Metabolomics analysis of TiO2 nanoparticles induced toxicological effects on rice (Oryza sativa L.)

Environmental Pollution 230 (2017) 302e310

Contents lists available at ScienceDirect

Environmental Pollution journal homepage: www.elsevier.com/locate/envpol

Metabolomics analysis of TiO2 nanoparticles induced toxicological effects on rice (Oryza sativa L.)* Biying Wu a, b, Lizhong Zhu a, b, *, X. Chris Le c a

Department of Environmental Science, Zhejiang University, Hangzhou, Zhejiang, 310058, China Zhejiang Provincial Key Laboratory of Organic Pollution Process and Control, Hangzhou, Zhejiang, 310058, China c Department of Laboratory Medicine and Pathology, University of Alberta, Edmonton, Alberta, T6G 2G3, Canada b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 9 February 2017 Received in revised form 7 June 2017 Accepted 19 June 2017

The wide occurrence and high environmental concentration of titanium dioxide nanoparticles (nanoTiO2) have raised concerns about their potential toxic effects on crops. In this study, we employed a GCMS-based metabolomic approach to investigate the potential toxicity of nano-TiO2 on hydroponicallycultured rice (Oryza sativa L.) after exposed to 0, 100, 250 or 500 mg/L of nano-TiO2 for fourteen days. Results showed that the biomass of rice was significantly decreased and the antioxidant defense system was significantly disturbed after exposure to nano-TiO2. One hundred and five identified metabolites showed significant difference compared to the control, among which the concentrations of glucose-6phosphate, glucose-1-phosphate, succinic and isocitric acid were increased most, while the concentrations of sucrose, isomaltulose, and glyoxylic acid were decreased most. Basic energy-generating ways including tricarboxylic acid cycle and the pentose phosphate pathway, were elevated significantly while the carbohydrate synthesis metabolism including starch and sucrose metabolism, and glyoxylate and dicarboxylate metabolism were inhibited. However, the biosynthetic formation of most of the identified fatty acids, amino acids and secondary metabolites which correlated to crop quality, were increased. The results suggest that the metabolism of rice plants is distinctly disturbed after exposure to nano-TiO2, and nano-TiO2 would have a mixed effect on the yield and quality of rice. © 2017 Elsevier Ltd. All rights reserved.

Keywords: Nanoparticles Titanium dioxide Metabolic profiling Rice

1. Introduction Titanium dioxide nanoparticle (nano-TiO2) is one of the most abundant engineering nanomaterials produced in the world and is widely used in cosmetics, drugs, dyes, and many other consumer products (Gottschalk et al., 2009; Sun et al., 2014). Annual global production of nano-TiO2 is more than 10,000 metric tons (U.S. EPA, 2009). Sun et al. (2014) have reported that the annual increase of nano-TiO2 in the environment is 21 ng/L in surface waters and 89 mg/kg in sludge-treated soil; the annual input concentration in the sediment can reach 1.9 mg/L in the U.S., Europe and Switzerland. Nano-TiO2 concentration in the environmental media, including surface water, sediment, natural and urban soils, was estimated to be in the range of mg/kg to low mg/kg, which is far

*

This paper has been recommended for acceptance by Baoshan Xing. * Corresponding author. Department of Environmental Science, Zhejiang University, Hangzhou, Zhejiang, 310058, China. E-mail address: [email protected] (L. Zhu). http://dx.doi.org/10.1016/j.envpol.2017.06.062 0269-7491/© 2017 Elsevier Ltd. All rights reserved.

higher than that of other nanoparticles (Gottschalk et al., 2009; Sun et al., 2014). The earliest report on the potential toxicity of nanoTiO2 was in 1985, where tumors were found in the lung of rats after a lifetime exposure to nano-TiO2 at 250 mg/L (Lee et al., 1985). In recent years, nano-TiO2 has drawn great public concerns because of its potential toxic effects on organisms and ecosystems (Godwin et al., 2009; Hext et al., 2005; Klaine et al., 2008). For many metal nanoparticles, the dissolved metal ions (Lin and Xing, 2008; Thuesombat et al., 2014) and oxidative stress induced by reactive oxygen species (ROS) (Burello and Worth, 2011; Xia et al., 2006) are usually considered to be critical factors in causing toxic effects. However, because of the low solubility of nano-TiO2, the specific photoelectric properties of nano-TiO2 that induced the production of ROS were seen as a major factor resulting in cell
ska et al., 2016; damage (Lee et al., 1985; Shukla et al., 2011; Szyman Xia et al., 2015). Li et al. (2012) found that nano-TiO2 could induce three types of ROS: superoxide radicals (O2
e), hydroxyl radicals (
OH), and singlet oxygen species(1O2) under UV irradiation. It was observed that nano-TiO2 generated the most
OH and 1O2 among seven experimental nanoparticles including nano-TiO2, nano-ZnO,

