure concentrations. Toxicol. Res. 1 (2), 116–130. Dalai, S., Pakrashi, S., Bhuvaneshwari, M., Iswarya, V., Chandrasekaran, N., Mukherjee, A., 2014. Toxic effect of Cr(VI) in presence of n-TiO2 and n-Al2 O3 particles towards freshwater microalgae. Aquat. Toxicol. 146, 28–37. Dick, C.A., Brown, D.M., Donaldson, K., Stone, V., 2003. The role of free radicals in the toxic and inflammatory effects of four different ultrafine particle types. Inhal. Toxicol. 15, 39–52. Ekstrand-Hammarström, B., Akfur, C.M., Andersson, P.O., Lejon, C., Österlund, L., Bucht, A., 2012. Human primary bronchial epithelial cells respond differently to titanium dioxide nanoparticles than the lung epithelial cell lines A549 and BEAS-2B. Nanotoxicology 6 (6), 623–634. Fargasova, A., 2001. Interactive effect of manganese, molybdenum, nickel, copper I and II: and vanadium on the freshwater alga Scenedesmus quadricauda. Bull. Environ. Contam. Toxicol. 67, 688–695. Ferguson, M.A., Hoffmann, M.R., Hering, J.G., 2005. TiO2 -photocatalyzed As(III) oxidation in aqueous suspensions: reaction kinetics and effects of adsorption. Environ. Sci. Technol. 39 (6), 1880–1886. Furman, O., Usenko, S., Lau, B.L.T., 2013. Relative importance of the humic and fulvic fractions of natural organic matter in the aggregation and deposition of silver nanoparticles. Environ. Sci. Technol. 47, 1349–1356. Ghosh, S., Mashayekhi, H., Pan, B., Bhowmik, P., Xing, B., 2008. Colloidal behavior of aluminum oxide nanoparticles as affected by pH and organic matter. Langmuir 24, 12385–12391. Gojova, A., Guo, B., Kota, R.S., Rutledge, J.C., Kennedy, I.M., Barakat, A.I., 2007. Induction of inflammation in vascular endothelial cells by metal oxide nanoparticles: effect of particle composition. Environ. Health Perspect. 115, 403–409. Gottschalk, F., Sonderer, T., Scholz, R.W., Nowack, B., 2009. Modeled environmental concentrations of engineered nanomaterials (TiO2 ZnO, Ag, CNT, fullerenes) for different regions. Environ. Sci. Technol. 43, 9216–9222. Gregory, M., Chen, Q., Kleinhammes, A., Wu, Y., 2008. The structure of multilayered titania nanotubes based on delaminated anatase. Chem. Phys. Lett. 460, 517–520. Gurr, J.R., Wang, A.S.S., Chen, K.C.H., Jan, Y., 2005. Ultrafine titanium dioxide particles in the absence of photoactivation can induce oxidative damage to human bronchial epithelial cells. Toxicology. 213, 66–73. Hartmann, N.B., Von der Kammer, F., Hofmann, T., Baalousha, M., Ottofuelling, S., Baun, A., 2010. Algal testing of titanium dioxide NPs-testing considerations: inhibitory effects and modification of cadmium bioavailability. Toxicology. 269, 190–197. Hirakawa, K., Mori, M., Yoshida, M., Oikawa, S., Kawanishi, S., 2004. Photo-irradiated titanium dioxide catalyzes site specific DNA damage via generation of hydrogen peroxide. Free Radical Res. 38 (5), 439–447. Holzinger, A., Lütz, C., 2006. Algae and UV irradiation: effects on ultrastructure and related metabolic functions. Micron 37, 190–207. Hund-Rinke, K., Simon, M., 2006. Ecotoxic effect of photocatalytic active nanoparticles (TiO2 ) on algae and daphnids. Environ. Sci. Pollut. Res. 3, 225–232. Hyung, H., Fortner, J.D., Hughes, J.B., Kim, J.H., 2007. Natural organic matter stabilizes carbon nanotubes in aqueous phase. Environ. Sci. Technol. 