Introduction

Nanoscale zerovalent iron (nZVI) has special properties such as larger specific surface area and higher surface reactivity in comparison to bulky particles. nZVI has been extensively used in in situ remediation (i.e., groundwater and soil) (Chowdhury et al. 2015; Tosco et al. 2014), and the removal efficiency of pollutants is regarded as considerable (Zhao et al. 2016). Quantitatively higher release of nZVI to the soils have raised great concerns on health and safety (Lankadurai et al. 2015; Nowack and Bucheli 2007), because most of the studies were performed on aquatic organisms (Chen et al. 2012; Marsalek et al. 2012) and not on edaphic species that is important for risk evaluation of nZVI application.

One of the important hazardous concerns is that how soil organisms are affected by high iron doses, particularly high reactive iron. nZVI was mainly intended to be used in the subsoil under saturated conditions. Furthermore, due to the aggregation and deposition characteristics (Phenrat et al. 2008), high concentrations of nZVI may possess the transportation ability towards unsaturated soil (i.e., surface layers). The highly redox-reactive nZVI and oxidized products may then penetrate into animal and plant cells and thus induce toxic effects. Therefore, elucidating the ecological hazard of nZVI in the soil environment is urgently needed (Ma et al. 2013; Saccà et al. 2014).

Among soil organisms, the earthworm Eisenia fetida has the ability to accumulate large quantities of contaminants and is considered as a model organism for toxicity assessment (OECD 1984). A recent study has used earthworms to determine the toxicity and bioavailability of engineered nanoparticles (ENPs) (Kwak and An 2015), while to the best of our knowledge, nZVI has been barely assessed in detail with regard to E. fetida. In fact, nZVI could result in cellular damage by the formation of reactive oxygen species (ROS) (Chen et al. 2012). Therefore, the objectives of the present study are to determine the biochemical and toxicological effects of nZVI to earthworms. To this end, we investigated the survival, growth, avoidance, respiration, ROS, superoxide dismutase (SOD), catalase (CAT), and malondialdehyde (MDA) content of E. fetida when subjected to nZVI exposure. Additionally, the skin damage of earthworms was also assessed based on histological examinations. The results and related findings would establish a more integrated understanding of potential ecological risk of nanomaterials in a soil system for its better diagnosis and remediation.

Materials and methods

Chemicals

nZVI was purchased from Chaowei Nanotechnology Co. Ltd., Shanghai, China. According to the manufacturer, nZVI was made of iron nanoparticles coated with 1-nm iron oxide shell. In addition to this, its specific surface area was 23 m2 g−1, and the purity was 99.9%. All reagents used in the study were of analytical grade.

Characterization of nZVI

The characterization of nZVI was conducted by transmission electron microscopy (TEM, JEOL JEM-1400). The particle size of nZVI ranged from 50 to 100 nm, confirming the information provided by the manufacturer. X-ray diffraction (XRD) analysis of nZVI was obtained by RIGAKU D/MAX-2550 VB/PC at 40 kV and 100 mA. The detailed data is shown in Fig. S1.

Soil and earthworm collection

Soil samples were collected from the surface layer (0–20 cm) in the center of East China University of Science and Technology, China. The soil was air-dried and sieved to 2 mm. Soil properties were assessed according to Nelson and Sommers (1982) and can be summarized as follows: silty clay loam; pH, 7.3; organic matter, 6.5%. Variation of oxidation-reduction potential (ORP) was monitored in soils for indicated days. Additionally, the background level of total Fe was 47.62 mg kg−1 in pristine soil by atomic absorption spectrometer (AAS) (Agilent, USA).

Mature (clitellate) specimens of E. fetida (0.55 ± 0.12 g) were bought from Yonghe Earthworm Culture Farm (Shanghai, China) and cultured in natural soils with cattle manure at room temperature (20 ± 1 °C) for about 28 days.

Acute toxicity tests

The whole experiment outline is depicted in Fig. 1.

Fig. 1
figure 1

Experimental outline

Full size image

The toxicity of nZVI to earthworms was investigated based on filter paper contact and natural soil tests as recommended by OECD (1984) with slight modifications. The nZVI-containing solutions were prepared in an anaerobic glovebox.

