Ecotoxicology DOI 10.

1007/s10646-013-1137-y

The use of elements as a substitute for biomass in toxicokinetic studies in small organisms
Nina Cedergreen Peter E. Holm Helle Marcussen

Accepted: 25 September 2013 Springer Science+Business Media New York 2013

Abstract Determining pollutant concentrations in the tissues of experimental test organisms is necessary for understanding uptake and excretion mechanisms of toxicants. Using small organisms can make the determination of organism biomass inaccurate. We here propose the use of selected tissue element contents as a proxy for tissue biomass. Forty different elements were determined in tissues of the two worm species Enchytraeus crypticus and Caenorhabditis elegans derived from cultures exposed to combinations of varying temperatures and sublethal concentrations of Cu and Cd. Three criteria were used to select good biomass indicators: The element concentration must (1) be present in concentrations above the limit of quantication of the analytical method, (2) must be stable and (3) must not be affected by the treatment. If the organisms are believed to have signicant amounts of soil in their gut, the element must also be present at higher concentrations in the tissue compared to the soil. The three elements K, Mg and P all lived up to the rst three criteria for both worm species, showing correlation coefcients between element content and tissue biomass of 0.97, 0.96 and 0.97 (n = 25) and 0.997, 0.998 and 0.992 (n = 10) for K, Mg and P in the E. crypticus and C. elegans, respectively. Only P would be an appropriate biomass indicator for organisms with a soil gut uptake assuming the tissue concentrations in soil eating organisms are similar to those measured in the present study. Using Mg as a biomass indicator on a verication dataset of Cu and Cd uptake in E. crypticus, compared to giving Cu and Cd content per individual organism, decreased the coefcient of variation from 31 21 to 21 17 % and from 34 22 to
N. Cedergreen (&) P. E. Holm H. Marcussen Department of Plant and Environmental Sciences, University of Copenhagen, Thorvaldsensvej 40, 1871 Frederksberg, Denmark e-mail: ncf@life.ku.dk

9.3 6.4 % for tissue Cu and Cd, respectively. We therefore conclude that the use of an element as a biomass indicator can reduce tissue concentration variability. Keywords Biomass indicator Uptake kinetic Metals Variability Dry weight

Introduction Determining pollutant concentrations in the tissues of experimental test organisms is necessary for understanding uptake and excretion mechanisms under different toxicant exposures and environmental conditions. Such determinations require sampling of several organisms over time for each treatment. Hence, small test organisms such as springtails, enchytraeids, mites and woodlice are often used for terrestrial test, while daphnids or gammarids are often preferred in aquatic experiments on toxicokinetics (Vijver et al. 2001; Spurgeon and Hopkin 1999; Ashauer et al. 2007; Heugens et al. 2003; Kramarz 1999; Lagisz et al. 2005). The most common way to express internal pollutant contents is to use concentrations per biomass fresh or dry weight (Vijver et al. 2001; Kramarz 1999; Lagisz et al. 2005), but also content per number of organisms have been used (Spurgeon and Hopkin 1999). For soil and sedimenteating animals fresh and dry weight measurements carries the risk of overestimating biomass, as the soil content of the gut will be included in the mass determination. In addition the error in weight determination due to gut content are likely to be very variable between individuals as gut content can vary considerably (Piearce 1978). Letting, for example, earthworms empty their gut in a non-soil environment prior to sampling (Lock et al. 2000; Spurgeon and Hopkin 1999), can introduce a bias, as they might also

