Toxicity of lead pollution to the collembolan Folsomia candida in Ferri-Udic Cam

Ying DINGZhu LI*Xin KELonghua WU and Shengpeng ZUOKey Laboratory of Soil Environment and Pollution RemediationInstitute of Soil ScienceChinese Academy of SciencesNanjing 0008(China)

2College of Environmental Science and Engineering,Anhui Provincial Engineering Laboratory of Water and Soil Pollution Control and Remediation,Anhui Normal University,Wuhu 241003(China)

3Institute of Plant Physiology and Ecology,Shanghai Institute for Biological Sciences,Chinese Academy of Sciences,Shanghai 200032(China)

ABSTRACTToxic effect of lead(Pb)pollution on Collembola in soils has seldom been studied in depth,which is especially true for growth responses since the juveniles are very small and numerous.A single species test was conducted using the collembolan Folsomia candida as the indicator species,and soil Pb pollution(at Pb concentrations of 0,300,600,1 200,2 400,and 4 800 mg kg-1)was simulated by mixing Pb(NO3)2 with a natural unpolluted soil.Adult survival,reproduction,body length,and body Pb concentration were determined.Adult survival and reproduction started to decrease at soil Pb concentrations of 1 200 and 600 mg kg-1,respectively.Lead concentrations in adult and juvenile bodies increased with increasing Pb concentration in soil.The percentage and individual abundances based on body length analysis showed that the larger adults(2.1–2.6 mm,P＜0.01)were more sensitive to soil Pb concentration than the intermediate(1.9–2.1 mm,P＜0.001)and smaller(1.7–1.9 mm,P＜0.001)adults.Similarly,the larger juveniles(1.0–1.4 mm,P＜0.001)were more sensitive to soil Pb concentration than the intermediate(0.6–1.0 mm,P＜0.001)and smaller(0.3–0.6 mm,P＜0.001)juveniles.For both adults or juveniles,the abundance and percentage of larger individuals declined significantly with increasing soil Pb concentration,suggesting that larger individuals were more sensitive to soil Pb pollution.Therefore,body length features,especially the percentage and abundance of larger individuals,would be potential bio-indicators of soil Pb pollution and possibly other types of soil pollution.

Key Words:bio-indicators,body length,Collembola,large individuals,soil Pb pollution

INTRODUCTION

Metal pollution is a widespread problem with potentially toxic metal(loid)s representing a substantial risk to aquatic and terrestrial ecosystems(Zhang and van Gestel,2017).Lead(Pb)-contaminated soils have aroused great concern and have received increasing interest from national and international regulatory organizations(Miglioriniet al.,2004).Lead pollution has increased considerably due to human activities such as transportation,industrial practices(e.g.,mining),and waste disposal.When Pb enters an agricultural soil,it may produce biological toxicity effects,resulting in declining crop yields and possibly wilting and death(Yanget al.,2001).It may also inhibit the growth and reproduction of soil animals,such as earthworms,springtails,and mites,leading to lower oviposition,egg hatching and survival rates,weight loss,and other effects(Xuet al.,2009).Soil Pb contamination is a particular challenge due to the long-term retention time of Pb in the environment,from 150 to 5 000 years(Kumaret al.,1995;Luoet al.,2014b).Consequently,Pb is ranked as the number two priority hazardous substance(second only to As)(Chenet al.,2002;Luoet al.,2015).Current legislation and assessment of Pb concentrations in soils are based mainly on the total concentration of Pb present in the soil(Davieset al.,2003;Luoet al.,2014b),but the effects within the terrestrial environment have not been fully studied,and the evaluation of soil toxicity is complicated due to the complex interactions among pollutants,organisms,and soil particles with different properties(van Gestel,2008).Few data on Pb toxicity to soil invertebrates are currently available(Zhang and van Gestel,2017).

Ecotoxicological tests are key methods for estimating the impact of soil pollution.They are often used as supplemental tools for soil chemical analysis in order to assess chemical toxic effects on living organisms(Buret al.,2012).Soil invertebrates which respond to a wide range of chemicals at various concentrations in soils have been used as reliable bioindicators of soil health in combination with chemical and physical properties(Fountain and Hopkin,2001;Sonet al.,2007).The widespread interest in developing soil invertebrate tests to assess ecotoxicological effects of chemicals has resulted in several proposals to standardize experiments with invertebrates(Fountain and Hopkin,2001),such as earthworms,enchytraeids,nematodes,and collembolans.Collembolans are regarded as ideal test animals for ecotoxicological research(ISO,1999;OECD,2009;Liuet al.,2018)due to their widespread distribution in different soil types,large abundance,and their role in the decomposition of organic matter,regulation of microbial activity,and nutrient cycling(Fountain and Hopkin,2001;Buret al.,2010;Luoet al.,2014a).Numerous studies have been conducted in relation to collembolan sensitivity and resistance to environmental contaminants,in particular to potentially toxic metals(Nursitaet al.,2005;Sørensen and Holmstrup,2005;Liuet al.,2018).The springtailFolsomia candidais often used as a biological indicator of pollution(Fountain and Hopkin,2001;Buret al.,2010;Luoet al.,2014b).

