Bull Environ Contam Toxicol DOI 10.1007/s00128-015-1485-9
The Effect of Metal Oxide Nanoparticles on Functional Bacteria and Metabolic Profiles in Agricultural Soil Hankui Chai • Jun Yao • Jingjing Sun • Chi Zhang • Wenjuan Liu • Mijia Zhu • Brunello Ceccanti
Received: 23 June 2014 / Accepted: 26 January 2015 Ó Springer Science+Business Media New York 2015
Abstract A significant knowledge gap in nanotechnology is the absence of standardized protocols for examining the effect of engineered nanoparticles on soil microorganisms. In this study, agricultural soil was exposed to ZnO, SiO2, TiO2 and CeO2 nanoparticles at 1 mg g-1. The toxicity effect was evaluated by thermal metabolism, the abundance of functional bacteria and enzymatic activity. ZnO and CeO2 nanoparticles were observed to hinder thermogenic metabolism, reduce numbers of soil Azotobacter, P-solubilizing and K-solubilizing bacteria and inhibit enzymatic activities. TiO2 nanoparticles reduced the abundance of functional bacteria and enzymatic activity. SiO2 nanoparticles slightly boosted the soil microbial activity. Pearson’s correlation analysis showed that thermodynamic parameters had a strong correlation with abundance of functional bacteria and enzymatic activity. These findings demonstrated that the combined approach of monitoring thermal metabolism, functional bacteria and enzymatic activity is feasible for testing the ecotoxicity of nanoparticles on agricultural soil. Keywords Metal oxide nanoparticles Metabolism Functional bacteria Enzymatic activity Toxicity
H. Chai J. Yao (&) J. Sun C. Zhang W. Liu M. Zhu School of Civil and Environmental Engineering, National ‘‘International Cooperation Based on Environment and Energy’’, University of Science and Technology Beijing, 30 Xueyuan Road, Haidian District, 100083 Beijing, People’s Republic of China e-mail: [email protected] B. Ceccanti Institute of Ecosystem Studies (ISE), National Research Council (CNR), Via Giuseppe Moruzzi 1, 56124 Pisa, Italy
Products that incorporate engineered nanoparticles are increasingly entering the market. It is estimated that the commercial value of this part of the agricultural market would amount to three trillion dollars by 2020 (Servin et al. 2013). Engineered nanoparticles would inevitably be introduced into the soil matrix with application of biosolids (Shah et al. 2014). More and more people are becoming aware of the potential ecotoxicity of engineered nanoparticles. Four types of metal oxide nanoparticles (ZnO, SiO2, TiO2 and CeO2) are being widely applied in textiles, aerospace, sensors, packaging, cosmetics and sunscreens (Keller et al. 2013). Proliferation of their use will inevitably lead to uncontrolled release of nanoparticles into soil with the potential to adversely affect soil microorganisms (Keller et al. 2013; Whiteley et al. 2013; Wiesner et al. 2011). It was predicted that soil ecosystems would be the largest recipients of engineered nanoparticles after water and air. Systematic screening of the effects of nanoparticles on the soil environment has so far resulted in scarce information. Soil microorganisms play vital roles in biogeochemical cycling and organic matter dynamics. The toxicity, metabolism and transport of nanoparticles are related to the humic acid, organic matter, pH and ionic strengths of the host soil (Tourinho et al. 2012; Wang et al. 2011; Zhu et al. 2014). Therefore, it is very important to understand the behavior of nanoparticles in soil and to evaluate the risk of nanoparticles in arable soil ecosystems or other real environmental scenarios (Shrestha et al. 2013). Free Ce was observed in soil spiked with CeO2 nanoparticles, indicating that these elements were taken up by soil microorganisms (Vittori Antisari et al. 2013). More recent work by Priester et al. (2012) showed that ZnO and CeO2 nanoparticles reduced the soil nitrogen-fixing capacity and amassed in soybean tissue (Priester et al. 2012). TiO2 and ZnO nanoparticles significantly decreased
123
Bull Environ Contam Toxicol
biodiversity of soil microorganisms in a dose-dependent manner (Ge et al. 2012). Some studies measured the effects of metal nanoparticles on the standard artificial OECD soil, showing nanoparticles affected the microbial diversity and exerted toxic effects on soil invertebrates (Nogueira et al. 2012; Tourinho et al. 2012). Abundant functional bacteria thrive in soil and are crucial to the ecosystem through debris decomposition and nutrient cycling. In particular, the Azotobacter may drive the soil nitrogen fixation rate; P- and K-solubilizing bacteria may boost uptake of essential elements by plants through solubilizing P from organic debris and releasing K from silicate in soil. These bacteria may be the most vulnerable to toxic substances in the environment. Establishment of principles and test procedures for comprehensive understanding of the effect of nanoparticles on soil microorganism is essential (Rahman et al. 2013). It was shown that any xenobiotic entering into the soil environment would result in a shift of the thermogenic metabolic characteristics of the total soil microbial activity and biomass (Sandy et al. 2010). If nanoparticles have toxic effects on soil microorganisms, they would reduce biomass and thermodynamic activity. Essential nutrients such as N, P and potassium (K) are required for soil fertility and productivity. Moreover, the beneficial effect of microorganism is highly dependent on the abundance and diversity of functional bacteria in soil. Soil enzyme activity can be disturbed through the toxic effects of contamination on soil microflora and could be a reliable indicator of the current microbiological state. A bacterial cell usually contains approximately 1,000 different enzymes; many of these associated with the cell membrane. Information on the toxic effects of contaminants can be provided by evaluating soil enzyme activity involving nutrient cycling and organic matter dynamics (Wang et al. 2011). We hypothesized that the microcalorimetry technique in conjunction with measurement of other specific enzyme activities would be an effective toolkit to reflect the toxic effect of nanoparticles on soil microorganisms. To test this hypothesis, a combined approach consisting of monitoring of thermal metabolism, abundance of functional microbial groups and soil enzyme activities was used to characterize the effect of metal oxide nanoparticles (ZnO, SiO2, TiO2 and CeO2) on agricultural soil.