As for the filter paper test, the experiments were conducted in 9-cm petri dishes, each of which contained a filter paper. Ultrapure water was exposed to nitrogen gas for 10 min. Stable suspensions of nZVI were prepared with ultrapure water (18.2 MΩ) by 30-min sonication in an ultrasonic bath to have a homogenous suspension. Then, 2-mL suspensions were added onto the filter paper. Five different treatments of nZVI were prepared as 100, 500, 1000, 1500, and 2000 μg cm−2. As a control, 2 mL of deionized water was added to another batch of dishes. Twenty replicates were used for each concentration treatment of nZVI. Each replicate contained one depurated earthworm. The dishes were capped loosely with medical gauze. Tests were conducted under the conditions: 20 ± 1 °C, 12/12 h day/dark. Mortality was recorded after the incubation periods of 24, 48, and 72 h. Worms were considered dead when they did not respond to gentle mechanical stimulus.

As for the natural soil test, 500 g of soil amended with nZVI was filled into 800-mL glass beakers. Ultrapure water was exposed to nitrogen gas for 10 min. Stable suspensions of nZVI were prepared with ultrapure water (18.2 MΩ) by sonication for 30 min in an ultrasonic bath to have a homogenous suspension. After that, the suspension was immediately hand mixed into a subsample of soil for each replicate with a glass rod. The concentrations were selected based on El-Temsah and Joner (2012), being 0, 100, 500, and 1000 mg kg−1 dry soil of nZVI. Three replicates were used for each dose of nZVI. Each replicate had ten depurated earthworms. The test was run at 20 ± 1 °C, 12/12 h day/dark. Water was added regularly to maintain the soil moisture to 65% of the total water holding capacity (WHC). According to Zhang (2003), nZVIs remain reactive for more than 1 month. Mortality was recorded after the incubation periods of 7, 14, 21, and 28 days. Worms were considered dead on being irresponsive to gentle mechanical stimulus.

Growth test

Experimental processing and concentration levels were referred to the present natural soil test. After 7, 14, 21, and 28 days of exposure, earthworms were removed from the substrate, washed in water, placed on moisture filter paper for 1 day to purge the guts, and weighed. Growth inhibition of earthworms after various exposure periods in each dose group was calculated according to the following equation:

$$ {GI}_{\mathrm{n}}=\left({\mathrm{W}}_0-{\mathrm{W}}_{\mathrm{t}}\right)/{\mathrm{W}}_0\times 100\% $$

where GIn is the growth inhibition for dose group n, W0 is the weight on day 0, and Wt is the weight after t days of exposure.

Avoidance test

The avoidance test was performed in two-sectioned vessels following the description provided by the ISO guideline (ISO 2007). Briefly, one side with 500 g of soil spiked with different levels of nZVI and the other with 500 g of control soil was included. Three replicates for each treatment were performed, with ten adult earthworms in each replicate. In order to prevent the escape of earthworms, the boxes were covered by perforated medical gauze. The test was run at 20 ± 1 °C, 12/12 h day/dark. After 48 h of incubation, the number of earthworms in each vessel was recorded. The mean percentages of net response (NR) were calculated using the following equation:

$$ NR=\left(C-T\right)/N\times 100\% $$

where C is the number of earthworms observed in the control soil, T is the number of earthworms observed in the test soil, and N is the total number of earthworms per replicate.

Respiratory test

The respiratory test was performed following a method described by Li et al. (2015) with slight modifications. Briefly, the test was conducted in a beaker (800 mL) filled with sterilized soils (100 g), and later, earthworms were exposed to the spiked soils for 4 weeks. The concentration levels were referred to above 28-day growth toxicity test. After the 28-day exposure, the earthworms were removed from the containers (two earthworms in a beaker), depurated and weighed, then placed in a flask, and the amount of CO2 was determined after 16 h. The CO2 was captured by a small beaker filled with 5 mL of 1 M NaOH solution that was determined by titrating with 0.1 M HCl on day 28. The final results were expressed as 10−5 mol CO2 per g of earthworms’ body weight per hour (Tang et al. 2016). Three replicates were set up for each of the four treatments.

