Lysosomal membrane response of the earthworm, Eisenia fetida, to arsenic species exposure in OECD soil

Abstract

Different arsenic species always coexist in environmental soils and show various degrees of toxicity. However, the biochemical impact on earthworms under exposure to different arsenic species remains unknown. In this study, earthworms (Eisenia fetida) were exposed to Organization for Economic Cooperation and Development (OECD) artificial soil contaminated by arsenite [As(III)], arsenate [As(V)], monomethylarsonate (MMA) and dimethylarsinate (DMA) respectively for 70 days. Aimed at the comparison of toxicity, the exposure concentrations of the four arsenic species were set at one-tenth of 14 d-LC₅₀. A biomarker of the lysosomal membrane stability, measured by neutral-red retention time (NRRT), was evaluated for toxicity assessment of the four arsenic species. Furthermore, the contents of total arsenic and arsenic species were analyzed in earthworms in order to not only evaluate the dose–response relationship between arsenic accumulation and NRRTs, but also to observe the effect of arsenic biotransformation on subcellular level toxicity. The results showed that there were clear dose–response relationships between the body burden of inorganic arsenic species and NRRTs, while the correlation coefficient between NRRT and As(III) was higher than that between NRRT and As(V). The NRRT assay results suggested that the toxicity of the four arsenic species was ranked as As(III) > As(V) > MMA > DMA. Therefore, arsenic speciation should be conducted in order to obtain accurate results during environmental assessment programs. Despite the low toxicity, the two organic arsenics could be transformed to highly toxic inorganic arsenic species through the demethylation process, after which adverse effects observable at the subcellular level in earthworms could thus be induced. Highly toxic inorganic arsenics were methylated and converted to less toxic organic species before excretion. Considered as the biotransformation pathway of arsenic in earthworms, the process mentioned above was validated in this study. Lysosomal membrane stability is a sufficiently sensitive biomarker for toxicity assessment of inorganic arsenic pollution, and especially suitable for the monitoring of As(III) species.

---

Introduction

Arsenic (As) is used in the process of hardening alloys and producing semiconductors, pigments, glass, pesticides, rodenticides and fungicides.¹ It is also used as an ingredient in drugs for the treatment of many diseases.² Because of its usefulness and exploitation, arsenic is now distributed throughout the environment in varying concentrations. It has been commonly and copiously added to soils as biocides/pesticides, wood preservatives or industrial wastes.³ As a consequence of its potential hazard to animals and plants through direct toxic action or via food-chain accumulation, arsenic has not only been widely investigated in many environmental monitoring projects, but has also been listed in a large number of quality standards and safety standards around the world.^4–10 However, with no differentiation for different arsenic species, nearly all the arsenic contents regulated in these documents refer to total arsenic or the content of one arsenic species.

Arsenic in environmental and biological systems appears in many different chemical forms that differ with regard to their physical, chemical and biological properties.¹¹ Due to the natural metabolic processes in the environment, arsenic occurs as different inorganic and organic species. The trivalent arsenic [arsenite, As(III)] and the pentavalent arsenic [arsenate, As(V)] are widely present in soil, freshwater and marine environments and are soluble over a wide range of pH and Eh conditions.³ In oxidizing environmental conditions, As(V) species are more stable and predominant, whereas in reducing environmental conditions, As(III) species are predominant. Inorganic arsenic species may be methylated as monomethylarsonate (MMA), dimethylarsinate (DMA) and trimethylarsine oxide (TMAO) by microorganisms, humans and animals.¹² Previous studies have shown that the toxicity of arsenic is determined by its chemical species.¹³ Inorganic arsenic As(III) and As(V) are the most toxic species, while As(III) is more toxic than arsenate As(V), the toxicity of organic arsenical species is lower, while trimethylated species are recognized to have the least toxicity.^14,15 As(III) can be toxic through, on the one hand, its interaction with sulfhydryl groups of proteins and enzymes to denature them within cells, and on the other hand, an increase of reactive oxygen species (ROS) in cells, consequently causing cell damage.¹⁶ As(V) is also known to inhibit many enzymes in organisms. Furthermore, because arsenate is hydrolyzed easily in the cell, it can prevent the subsequent transfer of phosphate to adenosine diphosphate (ADP) to form adenosine triphosphate (ATP) and thus deplete the cell of its energy.¹⁷ Although the toxic mechanisms of MMA and DMA are still unknown, they could be transformed into highly toxic inorganic arsenic species through the demethylation process, which can have a toxic influence on organisms. Therefore, different arsenic species should be investigated separately when studies such as toxicity assessment and bioaccumulation analysis are conducted for the arsenic element.

In many ecosystems, earthworms are key species in decomposer communities, thus bringing about great impacts on decomposition activity, nutrient mineralization and primary production.¹⁸ It is well known that earthworms can be affected by a range of contaminants while they can also accumulate some of these, e.g. metals/metalloids.^19,20 Therefore, they have been recommended as one of the most suitable bioindicator organisms for risk assessment in soil.^21,22 Eisenia fetida was chosen for this study due to the standardization of acute and chronic ecotoxicological assays. Considered as a suitable model species, it has been widely utilized in standard tests, namely for testing the adverse effects of chemicals on soil organisms.^23,24

There has been a noticeable increase in the use of biomarkers to assess the impacts of contaminants on terrestrial ecosystems over the past few years.^25–27 Theoretically, each type of biological response, ranging from the molecular to the community level, can be considered as a biomarker.²⁸ Accumulated metals and metalloids may damage earthworms at the subcellular level, after which organisms can be affected at higher biological levels such as reproduction and growth.^29–31 Lysosomes are among the subcellular systems which could be directly affected by pollution.^32,33 Therefore, the stability of lysosomal membranes in coelomocytes has gained attention as a subcellular biomarker of contamination stress.^24,34–36 There is one method that can be used to determine lysosomal membrane stability, namely the neutral-red retention time (NRRT) assay, which has already been proved to be reliable, dose-related and practical for use in marine³⁷ as well as terrestrial systems.^38,39 Although arsenic is the second most common inorganic constituent among soil contaminants, the NRRT assay has not been used to test the toxicity of different arsenic species in earthworms so far.