Materials and Methods The soil samples were collected from arable maize fields located in the Agricultural University of Hebei, China (38°460 5100 N, 115°330 3600 E). Initially, the soil samples were air dried and sieved through 2-mm mesh. The basic physicochemical properties of the soil were as follows: pH 7.46,
123
organic matter content 16.8 g kg-1, total N 0.82 mg kg-1, available P 13.3 mg kg-1 and available K 96.43 mg kg-1. Four types of nanoparticles were obtained from Boyugaoke Bio-Tech Co., China, including ZnO (15 nm, 15–25 m2 g-1), SiO2 (Amorphous, 9 nm, 50–60 m2 g-1), TiO2 (Anatase, 10 nm, [16 m2 g-1) and CeO2 (10 nm, 50–60 m2 g-1), with purity [99.9 %. Figure 1 shows the transmission electron microscopy (TEM) images of the four nanoparticles. To minimize the aggregation of nanoparticles, ZnO, SiO2, TiO2 and CeO2 nanoparticle powders were dispersed in an ultrasonic water bath at 25°C for 1 h. The above dispersions were added to soil (50 g dry weight) with a pipette and continuously stirred to achieve the homogeneous exposure dose at 1 mg g-1. Control soil had the same aliquot of water added without nanoparticles. All treatments were repeated in triplicate. To obtain the stabilization of microbial activity, each treatment was prepared in triplicate and incubated for 30 days at 70 % of water holding capacity and at 28°C. An isothermal TAM III microcalorimeter (TA instruments, New Castle, DE, USA) equipped with 12 independent channels was used to record continuous and simultaneous thermogenesis of soil microbial metabolism. Four-milliliter stainless steel ampoules sterilized by autoclaving were used in each channel. Each ampoule spiked with treatment soil (1 g) was placed inside the measuring position of the microcalorimeter. Nutrient solution (200 lL) containing 5.0 mg glucose and 5.0 mg ammonium sulfate was added in each ampoule to support the growth of soil microorganisms (Wang et al. 2010). Data were exported for further analysis and graphic presentation through Origin software 8.0 (OriginLab, Northampton, MA, USA). Thermogenic parameters, such as total heat output (Qtotal), growth rate constant (k), the value of peak height (Pmax) and corresponding time (Tmax), were calculated from each curve. The Qtotal was determined by the integration of each power–time curve. The k was calculated based on the heat output being proportional to the rate of bacterial growth. In the exponential growth phase, Pt which is equal to the first derivative of Qtotal with time (t), obeys the kinetics: Pt = P0 ekt. Pt is the output of power at time T, and the P0 is the power at the initial exponential growth stage. Enumeration of soil Azotobacter, P- and K-solubilizing bacteria were counted by suspension dilution methods on differentiating agar media. Ashby nitrogen-free culture medium was used to provide quantitative estimation of nitrogen-fixing bacteria (per L: mannitol 10 g, KH2PO4 0.2 g, NaCl 0.2 g, MgSO47H2O 0.2 g, CaCO3 5 g, CaSO4 0.1 g, agar 15 g, pH 7.0). P-solubilizing bacteria were counted using the following medium composition (per L: glucose 10.0 g, NaCl 0.3 g, (NH4)2SO4 0.5 g, KCl 0.3 g, FeSO47H2O 0.3 g, MgSO47H2O 0.3 g, MnSO44H2O 0.03 g, CaCO3 5.0 g, lecithin 0.2 g, agar 15 g, pH 7.2).
Bull Environ Contam Toxicol
Fig. 1 Transmission electron microscopy (TEM) images of four metal oxide nanoparticles: a ZnO; b SiO2; c TiO2; d CeO2
Numbers of K-dissolving bacteria were determined on the following medium (per L: sucrose 5.0 g, CaCO3 0.1 g, Na2HPO4 2 g, FeCl3 0.005 g, soil minerals 1 g, MgSO4 7H2O 0.5 g, agar 15.0 g, pH 7.5). The above media were sterilized at 120°C for 30 min. All the plates were incubated at 28°C for 7 days. Soil urease activity was determined by absorbance of ammonium released from urea hydrolysis at 578 nm (Wang et al. 2010). Catalase activity was determined by back-titrating residual peroxide with 0.1 mol L-1 KMnO4 solution (Du et al. 2011). Fluorescein diacetate hydrolysis activity was measured by absorbance of fluorescence generated by fluorescein at 490 nm (Yang et al. 2013). The availability of nanoparticles in soil was determined by CHCl3-labile metals and water extraction (Vittori Antisari et al. 2013). Subsamples (4 g) were divided into two portions. One sample portion (2 g) was fumigated with ethanol-free CHCl3 for 24 h at 25°C to lyse bacterial cells. After the fumigation step, the samples were shaken by
adding 5 mL 1 mol L-1 NH4NO3 for 1 h and filtered through a 0.45 lm polysulfone filter. After filtration, 0.1 mL HNO3 solution (1:10 v/v, trace-metal) was added to the extraction which was then stored at 4°C. The nonfumigated portion was extracted using the same procedure as the fumigated portion. The CHCl3-labile metals were calculated as the concentration in the fumigated portion minus the concentration in the non-fumigated portion. Water extraction was extracted with Milli-Q water. The soil samples were shaken with Milli-Q water (1:10 w/v) for 24 h. The extract was obtained by centrifugation for 15 min at 2,000 rpm and then filtered through a 0.45 lm polysulfone filter. The total concentrations of Zn, Si, Ti and Ce were determined by ICP-MS. Statistical analyses were carried out using SPSS 13.0 (SPSS International, USA). Differences between treatments and controls were tested with one-way analysis of variance. Duncan’s test was applied to determine the significant differences at the level of p \ 0.05. Correlation
123
Bull Environ Contam Toxicol
analysis was performed with two-tailed Pearson test, with p \ 0.05 and p \ 0.01 as the significant values.
