Pathways of trace metal uptake in the lugworm Arenicola marina

Aquatic Toxicology 92 (2009) 9–17

Contents lists available at ScienceDirect

Aquatic Toxicology journal homepage: www.elsevier.com/locate/aquatox

Pathways of trace metal uptake in the lugworm Arenicola marina M.C. Casado-Martinez a,∗ , B.D. Smith a , T.A. DelValls b , P.S. Rainbow a a b

Department of Zoology, The Natural History Museum, Cromwell Road, London SW7 5BD, United Kingdom Unesco UNITWIN Wicop Chair, Department of Physical-Chemistry, University of Cadiz, Poligono Industrial Rio San Pedro s/n, C.P. 11510 Puerto Real, Cadiz, Spain

a r t i c l e

i n f o

Article history: Received 10 November 2008 Received in revised form 19 December 2008 Accepted 23 December 2008 Keywords: Bioavailability Biodynamic model Polychaete Cadmium Silver Zinc

a b s t r a c t Radiotracer techniques were used to determine the rates of trace metal (Ag, Cd and Zn) uptake and elimination (33 psu, 10 ◦ C) from water and sediment by the deposit-feeding polychaete Arenicola marina, proposed as a test species for estuarine-marine sediments in whole-sediment toxicity tests. Metal uptake rates from solution increase with increasing dissolved metal concentrations, with uptake rate constants (± SE) (l g−1 d−1 ) of 1.21 ± 0.11 (Ag), 0.026 ± 0.002 (Zn) and 0.012 ± 0.001 (Cd). Assimilation efficiencies from ingested sediments were measured using a pulse-chase radiotracer feeding technique in two different lugworm populations, one from a commercial supplier (Blyth, Northumberland, UK) and the other a field-collected population from the outer Thames estuary (UK). Assimilation efficiencies ranged from 2 to 20% for Zn, 1 to 6% for Cd and 1 to 9% for Ag for the Northumberland worms, and from 3 to 22% for Zn, 6 to 70% for Cd and 2 to 15% for Ag in the case of the Thames population. Elimination of accumulated metals followed a two-compartment model, with similar efflux rate constants for Zn and Ag and lower rates of elimination of Cd from the slow pool. Efflux rate constants (± SE) of Zn and Ag accumulated from the dissolved phase were 0.037 ± 0.002 and 0.033 ± 0.006 d−1 whereas Cd was eliminated with an efflux rate constant one order of magnitude lower (0.003 ± 0.002 d−1 ). When metals were accumulated from ingested sediments, the efflux rate constants for the slow-exchanging compartment were of the same order of magnitude for the three metals, and of the same order of magnitude as those derived after the dissolved exposure for Zn and Ag (0.042 ± 0.004 and 0.056 ± 0.012 d−1 for Zn and 0.044 ± 0.012 and 0.069 ± 0.016 d−1 for Ag for the Northumberland and Thames populations, respectively). Cd accumulated from ingested sediments was eliminated with a rate constant not different from the fast-exchanging compartment after the water-only exposure (0.025 ± 0.012 and 0.020 ± 0.004 d−1 for the Northumberland and Thames populations, respectively). A biodynamic model was used to estimate the relative importance of the dissolved phase versus ingested sediment as source of metal for the worms, showing that more than 90% of the Zn and Cd and more than 70% of Ag in lugworms is accumulated from sediment ingestion at realistic environmental concentrations. The model also shows that metal accumulation is highly dependent on the ingestion rate and assimilation efficiency. © 2009 Elsevier B.V. All rights reserved.

1. Introduction In recent years the use of biodynamic models has gained importance for the study of trace metal bioaccumulation in marine invertebrates. This type of model requires experimental data measured under environmentally realistic conditions in order to derive physiological parameters for each metal-species. Luoma and Rainbow (2005) found that a biodynamic model accurately predicted metal accumulation for a wide range of metals, animals, and aquatic habitats based on data derived from 15 separate studies when either water, food or both is the major uptake pathway of metals. Compilations of such data on model input parameters are available for some species and metals, in the case of benthic invertebrates especially

∗ Corresponding author. Tel.: +44 20 7942 5565; fax: +44 20 7942 6126. E-mail address: [email protected] (M.C. Casado-Martinez). 0166-445X/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.aquatox.2008.12.010

for bivalves and crustaceans. In these last cases biodynamic modeling has also succeeded in explaining causality in whole-sediment copper toxicity tests (Simpson and King, 2005), highlighting the importance of feeding behaviour (ingestion rates, selectivity of feeding) and the partitioning of metal between the sediments and water (Kd ) in controlling the total copper exposure that these organisms would receive during whole-sediment toxicity tests. Despite the fact that deposit-feeding polychaetes are widely used test organisms in sediment toxicity tests (Thain and Bifield, 2001; ASTM, 2007; Alvarez-Guerra et al., 2007), few studies have dealt with exposure pathways. Available studies on the depositfeeding polychaetes Nereis succinea and Capitella sp. observed that more than 95% of the total amount of metal accumulated (Cd, Co, Se, Zn) arose from sediment ingestion as a result of high ingestion rates and reduced bioavailability of dissolved metals to these organisms (Selck et al., 1998; Wang et al., 1999). The lugworm Arenicola marina has been found to be sensitive and potentially useful for sediment

