- Email: [email protected]
Metal oxide nanoparticles with low toxicity
Journal of Photochemistry and Photobiology B: Biology 151 (2015) 17–24
Contents lists available at ScienceDirect
Journal of Photochemistry and Photobiology B: Biology journal homepage: www.elsevier.com/locate/jphotobiol
Metal oxide nanoparticles with low toxicity Alan Man Ching Ng a,b, Mu Yao Guo a,b, Yu Hang Leung a, Charis M.N. Chan c, Stella W.Y. Wong c, Mana M.N. Yung c, Angel P.Y. Ma c, Aleksandra B. Djurišic´ a,⇑, Frederick C.C. Leung c, Kenneth M.Y. Leung c, Wai Kin Chan d, Hung Kay Lee e a
Department of Physics, The University of Hong Kong, Pokfulam Road, Hong Kong Department of Physics, South University of Science and Technology of China, Shenzhen, China School of Biological Sciences, The University of Hong Kong, Pokfulam Road, Hong Kong d Department of Chemistry, The University of Hong Kong, Pokfulam Road, Hong Kong e Department of Chemistry, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong b c
a r t i c l e
i n f o
Article history: Received 26 May 2015 Received in revised form 26 June 2015 Accepted 29 June 2015 Available online 30 June 2015 Keywords: Metal oxide nanoparticles Toxicity Ecotoxicity
a b s t r a c t A number of different nanomaterials produced and incorporated into various products are rising. However, their environmental hazards are frequently unknown. Here we consider three different metal oxide compounds (SnO2, In2O3, and Al2O3), which have not been extensively studied and are expected to have low toxicity. This study aimed to comprehensively characterize the physicochemical properties of these nanomaterials and investigate their toxicity on bacteria (Escherichia coli) under UV illumination and in the dark, as well as on a marine diatom (Skeletonema costatum) under ambient illumination/dark (16–8 h) cycles. The material properties responsible for their low toxicity have been identified based on comprehensive experimental characterizations and comparison to a metal oxide exhibiting significant toxicity under illumination (anatase TiO2). The metal oxide materials investigated exhibited significant difference in surface properties and interaction with the living organisms. In order for a material to exhibit significant toxicity, it needs to be able to both form a stable suspension in the culture medium and to interact with the cell walls of the test organism. Our results indicated that the observed low toxicities of the three nanomaterials could be attributed to the limited interaction between the nanoparticles and cell walls of the test organisms. This could occur either due to the lack of significant attachment between nanoparticles and cell walls, or due to their tendency to aggregate in solution. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Toxicity of nanomaterials has been a rising concern with the increase of the production of various nanomaterials [1–8]. In particular, there is interest in elucidating the relationship between the structural properties and the activity of a nanomaterial [1], and developing predictive toxicological approaches to establish toxicity screening priorities [2]. Such predictive approaches can include prediction of oxidative stress (for example, the ability of a nanomaterial to generate reactive oxygen species (ROS) [2] or their chemical stability [2,5]. They can be semi-empirical approaches such as quantitative structure–activity relationship (QSAR) [2,5,6] or purely theoretical approaches based on fundamental properties of the nanomaterials. In spite of enormous interest in predictive toxicology of nanomaterials, to date there has been very limited progress on this ⇑ Corresponding author. E-mail address: [email protected] (A.B. Djurišic´). http://dx.doi.org/10.1016/j.jphotobiol.2015.06.020 1011-1344/Ó 2015 Elsevier B.V. All rights reserved.
issue. Furthermore, there is limited understanding of the mechanisms of toxicity of metal oxides and a large number of studies in the literature report incomplete characterization of the nanomaterials studied [9]. Different toxicity mechanisms, such as ROS production followed by oxidative stress/lipid peroxidation/cell wall damage, metal ion release, and interaction between the nanomaterial and cells, have been proposed by different research groups but as yet no conclusive and unambiguous identification of the mechanism of toxicity has been made [9]. This has an obvious implication on trying to understand and predict toxicity of nanomaterials. Due to the complexity of biological systems, it is difficult to theoretically predict the nanomaterial behavior in the environment, especially considering the fact that experimental studies often report contradictory results and the actual toxicity mechanism is unclear [9]. Thus, comprehensive experimental studies which would establish distinguishing features of interaction between toxic and non-toxic nanoparticles and living organisms are of significant interest in establishing the toxicity mechanisms
18
A.M.C. Ng et al. / Journal of Photochemistry and Photobiology B: Biology 151 (2015) 17–24
Fig. 1. TEM images of different nanoparticles. The insets show corresponding selected area electron diffraction patterns.
and possibly enable future predictions once more realistic models are developed. We have selected to study the following metal oxides: a-Al2O3, In2O3, and SnO2, which have not been extensively studied but are expected to result in low to moderate toxicity [10–20]. The majority of studies was performed on alumina which exhibited low to moderate toxicity according to literature reports [12,14,16–18,20], while there have been very few studies on tin oxide [18] and indium oxide [15]. On the other hand, anatase TiO2 is well studied and known to exhibit toxicity to various organisms under different illumination conditions [9]. Therefore, a comprehensive comparative study of anatase TiO2 compared to low toxicity metal oxides is expected to reveal distinguishing characteristics of nanomaterial/organism interactions resulting in toxicity. 2. Methods 2.1. Materials and characterization TiO2 (99%, average particle size APS 15 nm), SnO2 (99.5%, APS 55 nm), In2O3 (99.99%, APS 30–50 nm), and Al2O3 (99%, APS
30–40 nm) nanoparticles were obtained from Nanostructured & Amorphous Materials Inc. The particle morphology and structure was investigated by transmission electron microscopy (TEM) and selected area electron diffraction (SAED) using a Phillips Tecnai G2 20 S-TWIN TEM. Absorption measurements were performed using a Cary 50 Bio UV–Vis spectrophotometer. Aggregation sizes of the nanoparticles in 0.9% w/v sodium chloride solution were determined using ZETASIZER 3000HSA (Malvern Instruments Ltd.). For aggregate size determination in artificial seawater, nanoparticles at a concentration 100 mg/L were dispersed in filtered artificial seawater (salinity, 30 ± 0.5‰; pH, 8.0 ± 0.1; sea salt: Tropic Marine, Germany; filtered through 0.45 lm membrane filter). The aggregate size (average of three replicates) was determined using laser diffractometry (LD; LS 13 320 Series, Beckman Coulter Inc., Fullerton, USA). The reactive oxygen species generated by the nanoparticles were detected using electron spin resonance (ESR) spectroscopy at room temperature with the addition of a spin trap molecule [21]. Spin trap 5,5-Dimethyl-1-pyrroline N-oxide (DMPO) was purchased from Sigma–Aldrich Co., and the solution was prepared by adding 0.02 M DMPO to 1 mg/ml metal oxide nanoparticle
Table 1 Summary of nanoparticle characteristics and toxicity testing results in 0.9% w/v NaCl (SC) and artificial seawater (ASW), respectively, and toxicity testing results on the marine diatom in ASW. The toxicity endpoint is median inhibition concentration (IC50) after 72 h of exposure to the nanomaterial, and the IC50 values sharing with the same superscripted letter are statistically indifferent (based on overlapping of the 95% confidence intervals). IC50 values are given as mean value (n = 3), with confidence interval indicated in brackets. 1
Nanoparticle
Aggregation size in SC (lm), metal content (lg/L)
Aggregation size in ASW (lm), metal content (lg/L)
72-h IC50 (mg L
Control TiO2 SnO2 In2O3 Al2O3
–, <10 (Sn, In, Ti), <100 (Al) 0.6, <10 (Ti) 1.1, <10 (Sn) 0.1–0.5, <10 (In) 0.35 and 1.6, <100 (Al)
–, <10 (Sn, In, Ti), <100 (Al) 0.69, <10 (Ti) 0.42 and 1.8, <10 (Sn) 0.45 and 1.6, <10 (In) 0.67, <100 (Al)
– 353.3 (322.8–386.8)a 5200.0 (1389.0–19472.0)b 739.8 (580.0–943.7)c 2101.0 (1671.0–2642.0)b
) Skeletonema costatum
A.M.C. Ng et al. / Journal of Photochemistry and Photobiology B: Biology 151 (2015) 17–24
19
Fig. 2. ESR spectra of H2O2 with DMPO for different illumination conditions; ESR spectra of different nanoparticles with DMPO spin trap for UV illumination, 0.9% w/ v NaCl and ambient illumination, artificial seawater. The spectra have been vertically shifted for clarity.