Results and Discussions The microcalorimetry assay was used to study soil microbial activity in the presence and absence of four types of metal oxide nanoparticles (Fig. 2). As seen from the power–time curves, some differences were observed in nanoparticle treatments in comparison to the control. Soils spiked with ZnO and CeO2 nanoparticles exhibited low thermodynamic activity. More specifically, the initial power value of ZnO nanoparticles was significantly lower than the other treatments. Soils spiked with SiO2 and TiO2 nanoparticles had similar thermodynamic characteristics to the control. Table 1 presents the thermal parameters of all treatments to compare the toxic effect of the four metal oxide nanoparticles. ZnO and CeO2 nanoparticles had lower values of Pmax (852 lW and 898 lW for ZnO and CeO2 nanoparticles, respectively (p \ 0.05)) than the other treatments and the control. The Pmax of TiO2 and SiO2 nanoparticles was slightly higher than the control. The low k values for ZnO and CeO2 nanoparticles indicated that the soil microbial metabolism was inhibited at the exponential growth stage (p \ 0.05). The superoxide and reactive oxygen species generated by nanoparticles might have reduced the microbial biomass and activity. It has been demonstrated that TiO2 and ZnO nanoparticles reduced the substrate induced respiration of grassland soil (Ge et al. 2011). Thermogenesis involves many enzymes that drive glycolysis, the Krebs cycle and the electron transport chain. This variability in microbial activities of the soil might be caused by direct oxidative stress on enzymes (Thill et al. 2006). 1200 control
ZnO
SiO2
TiO2
CeO2
20
25
Power (µW)
1000 800 600 400 200 0 5
10
15
Time (h) Fig. 2 Power–time curves of soil microorganisms spiked with metal oxide nanoparticles. Values are means of three replicates
123
The total plate counts and enzymatic activity of soil samples exposed to different nanoparticles are displayed in Table 2. ZnO, CeO2 and TiO2 nanoparticles significantly reduced the numbers of Azotobacter, P- and K-solubilizing bacteria (p \ 0.05). No significant difference was observed for total plate counts between the SiO2 nanoparticles and the control. Reductions in enzymatic activities were more significant in response to ZnO, CeO2 and TiO2 nanoparticle treatments (p \ 0.05) than for SiO2 nanoparticles, compared with the control. This was probably a consequence of microbial cell absorption or uptake of free metals released by nanoparticles in soil. We observed Zn, Ce and Ti, but not Si, in the CHCl3-labile metal extractions and the H2O extractions. A significant part of the nanoparticles might be transferred or absorbed into microbial cells and ions might be released in soil. Nitrogen fixation of soybean nodules was reduced in organic farm soil contaminated with ZnO and CeO2 nanoparticles (Priester et al. 2012). It was shown that cell growth and nitrogen fixing capacity of Anabaena variabilis was hampered by TiO2 nanoparticles in the aquatic environment (Cherchi and Gu 2010). DNA-based fingerprinting analyses showed that TiO2 and ZnO nanoparticles induced a decrease in taxa associated with nitrogen cycling (Rhizobiales, Bradyrhizobiaceae, Bradyrhizobium) (Ge et al. 2012). Similar studies confirmed that Ag nanoparticles exerted adverse effects on nitrogen-cycling bacteria in agricultural soil (Yang et al. 2013). TiO2 and ZnO nanoparticles significantly decreased protease, catalase, and peroxidase activities in wheat soil, which might be due to dissolved ions in soil (Du et al. 2011). b-Glucosidase and phosphatase activities were inhibited in sandy loam treated with multi-walled carbon nanotubes (Chung et al. 2011). Pearson’s correlation analysis showed that thermodynamic parameters were strongly correlated with biochemical properties under the impact of nanoparticles (Table 3). Catalase activity and K-solubilizing bacteria were significantly correlated with the Pmax and the k. In the microenvironment of microcalorimeter ampoules, oxygen and nutrients are the limiting factors for the growth and reproduction of microorganisms. Therefore, heat output from soil samples is a 1:1 molar relationship with microbial growth rate. For this reason, the Pmax and k values can represent the soil microbial activity. The change in k and the Pmax of each experiment indicated the change in soil microbial activity. Soil enzyme activity involved in nutrient cycles is highly dependent on the total microbial activity and related to the abundance of functional bacteria. In this study, we found positive correlations for each thermodynamic parameter, and we also found positive correlations for functional bacteria and enzymatic activities. These findings indicated that the combined method is feasible to assess the effects of nanoparticles on the soil.
Bull Environ Contam Toxicol Table 1 Thermogenesis parameters of soil treated with different metal oxide nanoparticles Qtotala (J g-1)
Tmaxb (h)
Kd (h-1)
Pmaxc (lW)
Control
25 ± 1.5A
11 ± 0.3 A
985 ± 17.8 A
0.494 ± 0.020 A
ZnO
23 ± 1.2 A
12 ± 0.2 B
852 ± 21.3 D
0.419 ± 0.025 C
SiO2
25 ± 1.1 A
11 ± 0.3 A
TiO2
24 ± 1.2 A
11 ± 0.5 A
991 ± 2.3 AB
0.498 ± 0.026 A
CeO2
18 ± 0.5 B
10 ± 0.3 A
898 ± 20.3 C
0.456 ± 0.017 B
1,019 ± 2.0 B
0.526 ± 0.011 A
Values are means of three replicates ±SD, values labeled with different letters indicate statistically significant differences (p \ 0.05, Duncan’s test), a Total heat output calculated from power–time curve; b The corresponding time of the height of each peak; c The height of the peak; d The microbial growth rate constant
Table 2 Total plate counts of functional bacteria and enzyme activities, and the availability of nanoparticles in soil Control
ZnO
SiO2
TiO2
CeO2
Azotobacter (logCFU)
6.23 ± 0.027 A
6.09 ± 0.029 D
6.2 ± 0.017 AB
6.16 ± 0.012 BC
6.13 ± 0.030 CD
P-bacteria (logCFU)
6.55 ± 0.060 A
6.07 ± 0.091 D
6.54 ± 0.026 AB
6.28 ± 0.074 D
6.41 ± 0.028 BD
K-bacteria (logCFU)
5.52 ± 0.042 A
5.22 ± 0.033 C
5.53 ± 0.022 A
5.42 ± 0.020 B
5.35 ± 0.066 B
Urease (mg NH4?-N g-1 soil 24 h-1)
0.88 ± 0.015 A
0.53 ± 0.037 C
0.84 ± 0.007 B
0.82 ± 0.006 B
0.81 ± 0.004 B
Catalase (mL KMnO4 g
-1
-1
1.15 ± 0.010 A
1.01 ± 0.030 B
1.16 ± 0.003 A
1.13 ± 0.010 A
1.06 ± 0.020 B
FDA (lg Fluorescein g-1 soil 0.5 h-1) 21.35 ± 0.877 A
soil 0.5 h )
13.12 ± 0.018 C
21.05 ± 0.673 A
15.61 ± 1.905 B
13.41 ± 0.032 B
CHCl3-labile (lg kg-1)
ND
28.51 ± 0.907 A
ND
5.48 ± 0.991 C
12.96 ± 0.536 B
–
376.33 ± 3.512 A
ND
12.52 ± 1.284 C
12.96 ± 0.537 B
H2O-extraction (lg L-1)
FDA indicates the fluorescein diacetate hydrolysis activity. ND is not detectable (lower than the detection limit). The values of H2O-extraction for Zn, Ti and Ce are 1.20, 1.32 and 0.04 lg L-1, respectively. Columns with no common letter are significantly different
Table 3 Correlation matrix of variances of soil profiles in metal nanoparticles treatment
* Correlation is significant at the level of p \ 0.05; ** Correlation is significant at the level p \ 0.01
Qtoal
Pmax
k
Urease
0.148
0.816
0.833
Catalase
0.599
0.987**
0.974**
Urease
Catalase
FDA
Azotobacter
0.871
FDA
0.705
0.823
0.811
0.667
0.871
Azotobacter
0.566
0.837
0.812
0.828
0.910*
P-bacteria
0.269
0.716
0.754
0.840
0.800
0.871
0.906*
K-bacteria
0.562
0.923*
0.925*
0.858
0.965**
0.950*
0.965**
Acknowledgments This work is supported in part by grants from the International Joint Key Project from Chinese Ministry of Science and Technology (2010DFB23160), Key project from National Science Foundation of China (41430106), National Natural Science Foundation of China (41273092), Public Welfare Project of the Chinese Ministry of Environmental Protection (201409042), and Overseas, Hong Kong and Macau Young Scholars Collaborative Research Fund (41328005).