123

N. Cedergreen et al.

excrete the toxicant during the depuration time of typically 1248 h. Weighing very small animals like springtails and small worms such as nematodes, with masses around 40 lg dry weight and 10 lg fresh weight per specimen, respectively (Fountain and Hopkin 2001), requires very accurate balances and consistent methods for the error in weight determination to be tolerable. As the determination of toxicokinetic parameters, such as uptake and elimination rates, are very reliant on accurate and consistent measures of biomass, a precise measure of biomass is of the utmost importance. When measuring organic contaminants or various metabolites, the internal contaminant content is sometimes given per protein content or nucleic acid equivalent, in cases where biomass estimates are difcult to make (Schroeder et al. 2006; Stitt and Schulze 1994). For quantication of the biomass of fungi in ecological studies, for example, the fungi specic membrane component ergosterol has been used as a biomass indicator (Charcosset and Chauvet 2001). If the contaminant is a bioactive metal, however, the digestion of the sample, necessary to release all internal metal, will decompose any organic molecules that one could use as a measure of biomass. The aim of this study was therefore to investigate, if any elements, which would not be affected by a total digestion, could act as robust biomass indicators for the worms Enchytraeus crypticus and Caenorhabditis elegans. We aimed to identify elements to be used as biomass indicators that lived up to the following criteria: (1) the content of the biomass indicator is well above the limit of quantication (LOQ) for the amount of biomass proposed to be available (2) there is a strong correlation between the total amount of biomass indicator and the tissue dry weight and (3) this correlation (where the slope gives the tissue concentration of the biomass indicator), is independent of external stressors such as changes in temperatures or pollutant level. In addition, if the gut content of soil is believed to be signicant, the element concentration in the soil should ideally be much smaller than that of the tissue in order to avoid the bias mentioned for weighing worms with gut soil contents, where tissue weights and soils weights cannot be distinguished. In this study the tissue content of 40 elements were tested and evaluated as possible substitutes for tissue dry weight in worms exposed to combinations of variable temperatures and sublethal concentrations of copper (Cu) and cadmium (Cd).

Materials and methods Test organisms The enchytraeids, E. crypticus, were reared from cultures originally established with worms obtained by the

terrestrial ecotoxicology group at Vrije University, Amsterdam. The worms were kept in LUFA 2.2 standard soil (Landwirtschaftliche Untersuchungs- und Forschungsanstalt (LUFA), Speyer, Germany) and were fed weekly with a mixture of autoclaved oat grains and sh food (protein content 33.5 %). The LUFA 2.2 soil is a loamy sand soil with a pH of 5.5 1.1 (CaCl2), organic carbon content of 2.1 0.4 %, cation exchange capacity of 10 0.5 cmol(?) kg-1, and water holding capacity of 55 4 %. To test the correlation between biomass and element content in differentially stressed worms the following experiments were carried out: Samples of worms exposed to sublethal concentrations of Cu (150 mg added Cu kg-1 dry soil, added as CuCl2 stock in MilliQ water), and Cd (40 mg added Cd kg-1, added as CdCl2 stock in MilliQ water), or no metals were raised at three different temperatures (11, 18 and 25 C). After 24 days large numbers of adult worms (1001,000 per treatment) were collected, rinsed, frozen and freeze dried. The nematodes, C. elegans, of the N2 Bristol strain (obtained from the C. elegans Genetics Centre) were cultivated at 21 C 0.5 C in darkness on plates of a modied bacteriological agar [nematode growth medium (NGM) (Brenner 1974)] and fed Escherichia coli, of the uracil decient strain OP50. The K, Mg and P content in the agar was 25, 1 and 25 mM. The cultures were maintained by transferring a chunk of agar from the existing culture plates to freshly prepared NGM agar plates each week. Age synchronization of the cultures was done by washing gravid adults of culture plates with autoclaved M9 media (3 g KH2PO4, 6 g Na2HPO4, 5 g NaCl, 1 mL 1 M MgSO4 per L MilliQ water, H2O to 1 L and then bleaching them with a 0.5 M NaOH, 1 % sodium hypochlorite solution to obtain liquid cultures containing eggs (http://www.wormbook.org/chapters/www_strainmaintain/ strainmaintain.pdf). These where then spread thinly on OP50 inoculated agar plates, and left to hatch at 21 C. When the synchronized cultures had reached the L4 larval stage, they were suspended in M9 media and spread on test plates (90 mm diameter), with approximately 1,000 individuals per plate. The plates were either clean agar, or agar spiked with 2 mg Cu L-1 and 0.1 mg Cd L-1 from CuCl2 and CdCl2 stocks in MilliQ water. Three plates per replicate and three replicates per treatment were used. All plates contained 50 lM of 5-uorodeoxyuridine (FUdR) (98 %, Alfa Aesar) to prevent the test animals from reproducing (Gandhi et al. 1980). The plates were checked daily and when necessary a few drops of concentrated OP50 suspension was added to the plate and spread with an inoculation hoop. After seven days, the worms were gently washed off the agar plates with MilliQ water into a 15 mL centrifuge tube. They were gently shaken to wash off excess bacteria and spun down for two minutes at 500 rpm