In this study, earthworms (E. fetida) were exposed to OECD soil contaminated by four common arsenic species, including As(III), As(V), MMA and DMA. NRRT was selected as a potential biomarker in earthworms for toxicity assessment of the four arsenic species. The aim of this study was, therefore, to investigate and evaluate the possible use of lysosomal membrane response as a biomarker of the toxicity of different arsenic species in E. fetida by NRRT assay. Furthermore, in order to gain an understanding about the effects of arsenic bioaccumulation and biotransformation on NRRT, the contents of total arsenic and arsenic species in earthworms were thus characterized. Through these observations and related findings, a useful scientific basis will be established for soil ecological risk assessment of different arsenic species in terrestrial ecosystems.

Materials and methods

Soils and chemicals

Artificial soil, as specified in the OECD guidelines,²³ was used in this study. The composition of the artificial soil (dry weight) was 70% quartz sand, 20% kaolinite, and 10% finely ground sphagnum peat, with pH adjusted to 6.5 by addition of calcium carbonate.

Standard solutions of As(III) (1.011 μmol mL⁻¹), As(V) (0.233 μmol mL⁻¹), MMA (0.335 μmol mL⁻¹) and DMA (0.706 μmol mL⁻¹) were supplied by the China Standard Certification Center. Ultrapure water (18 MΩ) was obtained by using a Milli-Q water purification system (Millipore, USA). All glassware was cleaned by using 10% (v/v) nitric acid (Merck KGaA, Germany), followed by multiple rinses with ultrapure water. Reagents used in NRRT assays were obtained from Sigma-Aldrich China Co. (Shanghai, China), including neutral red and dimethyl sulfoxide (DMSO, >99% purity). All the reagents used for total arsenic and arsenic speciation analysis were of analytical reagent grade or better, while ultrapure water was used throughout. Working standard solutions of As(III), As(V), MMA and DMA were prepared by stepwise dilution of standard solutions. A NaBH₄ solution of 1.5% (m/v) in 0.5% NaOH (m/v) solution was prepared daily with sodium tetrahydroborate (Beijing Chemical Co., China). A magnesia mixture was prepared by dissolving 55 g of MgCl₂·6H₂O in 500 mL of ultrapure water, adding 140 g of NH₄Cl and 130 mL of concentrated ammonia solution, followed by dilution with water to 1 L. The phosphate solution was prepared by dissolving 0.158 g of K₂HPO₄ in 100 mL of ultrapure water. The chemicals used to prepare the magnesia mixture and phosphate solution were analytical grade reagents obtained from Beijing Chemical Co. (Beijing, China). Other chemicals used in the experiment include suprapur hydrochloric acid (Beijing Chemical Co., China), thioglycolic acid ethyl ester (Shanghai Chemical Reagent Company of the Medicine Group, China), and cyclohexane (Chemical Experimental Factory of Tianjin University, China).

Earthworms

The study protocol was approved by the Chinese Association for Laboratory Animal Science. Adult earthworms E. fetida with well-developed clitella were purchased from a local commercial earthworm farm company in Jinan, China. The earthworms were collected from a synchronized culture with the same age for the control group and each exposure group. The selected earthworms possessed a weight ranging from 350 to 450 mg, and were acclimated for 7 d to the artificial soil substrate under controlled conditions (20 °C, 12 h light/12 h dark, 80% humidity) before the experiment.

Experimental design

The toxicity levels of different arsenic species are quite different. Hence, the concentrations of spiked soils were designed based on the 14 d median lethal concentration (LC₅₀) of the four arsenic species for the purpose of comparing their toxicity levels, with these values obtained in an artificial soil test following the OECD guideline 207.²³ As shown in the results of our pre-experiment, the LC₅₀ for As(III), As(V), MMA and DMA following the standard toxicity test was 293, 352, 3425 and 3730 mg kg⁻¹, respectively. Therefore, based on the arsenic concentration in typical contaminated soils in China⁴⁰ as well as the limit value of arsenic in the Chinese Environmental Quality Standard for Soils (GB 15618-1995), accordingly, the concentrations of spiked soils were set at one-tenth of the 14 d-LC₅₀, namely 29.3, 35.2, 342.5 and 373.0 mg kg⁻¹ for the four arsenic species.

Spiking solutions were prepared using standard solutions and added to four soil samples so as to give soil arsenic concentrations of 29.3 mg kg⁻¹ As(III), 35.2 mg kg⁻¹ As(V), 342.5 mg kg⁻¹ MMA and 373.0 mg kg⁻¹ DMA, along with 70% water holding capacity. One thousand grams of As-spiked soil was placed in each polyethylene plastic container (30 × 25 × 20 cm) for 4 days prior to experimentation. To guarantee smooth ventilation, the containers were covered with a lid punched with holes. The test was maintained at 20 °C, 80% humidity and with a 12 h light/12 h dark cycle by an artificial climate chamber. The soil of the control group was prepared in a similar way with no pollutants added. Before introduction into the soils, earthworms were rinsed with ultrapure water to remove adhering soils or particles and then blotted dry. Forty earthworms with nearly equivalent masses were added into each container, with five earthworms removed from each treatment at t = 7, 14, 28, 42, 56 and 70 days following soil exposure. An appropriate amount of an arsenic free diet (3–5 g per box) was applied on the soil surface at the start of the test, followed by supplementation on the weekly basis. During the experiment, dead earthworms were removed immediately and the mortality of earthworms was less than 10% for each treatment.

Neutral-red retention time (NRRT)

Lysosomal membrane stability was assessed with the NRRT assay as described by Weeks and Svendsen.³⁸ Briefly, a volume of 20 μL of coelomic fluid containing coelomocytes was collected by inserting a hypodermic needle directly into the coelomic cavity posterior to the clitellum of the earthworm, after which the fluid was mixed with an equal volume of earthworm physiological Ringer solution. The stock solution of neutral red was obtained by dissolving 20 mg of neutral red powder in 1 mL DMSO. Next, the neutral red working solution of 80 μg mL⁻¹ was prepared by diluting 10 μL of the stock solution in 2.5 mL earthworm physiological Ringer solution. The working solution was renewed on an hourly basis during the measuring process, thus avoiding crystallization. Then, the 20 μL of coelomic fluid was placed onto a clean microscope slide for 30 s prior to the application of an equal volume of the neutral red working solution and a cover slip. Each slide was observed every 2 min under a light microscope at 400× magnification. The observation was stopped when the proportion of fully stained cytosols was over 50% of the total number of cells counted. This time was recorded as the NRRT. The coelomic fluid from each individual earthworm was collected and processed separately for each treatment at each sampling time point.