The toxicity of soil metals such as Pb can be evaluated on the basis of both the total and available metal concentrations,which may provide an understanding of metal(Pb)toxicity to soil invertebrates.Furthermore,the internal Pb concentration in animals may eliminate the effects from different routes of exposure and therefore serve as a superior indicator of exposure.Lead is a nonessential metal and is difficult for organisms to regulate after uptake.Thus,the internal Pb concentration may be a good estimator of its bioavailable fraction and its potential risk(Zhang and van Gestel,2017).However,the internal Pb concentration is seldom taken as a toxicity indicator because of methodological problems(Zhuet al.,2017).

Different chemicals may affect the life-cycle parameters of various arthropods differently,especially collembolans(Crommentuijnet al.,1993;Folker-Hansenet al.,1996).Survival and reproduction are widely used parameters in ecotoxicological research and evaluation index characteristics of soil metal contamination(van Straalenet al.,1994;Folker-Hansenet al.,1996).However,other life-cycle parameters such as body length are seldom studied because the measurement is difficult and generally considered to be less sensitive than the reproductive rate(Domeneet al.,2007).Nevertheless,a previous study has indicated that body length might be more sensitive and therefore useful for providing an early warning(Crommentuijnet al.,1993),and the body length of the juveniles ofF.fimetariacould be a steady parameter to assess the effects of the insecticide dimethoate(Folker-Hansenet al.,1996).Bengtssonet al.(1983)showed that the mean maximum lengths ofOnychiurus armatusfeeding on copper(Cu)and Pb were higher at moderate metal pollution in both F1 and F2 generations,but were significantly reduced when feeding on substrates containing≥90 mg metals(Cu or Pb)kg-1.These earlier studies all had different treatment methods and insights for body length parameters.However,there has been no in-depth study on the overall trends and regularity of changes in body length in contaminated soils.Body length might be an indicator with great potential for understanding the risks from potentially toxic elements and other pollutants,complementing the other indicators.

Adult survival and reproduction are the most widely used endpoints,and the latter is regarded as the single most important function in the life cycle of an organism(Linet al.,2019).The growth ofF.candidais a primary life-history characteristic,and hence,reduced growth is indicative of reduced fitness(Fountain and Hopkin,2001;Linet al.,2019).A study regarding the effects of metal Hg-contaminated food on the growth ofF.candidashowed that the growth rate(per day of exposure of body length)decreased at days 12,14,and 20 for the highest(4 mg kg-1),intermediate(2 and 3 mg kg-1),and lowest(1 mg kg-1)Hg concentration,respectively,when compared to the no Hg control(Cardosoet al.,2019).In this study,single-species testing was adopted using the standard collembolan speciesF.candida.Specifically,all individuals of adults and juveniles were separated according to their body length,and the individuals with different body lengths were analysed separately in order to determine whether and how different individuals of different sizes respond to Pb pollution.The Pb concentration in the body tissues of the animals was determined to ensure that the toxic effects on the animals were due to the Pb pollution.Overall,this study had two main purposes:firstly,to investigate how soil Pb pollution exerts toxic effects on collembolans and,consequently,influences ecosystem and secondly,to seek potential indicator features of the animals for ecological risk assessment of a Pb-contaminated soil on the basis of the toxicity and response characteristics to soil Pb pollution.

MATERIALS AND METHODS

Soil

The test soil was collected from a natural mountain area close to Nanjing City,Jiangsu Province,East China.It was a yellow-brown soil with a sandy loam texture(Ferri-Udic Cambosol).The soil was air-dried,plant debris,stones,and gravel were removed,and the soil was then sieved through a 2-mm nylon mesh.The soil properties were as follows:pH(H2O),5.1;water holding capacity(WHC),32%;organic matter(OM)content,12.3 g kg-1;and cation exchange capacity(CEC),15.2 cmolckg-1.Total nitrogen,phosphorus(P),and potassium(K)contents were 0.77,0.10,and 15.7 g kg-1,respectively.Available P content was 19.2 mg kg-1,and available potassium(K)was 100 mg kg-1.Total cadmium(Cd)content was 45.2 μg kg-1,and total Pb,zinc(Zn),and Cu were 18.7,42.2,and 3.59 mg kg-1,respectively.Selected physicochemical properties of the soil were determined according to Zhang and Gong(2012).

Chemical reagents and standard materials

Single-element stock solutions(Merck,Darmstadt,Germany)of 1 000 mg Pb L-1(superior grade purity;National Iron and Steel Materials Analysis Centre,Beijing,China)diluted with 10%(volume:volume)HNO3were used as the calibration solutions.Certified reference materials(CRM)GSB-28(prawn)and GSB-29(pork liver)for Pb were obtained from the Institute of Geophysical and Geochemical Exploration,Langfang,China.All standard materials were stored at 4°C.The extra Pb added to the test soil was added as Pb(NO3)2(reagent grade;Nanjing Chemical Reagent Company,Nanjing,China).