41, 179–184. ICIS, 2010. ICIS Chemical Report. ICIS. Jak, R.G., Maas, J.L., Scholten Th, M.C., 1996. Evaluation of laboratory derived toxic effect concentrations of a mixture of metals by testing fresh water plankton communities in exposures. Water Res. 30, 1215–1227. Ji, J., Long, Z., Lina, D., 2011. Toxicity of oxide nanoparticles to the green algae Chlorella sp. Chem. Eng. J. 170, 525–530. Jin, C., Tang, Y., Yang, F.G., Li, X.L., Xu, S., Fan, X.Y., Huang, Y.Y., Yang, Y.J., 2011. Cellular toxicity of TiO2 nanoparticles in anatase and rutile crystal phase. Biol. Trace Elem. Res. 141, 3–15. Kaegi, R., Ulrich, A., Sinnet, B., Vonbank, R., Wichser, A., Zuleeg, S., Simmler, H., Brunner, S., Vonmont, H., Burkhardt, M., Boller, M., 2008. Synthetic TiO2 nanoparticle emission from exterior facades into the aquatic environment. Environ. Pollut. 156, 233–239. Kang, N.K., Lee, B., Choi, G.G., Moon, M., Park, M.S., Lim, J., Yang, J.W., 2014. Enhancing lipid productivity of Chlorella vulgaris using oxidative stress by TiO2 nanoparticles. Kor. J. Chem. Eng. 31 (5), 861–867. Karakoti, A.S., Hench, L.L., Seal, S., 2006. The potential toxicity of nanomaterials—the role of surfaces. JOM 58, 77–82. Keller, A.A., Wang, H., Zhou, D., Lenihan, H.S., Cherr, G., Cardinal, B.J., Miller, R., Ji, Z., 2010. Stability and aggregation of metal oxide nanoparticles in natural aqueous matrices. Environ. Sci. Technol. 44, 1962–1967. Lee, W.M., An, Y.J., 2013. Effects of zinc oxide and titanium dioxide nanoparticles on green algae under visible, UVA, and UVB irradiations: no evidence of enhanced algal toxicity under UV pre-irradiation. Chemosphere 91 (4), 536–544. Lubick, N., 2008. Nanosilver toxicity: ions, nanoparticles- or both? Environ. Sci. Technol. 42 (23), 8617–8618. Malhi, G.S., 2012. The Chronic Toxicity of Titanium Dioxide Nanoparticles to the Freshwater Amphipod Hyalella azteca. Theses and Dissertations (Comprehensive). Paper 839. Marcone, G.P., Oliveira, Á.C., Almeida, G., Umbuzeiro, G.A., Jardim, W.F., 2012. Ecotoxicity of TiO2 to Daphnia similis under irradiation. J. Hazard. Mater. 211, 436–442. McGivney, E., 2007. Comparison of UV-C and Vacuum-UV induced advanced oxidation technology on the acture mortality of microalgae. TRITA-LWR Degree Project 13:24 p. 31. 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. Mittler, R., 2002. Oxidative stress, antioxidants and stress tolerance. Trends Plant Sci. 7 (9), 405–410. Mueller, N.C., Nowack, B., 2008. Exposure modeling of engineered nanoparticles in the environment. Environ. Sci. Technol. 42, 4447–4453. Navarro, E., Piccapietra, F., Wagner, B., Marconi, F., Kaegi, R., Odzak, N., Sigg, L., Behra, R., 2008. Toxicity of silver nanoparticles to Chlamydomonas reinhardtii. Environ. Sci. Technol. 