Total arsenic analysis

After the extraction of coelomocytes, each earthworm used to analyze arsenic body burden in tissue was placed on a moist filter paper for 48 h to allow for defecation. After rinsing the earthworms with ultrapure water several times, the earthworms were killed with liquid nitrogen before being kept at −80 °C for 24 h. Then the earthworm samples were dried by freeze drying at −40 °C for 72 h, and ground with an agate pestle and mortar to a fine powder. For total arsenic analysis, twenty milligrams of earthworm powder was transferred into a vial. Two milliliters of nitric acid was then added and heated at 80 °C for 8 h. The extract was cooled and diluted to 10 mL with ultrapure water, after which it was analyzed for total arsenic content using hydride generation atomic fluorescence spectrometry (HG-AFS).

A double channel atomic fluorescence spectrometer (AFS-920, Beijing Titan Analytical Instrument Co., Beijing, China) was used for total arsenic analysis. A high-intensity arsenic hollow cathode lamp (General Research Institute for Nonferrous Metals, Beijing, China) was employed. Argon (>99.995%, Jinan Gas Factory of Shandong Province, China) was used as a protective and purging gas. Table 1 presents a summary of the operating parameters for HG-AFS determination of total arsenic. The limit of detection (LOD) in solid samples was 0.06 mg kg⁻¹ for total arsenic. The quality assurance (QA) was checked by using a certified reference material (Scallop tissue, GBW 10024) provided by the Institute of Geophysical and Geochemical Exploration of China. The quantified mean value (n = 9) obtained for the reference material was 3.41 ± 0.36 mg kg⁻¹, which was in accordance with the certified value (3.6 ± 0.6 mg kg⁻¹). The concentrations of total arsenic were expressed as mg kg⁻¹ of dry weight.

Table 1 HG-AFS operating parameters

Parameters	Settings
Arsenic hollow cathode lamp	193.7 nm
Lamp current	60 mA
Atomizer height	8 mm
Integration time	3 s
Time delay	3 s
Negative high voltage of photomultiplier	270 V
Reductant	1.5% (m/v) NaBH4 in 0.5% (m/v) NaOH
Carrier argon flow	1000 mL min^−1
Shield argon flow	800 mL min^−1
Reading mode	Peak area

---

Arsenic speciation analysis

A microwave-assisted extraction method was adopted for the arsenic speciation analysis in earthworm samples.⁴¹ Twenty milligrams of homogenized earthworm powder was weighed and transferred into a PTFE microwave digestion vessel. Subsequently, 10 mL of ultrapure water was added into the vessel. The arsenic species were extracted by using a high-pressure microwave system (XT-9900A, Xintuo analytical instruments Co., China). The temperature program was as follows: set at 35 °C and held for 1 min initially, then increased up to 75 °C at 20 °C min⁻¹ and held for 3 min. After cooling to room temperature, the extract was filtered through a 0.45 μm syringe filter. The final solution was analyzed for the four arsenic species with the employment of a modified derivatization-gas chromatography (GC) method developed by Wang and Cui.⁴²

For the analysis of DMA, MMA and inorganic arsenic, as the sum of As(III) plus As(V), one aliquot of sample was derivatized by thioglycolic acid ethyl ester (TGE) and analyzed by GC (GC-2014, Shimadzu, Japan), with the procedure conducted as follows. The aqueous samples (5 mL) were placed in 10 mL capped glass tubes, followed by the addition of 50 μL 1 mol L⁻¹ HCl and 150 μL TGE subsequently. The tubes were closed and shaken intensively for 2 min at ambient temperature. After the addition of 2 mL of cyclohexane, shaking was continued for another 2 min. Finally, 2 μL of the upper organic phase of the reaction mixture was injected into the GC injection port. For the analysis of As(III), magnesia mixture and phosphate solution were added to another aliquot of sample to remove As(V) by co-precipitation with magnesium ammonium phosphate.⁴³ The precipitate was filtered off and As(III) was determined in the filtrate following the same procedure. The content of As(V) was calculated as the difference between total inorganic arsenic content and As(III) content. Table 2 summarizes the operating parameters of the GC system. The LOD in solid matrices was 0.15, 0.3 and 1.2 mg kg⁻¹ for DMA, MMA and inorganic arsenic species, respectively. The QA was checked by recovery experiments for six different earthworm samples spiked with the four arsenic species at three concentration levels (2.5, 5 and 15 mg kg⁻¹). The average recoveries ± standard deviations of As(III), As(V), MMA and DMA were 94.3 ± 2.8, 95.2 ± 5.6, 96.1 ± 6.3 and 98.1 ± 10.5, respectively. Moreover, there was no evidence of interconversion between species found during the analysis process, proving that the analytical methodology proposed in this study was accurate and reliable for the analysis of the four arsenic species in earthworm samples. The concentrations of different arsenic species were also expressed as mg kg⁻¹ of dry weight.

Table 2 GC operating parameters

Parameters	Settings
Detector	Flame ionization detector (FID)
GC column	Restek Rtx-5 30 m × 0.25 mm, 0.25 μm
Temperature program	50 °C (0.5 min), 20 °C min^−1 to 230 °C, 40 °C min^−1 to 300 °C (1 min)
Carrier nitrogen flow	3.0 mL min^−1
Injector temperature	240 °C
Detector temperature	300 °C
Sample volume	2 μL
Injection mode	Splitless
Reading mode	Peak area

---

Statistical analysis

All statistical analysis in this study was carried out using the SPSS 18.0 software (SPSS Inc., Chicago, USA). The NRRT data were analyzed by one-way analysis of variance (ANOVA) a priori test for the differences among all means. Tukey’s post hoc test was used to determine the significance of differences in relation to the control and between specific groups. A logistic–sigmoid regression was used to fit a dose–response curve to the body arsenic concentration versus NRRT relationship. The probability level used for the statistical significance was P < 0.05. All data are presented as mean ± standard deviation (SD).

Results

Arsenic analysis of earthworms

The initial arsenic concentration in the earthworms was 5.56 ± 1.09 mg kg⁻¹ dry weight. During the 70 day experimental period, the total arsenic concentration of the earthworms in control group varied slightly between 4.95 and 6.68 mg kg⁻¹. As(V) was the only detectable arsenic species in the control group with the contents ranging from 4.42 to 6.30 mg kg⁻¹.