B. Wu et al. / Environmental Pollution 230 (2017) 302e310

nano-Al2O3, nano-SiO2, nano-Fe2O3, nano-CeO2, and nano-CuO. Moreover, in all seven nanoparticles, the total concentration of all three types of ROS generated by nano-TiO2 was the highest. Studies on the phytotoxicity of nano-TiO2 found that nano-TiO2 could be transported and accumulated in plant tissue (Jacob et al., 2013; Servin et al., 2012, 2013), which affected the physiological metabolism in plants (Gardea-Torresdey et al., 2014; Ghosh et al., 2010; Miralles et al., 2012; Song et al., 2013a, 2013b; Xia et al., 2015; Zheng et al., 2007). Servin et al. (2012, 2013) observed that nano-TiO2 was absorbed and transported from root to leaf trichomes in cucumbers with no biotransformation, indicating that nano-TiO2 was transportable in plants. Additionally, the ROS induced by nano-TiO2 can break the cell membrane of plant, decrease the biomass, affect the enzymatic activities of the antioxidant defense system, decrease photosynthesis, and damage DNA (Gardea-Torresdey et al., 2014; Ghosh et al., 2010; Miralles et al., 2012; Song et al., 2013a, 2013b; Xia et al., 2015; Zheng et al., 2007). It was found that nano-TiO2 induced DNA damage, inhibited the growth and increased lipid peroxidation in Allium cepa root at the concentration of 319 mg/L and induced DNA damage in Nicotiana tabacum leaf at 157 mg/L (Ghosh et al., 2010). However, studies also explored the advantages or non-toxic of nano-TiO2 treatment on plants (Song et al., 2013b; Su et al., 2009; Yang et al., 2007; Zheng et al., 2007). Nanoanatase TiO2 was shown to promote photosynthesis, improve plant growth, promote the vigor of aged seeds, and induce chlorophyll biosynthesis in spinach (Yang et al., 2007). Song et al. found that nano-TiO2 was not toxic to the three plants species (oilseed rape, lettuce and kidney bean) in their experiment (Song et al., 2013b). Nevertheless, few studies focused on the mechanism of phytotoxicity of nano-TiO2 at the metabolite level. Metabolomics provide valuable information regarding the overall changes of small metabolites and biochemical pathways that might be altered in response to toxicants or environmental stressors (Zhang et al., 2013). It has also been shown as an effective tool to help understand the changes in chemical composition of crops with subsequent multivariate analysis (Hirai et al., 2004; Jonsson et al., 2005; Schauer and Fernie, 2006). GC-MS has been a long-standing approach used for metabolite profiling, offering a good balance between sensitivity and reliability to separate and identify the small molecular intracellular metabolites, and their intermediates in plant tissue (Lisec et al., 2006). Recently, metabolic responses to nano-TiO2 were studied in mouse fibroblast cells (Bo et al., 2014; Liu et al., 2012), HpeG2 cells (Kitchin et al., 2014), human gingivitis cells (Garcia-Contreras et al., 2015), C. elegans (Ratnasekhar et al., 2015), and earthworms (Whitfield Åslund et al., 2012). There are researches that used metabolomics to study the phytotoxicity of C60, nano-Cu, nano-Cu(OH)2, and nano-CdO (Du et al., 2016; Kim et al., 2016; Ve cerov
a et al., 2016; Zhao et al., 2016a, 2016b). In these researches, the metabolic changes on the root exudate, roots and leaves were indicated, which revealed important detoxification mechanisms defense the nanoparticles stress. However, there is still scant research that uses metabolomics to estimate the effects of nano-TiO2 on the crops. Serving as the initial step of the food chain, crops provide a major route of pollutant exposure to higher species, including humans (GardeaTorresdey et al., 2014; Miralles et al., 2012; Rico et al., 2011). Furthermore, these harmful effects can affect the yield and quality of crops. Previous researches on wheat, tomato and cucumber reported that nano-TiO2 could be detected in crop tissues and fruits, which reduced the biomass, and modified the nutritive element content and macromolecular composition (Du et al., 2011; Servin et al., 2013; Song et al., 2013a). Kim et al. (2016) reported that nano-TiO2 showed greater movement in the sediment than in the water in a paddy microcosm. Considering that rice (Oryza sativa L.)

303

is a staple food consumed by more than half of the world's population (Rico et al., 2013a), it is essential to understand the interaction of rice plant with nano-TiO2 to provide information on the metabolic influence of engineered nanoparticles (ENPs) on it. Therefore, the primary aim of this work is to investigate the metabolic response of rice to nano-TiO2 exposure. We measured the biomass and activities of the antioxidant defense system of roots and leaves of rice plants. Using GC-MS-based metabolomics, we evaluated the up- or down-regulation of small molecule metabolites and analyzed the connection among the metabolites using principal component analysis (PCA) and partial least-squares discriminant analysis (PLS-DA). Significantly affected metabolic pathways were also identified. This research contributes to elucidate the potential influence of nanoparticles on the yield and quality of crops. 2. Materials and methods 2.1. Preparation of TiO2 suspensions The hydrophilic nanosized TiO2 material (99.8% metals basis, anatase) used in this study was purchased from the Aladdin® Chemical Reagent Co., Ltd., China. The nano-TiO2 has a mean transmission electron microscopy (TEM) imaged size of 20 nm and particle size of 293 ± 17 nm in deionized water. A nano-TiO2 250 mg/L stock solution was characterized for monomer size and morphology by TEM (HRTEM, JEM, 2010; JEOL Ltd., Japan) and for hydrodynamic diameter by dynamic light scattering (Nano-ZS, Malvern Instruments, U.K.). The TEM images of nano-TiO2 are showed in Supporting Information (SI) Fig. S1. The surface area of nano-TiO2 was 79.5 m2/g, determined by the multipoint BrunauerEmmettTeller (BET) method (Yang et al., 2006). Nano-TiO2 was suspended in Hoagland nutrient solution (Lin and Xing, 2008). The experimental concentrations used in the study are 0, 100, 250, and 500 mg/L, which are higher than those observed in the soil, thus the obtained results should be considered as mechanism research rather than a simulation of processes in natural environ
ska et al., 2016; Xia et al., 2015). All ment (Song et al., 2013b; Szyman suspensions were stirred for 5 min, and then sonicated (100 w) for 30 min in a water-bath with continuous stirring. 2.2. Seed germination and plant growth Seeds of rice (Oryza sativa L.) were obtained from Zhejiang Academy of Agricultural Sciences (Hangzhou, China). Rice seeds were sterilized for 12 h in a 3% H2O2 solution, then rinsed 3 times with Millipore water, and finally placed on sterilized perlite growth substrate in Millipore water to grow. After 7 days, similar plants were selected and transferred to 2 L glass pots containing 1.8 L nano-TiO2 suspensions. The seedlings were fixed on the perforated sheet with a sponge, and their roots were submerged into the suspensions. Every pot had four bundles of seedlings, and each bundle contained six seedlings. The seedlings were grown in a greenhouse under natural light and temperature for 14 days. The absorption and transportation of nano-TiO2 in roots, stems and leaves of the exposed rice plants were observed by TEM. 2.3. Titanium (Ti) analysis At harvest, the roots and leaves were washed with deionized water and lyophilized. Dried tissues were ground by a highthroughput tissue grinder (TP-48P, ONEBIO, Shanghai, China), and then digested with a mixture of 5 mL of hydrofluoric acid (HF) and 5 mL of HNO3 (v/v 1:1) using a graphite digestion system (ZEROM, ProD 48, Changsha, China) at 120  C for 3 h (Schimpf et al., 2002).