10

M.C. Casado-Martinez et al. / Aquatic Toxicology 92 (2009) 9–17

monitoring, due to its abundance and key ecological role in the lower intertidal zone of sandy to muddy-sandy shores along Western European coasts, being a major prey for shorebirds and fish (Bat, 1998). However, the detailed ecotoxicology of this species is still not well documented. It is known that A. marina ingests large quantities of sediment that, after passing through the gut, is deposited on the sediment surface forming characteristic casts while water is pumped for ventilation (Riisgård and Banta, 1998). Bat (1998) observed an increase in the concentration of Cu, Zn and Cd in A. marina tissues with increasing metal sediment concentrations, and Packer et al. (1980) found a positive correlation between tissue concentrations and those in the sediment for Cd and Zn. Miramand et al. (1982) found that A. marina directly takes up americium and plutonium from sediment, although none of these experiments studied the uptake pathways and the proportion of metal from sediments that is bioavailable to lugworms. The objective of this study is to model the accumulation of Ag, Cd and Zn by the lugworm A. marina. We studied the assimilation efficiencies from ingested sediment and the uptake rates from the dissolved phase of Cd, Ag and Zn in this worm. Then a biodynamic model was used to compare quantitatively the contribution of sediment ingestion and dissolved metals in overall metal uptake by A. marina as a first approach to understanding the pathways of metal uptake in this keystone coastal polychaete. 2. Materials and methods 2.1. Organisms Lugworms A. marina (1.6–4.0 g wet weight) were purchased in November 2007 and January 2008 from SEABAIT Ltd. (Blyth, Northumberland, UK). The worms were transported in sea water with charcoal to adsorb any excess of mucus produced and arrived in the laboratory within 12–24 h in good condition. Once in the laboratory, the lugworms were maintained in approximately 10 cm of sediment from the shore at Two Tree Island, outer Thames estuary, covered by artificial sea water (TM: Tropic Marin, Tropicarium Buchschlag, Dreieich, Germany) at a salinity of 33 at 10 ◦ C. Artificial sea water was used in order to reduce and standardize chelation of dissolved trace metals and at the same time produce good replicability. For further experiments of assimilation from food, lugworms (0.9–6.6 g wet weight) were also collected by hand from the Thames estuary in May 2008. They were brought to the laboratory in TM and, once in the laboratory, they were handled identically as the worms from Northumberland. 2.2. Metals The metals considered in this study were the essential metal Zn and the non-essential metals Cd and Ag. The radioisotopes 65 Zn and 109 Cd (Brookhaven National Laboratory, New York, USA) and 110M Ag (Riso National Laboratoty, Denmark) were used to track metal accumulation in the experiments using a LKB Wallac 1480 Wizard gamma counter. A standard solution was used with each batch of measurements to determine the level of background radiation and any changes in the activity of the metals due to radioactivity decay during the experimental exposure. 2.3. Metal uptake and depuration from solution Metal uptake from the dissolved phase was determined by exposing lugworms at different concentrations of each metal individually over time in acid-washed plastic boxes (7 cm × 7 cm × 5 cm) at 10 ◦ C. Concentrations were 10, 40, 60 and 100 ␮g l−1 for Zn; 1, 3, 6 and 10 ␮g l−1 for Cd and 0.6, 2, 4 and 6 ␮g l−1 for Ag. These concentrations were labelled with 2.5 ␮Ci l−1