Table 3 shows the contents of As(III), As(V), MMA, DMA and total arsenic in the earthworms of the four treatment groups at the six sampling time points. The earthworms of group I and group II were exposed to soil spiked with 29.3 mg kg⁻¹ As(III) and 35.2 mg kg⁻¹ As(V), respectively. The total arsenic body burden of the earthworms was obviously increased by the increase of exposure time. When the earthworms were exposed to the two inorganic arsenic species for 7 days, the total arsenic levels in the tissue exceeded those in the soil, indicating the bioconcentration of arsenic in earthworms. The earthworms of group III and group IV were exposed to soil spiked with 342.5 mg kg⁻¹ MMA and 373.0 mg kg⁻¹ DMA, respectively. As shown from the total arsenic content of the earthworms, there was a quite different trend from the other two exposure groups: both increased at the beginning and then decreased with exposure time, with the appearance of maximum values on day 14. During the whole experiment, the arsenic body burden of group III and group IV did not exceed the concentration in the soil, indicating the biotransformation and excretion of organic arsenic species in earthworms.

Table 3 Mean concentration ± SD (mg kg⁻¹ dry wt) of four arsenic species and total arsenic in E. fetida following 70 days exposure to arsenic contaminated soils (n = 5). Earthworms of group I, II, III, IV were exposed to As(III), As(V), MMA and DMA, respectively^a

Experimental group	Exposure days	Arsenic in E. fetida (mg kg^−1 dry wt)
		As(III)	As(V)	MMA	DMA	Total As
a N.D. = below limit of quantification or not detected.
Group I	7 d	19.23 ± 3.34	9.27 ± 2.17	N.D.	N.D.	31.20 ± 3.81
	14 d	23.96 ± 5.23	13.58 ± 4.50	1.15 ± 0.12	3.48 ± 1.04	47.42 ± 4.39
	28 d	35.61 ± 6.19	19.09 ± 6.56	1.49 ± 0.48	6.42 ± 2.27	73.56 ± 8.18
	42 d	41.68 ± 4.80	23.44 ± 3.92	2.67 ± 0.73	8.72 ± 1.92	84.98 ± 8.43
	56 d	43.72 ± 7.88	26.63 ± 6.09	3.35 ± 0.60	10.25 ± 3.18	93.21 ± 6.92
	70 d	45.91 ± 6.03	28.42 ± 5.11	5.20 ± 1.01	13.66 ± 2.46	101.73 ± 9.89
Group II	7 d	10.84 ± 2.43	23.76 ± 3.32	N.D.	N.D.	37.56 ± 2.46
	14 d	13.73 ± 4.55	25.12 ± 3.68	1.26 ± 0.34	2.73 ± 0.83	50.47 ± 6.32
	28 d	22.66 ± 4.14	33.70 ± 2.70	2.45 ± 0.69	5.29 ± 0.79	74.53 ± 6.92
	42 d	24.81 ± 5.20	39.98 ± 5.68	3.17 ± 1.21	8.48 ± 2.97	86.28 ± 4.37
	56 d	25.35 ± 2.25	42.32 ± 5.71	4.05 ± 1.41	9.92 ± 1.75	91.49 ± 7.93
	70 d	26.44 ± 2.03	45.86 ± 4.35	5.83 ± 0.86	11.34 ± 3.19	99.38 ± 6.95
Group III	7 d	1.98 ± 0.53	7.25 ± 1.03	120.92 ± 11.78	42.71 ± 5.63	197.28 ± 9.27
	14 d	3.46 ± 0.66	6.82 ± 1.39	173.45 ± 9.21	66.24 ± 5.92	285.46 ± 12.09
	28 d	4.39 ± 1.35	8.77 ± 2.02	143.28 ± 10.24	64.89 ± 4.71	256.89 ± 8.25
	42 d	6.28 ± 1.01	12.18 ± 4.48	116.70 ± 9.42	56.81 ± 7.01	230.73 ± 11.93
	56 d	8.35 ± 1.59	14.62 ± 2.50	83.46 ± 6.98	49.08 ± 5.23	197.19 ± 6.17
	70 d	9.67 ± 2.05	18.59 ± 3.14	72.83 ± 5.69	40.74 ± 3.51	183.77 ± 9.16
Group IV	7 d	2.02 ± 0.77	7.08 ± 1.53	14.53 ± 2.82	155.28 ± 7.86	208.31 ± 11.58
	14 d	3.25 ± 0.54	7.32 ± 1.22	17.13 ± 3.10	213.76 ± 11.46	280.55 ± 10.21
	28 d	4.58 ± 1.20	8.96 ± 0.98	15.21 ± 1.64	164.67 ± 9.73	228.46 ± 10.50
	42 d	6.69 ± 1.47	11.99 ± 2.21	14.98 ± 5.07	135.84 ± 13.12	206.87 ± 11.41
	56 d	8.87 ± 2.32	15.43 ± 2.89	11.46 ± 3.48	107.46 ± 7.48	183.62 ± 7.41
	70 d	9.50 ± 1.25	19.47 ± 4.15	9.38 ± 1.06	89.01 ± 4.79	167.60 ± 9.13

---

The four experimental groups presented different arsenic speciation results. In group I, As(III) was the dominant species followed by As(V), while As(V) was the species with the highest content followed by As(III) in group II. The contents of MMA and DMA in the earthworms of group I and group II were less than those of the two inorganic arsenics. In comparison, the DMA concentrations were slightly higher than MMA during the whole experimental period except the 7th day, at which time period neither of the two organic species could be detected. In terms of the four arsenic species contents in group I and group II, the trend was similar; that is, the concentration increased with exposure time. However, in the later stage of the experiment, the increase in the As(III) and As(V) contents of group I and group II was not as obvious as that before day 42.