42 (23), 8959–8964. Nel, A., Xia, T., Madler, L., Li, N., 2006. Toxic potential of materials at the nanolevel. Science 311 (5761), 622–627. Nur, Y., Lead, J.R., Baalousha, M., 2014. Evaluation of charge and agglomeration behavior of TiO2 nanoparticles in ecotoxicological media. The Science of the Total Environment http://dx.doi.org/10.1016/j.scitotenv.2014.11.057 Olmedo, D.G., Tasat, D.R., Guglielmotti, M.B., Cabrini, R.L., 2005. Effects of titanium dioxide on the oxidative metabolism of alveolar macrophages: an experimental study in rats. J. Biomed. Mater. Res. A 73, 142–149. Organisation for Economic Cooperation and Development (OECD), 2011, Freshwater Alga and Cyanobacteria, Growth Inhibition Test. OECD Guidelines for the Testing of Chemicals, Test No. 201. OECD Publishing, Paris, France, 10.1787/9789264069923-en. Pakrashi, S., Dalai, S., Prathna, T.C., Trivedi, S., Myneni, R., Raichur, A.M., Chandrasekaran, N., Mukherjee, A., 2013. Cytotoxicity of aluminium oxide nanoparticles towards fresh water algal isolate at low exposure concentrations. Aquat. Toxicol. 132, 34–45. Pakrashi, S., Dalai, S., Sabat, D., Singh, S., Chandrasekaran, N., Mukherjee, A., 2011. Cytotoxicity of Al2 O3 nanoparticles at low exposure levels to a freshwater bacterial isolate. Chem. Res. Toxicol. 24 (11), 1899–1904. Panessa-Warren, B.J., Maye, M.M., Warren, J.B., Crosson, K.M., 2009. Single walled carbon nanotube reactivity and cytotoxicity following extended aqueous exposure. Environ. Pollut. 157 (4), 1140–1151. Pawley, J.B., 2006. Handbook of Biological Confocal Microscopy. Springerlink Publishers, Midtown Manhattan, New York City USA, pp. 20–42. Reyes-Coronado, D., Rodriguez-Gattorno, G., Espinosa-Pesqueira, M.E., Cab, C., De Coss, R., Oskam, G., 2008. Phase-pure TiO2 nanoparticles: anatase, brookite and rutile. Nanotechnology 19 (14), 145605. Sadiq, I.M., Pakrashi, S., Chandrasekaran, N., Mukherjee, A., 2011. Studies on toxicity of aluminum oxide (Al2 O3 ) nanoparticles to microalgae species: Scenedesmus sp and Chlorella sp. J. Nanopart. Res. 13, 3287–3299. Sayes, C.M., Wahi, R., Kurian, P.A., Liu, Y., West, J.L., Ausman, K.D., Warheit, D.B., Colvin, V.L., 2006. Correlating nanoscale titania structure with toxicity: a cytotoxicity and inflammatory response study with human dermal fibroblasts and human lung epithelial cells. Toxicol. Sci. 92 (1), 174–185. Scanlon, D.O., Dunnill, C.W., Buckeridge, J., Shevlin, S.A., Logsdail, A.J., Woodley, S.M., Catlow, C.R.A., Powell, M.J., Palgrave, R.G., Parkin, I.P., Watson, G.W., Keal, T.W., Sherwood, P., Walsh, A., Sokol, A.A., 2013. Band alignment of rutile and anatase TiO2 . Nat. Mater. 12, 798–801. Sevcu, A., El-Temsah, Y.S., Joner, E.J., Cernik, M., 2011. Oxidative stress induced in microorganisms by zero-valent iron nanoparticles. Microbes Environ. 26, 271–281. Sharma, V.K., 2009. Aggregation and toxicity of titanium dioxide nanoparticles in aquatic environment—a review. J. Environ. Sci. Health A Tox. Hazard. Subst. Environ. Eng. 44, 1485–1495. Small, C., Chen, S., Subbiah, J., Amb, C., Tsang, S., Lai, S., Reynolds, J., So, F., 2012. High-efficiency inverted dithienogermole-thienopyrrolodione-based polymer solar cells. Nat. Photonics 6, 115–120. V. Iswarya et al. / Aquatic Toxicology 161 (2015) 154–169 Sun, B., Smirniotis, P.G., 2003. Interaction of anatase and rutile TiO2 particles in aqueous photooxidation. Catal Today 88, 49–59. Suzuki, R., Ishimaru, T., 1990. An improved method for the determination of phytoplankton chlorophyll using N,N-dimethylformamide. J. Oceanogr. Soc. Jpn. 46, 190–194. Teisseire, H., Couderchet, M., Vernet, G., 1999. Phytotoxicity of diuron alone and in combination with copper or folpet on duckweed (Lemna minor). Environ. Pollut. 