Fig. 3. Absorption spectra of different nanoparticles. Emission spectrum of UV lamp used for illumination is also shown for comparison.
Table 2 Number of bacteria colonies for bacteria exposed to the different metal oxide nanoparticles in suspension. The bacteria count is an average value of three replicates. For each condition, mean values sharing the same superscripted letter are statistically indifferent (1-way analysis of variance followed by post hoc Tukey test, p < 0.05). Condition
Sample
Colony counts (Mean ± S.D.)
Dark
Control TiO2 SnO2 In2O3 Al2O3
406, 464, 602, 408, 544,
394, 560, 450, 376, 694,
378 526 574 336 544
(393 ± 14)ab (517 ± 49)abc (542 ± 81)bc (373 ± 36)a (594 ± 87)c
Survival rate (%) 100.0 131.6 137.9 94.9 151.1
Ambient
Control TiO2 SnO2 In2O3 Al2O3
556, 476, 592, 742, 612,
562, 614, 540, 420, 764,
564 594 742 622 720
(561 ± 4)a (561 ± 75)a (625 ± 105)a (595 ± 163)a (699 ± 78)a
100.0 100.0 111.4 106.1 124.6
UV
Control TiO2 SnO2 In2O3 Al2O3
417, 237, 381 (345 ± 95)a 0, 0, 0 (0 ± 0)b 192, 244, 169 (202 ± 38)c 703, 687, 769 (720 ± 43)d 372, 392, 417 (394 ± 23)a
100.0 0.0 58.6 208.7 114.2
Fig. 4. FTIR spectra of different nanoparticles before and after exposure to LPS. FTIR spectrum of pure LPS is also shown for comparison.
20
A.M.C. Ng et al. / Journal of Photochemistry and Photobiology B: Biology 151 (2015) 17–24
Fig. 5. Representative SEM images of bacterial cells exposed to different nanoparticles under UV illumination.
suspension in sodium chloride 0.9% w/v aqueous solution or artificial seawater. ESR measurements were performed using a Bruker EMX EPR spectrometer, with the following settings: microwave power 4 mW, microwave frequency 9.69 GHz, modulation frequency 100 kHz, modulation amplitude 1 G, time constant 41 ms, sweep time 42 s. The measurements were performed for ambient illumination (in artificial seawater) or under UV illumination (365 nm, Blak-RayÒ B-100 AP Lamp, 27 mW/cm2 for 2 min). For metal ion release analysis, nanoparticles were dispersed in different media for 24 h. The nanoparticles were then removed by centrifugation and filtering. The metal ion contents were analyzed by inductively coupled plasma mass spectrometry (ICP-MS) using Agilent 7500cx ICP-MS with EG020 standard.
2.2. Antibacterial activity testing Antibacterial testing was performed on a gram-negative bacterium Escherichia coli XL1-Blue (Stratagene), since E. coli is a commonly used organism for nanomaterial toxicity/antibacterial activity testing. Luria-Bertani (LB) broth (Affymetrix/USB) was used as the culture medium, and the incubation temperature was 37 °C. When the antibacterial activity test was done under UV illumination, 0.9% w/v Sodium Chloride (NaCl) solution was used as the dispersion medium for the bacteria [22]. 20 lL of bacteria suspension (109 CFU/ml) was added to 10 ml of nanoparticle
suspension in NaCl solution with a concentration of 0.1 mg/ml. The mixture was subjected to UV illumination (365 nm, Blak-RayÒ B-100 AP Lamp, 40 mW/cm2 measured by an optical power meter) for 20 min. Serial dilution of the bacterialnanoparticle suspension was performed and was then plated onto LB agar plates in triplicate. After incubating the agar plates at 37 °C over night (at least 16 h), number of colonies of bacteria formed was counted.
2.3. Ecotoxicity testing Ecotoxicity testing was done by performing acute algal growth inhibition tests on Skeletonema costatum (CCMP 1332; CCMP, USA) algae. The test procedures were modified from ASTM and OECD guidelines [23,24]. The nanomaterial dispersions for testing were prepared by dispersing 1 g of nanoparticles into 100 ml of filtered artificial seawater, followed by continuous stirring at room temperature for a minimum of 3 days before performing the experiments. For the experiments, initial algal concentrations of 105 cells/ml of S. costatum were exposed to different concentrations of the test chemicals dispersed in f2-Si medium in 24-well multidishes (Nunc; Naterville, USA) (n = 3–4). To evaluate possible shading effects (due to insolubility of metal oxides resulting in a cloudy appearance of the suspension), an experimental design adapted from Ref. [25] was utilized. The algae were incubated only
A.M.C. Ng et al. / Journal of Photochemistry and Photobiology B: Biology 151 (2015) 17–24
21
in the lower chamber of the stacked transparent multidish (with walls and undersides blackened so that only the light passing through the upper wells could reach the lower dish). For toxicity testing, the test compounds and the algae were mixed together in the lower plate wells, while the upper wells contained only f2-Si medium. For shading effects evaluation, the test materials and algae were separated in the upper and lower wells, respectively. Due to the tendency of nanoparticles to aggregate and settle down over time, the dishes were shaken on a titer plate shaker at 100 rpm for 15 min every 2 h. Incubation was performed in an environmental chamber at room temperature (25 ± 2 °C) for 72 h, with 16 h:8 h light:dark cycle. Average growth rate in the test period was calculated based on initial and final cell counts determined using a hemocytometer. 2.4. Characterization of the interaction between nanoparticles and bacterial and algae cells Scanning electron microscopy (SEM) was used to examine the interaction between the nanoparticles and cells. After exposure to nanoparticles, the bacterial cells were collected and fixed in 2.5% glutaraldehyde in cacodylate buffer (0.1 M sodium cacodylate–HCl buffer, pH 7.4) at 4 °C overnight. For the algal cells, fixation was performed in 2.5% glutaraldehyde in cacodylate buffer (0.1 M sodium cacodylate–HCl buffer, pH 7.4) at 4 °C for 72 h. The cells were then collected on a membrane (Millipore, pore size 0.8 lm), rinsed with cacodylate buffer and serially dehydrated with ethanol. The cells were dried by critical point drying method. Finally, a thin layer of Au was coated on the specimens, and SEM was performed using LEO 1530 FEG SEM. For the investigation of the attachment of lipopolysaccharide (LPS) on the nanoparticles surface, the nanoparticles were first added into an aqueous solution of LPS (from E. coli K-235, Sigma–Aldrich) and dispersed by sonication. The concentration of both LPS and nanoparticles was 1 mg/ml. The suspension was left in ambient for 2 h. The nanoparticles were then collected by centrifugation and cleaned with fresh de-ionized water for several times, followed by drying in ambient overnight. FTIR measurements were performed with PerkinElmer Spectrum Two IR spectrometer. For the interaction between nanoparticles and algal cell walls, the cell walls of S. costatum were extracted with a method similar to a paper reported [26]. The algal cells were firstly suspended in 1 mM CaCl2. Glass beads (0.3 mm in diameter) were added and mixed with the cells. The mixture was then vortexed at maximum speed for 6 min. The lysed cells were then collected by centrifugation at 600g for 5 min. The resultant cell pellet was then washed with 1 mM CaCl2 for several times until the supernatant was colorless. For FTIR measurement, the cell wall extract was resuspended in nanoparticle suspension in f2-Si with concentration of 1 mg/ml. The suspension mixture was left in ambient for 2 h. The nanoparticles were then collected by centrifugation, rinsed with de-ionized water and then dried in ambient overnight. The nanoparticles were mixed with KBr powder (infrared grade, Sigma–Aldrich) and pellets of the mixture were made. The measurement was performed on the pellets using PerkinElmer Spectrum Two IR spectrometer. 3. Results and discussion Fig. 1 shows the TEM images of the nanoparticles investigated. Their aggregation size in different media is summarized in Table 1. It can be observed that all the nanoparticles tend to aggregate into larger aggregates in both media investigated. The production of ROS under ambient and UV illumination was tested using electron spin resonance (ESR) spectroscopy and the obtained results are
Fig. 6. FTIR spectra of different nanoparticles before and after exposure to algal cell walls. FTIR spectrum of pure algal cell walls is also shown for comparison.