In the two organic arsenic treatments, MMA was the dominant species followed by DMA in group III, while DMA was the only species that showed a high proportion of total arsenic in group IV. Compared with the contents of the dominant organic arsenic species, the concentrations of As(III) and As(V) in the earthworms of group III and group IV were much lower. The change in DMA and MMA contents in group III and DMA contents in group IV followed a similar trend, that is, it increased at first, followed by a decrease with exposure time, while the maximum values appeared on day 14. The content of the two inorganic arsenic species in group III and group IV also showed a similar trend. That is, it slowly increased with exposure time, and the As(V) concentrations were slightly higher in comparison with As(III) during the whole experimental period. In the samples of group IV, however, MMA showed an irregular trend: its contents were 14.53, 17.13, 15.21, 14.98, 11.46 and 9.38 mg kg⁻¹ from day 7 to day 70, respectively.

Neutral-red retention time (NRRT)

Fig. 1 shows the NRRTs for all exposure treatments, including the control group during the 70 day experimental period. The NRRT values in the control group exhibited no obvious changes during the whole experiment. However, in the four exposure groups, the biomarker showed different response characteristics. On the whole, NRRTs showed a decreasing tendency with the increase of exposure time in all treatments with one exception; that is, the value of group IV on day 28 was slightly higher than the NRRT value of this group obtained on day 14. In the earthworms of group I and group II, the NRRT values were significantly lower compared with those of the other two groups. Although the NRRTs of group II were higher than those of group I, there were no significant differences between the two treatments. Compared to the control, the NRRTs of coelomocyte lysosomal membranes in group I and group II decreased significantly at each sampling time, along with a significant decrease of the NRRT values in group III and group IV on days 56 and 70. In addition, the NRRTs for earthworms of group III were somewhat lower than the biomarker values obtained in group IV and showed a significant decrease on day 42 compared with the control. Also, as shown from the NRRTs of group III and group IV, there were no significant differences between these two groups. Fig. 2 and 3 present the NRRTs for the earthworms of the four exposure groups at each sampling time point plotted against the corresponding body As(III) and As(V) concentration, respectively. There was an apparently clear relationship between the elevation of body inorganic arsenic concentration in earthworms and decreased neutral-red retention time. Based on the logistic–sigmoid regression, significant dose–response correlations could be shown [R² = 0.938 for As(III) and R² = 0.715 for As(V)], indicating that the lysosomal membranes of the earthworms were injured due to the inorganic arsenic species accumulated in the body.

Fig. 1 Mean ± SD (min) of neutral-red retention times measured for E. fetida following 70 days exposure to arsenic contaminated soils (n = 5). Earthworms of group I, II, III and IV were exposed to As(III), As(V), MMA and DMA, respectively.

Fig. 2 Logistic–sigmoid regression between body As(III) concentrations (mg kg⁻¹ dry wt, n = 5) and neutral-red retention times (min, n = 5) measured for E. fetida following 70 days exposure to arsenic contaminated soils.

Fig. 3 Logistic–sigmoid regression between body As(V) concentrations (mg kg⁻¹ dry wt, n = 5) and neutral-red retention times (min, n = 5) measured for E. fetida following 70 days exposure to arsenic contaminated soils.

Discussion

Exposure to arsenic is a global public health concern due to its wide distribution in the environment and its close association with numerous adverse effects.⁴⁴ As the toxicity of arsenic is determined by its chemical species, many studies have been conducted for arsenic speciation in soils. Yuan et al.⁴⁵ determined different arsenic species in many polluted soil samples and found that As(III) and As(V) were the major arsenic species in the soils. The concentrations of As(III) and As(V) were 0.59–0.72 mg kg⁻¹ and 61.7–76.9 mg kg⁻¹, respectively. Chatterjee and Mukherjee⁴⁶ collected soil samples from the ground of a chemical company producing Paris green and arsenical pesticides. The arsenic contaminated soils contained 16.4 mg kg⁻¹ As(III), 131 mg kg⁻¹ As(V), 51.2 mg kg⁻¹ MMA and 25.0 mg kg⁻¹ DMA. Chappell et al.⁴⁷ determined arsenic species in a contaminated soil sample and found that the concentrations of the different species were 3 mg kg⁻¹ for As(III), 942 mg kg⁻¹ for As(V) and 40 mg kg⁻¹ for organic arsenic. Liu et al.⁴⁸ developed an analytical method for precise and accurate quantification of different arsenic species in soils surrounding several swine farms in the Pearl River Delta in southern China. Relevant results showed that the environmental media were already impacted by organic arsenic feed additives. The average concentrations of arsenic species in the soils were 4.9 μg kg⁻¹ for As(III), 37.3 mg kg⁻¹ for As(V), 24.0 μg kg⁻¹ for MMA and 6.4 μg kg⁻¹ for DMA. As reported in many papers, trivalent and pentavalent arsenic are the most common chemical forms of arsenic in soil. However, a large amount of organic arsenic was also found in many cases.⁴⁹ Therefore, in this study four common arsenic species in soil have been selected, while the recommended soil toxicity test organism, E. fetida, was utilized so as to analyze the toxicity of different arsenic species by detecting its lysosomal membrane response. The exposure concentrations were carefully chosen based on the value of 14 d-LC₅₀ of the four arsenic species obtained in a pre-experiment following the OECD guideline 207. With the accumulation of arsenic species, arsenic biotransformation processes would occur during long-term exposure, due to which the contents of total arsenic and arsenic species in the earthworms were therefore determined.

Arsenic from soil is accumulated in earthworms mainly through ingestion and dermal contact in both the solid and aqueous phases.³ During a long-term exposure, however, the contents of different arsenic species in soil may change through the effects of phosphorus, aluminium, iron and organic matter in soil, as well as pH and redox potential.⁵⁰ Therefore, an artificial OECD soil was prepared in this study in order to avoid interference by these factors. The experimental soils of each group were also collected at each sampling time point for the analysis of the four arsenic species using the same analysis approach. With regard to the extraction procedure, it was performed according to a method developed by Button et al.⁵¹ The results showed that the concentrations of the arsenic species were basically unchanged during the entire experiment, with no obvious interconversion between species. Therefore, the variation of different arsenic species contents in the earthworms was only based on the bioaccumulation and biotransformation process.

Many studies have found the bioaccumulation of arsenic in earthworms sampled from contaminated soils.^20,52–54 In this study, there were significant elevations in the total arsenic concentrations in the earthworms of group I and group II with the increase of exposure time. The total arsenic levels of the two treatments exceeded those in the soil at each sampling time and, due to the increase with the exposure time, indicated that the accumulated arsenic was sequestered in tissues and was not readily excreted.⁵⁵ Comparatively, the total arsenic concentrations in the earthworms of group III and group IV did not exceed those in the soil during the whole experiment, indicating that the organic arsenics could be easily excreted by the earthworms, thus proving the avoidance of bioconcentration.