Animals

Individuals of the collembolanF.candidawere originally obtained from the Bioscience Department,Aarhus University,Denmark and have been cultured in our laboratory for＞10 years.The animals were reared in 90-mm-diameter Petri dishes covered with a 0.5–1 cm layer of a mixture of plaster of Paris and activated charcoal with a ratio of 9:1(weight:weight)at 20±15°C and 75%relative humidity(RH),with a light/dark cycle of 12 h:12 h(800 lux).Distilled water and a small amount of dried baker’s yeast(Angel Yeast Co.Ltd.,Yichang,China)were added as food on a weekly basis.Individuals synchronized at the age of 10–12 d were obtained according to standardized methods(ISO,1999;OECD,2009)and used for the experiment.

Experimental setup

The Pb(NO3)2was dissolved in deionized water to obtain a stock solution of Pb(30 g Pb L-1).Aliquots of 0,2,4,8,16,and 32 mL Pb stock solution were mixed with 200 g dry soil to produce a series of Pb-spiked soil samples with 0(control),300,600,1 200,2 400,and 4 800 mg kg-1Pb,respectively.The soil samples were adjusted with distilled water to a soil moisture content of 50%WHC.The spiked soil samples were aged in sealed plastic containers at 20±2°C for 48 h.Aliquots of 30 g soil and 10 synchronized 10–12 d oldF.candidawere transferred into each cylindrical glass container(5.5-cm height,11-cm inner diameter,and 0.5-cm thickness)and 15 mg yeast was added into each container.Ten replicates of the control soil and five replicates of the Pb-contaminated soil treatments were set up.The containers were covered with an opaque plastic film and incubated in the same culture conditions as described above for 28 d.The yeast was refreshed twice weekly,and the containers were weighed to maintain the water balance once a week.At the end of the experiment,the collembolans in the soil were extracted on an extractor for 48 h from 25 to 45°C with a rate of increase of 5°C every 12 h(Fountain and Hopkin,2001;Zhuet al.,2016b).Distilled water(10 mL)was added to each collecting beaker to keep the animals alive.

Determination of animal survival,reproduction,and body length

A drop of blue ink was added to each collecting beaker and gently shaken.The animals floating on the water surface were photographed(SZX12 stereo microscope,Olympus,Tokyo,Japan).The pictures were used to count the numbers of adults and juveniles for survival and reproduction rates and to measure body length.The survival,reproduction,and body length from the end of the posterior abdominal segment to the anterior margin of the head were measured using the Image-Pro Plus 1.47a software package(Folker-Hansenet al.,1996;Nursitaet al.,2005;Zhuet al.,2016a).The tests are valid if the number of dead/missing springtails is＜20%of the introduced animals in the control(Buchet al.,2016).

In order to see if and how individuals of different body sizes responded differently to Pb pollution,the body lengths of adults were divided into ranges by 0.1-mm intervals from the shortest to the longest individuals,viz.1.7–1.8,1.8–1.9,1.9–2.0,2.0–2.1,2.1–2.2,2.2–2.3,2.3–2.4,2.4–2.5,and 2.5–2.6 mm,and the body lengths of juveniles were divided into the ranges of 0.3–0.4,0.4–0.5,0.5–0.6,0.6–0.7,0.7–0.8,0.8–0.9,0.9–1.0,1.0–1.1,1.1–1.2,1.2–1.3,and 1.3–1.4 mm.

Determination of Pb in soil

Soil samples were air-dried and ground to pass through a 0.15-mm nylon sieve.For the analysis of soil total Pb concentration,approximately 0.20 g air-dried soil per sample was weighed into a 70-mL digestion vessel and digested with 10 mL mixed acid(HCl:HNO3,1:1,volume:volume)at 105°C in a drying oven for 6 h(Luoet al.,2014c;Wuet al.,2018).To determine the available Pb concentration in soil,2 g air-dried soil(passed through a 2-mm nylon sieve)was extracted with 20 mL 0.01 mol L-1CaCl2by shaking for 2 h at 20°C,centrifuging at 3 000 r min-1for 10 min,and filtering through a 0.45–mm membrane filter to obtain the exchangeable solution.Total and available Pb concentrations were determined using a Varian Spectra AA 220FS flame atomic absorption spectrophotometer and a Varian Spectra AA 220Z graphite furnace spectrophotometer(Varian,Palo Alto,USA).