22
A.M.C. Ng et al. / Journal of Photochemistry and Photobiology B: Biology 151 (2015) 17–24
Fig. 7. Representative SEM images of algal cells exposed to different nanoparticles under ambient illumination.
shown in Fig. 2. The only peaks observed correspond to the OH radicals based on the comparison with the ESR spectra of a dilute H2O2 solution [21]. For the generation of ROS under illumination, we need to consider the bandgap of the materials. The illumination energy (3.4 eV) is smaller than the bandgap of alumina (>8 eV, although in nanoparticles it can be as small as 3.7 eV) [27] and tin oxide (3.6 eV) [28], and larger than the bandgap of indium oxide (3.0 eV) [29] and anatase titania (3.2 eV) [30]. From the absorption spectra of the nanoparticles shown in Fig. 3, it can be observed that all oxides except alumina exhibit similar absorption, indicating that for SnO2 nanoparticles bandgap smaller than the bulk material can be observed, likely due to the presence of defects. In the 0.9% w/v NaCl solution under UV illumination, ROS are produced by TiO2, SnO2, and In2O3, with TiO2 exhibiting lower signal compared to the other two oxides. However, we cannot observe any correlation between the ROS production in NaCl solution (Fig. 2), aggregation size in NaCl solution (Table 1), and toxicity to E. coli bacteria (Table 2). Under ambient light, there is no significant difference in E. coli counts among the different nanomaterials, but there are significantly different effects of the nanomaterials on the bacteria in dark or under UV irradiation (Table 2). Only TiO2 shows significant toxicity under UV illumination (0% survival rate), followed by SnO2 which exhibits some toxicity (60% survival rate), while other two oxides show no toxicity to E. coli. In addition, no significant metal ion release is observed from any of the
nanoparticles considered (see Table 1). Thus, toxicity mechanism of SnO2 toward E. coli does not involve photocatalysis or metal ion release, possible mechanism could be mechanical damage to the bacteria by nanoparticles. To examine this possibility, we have examined the interaction between the nanoparticles and the lipopolysaccharide (LPS), a molecule which is the major constituent of E. coli outer membrane [31]. Obtained results are shown in Fig. 4, while the SEM images of the bacteria exposed to nanoparticles are shown in Fig. 5. In the case of TiO2, we can observe cell damage (holes in the membrane) and a number of larger and small aggregates sticking to the cell surface. For SnO2, we can see loose aggregates with some particles sticking to the cells, and the number of visibly damaged cells is lower compared to TiO2, in agreement with the antibacterial testing results. In the case of In2O3, we also see loose aggregates but they are rarely sticking to cell surfaces, while for alumina we observe coarser aggregates which also do not seem to be attached to cell surfaces. From the FTIR spectra of nanoparticles exposed to LPS (Fig. 4), we can observe significant changes in the case of TiO2 in the region 1000–1500 cm 1 which mainly corresponds to the vibrations of carbohydrates and vibrations of PO2 group [13,31,32]. In particular, the vibrations at 1230 cm 1 and 1070 cm 1 can be attributed to the symmetric and asymmetric vibrations of PO2 [31,32]. Small changes in peak shapes in the same spectral region are observed for SnO2, and no significant
A.M.C. Ng et al. / Journal of Photochemistry and Photobiology B: Biology 151 (2015) 17–24
Fig. 8. Photographs of nanoparticle suspensions (1 mg/ml) in artificial seawater observed at different times.
23
changes in this region are found for the other two oxides. Therefore, toxicity appears to be related to the ability of nanoparticles to interact with the cell walls. These observations are in agreement with a previous study on toxicity of alumina, which found that the toxicity could be attributed to the interactions between nanoparticles and cells [13]. In the case of algae, we also observe more significant effects of TiO2 compared to other nanomaterials (Table 1). The lowest value of 72 h-IC50 is observed for TiO2, followed by In2O3, Al2O3, and SnO2. Low toxicities of In2O3, Al2O3, and SnO2 materials are in agreement with the data reported in other studies. Although there have been studies reporting effects of a-alumina on reduction of embryo growth and hatchability in freshwater snails [10] and the inhibition of the growth of algae [13], the majority of the studies on the toxicity of alumina indicate that it has moderate [18] to low toxicity [12,14,16,17,20], as documented in studies on yeast [12], different cell cultures [14,20], plants [16], E. coli bacteria [17], and oxidative stress in human blood serum [19]. Exceptions can possibly occur in case of inhalation where inflammatory response can occur [11], which is also the case for In2O3 and indium tin oxide nanoparticles (with indium oxide less toxic compared to indium tin oxide) [15]. Tin oxide was also previously found to have moderate toxicity, somewhat higher than alumina [18]. However, reasons for the low toxicity of these materials have not been conclusively identified. Since oxidative stress is commonly proposed as the cause of toxicity [9,19], we have examined ROS production in the algae culture medium since ROS production is medium dependent [33]. In f2-Si medium, all oxides except alumina are producing ROS, with the signal intensity following the trend TiO2 > In2O3 > SnO2. Thus, in this case as well the toxicity cannot be fully explained by ROS production, nor by aggregation sizes. Therefore, we have also examined the interaction of nanoparticles with algae cell walls and algae cells, with FTIR spectra shown in Fig. 6 and SEM images shown in Fig. 7. In this case, however, we can observe that all the particles exhibit significant changes upon exposure to algae cell walls, including those exhibiting very low toxicity, such as SnO2 and alumina, while TiO2 which displays the lowest IC50 value also exhibits the smallest changes in the FTIR spectra. For other oxides, changes in the spectra are observed not only in the vibrations of PO2 group [31,32] but we can also observe significant changes in the region 2800–2950 cm 1, corresponding to CH3 group vibrations [32], as well as vibrations of water (scissoring mode at 1630 cm 1) [30] and amide II bands at 1540 cm 1. However, from the SEM images (Fig. 7) we can observe that for In2O3 there are large aggregates formed and very few nanoparticles stick to the algae cell surface. Adhesion of the nanoparticles to the algal cells can be observed for TiO2 and SnO2 nanoparticles, but in this case as well SnO2 nanoparticles aggregate into looser agglomerations compared to TiO2 nanoparticles. The attachment of some small particles to cell walls does not appear to be necessarily detrimental to cell survival since SnO2 exhibits the highest IC50 value. Similar observation was made previously for some CeO2 samples [33]. It has been proposed that the attachment of nanoparticles to the cell walls is dependent on the aggregation tendency [13]. However, the aggregation tendency we observed in SEM images is not consistent with the measured aggregation sizes. This is due to the fact that aggregation size measurement characterizes agglomeration of nanoparticles in a medium at a certain specific point in time (in our study, immediately after dispersing the nanoparticles, to ensure reproducible results). However, during the course of toxicity testing, the particles may exhibit sedimentation at different rates which would affect the outcome of the experiment. It can be observed that TiO2
24
A.M.C. Ng et al. / Journal of Photochemistry and Photobiology B: Biology 151 (2015) 17–24
particles remain suspended in the medium for a long time after the other oxide particles exhibit significant sedimentation, as shown in Fig. 8. This phenomenon could be avoided by adding dispersing agents, but that could affect the test result due to artifacts induced by a dispersing agent [9] and the obtained results would not apply to environmentally relevant conditions. Therefore, we can conclude that electron microscopy is an essential tool for characterizing the interaction between the nanoparticles and the cells, since the true size and shape of aggregates relevant to the conditions under which the experiment is conducted can be observed.
[11]
[12]
[13]
[14]
[15]
4. Conclusions We have examined the toxicity of metal oxides expected to show low toxicity (SnO2, alumina, In2O3) and compared them to a material which is known to exhibit significant toxicity, anatase TiO2. We found that the studied three metal oxides indeed exhibit much lower toxicity compared to titania, and that their low toxicity can be attributed to their tendency to aggregate in solution which results in limited interaction between the nanoparticles and cell walls of the test organisms. Even in the cases where the material has the ability to strongly interact with the cell walls (SnO2, alumina, or In2O3 and algae cell walls), the aggregation in the test medium may prevent the significant interaction between the nanoparticles and the cells. The ability to interact with the cell walls (for example TiO2 and LPS), in particular involving the phosphate groups in the cell walls, and the slow sedimentation rate in the test medium are both essential to observe the toxicity of a nanomaterial.
[16]
Acknowledgments
[22]
Financial support from the project RGC CRF CityU6/CRF/08 and the Strategic Research Theme, University Development Fund, and Small Project Funding of the University of Hong Kong is acknowledged. The authors thank ALS Technichem (HK) Pty Ltd for metal ion release measurements.