304

B. Wu et al. / Environmental Pollution 230 (2017) 302e310

The standard reference materials (Orange leaves, GBW10020, China) were digested and analyzed using the same method as the samples. Titanium element was analyzed using inductively coupled plasma mass spectrometry (ICP-MS) (NexION 300X PerkinElmer, USA). The recovery rate of titanium in orange leaves was in the range of 84.7 ± 5.8%, which meant the method was reliable. The recovery rate of titanium of ICP-MS method was between 90 and 99%. The dissolved titanium ion content in the hydroponic media was also detected (SI Table S2). 2.4. Antioxidative enzyme activities and MDA content The enzyme activities of catalase (CAT), peroxidase (POD), and superoxide dismutase (SOD) were measured by Zhang's method (Zhang and Kirkham, 1994). The content of malondialdehyde (MDA) was measured using Sudhakar's method (Sudhakar et al., 2001). The specific methods are described in SI Text S1. All assays were measured using a UVevis Spectrometer (UV1800, SHIMADZU, Japan) at 25  C. 2.5. Metabolites extraction and derivatization As the leaf is the major photosynthesis organ to manufacture and provide substance stored in fruit, we did metabolomics to analyze the disturbance induced by nano-TiO2 among the metabolism in rice leaves. Fresh rice leaves were immediately frozen in liquid nitrogen and freeze-dried, then grounded to powder, and stored at 80  C before analysis. Every sample contained 10.0 mg of rice leaf powder. The samples were extracted by ultrasonication with a single-phase solvent methanol:chloroform:water mixture (2 mL, 5:2:2, v/v/v), centrifuged to get supernatant, and freeze-dried. For derivatization, 60 mL methoxyamine hydrochloride in pyridine (20 mg/L) was added to the dried sample, and the mixture was incubated at 37  C for 90 min. Then 80 mL of N-methyl-N-(trimethylsilyl)trifluoroacetamide (MSTFA) was added. The samples were incubated at 37  C for 30 min to be silylated. The methods for metabolite extraction and derivatization were previously reported by Zhang et al. (2013), and the details are provided in the SI Text S2. 2.6. GC-MS conditions and data processing Metabolite profiling was performed on a GC (Agilent 7890B, Agilent) linked to a quadrupole MS (Agilent 5977A, Agilent). Derivative samples (1 mL) were injected into the GC column in a split mode (1:15). The column was a DB-5 MS capillary column (30 m, 0.25 mm i.d., 0.25 mm film thickness). Helium was maintained at a constant flow rate of 1.0 mL/min as the carrier gas. The transfer line and the ion source temperatures were 280 and 230  C respectively. The injection temperature was 290  C. The oven temperature was maintained at 70  C for 4 min, then increased at a rate of 15  C/min to 300  C, holding for 5 min. The solvent delay was set at 5 min. The electron energy of electron impact ionization (EIþ) was set as 70 eV. Mass spectra were acquired using full scan monitoring mode, and the mass scan range was from m/z 33 to 600. The metabolites were extracted through the Mass Hunter Qualitative Analysis software B.07.00 (Agilent, USA), then annotated and confirmed by comparison with the NIST 05 library. PCA and PLS-DA were performed using SIMCA-P 11.5 (Umetrics, Umeå, Sweden) after the data matrix was mean centered and Pareto (Par) scaled with the SIMCA-P 11.5. MetaboAnalyst 3.0 (http://www. metaboanalyst.ca/MetaboAnalyst/) was used to identify metabolic pathways in rice leaves with major perturbed. The pathway impact value threshold was set as 0.2. A basic network of metabolites was built through the GAM web-service (https://artyomovlab.wustl.

edu/shiny/gam/) and Cytoscape 3.4.0 (http://www.cytoscape.org/) to visualize the connection among the metabolites (Sergushichev et al., 2016). 2.7. Statistical analyses of other data All experiments were performed in at least triplicates; the metabolite assay was performed in six replicate samples. The results were presented as the mean ± SD (standard deviation). The Student's t-test was used to compare the data using SPSS Statistics 17.0 (IBM, New York), and statistical significance was accepted at P < 0.05. The heat map was drawn with R software (https://www.rproject.org/). 3. Results 3.1. Biomass and titanium accumulation of rice tissues After 14 days of exposure to nano-TiO2, the rice seedlings exhibited a reduction in biomass compared to the control (Fig. 1a). The decreases in root biomass of rice exposed to 100, 250 and 500 mg/L of nano-TiO2 were 23.3%, 44.4%, and 42.4%, respectively, being significantly smaller than control. Nano-TiO2 also exhibited a negative effect on the biomass of leaves. The biomass of seedling leaves was significantly reduced by 26.0% and 30.1%, when exposed to nano-TiO2 concentrations of 250 and 500 mg/L, respectively. The accumulation of titanium in rice roots and leaves varied greatly with the treated concentrations of nano-TiO2 (Fig. 1b). The titanium contents in rice exposed to 100, 250, and 500 mg/L nanoTiO2 were much higher than the control in both roots and leaves. The mean concentration of titanium in roots increased by 782%, 964%, and 1519%, and in leaves increased by 30.9%, 140%, and 89.8%, compared with that in control. The contents of titanium in roots were much higher than leaves, which was also found in other re
et al., 2017; Larue et al., 2016). The reason searches (Korenkova could be that the part of nano-TiO2 absorbed to the epidermal cells is persistent and remains stuck to the root surface after rinse (Larue et al., 2016; Tan et al., 2017). Unlike roots, where titanium content increased with the treated concentration of nano-TiO2, leaves showed lower titanium content in 500 mg/L nano-TiO2 than in 250 mg/L nano-TiO2. As the nano-TiO2 is hardly dissolved in culture (Table S2), the increased content of titanium in leaves suggested that the nano-TiO2 was absorbed in rice. The TEM images showed that there were nanoparticles gathered on the root surface, and subsequently absorbed and accumulated by root cells. A small amount of the particles was transported through stems (Fig. S2). 3.2. Antioxidative enzyme activities and MDA content As shown in Fig. 2, all nano-TiO2 treatments presented an inhibition of POD activity in both roots and leaves, in which the degree of severity increased with increasing concentrations of nanoTiO2. The activities of CAT in leaves and SOD in both roots and leaves presented an earlier increase and later decrease trend, while there were no significant differences in the CAT activities in roots. The MDA content fluctuated around the content of control group, and only the decrease of 47.6% on the roots exposed to 500 mg/L nano-TiO2 was statistically significant (P < 0.05). 3.3. Effect of nano-TiO2 on metabolic profile of rice leaves In the present study, the GC-MS analysis yielded 127 peaks and was processed through Mass Hunter software, and 105 of them varied significantly between experimental and control groups. All detected metabolites are presented in the Supporting Information

2.1

root leaf

a

a

A a

b

B

1.4

C

C

b

0.7

0.0

0

100

250

500

nano-TiO2 concentration (mg/L)

Ti concentration (mg/Kg dried tissue)

Biomass (g/six seedlings)

B. Wu et al. / Environmental Pollution 230 (2017) 302e310

2500

305

root leaf

2000

D C

1500

B

1000

100

A

50 0

b

a 0

b 100

c

250

bb 500

nano-TiO2 concentration (mg/L)

Fig. 1. Biomass of and Ti content in rice roots and leaves after 14-day exposure to different concentrations of TiO2. (a) Biomass. (b) Ti content. Values are means ± SE (n ¼ 4). Means with the same letter are not significantly different (p < 0.05).

