Environ. Sci. Technol. 2009, 43, 3933–3940
Ecotoxicity of Silver Nanoparticles on the Soil Nematode Caenorhabditis elegans Using Functional Ecotoxicogenomics JI-YEON ROH,† SANG JUN SIM,‡ JONGHEOP YI,§ KWANGSIK PARK,| KYU HYUCK CHUNG,⊥ DONG-YOUNG RYU,# AND J I N H E E C H O I * ,† Faculty of Environmental Engineering, College of Urban Science, University of Seoul, 90 Jeonnong-dong, Dongdaemun-gu, Seoul 130-743, Korea, Department of Chemical Engineering, Sungkyunkwan University, Changan-gu, Suwon 440-746, South Korea, School of Chemical and Biological Engineering, Institute of Chemical Processes, Seoul National University, Shinlim-dong, Kwanak-ku, Seoul 151-744, Korea, College of Pharmacy, Dongduk Women’s University, 23-1, Wolgok-dong, Seongbuk-gu, Seoul 136-714, Korea, College of Pharmacy, Sungkyunkwan University, 300 Cheoncheon dong, Jangan-gu, Suwon, Gyeonggi-do 440-746, Korea, and College of Veterinary Medicine, BK21 Program for Veterinary Science, Seoul National University, Shinlim-dong, Kwanak-ku, Seoul, 151-742, Korea
Received December 8, 2008. Revised manuscript received February 28, 2009. Accepted March 11, 2009.
In the present study, the ecotoxicity of silver nanoparticles (AgNPs) was investigated in Caenorhabditis elegans using survival, growth, and reproduction, as the ecotoxicological endpoints, as well as stress response gene expression. Whole genome microarray was used to screen global changes in C. elegans transcription profiles after AgNPs exposure, followed by quantitative analysis of selected genes. The integration of gene expression with organism and population level endpoints was investigated using C. elegans functional genomics tools, to test the ecotoxicological relevance of AgNPsinduced gene expression. AgNPs exerted considerable toxicity in C. elegans, most clearly as dramatically decreased reproduction potential. Increased expression of the superoxide dismutases-3 (sod-3) and abnormal dauer formation protein (daf12) genes with 0.1 and 0.5 mg/L of AgNPs exposures occurred concurrently with significant decreases in reproduction ability. Overall results of functional genomic studies using mutant analyses suggested that the sod-3 and daf-12 gene expressions may have been related to the AgNPs-induced reproductive failure in C. elegans and that oxidative stress may have been an important mechanism in AgNPs toxicity. * Corresponding author phone: 82-2-2210-5622; fax: 82-2-22442245; e-mail:
[email protected]. † Faculty of Environmental Engineering, University of Seoul. ‡ Department of Chemical Engineering, Sungkyunkwan University. § School of Chemical and Biological Engineering, Seoul National University. | Dongduk Women’s University. ⊥ College of Pharmacy, Sungkyunkwan University. # College of Veterinary Medicine, Seoul National University. 10.1021/es803477u CCC: $40.75
Published on Web 04/21/2009
2009 American Chemical Society
Introduction Silver nanoparticles (AgNPs) have a wide range of current and potential future applications, including spectrally selective coatings for solar energy absorption (1), chemical catalysts (2), surface-enhanced Raman scattering for imaging (3), and in particular, antimicrobial sterilization (4), which has made them one of the most commonly used nanomaterials (5). However, these same, effective, biocidal properties have the potential to adversely affect beneficial bacteria in the environment, especially in the soil and water (6). Investigation is increasing into the potential toxic mechanisms and long-term effects by which these nanomaterials could pose environmental risks through widespread production and use (7, 8). Widely used NPs, such as AgNPs, will most likely enter the environment and may produce a physiological response in certain organisms, possibly altering their fitness, and ultimately might change their populations or community densities. Research and literature regarding the ecotoxicity of NPs is still emerging and gaps exist in our knowledge of this area. Recent reports regarding NPs toxicity come from mammalian studies of respiratory exposure or in vitro assays with mammalian cells (9, 10) and ecotoxicological NP studies are increasing, with most of the available data involving freshwater species and species used for regulatory toxicology studies, such as Daphnia magna, Oncorhynchus mykiss, and Chlorella kessleri (11-14). Unfortunately, few studies have involved terrestrial organisms (15), even fewer specifically used AgNPs. Caenorhabditis elegans, a free-living nematode mainly found in the liquid phase of soils, is the first multicellular organism whose genome has been completely sequenced. The genome’s unexpectedly high level of conservation relative to the vertebrate genome makes C. elegans an ideal system for biological studies in areas such as genetics, molecular biology, and developmental biology, and functional genomic tools (gene knock out and RNAi) are available to readily study sublethal effects on the animal’s metabolism and physiology (16). Recently, this species has also been used as an animal model for ecotoxicological studies due to its abundance in soil ecosystems, convenient handling in the laboratory, and sensitivity to various types of stresses, utilizing various exposure media, including soil and water (17-21). In this study, AgNPs ecotoxicity on C. elegans was investigated using