[23]
[17]
[18]
[19]
[20]
[21]
[24] [25] [26]
References [1] P.J.J. Alvarez, V. Colvin, J. Lead, V. Stone, Research priorities to advance ecoresponsible nanotechnology, ACS Nano 3 (2009) 1616–1619. [2] H. Meng, T. Xia, S. George, A.E. Nel, A predictive toxicological paradigm for the safety assessment of nanomaterials, ACS Nano 3 (2009) 1620–1627. [3] A.S. Barnard, Computational strategies for predicting the potential risks associated with nanotechnology, Nanoscale 1 (2009) 89–95. [4] A.S. Barnard, How can ab initio simulations address risks in nanotech?, Nat Nanotechnol. 4 (2009) 332–335. [5] T. Puzyn, B. Rasulev, A. Gajewitz, X. Hu, T.P. Dasari, A. Michalkova, H.M. Hwang, A. Toropov, D. Leszczynska, J. Lesczynski, Using nano-QSAR to predict the cytotoxicity of metal oxide nanoparticles, Nat. Nanotechnol. 6 (2011) 175–178. [6] D. Fourches, D. Pu, C. Tassa, R. Weissleder, S.Y. Shaw, R.J. Mumper, A. Tropsha, Quantitative nanostructure–activity relationship modeling, ACS Nano 4 (2010) 5703–5712. [7] A.S. Barnard, One-to-one comparison of sunscreen efficacy, aesthetics and potential nanotoxicity, Nat. Nanotechnol. 5 (2010) 271–274. [8] E. Burello, A.P. Worth, Theoretical framework for predicting the oxidative stress potential of oxide nanoparticles, Nanotoxicology 5 (2011) 228–235. [9] A.B. Djurišic´, Y.H. Leung, A.M.C. Ng, X.Y. Xu, P.K.H. Lee, N. Degger, R.S.S. Wu, Toxicity of metal oxide nanoparticles: mechanisms, characterization, and avoiding experimental artefacts, Small 11 (2015) 26–44. [10] N. Musee, P.J. Oberholster, L. Sikhwivhilu, A.-M. Botha, The effects of engineered nanoparticles on survival, reproduction, and behaviour of
[27]
[28]
[29]
[30]
[31]
[32]
[33]
freshwater snail, Physa acuta (Draparnaud, 1805), Chemosphere 81 (2010) 1196–1203. C. Som, P. Wick, H. Krug, B. Nowack, Environmental and health effects of nanomaterials in nanotextiles and façade coatings, Environ. Int. 37 (2011) 1131–1142. C. García-Saucedo, J.A. Field, L. Otero-Gonzalez, R. Sierra-Álvarez, Low toxicity of HfO2, SiO2, Al2O3 and CeO2 nanoparticles to the yeast, Saccharomyces cerevisiae, J. Hazard. Mater. 192 (2011) 1572–1579. I.M. Sadiq, S. Pakrashi, N. Chandrasekaran, A. Mukherjee, Studies on toxicity of aluminum oxide (Al2O3) nanoparticles to microalgae species: Scenedesmus sp. and Chlorella sp., J. Nanopart. Res. 13 (2011) 3287–3299. D. Yoon, D. Woo, J.H. Kim, M.K. Kim, T. Kim, E.S. Hwang, S. Baik, Agglomeration, sedimentation, and cellular toxicity of alumina nanoparticles in cell culture medium, J. Nanopart. Res. 13 (2011) 2543–2551. K. Nagano, T. Nishizawa, Y. Eitaki, M. Ohnishi, T. Noguchi, H. Arito, S. Fukushima, Pulmonary toxicity in mice by 2- and 13-week inhalation exposures to indium-tin oxide and indium oxide aerosols, J. Occup. Health 53 (2011) 234–239. C.W. Lee, S. Mahendra, K. Zodrow, D. Li, Y.C. Tsai, J. Braam, P.J.J. Alvarez, Developmental phytotoxicity of metal oxide nanoparticles to Arabidopsis thaliana, Environ. Toxicol. Chem. 29 (2010) 669–675. I.M. Sadiq, B. Chowdhury, N. Chandrasekaran, A. Mukherjee, Antimicrobial sensitivity of Escherichia coli to alumina nanoparticles, Nanomed. Nanotechnol. Biol. Med. 5 (2009) 282–286. S. Lanone, F. Rogerieux, J. Geys, A. Dupont, E. Maillot-Marechal, J. Boczkowski, G. Lacroix, P. Hoet, Comparative toxicity of 24 manufactured nanoparticles in human alveolar epithelial and macrophage cell lines, Part. Fibre Toxicol. 6 (2009) 14. 1–12. E.J. Rogers, S.F. Hsieh, N. Organti, D. Schmidt, D. Bello, A high throughput in vitro analytical approach to screen for oxidative stress potential exerted by nanomaterials using a biologically relevant matrix: human blood serum, Toxicol. In Vitro 22 (2008) 1639–1647. J. Sun, S.C. Wang, D. Zhao, F.H. Hun, L. Weng, H. Liu, Cytotoxicity, permeability, and inflammation of metal oxide nanoparticles in human cardiac microvascular endothelial cells, Cell Biol. Toxicol. 27 (2011) 333–342. M. Kamibayashi, S. Oowada, H. Kameda, T. Okada, O. Inanami, S. Ohta, T. Ozawa, K. Makino, Y. Kotake, Synthesis and characterization of a practically better DEPMPO-type spin trap, 5-(2,2-dimethyl-1,3-propoxy cyclophosphoryl)-5-methyl-1-pyrroline N-oxide (CYPMPO), Free Radic. Res. 40 (2006) 1166–1172. T.P.T. Cushnie, P.K.J. Robertson, S. Officer, P.M. Pollard, C. McCullagh, J.M.C. Robertson, Variables to be considered when assessing the photocatalytic destruction of bacterial pathogens, Chemosphere 74 (2009) 1374–1378. Environmental Toxicology Standards, ASTM Standard E1218, Guide for Conducting Static Toxicity Tests with Microalgae, ASTM International, Philadelphia, PA, 1993. OECD, Test No. 201: Freshwater Alga and Cyanobacteria, Growth Inhibition Test, OECD Guidelines for the Testing of Chemicals, Paris, France, 1984. K. Hund-Rinke, M. Simon, Ecotoxic effect of photocatalytic active nanoparticles (TiO2) on algae and daphnids, Environ. Sci. Pollut. Res. 13 (2006) 225–232. K. Kröger, C. Bergsdorf, M. Sumper, A new calcium binding glycoprotein family constitutes a major diatom cell wall component, EMBO J. 13 (1994) 4676– 4683. I.L. Liu, P. Shen, S.Y. Chen, Formation of ultrafine and dense a-Al2O3 nanoparticles via kinetic phase change in a dynamic process, J. Nanopart. Res. 12 (2010) 2929–2940. A. Kar, M.A. Stroscio, M. Dutta, J. Kumari, M. Meyyapan, Growth and properties of tin oxide nanowires and the effect of annealing conditions, Semicond. Sci. Technol. 25 (2010) 024012. 1–9. P.D.C. King, T.D. Veal, Ch.Y. Wang, D.J. Payne, A. Bourlange, H. Zhang, G.R. Bell, V. Cimalla, O. Ambacher, R.G. Egdell, F. Bechstedt, C.F. McConville, Band gap, electronic structure, and surface electron accumulation of cubic and rhombohedral In2O3, Phys. Rev. B 79 (2009) 205211. 1–10. G.K. Mor, O.K. Varghese, M. Paulose, C.A. Grimes, Transparent highly ordered TiO2 nanotube arrays via anodization of titanium thin films, Adv. Funct. Mater. 15 (2005) 1291–1296. J. Kiwi, V. Nadtochenko, Evidence for the mechanism of photocatalytic degradation of the bacterial wall membrane at the TiO2 interface by ATRFTIR and laser kinetic spectroscopy, Langmuir 21 (2005) 4631–4641. A. Hsu, F.Z. Liu, Y.H. Leung, A.P.Y. Ma, A.B. Djurišic´, F.C.C. Leung, W.K. Chan, H.K. Lee, Is the effect of surface modifying molecules on antibacterial activity universal for a given materials?, Nanoscale 6 (2014) 10323–10331 Y.H. Leung, M.M.N. Yung, A.M.C. Ng, A.P.Y. Ma, S.W.Y. Wong, C.M.N. Chan, Y.H. Ng, A.B. Djurišic´, M.Y. Guo, M.T. Wong, F.C.C. Leung, W.K. Chan, K.M.Y. Leung, H.K. Lee, Toxicity of CeO2 nanoparticles – the effect of nanoparticle properties, J. Photochem. Photobiol. B 145 (2015) 48–59.