for Zn and Cd, and 0.25 ␮Ci l−1 for Ag, and brought to the final concentration through the addition of non-labelled metal. Exposure chambers used for the experiments were soaked in the corresponding metal solution for 72 h in order to saturate adsorption sites. That same day, the organisms were acclimated to the exposure conditions by transferring them to plastic boxes with 50 ml of artificial sea water. This also allowed the lugworms to empty the gut. On the day of exposure, the worms were transferred to fresh radiolabelled metal solutions in the pre-soaked exposure chambers and were left to stand for 2 h, in order to make an assessment of any adsorption of labelled metal onto the body surface. The radioactive content of each worm was then counted at intervals of 24 h for 96 h; after each count the worms were returned to exposure chambers containing 50 ml of renewed radiolabelled solution. The uptake rate (Iw ) was calculated for each individual as the slope of the best-fit regression line of labelled metal concentrations in the lugworm (expressed as ␮g metal g−1 d.w.) against time up to 96 h. The uptake rate constant ku was then calculated as the slope of the best-fit regression line correlating individual worm uptake rates and the corresponding exposure concentrations without the y-intercept set through zero. After the 96 h exposure counts (100%), the worms were transferred to 50 ml of TM without added metal, and the radioactive content of the lugworms was measured after 2 h to account for any loss of weakly bound metals. The loss during the initial 2 h depuration is considered to be the adsorbed metal which has not crossed the cell membrane and has not therefore been taken up by the organism (Luoma and Rainbow, 2008). Over the following two weeks the radioactive counts of the worms were measured at intervals of 48–72 h, after which they were transferred to fresh TM. kew was calculated as the slope of the best-fit regression line between time and the loge -transformed percentage of metal retained in the lugworm. 2.4. Radioactive pulse-chase feeding and depuration Due to the feeding biology of A. marina, which feeds by swallowing subsurface sediment while burrowed, it was not possible to offer radiolabelled sediment as a traditional food source without exposing the worm to radiolabelled metal in sediment porewater. To avoid metal uptake from any other route apart from gut passage, up to 10 worms were fed individually by injecting a small quantity of radiolabelled sediment into the proboscis after previous stimulation with clean sediment. For each metal separately, 0.25 g of wet sediment from the site of collection in the outer Thames estuary was labelled with 2 ␮Ci of radiotracer and left to stand at 4 ◦ C for at least 1 month. This sediment was characterised by approximately 50% of particles < 125 ␮m and 3% weight of organic matter (measured as loss of weight after ignition at 550 ◦ C). The worms had spent the previous 72 h acclimating to the exposure conditions in plastic boxes with 50 mL of artificial sea water. This also allowed the lugworms to empty the gut. After the meal, lugworms were then washed with TM, and the radioactive content counted to determine the amount of labelled metal ingested. Initially the worms were not directly introduced into sediment to facilitate the collection of casts for counting for radiolabelled metals. The lugworms were then placed individually in TM and were fed unlabelled sediment following the same technique. In subsequent experiments worms directly introduced into clean sediment showed similar metal assimilation and elimination results to those obtained when lugworms were not buried in sediment immediately after feeding. The radioactivity contents of water, worms and casts (if produced) were counted every 48–72 h during the following two weeks. Assimilation efficiency (AE) was calculated as the percentage of ingested labelled metal that was retained in the worms after egestion of unassimilated metal. Worms defaecated non-assimilated metal as a single or multiple casts produced at different times. We considered the

M.C. Casado-Martinez et al. / Aquatic Toxicology 92 (2009) 9–17

gut passage time (GPT) as the time between the introduction of the radiolabelled sediment into the lugworm and the time to produce the last cast with a significant amount of counts. For the calculation of kef after the ingestion of sediment-bound metal, a similar approach to that used for water-only exposure was followed, considering as 100% the radiolabelled count used for the calculation of AE. 2.5. Data analysis Regression analyses, a priori analysis of variance (ANOVA) and post hoc a posteriori ANOVA using Tukey’s HSD test for unequal numbers were carried out using STATISTICA. Percentage data were arcsine-transformed before statistical analysis. All errors are expressed as standard errors (SE).

11

3. Results 3.1. Assimilation and depuration of metals from the dissolved phase Metal accumulation with exposure time and dissolved ambient concentration was best described by linear functions for all exposure conditions. Under the experimental conditions followed in this study (33 psu, 10 ◦ C), the amount of trace metal taken up from solution by lugworms increased linearly (Fig. 1), indicating that there was no saturation of uptake over the range of concentrations used to derive ku . The uptake rates from water for all three metals displayed a significant (p < 0.001) positive linear relationship

2.6. Biodynamic modeling of trace element accumulation in lugworms According to the dynamic multi-pathway bioaccumulation model, net accumulation is the result of a balance between uptake rate from diet, uptake rate from solution, and loss rates after uptake from either source (Luoma and Rainbow, 2005). Metal concentrations at steady-state can be determined by the equation CSS =

(ku × CW )/kew + (AE × IR × CF )/kef g

ku , kew , AE and kef have been calculated in this study, whereas values for ingestion (IR) and growth rates (g) have been obtained from the available literature. Ingestion rates were calculated using the empirical model developed by Cammen (1980) for aquatic deposit feeders and detritivores. This model is based on observations on the inverse correlation of ingestion rates of total dry material with the organic matter of the food, organic matter ingestion being a function of body size. The final equation describing this relationship is: IR = 0.435 W 0.771 OM−0.920 where IR is expressed in mg total dry matter ingested per day, W is mg dry body weight, and OM is the fraction of organic material in the food. We used a standard size of organism of 300 mg and a proportion of OM of 3%, which is the value measured in the sediment used in the experiments to measure metal assimilation from food. The calculated IR was then expressed as grams of OM ingested per gram of organism per day, considering that lugworms feed preferentially on the organic fraction. A growth rate constant (g) for A. marina of 0.02 d−1 has been estimated for a control population feeding on sand, although it may reach values of 0.06 and 0.07 d−1 when sand is enriched with additional food (WO/2003/007701), these values being very similar to those reported for Abarenicola pacifica (0.023–0.052 d−1 ; Taghon, 1988). According to the variability of these parameters, we performed a sensitivity analysis to evaluate the effects of changes in IR, AE and Kd on the relative importance of sediment versus dissolved metal uptake in lugworms. Parameters were adjusted individually, while holding all other parameters at average values. We considered ranges of sediment metal concentrations normally found in the literature for estuaries and coastal waters (Luoma and Rainbow, 2008): sediment concentrations were adjusted to 200 ␮g Zn g−1 , 1.5 ␮g Cd g−1 and 0.75 ␮g Ag g−1 ; and concentrations in sea water were considered to be 2.5 ␮g Zn l−1 , 0.035 ␮g Cd l−1 and 0.001 ␮g Ag l−1 . The range of variation of Kd within each metal was that found in estuaries (Turner and Millward, 2002; Luoma and Rainbow, 2008).