123

Elements as a substitute for biomass

(21.5 g) (Sigma, Centrifuge 2K15). The supernatant was removed with a pipette and approximately 10 mL of MilliQ water was added, the worms were gently shaken, spun down and the procedure was repeated another two times after which the nematodes were frozen and freeze dried. Verication dataset To verify the use of an element as a biomass indicator, Mg was used as a proxy for biomass in a study of uptake kinetics of Cu and Cd in E. crypticus grown at three temperatures over 24 days. Two times ve worms were taken out at each sampling time into oven dried preweighed eppendorph tubes. They were dried at 80 C and re-weighed on a ve-decimal balance (Sartorious MC210S). As the worm dry weight calculated after extracting the weight of the eppendorf tube was negative in approximately 30 % of the samples, it was concluded that accurate dry weight determination was not possible for the worms, which adhered to the inside of the eppendorf tubes. Hence, internal Cu and Cd concentrations were given per worm. Worm dry weight was later calculated from the element contents to be in the range of 12 mg per worm. The dataset comprised 52 and 47 measurements of Cu and Cd tissue concentrations. Uptake kinetics based on these data were compared to those based on Cu and Cd per lg sample Mg, assuming Mg can be used as a proxy for biomass. The Cu and Cd uptake kinetics, their dependence on growth temperatures and Cu and Cd availability in the soil and the resulting toxicity have been published in Cedergreen et al. 2013. The content of other elements than Cu and Cd and the variance of the data when expressed per worm versus when expressed per Mg content, has not been previously published. Chemical analyses To establish the correlation between the elements and the biomass, digestion of 1-20 mg worm were carried out using 0.250 mL conc. HNO3 (Baker Instra-Analyzed, J.T. Baker) and 0.125 mL 30 % H2O2 (puriss. p.a., SigmaAldrich) in a microwave assisted system operated at 350 W for 20 min with a 10 min ramp and 450 W for 30 min with a 5 min ramp (Multiwave 3000, rotor 64MG5, Anton Paar GmbH). The digestion was done for 25 and 12 samples of E. crypticus and C. elegans, respectively, grown as described above. Blanks and bovine liver standard reference material NIST 1577c (National Institute of Standards and Technology (NIST)) were included in each digestion for quality assurance. The concentrations of 40 elements (Ag, Al, As, Ba, Be, Ce, Co, Cr, Cs, Cu, Dy, Er, Eu, Fe, Gd, Ho, K, La, Li, Lu, Mg, Mn, Mo, Nd, Ni, P, Pb, Pr, Sb, Sc, Se, Sm, Tb, Th,