Contents lists available at ScienceDirect
Journal of Photochemistry and Photobiology B: Biology journal homepage: www.elsevier.com/locate/jphotobiol
Metal oxide nanoparticles with low toxicity Alan Man Ching Ng a,b, Mu Yao Guo a,b, Yu Hang Leung a, Charis M.N. Chan c, Stella W.Y. Wong c, Mana M.N. Yung c, Angel P.Y. Ma c, Aleksandra B. Djurišic´ a,⇑, Frederick C.C. Leung c, Kenneth M.Y. Leung c, Wai Kin Chan d, Hung Kay Lee e a
Department of Physics, The University of Hong Kong, Pokfulam Road, Hong Kong Department of Physics, South University of Science and Technology of China, Shenzhen, China School of Biological Sciences, The University of Hong Kong, Pokfulam Road, Hong Kong d Department of Chemistry, The University of Hong Kong, Pokfulam Road, Hong Kong e Department of Chemistry, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong b c
a r t i c l e
i n f o
Article history: Received 26 May 2015 Received in revised form 26 June 2015 Accepted 29 June 2015 Available online 30 June 2015 Keywords: Metal oxide nanoparticles Toxicity Ecotoxicity
a b s t r a c t A number of different nanomaterials produced and incorporated into various products are rising. However, their environmental hazards are frequently unknown. Here we consider three different metal oxide compounds (SnO2, In2O3, and Al2O3), which have not been extensively studied and are expected to have low toxicity. This study aimed to comprehensively characterize the physicochemical properties of these nanomaterials and investigate their toxicity on bacteria (Escherichia coli) under UV illumination and in the dark, as well as on a marine diatom (Skeletonema costatum) under ambient illumination/dark (16–8 h) cycles. The material properties responsible for their low toxicity have been identified based on comprehensive experimental characterizations and comparison to a metal oxide exhibiting significant toxicity under illumination (anatase TiO2). The metal oxide materials investigated exhibited significant difference in surface properties and interaction with the living organisms. In order for a material to exhibit significant toxicity, it needs to be able to both form a stable suspension in the culture medium and to interact with the cell walls of the test organism. Our results indicated that the observed low toxicities of the three nanomaterials could be attributed to the limited interaction between the nanoparticles and cell walls of the test organisms. This could occur either due to the lack of significant attachment between nanoparticles and cell walls, or due to their tendency to aggregate in solution. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Toxicity of nanomaterials has been a rising concern with the increase of the production of various nanomaterials [1–8]. In particular, there is interest in elucidating the relationship between the structural properties and the activity of a nanomaterial [1], and developing predictive toxicological approaches to establish toxicity screening priorities [2]. Such predictive approaches can include prediction of oxidative stress (for example, the ability of a nanomaterial to generate reactive oxygen species (ROS) [2] or their chemical stability [2,5]. They can be semi-empirical approaches such as quantitative structure–activity relationship (QSAR) [2,5,6] or purely theoretical approaches based on fundamental properties of the nanomaterials. In spite of enormous interest in predictive toxicology of nanomaterials, to date there has been very limited progress on this ⇑ Corresponding author. E-mail address: [email protected] (A.B. Djurišic´). http://dx.doi.org/10.1016/j.jphotobiol.2015.06.020 1011-1344/Ó 2015 Elsevier B.V. All rights reserved.
issue. Furthermore, there is limited understanding of the mechanisms of toxicity of metal oxides and a large number of studies in the literature report incomplete characterization of the nanomaterials studied [9]. Different toxicity mechanisms, such as ROS production followed by oxidative stress/lipid peroxidation/cell wall damage, metal ion release, and interaction between the nanomaterial and cells, have been proposed by different research groups but as yet no conclusive and unambiguous identification of the mechanism of toxicity has been made [9]. This has an obvious implication on trying to understand and predict toxicity of nanomaterials. Due to the complexity of biological systems, it is difficult to theoretically predict the nanomaterial behavior in the environment, especially considering the fact that experimental studies often report contradictory results and the actual toxicity mechanism is unclear [9]. Thus, comprehensive experimental studies which would establish distinguishing features of interaction between toxic and non-toxic nanoparticles and living organisms are of significant interest in establishing the toxicity mechanisms
18
A.M.C. Ng et al. / Journal of Photochemistry and Photobiology B: Biology 151 (2015) 17–24
Fig. 1. TEM images of different nanoparticles. The insets show corresponding selected area electron diffraction patterns.
and possibly enable future predictions once more realistic models are developed. We have selected to study the following metal oxides: a-Al2O3, In2O3, and SnO2, which have not been extensively studied but are expected to result in low to moderate toxicity [10–20]. The majority of studies was performed on alumina which exhibited low to moderate toxicity according to literature reports [12,14,16–18,20], while there have been very few studies on tin oxide [18] and indium oxide [15]. On the other hand, anatase TiO2 is well studied and known to exhibit toxicity to various organisms under different illumination conditions [9]. Therefore, a comprehensive comparative study of anatase TiO2 compared to low toxicity metal oxides is expected to reveal distinguishing characteristics of nanomaterial/organism interactions resulting in toxicity. 2. Methods 2.1. Materials and characterization TiO2 (99%, average particle size APS 15 nm), SnO2 (99.5%, APS 55 nm), In2O3 (99.99%, APS 30–50 nm), and Al2O3 (99%, APS
30–40 nm) nanoparticles were obtained from Nanostructured & Amorphous Materials Inc. The particle morphology and structure was investigated by transmission electron microscopy (TEM) and selected area electron diffraction (SAED) using a Phillips Tecnai G2 20 S-TWIN TEM. Absorption measurements were performed using a Cary 50 Bio UV–Vis spectrophotometer. Aggregation sizes of the nanoparticles in 0.9% w/v sodium chloride solution were determined using ZETASIZER 3000HSA (Malvern Instruments Ltd.). For aggregate size determination in artificial seawater, nanoparticles at a concentration 100 mg/L were dispersed in filtered artificial seawater (salinity, 30 ± 0.5‰; pH, 8.0 ± 0.1; sea salt: Tropic Marine, Germany; filtered through 0.45 lm membrane filter). The aggregate size (average of three replicates) was determined using laser diffractometry (LD; LS 13 320 Series, Beckman Coulter Inc., Fullerton, USA). The reactive oxygen species generated by the nanoparticles were detected using electron spin resonance (ESR) spectroscopy at room temperature with the addition of a spin trap molecule [21]. Spin trap 5,5-Dimethyl-1-pyrroline N-oxide (DMPO) was purchased from Sigma–Aldrich Co., and the solution was prepared by adding 0.02 M DMPO to 1 mg/ml metal oxide nanoparticle
Table 1 Summary of nanoparticle characteristics and toxicity testing results in 0.9% w/v NaCl (SC) and artificial seawater (ASW), respectively, and toxicity testing results on the marine diatom in ASW. The toxicity endpoint is median inhibition concentration (IC50) after 72 h of exposure to the nanomaterial, and the IC50 values sharing with the same superscripted letter are statistically indifferent (based on overlapping of the 95% confidence intervals). IC50 values are given as mean value (n = 3), with confidence interval indicated in brackets. 1
Nanoparticle
Aggregation size in SC (lm), metal content (lg/L)
Aggregation size in ASW (lm), metal content (lg/L)
72-h IC50 (mg L
Control TiO2 SnO2 In2O3 Al2O3
–, <10 (Sn, In, Ti), <100 (Al) 0.6, <10 (Ti) 1.1, <10 (Sn) 0.1–0.5, <10 (In) 0.35 and 1.6, <100 (Al)
–, <10 (Sn, In, Ti), <100 (Al) 0.69, <10 (Ti) 0.42 and 1.8, <10 (Sn) 0.45 and 1.6, <10 (In) 0.67, <100 (Al)
– 353.3 (322.8–386.8)a 5200.0 (1389.0–19472.0)b 739.8 (580.0–943.7)c 2101.0 (1671.0–2642.0)b
) Skeletonema costatum
A.M.C. Ng et al. / Journal of Photochemistry and Photobiology B: Biology 151 (2015) 17–24
19
Fig. 2. ESR spectra of H2O2 with DMPO for different illumination conditions; ESR spectra of different nanoparticles with DMPO spin trap for UV illumination, 0.9% w/ v NaCl and ambient illumination, artificial seawater. The spectra have been vertically shifted for clarity.