Biochemical indices assay

The procedure implemented was that employed by Zhang et al. (2015a) with slight modifications. Experimental processing and concentration levels were referred to the above 28-day growth toxicity test. After 28 days of exposure, two depurated earthworms were homogenized using a glass homogenizer under ice-cold conditions in Tris–HCl buffer (1:4, w/v) and centrifuged at 10,000×g at 4 °C for 15 min. The supernatants were used to assess SOD and CAT activities as well as MDA contents by the commercial kits (Nanjing Jiancheng, Nanjing, China). Three runs of each treatment were conducted.

ROS detection

ROS detection was conducted using the DCFH-DA method according to Zhao et al. (2013) with slight modifications. After 28 days, two depurated earthworms were homogenized using a glass homogenizer under ice-cold conditions in potassium phosphate buffer (pH = 7.4) and centrifuged at 1000×g at 4 °C for 10 min. Then, the supernatants were re-centrifuged at 12,000×g at 4 °C for 20 min to obtain the mitochondrial protein, which was incubated at 37 °C with DCFH-DA solution for 30 min in a water bath. The fluorescence of samples was monitored at an excitation wavelength of 485 nm and an emission wavelength of 530 nm using a Varian Cary Eclipse fluorescent spectrophotometer.

Histology

The whole procedure was conducted based on the previous reports by Chen et al. (2011) and Zhang et al. (2015b) with slight modifications. Experimental processing and concentration levels were referred to the present natural soil test. After 28 days, the depurated earthworms were fixed with formalin, embedded in paraffin, and 7 μm transverse sections were stained with hematoxylin-eosin (HE) for light microscope observation (Nikon Eclipse 80i).

Data analysis

All data analysis was conducted using SPSS 16.0 (SPSS, Chicago, USA). Significant differences on growth, avoidance, respiration, ROS, SOD, and CAT activities and MDA contents were assessed by one-way analysis of variance followed by Duncan’s post hoc test. Figures were completed using Origin 8.0 (Origin Lab, Northampton, MA, USA). Pearson’s correlation analysis was performed to identify significant correlations between the measured variables.

Results and discussion

Acute toxicity

The acute toxic effects of nZVI on earthworms evaluated by the filter paper contact and soil tests are shown in Fig. S2. In the case of the filter paper contact test, the results showed that after 24-h incubation, the mortality (>10%) was observed in earthworms exposed to nZVI (>1000 μg cm−2). Afterward, it was aggravated with increasing nZVI dosage and exposure time. In the natural soil assay, the mortality of earthworms was lower than 50% in different treated groups throughout the entire incubation. Seventy-nine percent of mortality was reported by El-Temsah and Joner (2012) for E. fetida exposed to 500 mg kg−1 nZVI in sandy loam soil after 14-day exposure. The discrepancy might be because nZVI toxicity was strongly affected by soil properties. Chen et al. (2012) reported that ORP was an absolutely crucial parameter to elucidate the toxicity mechanisms, and ORP values (Fig. S3) in the soil could verify the corresponding indicators in the present study. In addition, we noticed that the toxicity of nZVI was different from that observed in the filter paper contact assay, which could be due to the bioavailability of pollutants modified by soil physicochemical characteristics. Therefore, the exposure media should be taken into consideration when evaluating the toxicity of nanomaterials (Park et al. 2013).

Impacts on growth

Effects of nZVI on earthworms’ growth are shown in Fig. 2. In general, the growth was induced (>5%) when earthworms were exposed to the control and 100 mg kg−1 nZVI groups. The inhibitory effects of nZVI (1000 mg kg−1) appeared to be time-dependent, and the growth was inhibited by approximately 15% at the highest level of nZVI. It has been shown that the inhibition resulted from the depletions of glycogen and lipid contents and decrease in protein content (Liu et al. 2015; Ribeiro et al. 2001). In the present study, the reason might be the same, i.e., the declining energy reserves of the earthworms.