References Blagodatskaya E, Kuzyakov Y (2013) Active microorganisms in soil: critical review of estimation criteria and approaches. Soil Biol Biochem 67:192–211 Cherchi C, Gu AZ (2010) Impact of titanium dioxide nanomaterials on nitrogen fixation rate and intracellular nitrogen storage in Anabaena variabilis. Environ Sci Technol 44:8302–8307
P-bacteria
0.946* 0.918*
Chung H, Son Y, Yoon TK, Kim S, Kim W (2011) The effect of multi-walled carbon nanotubes on soil microbial activity. Ecotoxicol Environ Saf 74:569–575 Du W, Sun Y, Ji R, Zhu J, Wub J, Guo H (2011) TiO2 and ZnO nanoparticles negatively affect wheat growth and soil enzyme activities in agricultural soil. J Environ Monit 13:822–828 Ge Y, Schimel JP, Holden PA (2011) Evidence for negative effects of TiO2 and ZnO nanoparticles on soil bacterial communities. Environ Sci Technol 45:1659–1664 Ge Y, Schimel JP, Holden PA (2012) Identification of soil bacteria susceptible to TiO2 and ZnO nanoparticles. Appl Environ Microbiol 78:6749–6758 Keller A, McFerran S, Lazareva A, Suh S (2013) Global life cycle releases of engineered nanomaterials. J Nanoparticle Res 15:1–17 Nogueira V et al (2012) Impact of organic and inorganic nanomaterials in the soil microbial community structure. Sci Total Environ 424:344–350
123
Bull Environ Contam Toxicol Priester JH et al (2012) Soybean susceptibility to manufactured nanomaterials with evidence for food quality and soil fertility interruption. Proc Natl Acad Sci USA 109:E2451–E2456 Rahman MM, Lee JK, Jeong J, Seo YR (2013) Usage of nanoparticles with their potential toxicity assessment and regulatory guidelines. Toxicol Environ Health Sci 5:49–54 Sandy EH et al (2010) A comparative cytotoxicity study of isomeric alkylphthalates to metabolically variant bacteria. J Hazard Mater 182:631–639 Servin AD et al (2013) Synchrotron verification of TiO2 accumulation in cucumber fruit: a possible pathway of TiO2 nanoparticle transfer from soil into the food chain. Environ Sci Technol 47:11592–11598 Shah V, Jones J, Dickman J, Greenman S (2014) Response of soil bacterial community to metal nanoparticles in biosolids. J Hazard Mater 274:399–403 Shrestha B, Acosta-Martinez V, Cox SB, Green MJ, Li S, CanasCarrell JE (2013) An evaluation of the impact of multiwalled carbon nanotubes on soil microbial community structure and functioning. J Hazard Mater 261:188–197 Thill A, Zeyons O, Spalla O, Chauvat F, Rose J, Auffan M, Flank AM (2006) Cytotoxicity of CeO2 nanoparticles for Escherichia coli. Physico-chemical insight of the cytotoxicity mechanism. Environ Sci Technol 40:6151–6156 Tourinho PS, van Gestel CA, Lofts S, Svendsen C, Soares AM, Loureiro S (2012) Metal-based nanoparticles in soil: fate, behavior, and effects on soil invertebrates. Environ Toxicol Chem/SETAC 31:1679–1692
123
Vittori Antisari L, Carbone S, Gatti A, Vianello G, Nannipieri P (2013) Toxicity of metal oxide (CeO2, Fe3O4, SnO2) engineered nanoparticles on soil microbial biomass and their distribution in soil. Soil Biol Biochem 60:87–94 Wang F, Yao J, Chen H, Chen K, Trebse P, Zaray G (2010) Comparative toxicity of chlorpyrifos and its oxon derivatives to soil microbial activity by combined methods. Chemosphere 78:319–326 Wang Z, Li J, Zhao J, Xing B (2011) Toxicity and internalization of CuO nanoparticles to prokaryotic alga Microcystis aeruginosa as affected by dissolved organic matter. Environ Sci Technol 45:6032–6040 Whiteley CM, Dallavalle M, Jones KC, Sweetman A (2013) Challenges in assessing release, exposure and fate of silver nanoparticles within the UK environment. Environ Sci Process Impacts 15:2050–2058 Wiesner MR et al (2011) Meditations on the ubiquity and mutability of nano-sized materials in the environment. ACS Nano 5:8466–8470 Yang Y, Wang J, Xiu Z, Alvarez PJ (2013) Impacts of silver nanoparticles on cellular and transcriptional activity of nitrogencycling bacteria. Environ Toxicol Chem 32:1488–1494 Zhu M, Wang H, Keller AA, Wang T, Li F (2014) The effect of humic acid on the aggregation of titanium dioxide nanoparticles under different pH and ionic strengths. Sci Total Environ 487:375–380
The Effect of Metal Oxide Nanoparticles on Functional Bacteria and Metabolic Profiles in Agricultural Soil Hankui Chai • Jun Yao • Jingjing Sun • Chi Zhang • Wenjuan Liu • Mijia Zhu • Brunello Ceccanti
Received: 23 June 2014 / Accepted: 26 January 2015 Ó Springer Science+Business Media New York 2015
Abstract A significant knowledge gap in nanotechnology is the absence of standardized protocols for examining the effect of engineered nanoparticles on soil microorganisms. In this study, agricultural soil was exposed to ZnO, SiO2, TiO2 and CeO2 nanoparticles at 1 mg g-1. The toxicity effect was evaluated by thermal metabolism, the abundance of functional bacteria and enzymatic activity. ZnO and CeO2 nanoparticles were observed to hinder thermogenic metabolism, reduce numbers of soil Azotobacter, P-solubilizing and K-solubilizing bacteria and inhibit enzymatic activities. TiO2 nanoparticles reduced the abundance of functional bacteria and enzymatic activity. SiO2 nanoparticles slightly boosted the soil microbial activity. Pearson’s correlation analysis showed that thermodynamic parameters had a strong correlation with abundance of functional bacteria and enzymatic activity. These findings demonstrated that the combined approach of monitoring thermal metabolism, functional bacteria and enzymatic activity is feasible for testing the ecotoxicity of nanoparticles on agricultural soil. Keywords Metal oxide nanoparticles Metabolism Functional bacteria Enzymatic activity Toxicity