Fig. 3. Absorption spectra of different nanoparticles. Emission spectrum of UV lamp used for illumination is also shown for comparison.
Table 2 Number of bacteria colonies for bacteria exposed to the different metal oxide nanoparticles in suspension. The bacteria count is an average value of three replicates. For each condition, mean values sharing the same superscripted letter are statistically indifferent (1-way analysis of variance followed by post hoc Tukey test, p < 0.05). Condition
Sample
Colony counts (Mean ± S.D.)
Dark
Control TiO2 SnO2 In2O3 Al2O3
406, 464, 602, 408, 544,
394, 560, 450, 376, 694,
378 526 574 336 544
(393 ± 14)ab (517 ± 49)abc (542 ± 81)bc (373 ± 36)a (594 ± 87)c
Survival rate (%) 100.0 131.6 137.9 94.9 151.1
Ambient
Control TiO2 SnO2 In2O3 Al2O3
556, 476, 592, 742, 612,
562, 614, 540, 420, 764,
564 594 742 622 720
(561 ± 4)a (561 ± 75)a (625 ± 105)a (595 ± 163)a (699 ± 78)a
100.0 100.0 111.4 106.1 124.6
UV
Control TiO2 SnO2 In2O3 Al2O3
417, 237, 381 (345 ± 95)a 0, 0, 0 (0 ± 0)b 192, 244, 169 (202 ± 38)c 703, 687, 769 (720 ± 43)d 372, 392, 417 (394 ± 23)a
100.0 0.0 58.6 208.7 114.2
Fig. 4. FTIR spectra of different nanoparticles before and after exposure to LPS. FTIR spectrum of pure LPS is also shown for comparison.
20
A.M.C. Ng et al. / Journal of Photochemistry and Photobiology B: Biology 151 (2015) 17–24
Fig. 5. Representative SEM images of bacterial cells exposed to different nanoparticles under UV illumination.
suspension in sodium chloride 0.9% w/v aqueous solution or artificial seawater. ESR measurements were performed using a Bruker EMX EPR spectrometer, with the following settings: microwave power 4 mW, microwave frequency 9.69 GHz, modulation frequency 100 kHz, modulation amplitude 1 G, time constant 41 ms, sweep time 42 s. The measurements were performed for ambient illumination (in artificial seawater) or under UV illumination (365 nm, Blak-RayÒ B-100 AP Lamp, 27 mW/cm2 for 2 min). For metal ion release analysis, nanoparticles were dispersed in different media for 24 h. The nanoparticles were then removed by centrifugation and filtering. The metal ion contents were analyzed by inductively coupled plasma mass spectrometry (ICP-MS) using Agilent 7500cx ICP-MS with EG020 standard.
2.2. Antibacterial activity testing Antibacterial testing was performed on a gram-negative bacterium Escherichia coli XL1-Blue (Stratagene), since E. coli is a commonly used organism for nanomaterial toxicity/antibacterial activity testing. Luria-Bertani (LB) broth (Affymetrix/USB) was used as the culture medium, and the incubation temperature was 37 °C. When the antibacterial activity test was done under UV illumination, 0.9% w/v Sodium Chloride (NaCl) solution was used as the dispersion medium for the bacteria [22]. 20 lL of bacteria suspension (109 CFU/ml) was added to 10 ml of nanoparticle
suspension in NaCl solution with a concentration of 0.1 mg/ml. The mixture was subjected to UV illumination (365 nm, Blak-RayÒ B-100 AP Lamp, 40 mW/cm2 measured by an optical power meter) for 20 min. Serial dilution of the bacterialnanoparticle suspension was performed and was then plated onto LB agar plates in triplicate. After incubating the agar plates at 37 °C over night (at least 16 h), number of colonies of bacteria formed was counted.
2.3. Ecotoxicity testing Ecotoxicity testing was done by performing acute algal growth inhibition tests on Skeletonema costatum (CCMP 1332; CCMP, USA) algae. The test procedures were modified from ASTM and OECD guidelines [23,24]. The nanomaterial dispersions for testing were prepared by dispersing 1 g of nanoparticles into 100 ml of filtered artificial seawater, followed by continuous stirring at room temperature for a minimum of 3 days before performing the experiments. For the experiments, initial algal concentrations of 105 cells/ml of S. costatum were exposed to different concentrations of the test chemicals dispersed in f2-Si medium in 24-well multidishes (Nunc; Naterville, USA) (n = 3–4). To evaluate possible shading effects (due to insolubility of metal oxides resulting in a cloudy appearance of the suspension), an experimental design adapted from Ref. [25] was utilized. The algae were incubated only
A.M.C. Ng et al. / Journal of Photochemistry and Photobiology B: Biology 151 (2015) 17–24
21
in the lower chamber of the stacked transparent multidish (with walls and undersides blackened so that only the light passing through the upper wells could reach the lower dish). For toxicity testing, the test compounds and the algae were mixed together in the lower plate wells, while the upper wells contained only f2-Si medium. For shading effects evaluation, the test materials and algae were separated in the upper and lower wells, respectively. Due to the tendency of nanoparticles to aggregate and settle down over time, the dishes were shaken on a titer plate shaker at 100 rpm for 15 min every 2 h. Incubation was performed in an environmental chamber at room temperature (25 ± 2 °C) for 72 h, with 16 h:8 h light:dark cycle. Average growth rate in the test period was calculated based on initial and final cell counts determined using a hemocytometer. 2.4. Characterization of the interaction between nanoparticles and bacterial and algae cells Scanning electron microscopy (SEM) was used to examine the interaction between the nanoparticles and cells. After exposure to nanoparticles, the bacterial cells were collected and fixed in 2.5% glutaraldehyde in cacodylate buffer (0.1 M sodium cacodylate–HCl buffer, pH 7.4) at 4 °C overnight. For the algal cells, fixation was performed in 2.5% glutaraldehyde in cacodylate buffer (0.1 M sodium cacodylate–HCl buffer, pH 7.4) at 4 °C for 72 h. The cells were then collected on a membrane (Millipore, pore size 0.8 lm), rinsed with cacodylate buffer and serially dehydrated with ethanol. The cells were dried by critical point drying method. Finally, a thin layer of Au was coated on the specimens, and SEM was performed using LEO 1530 FEG SEM. For the investigation of the attachment of lipopolysaccharide (LPS) on the nanoparticles surface, the nanoparticles were first added into an aqueous solution of LPS (from E. coli K-235, Sigma–Aldrich) and dispersed by sonication. The concentration of both LPS and nanoparticles was 1 mg/ml. The suspension was left in ambient for 2 h. The nanoparticles were then collected by centrifugation and cleaned with fresh de-ionized water for several times, followed by drying in ambient overnight. FTIR measurements were performed with PerkinElmer Spectrum Two IR spectrometer. For the interaction between nanoparticles and algal cell walls, the cell walls of S. costatum were extracted with a method similar to a paper reported [26]. The algal cells were firstly suspended in 1 mM CaCl2. Glass beads (0.3 mm in diameter) were added and mixed with the cells. The mixture was then vortexed at maximum speed for 6 min. The lysed cells were then collected by centrifugation at 600g for 5 min. The resultant cell pellet was then washed with 1 mM CaCl2 for several times until the supernatant was colorless. For FTIR measurement, the cell wall extract was resuspended in nanoparticle suspension in f2-Si with concentration of 1 mg/ml. The suspension mixture was left in ambient for 2 h. The nanoparticles were then collected by centrifugation, rinsed with de-ionized water and then dried in ambient overnight. The nanoparticles were mixed with KBr powder (infrared grade, Sigma–Aldrich) and pellets of the mixture were made. The measurement was performed on the pellets using PerkinElmer Spectrum Two IR spectrometer. 3. Results and discussion Fig. 1 shows the TEM images of the nanoparticles investigated. Their aggregation size in different media is summarized in Table 1. It can be observed that all the nanoparticles tend to aggregate into larger aggregates in both media investigated. The production of ROS under ambient and UV illumination was tested using electron spin resonance (ESR) spectroscopy and the obtained results are
Fig. 6. FTIR spectra of different nanoparticles before and after exposure to algal cell walls. FTIR spectrum of pure algal cell walls is also shown for comparison.