Determination of Pb in body tissue

The extracted animals were placed on moist fliter paper for 92 h to void their guts.The animals were ultrasonically cleaned for 15 min in a container with 10 mL ultrapure water(Zödl and Wittmann,2003;Zhuet al.,2017)and rinsed at least three times with ultrapure water.The cleaned animals were collected and stored at-80°C for 1 d and freeze-dried at-18°C for 2 d(Ardestani and van Gestel,2013;Zhuet al.,2017).The dried animals were stored in desiccators at 4°C until analysis.They were weighed in a Model XS3DU automatic electronic balance(precision±1 μg,XS3DU,Mettler Toledo,Columbus,USA)when analysed.A portion of sample(50–100 μg)was placed in a small glass test tube(Pyrex glass,25.0-mm length,4.2-mm inner diameter,and 1.3-mm thickness),and 50 μL digestion reagent(HNO3:H2O2,3:1,volume:volume)was added to the test tube.The five small glass digestion tubes were placed in a 5-mL blank Pyrex beaker which was embedded in a 70-mL Teflon steel digestion vessel at 105°C and high pressure sealed in a drying oven for 4 h.After digestion and cooling,the beakers were transferred to an electric heating plate and heated to dryness at 120°C.The samples were stored in a refrigerator at 4°C until analysis.They were dissolved in 100 μL of 0.1 mol L-1HNO3and thoroughly mixed at least 1 h prior to the analysis and determination of Pb in theF.candidatissues(Zhuet al.,2017).A Thermo SOLAAR atomic absorption spectrophotometer(graphite furnace,ice 3500 AA System,Thermo Fisher Scientific,Waltham,USA)was used to determine the concentration of Pb in the body tissues.Using 2%NH4H2PO4as matrix modifier,the volume of the solution was 20.0 μL and the matrix modifier volume was 5.0 μL.The quality of the animal body tissue Pb analysis was controlled by analysing replicate blank samples and certified reference materials.Measured Pb recoveries of the reference materials were always 100%±15%.

Statistical analysis

The data were analysed using the STATISTICA v.7.0 software package(StatSoft,Tulsa,USA).One-way analysis of variance and Tukey’s honestly significant difference comparison of means for all pairs were used,and the data were expressed as means±standard errors(SEs).The half-maximal effective concentration(EC50)values forF.candidasurvival and reproduction were calculated with Probit Analyze in the SPSS 21.0 software package.All data are presented using OriginPro 9.1.The bioaccumulation factor(BAF)was calculated as the ratio of Pb concentration in collembolan to the total Pb concentration in soil(Luoet al.,2014a).

RESULTS

Soil Pb concentration

The Pb concentrations determined in the soil at the end of the experiment were comparable to the experimental nominal added Pb concentrations(Table I).The CaCl2-extractable Pb concentration increased with increasing nominal added Pb concentration.

Animal Pb concentration

The Pb concentrations in body tissues of both adults(F3,17=63.43,P＜0.001)and juveniles(F5,24=6 424,P＜0.001)ofF.candidaincreased with increasing added soil Pb concentration after 28 d of exposure.The Pb concentration in adults at 1 200 mg kg-1Pb added was 21 times that in the control(Fig.1a).There were no data for adults at 2 400 or 4 800 mg kg-1Pb added because all the adults died in these two treatments(see below).The Pb concentrations in juveniles at 1 200,2 400,and 4 800 mg kg-1Pb added were 10.5,286,and 488 times the control,respectively(Fig.1b).Except for the control,the BAF values of Pb from soil to body of adults and juveniles were 0.02–0.42 and 0.02–0.05,respectively.

Fig.1 Lead concentration in adult issue(a)and juvenile tissue(b)of F.candida after 28-d exposure to Pb-polluted soil with nominal added Pb concentrations of 0,300,600,1 200,2 400,and 4 800 mg kg-1.Bars indicate standard errors of means(n=5).

Animal survival and reproduction

After 28 d of exposure,the adult survival of 94%exceeded the validity criterion of 80%(Fig.2a),and there were more than 100 juveniles in each control treatment,which suggested the test was valid(Fig.2b),as set by theISO and OECD standards(ISO,1999;OECD,2009).Adult survival did not change significantly within the added Pb concentration range of 0 to 1 200 mg kg-1,with survival rates of 88%–98%,but then decreased sharply to 0 at added Pb concentrations of 2 400 and 4 800 mg kg-1(F5,28=334.6,P＜0.001)(Fig.2a).Reproduction decreased with increasing added Pb concentration above 1 200 mg kg-1(F5,28=40.79,P＜0.001)(Fig.2b).The estimated median lethal concentration(LC50)value for adult survival was 256 mg kg-1(169–456 mg kg-1)based on the 0.01 mol L-1CaCl2extractable Pb concentration,and the estimated EC50value for reproduction was 1 322 mg kg-1(1 175–1 520 mg kg-1)based on the actual total Pb concentration.

Fig.2 Survival(adult individuals,a)and reproduction(juvenile individuals,b)of F.candida after 28-d exposure to Pb-polluted soil with nominal added Pb concentrations of 0,300,600,1 200,2 400,and 4 800 mg kg-1.Bars indicate standard errors of means(n=5).