Fig. 1. Arenicola marina: relationships between accumulated labelled metal concentration (mean ± SD, n = 6; ␮g metal g−1 dry wt) and exposure time at different dissolved exposure concentrations (33 psu, 10 ◦ C).

12

M.C. Casado-Martinez et al. / Aquatic Toxicology 92 (2009) 9–17

Fig. 2. Arenicola marina: relationships between uptake rates of Zn, Cd and Ag from solution and dissolved metal concentration (33 psu, 10 ◦ C). Equations describing the best-fit linear relationship between metal uptake rates (Iw , ␮g g−1 d−1 ) and dissolved metal concentration (Cw , ␮g l−1 ) are: Zn, Iw = 0.026Cw + 0.105 (r2 = 0.91); Cd, Iw = 0.012Cw + 0.272 (r2 = 0.86); Ag, Iw = 1.21Cw + 0.272 (r2 = 0.84).

with the dissolved concentrations (Fig. 2). The metal uptake rate constant (ku ± SE), described as the slope of the best-fit regression between the uptake rate and metal concentration in the dissolved phase (Cw ), was greatest for Ag (1.21 ± 0.11 l g−1 d−1 ) followed by Zn (0.026 ± 0.002 l g−1 d−1 ) and Cd (0.012 ± 0.001 l g−1 d−1 ). After 96 h exposure to the dissolved metal concentrations, lugworms were transferred to clean sea water and were counted regularly in order to assess efflux rate constants and biological retention half-lives of the assimilated metals. After the initial 2 h depuration lugworms still contained 100.3 ± 1.2% of the total accumulated Zn, showing that the percentage of contribution of surface

Fig. 3. Arenicola marina: percentage of metal retained in lugworms after uptake from solution (96 h exposure) to different dissolved ambient concentrations.

adsorbed zinc was negligible and that almost all the Zn accumulated by lugworms had therefore been taken up by the worm (Luoma and Rainbow, 2008). The highest desorption was observed at the highest exposure concentration, although adsorption only accounted for a maximum of 4% of the total Zn. The depuration of Zn was characterised by a two-compartmental loss (Fig. 3). The lugworms initially eliminated 12–30% of the accumulated Zn rapidly from a ‘fast’ compartment. Lugworms exposed to the highest concentration of dissolved Zn depurated a significantly (p < 0.01) higher proportion of Zn through the fast compartment than at lower exposure concentrations (12 ± 3%, 21 ± 3%, 13 ± 3% and 30 ± 4% at exposure concentrations 10, 40, 60 and 100 ␮g l−1 , respectively). Nevertheless, the kew s for the fast compartment did not differ significantly between exposures,

M.C. Casado-Martinez et al. / Aquatic Toxicology 92 (2009) 9–17

13

Table 1 Compartmental analysis of metal depuration in the lugworm Arenicola marina after exposure to dissolved metal concentrations for 96 h (mean ± SE). kew (d−1 )

Metal

Compartment (d)

Percent in slow compartment

Zn

0–3 4–14

82.9 ± 2.0

0.097 ± 0.006 0.037 ± 0.002

7.1 18.7

92.4 ± 1.6

0.018 ± 0.002 0.003 ± 0.002

38.5 213c

70.2 ± 3.2

0.080 ± 0.006 0.033 ± 0.006

8.6 21.0

Cd

a b

Ag

0–5/7d 5/7–14e

t1/2 (d)

a Calculated using data from fast pool at concentrations 1, 3 and 6 ␮g l−1 ; compartment covering 0–14, 0–3 and 0–5 d, respectively. b Calculated using data from slow pool at 3, 6 and 10 ␮g l−1 ; compartment covering 3–14, 5–14 and 0–14 d, respectively. c Only calculated when Kew is < 0 (n = 8). d Calculated using data from fast pool at concentrations 0.6 and 2 ␮g l−1 ; compartment covering 0–5/7 and 0–5 d, respectively. e Calculated using data from slow pool at 0.6, 2, 4 and 6 ␮g l−1 ; compartment covering 5/7–14 at 0.6 ␮g l−1 , 5–14 d at 2 ␮g l−1 and 0–14 d at 4 and 6 ␮g l−1 .