survival, growth, and reproduction, as well as, stress induced gene expression, as toxic endpoints. Global changes in the nematode transcription profile following AgNPs exposure were detected using whole genome microarrays and, subsequently, selected genes analyzed by quantitative real-time PCR (qRT-PCR). DNA microarray applications for detecting changes in gene expression profiles following pollutant exposure appear to provide a more comprehensive, and sensitive insight into toxicity than typical ecotoxicological parameters, such as mortality, growth, or reproduction (22, 23). This “exotoxicogenomic” strategy aims to develop new predictive models for identifying environmental or human health hazards and identify precise and rapid molecular biomarkers of chemical exposure. Microarray-based gene expression assays have the potential to describe the interactions of suites of genes comprising multiple metabolic pathways, possibly revealing system-wide perturbations in functions associated with stress. They are particularly interesting for the toxicity screening of new chemicals, such as NPs, with largely unknown modes of action. VOL. 43, NO. 10, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3933
In this study, the integration of gene expression with organismal and population level endpoints was undertaken using the available C. elegans functional genomic tools, to examine the ecotoxicological relevance of detected AgNPsinduced gene expression. Such functional genomic studies can provide experimental evidence of the causal relationships between the observed altered physiological indicators and gene expression. C. elegans functional genomic tools can aid in the assessment of the physiological impact of up- or downregulated gene expression and can provide indicators of the mode of action from the level of a single gene to the whole organism (21). As such analyses can be validated using in vivo mutational approaches (24, 25), here, the biological roles of AgNPs-responsive genes in the organisms’ defense against AgNPs toxicity were studied using the selected mutant strains, such as, mtl-2 (gk125), sod-3 (gk235), and daf-12 (rh286). To compare the toxicity of AgNPs to that of Ag ions, toxicity of Ag ion was also investigated in C. elegans using the same toxic endpoints used for AgNPs toxicity assay.
Materials and Methods Organisms. C. elegans were grown in Petri dishes on nematode growth medium (NGM) and fed OP50 strain Escherichia coli according to a standard protocol (26) and young adults (3 days) from an age-synchronized culture were used in all the experiments. To produce age-synchronized cultures, at 2 to 3 days, eggs from mature adults were isolated using a 10% hypochlorite solution, followed by a rinse with M9 buffer (27), and the eggs allowed to hatch on agar plates with a food source, resulting in synchronized adult worm populations. In addition to wild type animal (N2 var. Bristol), mtl-2 (gk125), sod-3 (gk235), and daf-12 (rh286) mutant strains were used. (Supporting Information Table 1). Wild type and mutant strains were provided by the Caenorhabditis Genetics Center at the University of Minnesota. AgNPs and Ag Ion Preparation and Exposure to C. elegans. AgNPs (size 2-fold; Supporting Infor-
FIGURE 1. Characterization of AgNPs in test media using TEM (transmission electron microscopy, A); darkfiled microscope (B); and DLS (dynamic light scattering) spectrometer (C).
FIGURE 2. Images of AgNPs in C. elegans using darkfiled microscope. C. elegans was exposed to AgNPs for 24 h and incorporation of AgNPs into the transparent worm body was investigated using an inverted, dark-field microscope. mation Table 4) and 26 of the upregulated and 685 of the downregulated genes were annotated. The functional analysis of these genes was determined by submitting these microarray results for interpretation using the Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases (Supporting Information Table 5). Thirteen GO categories, such as nuclear signaling and transport pathways, were found to be significantly represented within the upregulated genes following the 24 h AgNPs exposure, whereas 149 GOs, including behavior-related pathways such as response to stimulus, locomotive behavior, feeding behavior, and reproductive behavior pathways were significantly represented within the downregulated genes. According to the KEGG database, 15 up- and 44 downregulated AgNPs-responsive C. elegans gene probes have been mapped to known metabolic pathways. Quantitative Assessment of AgNPs-Responsive Gene Expression. qRT-PCR was performed for 16 genes chosen from the microarray data and for 10 genes chosen for their known involvement in stress response pathways (Figure 3A; Raw data, Supporting Information Tables 6). The 16 genes were heat shock proteins (hsp-16.1, hsp-16.2, hsp-16.41, and hsp-70), collagens (col-131 and col-101), P-glycoproteinrelated protein (pgp-14), ALG-1 interacting protein (ain-1), uncoordinated protein (unc-130), adaptin-related protein (apt-10), osmotic avoidance abnormal (osm-11), Abnormal cell LINeage (lin-10), conserved oligomeric golgi component (cogc-4), mammalian WNK-type protein kinase homologue (wnk-1), and two nonannotated genes, M162.5 and F02A9.4; and the 10 known stress-response genes were metallothioneins (mtl-1 and mtl-2), superoxide dismutases (sod-1 and sod-3), glutathione-S-transferase (gst-1), catalases (ctl-2, and ctl-3), aging alteration protein (age-1), and abnormal dauer formation proteins (daf-12 and daf-21). The qRT-PCR results showed that the selected genes from the microarray results tended to yield expression levels lower than those measured using the microarray and also revealed that, of the 26 tested genes, the expression of four genes (M162.5, mtl-2, sod-3, and daf-12) was upregulated by AgNPs exposure. In par-