22
A.M.C. Ng et al. / Journal of Photochemistry and Photobiology B: Biology 151 (2015) 17–24
Fig. 7. Representative SEM images of algal cells exposed to different nanoparticles under ambient illumination.
shown in Fig. 2. The only peaks observed correspond to the OH radicals based on the comparison with the ESR spectra of a dilute H2O2 solution [21]. For the generation of ROS under illumination, we need to consider the bandgap of the materials. The illumination energy (3.4 eV) is smaller than the bandgap of alumina (>8 eV, although in nanoparticles it can be as small as 3.7 eV) [27] and tin oxide (3.6 eV) [28], and larger than the bandgap of indium oxide (3.0 eV) [29] and anatase titania (3.2 eV) [30]. From the absorption spectra of the nanoparticles shown in Fig. 3, it can be observed that all oxides except alumina exhibit similar absorption, indicating that for SnO2 nanoparticles bandgap smaller than the bulk material can be observed, likely due to the presence of defects. In the 0.9% w/v NaCl solution under UV illumination, ROS are produced by TiO2, SnO2, and In2O3, with TiO2 exhibiting lower signal compared to the other two oxides. However, we cannot observe any correlation between the ROS production in NaCl solution (Fig. 2), aggregation size in NaCl solution (Table 1), and toxicity to E. coli bacteria (Table 2). Under ambient light, there is no significant difference in E. coli counts among the different nanomaterials, but there are significantly different effects of the nanomaterials on the bacteria in dark or under UV irradiation (Table 2). Only TiO2 shows significant toxicity under UV illumination (0% survival rate), followed by SnO2 which exhibits some toxicity (60% survival rate), while other two oxides show no toxicity to E. coli. In addition, no significant metal ion release is observed from any of the
nanoparticles considered (see Table 1). Thus, toxicity mechanism of SnO2 toward E. coli does not involve photocatalysis or metal ion release, possible mechanism could be mechanical damage to the bacteria by nanoparticles. To examine this possibility, we have examined the interaction between the nanoparticles and the lipopolysaccharide (LPS), a molecule which is the major constituent of E. coli outer membrane [31]. Obtained results are shown in Fig. 4, while the SEM images of the bacteria exposed to nanoparticles are shown in Fig. 5. In the case of TiO2, we can observe cell damage (holes in the membrane) and a number of larger and small aggregates sticking to the cell surface. For SnO2, we can see loose aggregates with some particles sticking to the cells, and the number of visibly damaged cells is lower compared to TiO2, in agreement with the antibacterial testing results. In the case of In2O3, we also see loose aggregates but they are rarely sticking to cell surfaces, while for alumina we observe coarser aggregates which also do not seem to be attached to cell surfaces. From the FTIR spectra of nanoparticles exposed to LPS (Fig. 4), we can observe significant changes in the case of TiO2 in the region 1000–1500 cm 1 which mainly corresponds to the vibrations of carbohydrates and vibrations of PO2 group [13,31,32]. In particular, the vibrations at 1230 cm 1 and 1070 cm 1 can be attributed to the symmetric and asymmetric vibrations of PO2 [31,32]. Small changes in peak shapes in the same spectral region are observed for SnO2, and no significant
A.M.C. Ng et al. / Journal of Photochemistry and Photobiology B: Biology 151 (2015) 17–24
Fig. 8. Photographs of nanoparticle suspensions (1 mg/ml) in artificial seawater observed at different times.
23
changes in this region are found for the other two oxides. Therefore, toxicity appears to be related to the ability of nanoparticles to interact with the cell walls. These observations are in agreement with a previous study on toxicity of alumina, which found that the toxicity could be attributed to the interactions between nanoparticles and cells [13]. In the case of algae, we also observe more significant effects of TiO2 compared to other nanomaterials (Table 1). The lowest value of 72 h-IC50 is observed for TiO2, followed by In2O3, Al2O3, and SnO2. Low toxicities of In2O3, Al2O3, and SnO2 materials are in agreement with the data reported in other studies. Although there have been studies reporting effects of a-alumina on reduction of embryo growth and hatchability in freshwater snails [10] and the inhibition of the growth of algae [13], the majority of the studies on the toxicity of alumina indicate that it has moderate [18] to low toxicity [12,14,16,17,20], as documented in studies on yeast [12], different cell cultures [14,20], plants [16], E. coli bacteria [17], and oxidative stress in human blood serum [19]. Exceptions can possibly occur in case of inhalation where inflammatory response can occur [11], which is also the case for In2O3 and indium tin oxide nanoparticles (with indium oxide less toxic compared to indium tin oxide) [15]. Tin oxide was also previously found to have moderate toxicity, somewhat higher than alumina [18]. However, reasons for the low toxicity of these materials have not been conclusively identified. Since oxidative stress is commonly proposed as the cause of toxicity [9,19], we have examined ROS production in the algae culture medium since ROS production is medium dependent [33]. In f2-Si medium, all oxides except alumina are producing ROS, with the signal intensity following the trend TiO2 > In2O3 > SnO2. Thus, in this case as well the toxicity cannot be fully explained by ROS production, nor by aggregation sizes. Therefore, we have also examined the interaction of nanoparticles with algae cell walls and algae cells, with FTIR spectra shown in Fig. 6 and SEM images shown in Fig. 7. In this case, however, we can observe that all the particles exhibit significant changes upon exposure to algae cell walls, including those exhibiting very low toxicity, such as SnO2 and alumina, while TiO2 which displays the lowest IC50 value also exhibits the smallest changes in the FTIR spectra. For other oxides, changes in the spectra are observed not only in the vibrations of PO2 group [31,32] but we can also observe significant changes in the region 2800–2950 cm 1, corresponding to CH3 group vibrations [32], as well as vibrations of water (scissoring mode at 1630 cm 1) [30] and amide II bands at 1540 cm 1. However, from the SEM images (Fig. 7) we can observe that for In2O3 there are large aggregates formed and very few nanoparticles stick to the algae cell surface. Adhesion of the nanoparticles to the algal cells can be observed for TiO2 and SnO2 nanoparticles, but in this case as well SnO2 nanoparticles aggregate into looser agglomerations compared to TiO2 nanoparticles. The attachment of some small particles to cell walls does not appear to be necessarily detrimental to cell survival since SnO2 exhibits the highest IC50 value. Similar observation was made previously for some CeO2 samples [33]. It has been proposed that the attachment of nanoparticles to the cell walls is dependent on the aggregation tendency [13]. However, the aggregation tendency we observed in SEM images is not consistent with the measured aggregation sizes. This is due to the fact that aggregation size measurement characterizes agglomeration of nanoparticles in a medium at a certain specific point in time (in our study, immediately after dispersing the nanoparticles, to ensure reproducible results). However, during the course of toxicity testing, the particles may exhibit sedimentation at different rates which would affect the outcome of the experiment. It can be observed that TiO2
24
A.M.C. Ng et al. / Journal of Photochemistry and Photobiology B: Biology 151 (2015) 17–24
particles remain suspended in the medium for a long time after the other oxide particles exhibit significant sedimentation, as shown in Fig. 8. This phenomenon could be avoided by adding dispersing agents, but that could affect the test result due to artifacts induced by a dispersing agent [9] and the obtained results would not apply to environmentally relevant conditions. Therefore, we can conclude that electron microscopy is an essential tool for characterizing the interaction between the nanoparticles and the cells, since the true size and shape of aggregates relevant to the conditions under which the experiment is conducted can be observed.
[11]
[12]
[13]
[14]
[15]
4. Conclusions We have examined the toxicity of metal oxides expected to show low toxicity (SnO2, alumina, In2O3) and compared them to a material which is known to exhibit significant toxicity, anatase TiO2. We found that the studied three metal oxides indeed exhibit much lower toxicity compared to titania, and that their low toxicity can be attributed to their tendency to aggregate in solution which results in limited interaction between the nanoparticles and cell walls of the test organisms. Even in the cases where the material has the ability to strongly interact with the cell walls (SnO2, alumina, or In2O3 and algae cell walls), the aggregation in the test medium may prevent the significant interaction between the nanoparticles and the cells. The ability to interact with the cell walls (for example TiO2 and LPS), in particular involving the phosphate groups in the cell walls, and the slow sedimentation rate in the test medium are both essential to observe the toxicity of a nanomaterial.
[16]
Acknowledgments
[22]
Financial support from the project RGC CRF CityU6/CRF/08 and the Strategic Research Theme, University Development Fund, and Small Project Funding of the University of Hong Kong is acknowledged. The authors thank ALS Technichem (HK) Pty Ltd for metal ion release measurements.
[23]
[17]
[18]
[19]
[20]
[21]
[24] [25] [26]
References [1] P.J.J. Alvarez, V. Colvin, J. Lead, V. Stone, Research priorities to advance ecoresponsible nanotechnology, ACS Nano 3 (2009) 1616–1619. [2] H. Meng, T. Xia, S. George, A.E. Nel, A predictive toxicological paradigm for the safety assessment of nanomaterials, ACS Nano 3 (2009) 1620–1627. [3] A.S. Barnard, Computational strategies for predicting the potential risks associated with nanotechnology, Nanoscale 1 (2009) 89–95. [4] A.S. Barnard, How can ab initio simulations address risks in nanotech?, Nat Nanotechnol. 4 (2009) 332–335. [5] T. Puzyn, B. Rasulev, A. Gajewitz, X. Hu, T.P. Dasari, A. Michalkova, H.M. Hwang, A. Toropov, D. Leszczynska, J. Lesczynski, Using nano-QSAR to predict the cytotoxicity of metal oxide nanoparticles, Nat. Nanotechnol. 6 (2011) 175–178. [6] D. Fourches, D. Pu, C. Tassa, R. Weissleder, S.Y. Shaw, R.J. Mumper, A. Tropsha, Quantitative nanostructure–activity relationship modeling, ACS Nano 4 (2010) 5703–5712. [7] A.S. Barnard, One-to-one comparison of sunscreen efficacy, aesthetics and potential nanotoxicity, Nat. Nanotechnol. 5 (2010) 271–274. [8] E. Burello, A.P. Worth, Theoretical framework for predicting the oxidative stress potential of oxide nanoparticles, Nanotoxicology 5 (2011) 228–235. [9] A.B. Djurišic´, Y.H. Leung, A.M.C. Ng, X.Y. Xu, P.K.H. Lee, N. Degger, R.S.S. Wu, Toxicity of metal oxide nanoparticles: mechanisms, characterization, and avoiding experimental artefacts, Small 11 (2015) 26–44. [10] N. Musee, P.J. Oberholster, L. Sikhwivhilu, A.-M. Botha, The effects of engineered nanoparticles on survival, reproduction, and behaviour of
[27]
[28]
[29]
[30]
[31]
[32]
[33]
freshwater snail, Physa acuta (Draparnaud, 1805), Chemosphere 81 (2010) 1196–1203. C. Som, P. Wick, H. Krug, B. Nowack, Environmental and health effects of nanomaterials in nanotextiles and façade coatings, Environ. Int. 37 (2011) 1131–1142. C. García-Saucedo, J.A. Field, L. Otero-Gonzalez, R. Sierra-Álvarez, Low toxicity of HfO2, SiO2, Al2O3 and CeO2 nanoparticles to the yeast, Saccharomyces cerevisiae, J. Hazard. Mater. 192 (2011) 1572–1579. I.M. Sadiq, S. Pakrashi, N. Chandrasekaran, A. Mukherjee, Studies on toxicity of aluminum oxide (Al2O3) nanoparticles to microalgae species: Scenedesmus sp. and Chlorella sp., J. Nanopart. Res. 13 (2011) 3287–3299. D. Yoon, D. Woo, J.H. Kim, M.K. Kim, T. Kim, E.S. Hwang, S. Baik, Agglomeration, sedimentation, and cellular toxicity of alumina nanoparticles in cell culture medium, J. Nanopart. Res. 13 (2011) 2543–2551. K. Nagano, T. Nishizawa, Y. Eitaki, M. Ohnishi, T. Noguchi, H. Arito, S. Fukushima, Pulmonary toxicity in mice by 2- and 13-week inhalation exposures to indium-tin oxide and indium oxide aerosols, J. Occup. Health 53 (2011) 234–239. C.W. Lee, S. Mahendra, K. Zodrow, D. Li, Y.C. Tsai, J. Braam, P.J.J. Alvarez, Developmental phytotoxicity of metal oxide nanoparticles to Arabidopsis thaliana, Environ. Toxicol. Chem. 29 (2010) 669–675. I.M. Sadiq, B. Chowdhury, N. Chandrasekaran, A. Mukherjee, Antimicrobial sensitivity of Escherichia coli to alumina nanoparticles, Nanomed. Nanotechnol. Biol. Med. 5 (2009) 282–286. S. Lanone, F. Rogerieux, J. Geys, A. Dupont, E. Maillot-Marechal, J. Boczkowski, G. Lacroix, P. Hoet, Comparative toxicity of 24 manufactured nanoparticles in human alveolar epithelial and macrophage cell lines, Part. Fibre Toxicol. 6 (2009) 14. 1–12. E.J. Rogers, S.F. Hsieh, N. Organti, D. Schmidt, D. Bello, A high throughput in vitro analytical approach to screen for oxidative stress potential exerted by nanomaterials using a biologically relevant matrix: human blood serum, Toxicol. In Vitro 22 (2008) 1639–1647. J. Sun, S.C. Wang, D. Zhao, F.H. Hun, L. Weng, H. Liu, Cytotoxicity, permeability, and inflammation of metal oxide nanoparticles in human cardiac microvascular endothelial cells, Cell Biol. Toxicol. 27 (2011) 333–342. M. Kamibayashi, S. Oowada, H. Kameda, T. Okada, O. Inanami, S. Ohta, T. Ozawa, K. Makino, Y. Kotake, Synthesis and characterization of a practically better DEPMPO-type spin trap, 5-(2,2-dimethyl-1,3-propoxy cyclophosphoryl)-5-methyl-1-pyrroline N-oxide (CYPMPO), Free Radic. Res. 40 (2006) 1166–1172. T.P.T. Cushnie, P.K.J. Robertson, S. Officer, P.M. Pollard, C. McCullagh, J.M.C. Robertson, Variables to be considered when assessing the photocatalytic destruction of bacterial pathogens, Chemosphere 74 (2009) 1374–1378. Environmental Toxicology Standards, ASTM Standard E1218, Guide for Conducting Static Toxicity Tests with Microalgae, ASTM International, Philadelphia, PA, 1993. OECD, Test No. 201: Freshwater Alga and Cyanobacteria, Growth Inhibition Test, OECD Guidelines for the Testing of Chemicals, Paris, France, 1984. K. Hund-Rinke, M. Simon, Ecotoxic effect of photocatalytic active nanoparticles (TiO2) on algae and daphnids, Environ. Sci. Pollut. Res. 13 (2006) 225–232. K. Kröger, C. Bergsdorf, M. Sumper, A new calcium binding glycoprotein family constitutes a major diatom cell wall component, EMBO J. 13 (1994) 4676– 4683. I.L. Liu, P. Shen, S.Y. Chen, Formation of ultrafine and dense a-Al2O3 nanoparticles via kinetic phase change in a dynamic process, J. Nanopart. Res. 12 (2010) 2929–2940. A. Kar, M.A. Stroscio, M. Dutta, J. Kumari, M. Meyyapan, Growth and properties of tin oxide nanowires and the effect of annealing conditions, Semicond. Sci. Technol. 25 (2010) 024012. 1–9. P.D.C. King, T.D. Veal, Ch.Y. Wang, D.J. Payne, A. Bourlange, H. Zhang, G.R. Bell, V. Cimalla, O. Ambacher, R.G. Egdell, F. Bechstedt, C.F. McConville, Band gap, electronic structure, and surface electron accumulation of cubic and rhombohedral In2O3, Phys. Rev. B 79 (2009) 205211. 1–10. G.K. Mor, O.K. Varghese, M. Paulose, C.A. Grimes, Transparent highly ordered TiO2 nanotube arrays via anodization of titanium thin films, Adv. Funct. Mater. 15 (2005) 1291–1296. J. Kiwi, V. Nadtochenko, Evidence for the mechanism of photocatalytic degradation of the bacterial wall membrane at the TiO2 interface by ATRFTIR and laser kinetic spectroscopy, Langmuir 21 (2005) 4631–4641. A. Hsu, F.Z. Liu, Y.H. Leung, A.P.Y. Ma, A.B. Djurišic´, F.C.C. Leung, W.K. Chan, H.K. Lee, Is the effect of surface modifying molecules on antibacterial activity universal for a given materials?, Nanoscale 6 (2014) 10323–10331 Y.H. Leung, M.M.N. Yung, A.M.C. Ng, A.P.Y. Ma, S.W.Y. Wong, C.M.N. Chan, Y.H. Ng, A.B. Djurišic´, M.Y. Guo, M.T. Wong, F.C.C. Leung, W.K. Chan, K.M.Y. Leung, H.K. Lee, Toxicity of CeO2 nanoparticles – the effect of nanoparticle properties, J. Photochem. Photobiol. B 145 (2015) 48–59.