TABLE INominal added soil Pb concentrations at the start and the total and CaCl2-extractable Pb concentrations in soil determined at the end of the experiment

Mean body length of all individuals

Adult mean body length of all individuals in one container did not differ significantly with increasing added soil Pb concentration at lower Pb concentrations,while the body length under the 600 mg kg-1Pb treatment(2.17 mm)exceeded that under the 1 200 mg kg-1Pb treatment by 8.41%(2.01 mm)(F3,21=3.433,P＜0.05)(Fig.3a).All adults died at 2 400 and 4 800 mg kg-1.The mean body length of juveniles was generally not significantly different among different Pb-polluted soils,with only that at 1 200 mg kg-1Pb being lower than both at 0 and 2 400 mg kg-1Pb(F5,28=3.926,P＜0.01)(Fig.3b).

Fig.3 Mean body lengths of adults(a)and juveniles(b)of F.candida after 28-d exposure to Pb-polluted soil with nominal added Pb concentrations of 0,300,600,1 200,2 400,and 4 800 mg kg-1.Bars indicate standard errors of means(n=5).The asterisks*and**indicate significant differences at P＜0.05 and P＜0.01,respectively.

Percentage distribution of body length in different size ranges

Generally,with increasing added soil Pb concentration,the percentage of larger individuals with longer body length decreased,while the percentage of shorter ones increased or remained unchanged(Fig.4).The percentage of adult individuals with body length ranging from 2.1 to 2.6 mm showed similar response patterns to soil Pb concentrations,initially increasing slightly at 600 mg kg-1Pb and then decreasing at 1 200 mg kg-1Pb when compared to the control(Fig.4a).The percentage of adults in the range of 1.9–2.1 mm was almost unchanged(although slightly decreased at 600 mg kg-1),while those in the range of 1.7–1.9 mm generally increased with increasing added soil Pb concentration.Thus,the individuals in the range of 2.1–2.6 mm were grouped together as“larger adults”,those in the range of 1.9–2.1 mm as“intermediate adults”,and those in the range of 1.7–1.9 mm as“smaller adults”.

Similarly,with increasing added soil Pb concentration,the percentage of larger juvenile individuals with body length in the range of 1.0–1.4 mm decreased at 600,1 200,2 400,and 4 800 mg kg-1Pb compared with the control,but increased slightly at 300 mg kg-1Pb(Fig.4b).The percentage in the range of 0.6–1.0 mm remained almost unchanged from 0 to 1 200 mg kg-1Pb(with only slight decrease at 300 mg kg-1Pb)and then roughly doubled at 2 400 and 4 800 mg kg-1Pb.The percentage in the range of 0.3–0.6 mm increased slightly from 0 to 1 200 mg kg-1Pb and then decreased to almost 0%at 2 400 and 4 800 mg kg-1Pb.Thus,the individuals in the range of 1.0–1.4 mm were grouped as“larger juveniles”,those in the range of 0.6–1.0 mm as“intermediate juveniles”,and those in the range of 0.3–0.6 mm as“smaller juveniles”.

Fig.4 Percentage distribution of different size ranges of body length by 0.1-mm increments from the shortest to longest individuals,with 1.7–2.6 mm for adults(a)and 0.3–1.4 mm for juveniles(b),of F.candida after 28-d exposure to Pb-polluted soil with nominal added Pb concentrations of 0,300,600,1 200,2 400,and 4 800 mg kg-1.

Individual abundances in different size ranges of body length

Adult individual abundances in more than half of the different size ranges of body length were significantly different among the treatments,except those in the ranges of 1.7–1.8,2.2–2.3,2.4–2.5,and 2.5–2.6 mm(Table II).Generally,the abundances in the ranges below 2.0–2.1 mm did not change with increasing added soil Pb concentration from 0 to 1 200 mg kg-1,but the abundances in the ranges of 1.7–1.8 and 1.8–1.9 mm were lower than those in the ranges of 1.9–2.0 and 2.0–2.1 mm(Table III).In contrast,the abundances in the ranges above 2.1–2.2 mm decreased with increasing added soil Pb concentration from 0 to 1 200 mg kg-1.No adults survived in any of the ranges from 1.7–1.8 to 2.5–2.6 mm at 2 400 and 4 800 mg kg-1Pb.