The change in the different arsenic species in the four experimental groups was based on the biotransformation of arsenic in earthworms. A biotransformation pathway for this explanation has been proposed by Langdon et al.³ and developed by Watts et al.⁵⁶ There were four steps included: (1) As(V) is reduced to As(III);⁵⁷ (2) sulfur-rich proteins such as metallothioneins (MTs) form complexes with As(III);⁵⁸ (3) As(III) is methylated to MMA, followed by the formation of DMA;⁵⁶ (4) AsB is produced along a complex pathway involving DMA.⁵⁹ At last, AsB and other organic arsenic species are excreted from the earthworm through mucus, casts and urine, which would thus decrease the arsenic body burden. This biotransformation mode could help explain the decrease in the contents of total arsenic with exposure time after the 14th day in group III and group IV.

The biotransformation of highly toxic inorganic arsenic to the less toxic organic species AsB has been considered as a mode of mitigating arsenic toxicity in earthworms,⁵² which was also verified in this study. As shown in the arsenic speciation results of group II, a large amount of As(III) can be detected, implying the occurrence of the first step of arsenic biotransformation, namely, the reduction of As(V). The trivalent arsenic species is essential for the arsenic detoxification process for the reason that arsenic is only methylated in the As(III) form.⁵⁷ In addition, As(III) can be biotransformed to decrease its toxicity by complexing with sulfur-rich proteins. Therefore, the reduction of As(V) is a critical step for earthworms to metabolize arsenic. However, the high contents of As(V) in group I were mainly generated from the oxidation of As(III), which would not facilitate the detoxification process in spite of the fact that As(V) is generally thought to be less toxic than As(III). The contents of MMA and DMA in the earthworms of group I and group II changed from undetectable to detectable and then kept increasing with exposure time, indicating that the third step of arsenic biotransformation, the methylation of inorganic arsenics, occurred in the earthworms. Based on the arsenic speciation results of group III, a high level of DMA apart from MMA was detected, indicating that MMA can be methylated to DMA, serving as further evidence of the third step of the biotransformation pathway. However, the contents of As(III) and As(V) exhibited a gradual increase with exposure time, which may come from the demethylation process of organic arsenics. The same phenomenon could also be observed in group IV. The demethylation process could greatly increase the toxicity of arsenic, which may account for the death of half the earthworms in the pre-experiment. The arsenic speciation results of group IV showed that DMA was the only species detected at high concentration in the earthworms, implying the occurrence of the fourth step of arsenic biotransformation, namely, the formation of AsB. In another study, we further proved the presence of AsB in earthworms exposed to different arsenic species ​(data not shown).

An important objective in ecotoxicology studies is to provide information to assess ecological risk.⁶⁰ In previous studies, biomarkers have been developed to test the toxicity of substances to earthworms, including survival, growth, reproduction, behavior, etc.^61,62 These bioindicators are often insensitive and fail to reflect the relevant ecological effect of toxicity, especially under the effect of a low concentration of contaminants. In contrast, lysosomal membrane response is regarded as a general subcellular biomarker for the action of toxic metals, and the NRRT assay has been found to be reliable, dose-related, and practical in assessing the adverse effects of anthropogenic heavy metal pollution for different earthworm species.^35,63,64 The reduced NRRT indicates that exposure has induced a physiological response.³⁸ In previous toxicity studies relating to arsenic, only one arsenic species was considered (in most cases it was pentavalent arsenic) without differentiating arsenic species; however, in this study, the possible use of NRRT has been investigated and evaluated as a biomarker for the toxicity of four different arsenic species, as well as a comparison of toxicity between the arsenic species.

The NRRT results in Fig. 1 show that there was no significant decrease in the control group with exposure time. Comparatively, the NRRT values showed a significant decrease in the earthworms of group I and group II during the whole experiment, followed with a gradual decrease with exposure time. In another study conducted by Lee and Kim,³⁶ E. fetida earthworms were exposed to sandy soils contaminated with As(V) for 28 days, and a reduction of the NRRTs could also be observed with the increase of exposure time. According to the trend of the NRRTs in the earthworms of group III and group IV, the responses of the lysosomal membrane did not decrease significantly compared with the control before day 42 in spite of the accumulation of large amounts of organic arsenic, suggesting that MMA and DMA could hardly damage the integrity of the lysosomal membrane. However, the NRRT values of group III and group IV decreased significantly in the later stage of the experiment, while the contents of As(III) and As(V) gradually increased with the occurrence of the demethylation process. This phenomenon implied that even under a severe contamination of MMA and DMA, it was inorganic arsenics generated from demethylation, rather than organic arsenics themselves, that could induce toxic effects on organisms.

The regression analysis results shown in Fig. 2 and 3 show that there were significant correlations between the reduction in NRRTs and the elevation of the content of the two inorganic arsenic species in the earthworms in the four treatments, indicating dose–response relationships. Moreover, the correlation coefficient between NRRTs and As(III) body burden was much higher in comparison with that between NRRTs and As(V), indicating that the lysosomal membrane response of the earthworm was more sensitive to As(III) species. Hence, NRRT was proved to be more useful to reflect the toxic effect of the As(III) species. On the other hand, it can be inferred that the lysosome membrane stability is affected more by As(III) compared with other arsenic species, which could be a possible explanation for the mechanism that As(III) is more toxic than As(V) for many biological species.

The exposure to toxic substances damages the lysosomal membrane and increases its permeability, leading to the decrease of NRRTs in organisms.³⁵ Therefore, the NRRT biomarker can be used to compare the toxicity of different contaminants; that is, lower NRRT values imply higher toxicity, and vice versa.⁶⁵ As can be seen from Fig. 1, the NRRT values of the four exposure groups were ranked as group I < group II < group III < group IV at each sampling time point. Therefore, the toxicity of the four arsenic species could be ranked as As(III) > As(V) > MMA > DMA. Despite the great difference between the 14 d-LC₅₀ of MMA and DMA, the toxic effect of the two species on E. fetida at a subcellular level showed no significant difference during the 70 day exposure period. These results proved the low toxicity of MMA and DMA when the soil concentration was set at one-tenth of 14 d-LC₅₀; moreover, the pollution stress was mainly caused by inorganic arsenics generated from the demethylation process. Consequently, there was no dose–response relationship between NRRTs and organic arsenic accumulation in the earthworms. In real environmental soils, the contents of MMA and DMA are generally lower compared with the concentrations of the two inorganic arsenic species in this study; therefore, it can be thought that the organic arsenics are nontoxic in most cases. During long-term exposure of MMA and DMA, however, inorganic arsenics could be generated through demethylation and accumulate in organisms,³ thus causing damage to the integrity of the lysosomal membrane and other biomarkers. This may explain why half of the earthworms were dead on the 14th day in the pre-experiment, during which period the exposure concentrations of DMA and MMA were ten times higher than the concentrations of the two arsenic species set in this study. Therefore, the MMA and DMA concentrations should be considered, especially in long-term monitoring programs. In this study, the spiked concentrations of As(III) and As(V) were close to each other. Although the differences of NRRT values between group I and group II were not significant, the results of the NRRT assay and logistic–sigmoid regression suggested that the trivalent arsenic was more toxic than the pentavalent arsenic, which agreed with previous toxicology research.⁵⁰ In real soils, different arsenic species always coexist with each other. Hence, it is not accurate for environmental quality assessment to only consider total arsenic or one arsenic species. For future research, further investigation shall therefore be conducted on the toxicology of different arsenic species, so as to provide evidence for accurate risk assessment.

Conclusions

In this study, we investigated the lysosomal membrane responses of the earthworm, E. fetida, to the exposure of four main arsenic species in OECD artificial soil. The results showed that the neutral-red retention time assay was sufficiently sensitive for toxicity assessment of inorganic arsenic pollution and it was affected more by As(III) compared with As(V). The NRRT values suggested that the toxicity of the four arsenic species was ranked as As(III) > As(V) > MMA > DMA. Despite the low toxicity of the two organic arsenics, they could be transformed into highly toxic inorganic arsenic species through the demethylation process, which might alter lysosomal membrane stability and induce adverse effects observable at the subcellular level in earthworm. Arsenic speciation analysis results validated the biotransformation pathway of arsenic in earthworms. Highly toxic inorganic arsenics were methylated and converted to less toxic organic species before excretion, which was considered as a mode of mitigating arsenic toxicity in earthworm. As different arsenic species always coexist in real environmental soils and show different degrees of toxicity, arsenic speciation should be considered in order to obtain accurate results in environmental assessment programs.

Conflict of interest

The authors declare that there are no conflicts of interest.

Acknowledgements

This work was financially supported by Environmental Protection and Public Welfare Industry Research Special: the remediation technologies and demonstration for the combined pollution of the oil-heavy metals in the saline soil (No. 201109022) and National High-tech Research and Development Projects (National 863 Projects): the key technology of efficient exploiting deep brine in the Yellow River delta (No. 2012AA061705). We are grateful to Dr Jing Wang at Shandong University and Prof. Fanping Meng at Ocean University of China for their assistance in the experiment.