and a mean kew of 0.097 ± 0.006 d−1 is given in Table 1, corresponding to a half-life of 7.1 d. Similarly there was no significant difference between exposures for the efflux rate constants for the remaining metal in the slow compartment, and again a mean kew (0.037 ± 0.002 d−1 ) is given in Table 1, corresponding to a half-life of 18.7 ± d. The proportion of the total accumulated Ag that was eliminated during the initial 2 h depuration increased with increasing exposure concentrations. 5–10% of the total Ag was eliminated during the initial 2 h depuration in clean sea water for the 0.6 and 2 ␮g l−1 exposure concentrations, whereas at higher exposure concentrations desorption accounts for a higher proportion of the total Ag accumulated, with 83 ± 3 and 84 ± 2% of the total Ag accumulated at 4 and 6 ␮g l−1 , respectively remaining after the initial 2 h of depuration in clean sea water. At 0.6 and 2 ␮g l−1 exposure concentrations, lugworms eliminated approximately 20% of the remaining Ag from a fast compartment during the following 5–7 days with an efflux rate constant of 0.080 ± 0.006 d−1 , corresponding to a mean halflife of 8.6 ± d. This fast compartment could not be distinguished at 4 and 6 ␮g l−1 , and the remaining Ag after desorption was eliminated with an efflux rate constant that was not significantly different from kew s for the fast pool at 0.6 and 2 ␮g l−1 , and a mean kew of 0.033 ± 0.006 d−1 is given is Table 1, corresponding to a half-life of 21.0 ± d. The elimination pattern of Cd during the period of observations differed depending on the dissolved ambient concentrations during the uptake phase (Fig. 3), although the proportion of Cd eliminated through desorption was similar across the range of exposure concentrations (3.4 ± 0.6%). Lugworms exposed to the lowest and intermediate concentrations (1 to 6 ␮g l−1 ) are able to eliminate up to 10% of the Cd accumulated with an efflux rate constant of 0.018 ± 0.002 d−1 from a ‘fast’ compartment, although the time needed to eliminate the Cd accumulated in this compartment was 14 d when it was accumulated at 1 ␮g Cd l−1 , 3 d at 3 ␮g l−1 and 5 d at 6 ␮g l−1 . The efflux rate constants for the ‘slow’ compartment were not statistically different from 0 (p > 0.01), and corresponded to retention half-lives higher than 100 d. After 14 d of depuration in clean sea water the Cd remaining in the lugworms was between 84 and 91%. Body size did not have any significant effect on metal efflux in this study. 3.2. Assimilation and depuration from sediment particles Significant egestion through the production of radiolabelled casts was evident just after 2 h of depuration for some worms

Fig. 4. Arenicola marina: percentage of metal retained in lugworms after uptake from ingested sediment.

whereas others took up to one week to produce the last cast with a significant amount of radiolabelled metal (Fig. 4). No statistical difference in GPT was found between metals for each site and between sites for each metal. Average GPT was 71 ± 14 h for Zn, 128 ± 21 h for Cd and 96 ± 12 h for Ag. Lugworms from Northumberland assimilated Zn more efficiently than Cd or Ag (significant at p < 0.05), with AEs of 8.5 ± 1.7%, 3.3 ± 0.6% and 3.5 ± 0.9%, respectively (Table 2). The experiments carried out with worms collected from the outer Thames resulted in a higher variation between individuals and higher AEs for the three metals studied than for Northumberland worms (12.5 ± 2.4% for Zn, 43.6 ± 8.5% for Cd and 8.0 ± 2.2% for Ag). Only AEs of Cd were statistically different between the two populations (p = 0.00018).

14

M.C. Casado-Martinez et al. / Aquatic Toxicology 92 (2009) 9–17

Table 2 Compartmental analysis of metal depuration in two populations of the lugworm Arenicola marina following feeding on radiolabelled (Zn, Cd and Ag) sediment (mean ± SE, n = 7–10). Metal

Population

AE (%)

kef (d−1 )

t1/2 (d)

Zn

Northumberland Thames

8.5 ± 1.7 12.5 ± 2.4

0.042 ± 0.004 0.056 ± 0.012

16.5 12.3

Cd

Northumberland Thames

3.3 ± 0.6a 43.6 ± 8.5a

0.025 ± 0.012 0.020 ± 0.004

27.8 34.0

Ag

Northumberland Thames

3.5 ± 0.9 8.0 ± 2.2

0.044 ± 0.012 0.069 ± 0.016

15.8 10.1

a

Statistically significant difference at p < 0.01.

After the initial defaecation of radiolabelled metals through casting, Ag and Zn assimilated from ingested sediment were eliminated with an efflux rate constant not significantly different from that of metals accumulated from the dissolved phase held in the slow pool (p > 0.05). The Zn assimilated from sediment ingestion was eliminated with an efflux rate constant of 0.042 ± 0.004 and 0.056 ± 0.012 d−1 for Northumberland and Thames worms, respectively, corresponding to mean biological half-lives of 16.5 d and 12.3 d (Table 2). The Ag assimilated from food was eliminated by Northumberland worms with an efflux rate constant (0.044 ± 0.012 d−1 ) similar to that observed for Zn, whereas slightly higher values were observed for Thames worms (0.069 ± 0.016 d−1 ). As for elimination after uptake from the dissolved phase, the lowest values of kef were obtained for Cd. The Cd assimilated from sediment ingestion was eliminated with an efflux rate constant of 0.025 ± 0.012 and 0.020 ± 0.004 d−1 for Northumberland and Thames worms, corresponding to mean biological half-lives of 27.8 and 34.0 d, respectively (Table 2). Cd kef values were not statistically different from kew from the fast pool (p > 0.05). There was no significant difference between the efflux rate constants for the two populations studied. Metal assimilation and efflux from sediment ingestion was neither associated with the metal concentration in sediment nor with the body size of the worms. 3.3. Biodynamic modeling of trace metal accumulation in lugworms The modeling results showed that, for most of the Kd range investigated, over 95% of the body burden of Zn and 90% of the Cd in lugworms is obtained from sediment ingestion for an standard sediment concentration of 200 ␮g g−1 Zn, 1.5 ␮g g−1 Cd and 0.75 ␮g g−1 Ag. Ag shows a higher contribution of the dissolved phase, providing up to 30% of the total Ag accumulated in lugworms (Fig. 5). We then performed a sensitivity analysis to evaluate how variations in the physiological parameters IR and AE modify the contribution of sediment ingestion to the total metal body burden in lugworms (Fig. 6). The ingestion rate of organic matter depends on the dry weight of the worm (see Section 2.6 above after Cammen, 1980) and therefore the dry weight of the worm might be expected to affect the contribution of sediment metal to total metal accumulated. Fig. 6a shows that in fact there was no effect of changes of dry weight of worm on this contribution for Zn in either populations, nor for Cd for the Thames population. For Cd in the Northumberland population and for Ag in both populations, there was increased accumulation of metal from sediment with changes in dry weight from 0.8 to 0.1 g (Fig. 6a), a 10% increase for Cd and increases of 10% for Ag in Thames and Northumberland populations. When OM content in the sediment increases from 3–15%, the proportion of metal accumulated from sediment increases by 1% for Cd and Zn and 3% for Ag for a standard sized worm (0.3 g dry weight) in both populations (Fig. 6b).