106, 39–45. Termtanun, M., 2013. Photocatalytic degradation of pesticides using TiO2 nanoparticles. In: Doctoral Dissertation. University of Nottingham, Available online at: http://eprints.nottingham.ac.uk /13827/2/Thesis MUTSEE2 pdf. Tucci, P., Porta, G., Agostini, M., Dinsdale, D., Iavicoli, I., Cain, K., Finazzi-Agró, A., Melino, G., Willis, A., 2013. Metabolic effects of TiO2 nanoparticles, a common component of sunscreens and cosmetics, on human keratinocytes. Cell Death Dis. 4, 549. Utgikar, V.P., Chaudhary, N., Koeniger, A., Tabak, H.H., Haines, J.R., Govind, R., 2004. Toxicity of metals and metal mixtures: analysis of concentration and time dependence for zinc and copper. Water Res. 38, 3651–3658. Van Breusegem, F., Dat, J.F., 2006. Reactive oxygen species in plant cell death. Plant Physiol. 141, 384–390. Varela-Valencia, R., Gómez-Ortiz, N., Oskam, G., De Coss, R., Rubio-Piña, J., Del Río-García, M., Albores-Medina, A., Zapata-Perez, O., 2014. The effect of titanium dioxide nanoparticles on antioxidant gene expression in tilapia (Oreochromis niloticus). J. Nanopart. Res. 16 (4), 1–12. Vileno, B., Lekka, M., Sienkiewicz, A., Jeney, S., Stoessel, G., Lekki, J., Forró, L., Stachura, Z., 2007. Stiffness alterations of single cells induced by UV in the presence of nanoTiO2 . Environ. Sci. Technol. 41, 5149–5153. 169 Wang, B.Q., Jing, L.Q., Qu, Y.C., Li, S.D., Jiang, B.J., Yang, L.B., Xin, B.F., Fu, H.G., 2006. Enhancement of the photocatalytic activity of TiO2 nanoparticles by surface-capping DBSgroups. Appl. Surf. Sci. 252, 2817–2825. Wang, H., Joseph, J.A., 1999. Quantifying cellular oxidative stress by Dichlorofluorescein assay using microplate reader. Free Radical Bio. Med. 27, 612–616. Wang, J., Zhang, X., Chen, Y., Sommerfeld, M., Hu, Q., 2008. Toxicity assessment of manufactured nanomaterials using the unicellular green alga Chlamydomonas reinhardtii. Chemosphere 73 (7), 1121–1128. Winkler, J., 2003. Production of Titanium Dioxide Pigments, European Coatings Literature. Vincentz, pp. 37–40. Xia, T., Kovochich, M., Brant, J., Hotze, M., Sempf, J., Oberley, T., Sioutas, C., Yeh, J.I., Wiesner, M.R., Nel, A.E., 2006. Comparison of the abilities of ambient and manufactured nanoparticles to induce cellular toxicity according to an oxidative stress paradigm. Nano Lett. 6 (8), 1794–1807. Yeo, M.K., Kang, M., 2006. Photodecomposition of bisphenol A on nanometer-sized TiO2 thin film and the associated biological toxicity to zebrafish (Danio rerio) during and after photocatalysis. Water Res. 40, 1906–1914. Yu, H., Pan, J., Bai, Y., Zong, X., Li, X., Wang, L., 2013. Hydrothermal synthesis of a crystalline rutile TiO2 nanorod based network for efficient dye-sensitized solar cells. Chemistry 19 (40), 13569–13574. Zou, X., Shi, J., Zhang, H., 2014. Coexistence of Silver and Titanium Dioxide Nanoparticles: Enhancing or Reducing Environmental Risks? Aquat. Toxicol 154, 168–175.