Fig. 2
figure 2

Growth responses of Eisenia fetida exposed to various concentrations of nZVI for indicated days

Full size image

Impacts on avoidance behavior

The results of the avoidance test are presented in Fig. 3. At the lowest nZVI concentration (100 mg kg−1), the earthworms exhibited low avoidance behavior, i.e., their numbers were higher in the spiked soils than in the control samples. Similarly, Amorim and Scott-Fordsmand (2012) showed that Cu-NP exposure (30 mg kg−1) resulted in low avoidance behavior. When nZVI concentration reached 1000 mg kg−1, the avoidance response reached 40%. The NR values were less than 80%, which indicated that no obvious avoidance response occurred after 48-h exposure to nZVI. Nevertheless, the NR in the treatments was significantly higher (P < 0.05) than in the controls, illustrating that it could mean an advantage in case of spilling of nZVI in natural soils (Hu et al. 2014).

Fig. 3
figure 3

Avoidance responses of Eisenia fetida exposed to various concentrations of nZVI after 48 h of incubation

Full size image

Impacts on respiration

The effects of nZVI on the respiration of earthworms were determined based upon the principle of CO2 production. As illustrated in Fig. 4, there were no significant (P > 0.05) changes when earthworms were exposed to 100 mg kg−1 nZVI. However, exposures to 500 and 1000 mg kg−1 nZVI significantly inhibited the respiratory activity (P < 0.05) with inhibition rates of 18.1 and 24.5%, respectively. Li et al. (2015) reported that the CO2 production of earthworms was decreased after exposure to enrofloxacin-spiked soils, and it was correlated to the inhibition in earthworms’ metabolism. In the present study, significant decreases in respiration after exposure to higher levels of nZVI suggest the inhibition in earthworms’ metabolism probably leading to lower vital activity. Additionally, the result of respiration was well correlated with growth, which could be ascribed to energy reserves of soil invertebrates (Khalil 2015).

Fig. 4
figure 4

Respiration responses of Eisenia fetida exposed to various concentrations of nZVI for 28 days

Full size image

Impacts on antioxidant enzymes, MDA and ROS levels

The living organisms have the ability to protect themselves from ROS damage through the antioxidant enzymes (i.e., SOD and CAT) (Foyer and Shigeoka 2011; Hao et al. 2009). The impacts on SOD and CAT activities of E. fetida after 7, 14, 21, and 28 days of exposure are shown in Fig. 5. The activity of SOD was clearly (P < 0.05) induced after exposure to different levels of nZVI after a short-term incubation (7 days), with activation rates of 54.6, 29.4, and 50.0%, respectively, which could be considered as a protection mechanism to the chemical stress (Hao et al. 2009). After a long-term incubation (28 days), SOD activity was significantly (P < 0.05) inhibited in the low (100 mg kg−1 nZVI) and high (1000 mg kg−1 nZVI) tested groups, with inhibition rates of 24.0 and 20.6%, respectively. The reason might be due to the overwhelming effect on SOD by the excess of reactive superoxide anion induced by nZVI (Foyer and Shigeoka 2011; Hu et al. 2016). On contrary to the SOD activity, CAT was significantly increased (P < 0.05) in only one treatment (1000 mg kg−1 nZVI) after short-term exposure (14 days), while it was significantly (P < 0.05) induced after a long-term exposure (21 days), indicating that the CAT has enough time to depose of excess H2O2. Apart from that, momentous enhancement of the CAT activity resulting from the early activation of SOD pointed out their complementary mechanism in earthworms.

Fig. 5
figure 5

SOD and CAT activities responses of Eisenia fetida exposed to various concentrations of nZVI for indicated days

Full size image

Lipid peroxidation (LPO), regarded as the biomarker to assess the oxidative stress, was determined by evaluating earthworm’s MDA content (Dotan et al. 2004). As depicted in Fig. 6, there was a clear dose- and time-response pattern of MDA content in earthworms during the whole incubation, and the peak was observed at 1000 mg kg−1 nZVI after 28 days of incubation, increasing by 18.9% relative to controls. Increases in the MDA contents in earthworms exposed to the pollutant would suggest that the organisms were attacked by excessive ROS (Zhao et al. 2013), which is evidenced from Fig. 7. It was demonstrated that there was a clear dose-dependent pattern of ROS after exposure to nZVI. In addition, the results obtained from this study showed that the generation of ROS might also cause oxidative perturbations in earthworms.