Fig. 5. Percentage of total metal uptake obtained from sediment ingestion for Arenicola marina calculated by a biodynamic model for different combinations of sediment and water concentrations, calculated for the low range of Kd , for Northumberland and Thames populations (filled and empty symbols, respectively), for sediment concentrations of 200 ␮g g−1 Zn, 1.5 ␮g g−1 and 0.75 ␮g g−1 Ag.

The most significant determinant of the relative contribution of dissolved versus sediment ingestion in our biodynamic model was AE. For the pair of sediment-water concentrations, the contribution of the dissolved phase in the total body burden may become significant when the AE takes on values below 5% (Fig. 6c). The differences in this parameter for the two populations studied increases the relative contribution of sediment by approximately 10% for Cd and Ag in the Thames population in relation to the Northumberland ones independently of the IR (Fig. 6). 4. Discussion In this study we considered firstly the uptake of metals from solution, and for this purpose we exposed lugworms to a range of concentrations of each metal in sea water. A similar approach was followed by Bernds et al. (1998) to verify toxicokinetic models for A. marina from the German Wadden Sea. In that study, Zn internal concentrations in lugworms increased approximately from 90 to 150 ␮g g−1 when they were exposed to 300 ␮g l−1 Zn for 5–10 days, which is equivalent to an uptake rate constant between 0.02 and 0.04 l g−1 d−1 , similar to the uptake rate constant calculated in this study for lugworms exposed for 4 days to concentrations below 100 ␮g l−1 . For Cd, lugworms exposed to 20 ␮g l−1 Cd increased from approximately 0.8 to 5.5 ␮g l−1 in 7 days (Bernds et al., 1998), which corresponds to an uptake rate constant of 0.03 l g−1 d−1 , slightly higher than our ku calculated for lugworms exposed for 4 days to concentrations below 10 ␮g l−1 . When the uptake rate constants describing metal uptake from solution by A. marina are compared with those developed for other deposit-feeding marine invertebrates under similar experimental conditions, a good agreement is observed (Table 3). Cd uptake rate constants are very similar in A. marina and N. succinea (Wang et al., 1999), whereas ku s for Ag and Zn are slightly lower in A. marina. At similar experimental conditions, uptake rate constants from solution for A. marina are in the range of values obtained for different populations of the polychaete Nereis diversicolor from metal contaminated estuaries (Rainbow et al., in press), showing little interspecific variation between these two species. The rank order of uptake rate constants from solution was also consistent with that common among most marine invertebrates (Ag > Zn > Cd; Luoma and Rainbow, 2005). Food has not been considered to be an important source for metal uptake in aquatic invertebrates until recently (Wang, 2002),

M.C. Casado-Martinez et al. / Aquatic Toxicology 92 (2009) 9–17

Fig. 6. Percentages of total metal accumulation obtained from sediment ingestion for Arenicola marina, calculated by a biodynamic model for different IR calculated for ranges of different organisms’ weight, percentage of organic matter content in sediment, and AE. Estimations for Northumberland and Thames populations are represented by filled and empty symbols, respectively.

when the techniques to measure chemical assimilation had been further developed and standardized (Wang and Fisher, 1999). AE measurements in deposit-feeding invertebrates have been carried out on a limited number of species compared to filter-feeders (Table 4). A. marina assimilates the essential metal Zn from sediment with a higher efficiency than the non-essential Cd and Ag, as has been observed in other aquatic invertebrates (Wang and Fisher, 1999), although lugworms assimilate Zn and Ag from sediment ingestion less efficiently than other polychaetes such as N.