TABLE IIF and P values of one-way analysis of variance of abundance of adult and juvenile individuals in different size ranges of body length of F.candida after 28-d exposure to Pb-polluted soil with nominal added Pb concentrations of 0,300,600,1 200,2 400,and 4 800 mg kg-1

In contrast,juvenile individual abundances were significantly different among soil Pb concentration treatments in most size ranges of body length,except those in the ranges of 0.3–0.4,1.2–1.3,and 1.3–1.4 mm(Table II).The abundances in the ranges from 0.3–0.4 to 0.5–0.6 mm did not change from 0 to 1 200 mg kg-1Pb,but then sharply declined to zero or extremely low values at 2 400 and 4 800 mg kg-1Pb.The abundances in ranges from 0.6–0.7 to 0.9–1.0 mm generally decreased with increasing added soil Pb concentrations from 0 to 4 800 mg kg-1,while the abundances in ranges from 1.0–1.1 to 1.3–1.4 mm generally decreased from 0 to 1 200 mg kg-1Pb and then sharply declined to extremely low values or zero at 2 400 and 4 800 mg kg-1Pb(Table III).According to these analyses,the adult individual abundances in the range of 1.7–1.9 mm were grouped together as“smaller adults”,those in the range of 1.9–2.1 mm as“intermediate adults”,and those in the range of 2.1–2.6 mm as“larger adults”.Similarly,the juvenile abundances in the range of 0.3–0.6 mm were grouped as“smaller juveniles”,those of 0.6–1.0 mm as“intermediate juveniles”,and those of 1.0–1.4 mm as“larger juveniles”.These ranges were consistent with those for the percentage distribution of body length(Fig.4).

All the larger adults(F5,29=4.407,P=0.004),intermediate adults(F5,29=12.39,P=0.000),and smaller adults(F5,29=7.604,P=0.000 1)were significantly different among the soil Pb concentration treatments,but the patterns of change in the three types of adults were different(Fig.5).The individual abundance of the larger adults was higher in the control and lower soil Pb concentration treatments,but it declined from 600 mg kg-1Pb compared to the control,declining by 52.9%for the 1 200 mg kg-1Pb treatment and finally reaching 0 at 2 400 and 4 800 mg kg-1Pb(Fig.5a).The abundance of intermediate adults was also higher in the control and lower Pb concentration treatments,but declined from 1 200 mg kg-1Pb,directly to 0 at 2 400 and 4 800 mg kg-1Pb(Fig.5b).The abundance of smaller adults in the control was much lower than that of the larger or intermediate adults,and it increased with added soil Pb concentration from being lower in the control to being the highest at 1 200 mg kg-1.Then,it declined sharply at 2 400 and 4 800 mg kg-1Pb(Fig.5c).

Fig.5 Abundance of adult(a–c)and juvenile(d–f)individuals in three size groups(smaller,intermediate,and larger)of body length of F.candida after 28-d exposure to Pb-polluted soil with nominal added Pb concentrations of 0,300,600,1 200,2 400,and 4 800 mg kg-1.Bars indicate standard errors of means(n=5).

TABLE IIIAbundance of adult and juvenile individuals in different size ranges of body length of F.candida after 28-d exposure to Pb-polluted soil with nominal added Pb concentrations of 0(Pb0),300(Pb300),600(Pb600),1 200(Pb1200),2 400(Pb2400),and 4 800(Pb4800)mg kg-1

As for juveniles,although the individual abundances of all the larger juveniles(F5,28=10.02,P=0.000),intermediate juveniles(F5,28=20.28,P=0.000),and smaller juveniles(F5,28=7.180,P=0.000 2)decreased with increasing added soil Pb concentration,the patterns of change in the three types of juveniles were also different.The individual abundance of the larger juveniles was higher in the control and lower soil Pb concentration treatments and declined by 22.0%,91.5%,99.3%,and 99.5%at 300,600,1 200,2 400,and 4 800 mg kg-1Pb,respectively,compared to the control(Fig.5d).The abundance of the intermediate juveniles was also higher in the control and lower soil Pb concentration,but declined from 600 mg kg-1Pb,declining by 51.8%,97.2%,and 98.0%at 1 200,2 400,and 4 800 mg kg-1Pb,respectively(Fig.5e).Similar to the intermediate juveniles,the abundance of the smaller juveniles was higher in the control and lower soil Pb concentration treatments,but declined from 600 mg kg-1Pb,declining by 14.4%,100%,and 99.8%at 1 200,2 400,and 4 800 mg kg-1Pb,respectively(Fig.5f).

DISCUSSION

Generally,adult survival,reproduction,and growth,as well as the Pb concentration,in body tissues all changed with increasing soil Pb concentration.However,the nature of the changes and the sensitivity of the reaction were different.

Lead accumulation in collembolan

Lead concentrations in the body tissues of both adults and juveniles increased linearly with increasing soil Pb concentration.These results are in agreement with other published studies on Pb contaminated soil(Bongerset al.,2004;Buret al.,2012;Luoet al.,2014a)and soils contaminated with other potentially toxic elements such as Cd(Crommentuijnet al.,1993;van Gestel and Hensbergen,1997;Ardestaniet al.,2013;Zhuet al.,2016a,b)and Cu(Pedersenet al.,2000).The Pb accumulated byF.candidais stored mainly in the midgut epithelial cells and can be excreted by exfoliating the epithelial cells of the middle intestine by the process of moulting(Fountain and Hopkin,2001;Ardestani and van Gestel,2013).The high Pb concentration inF.candidain the present study indicates that soil Pb can enter and accumulate in the animal body and exert toxicity to the animal.The BAF values of adults(0.02–0.42)and juveniles(0.02–0.05)were quite low and similar to those previously reported by Fountain and Hopkin(2001)(0.04–0.6)and Luoet al.(2014a)(0.07–0.2).Fountain and Hopkin(2001)showed that the BAF values of Pb are extremely low compared with other metals such as Cd(0.05–0.29),Cu(0.05–18.1),and Zn(0.07–3.79).The BAF values for Pb are often very low in most studies,and this may be attributed to the low Pb bioavailability in soil(Luoet al.,2014a).

Toxic effects of soil Pb contamination on adult survival and reproduction

There was no effect on adult survival at 300,600,1 200 mg kg-1Pb,while no adult springtail survived at 2 400 and 4 800 mg kg-1Pb.Nursitaet al.(2005)reported thatProisotoma minutaTullberg(Collembola)survival showed no significant differences at any Pb contamination level(300,750,1 500,and 3 000 mg kg-1).Some studies also showed thatF.candidasurvival appears to be insensitive to Pb contamination(Mentaet al.,2006;Luoet al.,2014a).There was no clear toxic effect and no significant change in survival whenF.candidawas fed with Pb-contaminated yeast food(49 200 mg kg-1)(Fountain and Hopkin,2001).This may be due to the different exposure pathways to the Pb-contaminated food,since food exposure is much less toxic than soil exposure because of the ability to detect and avoid the contaminated food and the exuviation to excrete the accumulated metals in the midgut epithelial cells(Humbert,1978;Joosse and Buker,1979;Pawertet al.,1996;Fountain and Hopkin,2001).Lead toxicity to soil invertebrates is related to the concentration and also potentially related to the metal salt anions(counterions)(Zhang and van Gestel,2017).Bongerset al.(2004)reported that the toxicity of Pb(NO3)2was stronger than PbCl2forF.candida,which might partly explain the survival ofF.candidabeing sensitive to Pb contamination in present study.

The reproduction ofF.candidaalso decreased with increasing soil Pb concentration.As expected,reproduction was more sensitive to Pb pollution than adult survival,showing that the decrease in reproduction started at 600 mg kg-1Pb which was lower than that for adult survival(1 200 mg kg-1Pb).Similar to present study,other studies showed that the reproduction of the collembolansSinella coecaandF.candidadeclined at 1 000 mg kg-1Pb in soil(Mentaet al.,2006)and that the amount of the springtailParonychiurus kimiin the sublethal endpoints(population)decreased to almost zero at 1 312 mg kg-1Pb(Sonet al.,2007).It was reported that the reproduction ofF.candidadeclined significantly at a total soil Pb concentration of 3 877 mg kg-1as nitrate salt in spiked soils(Locket al.,2006),a much higher Pb concentration than that used in the present study.The difference in toxicity may be due,at least in part,to the different soils used.In the study of Locket al.(2006),the soil used had higher soil pH(7.3)than the pH of the soil in present study(5.1),and higher soil pH might reduce the toxicity of Pb.Buret al.(2012)found a significant decrease in reproduction at the highest Pb concentration in three soils with the same series of Pb-spiked concentrations(0,300,600,1 200,2 400,and 4 800 mg kg-1)but with different soil pH and soil organic matter contents.

Toxic effects of soil Pb contamination on body length

Body length has seldom been included in past toxicology studies,especially for soil Pb pollution(Crommentuijnet al.,1993;Tranviket al.,1993;Folker-Hansenet al.,1996;Fountain and Hopkin,2001;Domeneet al.,2007;Buret al.,2010;Zhuet al.,2016b).The main reason is that the body length is difficult to determine,especially in juveniles,which are very small and numerous.The conventional method is to measure the body length of all individuals and then calculate the mean value as an index for comparison between treatments.In the present study,the mean body length of both adults and juveniles generally did not decrease with increasing soil Pb concentration(Fig.3),suggesting that the mean body length is not sensitive to soil Pb.Previous studies also showed that the mean body length was not sensitive enough(Domeneet al.,2007;Buret al.,2010,2012;Zhuet al.,2016b).Springtails can resist pollution by sacrificing the energy needed for growth to maintain reproductive capacity in order to ensure their population is large enough(Hubbell,1971).Thus,the number of individuals larger in body size could start declining at lower pollution levels and decline continuously with increasing pollutant concentration,while smaller individuals might remain unchanged or increase at the lower pollution level and then decline at higher metal concentrations.Consequently,the overall mean body length will not necessarily decrease at this point.

To search for more sensitive body length features than the mean body length,the individuals were separated into different ranges according to their body length to understand if and how individuals differing in body size responded differently to Pb pollution.Based on the analysis of percentage distribution at 0.1-mm intervals,it was found that the percentage of larger adult individuals with body length in the range of 2.1–2.6 mm had similar response patterns and generally decreased with soil Pb concentration,while those of intermediate adults of 1.9–2.1 mm were almost unchanged and those of smaller adults of 1.7–1.9 mm generally increased.This suggests that the percentage of adults of different body lengths may have different sensitivity to soil Pb pollution,with larger adults being more sensitive than the intermediate and smaller adults.Similarly,it was found that,with increasing added soil Pb concentration,the percentage of larger juveniles of 1.0–1.4 mm decreased gradually from 600 mg kg-1Pb until the highest Pb concentration.In contrast,the percentage of intermediate juveniles of 0.6–1.0 mm was unchanged at Pb concentration≤1 200 mg kg-1and then increased,while the percentage of smaller juveniles of 0.3–0.6 mm increased slightly initially at 1 200 mg kg-1and then decreased.This also suggests that the percentage of juveniles of different body lengths may have different sensitivities,with the larger juveniles being more sensitive than the intermediate and smaller ones.

In order to further analyse body length features,the individual abundance difference in body length was analysed.Similarly,based on the analysis of individual abundance in different body length ranges at 0.1-mm intervals,the individual abundances in the ranges with similar response patterns to Pb concentrations were grouped together.Consistent with the percentage distribution,the individual abundances of larger adults with body length in the range of 2.1–2.6 mm had similar response patterns and decreased at lower(600 mg kg-1)added soil Pb concentrations.Contrastingly,those of intermediate adults of 1.9–2.1 mm decreased at higher(1 200 mg kg-1)Pb concentrations,while those of smaller adults of 1.7–1.9 mm were much lower than those of large or intermediate adults and increased slightly initially and then declined.Similarly,the individual abundance of larger juveniles of 1.0–1.4 mm declined at low Pb concentrations(300 mg kg-1)and then gradually decreased with increasing Pb concentration,while those of the intermediate(0.6–1.0 mm)and smaller(0.3–0.6 mm)juveniles declined at higher Pb concentration(600 mg kg-1)and then declined gradually.These results suggest that the abundances of individuals of different body lengths can also be different in sensitivity to soil Pb pollution,and the larger individuals are more sensitive than the intermediate and smaller ones both for adults and juveniles.

Considering these results,it seems that the larger the individuals in body size,the earlier(at the lower Pb concentration)the individual abundance decline,i.e.,features of larger individuals are more sensitive to Pb pollution.The juveniles may be more sensitive than the adults because the abundances(or even the percentage distributions)of juveniles declined earlier or changed more gradually with increasing Pb concentration than those of the adults.This may be due to a relatively larger surface area to volume ratios of juveniles than adults,thus being more readily exposed to contaminants(Folker-Hansenet al.,1996;Scott-Fordsmandet al.,1997,1999).On the other hand,the newly formed body surface waxes of juveniles are thinner and more susceptible to contamination than the body surfaces of adults(Noble-Nesbitt,1963).Crouau and Moïa(2006)observed a positive correlation between the length of animals and the number of juveniles,concluding that body length differences in polluted and control soils had a tendency to decrease with time.Once maximum length was reached,animals could not grow anymore and the smaller individuals in polluted soils could catch up with those in the control.Crommentuijnet al.(1997)suggested that larger individuals are basically consistent with reproduction,showing that a decline in larger individuals may finally cause a decrease in reproduction.Some studies have also indicated that reproduction may be influenced indirectly when metal exposure leads to growth retardation.The elongated juvenile period and decreased number of animals reaching the reproductive state may be explained by delayed growth(Smitet al.,2004).This may suggest that growth and reproduction may be complementary and suitable for use in ecotoxicological evaluation.

CONCLUSIONS

Soil Pb contamination can strongly reduce adult survival,reproduction,and growth(measured by body length)of the soil collembolanF.candidaand thus exert toxic effects on the animals.Lead can enter the body tissues of the animals,and the Pb concentration in the body coincides with the Pb concentration in soil.This may represent an important pathway causing toxicity to the collembolans.Reproduction is generally more sensitive than adult survival.In terms of growth,the mean body length of all individuals is generally insensitive to Pb pollution levels,but the percentage and abundance of individuals with different body lengths are quite different in their sensitivity,especially at low Pb contamination levels.The percentage and abundance of larger individuals are more sensitive than those of smaller ones in both adults and juveniles,but juveniles are more sensitive than adults.Therefore,the percentage and abundance,especially of larger individuals,could be proposed as an ecotoxicological test parameter and would be a potential index for the ecotoxicity assessment of soil pollution of Pb and possibly other potentially toxic elements.Since Pb pollution can be harmful to soil springtails and therefore disturb their ecological function,the test procedures should be conducted strictly according to the standards published in the OECD guidance.

ACKNOWLEDGEMENTS

This research was supported by the National Key R&D Program of China(No.2018YFC1802602)and the National Natural Science Foundation of China(No.41977136).The authors thank the anonymous reviewers and Prof.Peter Christie from the Institute of Soil Science,Chinese Academy of Sciences for their comments,which helped improve the manuscript.