ticular, the increased expression of mtl-2 and sod-3 occurred in a AgNPs concentration dependent manner. Of the four upregulated genes, functional analysis was carried out on the three genes (mtl-2, sod-3, and daf-12) with an identified biological role. Functional Analysis of AgNPs-Responsive Genes Using Mutant Strains. Organismal and population-relevant toxicity endpoints were examined after short (24 h) and long-term (72 h) exposures to AgNPs in the wild type (Table 1; Raw data, Supporting Information Table 6). Short-term survival and growth experiments were compared with the sensitivity of physiological level responses of C. elegans to AgNPs exposure with those at the molecular level (gene expression). Short-term testing only provided a snapshot of the physiological status, thus longer term testing was conducted for the effects on reproductive potential. Although AgNPs exposure did not affect the survival and growth of the wild type, reproduction was seriously affected, with the number of offspring per individual dramatically decreased (∼70% of the controls in 0.1 and 0.5 mg/L AgNPs). Increased expressions of sod-3 and daf-12 genes after a 24 h exposure to 0.1 and 0.5 mg/L of AgNPs occurred concurrently with the dramatic decrease in reproduction. A Pearson correlation test to identify any correlation between increased gene expressions and higher level effects (Supporting Information Table 7) showed that, indeed, the sod-3 and daf-12 gene expressions significantly correlated with the observed effect on reproduction (r2 ) 0.999 and 0.993, respectively; p < 0.01). C. elegans functional genomics tools were used to investigate the biological meaning and/or the ecophysiological consequences of the increased expression of mtl-2, sod-3, and daf-12 genes caused by AgNPs exposure. The survival, growth, and reproduction of mtl-2 (gk125), sod-3 (gk235), and daf-12 (rh286) mutant strains in response to AgNPs exposure were examined and compared with the wild type (Table 1). The mutant strains’ survival and growth response were not different from the wild type, but the reproductive responses of the mtl-2 (gk125) and sod-3 (gk235) mutants were less sensitive (∼40-60% less at 0.1 mg/L and ∼10% at 0.5 mg/L) to AgNPs exposure than the wild type, while the response of the daf-12 (rh286) mutant was similar to the wild type. Expression of Stress Genes in Response to AgNPs Exposure in Wild Type and Mutant C. elegans. To understand the mechanism of the different sensitivities of reproductive potential to AgNPs exposure in the mutant strains, the mRNA levels of 26 potential AgNPs-responsive genes were measured in the mutant strains (Figure 4; Raw data, Supporting Informaton Table 6) and compared to the wild type (Figure 3A). Predictably, the mtl-2, sod-3, and daf-12 genes were not detected in the mtl-2 (gk125), sod-3 (gk235), and daf-12 (rh286) mutants, respectively, but notably, the mtl-2 gene was not detected in the sod-3 (gk235) mutant. VOL. 43, NO. 10, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3935
FIGURE 3. Quantitative assessment of AgNPs-(A) and AgNO3- responsive gene expression in wild type C. elegans (B). The wild type of C. elegans were exposed to AgNPs for 24 h and qRT-PCR was performed for gene expression analysis. Densitometric values of stress-response gene expression were normalized using Actin mRNA. Data are presented in arbitrary unit compared to control (control ) 1; replicates number ) 3; mean ( standard error of the mean; *p < 0.05). Considerable differences in these genes’ expression patterns toward AgNPs exposure were found, most notably, in the levels of hsps, M162.5, col-131, fgp-14, ain-1, apt-10, lin-10, cogc-1, and wnk-1, which were significantly increased in the sod-3 (gk235) mutant, a response distinctively different from the wild type. In contrast, the expression of only a small number of genes was changed in the mtl-2 (gk125) and daf12(rh286) mutants. Comparison of Toxicity of AgNPs and Ag Ions. The toxicities of AgNPs and Ag ions in C. elegans were compared using the same toxic endpoints used to investigate AgNPs toxicity, which were stress-related gene expression (Figure 3B; raw data, Supporting Information Table 6), survival, growth, and reproduction (Table 1; raw data, Supporting Information Table 6). Ag ion exposure did not alter the worms’ mortality rate or growth, but it seriously decreased reproduction potential, similar to the effects of AgNPs exposure (Table 1). The degree of reproduction potential decrease, however, was more significant with AgNPs exposure than with Ag ions. Gene expression patterns from Ag ion exposure also showed different tendencies compared to AgNPs exposure. Here, Ag ion exposure at the two highest concentrations (0.1 and 0.5 mg/L) induced the expression of hsp gene groups (hsp-16.1, hsp-16.2, hsp-16.41, and hsp-70) more than 2-fold greater than the control. 3936