15

succinea (Wang et al., 1999) and N. diversicolor (Rainbow et al., in press). Cd AEs for Northumberland worms were always among the lowest observed in marine invertebrates, whereas the population from the outer Thames assimilated Cd more efficiently than the population from Northumberland, with AEs more similar to those of other marine deposit-feeding invertebrates. Both populations of lugworms were studied under the same experimental conditions, and the only variant between batches of experiments was the radiolabelling time of sediments. The labelling time may significantly affect AEs in animals feeding on sediments because metals associate with the different sediment components, i.e. easily exchanging pools, iron and manganese oxides, carbonates, labile and possibly refractory organic matter, and sulfides, as a function of labelling time (Jannasch et al., 1988). AEs in marine bivalves are two to five times lower for sediments radiolabelled for 6 months than for the same sediments labelled for 3 d (Wang and Fisher, 1999). AEs of Cd for N. succinea were reduced significantly when ageing was increased from one week to 2 months whereas for other metals such as Ag, Se and Zn AEs seem not to be affected by radiolabelling time (Wang et al., 1999). AEs for the Northumberland population were derived from experiments carried out during the third month after radiolabelling, whereas AEs for the outer Thames population were fed on the sediments the following month. Another source of this observed variability in Cd AE might be physiological variability between the Northumberland and outer Thames lugworms, for example in the factors that control the digestive release and uptake of dietary trace metals in the gut (Luoma and Rainbow, 2008). Several studies have explored differences in metal uptake between populations at sites with different degrees of metal contamination (Wang and Rainbow, 2005). It is difficult, however, to make generalizations of how a history of metal exposure affects dietary uptake (Luoma and Rainbow, 2008). Metal (Cd, Zn) dietary uptake increased in the green mussel Perna viridis and the clam Ruditapes philippinarum after prior metal exposure, in correlation with increasing MT concentrations (Blackmore and Wang, 2002), but Rainbow et al. (2004) found no significant differences of the AEs of Cd, Zn and Ag between populations of the barnacle Balanus amphitrite with different ambient metal exposures. These authors suggested that the induction of metallothioneins as a detoxification mechanism in R. philippinarum, as opposed to a reliance on metal-rich granules as in barnacles, could be associated with this difference in metal assimilation (Wang and Rainbow, 2005). The polychaete Capitella capitata does bind accumulated trace metals to both MT and insoluble forms (Goto and Wallace, 2007), as does the polychaete N. diversicolor (Amiard et al., 2006), although a consistent pattern of variation in any of the biodynamic parameters controlling Ag, Cd and Zn has not been observed among six different populations of this last species from metal-contaminated estuaries (Rainbow et al., in press). Whereas the process of contaminant desorption in gut fluids of lugworms has been studied in-depth, demonstrating that lugworms rely strongly on amino acids for metal solubilisation as the first step for metal assimilation through the gut (Mayer et al., 1996; Chen and Mayer, 1999), metal incorporation, handling and detoxification strategies remain undetermined, as any presence of metallothionein-like proteins in A. marina is still poorly investigated. The accumulation of trace metal by an organism is the net result of two processes—the uptake of trace metal integrated across all uptake routes and the excretion of that metal summated across all excretion routes. In A. marina, efflux rates for Ag and Zn fall in the range of 0.01–0.05 per day, common values for many species (Tables 3 and 4). Zn and Ag are eliminated by A. marina with similar efflux rates independently of the source of metal, which suggests that it is not likely that different mechanisms of metal elimination exist in this species depending on the route of metal uptake.

16

M.C. Casado-Martinez et al. / Aquatic Toxicology 92 (2009) 9–17

Table 3 Uptake and efflux rate constants of metals from solution reported for marine deposit-feeding invertebrates. Animal

ku

Arenicola marinaa Nereis diversicolorb Nereis diversicolorc Nereis diversicolorc Nereis succinead Eunice sp.e a b c d e

kew

Reference

Ag

Cd

Zn

Ag

Cd

Zn

1.214 1.780–7.335

0.0120 0.0087–0.0267 0.016–0.078 0.005–0.091 0.010 0.0023

0.0260 0.0173–0.1021 0.041–0.088 0.009–0.031 0.064

0.0327 0.0172–0.0518

0.0029 0.0184–0.0336

0.0374 0.0235–0.0393

1.853

0.00288

This study Rainbow et al., in press Geffard et al., 2005 Zhou et al., 2003 Wang et al., 1999 Alquezar et al., 2007

Salinity 33, 10 ◦ C. Ranges calculated for two populations with different metal pre-exposure conditions; salinity 16, 10 ◦ C. Salinity 16, 10 ◦ C. Salinity 28, 12 ◦ C. Salinity 31, 18 ◦ C.