Fig. 6
figure 6

MDA contents responses of Eisenia fetida exposed to various concentrations of nZVI for indicated days

Full size image
Fig. 7
figure 7

ROS levels in Eisenia fetida exposed to nZVI for 28 days

Full size image

Previous studies have demonstrated that the measurement of biological indicators could be useful in environmental pollutant monitoring. Therefore, it is vital to establish that the specific responses were dose-responsive and sensitive to the pollutant of interest (Wang et al. 2016; Li et al. 2016). In this study, the correlation analysis was performed in order to compare the sensitivity of earthworm’s biomarkers to nZVI toxic effects (Table 1). Growth, respiration, MDA, and ROS were well correlated with each other (R > 0.8; P < 0.05). Additionally, a negative correlation (R = −0.974; P < 0.05) was observed between CAT and SOD activity. We can conclude that increasing nZVI contents in soils toxically affected earthworms. This could lead to ROS generation and subsequent production of MDA (R = 0.880), which inhibited the growth (R = 0.842) and respiration (R = 0.855) of earthworms. Hence, ROS might be the most sensitive parameter.

Table 1 Pearson’s correlation coefficients between pairs of parameters measured in earthworms after 28 days of exposure to nZVI in natural soils (R value)
Full size table

According to the previous reports (Adeleye et al. 2016; Clark et al. 2003; Matheson and Tratnyek 1994), non-nano-ZVI particles have been proved non-toxic, indicating that they would not exert significant toxic effects on earthworms. Therefore, we could speculate that the observed results in the current study were mainly nano-specific. Furthermore, Lee et al. (2008) observed that Fe ions may have ability to enter the cells to induce oxidative stress and further damage cell membranes. The hypothesized mechanisms may be that nZVI attached to the surfaces of earthworms and transferred electrons to different biochemicals, leading to some undesired reactions.

Histological effects

The histopathological examination of transverse sections of E. fetida after the 28-day exposure is shown in Fig. 8. When the earthworms were exposed to the control group, normal skin structure including epidermal, circular, and longitudinal muscles were observed. The addition of 100 mg kg−1 nZVI barely instigated any distinctive change in earthworms. However, the aggravated structures of circular muscle layer and disintegration of longitudinal muscle layer were observed at 500 and 1000 mg kg−1 nZVI-treated groups. Similar conclusions were proved by SEM results (Fig. S4). Figure S4a–g illustrated the toxic impacts of different levels of nZVI on earthworms’ skin. The intersegmental furrows or setae in earthworms in the control and 100 mg kg−1 nZVI tested groups showed normal architecture (Fig. S4a–c). However, various anomalies appeared in earthworms after being exposed to the higher nZVI-treated groups (500 and 1000 mg kg−1) (Fig. S4d–g). For example, irregular folds, disorganized texture, dehydration symptoms, and lacerations were observed clearly in the intersegmental furrows or setae. The present study demonstrated that ROS was induced after exposure to nZVI, which may be responsible for the degeneration of skin in earthworms. Further investigations are needed to demonstrate the potential mechanisms.

Fig. 8
figure 8

Histopathological examination of Eisenia fetida exposed to nZVI for 28 days

Full size image

Conclusions

In the present work, multiple biomarkers, including mortality, growth, respiration, avoidance behavior, enzymatic activities (SOD and CAT), MDA content, ROS intensity, and histopathological examination, were investigated to assess the potential toxicological effects of nZVI on earthworms. The adverse impacts of nZVI on the individual parameters of the earthworms, such as growth, respiration, and behavioral parameters, were clearly observed. The biochemical parameters, including SOD and CAT, clearly revealed that oxidative stress was induced by nZVI in earthworms. The accumulation of ROS was obviously induced by nZVI, followed by a significant increase in the MDA level. Furthermore, we observed that the biomarkers differed in their sensitivity to the nZVI exposure, and ROS seemed to be the most sensitive endpoint. The produced ROS might be responsible for the degeneration of skin cells, generating various anomalies in histopathological observations. These comprehensive biomarker responses will be useful in environmental monitoring and could be a useful approach in ecotoxicological studies.