Tm, U, V, Y, Yb and Zn) were determined by inductively coupled plasma mass spectrometry (ICP-MS) (Agilent 7500C, Agilent Technologies, Manchester, UK) equipped with an octopole reaction system (ORS). External calibration was applied and drift within 10 % was corrected for by recalibration. Cadmium was determined by graphite furnace atomic absorption spectrometry (GFAAS) using Zeeman-effect background correction (Perkin Elmer 5100 AAS, HGA-600 graphite furnace) applying a 0.1 % Pd and 0.2 % Mg modier (Merck). Limit of detection and quantication were calculated as three and ten times the standard deviation of at least 8 replicate analyses of the calibration blank. To further investigate, whether the selected biomass indicator Mg could be used in studies of Cu and Cd in enchytraeids experiments where the total sample mass is too low to be determined by weighting, the following analysis was carried out for the verication dataset. Five worms per sample were digested with the same digestion procedure and quality assurance as described above. The concentrations of Cu or Cd were determined by graphite furnace atomic absorption spectrometry using Zeemaneffect background correction (Perkin Elmer 5100 AAS, HGA-600 graphite furnace). Magnesium concentrations were determined by ame atomic absorption spectrometry. The determined element concentrations in digestion blanks were all below the detection limit or less than 1 % of the sample concentrations. Determined concentrations of Mg, K, Co, Cu, Zn, Se and Mo in the digested NIST 1577c standard were within 10 % of the certied range and the concentration of the other certied elements in the reference material were below the limit of quantication. These results show that the digestion and analytical procedure in general is accurate and reproducible. Data treatment All elements with any samples below LOQ were discarded as potential indicators of biomass together with Cu and Cd, which formed part of the treatment. Then the element concentration per dry weight (DW) was calculated and the variance given as the coefcient of variation (CV) (standard deviation/average) was compared between elements, with the aim of nding the most stable tissue element concentration. To test for treatment effects, the element concentration for the three elements with the smallest CV was tested with a one-way ANOVA for C. elegans and with a two-way ANOVA for E. crypticus using temperature and metal treatment as the two independent variables. Finally, to test the consequences of using Mg as a biomass indicator, the CV of Cu and Cd per Mg content was compared to that of Cu and Cd per worm for the verication dataset (Cedergreen et al. 2013).

123

N. Cedergreen et al.

Results and discussion Enchytraeus crypticus samples had concentrations below the LOQ for the following elements: Ag, As, Be, Cs, Dy, Er, Eu, Gd, Ho, Lu, Mo, Se, Sm, Tb, Tm and Yb. For C. elegans the elements with samples below the LOQ were the same as for the enchytraeids with the addition of: La, Li, Nd, Pr, Sb, Se, Th, U and Y. These elements together with the two amended treatment metals/elements: Cu and Cd, were excluded for further investigation. A good element indicator of biomass for small organisms must be present in concentrations which allow quantication with a reasonable accuracy. To get a measure of the quantity of each element in digested samples relative to what is quantiable by the analytical method, the ratio of the observed smallest element concentration determined in a sample of 120 mg relative to the LOQ of that element was used. For the enchytraeids, the ve elements present in the highest quantity relative to what could be determined were Fe, Al, Mn, P and Mg, whereas for the nematodes it was P, Mg, Mn, Al and K (Table 1 and 2). For the lowest biomasses used in this study (2.8 mg DW for E. crypticus and 1.9 mg DW for C. elegans), these elements were present in 188813 fold and 671,380 fold the LOQ for E. crypticus and C. elegans, respectively. This means that the minimum biomass that can be quantied by use of a biomass indicator within their LOQ would be in the range of 1045 lg DW for the enchytraeids and 485 lg DW for the nematodes, depending on the choice of element. Such biomasses are in the very low range of what is accurately quantiable using even the most sensitive balances. Hence, using one of these elements as a biomass indicator, in the case of organisms within this size range, could possibly provide a more accurate estimate of biomass compared to what can be obtained using a balance. Or it could provide a suitable alternative in the laboratories that do not have access to balances that can weigh such small biomasses accurately. A good biomass indicator must, however, also be very stable and unaffected by the growth conditions of the organisms in question. Different metal stressors are known to affect the uptake and excretion of other elements (Goyer 1997). Hence, if working with pollutant stressors, or other stressors, it is important to ensure that the biomass indicators are not prone to covariate with the treatments, as this could cause a bias in the nal results. In the presented example, the three elements that were most stable biomass indicators across treatments for both enchytraeids (Table 1) and nematodes (Table 2) were K, Mg and P. For the enchytraeids, Co was also very stable, but as it is present in lower concentrations, its detection may not always be possible in studies with limited amount of biomass. For K, Mg and P measured on biomasses in the range of 120 mg DW, the CV of the element to biomass ratio (tissue concentration) was B26 % for the enchytraeids and B29 % for

Table 1 A measure of the sample minimum element concentration (Min) relative to limit of quantication (LOQ) is given together with the average element concentration given per dry weight SD and the corresponding CV for E. crypticus (n = 25) Element K Co P Mg Zn U Ca Pb V Ba Sr Mn Fe Al Li Th Pr Y La Ce Nd Sb Ni Cr Min:LOQ ratio 28 12 456 187 45 2 15 19 99 31 79 570 813 809 10 2 5 18 16 34 3 1 6 8 Element conc. (lg g-1) 4,308 854 0.53 0.11 4,722 1,022 543 139 82 23 0.29 0.08 1,153 382 4.1 1.5 5.6 2.0 16 6 4.2 1.7 55 23 855 376 1,504 681 1.5 0.7 0.36 0.18 0.29 0.15 0.86 0.46 1.37 0.73 2.5 1.4 1.2 0.6 0.10 0.06 2.4 1.8 3.9 3.2 CV % 20 21 22 26 28 28 33 36 36 37 39 42 44 45 46 50 53 53 53 55 55 57 76 80

Data are sorted with increasing CV for the element concentration

the nematodes (Tables 1, 2). In addition all three elements were within the top ve candidates with regards to the ratio between the minimum concentration and the LOQ, with the exception of K for the enchytraeids which were number 11 on the list with a minimum concentration to LOQ ratio of 28. The Pearson correlation coefcients between these elements and biomasses were 0.97, 0.96 and 0.97 for K, Mg and P in the enchytraeids (n = 25) (Fig. 1a), while in nematodes they were 0.73, 0.72 and 0.71 for K, Mg and P (n = 12). The relatively poor correlation between elements and biomass for the nematodes was primarily due to two outliers (Fig. 1b). If these were removed, the correlation coefcients were 0.997, 0.998 and 0.992 for K, Mg and P (n = 10). With such high correlation coefcients, treatment effects, if any, will be very minor. For the nematodes, an ANOVA analysis on the element to biomass ratio of the three treatments: control, Cu and Cd, showed, as expected, no signicant difference between treatments (p = 0.85, 0.93 and 0.94 for K, Mg and P). The two-way ANOVA on the element to biomass ratio made on the enchytraeid data,

123

Elements as a substitute for biomass Table 2 A measure of the sample minimum element concentration (Min) relative to Limit of Quantication (LOQ) is given together with the average element concentration given per dry weight SD and the corresponding CV for C. elegans (n = 12) Element K Mg P Mn Fe Zn Sr Co Ca Th V Ce Cr Ni Ba Al Min:LOQ ratio 67 572 1,380 133 63 33 25 1 38 3 5 1 1 1 3 100 Element conc. (lg g-1) 13,033 3,743 1,946 565 17,641 5,154 14 4 102 33 69 22 0.98 0.34 0.03 0.01 2,957 1,255 0.50 0.23 0.39 0.23 0.05 0.04 0.60 0.57 0.81 0.85 2.2 2.6 179 224 CV % 29 29 29 30 32 32 35 41 42 46 59 85 96 106 114 125

Data are sorted with increasing CV for the element concentration

however, showed a signicant effect of treatment (control, Cu and Cd treatment) (p \ 0.001), but not for temperature for K and Mg (11, 18 and 25 C) (p = 0.38 and 0.80), but for P (p = 0.002). The analyses showed signicant interactions for all elements (p \ 0.1). Performing a one-way ANOVA on treatment effects using a post hoc Tukey test showed that there was no difference between Cu and Cd treated worms for any of the elements, but that two high control values of the 11 C made the control treatment signicantly higher than the metal treated samples. Considering the ne correlations between the contents of selected elements and the biomass, and corresponding low

CV of element concentrations, we do not consider the ANOVA results problematic in the sense that it should disregard the use of the elements as biomass indicators. They do, however, show the importance of checking for treatment effects to be sure to identify if potential outliers could be treatment related. Neither nematodes nor the enchytraeids are believed to have a gut content affecting the weight measurements. The nematodes, because they grew on agar, the enchytraeids because reviews of enchytraeids feeding habits suggest that the small worms feed selectively on decomposed plant material and microorganisms and that the ingestion of large amounts of inorganic soil is unlikely (OConnor 1967; Rombke 2003). For organisms where the gut content of soil elements could be of quantitative importance, the element concentration in the soil should ideally be much lower than that of the tissue concentration in order to avoid a bias from the soil gut content. The three elements, K, Mg and P are present in soils in average concentration of around 15, 9 and 0.43 mg g-1 soil (Sposito 1999). Assuming that in average 10 % of the weight of for example an earthworm consisted of soil and that the earthworm had a tissue mineral composition of a nematode (Table 1), this would mean that 11, 34 and 0.27 % of the K, Mg and P, respectively came from the gut soil content. Hence, for such a case, only P would be suitable as a biomass indicator. The example shows the importance of including the relationship between soil and tissue content in cases where soil gut content might be of signicance. Finally using Mg as a biomass indicator for studying uptake kinetics of Cu and Cd, as opposed to using the Cu and Cd content per worm, showed a much smaller variation between replicates. For the 28 and 21 combinations of temperature and time where two or more samples were sampled in the Cu and Cd uptake experiments, respectively, the CV % decreased from in average 31 21 to 21 17 % for internal Cu concentrations and from 34 22 to

Fig. 1 Correlations between the tissue content of the three elements: Mg (black symbols), P (grey symbols) and K (open symbols) as a function of dry weight of the enchytraeid Enchytraeus crypticus (a) and the nematode Caenorhabditis elegans (b) for worms treated with either Cu (squares), Cd (circles) or without metals (triangles).

The correlations are given as full lines (Mg), broken lines (P) or dotted lines (K). Correlation coefcients are given in the text. The outlier samples marked with an asterisk are excluded from the correlations

123

N. Cedergreen et al.

Fig. 2 Uptake kinetics of Cd in Enchytraeus crypticus at three different temperatures given either as Cd content per individual (a) or as Cd per Mg content (b). The variation between replicates is markedly decreased using the biomass indicator, Mg. Using the correlation between Mg and biomass established in this study

(R2 = 0.96) makes it easy to convert the Mg to biomass. Data is described with a two parameter saturation kinetic model, y = a(1 - e-bx) where a is proportional to the slope of the curves while b denotes its maximum. Data is from Cedergreen et al. (2013)

9.3 6.4 % for internal Cd concentrations, when using the biomass indicator Mg rather than giving the tissue concentrations per individual. The kinetic curves for Cd are shown in Fig. 2. The reason for the slight shift in the three temperature dependent uptake kinetic curves, relative to each other, is most likely a reection of the worms varying slightly in size depending on growth temperature (pers obs.). CVs in the 30 % range and up to [100 % is, from a visual assessment of uptake kinetic curves, not an uncommon range (Amorim et al. 2005, 2011; Lock et al. 2000; Spurgeon and Hopkin 1999; Alvarez et al. 2006; Broerse et al. 2012). We therefore suggest that in many cases the use of an element as a biomass indicator can reduce that variability, leading to more accurate and reproducible estimates of uptake and depuration kinetics of metals and organic pollutants in small organisms. Until the generality of element to dry weight ratios across genotypes and between different laboratories are veried, however, we suggest that laboratory specic element to dry weight correlations are made for the organisms used in a particular study.
Acknowledgments We wish to thank Nils J. Nrhave for giving us access to the nematode data and to help with the metal analyses and Birgitte Boje Rasmussen for carrying out GFAAS and ICP-MS analyses. The work was supported by a grant given by the Department of Basic Sciences and Environment at the Faculty of Life Science, University of Copenhagen. Conict of interest of interest. The authors declare that they have no conict

References
Alvarez OA, Jager T, Redondo EM, Kammenga JE (2006) Physiological modes of action of toxic chemicals in the nematode Acrobeloides nanus. Environ Toxicol Chem 25:32303237

Amorim MJB, Rombke J, Soares AMVM (2005) Avoidance behaviour of Enchytraeus albidus: effects of benomyl, carbendazim, phenmedipham and different soil types. Chemosphere 59:501510 Amorim MJ, Oliveira E, Teixeira AS, Gravato CS, Loureiro S, Guilhermino LC, Van Gestel CA, Soares AMV (2011) Toxicity and Bioaccumulation of Phenanthrene in Enchytraeus Albidus (Oligochaeta: Enchytraeidae). Environ Toxicol Chem 30:967972 Ashauer R, Boxall ABA, Brown CD (2007) Modeling combined effects of pulsed exposure to carbaryl and chlorpyrifos on Gammarus pulex. Environ Sci Technol 41:55355541 Brenner S (1974) Genetics of Caenorhabditis Elegans. Genetics 77:7194 Broerse M, Oorsprong H, Van Gestel CA (2012) Cadmium affects toxicokinetics of pyrene in the collembolan Folsomia candida. Ecotoxicology 21:795802 Cedergreen N, Nrhave N, Nielsen K, Johansson HKL, Marcussen H, Svendsen C, Spurgeon D (2013) Low temperatures enhance the toxicity of Cu and Cd to Enchytraeus crypticus through different mechanisms. Environ Toxicol Chem 32:22742283 Charcosset JY, Chauvet E (2001) Effect of culture conditions on ergosterol as an indicator of biomass in the aquatic hyphomycetes. App Environ Microbiol 67:20512055 Fountain MT, Hopkin SP (2001) Continuous monitoring of Folsomia candida (Insecta: Collembola) in a metal exposure test. Ecotoxicol Environ Saf 48:275286 Gandhi S, Santelli J, Mitchell DH, Stiles JW, Raosanadi D (1980) A simple method for maintaining large, aging populations of Caenorhabditis Elegans. Mech Ageing Dev 12:137150 Goyer RA (1997) Toxic and essential metal interactions. Annu Rev Nutr 17:3750 Heugens EHW, Jager T, Creyghton R, Kraak MHS, Hendriks AJ, van Straalen NM, Admiraal W (2003) Temperature-dependent effects of cadmium on Daphnia magna: accumulation versus sensitivity. Environ Sci Technol 37:21452151 Kramarz P (1999) Dynamics of accumulation and decontamination of cadmium and zinc in carnivorous invertebrates. 1. The ground beetle, Poecilus cupreus L. Bull Environ Contam Toxicol 63:531537 Lagisz M, Kramarz P, Niklinska M (2005) Metal kinetics and respiration rates in F-1 generation of carabid beetles (Pterostichus oblongopunctatus F.) originating from metal-contaminated and reference areas. Arch Environ Contam Toxicol 48:484489 Lock K, Janssen CR, De Coen WM (2000) Multivariate test designs to asses the inuence of zinc and cadmium bioavailability in

123

Elements as a substitute for biomass soils on the toxicity to Enchytraeus albidus. Environ Toxicol Chem 19:26662671 OConnor FB (1967) The Enchytraeidae. In: Burges A, Raw F (eds) Soil biology, vol 8, 1st edn. Academic Press, London, pp 213257 Piearce TG (1978) Gut content of some lumbricid earthworms. Pedobiologia 18:153157 Rombke J (2003) Ecotoxicological laboratory tests with enchytraeids: a review. Pedobiologia 47:607616 Schroeder A, Mueller O, Stocker S, Salowsky R, Leiber M, Gassmann M, Lightfoot S, Menzel W, Granzow M, Ragg T (2006) The RIN: an RNA integrity number for assigning integrity values to RNA measurements. BMC Mol Biol 7:3 Sposito G (1999) The Chemistry of Soils. Oxford University Press, Oxford Spurgeon DJ, Hopkin SP (1999) Comparisons of metal accumulation and excretion kinetics in earthworms (Eisenia fetida) exposed to contaminated eld and laboratory soils. App Soil Ecol 11:227243 Stitt M, Schulze D (1994) Does Rubisco control the rate of photosynthesis and plant growth? An exercise in molecular ecophysiology. Plant Cell Environ 17:465487 Vijver M, Jager T, Posthuma L, Peijnenburg W (2001) Impact of metal pools and soil properties on metal accumulation in Folsomia candida (Collembola). Environ Toxicol Chem 20:712720

123