The elimination of Cd was reduced compared to that observed for Zn and Ag. Lugworms only eliminated 20% of the total Cd accumulated from the dissolved phase at a similar rate to that observed for Cd accumulated through sediment ingestion over the period of observations. This efflux rate is lower than that observed for the other metals studied and in the lowest range of values recently reported by Rainbow et al. (in press) for the polychaete N. diversicolor, which is in the range common for the so called “strong bioaccumulators” (0.01–0.001 per day; Luoma and Rainbow, 2008). In terms of biological half-lives, these values of efflux rates imply that A. marina is able to eliminate up to 50% of the Ag and Zn accumulated in its tissues in less than 20 days independently of the route of metal uptake, whereas up to one year is necessary for lugworms to eliminate this same proportion of metal when the main source of Cd is the dissolved phase. These experimental data strongly suggest a limited capacity of A. marina to depurate Cd, although it is still to be investigated whether all Cd incorporated through sediment ingestion undergoes this elimination pattern or rather, as observed for the dissolved phase, some proportion of the accumulated Cd is held in a slow-exchanging compartment of delayed appearance not observed in this study. In this study, we quantified the rate constants of metal uptake and elimination from solution and sediment ingestion (ku , kew , AE and kef ) following standard laboratory techniques, and the ranges of values obtained were used to model the pathways of metal accumulation for this species. The modeling results for a standard concentration of these metals in sediments and a range of concentrations in water defined by environmentally realistic Kd s, found in estuaries, showed that almost the total amount of Zn and Cd accumulated in lugworms is obtained from sediment ingestion whereas a significant (but minor) contribution of metal in solution occurs only in the case of Ag. These results are in agreement with previous studies of Cd uptake in the polychaete Capitella sp., which demonstrated that this species accumulates more than 95% of the total Cd from sediments (Selck et al., 1998), and studies on metal accumulation in the polychaete N. succinea (Wang et al., 1999), for which sediment ingestion is the major route of uptake for Cd, Zn, Se and Co with a higher contribution (5 to 35%) from the dissolved phase for Ag.

Variations in the ingestion rate for different sized organisms and sediments with an organic matter content in the range found in the field do not modify the relative contribution of sediment ingestion in the cases of Zn and Cd. Only different metal AEs at the bottom of the range produced large changes in this profile. It is important to notice that this increase in the contribution of the dissolve phase may be biased by the selection of abnormally high metal concentrations in sea water to run the model. We chose these specific concentrations according to those compiled by Luoma and Rainbow (2008). These concentrations were the highest measured in subsurface sea water samples from estuary mouths and near coastal sites around England and Wales in the early 1990s and concentrations in many of these water bodies may have declined in recent years. Moreover, for these concentrations to occur with the sediment concentrations selected, the distribution coefficients Kd would need to adopt abnormally low values. Because this exposure profile has been described in terms of a biodynamic model using laboratory-based measurements of key model parameters, its validity depends on the applicability of these parameters to natural conditions. The parameters of metal uptake from solution were derived after exposure of individuals to metal solutions in artificial sea water. Whereas this approach was selected because of its simplicity and can be readily compared with the values available in the literature, lugworms are very unlikely to be exposed to sea water out of their burrows. Variation in the values obtained for metal uptake rates in the lugworms probably do not reflect possible variations in the physiological parameters associated with the absence of the sediment burrow in the water-only exposures in contrast to typical field exposure of lugworms to an undetermined mixture of overlying sea water and total pore water from the surrounding area. The most determinant parameter in the model was AE. A. marina probably adapts its IR depending on the nutritive value of the sediment, contamination and other parameters such as temperature (Cammen, 1980; Retraubun et al., 1996; Bat and Raffaelli, 1998), so a large number of combinations of AEs and IRs could be expected in environments with different sediment types and metal loads. In conclusion, the quantification of the physiological parameters controlling metal uptake in lugworms did not show conspicuous

Table 4 Ranges of assimilation efficiencies and efflux rate constants measured for marine polychaetes after taking up radiolabelled metal from ingested sediment. Animal

AE

Arenicola marinaa Nereis succineab Nereis diversicolorc a b c

◦

kef

Reference

Ag

Cd

Zn

Ag

Cd

Zn

0.8–15 16–30 25–77

0.8–69 17–29 50–73

1.7–22 21–53 30–65

0.044–0.069

0.020–0.025

0.042–0.056

0.044–0.084

0.012–0.026

0–0.071

Salinity 33, 10 C; ranges of values calculated for two populations. Salinity 28, 12 ◦ C; feeding on oxic sediments. Salinity 16, 10 ◦ C; feeding on sediments; ranges calculated for different populations with different metal pre-exposure conditions.

This study Wang et al., 1999 Rainbow et al., in press

M.C. Casado-Martinez et al. / Aquatic Toxicology 92 (2009) 9–17

differences from the values for other marine deposit-feeding species, and supports the intuitive perception that the prevalent route of exposure to contaminants is sediment ingestion. Without doubt, and based on these results, the key to understanding the prevailing pathways of metal uptake in polychaetes relies on sound estimations of ingestion rates and assimilation efficiencies under different conditions occurring in nature. Further knowledge of the mechanisms of storage and possibly detoxification of trace metals depending on the route of metal uptake is necessary for understanding accumulation patterns and possible ecotoxicological effects of the accumulated concentrations in this organism (Wang and Rainbow, 2006). Acknowledgements This project was financially supported by the Ramon Areces Foundation through grant to M.C. Casado-Martinez and the European Community’s Seventh Framework Program through a Marie Curie Intra-European Fellowship (FP7/2007-2013 under grant agreement n◦ PIEF-GA-2008-219781). This project was also supported by the UNITWIN WiCoP environmental quality assessment program at University of Cadiz. References Alquezar, R., