Ecotoxicity stress and bioaccumulation in Eisenia fetida earthworms exposed to vanadium pentoxide in soil

As an important commercial form of vanadium, vanadium pentoxide (V2O5) is widely used in various modern industries, and its environmental impacts and ecotoxicity have been extensively studied. In this research, the ecotoxicity of V2O5 to earthworms (Eisenia fetida) in soil was tested by exposure to V2O5 at a series of doses, and biochemical response indices, such as the superoxide dismutase (SOD) and catalase (CAT) enzyme activity and malondialdehyde (MDA) content, were analysed to determine the mechanism by which antioxidant enzymes respond to V2O5 exposure. The bioaccumulation factor (BAF) of vanadium pentoxide in the earthworms and soil was also measured to explore the bioaccumulation process of V2O5 in the test period. The results showed that the acute and subchronic lethal toxicity values of V2O5 towards E. fetida were 21.96 mg/kg (LC50, 14 days) and 6.28 mg/kg (LC10, 28 days), respectively. For the antioxidant enzymes, SOD and CAT were synchronously induced or inhibited within the time period, and the enzyme activity had a dose–effect relationship with the V2O5 concentration. MDA analysis indicated that lipid peroxidation in earthworms mainly occurred at the early stage and was eliminated slowly in the later stage during the test time. In addition, the BAFs were much less than 1, which indicated that V2O5 did not easily accumulate in earthworms, and the BAF was positively correlated with the exposure time and negatively linearly correlated with the V2O5 concentration in the soil. These results indicated that the bioconcentration and metabolic mechanism of V2O5 in earthworms differed with the different exposure concentrations, and bioaccumulation became balanced after 14–28 days in earthworms exposed to a relatively lower dose of V2O5. The analysis of the integrated biomarker response (IBR) index indicated that the trends of IBR values were positively related to the changing V2O5 concentration, and the IBR index could reflect the organism’s sensitivity to the external stimulus of V2O5. The toxicity of V2O5 is mainly caused by V5+, which is also an important factor in formulating guidelines regarding vanadium levels in soil, and the test earthworm species (Eisenia fetida) is a sensitive biological indicator for risk assessments of vanadium oxidation in the soil.


Introduction
With the development of modern industry, vanadium and vanadium pentoxide (V 2 O 5 ) have been used in various products. Vanadium is primarily employed in steel to improve its strength and corrosion resistance. It is also used in many other products, such as pigments, medicines and catalysts (Huang et al. 2015;IARC 2006;Schlesinger et al. 2017). With tougher requirements for oxynitride control for industries and vehicles, vanadium-based selective catalyst reduction (V-SCR) has been widely used in recent years (Hoej and Beier 2014). Therefore, a considerable amount of vanadium compounds could enter the soil, water or air and may cause environmental pollution with the extensive Responsible Editor: Chris Lowe use of vanadium products (Hindy et al. 1990;Imtiaz et al. 2015;Xiao et al. 2017). The mean soil background level of vanadium is approximately 36 mg/kg in the Unites States, 82 mg/kg in China, and 94 mg/kg in Japan (Aihemaiti et al. 2020). China is one of the main producers of vanadium, and mean soil vanadium levels in various provinces range from 54.6 to 118.4 mg/kg (Yang et al. 2017). V 2 O 5, as the main form of high valance vanadium oxide, is suspected to be carcinogenic to humans according to the China National Occupational Health Standard GBZ 2.1-2019 (NHC 2019), and the International Association for Research on Cancer (IARC) classifies the compound into Group B2, members of which may cause cancer to humans (Assem and Levy 2009;IARC 2006). The American Conference of Governmental Industrial Hygienists (ACGIH) classified V 2 O 5 as an A3 compound, members of which are confirmed animal carcinogens with unknown relevance to humans (ACGIH 2009). Acute exposure of humans, primates and rodents to V 2 O 5 has been confirmed to have adverse respiratory effects (USEPA 2011), human exposure to dust containing high levels of V 2 O 5 could induce wheezing, chest pain, bronchitis, and impaired lung function symptoms (Hauser et al. 2001), and the toxicity of V 2 O 5 to plants has also been studied (Fortoul et al. 2014;Smith et al. 2013). However, studies on the toxicity and risk to terrestrial soil animals have rarely been reported.
Soil organisms are generally exposed to various environmental pollutants, and earthworms are important soil mollusc species that play key roles in improving soil fertility (Lee 1985;Kula and Larink 1997). Earthworms have been exposed to chemicals for soil risk assessment due to their natural role in recycling nutrients for the food chain (Gestel and Dis 1988), and earthworms are recognized as test organisms for soil risk evaluation under the guidelines from the Organization for Economic Co-operation and Development (OECD 2004). As a group of biomarkers of external chemical toxicity, the antioxidant defence system, which is made up of antioxidant enzymes and nonenzymatic substances, is often used in early warning and ecotoxicity stress evaluations of low-concentration pollutants (Oost et al. 2003;Li et al. 2019). Antioxidant enzymes such as superoxide dismutase (SOD) and catalase (CAT) play a main role in eliminating or detoxifying free radicals and reactive oxygen species (ROS), which result in environmental contamination (Tripathi et al. 2006;Asagba et al. 2008). Moreover, as an important index of lipid peroxidation, the level of malondialdehyde (MDA) is also a sensitive indicator of cellular oxidative damage (Avci et al. 2005;Chen et al. 2012;Aihemaiti et al. 2020). Some evidence has shown that antioxidant enzymes are effective biomarkers of toxicological stress responses (Song et al. 2009;Xue et al. 2009;Zhang et al. 2009). Considering the potential ecotoxicity of vanadium in soil, the toxic stress effects and bioaccumulation of vanadium in earthworms should be tested to predict the possible effects of soil contamination on the food chain.
In this study, the mechanism of ecotoxicity stress caused by V 2 O 5 in the earthworm, including the analysis of the integrated biomarker response (IBR) index, the activity of catalase (CAT) and superoxide dismutase (SOD), the content of malondialdehyde (MDA), and the bioaccumulation factor (BAF), was investigated. The toxic effects of V 2 O 5 on the terrestrial earthworm (Eisenia fetida) were measured to reveal the acute and subchronic toxicity, including oxidative stress, of V 2 O 5 to the species and to understand the bioaccumulation process of V 2 O 5 in soil organisms.

Test organism
Earthworms (Eisenia fetida) were purchased from a commercial breeder in Beijing, China. Adult worms with welldeveloped clitella, above 2 months of age, and body weight of 300-500 mg were used for toxicity tests. Before the experiment, all test earthworms were bred on dry cow dung for 7 days in test artificial soil. The characteristics of the artificial soil were as follows: temperature of 20 ± 1 °C and moisture content of 35 ± 2%.

Test soil
The preparation of an artificial soil was conducted according to the OECD guidelines for testing chemicals No. 207 and No. 222 (OECD 2004). The artificial soil consisted of a mixture of 70% quartz sand, 20% kaolin and 10% Turfy soil (measured by dry mass), adjusted with carbonate to a pH of 6.0 ~ 6.5 and adjusted with ultrapure water to a moisture content of approximately 35% of the maximum water holding capacity. The concentration of V 2 O 5 in artificial soil was below the detection limit.

Eco-toxicity test
According to the guidelines for earthworm acute toxicity tests (OECD 2004), V 2 O 5 exposure was performed in artificial soil based on pre-experiment results. Appropriate amount of V 2 O 5 was added to 10 g fine ground quartz sand, and mixed thoroughly. Then, 490 g (dry weight) artificial soil was gradually added to the above mixture to obtain the required V 2 O 5 concentrations in a series as follows: 0, 2. 94, 5.00, 8.50, 14.45, 24.57, 41.79, 71.05 mg/kg soil. The moisture content was adjusted to 35% before the soil was put into the test boxes. Each treatment group included 3-4 repeats. Then, 10 earthworms were added to 500 g of soil and cultivated in an artificial climate incubator under controlled temperature (20 ± 1 °C), humidity (80%), light intensity (500-600 lx), etc. with deionized water blank groups. The toxicity test period was 28 d, and the soil moisture content was kept at the initial level by spraying ultrapure water regularly. During the exposure period from 14 to 28 days, the test earthworms were counted to derive the 50% and 10% lethal concentration (LC 50 /LC 10 ) values for exposure to V 2 O 5 . The mortality of the earthworms in the control were less than 10% during the test.

Biochemical enzyme assays
The enzyme activity measurements were taken at the same time as in the above earthworm toxicity test. During the exposure periods of 3, 7, 14, and 28 d, the test earthworms were collected from each treatment group for biochemical and metal determination. The gut contents of selected earthworms were emptied by placing the animals in separate beakers with filter paper for 24 h. Earthworm tissues were extracted and homogenized, and the SOD and CAT activities and MDA content were measured. The extracts were obtained from the corresponding enzyme activity test kits with extraction solution, and the test earthworms frozen in liquid nitrogen were removed and homogenized in an icecold extraction solution at a ratio of earthworm weight to extraction solution of 1:10 (w/v). The homogenates were centrifuged at 8000 g for 10 min at 4 °C. After centrifugation, the supernatant was removed for determination of the SOD and CAT activity and MDA content. CAT and SOD activities were measured as described previously (Lyttle and DeSombre 1977;McCord and Fridovich 1969), and lipid peroxidation was determined by measuring the MDA content with the thiobarbituric acid test (Ohkawa et al. 1979).

Analysis of V 2 O 5 in soil and earthworms
The collected earthworms were cleaned and put on damp filter paper for 24-48 h to empty their intestines (Cao et al. 2017). The worms and soils were freeze-dried and stored at -80 °C. The concentration of vanadium in the soil and worms was determined according to the following steps: the soils were soaked in 5 ml HNO 3 and 1 ml HCl overnight and digested at 180 ℃ for 8 h (MEE 2016). During this period, the H 2 O 2 was decreased, the samples were cooled, and the digested solution was stabilized to a volume of 50 ml with HNO 3 . The earthworms were soaked in 5 ml HNO 3 and 2 ml HClO 4 overnight, digested at 180 ℃ for 8 h, cooled, and the digested solution was stabilized to a volume of 50 ml with HNO 3 (Ji et al. 2011). Finally, both digested solutions were analysed by inductively coupled plasma-mass spectrometry (ICP-MS 7900, Agilent). In addition, the amount of vanadium measured was converted to the concentration of vanadium pentoxide according to the test substance. For quality assurance and quality control (QA/QC), procedural blanks, sample duplicates, and recovery rate tests were performed during sample analysis. The CVs of triplicates ranged from 3.31% to 9.82%, and the recovery rate ranged from 92.7% to 101.7%. The limit of detection (LOD) was 0.04 mg/kg, which was defined as 3-fold standard deviation (SD) of 7 blanks.

Statistical analysis
Data were first tested for homogeneity of variances and expressed in terms of the mean ± standard deviation (SD). One-way analysis of variance (ANOVA) was used for statistical analysis. The least significant differences (LSD) test was used to determine significant differences between different treatments, and P < 0.05 was considered statistically significant using SPSS 20.0 for data analysis. The calculation of the integrated biomarker response (IBR) index was based on a previously published method (Pytharopoulou et al. 2008;Oliveira et al. 2010). Test data were standardized, star charts were used to visualize the integrated results of the biomarkers, and the IBR values were calculated and analysed; the larger the IBR value was, the more the biological influence. The IBR index was calculated with the following steps: (1) calculation of the total mean value (m), the standard deviation (s) of the test time points, and the mean value (X) of the enzyme biomarker at each time point; (2) the mean value X of the time point was normalized by the formula Y = (X-m)/s; (3) if the test biomarker was activated, then Z = Y, and if it was not activated, then Z = -Y, and |Xmin| represented the absolute minimum value of the biomarker uniformity data for the time point. The biomarker score S i for the time point was calculated as S i = Z +|Xmin|. The S i value of the biomarker for a specific period is represented by the length of the radiation in the star chart (Li et al. 2020). The IBR value for a specific period was obtained by calculating the area of the star chart, which shows the total area A i of the triangle enclosed by the radiation of the adjacent biomarker, and analytical indices were described as follows: where n is the number of selected biomarkers, α is the angle between two nearby radiation lines, and α = 2π/n, S i+1 = S i .

Acute and subchronic toxicity of V 2 O 5 to earthworms
It is important to perform toxicity tests in the laboratory to correctly assess the chemical risks associated with contaminated sites. In this work, all test earthworms had a mortality of less than 5%, which occurred in the blank group after 14 days of exposure, and no new deaths occurred after 14-28 days. The earthworms exposed to V 2 O 5 showed a correlation between concentration and the lethal effects, as shown in Fig. 1. The acute 50% lethal concentration (LC 50 ) (1) β = aratan[sinαS i+1 ∕(S i − cosα S i+1 )] (2) A i = S i ∕2sinβ (S i cosβ + S i+1 sinβ) Ai of the V 2 O 5 to test earthworms was 21.96 mg/kg in 14 days, and the earthworm mortality increased from 56.67% to 100% with V 2 O 5 concentrations from 24.57 mg/kg to 41.76 mg/ kg. Additionally, the subchronic LC 50 was 18.71 mg/kg in 28 days. The acute and subchronic toxicity values of earthworms were calculated with lethal effects as an endpoint (Fig. 1). The acute toxicity value of LC 10 at 14 days was 7.32 mg/kg, and the subchronic toxicity value of LC 10 at 28 days was 6.28 mg/kg. However, the soil quality guideline value for vanadium released by the Canadian Council of Ministers of the Environment for environmental health (CCME 2007) was 130 mg/kg, which is less than the interim soil quality criterion. Therefore, this test result showed that the toxicity stress from vanadium oxide V 2 O 5 was greater than that of vanadium to the soil organism. It is known that vanadium (V) exists in two oxidation states in soil: V 4+ and V 5+ (Khan et al. 2011). The results also showed that the toxicity stress of vanadium oxide was mainly caused by V 5+ from V 2 O 5 . Furthermore, the vanadium oxide state of V 5+ is the main toxic form, and this is an important factor when formulating the guideline values for vanadium in soil. Moreover, the test earthworm species (Eisenia fetida) may be a sensitive biological indicator for toxicity assessments of vanadium oxide forms in the soil. Table 1 shows that the measured concentrations differed from the spiked concentrations by less than 20%, and the V 2 O 5 concentration ratio from earthworms to soil was used to calculate the bioaccumulation factor (BAF). The V 2 O 5 concentrations in earthworms ranged from 0.12 to 0.73 mg/ kg, becoming higher with increasing exposure time and spiked concentration after 3, 7, 14, and 28 d of exposure. However, the concentration in earthworms decreased first and then increased with longer exposure times in some lower concentration groups, such as 2.94, 5.00, and 8.50 mg/kg. Whereas, in the higher concentration group, such as the 14.45 and 24.57 mg/kg group, the bioaccumulation increased with the exposure time. In addition, the content of V 2 O 5 in

Biochemical enzyme effects of V 2 O 5 on earthworms
Early biochemical reactions are important information for evaluating the potential adverse impact of chemicals on the environment (Gao et al. 2007). The biochemical effects of V 2 O 5 exposure on the activities of two antioxidant enzymes (SOD and CAT) and the MDA content in earthworms are shown in Fig. 2. Exogenous pollutants can induce the production of ROS in organisms. ROS can cause oxidative stress, which leads to protein denaturation, enzyme inactivation, biofilm damage, DNA replication errors, and even cell death (Silva et al. 1999;Halliwell and Whiteman 2004;Asagba et al. 2008), as the ROS level exceeds the ability of the antioxidant enzyme defence system to resist external pressure. Figure 2 (1) shows that the SOD activity of earthworms exposed to V 2 O 5 decreased on the 3rd day. Then, the activity increased in the test groups on the 7th and 14th days, except that the higher concentration group of 24.57 mg/kg had a slight decrease on the 14th day, and all concentration groups showed inhibition of SOD activity on the 28th day. Superoxide anion free radicals are catalysed by SOD enzymes to convert hydrogen peroxide (H 2 O 2 ) to oxygen in cells for removal, so this enzyme plays an important role in protecting the dynamic balance of oxygen free radical  (3). Data are presented as the means ± standard deviations (SDs). Capital A represents significant differences compared with the blank control group at p < 0.01. Lowercase letters represent significant differences compared with the a (0 mg/kg), b (5.00 mg/kg), c (14.45 mg/kg), and d (8.50 mg/kg) control groups at p < 0.05 generation in organisms (Shao et al. 2018). In this study, SOD activity was inhibited by V 2 O 5 exposure for the first 3 days. Then, SOD activity was increased at almost all doses at 7-14 days and inhibited after 14-28 days of exposure. It can be speculated that this reactive mechanism may be related to the effect of vanadium on the metal prosthetic group of SOD, which could affect SOD activity by changing the molecular structure of the enzyme protein. As the test time increased, the oxidative stress of vanadium increased on the 7th and 14th days, which might have caused increases in SOD products. However, the regulation of enzyme activity in the test earthworm was limited during V 2 O 5 exposure for 14-28 days, it did not increase indefinitely, and finally, the enzyme activity was inhibited. The enzyme activity was generally attributed to the special physiological effect of vanadium on the test organisms, and the enzyme activity tendency indicated that SOD activity had a certain relationship with the dose at the test time.
As shown in Fig. 2 (2), CAT activity was increased by V 2 O 5 at relatively lower doses of V 2 O 5, such as 5.00-14.45 mg/kg, and inhibited at a higher dose of 24.57 mg/kg at the test time of 28 days. The response trends of the CAT activity in different test groups were similar at 5.00-14.45 mg/kg from 3-28 days. CAT was inhibited in the first 3 days, induced from 7 to 14 days with a peak concentration on the 14th day, and then inhibited on the 28th day. The main product in the catalysis process of SOD is hydrogen peroxide, which has an adverse effect on cells (Saint-Denis et al. 2001). H 2 O 2 is eliminated or converted into H 2 O and O 2 by the antioxidant defence effects of CAT. The CAT activity of the test earthworm had a response mechanism similar to that of SOD under V 2 O 5 exposure in the soil. Moreover, the CAT enzyme activity was promoted at relative lower-dose exposure over the 28 days process compared with the inhibitory effect in the higher-dose V 2 O 5 (24.57 mg/kg) group from 7-28 days. In addition, the CAT activity showed a decreasing trend within the same exposure time. The lower-dose (2.94 mg/kg) group had a larger peak than the higher-dose (24.57 mg/kg) group with lower CAT activity. This result indicated that the V 2 O 5 dose had a significant influence on the activity of CAT, and CAT activity was inhibited in a relatively short time at a higher dose.
The level of malondialdehyde (MDA) is a main indicator of lipid peroxidation, and the oxidative stress induced by V 2 O 5 is presented in Fig. 2 (3). The MDA content from the higher-dose groups such as 14.45 mg/kg and 24.57 mg/kg increased noticeably on the 3rd day by 2.69-fold and 2.81fold compared to the control groups. MDA showed a positive trend with increasing V 2 O 5 concentration in the same time period. In the subsequent exposure time (3-28 days), the MDA content from the same V 2 O 5 concentration group showed a downwards trend, while it still increased compared to the control group. The MDA content trend was related to the delayed increase in SOD and CAT activity, as shown in Fig. 2 (1) and (2). It could be speculated that SOD or CAT decreased with ROS or MDA increased under the V 2 O 5 exposure early on first 3 days, then the antioxidant defence system cleared the ROS and the MDA decreased with the subsequent V 2 O 5 exposure time. In addition, the MDA content trends indicated that the lipid peroxidation of earthworms under V 2 O 5 exposure mainly occurred during the early stage and gradually weakened in the later stage. This may be an effective process in which antioxidants or detoxification system enzymes scavenge free radicals and induce self-repair in earthworms. During the same exposure time, the MDA content showed an increasing trend, which indicated that the V 2 O 5 dose could obviously influence the MDA content, and MDA was increased in a relatively short time at a higher dose.

Analysis of the IBR
Because various enzymes from different stress conditions showed inhibitory or inducing effects and the responses were not synchronized, the sensitivity of enzyme activity was different in the organism system. It was difficult to quantitatively evaluate the comprehensive effect of lethal toxicity by considering only a single enzyme activity. The integrated biomarker response (IBR) index can be quantified to describe the incorporated biological effects of various biological indicators by integrating relative enzymes or proteomic biomarkers. The IBR index method has been extensively used to analyse the impact of environmental pollutants on a variety of biomarkers to determine the impact of environmental contaminants on organisms (Liu et al. 2015;Cao et al. 2019). In this study, Fig. 3 shows that the IBR index assessment method was appropriate to integrate the active effects of the three selected biomarkers (SOD, CAT and MDA) in different time periods and to evaluate the toxicity of V 2 O 5 to earthworms. In the same time periods, the IBR values were positively related to the changing V 2 O 5 concentrations, and the IBR value was the largest at the highest test concentration of 24.57 mg/kg. Over a range of time periods, the trend of IBR value had no distinct linear relationship with the test time extension, and it was relatively higher on the 3rd day and the 28th day. As the IBR values were significantly higher than those of the control group, the test earthworms had a significant stress response to V 2 O 5 exposure, and the higher the V 2 O 5 concentration was, the greater the toxicity to the organism. To a certain extent, this IBR index could also reflect the organism's sensitivity to defensive stress from external stimuli of V 2 O 5 . At the same concentration, the IBR values on the 7th and 14th days were lower than those on the 3rd day. This may have been due to the enzyme activity and MDA response to V 2 O 5 stress at the test time. The main reaction process predicted that the enzyme activity gradually recovered after 3 days of V 2 O 5 exposure, while the IBR value decreased to a certain extent. The IBR value increased after 14 days, which indicated that the organism's enzymes had an obvious stress response under prolonged V 2 O 5 exposure. As shown in Fig. 3, the IBR value was the largest on the 14th day under the highest concentration of 24.57 mg/kg. The reason for this phenomenon may be that the relatively higher stress and longer exposure time induced a difference in the sensitivity of the enzymes to the V 2 O 5 stress response. In terms of the relationship between the time effect and dose effect, the higher the concentration of V 2 O 5 was, the larger the biological toxicity impact. However, at the same exposure concentration, the largest IBR impact was not observed at the longest test time (the 28th day), which indicated that the V 2 O 5 concentration had a more visible influence on the oxidative stress of the earthworms than time.

Conclusion
This investigation can contribute to the evaluation of the adverse effects of V 2 O 5 on the soil environment. The results showed that the acute and subchronic toxicity values of V 2 O 5 to earthworms were 21.96 mg/kg (LC 50 , 14 days) and 6.28 mg/kg (LC 10 , 28 days), respectively. The toxicity of V 2 O 5 is mainly caused by V 5+ , which is an important factor when formulating guideline values for vanadium oxide in soil. Earthworms (Eisenia fetida) may be a sensitive biological indicator for risk assessments of vanadium oxide in the soil. SOD and CAT were synchronously induced or inhibited within the time period, and the enzyme activity had a dose-effect relationship with the V 2 O 5 concentration. MDA analysis indicated that lipid peroxidation of V 2 O 5 in earthworms mainly occurred at the early stage and was eliminated slowly in the later stage.
The BAF values indicated that V 2 O 5 did not easily accumulate in earthworms. The BAF of V 2 O 5 was positively correlated with exposure time and negatively linearly correlated with the V 2 O 5 concentration in the soil. Bioaccumulation characteristics in earthworms may be a sensitive predictor for describing the pollution risks posed by V 2 O 5 . The IBR method was appropriate to assess the active effects of the selected biomarkers (SOD, CAT and MDA) in different time periods, and it could evaluate the toxicity of V 2 O 5 to earthworms. For the same test periods, the trends of IBR values were positively related to the changing V 2 O 5 concentrations.