(Table S1). Volcano plots are provided to visualize changes in metabolites as compared to the control (Fig. S3). The results showed that the concentrations of sugars and organic acids had significantly changed in the rice of the groups exposed to nano-TiO2. The concentrations of glucose-6-phosphate, glucose-1-phosephate, succinic acid and isocitric acid were increased the most in all experimental groups. In the group exposed to 500 mg/L nano-TiO2, the concentrations of these metabolites had increased by 311%, 275%, 358% and 275% of that in control group, respectively. The sucrose, isomaltulose, and glyoxylic acid were the most inhibited metabolites, and in the group exposed to 500 mg/L nano-TiO2, their concentrations were only 70.1%, 82.7%, and 40.7% of that in control, respectively. Contents of most

identified amino acids and fatty acids, which transforms into protein and plant oil, were increased in all exposure groups. The increase of isoleucine and octadecanoic acid content in the 500 mg/L group were the most remarkable as they increased to 260% and 163% compared to control, respectively. The contents of most secondary metabolites were also increased. In the group with the treatment of 500 mg/L nano-TiO2, the vitamin C, E, and B5 were increased to 220%, 180%, and 266%, respectively, compared to control. All independent samples were regrouped using cluster analysis based on the similarity of the metabolite abundance profiles with the PCA model. The results showed that four principle components cumulatively accounted for 78.1% of the total variance with the first

Fig. 2. Parameters of the antioxidant defense system in rice roots and leaves after 14-day exposure to different concentrations of TiO2. (a) Malonaldehyde (MDA) contents. (b) Superoxide dismutase (SOD) activity. (c) Peroxidase (POD) activity. (d) Catalase (CAT) activity. One unit of SOD activity is the amount of enzyme which caused a 50% decrease in the rate of NBT reduction. One unit of POD activity was defined as an absorbance change of 0.01 units per min at 470 nm. One unit of CAT is the amount of enzyme necessary to decompose 1 mmol of H2O2 per minute. Values are means ± SE (n ¼ 3). Means with the same letter are not significantly different (p < 0.05).

306

B. Wu et al. / Environmental Pollution 230 (2017) 302e310

principle components (t1) explained 45.6% of the total variance and t2 explained 12.7%. The cumulative values of PCA model with R2Xcum ¼ 0.781, R2Ycum ¼ 0.979, Q2Ycum ¼ 0.921 showed that the fitting results were acceptable (Fig. 3a). As shown in Fig. 3a, clear separation of four groups represented the significant different of the metabolic flux between different treatments. Using parameters of variable importance in projection (VIP) score, 58 metabolites were identified to be responsible for the separation for their VIP score >1 (Ratnasekhar et al., 2015; Zhang et al., 2013). The top 16 metabolites including sugars, organic acids, and aldose phosphates were shown as putative marker metabolites in the Boxwhisker Plot (Fig. 3b). They are also presented on the correlation loading plot to describe the relationship between the change of these metabolites and different concentrations of nano-TiO2 (Fig. 3c). The relationships including all identified metabolites are presented in Fig. S5. In Fig. 3c, each black point represents a single metabolite, while the red points represent the responses of groups which mean the overall change of the metabolites in each group. The distance between metabolites and responses represents the correlation that

metabolites situated near responses are positively correlated with them, and those situated opposite are negatively correlated with the responses. The distance between metabolites represents their relation to each other. Among all identified metabolites, most sugars and organic acids were situated close together, and the concentrations of them increased with the increase of nano-TiO2 concentration (Fig. S4). However, in Fig. S4, metabolites such as glyoxylic acid, d-mannose-6-phosphate, sucrose, isomaltulose, and maltose, etc., which situated opposite to them, were positively correlated with responses to low concentrations of nano-TiO2 and the control, indicating that they were down-regulated with the increase of nano-TiO2 concentration. The relevant pathways involved in rice against nano-TiO2 exposure are shown in Fig. 4, including fatty acid, sugar, amino acid and organic acid metabolism. The relationships among the important metabolic pathways were shown in Fig. 5. The metabolite network contains some critical metabolites in the metabolic pathways in rice in the 500 mg/L nano-TiO2 group (Fig. 6). Red points indicate promotion while green indicates inhibition. The contents of these metabolites shown in the network are presented in

Fig. 3. PCA and PLS-DA results obtained from GC-MS data of metabolites extracted from rice leaves exposed to 100, 250 and 500 mg/L nano-TiO2 and control. (a) Scores plot (PC1 vs PC2) of principal component analysis (PCA). (b) The box plot of top 16 significantly different metabolites among the four groups of rice through the PLS-DA analysis. (c) Loading plot of PLS-DA results, demonstrating discrimination of 16 key metabolites. Every black point represents a kind of metabolite. Different colors are used to distinguish the metabolites situated between two red arrows. Red points represent the response which means the overall change of the metabolites in a group. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

B. Wu et al. / Environmental Pollution 230 (2017) 302e310

307

Fig. 4. Summary of the pathway analysis of rice exposed to nano-TiO2 with the MetaboAnalyst 3.0. Every point represents a metabolic pathway colored according to the log of Pvalue. The redder color represents more significant effect, and the size is determined by the degree of influence. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 5. Metabolic pathways that were significantly changed under the exposure to nano-TiO2. The central arrows represent energy metabolism, including the four rectangles, glycolysis and pentose phosphate pathway, citrate cycle, and glyoxylate and dicarboxylate metabolism. The four ellipses represent the significantly affected biosynthesis metabolisms based on the energy metabolism, including galactose metabolism, starch and sucrose metabolism, amino acid metabolism, and fatty acid metabolism. The red represents promotion and the green represents inhibition. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Table S3. The relative abundances of fluctuated metabolites in all experimental groups are presented on the heat map (Fig. S5). 4. Discussion The biomass of both roots and leaves of rice were significantly decreased under the exposure of nano-TiO2. Exposed rice absorbed

titanium from roots and transported to stems and leaves (Fig. 1b). Similarly, it was previously reported that nano-TiO2 exposure
ska reduced the root biomass of Arabidopsis thaliana plants (Szyman et al., 2016). It was also found that nano-TiO2 affected the antioxidant response in Arabidopsis thaliana plants in that research. Previous studies have reported that nano-TiO2 has a great capacity to generate ROS in living cell (Shukla et al., 2011; Li et al., 2012). This is

308

B. Wu et al. / Environmental Pollution 230 (2017) 302e310

Fig. 6. The metabolic network that generated from GC-MS data of rice exposed to 500 mg/L nano-TiO2 by GAM online (https://artyomovlab.wustl.edu/shiny/gam/) and modified using Cytoscape 3.4.0. Each point represents a metabolite. The size was scaled by the log of P-value. The FC represents the relative contents of metabolites. The points were colored according to the relative quantity (logFC 2 ) of control. The red means up-regulation and the green means down-regulation. The line represents the connection between two metabolites, and the arrows present the transformation direction. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

consistent with our results, where the results of the increase of both salicylic acid and antioxidants suggest there was ROS-induced stress in rice (Table 1). Subsequently, there would be a promotion in enzymatic and non-enzymatic antioxidant defense systems which work in concert to control the uncontrolled oxidation and protect plant cells from oxidative damage by scavenging of ROS (Gill and Tuteja, 2010). In this study, the activities of CAT and SOD exhibited an upward trend initially which then descended. This indicates that low concentrations of nano-TiO2 have a stimulatory effect on antioxidant enzymes which is consistent with similar results about nano-CeO2 reported by Rico et al. (2013b). However, nano-TiO2 obviously inhibited the POD activity. This may occur if the enhanced ROS level was beyond the scavenging ability of the enzymes. This downward trend was also found in marine microalgae succeeding an upward trend in a short time period with the addition of nano-TiO2 (Xia et al., 2015). Considering that the accumulation of salicylic acid improved chilling tolerance in lemon fruit but inhibited the activity of POD (Siboza et al., 2014), the decrease may also relate to the increase of salicylic acid.

Table 1 The relative quantities of nine non-enzymatic antioxidants and salicylic acid. Metabolite

100 mg/L

250 mg/L

500 mg/L

Linolenic acid Linoleic acid Glutamine Glycine Proline Ascorbic acid Tocopherol Cinnamic acid 3,7,30 ,40 -Tetrahydroxyflavone Salicylic acid

1.032 1.202* 1.054 0.9736 1.014 1.521** 1.433** 1.208* 1.068 2.005**

1.439** 1.741** 1.739** 1.831** 1.148** 1.318** 1.460* 1.143* 1.294** 1.290**

1.478** 2.155** 1.562** 1.902** 1.127** 2.202** 1.796** 1.231** 1.476** 2.971**

*Significant difference when p < 0.05. **Significant difference when p < 0.01.

The relative quantities of nine detected non-enzymatic antioxidants including unsaturated fatty acids, amino acids, and secondary metabolites increased in rice leaves under the exposure of nano-TiO2 (Table 1). The increases of glycine and glutamine may result from the promotion of the biosynthesis of glutathione (GSH), which plays a critical role in maintaining the intracellular redox state (Gill and Tuteja, 2010), as a response to oxidative stress. According to Hare and Cress (1997), the accumulation of proline in higher plants is viewed as a stress-induced response as it reduces stress-induced cellular acidification or prime oxidative respiration to provide energy for recovery. Previous studies also reported that the secondary metabolism in plants associated with the defense reaction is stimulated by abiotic stress (Bennett and Wallsgrove, 1994; Schützendübel and Polle, 2002; Zhao et al., 2016a). The content of ascorbic acid (VC), tocopherol (VE), cinnamic acid and other detected sterols and pantothenic acid (VB5) were significantly increased as antioxidant properties to suppress oxidative stress induced by nano-TiO2. Linoleic acid and linolenic acid are the most abundant fatty acids in plant membrane lipids and are major substrates for lipoxygenase to start a cascade of reactions in lipoxygenase pathway (Moore, 1989). As the capacity of nano-TiO2 to induce ROS production was mentioned before (Servin et al., 2012; Shukla et al., 2011), the increases in linoleic acid and linolenic acid in treated rice leaves suggest potential membrane lipid peroxidation (Xu et al., 2011). This may be a defense mechanism to protect the cell membrane under the abiotic stress (Allakhverdiev
pez-Pe
rez et al., 2009). However, when roots were et al., 2001; Lo exposed to 500 mg/L nano-TiO2, there was a significant decrease in MDA content compared to the control. This indicates that the collaboration of both enzymatic antioxidants and non-enzymatic antioxidants reduced the levels of reactive oxygen species and membrane damage in rice (Gill and Tuteja, 2010; Gunes et al., 2007). The activations of the antioxidant defense system were based on the changes in metabolic pathways, including energy metabolism

B. Wu et al. / Environmental Pollution 230 (2017) 302e310

and anabolism. The stress of ROS drove the elevation of energy metabolism which can lead to the reduction in biomass. As shown in Fig. 6, energy metabolisms based on sugar decomposition (glycolysis, TCA cycle, and pentose phosphate pathway) were significantly elevated when exposed of nano-TiO2. Three critical phosphohexoses, glucose-1-phosphate, glucose-6-phosphate and fructose-6-phosphate, were increased significantly, while glucose and fructose were stable and sucrose and isomaltulose were decreased. This indicates that glycolysis was up-regulated, while starch and sucrose metabolism was down-regulated. The pentose phosphate pathway, as an alternative pathway of oxidative decomposition of glucose, was also elevated. Citric acid increased by 1.3, 1.5 and 3.3 fold when exposed to 100, 250 and 500 mg/L of nano-TiO2, respectively in TCA cycle, which is the next step of glycolysis. Similar trends were found in the levels of isocitric, malic, succinic, and oxalic acid, which demonstrate that promotion occurred in whole TCA cycle. Meanwhile, the decrease of glyoxylic acid indicates the down-regulation of glyoxylate and dicarboxylate metabolism which is a pathway that synthesized carbohydrates from fatty acids. Conflicting results that the TCA cycle was inhibited were found in a study using a human gingivitis model, which may result from the different anti-adversity ability between individual plant and human cells (Garcia-Contreras et al., 2015). The elevations of glycolysis, the TCA cycle, and pentose phosphate metabolism demonstrate stimulating effects of nano-TiO2. With the promotion of glycolytic and TCA cycle pathways, the biosyntheses of numerous compounds such as secondary metabolites, amino acids, nucleic acids, and fatty acids were also elevated (Sweetlove et al., 2010). However, the decreases of sucrose and isomaltulose indicate that the energy expenditure exceeded the substance accumulation, which led to the decrease of biomass. Moreover, as most of the metabolites were increased, it is interesting to find a decrease in the content of shikimic acid, which acts as a critical molecule in the biosynthesis of numerous secondary metabolites. Salicylic and cinnamic acid levels were increased. Considering that the shikimic acid pathway is important in generating secondary metabolites, this result suggests that nano-TiO2 induced the shikimic acid routed in the shikimic acid pathway employed for the biosynthesis and accumulation of cinnamic acid and salicylic acid (Scalabrin et al., 2016). Those effects on metabolite content can affect the quality of rice. As proteins, lipids, non-structural carbohydrates, and antioxidants were belong to quality parameters of crops (Wang and Frei, 2011), the increases of amino acids, fatty acids, and secondary metabolites mean that nutritional value of rice was improved in terms of proteins, lipids, and antioxidants content. However, the amount of sugars was decreased. As the sucrose is the principal transport form of assimilates in most plants (Riesmeier et al., 1994) and the starch was the vital nutrient of rice, the down-regulation of sucrose and isomaltulose suggests that the nano-TiO2 could also have a negative effect on the yield and quality of rice. Besides, rice upon the nanoTiO2 exposure was found to have an accumulation of titanium in plant tissues in this research, which means there could be a potential risk to food safety and crop quality under long-term exposure to nano-TiO2. 5. Conclusion Under the exposure to nano-TiO2, the biomass of rice was decreased and the antioxidant defense system was elevated. The results of analysis of metabolomics data show that the metabolic flux in rice changed towards the energy metabolism and synthesis of antioxidants to combat the stress induced by nano-TiO2. The effects that contents of proteins, lipids, and antioxidants in rice increased, while the content of carbohydrate decreased may result

309

in changes in the yield and quality of rice. This finding can be used to provide information on the potential influence of nanoparticles on the yield and quality of crops grown in exposure situations. Further studies are needed to elucidate the effects on the gene expression of rice, and ultimately on grain production. Acknowledgment This work was supported by the National Natural Science Foundation of China (21520102009 and 21621005). We thank Aleksandra Popowich and Qingqing Liu of the University of Alberta for helpful editing of the manuscript. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.envpol.2017.06.062. References Allakhverdiev, S.I., Kinoshita, M., Inaba, M., Suzuki, I., Murata, N., 2001. Unsaturated fatty acids in membrane lipids protect the photosynthetic machinery against salt-induced damage in synechococcus. Plant Physiol. 125, 1842e1853. Bennett, R.N., Wallsgrove, R.M., 1994. Secondary metabolites in plant defence mechanisms. New Phytol. 127, 617e633. Bo, Y., Jin, C.Y., Liu, Y., Yu, W.J., Kang, H., 2014. Metabolomic analysis on the toxicological effects of TiO2 nanoparticles in mouse fibroblast cells: from the perspective of perturbations in amino acid metabolism. Toxicol. Mech. Method 24, 1e28. Burello, E., Worth, A.P., 2011. A theoretical framework for predicting the oxidative stress potential of oxide nanoparticles. Nanotoxicology 5, 228e235. Du, W.C., Sun, Y.Y., Ji, R., Zhu, J.G., Wu, J.C., Guo, H.Y., 2011. TiO2 and ZnO nanoparticles negatively affect wheat growth and soil enzyme activities in agricultural soil. J. Environ. Monit. 13, 822e828. Du, C.L., Zhang, B., He, Y.L., Hu, C.Y., Qin, X.N., Zhang, H., Ong, C.N., Lin, Z.F., 2016. Biological effect of aqueous C60 aggregates on Scenedesmus obliquus revealed by transcriptomics and non-targeted metabolomics. J. Hazard Mater 324, 221e229. Garcia-Contreras, R., Sugimoto, M., Umemura, N., Kaneko, M., Hatakeyama, Y., Soga, T., Tomita, M., Scougall-Vilchis, R.J., Contreras-Bulnes, R., Nakajima, H., Sakagami, H., 2015. Alteration of metabolomic profiles by titanium dioxide nanoparticles in human gingivitis model. Biomaterials 57, 33e40. Gardea-Torresdey, J.L., Rico, C.M., White, J.C., 2014. Trophic transfer, transformation, and impact of engineered nanomaterials in terrestrial environments. Environ. Sci. Technol. 48, 2526e2540. Ghosh, M., Bandyopadhyay, M., Mukherjee, A., 2010. Genotoxicity of titanium dioxide (TiO2) nanoparticles at two trophic levels: plant and human lymphocytes. Chemosphere 81, 1253e1262. Gill, S.S., Tuteja, N., 2010. Reactive oxygen species and antioxidant machinery in abiotic stress tolerance in crop plants. Plant Physiol. Bioch 48, 909e930. Godwin, H.A., Chopra, K., Bradley, K.A., Cohen, Y., Harthorn, B.H., Hoek, E.M., Holden, P., Keller, A.A., Lenihan, H.S., Nisbet, R.M., 2009. The university of California center for the environmental implications of nanotechnology. Environ. Sci. Technol. 43, 6453e6457. 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, 9216e9222. Gunes, A., Inal, A., Alpaslan, M., Eraslan, F., Bagci, E.G., Cicek, N., 2007. Salicylic acid induced changes on some physiological parameters symptomatic for oxidative stress and mineral nutrition in maize (Zea mays L.) grown under salinity. J. Plant Physiol. 164, 728e736. Hare, P.D., Cress, W.A., 1997. Metabolic implications of stress-induced proline accumulation in plants. Plant Growth Regul. 21, 79e102. Hext, P.M., Tomenson, J.A., Thompson, P., 2005. Titanium dioxide: inhalation toxicology and epidemiology. Ann. Occup. Hyg. 49, 461e472. Hirai, M.Y., Yano, M., Goodenowe, D.B., Kanaya, S., Kimura, T., Awazuhara, M., Arita, M., Fujiwara, T., Saito, K., 2004. Integration of transcriptomics and metabolomics for understanding of global responses to nutritional stresses in Arabidopsis thaliana. P. Natl. Acad. Sci. U. S. A. 101, 10205e10210. Jacob, D.L., Borchardt, J.D., Navaratnam, L., Otte, M.L., Bezbaruah, A.N., 2013. Uptake and translocation of Ti from nanoparticles in crops and wetland plants. Int. J. Phytoremediat 15, 142e153. Jonsson, P., Johansson, A.I., Gullberg, J., Trygg, J.A.J., Grung, B., Marklund, S., € stro € m, M., Antti, H., Moritz, T., 2005. High-throughput data analysis for Sjo detecting and identifying differences between samples in GC/MS-Based metabolomic analyses. Anal. Chem. 77, 5635e5642. Kim, J.I., Park, H.G., Chang, K.H., Nam, D.H., Yeo, M.K., 2016. Trophic transfer of nanoTiO2 in a paddy microcosm: a comparison of single-dose versus sequential multi-dose exposures. Environ. Pollut. 212, 316e324. Kitchin, K., Grulke, E., Robinette, B., Castellon, B., 2014. Metabolomic effects in