References

1. A. A. Duker, E. J. M. Carranza and M. Hale, Environ. Int., 2005, 31, 631–641.
2. J. Liu, Y. Lu, Q. Wu, R. A. Goyer and M. P. Waalkes, J. Pharmacol. Exp. Ther., 2008, 326, 363–368.
3. C. J. Langdon, T. G. Piearce, A. A. Meharg and K. T. Semple, Environ. Pollut., 2003, 124, 361–373.
4. J.-Y. Kim, K.-W. Kim, J. S. Ahn, I. Ko and C.-H. Lee, Environ. Geochem. Health, 2005, 27, 193–203.
5. S. Kapaj, H. Peterson, K. Liber and P. Bhattacharya, J. Environ. Sci. Health, Part A: Toxic/Hazard. Subst. Environ. Eng., 2006, 41, 2399–2428.
6. A. H. Smith, P. A. Lopipero, M. N. Bates and C. M. Steinmaus, Science, 2002, 296, 2145–2146.
7. T. N. Hartley, A. J. Macdonald, S. P. McGrath and F.-J. Zhao, Environ. Pollut., 2013, 180, 259–264.
8. J. C. Kwon, J.-S. Lee and M. C. Jung, Appl. Geochem., 2012, 27, 1020–1026.
9. X. Liu, W. Zhang, Y. Hu, E. Hu, X. Xie, L. Wang and H. Cheng, Chemosphere, 2015, 119, 273–281.
10. Y. Jiang, X. Zeng, X. Fan, S. Chao, M. Zhu and H. Cao, Ecotoxicol. Environ. Saf., 2015, 122, 198–204.
11. Y. Wang, S. Wang, P. Xu, C. Liu, M. Liu, Y. Wang, C. Wang, C. Zhang and Y. Ge, Rev. Environ. Sci. Bio/Technol., 2015, 14, 427–451.
12. P. Wang, G. Sun, Y. Jia, A. A. Meharg and Y. Zhu, J. Environ. Sci., 2014, 26, 371–381.
13. Z. Gong, X. Lu, M. Ma, C. Watt and X. C. Le, Talanta, 2002, 58, 77–96.
14. H. M. Anawar, Talanta, 2012, 88, 30–42.
15. L. Liu, B. He, Z. Yun, J. Sun and G. Jiang, J. Chromatogr. A, 2013, 1304, 227–233.
16. N. C. Sumedha and S. Miltonprabu, Int. J. Biol Pharmaceut. Res., 2013, 4, 507–515.
17. K. A. Winship, Adverse Drug React. Acute Poisoning Rev., 1983, 3, 129–160.
18. D. C. Coleman and E. R. Ingham, Biogeochemistry, 1988, 5, 3–5.
19. D. Kennette, W. Hendershot, A. Tomlin and S. Sauvé, Appl. Soil Ecol., 2002, 19, 191–198.
20. A. Geiszinger, W. Goessler, D. Kuehnelt, K. Francesconi and W. Kosmus, Environ. Sci. Technol., 1998, 32, 2238–2243.
21. N.-W. Xiao, Y. Song, F. Ge, X.-H. Liu and Z.-Y. Ou-Yang, Chemosphere, 2006, 65, 907–912.
22. W. Zhang, K. Liu, J. Li, L. Chen and K. Lin, Ecotoxicol. Environ. Saf., 2015, 113, 45–51.
23. OECD, OECD Guidelines for the Testing of Chemicals, Organization for Economic Co-operation and Development, 1984.
24. L. Chen, X. Lu and Y. Ma, Bull. Environ. Contam. Toxicol., 2012, 89, 1161–1164.
25. P. Muangphra, S. Sengsai and R. Gooneratne, Environ. Toxicol., 2013, 30, 597–606.
26. S. Hattab, I. Boughattas, H. Boussetta, A. Viarengo, M. Banni and S. Sforzini, Ecotoxicol. Environ. Saf., 2015, 122, 76–82.
27. A. Colacevich, M. J. Sierra, F. Borghini, R. Millán and J. C. Sanchez-Hernandez, J. Hazard. Mater., 2011, 194, 135–143.
28. C. Svendsen and J. M. Weeks, Ecotoxicol. Environ. Saf., 1997, 36, 72–79.
29. M. N. Moore, Mar. Pollut. Bull., 1985, 16, 134–139.
30. J. J. Scott-Fordsmand, J. M. Weeks and S. P. Hopkin, Environ. Toxicol. Chem., 2000, 19, 1774–1780.
31. D. J. Spurgeon, S. R. Stürzenbaum, C. Svendsen, P. K. Hankard, A. J. Morgan, J. M. Weeks and P. Kille, Comp. Biochem. Physiol., Part C: Toxicol. Pharmacol., 2004, 138, 11–21.
32. A. Viarengo, M. N. Moore, G. Mancinelli, A. Mazzucotelli, R. K. Pipe and S. V. Farrar, Mar. Biol., 1987, 94, 251–257.
33. Q. Zhou, J. Zhang, J. Fu, J. Shi and G. Jiang, Anal. Chim. Acta, 2008, 606, 135–150.
34. A. Rocco, J. J. Scott-Fordsmand, G. Maisto, S. Manzo, A. Salluzzo and J. Jensen, Ecotoxicol. Environ. Saf., 2011, 74, 984–988.
35. C. Svendsen, D. J. Spurgeon, P. K. Hankard and J. M. Weeks, Ecotoxicol. Environ. Saf., 2004, 57, 20–29.
36. B. T. Lee and K. W. Kim, Environ. Toxicol., 2009, 24, 369–376.
37. D. M. Lowe, M. N. Moore and B. M. Evans, Mar. Ecol.: Prog. Ser., 1992, 91, 135–140.
38. J. M. Weeks and C. Svendsen, Environ. Toxicol. Chem., 1996, 15, 1801–1805.
39. H. Qiu, M. G. Vijver, E. He and W. J. Peijnenburg, Environ. Sci. Technol., 2013, 47, 4796–4803.
40. L. Zhao, Y. Xu, H. Hou, Y. Shangguan and F. Li, Sci. Total Environ., 2014, 468, 654–662.
41. J. A. Brisbin and J. A. Caruso, Analyst, 2002, 127, 921–929.
42. Z.-F. Wang and Z.-J. Cui, Chin. Chem. Lett., 2016, 27, 241–246.
43. S. Fukui, T. Hirayama, M. Nohara and Y. Sakagami, Talanta, 1983, 30, 89–93.
44. N. L. Dangleben, C. F. Skibola and M. T. Smith, Environ. Health, 2013, 12, 73–87.
45. C.-G. Yuan, B. He, E.-L. Gao, J.-X. Lü and G.-B. Jiang, Microchim. Acta, 2007, 159, 175–182.
46. A. Chatterjee and A. Mukherjee, Sci. Total Environ., 1999, 225, 249–262.
47. J. Chappell, B. Chiswell and H. Olszowy, Talanta, 1995, 42, 323–329.
48. X. Liu, W. Zhang, Y. Hu and H. Cheng, Microchem. J., 2013, 108, 38–45.
49. H. Yang, M. He and X. Wang, Environ. Geochem. Health, 2015, 37, 21–33.
50. M. Bissen and F. H. Frimmel, Acta Hydrochim. Hydrobiol., 2003, 31, 9–18.
51. M. Button, G. R. T. Jenkin, C. F. Harrington and M. J. Watts, J. Environ. Monit., 2009, 11, 1484–1491.
52. C. J. Langdon, T. G. Piearce, S. Black and K. T. Semple, Soil Biol. Biochem., 1999, 31, 1963–1967.
53. E. Fischer and L. Koszorus, Pedobiologia, 1992, 36, 172–178.
54. A. Romero-Freire, F. J. M. Peinado, M. D. Ortiz and C. A. M. van Gestel, Environ. Sci. Pollut. Res., 2015, 22, 15016–15028.
55. A. A. Meharg, R. F. Shore and K. Broadgate, Environ. Toxicol. Chem., 1998, 17, 1124–1131.
56. M. J. Watts, M. Button, T. S. Brewer, G. R. T. Jenkin and C. F. Harrington, J. Environ. Monit., 2008, 10, 753–759.
57. K. J. Irgolic, Arsenic in the environment, in Frontiers in bioinorganic chemistry, ed. A. V. Xavier, VCH, Weinheim and Deerfield Beach, 1986, pp. 399–408.
58. A. J. Morgan, C. Winters and A. Yarwood, Cell Biol. Int., 1994, 18, 911–914.
59. C. J. Langdon, A. A. Meharg, J. Feldmann, T. Balgar, J. Charnock, M. Farquhar, T. G. Piearce, K. T. Semple and J. Cotter-Howells, J. Environ. Monit., 2002, 4, 603–608.
60. R. Wilson and E. A. Crouch, Science, 1987, 236, 267–270.
61. K. Lock and C. R. Janssen, Bull. Environ. Contam. Toxicol., 2002, 68, 760–765.
62. D. J. Spurgeon, C. Svendsen, V. R. Rimmer, S. P. Hopkin and J. M. Weeks, Environ. Toxicol. Chem., 2000, 19, 1800–1808.
63. C. Svendsen, A. A. Meharg, P. Freestone and J. M. Weeks, Appl. Soil Ecol., 1996, 3, 99–107.
64. M. Gao, Q. Zhou, W. Song and X. Ma, Environ. Toxicol. Pharmacol., 2014, 37, 689–696.
65. M. Gao, Y. Qi, W. Song and Q. Zhou, Chemosphere, 2015, 139, 229–234.

---