H. Chai J. Yao (&) J. Sun C. Zhang W. Liu M. Zhu School of Civil and Environmental Engineering, National ‘‘International Cooperation Based on Environment and Energy’’, University of Science and Technology Beijing, 30 Xueyuan Road, Haidian District, 100083 Beijing, People’s Republic of China e-mail: [email protected] B. Ceccanti Institute of Ecosystem Studies (ISE), National Research Council (CNR), Via Giuseppe Moruzzi 1, 56124 Pisa, Italy
Products that incorporate engineered nanoparticles are increasingly entering the market. It is estimated that the commercial value of this part of the agricultural market would amount to three trillion dollars by 2020 (Servin et al. 2013). Engineered nanoparticles would inevitably be introduced into the soil matrix with application of biosolids (Shah et al. 2014). More and more people are becoming aware of the potential ecotoxicity of engineered nanoparticles. Four types of metal oxide nanoparticles (ZnO, SiO2, TiO2 and CeO2) are being widely applied in textiles, aerospace, sensors, packaging, cosmetics and sunscreens (Keller et al. 2013). Proliferation of their use will inevitably lead to uncontrolled release of nanoparticles into soil with the potential to adversely affect soil microorganisms (Keller et al. 2013; Whiteley et al. 2013; Wiesner et al. 2011). It was predicted that soil ecosystems would be the largest recipients of engineered nanoparticles after water and air. Systematic screening of the effects of nanoparticles on the soil environment has so far resulted in scarce information. Soil microorganisms play vital roles in biogeochemical cycling and organic matter dynamics. The toxicity, metabolism and transport of nanoparticles are related to the humic acid, organic matter, pH and ionic strengths of the host soil (Tourinho et al. 2012; Wang et al. 2011; Zhu et al. 2014). Therefore, it is very important to understand the behavior of nanoparticles in soil and to evaluate the risk of nanoparticles in arable soil ecosystems or other real environmental scenarios (Shrestha et al. 2013). Free Ce was observed in soil spiked with CeO2 nanoparticles, indicating that these elements were taken up by soil microorganisms (Vittori Antisari et al. 2013). More recent work by Priester et al. (2012) showed that ZnO and CeO2 nanoparticles reduced the soil nitrogen-fixing capacity and amassed in soybean tissue (Priester et al. 2012). TiO2 and ZnO nanoparticles significantly decreased
123
Bull Environ Contam Toxicol
biodiversity of soil microorganisms in a dose-dependent manner (Ge et al. 2012). Some studies measured the effects of metal nanoparticles on the standard artificial OECD soil, showing nanoparticles affected the microbial diversity and exerted toxic effects on soil invertebrates (Nogueira et al. 2012; Tourinho et al. 2012). Abundant functional bacteria thrive in soil and are crucial to the ecosystem through debris decomposition and nutrient cycling. In particular, the Azotobacter may drive the soil nitrogen fixation rate; P- and K-solubilizing bacteria may boost uptake of essential elements by plants through solubilizing P from organic debris and releasing K from silicate in soil. These bacteria may be the most vulnerable to toxic substances in the environment. Establishment of principles and test procedures for comprehensive understanding of the effect of nanoparticles on soil microorganism is essential (Rahman et al. 2013). It was shown that any xenobiotic entering into the soil environment would result in a shift of the thermogenic metabolic characteristics of the total soil microbial activity and biomass (Sandy et al. 2010). If nanoparticles have toxic effects on soil microorganisms, they would reduce biomass and thermodynamic activity. Essential nutrients such as N, P and potassium (K) are required for soil fertility and productivity. Moreover, the beneficial effect of microorganism is highly dependent on the abundance and diversity of functional bacteria in soil. Soil enzyme activity can be disturbed through the toxic effects of contamination on soil microflora and could be a reliable indicator of the current microbiological state. A bacterial cell usually contains approximately 1,000 different enzymes; many of these associated with the cell membrane. Information on the toxic effects of contaminants can be provided by evaluating soil enzyme activity involving nutrient cycling and organic matter dynamics (Wang et al. 2011). We hypothesized that the microcalorimetry technique in conjunction with measurement of other specific enzyme activities would be an effective toolkit to reflect the toxic effect of nanoparticles on soil microorganisms. To test this hypothesis, a combined approach consisting of monitoring of thermal metabolism, abundance of functional microbial groups and soil enzyme activities was used to characterize the effect of metal oxide nanoparticles (ZnO, SiO2, TiO2 and CeO2) on agricultural soil.
Materials and Methods The soil samples were collected from arable maize fields located in the Agricultural University of Hebei, China (38°460 5100 N, 115°330 3600 E). Initially, the soil samples were air dried and sieved through 2-mm mesh. The basic physicochemical properties of the soil were as follows: pH 7.46,
123
organic matter content 16.8 g kg-1, total N 0.82 mg kg-1, available P 13.3 mg kg-1 and available K 96.43 mg kg-1. Four types of nanoparticles were obtained from Boyugaoke Bio-Tech Co., China, including ZnO (15 nm, 15–25 m2 g-1), SiO2 (Amorphous, 9 nm, 50–60 m2 g-1), TiO2 (Anatase, 10 nm, [16 m2 g-1) and CeO2 (10 nm, 50–60 m2 g-1), with purity [99.9 %. Figure 1 shows the transmission electron microscopy (TEM) images of the four nanoparticles. To minimize the aggregation of nanoparticles, ZnO, SiO2, TiO2 and CeO2 nanoparticle powders were dispersed in an ultrasonic water bath at 25°C for 1 h. The above dispersions were added to soil (50 g dry weight) with a pipette and continuously stirred to achieve the homogeneous exposure dose at 1 mg g-1. Control soil had the same aliquot of water added without nanoparticles. All treatments were repeated in triplicate. To obtain the stabilization of microbial activity, each treatment was prepared in triplicate and incubated for 30 days at 70 % of water holding capacity and at 28°C. An isothermal TAM III microcalorimeter (TA instruments, New Castle, DE, USA) equipped with 12 independent channels was used to record continuous and simultaneous thermogenesis of soil microbial metabolism. Four-milliliter stainless steel ampoules sterilized by autoclaving were used in each channel. Each ampoule spiked with treatment soil (1 g) was placed inside the measuring position of the microcalorimeter. Nutrient solution (200 lL) containing 5.0 mg glucose and 5.0 mg ammonium sulfate was added in each ampoule to support the growth of soil microorganisms (Wang et al. 2010). Data were exported for further analysis and graphic presentation through Origin software 8.0 (OriginLab, Northampton, MA, USA). Thermogenic parameters, such as total heat output (Qtotal), growth rate constant (k), the value of peak height (Pmax) and corresponding time (Tmax), were calculated from each curve. The Qtotal was determined by the integration of each power–time curve. The k was calculated based on the heat output being proportional to the rate of bacterial growth. In the exponential growth phase, Pt which is equal to the first derivative of Qtotal with time (t), obeys the kinetics: Pt = P0 ekt. Pt is the output of power at time T, and the P0 is the power at the initial exponential growth stage. Enumeration of soil Azotobacter, P- and K-solubilizing bacteria were counted by suspension dilution methods on differentiating agar media. Ashby nitrogen-free culture medium was used to provide quantitative estimation of nitrogen-fixing bacteria (per L: mannitol 10 g, KH2PO4 0.2 g, NaCl 0.2 g, MgSO47H2O 0.2 g, CaCO3 5 g, CaSO4 0.1 g, agar 15 g, pH 7.0). P-solubilizing bacteria were counted using the following medium composition (per L: glucose 10.0 g, NaCl 0.3 g, (NH4)2SO4 0.5 g, KCl 0.3 g, FeSO47H2O 0.3 g, MgSO47H2O 0.3 g, MnSO44H2O 0.03 g, CaCO3 5.0 g, lecithin 0.2 g, agar 15 g, pH 7.2).
Bull Environ Contam Toxicol
Fig. 1 Transmission electron microscopy (TEM) images of four metal oxide nanoparticles: a ZnO; b SiO2; c TiO2; d CeO2
Numbers of K-dissolving bacteria were determined on the following medium (per L: sucrose 5.0 g, CaCO3 0.1 g, Na2HPO4 2 g, FeCl3 0.005 g, soil minerals 1 g, MgSO4 7H2O 0.5 g, agar 15.0 g, pH 7.5). The above media were sterilized at 120°C for 30 min. All the plates were incubated at 28°C for 7 days. Soil urease activity was determined by absorbance of ammonium released from urea hydrolysis at 578 nm (Wang et al. 2010). Catalase activity was determined by back-titrating residual peroxide with 0.1 mol L-1 KMnO4 solution (Du et al. 2011). Fluorescein diacetate hydrolysis activity was measured by absorbance of fluorescence generated by fluorescein at 490 nm (Yang et al. 2013). The availability of nanoparticles in soil was determined by CHCl3-labile metals and water extraction (Vittori Antisari et al. 2013). Subsamples (4 g) were divided into two portions. One sample portion (2 g) was fumigated with ethanol-free CHCl3 for 24 h at 25°C to lyse bacterial cells. After the fumigation step, the samples were shaken by
adding 5 mL 1 mol L-1 NH4NO3 for 1 h and filtered through a 0.45 lm polysulfone filter. After filtration, 0.1 mL HNO3 solution (1:10 v/v, trace-metal) was added to the extraction which was then stored at 4°C. The nonfumigated portion was extracted using the same procedure as the fumigated portion. The CHCl3-labile metals were calculated as the concentration in the fumigated portion minus the concentration in the non-fumigated portion. Water extraction was extracted with Milli-Q water. The soil samples were shaken with Milli-Q water (1:10 w/v) for 24 h. The extract was obtained by centrifugation for 15 min at 2,000 rpm and then filtered through a 0.45 lm polysulfone filter. The total concentrations of Zn, Si, Ti and Ce were determined by ICP-MS. Statistical analyses were carried out using SPSS 13.0 (SPSS International, USA). Differences between treatments and controls were tested with one-way analysis of variance. Duncan’s test was applied to determine the significant differences at the level of p \ 0.05. Correlation
123
Bull Environ Contam Toxicol
analysis was performed with two-tailed Pearson test, with p \ 0.05 and p \ 0.01 as the significant values.
Results and Discussions The microcalorimetry assay was used to study soil microbial activity in the presence and absence of four types of metal oxide nanoparticles (Fig. 2). As seen from the power–time curves, some differences were observed in nanoparticle treatments in comparison to the control. Soils spiked with ZnO and CeO2 nanoparticles exhibited low thermodynamic activity. More specifically, the initial power value of ZnO nanoparticles was significantly lower than the other treatments. Soils spiked with SiO2 and TiO2 nanoparticles had similar thermodynamic characteristics to the control. Table 1 presents the thermal parameters of all treatments to compare the toxic effect of the four metal oxide nanoparticles. ZnO and CeO2 nanoparticles had lower values of Pmax (852 lW and 898 lW for ZnO and CeO2 nanoparticles, respectively (p \ 0.05)) than the other treatments and the control. The Pmax of TiO2 and SiO2 nanoparticles was slightly higher than the control. The low k values for ZnO and CeO2 nanoparticles indicated that the soil microbial metabolism was inhibited at the exponential growth stage (p \ 0.05). The superoxide and reactive oxygen species generated by nanoparticles might have reduced the microbial biomass and activity. It has been demonstrated that TiO2 and ZnO nanoparticles reduced the substrate induced respiration of grassland soil (Ge et al. 2011). Thermogenesis involves many enzymes that drive glycolysis, the Krebs cycle and the electron transport chain. This variability in microbial activities of the soil might be caused by direct oxidative stress on enzymes (Thill et al. 2006). 1200 control
ZnO
SiO2
TiO2
CeO2
20
25
Power (µW)
1000 800 600 400 200 0 5
10
15
Time (h) Fig. 2 Power–time curves of soil microorganisms spiked with metal oxide nanoparticles. Values are means of three replicates
123
The total plate counts and enzymatic activity of soil samples exposed to different nanoparticles are displayed in Table 2. ZnO, CeO2 and TiO2 nanoparticles significantly reduced the numbers of Azotobacter, P- and K-solubilizing bacteria (p \ 0.05). No significant difference was observed for total plate counts between the SiO2 nanoparticles and the control. Reductions in enzymatic activities were more significant in response to ZnO, CeO2 and TiO2 nanoparticle treatments (p \ 0.05) than for SiO2 nanoparticles, compared with the control. This was probably a consequence of microbial cell absorption or uptake of free metals released by nanoparticles in soil. We observed Zn, Ce and Ti, but not Si, in the CHCl3-labile metal extractions and the H2O extractions. A significant part of the nanoparticles might be transferred or absorbed into microbial cells and ions might be released in soil. Nitrogen fixation of soybean nodules was reduced in organic farm soil contaminated with ZnO and CeO2 nanoparticles (Priester et al. 2012). It was shown that cell growth and nitrogen fixing capacity of Anabaena variabilis was hampered by TiO2 nanoparticles in the aquatic environment (Cherchi and Gu 2010). DNA-based fingerprinting analyses showed that TiO2 and ZnO nanoparticles induced a decrease in taxa associated with nitrogen cycling (Rhizobiales, Bradyrhizobiaceae, Bradyrhizobium) (Ge et al. 2012). Similar studies confirmed that Ag nanoparticles exerted adverse effects on nitrogen-cycling bacteria in agricultural soil (Yang et al. 2013). TiO2 and ZnO nanoparticles significantly decreased protease, catalase, and peroxidase activities in wheat soil, which might be due to dissolved ions in soil (Du et al. 2011). b-Glucosidase and phosphatase activities were inhibited in sandy loam treated with multi-walled carbon nanotubes (Chung et al. 2011). Pearson’s correlation analysis showed that thermodynamic parameters were strongly correlated with biochemical properties under the impact of nanoparticles (Table 3). Catalase activity and K-solubilizing bacteria were significantly correlated with the Pmax and the k. In the microenvironment of microcalorimeter ampoules, oxygen and nutrients are the limiting factors for the growth and reproduction of microorganisms. Therefore, heat output from soil samples is a 1:1 molar relationship with microbial growth rate. For this reason, the Pmax and k values can represent the soil microbial activity. The change in k and the Pmax of each experiment indicated the change in soil microbial activity. Soil enzyme activity involved in nutrient cycles is highly dependent on the total microbial activity and related to the abundance of functional bacteria. In this study, we found positive correlations for each thermodynamic parameter, and we also found positive correlations for functional bacteria and enzymatic activities. These findings indicated that the combined method is feasible to assess the effects of nanoparticles on the soil.
Bull Environ Contam Toxicol Table 1 Thermogenesis parameters of soil treated with different metal oxide nanoparticles Qtotala (J g-1)
Tmaxb (h)
Kd (h-1)
Pmaxc (lW)
Control
25 ± 1.5A
11 ± 0.3 A
985 ± 17.8 A
0.494 ± 0.020 A
ZnO
23 ± 1.2 A
12 ± 0.2 B
852 ± 21.3 D
0.419 ± 0.025 C
SiO2
25 ± 1.1 A
11 ± 0.3 A
TiO2
24 ± 1.2 A
11 ± 0.5 A
991 ± 2.3 AB
0.498 ± 0.026 A
CeO2
18 ± 0.5 B
10 ± 0.3 A
898 ± 20.3 C
0.456 ± 0.017 B
1,019 ± 2.0 B
0.526 ± 0.011 A
Values are means of three replicates ±SD, values labeled with different letters indicate statistically significant differences (p \ 0.05, Duncan’s test), a Total heat output calculated from power–time curve; b The corresponding time of the height of each peak; c The height of the peak; d The microbial growth rate constant
Table 2 Total plate counts of functional bacteria and enzyme activities, and the availability of nanoparticles in soil Control
ZnO
SiO2
TiO2
CeO2
Azotobacter (logCFU)
6.23 ± 0.027 A
6.09 ± 0.029 D
6.2 ± 0.017 AB
6.16 ± 0.012 BC
6.13 ± 0.030 CD
P-bacteria (logCFU)
6.55 ± 0.060 A
6.07 ± 0.091 D
6.54 ± 0.026 AB
6.28 ± 0.074 D
6.41 ± 0.028 BD
K-bacteria (logCFU)
5.52 ± 0.042 A
5.22 ± 0.033 C
5.53 ± 0.022 A
5.42 ± 0.020 B
5.35 ± 0.066 B
Urease (mg NH4?-N g-1 soil 24 h-1)
0.88 ± 0.015 A
0.53 ± 0.037 C
0.84 ± 0.007 B
0.82 ± 0.006 B
0.81 ± 0.004 B
Catalase (mL KMnO4 g
-1
-1
1.15 ± 0.010 A
1.01 ± 0.030 B
1.16 ± 0.003 A
1.13 ± 0.010 A
1.06 ± 0.020 B
FDA (lg Fluorescein g-1 soil 0.5 h-1) 21.35 ± 0.877 A
soil 0.5 h )
13.12 ± 0.018 C
21.05 ± 0.673 A
15.61 ± 1.905 B
13.41 ± 0.032 B
CHCl3-labile (lg kg-1)
ND
28.51 ± 0.907 A
ND
5.48 ± 0.991 C
12.96 ± 0.536 B
–
376.33 ± 3.512 A
ND
12.52 ± 1.284 C
12.96 ± 0.537 B
H2O-extraction (lg L-1)
FDA indicates the fluorescein diacetate hydrolysis activity. ND is not detectable (lower than the detection limit). The values of H2O-extraction for Zn, Ti and Ce are 1.20, 1.32 and 0.04 lg L-1, respectively. Columns with no common letter are significantly different
Table 3 Correlation matrix of variances of soil profiles in metal nanoparticles treatment
* Correlation is significant at the level of p \ 0.05; ** Correlation is significant at the level p \ 0.01
Qtoal
Pmax
k
Urease
0.148
0.816
0.833
Catalase
0.599
0.987**
0.974**
Urease
Catalase
FDA
Azotobacter
0.871
FDA
0.705
0.823
0.811
0.667
0.871
Azotobacter
0.566
0.837
0.812
0.828
0.910*
P-bacteria
0.269
0.716
0.754
0.840
0.800
0.871
0.906*
K-bacteria
0.562
0.923*
0.925*
0.858
0.965**
0.950*
0.965**
Acknowledgments This work is supported in part by grants from the International Joint Key Project from Chinese Ministry of Science and Technology (2010DFB23160), Key project from National Science Foundation of China (41430106), National Natural Science Foundation of China (41273092), Public Welfare Project of the Chinese Ministry of Environmental Protection (201409042), and Overseas, Hong Kong and Macau Young Scholars Collaborative Research Fund (41328005).
References Blagodatskaya E, Kuzyakov Y (2013) Active microorganisms in soil: critical review of estimation criteria and approaches. Soil Biol Biochem 67:192–211 Cherchi C, Gu AZ (2010) Impact of titanium dioxide nanomaterials on nitrogen fixation rate and intracellular nitrogen storage in Anabaena variabilis. Environ Sci Technol 44:8302–8307
P-bacteria
0.946* 0.918*
Chung H, Son Y, Yoon TK, Kim S, Kim W (2011) The effect of multi-walled carbon nanotubes on soil microbial activity. Ecotoxicol Environ Saf 74:569–575 Du W, Sun Y, Ji R, Zhu J, Wub J, Guo H (2011) TiO2 and ZnO nanoparticles negatively affect wheat growth and soil enzyme activities in agricultural soil. J Environ Monit 13:822–828 Ge Y, Schimel JP, Holden PA (2011) Evidence for negative effects of TiO2 and ZnO nanoparticles on soil bacterial communities. Environ Sci Technol 45:1659–1664 Ge Y, Schimel JP, Holden PA (2012) Identification of soil bacteria susceptible to TiO2 and ZnO nanoparticles. Appl Environ Microbiol 78:6749–6758 Keller A, McFerran S, Lazareva A, Suh S (2013) Global life cycle releases of engineered nanomaterials. J Nanoparticle Res 15:1–17 Nogueira V et al (2012) Impact of organic and inorganic nanomaterials in the soil microbial community structure. Sci Total Environ 424:344–350
123
Bull Environ Contam Toxicol Priester JH et al (2012) Soybean susceptibility to manufactured nanomaterials with evidence for food quality and soil fertility interruption. Proc Natl Acad Sci USA 109:E2451–E2456 Rahman MM, Lee JK, Jeong J, Seo YR (2013) Usage of nanoparticles with their potential toxicity assessment and regulatory guidelines. Toxicol Environ Health Sci 5:49–54 Sandy EH et al (2010) A comparative cytotoxicity study of isomeric alkylphthalates to metabolically variant bacteria. J Hazard Mater 182:631–639 Servin AD et al (2013) Synchrotron verification of TiO2 accumulation in cucumber fruit: a possible pathway of TiO2 nanoparticle transfer from soil into the food chain. Environ Sci Technol 47:11592–11598 Shah V, Jones J, Dickman J, Greenman S (2014) Response of soil bacterial community to metal nanoparticles in biosolids. J Hazard Mater 274:399–403 Shrestha B, Acosta-Martinez V, Cox SB, Green MJ, Li S, CanasCarrell JE (2013) An evaluation of the impact of multiwalled carbon nanotubes on soil microbial community structure and functioning. J Hazard Mater 261:188–197 Thill A, Zeyons O, Spalla O, Chauvat F, Rose J, Auffan M, Flank AM (2006) Cytotoxicity of CeO2 nanoparticles for Escherichia coli. Physico-chemical insight of the cytotoxicity mechanism. Environ Sci Technol 40:6151–6156 Tourinho PS, van Gestel CA, Lofts S, Svendsen C, Soares AM, Loureiro S (2012) Metal-based nanoparticles in soil: fate, behavior, and effects on soil invertebrates. Environ Toxicol Chem/SETAC 31:1679–1692
123
Vittori Antisari L, Carbone S, Gatti A, Vianello G, Nannipieri P (2013) Toxicity of metal oxide (CeO2, Fe3O4, SnO2) engineered nanoparticles on soil microbial biomass and their distribution in soil. Soil Biol Biochem 60:87–94 Wang F, Yao J, Chen H, Chen K, Trebse P, Zaray G (2010) Comparative toxicity of chlorpyrifos and its oxon derivatives to soil microbial activity by combined methods. Chemosphere 78:319–326 Wang Z, Li J, Zhao J, Xing B (2011) Toxicity and internalization of CuO nanoparticles to prokaryotic alga Microcystis aeruginosa as affected by dissolved organic matter. Environ Sci Technol 45:6032–6040 Whiteley CM, Dallavalle M, Jones KC, Sweetman A (2013) Challenges in assessing release, exposure and fate of silver nanoparticles within the UK environment. Environ Sci Process Impacts 15:2050–2058 Wiesner MR et al (2011) Meditations on the ubiquity and mutability of nano-sized materials in the environment. ACS Nano 5:8466–8470 Yang Y, Wang J, Xiu Z, Alvarez PJ (2013) Impacts of silver nanoparticles on cellular and transcriptional activity of nitrogencycling bacteria. Environ Toxicol Chem 32:1488–1494 Zhu M, Wang H, Keller AA, Wang T, Li F (2014) The effect of humic acid on the aggregation of titanium dioxide nanoparticles under different pH and ionic strengths. Sci Total Environ 487:375–380