310

B. Wu et al. / Environmental Pollution 230 (2017) 302e310

HepG2 cells exposed to four TiO2 and two CeO2 nanomaterials. Environ.Sci. Nano 1, 466e477. Klaine, S.J., Alvarez, P.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, 1825e1851. 
, L., Sebesta, Korenkova M., Urík, M., Kolen cík, M., Kratosov
a, G., Bujdos, M., V
avra, I., Dobro cka, E., 2017. Physiological response of culture media-grown barley (Hordeum vulgare L.) to titanium oxide nanoparticles. Acta Agr. Scand. B-S p. 67, 285e291. Larue, C., Castillo-Michel, H., Stein, R.J., Fayard, B., Pouyet, E., Villanova, J., Magnin, V., Pradas del Real, A.-E., Trcera, N., Legros, S., Sorieul, S., Sarret, G., 2016. Innovative combination of spectroscopic techniques to reveal nanoparticle fate in a crop plant. Spectrochim. Acta B 119, 17e24. Lee, K.P., Trochimowicz, H.J., Reinhardt, C.F., 1985. Pulmonary response of rats exposed to titanium dioxide (TiO2) by inhalation for two years. Toxicol. Appl. Pharmacol. 79, 179e192. Li, Y., Zhang, W., Niu, J., Chen, Y., 2012. Mechanism of photogenerated reactive oxygen species and correlation with the antibacterial properties of engineered metal-oxide nanoparticles. Acs Nano 6, 5164e5173. Lin, D.H., Xing, B.S., 2008. Root uptake and phytotoxicity of ZnO nanoparticles. Environ. Sci. Technol. 42, 5580e5585. Lisec, J., Schauer, N., Kopka, J., Willmitzer, L., Fernie, A.R., 2006. Gas chromatography mass spectrometry-based metabolite profiling in plants. Nat. Protoc. 1, 387e396. Liu, Y.M., Cheng, Y., Chen, T.L., Zhang, Y.N., Wang, X.Y., Zhao, A.H., Jia, W., Bo, Y., Jin, C.Y., 2012. GC/TOFMS analysis of endogenous metabolites in mouse fibroblast cells and its application in TiO2 nanoparticle-Induced cytotoxicity study. Chromatographia 75, 1301e1310.
pez-Pe
rez, L., Martínez-Ballesta, M.D.C., Maurel, C., Carvajal, M., 2009. Changes in Lo plasma membrane lipids, aquaporins and proton pump of broccoli roots, as an adaptation mechanism to salinity. Phytochemistry 70, 492e500. Miralles, P., Church, T.L., Harris, A.T., 2012. Toxicity, Uptake, and translocation of engineered nanomaterials in vascular plants. Environ. Sci. Technol. 46, 9224e9239. Moore, T.S., 1989. Lipid metabolism in plants. Crit. Rev. Plant Sci. 8, 1e43. Ratnasekhar, C., Sonane, M., Satish, A., Mudiam, M.K.R., 2015. Metabolomics reveals the perturbations in the metabolome of Caenorhabditis elegans exposed to titanium dioxide nanoparticles. Nanotoxicology 9, 1e11. Rico, C.M., Majumdar, S., Duarte-Gardea, M., Peralta-Videa, J.R., GardeaTorresdey, J.L., 2011. Interaction of nanoparticles with edible plants and their possible implications in the food chain. J. Agric. Food Chem. 59, 3485e3498. Rico, C.M., Morales, M.I., Barrios, A.C., Mccreary, R., Hong, J., Lee, W.Y., Nunez, J., Peralta-Videa, J.R., Gardea-Torresdey, J.L., 2013a. Effect of cerium oxide nanoparticles on the quality of rice (Oryza sativa L.) grains. J. Agric. Food Chem. 61, 11278e11285. Rico, C.M., Morales, M.I., Mccreary, R., Castillo-Michel, H., Barrios, A.C., Hong, J., Tafoya, A., Lee, W.Y., Varela-Ramirez, A., Peralta-Videa, J.R., GardeaTorresdey, J.L., 2013b. Cerium oxide nanoparticles modify the antioxidative stress enzyme activities and macromolecule composition in rice seedlings. Environ. Sci. Technol. 47, 14110e14118. Riesmeier, J.W., Willmitzer, L., Frommer, W.B., 1994. Evidence for an essential role of the sucrose transporter in phloem loading and assimilate partitioning. Embo J. 13, 1e7. Scalabrin, E., Radaelli, M., Capodaglio, G., 2016. Simultaneous determination of shikimic acid, salicylic acid and jasmonic acid in wild and transgenic Nicotiana langsdorffii plants exposed to abiotic stresses. Plant Physiol. Bioch 103, 53e60. Schauer, N., Fernie, A.R., 2006. Plant metabolomics: towards biological function and mechanism. Trends Plant Sci. 11, 508e516. Schimpf, S., Lucas, M., Mohr, C., Rodemerck, U., Brückner, A., Radnik, J., Hofmeister, H., Claus, P., 2002. Supported gold nanoparticles: in-depth catalyst characterization and application in hydrogenation and oxidation reactions. Catal. Today 72, 63e78. Schützendübel, A., Polle, A., 2002. Plant responses to abiotic stresses: heavy metalinduced oxidative stress and protection by mycorrhization. J. Exp. Bot. 53, 1351e1365. Sergushichev, A.A., Loboda, A.A., Jha, A.K., Vincent, E.E., Driggers, E.M., Jones, R.G., Pearce, E.J., Artyomov, M.N., 2016. GAM: a web-service for integrated transcriptional and metabolic network analysis. Nucleic Acids Res. 44, W194eW200. Servin, A.D., Castillo-Michel, H., Hernandez-Viezcas, J.A., Diaz, B.C., PeraltaVidea, J.R., Gardea-Torresdey, J.L., 2012. Synchrotron micro-XRF and microXANES confirmation of the uptake and translocation of TiO2 nanoparticles in cucumber (Cucumis sativus) plants. Environ. Sci. Technol. 46, 7637e7643. Servin, A.D., Morales, M.I., Castillo-Michel, H., Hernandez-Viezcas, J.A., Munoz, B., Zhao, L.J., Nunez, J.E., Peralta-Videa, J.R., Gardea-Torresdey, J.L., 2013. Synchrotron verification of TiO2 accumulation in cucumber fruit: a possible pathway of TiO2 nanoparticle transfer from soil into the food chain. Environ. Sci. Technol. 47, 11592e11598.

Shukla, R.K., Sharma, V., Pandey, A.K., Singh, S., Sultana, S., Dhawan, A., 2011. ROSmediated genotoxicity induced by titanium dioxide nanoparticles in human epidermal cells. Toxicol. Vitro 25, 231e241. Siboza, X.I., Bertling, I., Odindo, A.O., 2014. Salicylic acid and methyl jasmonate improve chilling tolerance in cold-stored lemon fruit (Citrus limon). J. Plant Physiol. 171, 1722e1731. Song, U., Jun, H., Waldman, B., Roh, J., Kim, Y., Yi, J., Lee, E.J., 2013a. Functional analyses of nanoparticle toxicity: a comparative study of the effects of TiO2 and Ag on tomatoes ( Lycopersicon esculentum ). Ecotoxicol. Environ. Saf. 93, 60e67. Song, U., Shin, M., Lee, G., Roh, J., Kim, Y., Lee, E.J., 2013b. Functional analysis of TiO2 nanoparticle toxicity in three plant species. Biol. Trace Elem. Res. 155, 93e103. Su, M.Y., Liu, H.T., Chao, L., Qu, C.X., Zheng, L., Hong, F.S., 2009. Promotion of nanoanatase TiO2 on the spectral responses and photochemical activities of D1/D2/ Cyt b559 complex of spinach. Spectrochim. Acta A 72, 1112e1116. Sudhakar, C., Lakshmi, A., Giridarakumar, S., 2001. Changes in the antioxidant enzyme efficacy in two high yielding genotypes of mulberry (Morus alba L.) under NaCl salinity. Plant Sci. 161, 613e619. Sun, T.Y., Gottschalk, F., Hungerbühler, K., Nowack, B., 2014. Comprehensive probabilistic modelling of environmental emissions of engineered nanomaterials. Environ. Pollut. 185, 69e76. Sweetlove, L.J., Beard, K.F., Nunes-Nesi, A., Fernie, A.R., Ratcliffe, R.G., 2010. Not just a circle: flux modes in the plant TCA cycle. Trends Plant Sci. 15, 462e470.

ska, R., Kołodziej, K., Slesak, Szyman I., Zimak-Piekarczyk, P., Orzechowska, A., _ ˛ dło, A., Habina, I., Knap, W., Burda, K., Kruk, J., 2016. Titanium Gabruk, M., Za dioxide nanoparticles (100e1000mg/l) can affect vitamin E response in Arabidopsis thaliana. * Environ. Pollut. 213, 957e965. Tan, W.J., Du, W.C., Barrios, A.C., Armendariz Jr., R., Zuverza-Mena, N., Ji, Z.X., Chang, C.H., Zink, J.I., Hernandez-Viezcas, J.A., Peralta-Videa, J.R., GardeaTorresdey, J.L., 2017. Surface coating changes the physiological and biochemical impacts of nano-TiO2 in basil (Ocimum basilicum) plants. Environ. Pollut. 222, 64e72. Thuesombat, P., Hannongbua, S., Akasit, S., Chadchawan, S., 2014. Effect of silver nanoparticles on rice ( Oryza sativa L. cv. KDML 105 ) seed germination and seedling growth. Ecotox. Environ. Safe 104, 302e309. U.S. EPA, 2009. External Review Draft, Nanomaterial Case Studies: Nanoscale Titanium Dioxide in Water Treatment and in Topical Sunscreen. U.S. Environmental Protection Agency, Washington, DC. EPA/600/R-09/057.
, K., Ve Ve cerova cera, Z., Do cekal, B., Oravec, M., Pompeiano, A., Tríska, J., Urban, O., 2016. Changes of primary and secondary metabolites in barley plants exposed to CdO nanoparticles. * Environ. Pollut. 218, 207e218. Wang, Y., Frei, M., 2011. Stressed food e the impact of abiotic environmental stresses on crop quality. Agr. Ecosyst. Environ. 141, 271e286. Whitfield Åslund, M.L., Mcshane, H., Simpson, M.J., Simpson, A.J., Whalen, J.K., Hendershot, W.H., Sunahara, G.I., 2012. Earthworm sublethal responses to titanium dioxide nanomaterial in soil detected by 1H NMR metabolomics. Environ. Sci. Technol. 46, 1111e1118. 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, 1794e1807. Xia, B., Chen, B.J., Sun, X.M., Qu, K.M., Ma, F.F., Du, M.R., 2015. Interaction of TiO2 nanoparticles with the marine microalga Nitzschia closterium : growth inhibition, oxidative stress and internalization. Sci. Total Environ. 508, 525e533. Xu, L., Han, L., Huang, B., 2011. Membrane fatty acid composition and saturation levels associated with leaf dehydration tolerance and post-drought rehydration in Kentucky Bluegrass. Crop Sci. 51, 273e281. Yang, K., Zhu, L.Z., Xing, B.S., 2006. Adsorption of polycyclic aromatic hydrocarbons by carbon nanomaterials. Environ. Sci. Technol. 40, 1855e1861. Yang, F., Liu, C., Gao, F.Q., Su, M.Y., Wu, X., Zheng, L., Hong, F.S., Yang, P., 2007. The improvement of spinach growth by nano-anatase TiO2 treatment is related to nitrogen photoreduction. Biol. Trace Elem. Res. 119, 77e88. Zhang, J., Kirkham, M.B., 1994. Drought-stress-induced changes in activities of superoxide dismutase, catalase, and peroxidase in wheat species : environmental and stress responses : proteins, enzymes and metabolism. Plant Cell Physiol. 35, 785e791. Zhang, L., Wang, X.Y., Guo, J.Z., Xia, Q.L., Zhao, G., Zhou, H.N., Xie, F.W., 2013. Metabolic profiling of Chinese tobacco leaf of different geographical origins by GC-MS. J. Agric. Food Chem. 61, 2597e2605. Zhao, L.J., Huang, Y.X., Hu, J., Zhou, H.J., Adeleye, A.S., Keller, A.A., 2016a. 1H NMR and GC-MS based metabolomics reveal defense and detoxification mechanism of cucumber Plant under nano-Cu stress. Environ. Sci. Technol. 50, 2000e2010. Zhao, L.J., Ortiz, C., Adeleye, A.S., Hu, Q.R., Zhou, H.J., Huang, Y.X., Keller, A.A., 2016b. Metabolomics to detect response of Lettuce (Lactuca sativa) to Cu(OH)2 nanopesticides: oxidative stress response and detoxification mechanisms. Environ. Sci. Technol. 50, 9697e9707. Zheng, L., Su, M.Y., Liu, C., Chen, L., Hua, H., Wu, X., Liu, X.Q., Yang, F., Gao, F.Q., Hong, F.H., 2007. Effects of nanoanatase TiO2 on photosynthesis of spinach chloroplasts under different light illumination. Biol. Trace Elem. Res. 119, 68e76.