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 10, 2009
Discussion Preparation, dosing, and maintenance of NPs within the test medium are important factors when investigating the potential harmful effect of NP exposure in the environment and, among the various physicochemical properties of NPs, the aggregation process is an important factor influencing toxicity. Relatively uniform test solutions of NPs can be achieved by a chemical dispersant or physical methods (30), but these methods, particularly use of chemical dispersant, are ecologically irrelevant as NPs discharged to the environment are not likely to accompany or encounter effective dispersants. Thus, to be realistic and relevant, the present toxicity test design used only physical methods, such as, sonication, stirring, and filtration processes for creating homogeneous AgNPs dispersions; these methods were selected from our previous work for optimizing aqueous AgNPs dispersions (31). As AgNPs exist both individually and in aggregates, their states in the test media, as well as in C. elegans were characterized by image analysis, using TEM and dark field microscopy, to determine the size and the state of the AgNPs, as well as the uptake and the distribution of AgNPs in the C. elegans bodies (Figures 1 and 2). The TEM provided information on the size and shape of nanoparticles; however, it could not provide information on whether the nanoparticles existed in single or aggregated forms in the test medium, as the nanoparticles form aggregates when dried
TABLE 1. Ecotoxicological indicators investigated after exposure to AgNPs and AgNO3in wild type(N2) and mutant strains(mtl-2(gk125), sod-3(gk235) and daf-12(rh286)). Survival was investigated by counting the number of alive individuals compared to total introduced worms; growth was investigated by measuring the body length; reproduction was investigated by counting the number of offspring per individual. Results were expressed as the mean value compared to control (control = 1; replicates number = 5; mean ± standard error of mean) AgNPs (mg/L) exposure duration 24 h
parameters survival
growth
72 h
reproduction
strains
0.05
0.1
0.5
wild type(N2) mtl-2(gk125) sod-3(gk235) daf-12(rh286) wild type(N2) mtl-2(gk125) sod-3(gk235) daf-12(rh286) wild type(N2) mtl-2(gk125) sod-3(gk235) daf-12(rh286)
1.00 ( 0.000 1.00 ( 0.000 1.00 ( 0.000 1.00 ( 0.000 0.97 ( 0.018 0.98 ( 0.007 1.00 ( 0.021 0.99 ( 0.008 0.83 ( 0.052a 0.79 ( 0.019a 1.06 ( 0.053 0.54 ( 0.075b
1.00 ( 0.000 1.00 ( 0.000 1.00 ( 0.000 1.00 ( 0.000 0.96 ( 0.022 0.97 ( 0.015 1.00 ( 0.018 1.00 ( 0.006 0.32 ( 0.023b 0.79 ( 0.087a 0.88 ( 0.040a 0.35 ( 0.036b
0.98 ( 0.026 0.98 ( 0.000 0.98 ( 0.026 1.00 ( 0.000 0.94 ( 0.011 1.04 ( 0.007 1.02 ( 0.009 0.97 ( 0.003 0.32 ( 0.018b 0.41 ( 0.015b 0.40 ( 0.035b 0.25 ( 0.013b
AgNO3(mg/L) exposure duration
parameters
strains
0.05
0.1
0.5
24 h
survival growth reproduction
wild type(N2) wild type(N2) wild type(N2)
1.00 ( 0.000 0.99 ( 0.004 0.84 ( 0.119
1.00 ( 0.000 0.99 ( 0.000 0.51 ( 0.175b
1.00 ( 0.000 0.97 ( 0.016 0.40 ( 0.122b
72 h a
p < 0.05.
b
p < 0.01.
on the microscopic observation slide. The DLS result suggests that the AgNPs tended to exist as single particles in the test medium. Clear observation of AgNPs, afforded by the transparent C. elegans bodies, allowed measurement of NPs through light scattering without any pretreatment and thus, dark field microscopy can be used for the initial screening index for the study of nanotoxicity in a transparent animal model, such as C. elegans (32). The microarray profile of gene expression in AgNPs exposed C. elegans provided an overview of the worms’ molecular responses, revealing seriously affected transcriptional processes, with more genes downregulated than upregulated following a 24 h AgNPs exposure (1217 downversus 415 upregulated genes), and about 57% of the downregulated and 17% of the upregulated genes were annotated. The nonannotated genes may have represented expressed sequence tags (ESTs) that contain mainly untranslated regions or they may have been C. elegans specific genes. A major difficulty in analyzing expression data in this study was the relatively limited amount of gene annotation available, making the placement of the results in a biological context challenging. As C. elegans genomics in the ecotoxicological context continues to grow, the ability to formulate meaningful conclusions will be greatly enhanced by C. elegans functional genomic analyses. Here, GO and KEGG analyses revealed that, for most of the AgNPs-responsive genes, GO categories have not been assigned and the biochemical pathways not mapped. Only 162 (∼10%) of the 1632 genes whose expression significantly changed following a 24 h AgNPs exposure have been assigned GO terms and, similarly, only 59 (about 4%) of the gene probes have been mapped to known metabolic pathways included in the KEGG database. Although the effects of various physical and chemical stressors on individual gene expression has been intensively studied in C. elegans, the biological consequences of global transcriptional changes have been largely unexplored, particularly not changes relating to NP exposure. After exploring global transcriptional changes, further studies on the biological consequences of individual gene expression changes caused by AgNPs are warranted.
qRT-PCR assays revealed that increased expression was the most pronounced in the mtl-2, sod-3, and daf-12 genes (Figure 3A). Being metallic NPs, the induction of the mtl gene by AgNPs was expected (33, 34) increased expression of mtl-2, but not mtl-1 was also observed previously here regarding cadmium exposure (18). Oxidative stress as a toxic mechanism of AgNPs was reported in microorganisms (35) and reactive oxygen species (ROS) generated by AgNPs or Ag ions was reported to be responsible for observed bactericidal activity (36). Moreover, a recent study of the mechanisms of AgNPs toxicity using stress-specific, bioluminescent bacteria demonstrated toxicity of AgNPs via oxidative damage (37). In the present study, increased sod-3 gene expression in exposed worms was considered to confirm the involvement of oxidative stress in AgNPs-induced toxicity in this test system, as the enzymes involved in the breakdown of ROS play an important role in ROS related toxicity. Here, increased expression was observed in sod-3 (mitochondrial MnSOD), but not in sod-1 (cytosolic CuZnSOD) and the underlying mechanism for this, as well as, the involvement of the sod-3 gene in these worms’ reproductive failure, discussed below, merits further investigation. From egg through adults C. elegans has six life stages including an option for dauer formation, an alternative developmental stage of C. elegans, in which L1 and L2 stage animals have the option to divert their development under unfavorable environmental conditions. Signals for the dauer formation decision are integrated via daf genes, with two types of daf gens investigated in this study; daf-12 and daf-21. The former encodes a nuclear receptor regulating dauer formation, apparently involved in selecting stage-appropriate developmental programs as dafdefective mutants bypass dauer formation (38, 39). The latter encodes a member of the hsp90 family of molecular chaperones and daf-21 activity is required for larval development (40, 41). Increased daf-12 expression observed here suggested that AgNPs exposure may have acted as an environmental stressor that regulated C. elegans dauer formation, whereas unchanged daf-21 expression with AgNPs exposure suggested that AgNPs-induced alterations in reproduction may have been due to a lack of daf-21 activity. VOL. 43, NO. 10, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3937
FIGURE 4. Quantitative assessment of AgNPs-reponsive gene expression in mutant types C. elegans (mtl-2(gk125), sod-3(gk235), daf-21(rh286)). The mutant types of C. elegans were exposed to AgNPs for 24 h and qRT-PCR was performed for gene expression analysis. Densitometric values of stress-response gene expression were normalized using Actin mRNA. Data are presented in arbitrary unit compared to control (control ) 1; replicates number ) 3; mean ( standard error of the mean; *p < 0.05). Human health risk assessments use molecular biomarkers of human physiology (i.e., specific gene expression) to help understand individual health, but in ecological risk assessment, biomarkers are useful only when they can predict the effects on survival, growth, or reproduction. In most cases, survival, growth, or reproduction, which control population sizes of organisms, are widely accepted endpoints in ecotoxicity monitoring and ecological risk assessment, even though their utility for assessing NPs toxicity has yet to be 3938
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 10, 2009
convincingly demonstrated (42, 43). Linking molecular changes with relevant ecological responses will greatly improve the predictive powers of tests based on molecular responses and remains one of the great challenges in ecotoxicology (44). Gene expression profiles and adverse outcomes need to be conclusively linked at the individual and/or population level, but few NPs ecotoxicological studies have sought to demonstrate direct experimental relationships between molecular/biochemical effects and the conse-
quences at higher levels of biological organization. Using biochemical and organismal level endpoints, a recent ecotoxicity study of the effects of titanium dioxide NPs on terrestrial isopods, Porcellio scaber, showed that the activities of antioxidant enzymes, such as catalase and glutathione-S transferase, were affected, but higher level endpoints, including weight change and survival, were not affected (15). In the present study, among the ecotoxicological parameters tested (survival, growth, and reproduction), serious effects on reproduction were seen following 72 h AgNPs exposures (Table 1) and this method may provide an integrated approach to predicting population responses of nematodes to NPs. Environmental AgNPs exposure may seriously affect C. elegans populations as reproductive failure may induce significant disturbances in the population, most probably population density decreases. The present analysis examining correlations between gene expression and higher level effects revealed that the expressions of the sod-3 and daf-12 genes highly correlated with reproduction (Supporting Information), but statistical analyses only indicated a correlation, not a causal relationship. The most significant differences between the mutant strain responses and the wild type toward AgNPs exposure were in reproduction (Table 1). The less significant decreases in reproduction in the exposed mtl-2 (gk125) and sod-3 (gk235) mutants suggested that loss of gene function may actually enhance the worms’ reproductive potential. Many genes were upregulated in the sod-3 (gk235) mutant with AgNPs exposure (Figure 4), which may have indicated the importance of this gene in AgNPs-induced toxicity. Induced stress genes, including hsps, may have been a compensatory defense mechanism against oxidative stress in the absence of an important antioxidant enzyme gene, such as sod-3. Although the biocidal activity of AgNPs may be due to oxidative stress, but the toxic mechanism of AgNPs is largely unknown. Further studies of the mechanism by which the mitochondrial sod-3 gene is involved in these worms’ reproductive pathway are warranted to better understand oxidative stress-related toxicity with AgNPs exposure. From the present results, functional genomics using mutant strains appears to be an ideal tool for biomarker discovery in ecotoxicological research as it revealed the physiological meaning or function of observed altered gene expressions and thus helped in identifying the ecological relevance of certain molecular biomarkers. Moreover, in contrast to most functional genomics studies using cells or tissues with a focus on biochemical, physilogical pathways, this work represents a more holistic usage of a functional genomics method to identify biomarkers for enabling the monitoring of the overall fitness of an intact organism in an ecotoxicological context. There have been discussions regarding the comparative toxicity of AgNPs and Ag ions (45, 46), the latter’s bactericidal action having been studied previously (47, 48). The extremely small size of NPs have properties different from Ag ions, largely due to their relatively large surface area and related higher reactivity (49). Recently, Hwang et al. (37) demonstrated that AgNPs generate Ag ions and, subsequently, superoxide radicals, partially responsible for observed biocidal effect. The present study comparing the toxicity of AgNPs and Ag ions (Table 1, Figure 3B) suggested that AgNPs were slightly more toxic than Ag ions in terms of reproduction potential, and also it appeared that different mechanisms exerted toxicity with AgNPs and with Ag ions, as stress-related gene expression patterns were different between these two groups. As it appeared that the biocidal effects of AgNPs might be partially due to Ag ion generation, further studies of this aspect of toxicity are required. In conclusion, AgNPs exert considerable toxicity in C. elegans, particularly in reproduction potential. The results of functional genomics analyses suggested that the sod-3
and daf-12 gene expressions may have been related to AgNPsinduced reproductive failure in these worms and that oxidative stress was an important mechanism in AgNPsinduced toxicity. This study also suggested that the interpretation of microarray and subsequent quantitative gene expression data was greatly enhanced by linking them with organism and population level experiments.
Acknowledgments This work was accomplished through the generous support of the Ministry of Environment as “The Eco-technopia 21 project”.
Supporting Information Available Table 1 is a brief description on the mutant strains used in this study. Table 2 is an overview of the experimental scheme. Table 3 is a lists of the primers used for gene expression analysis using qRT-PCR. Table 4 is a list of the genes that were up- or down-regulated (g2 fold) following 24 h AgNPs exposure. Table 5 is a list displaying the GO categories significantly enriched with the up- or down-regulated genes and up- or down-regulated AgNPs-responsive genes that have been mapped to known metabolic pathways in the KEGG database. Table 6 is table displaying raw values of qRT-PCR (Figures 3-4) and ecotoxicity indicators used for data normalization (Table 1). Table 7 is a table displaying the pearson coefficient of correlation between gene expression and organism/population level indivicators. This material is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) Rand, R. B.; Peumans, P.; Forrest, S. R. Long-range absorption enhancement in organic tandem thin-film solar cells containing silver nanoclusters. J. Appl. Phys. 2004, 96, 7519–7526. (2) Zhai, H. J.; Sun, D. W.; Wang, H. S. Catalytic properties of silica/ silver nanocomposites. J. Nanosci. Nanotechnol. 2006, 6, 1968– 1972. (3) Yamamoto, S.; Watarai, H. Surface-enhanced Raman spectroscopy of dodecanethiol-bound silver nanoparticles at the liquid/ liquid interface. Langmuir. 2006, 22, 6562–6569. (4) Pal, S.; Tak, Y. K.; Song, J. M. Does the antibacterial activity of silver nanoparticles depend on the shape of the nanoparticle? A study of the Gram-negative bacterium Escherichia coli. Appl. Environ. Microbiol. 2007, 73, 1712–1720. (5) Maynard, A. D.; Michelson, E. the Nanotechnology Consumer Products Inventory; Woodrow Wilson International Center for Scholars: Washington DC, 2006; online at http://www. nanotechproject.org/consumerproducts. (6) Griffitt, R. J.; Luo, J.; Gao, J.; Bonzongo, J. C.; Barber, D. S. Effects of particle composition and species on toxicity of metallic nanomaterials in aquatic organisms. Environ. Toxicol. Chem. 2008, 27, 1972–1978. (7) Hoet, P. H. M.; Bru ¨ ske-Hohlfeld, I.; Salata, O. V. Nanoparticles - known and unknown health risks. J. Nanobiotechnol. 2004, 2, 2–26. (8) Handy, R. D.; Owen, R.; Valsami-Jones, E. The ecotoxicology of nanoparticles and nanomaterials: current status, knowledge gaps, challenges, and future needs. Ecotoxicology 2008, 17, 315–325. (9) Donaldson, K. Resolving the nanoparticles paradox. Nanomed 2006, 1, 229–234. (10) Cha, K. E.; Myung, H. Cytotoxic effects of nanoparticles assessed in vitro and in vivo. J. Microbiol. Biotechnol. 2007, 17, 1573– 1578. (11) Hund-Rinke, K.; Simon, M. Ecotoxic effect of photocatalytic active nanoparticles (TiO2) on algae and daphnids. Environ. Sci. Pollut. Res. Int. 2006, 13, 225–232. (12) Lovern, S. B.; Klaper, R. Daphnia magna mortality when exposed to titanium dioxide and fullerene (C60) nanoparticles. Environ. Toxicol. Chem. 2006, 25, 1132–1137. (13) Smith, C. J.; Shaw, B. J.; Handy, R. D. Toxicity of single walled carbon nanotubes to rainbow trout, (Oncorhynchus mykiss): respiratory toxicity, organ pathologies, and other physiological effects. Aquat. Toxicol. 2007, 82, 94–109. VOL. 43, NO. 10, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3939
(14) Fujiwara, K.; Suematsu, H.; Kiyomiya, E.; Aoki, M.; Sato, M.; Moritoki, N. Size-dependent toxicity of silica nano-particles to Chlorella kessleri. J. Environ. Sci. Health., Part A. Toxic/Hazard. Subst. Environ. Eng. 2008, 43, 1167–1173. (15) Jemec, A.; Drobne, D.; Remsˇkar, M.; Sepcˇic´, K.; Tisˇler, T. Effects of ingested nano-sized titanium dioxide on terrestrial isopods Porcellio scaber Environ. Toxicol. Chem. 2008, DOI: 10.1897/ 08-036.1. (16) Menzel, R.; Yeo, H. L.; Rienau, S.; Li, S.; Steinberg, C. E.; Stu ¨ rzenbaum, S. R. Cytochrome P450s and short-chain dehydrogenases mediate the toxicogenomic response of PCB52 in the nematode Caenorhabditis elegans. J. Mol. Biol. 2007, 370, 1–3. (17) Jonker, M. J.; Sweijen, R. A.; Kammenga, J. E. Toxicity of simple mixtures to the nematode Caenorhabditis elegans in relation to soil sorption. Environ. Toxicol. Chem. 2004, 23, 480–488. (18) Roh, J. Y.; Lee, J.; Choi, J. Assessment of stress-related gene expression in the heavy metal-exposed nematode Caenorhabditis elegans: a potential biomarker for metal-induced toxicity monitoring and environmental risk assessment. Environ. Toxicol. Chem. 2006, 25, 2946–2956. (19) Roh, J. Y.; Choi, J. Ecotoxicological evaluation of chlorpyrifos exposure on the nematode Caenorhabditis elegans. Ecotoxicol. Environ. Saf. 2008, 71, 483–489. (20) Roh, J. Y.; Jung, I. H.; Lee, J. Y.; Choi, J. Toxic effects of di(2ethylhexyl)phthalate on mortality, growth, reproduction and stress-related gene expression in the soil nematode Caenorhabditis elegans. Toxicology 2007, 237, 126–133. (21) Leung, M. C.; Williams, P. L.; Benedetto, A.; Au, C.; Helmcke, K. J.; Aschner, M.; Meyer, J. N. Caenorhabditis elegansan: An emerging model in biomedical and environmental toxicology. Toxicol. Sci. 2008, 106, 5–28. (22) Hamadeh, H. K.; Bushel, P.; Paules, R.; Afshari, C. A. Discovery in toxicology: mediation by gene expression array technology. J. Biochem. Mol. Toxicol. 2001, 15, 231–242. (23) Menzel, M.; Ro¨del, J.; Kulas, J.; Steinberg, C. E. CYP35: Xenobiotically induced gene expression in the nematode Caenorhabditis elegans. Arch. Biochem. Biophys. 2005, 438, 93– 102. (24) McElwee, J. J.; Schuster, E.; Blanc, E; Thomas, J. H.; Gems, D. Shared transcriptional signature in Caenorhabditis elegans Dauer larvae and long-lived daf-2 mutants implicates detoxification system in longevity assurance. J. Biol. Chem. 2004, 279, 44533–44543. (25) Yanase, S.; Ishii, N. Hyperoxia exposure induced hormesis decreases mitochondrial superoxide radical levels via Ins/IGF-1 signaling pathway in a long-lived age-1 mutant of Caenorhabditis elegans. J. Radiat. Res. 2008, 49, 211–218. (26) Brenner, S. The genetics of Caenorhabditis elegans. Genetics 1974, 77, 71–94. (27) Hitchcock, D. R.; Law, S. E.; Wu, J.; Williams, P. L. Determining toxicity trends in the ozonation of synthetic dye wastewaters using the nematode Caenorhabditis elegans. Arch. Environ. Contam. Toxicol. 1998, 34, 259–264. (28) Williams, P. L.; Dusenbery, D. B. Aquatic toxicity testing using the nematode Caenorhabditis elegans. Environ. Toxicol. Chem. 1990, 9, 1285–1290. (29) Dhawan, R.; Dusenbery, D. B.; Williams, P. L. Comparison of lethality, reproduction, and behavior as toxicological endpoints in the nematode Caenorhabditis elegans. J. Toxicol. Environ. Health., Part Aq 1999, 58, 451–462. (30) Crane, M.; Handy, R. D.; Garrod, J.; Owen, R. Ecotoxicity test methods and environmental hazard assessment for engineered nanoparticles. Ecotoxicology 2008, 17, 421–437. (31) Bae, E. J.; Lee, J. J.; Kim, Y. H.; Choi, K. H.; Yi, J. H. Physicochemical Analysis of Silver Nanoparticle Suspension for Nanotoxicity Evaluation. J. Nanopart. Res. 2009, . submitted.
3940
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 10, 2009
(32) Hu, M.; Novo, C.; Funston, A.; Wang, H.; Staleva, H.; Zou, S.; Mulvaney, P.; Xia, Y.; Hartland, G. V. Dark-field microscopy studies of single metal nanoparticles: understanding the factors that influence the linewidth of the localized surface plasmon resonance. J. Mater. Chem. 2008, 18, 1949–1960. (33) Swain, S. C.; Keusekotten, K.; Baumeister, R.; Stu ¨rzenbaum, S. R. C. elegans metallothioneins: new insights into the phenotypic effects of cadmium toxicosis. J. Mol. Biol. 2004, 341, 951–959. (34) Williams, P. L.; Ma, H.; Glenn, T. C.; Jagoe, C. H.; Jones, K. L. A transgenic strain of the nematode Caenorhabditis elegans as a biomonitor for heavy metal contamination. Environ. Toxicol. Chem. 2009, DOI: 10.1897/08-496.1. (35) Choi, O.; Hu, Z. Size dependent and reactive oxygen species related nanosilver toxicity to nitrifying bacteria. Environ. Sci. Technol. 2008, 42, 4583–4588. (36) Kim, D. W.; Hong, G. H.; Lee, H. H.; Choi, S. H.; Chun, B. G.; Won, C. K.; Hwang, I. K.; Won, M. H. Effect of colloidal silver against the cytotoxicity of hydrogen peroxide and naphthazarin on primary cultured cortical astrocytes. Int. J. Neurosci 2007, 117, 387–400. (37) Hwang, E. T.; Lee, J. H.; Chae, Y. J.; Kim, Y. S.; Kim, B. C.; Sang, B. I.; Gu, M. B. Analysis of the toxic mode of action of silver nanoparticles using stress-specific bioluminescent bacteria. Small 2008, 4, 746–750. (38) Riddle, D. L.; Albert, P. S. Genetic and Environmental Regulation of Dauer Larva Development in C. elegans II; Cold Spring Harbor Laboratory Press.: New York, 1997; pp. 739-768. (39) Antebi, A.; Culotti, J. G.; Hedgecock, E. M. daf-12 regulates developmental age and the dauer alternative in Caenorhabditis elegans. Development 1998, 125, 1191–1205. (40) Birnby, D. A.; Link, E. L.; Vowels, J. J.; Tian, H. Z.; Colacurcio, P. L.; Thomas, J. H. A transmembrane guanylyl cyclase (DAF11) and Hsp90 (DAF-21) regulate a common set of chemosensory behaviors in Caenorhabditis elegans. Genetics 2000, 155, 85– 104. (41) Inoue, T.; Hirata, K.; Kuwana, Y.; Fujita, M.; Miwa, J.; Roy, R.; Yamaguchi, Y. Cell cycle control by daf-21/Hsp90 at the first meiotic prophase/metaphase boundary during oogenesis in Caenorhabditis elegans. Dev., Growth. Differ. 2006, 48, 25–32. (42) Newman, M. C. Population Ecotoxicology; John Wiley and Sons Inc: Chichester, UK, 2001. (43) Crane, M.; Handy, R. D.; Garrod, J.; Owen, R. Ecotoxicity test methods and environmental hazard assessment for engineered nanoparticles. Ecotoxicology 2008, 17, 421–437. (44) Desalvo, M. K.; Voolstra, C. R.; Sunagawa, S.; Schwarz, J. A.; Stillman, J. H.; Coffroth, M. A.; Szmant, A. M.; Medina, M. Differential gene expression during thermal stress and bleaching in the Caribbean coral Montastraea faveolata. Mol. Ecol. 2008, 17, 3952–3971. (45) Ji, J. H.; Jung, J. H.; Kim, S. S.; Yoon, J. U.; Park, J. D.; Choi, B. S.; Chung, Y. H.; Kwon, I. H.; Jeong, J.; Han, B. S.; Shin, J. H.; Sung, J. H.; Song, K. S.; Yu, I. J. Twenty-eight-day inhalation toxicity study of silver nanoparticles in Sprague-Dawley rats. Inhalation Toxicol. 2007, 19, 857–871. (46) Hidalgo, E.; Domı´nguez, C. Study of cytotoxicity mechanisms of silver nitrate in human dermal fibroblasts. Toxicol. Lett. 1998, 98, 169–179. (47) Clement, J. L.; Jarrett, P. S. Antibacterial silver. Met.-Based Drugs 1994, 1, 467–482. (48) Silver, S. Bacterial resistances to toxic metal ions-A review. Gene 1996, 179, 9–19. (49) Asharani, P. V.; Wu, Y. L.; Gong, Z.; Valiyaveettil, S. Toxicity of silver nanoparticles in zebrafish models. Nanotechnology 2008, 19, 255102–255110.
ES803477U
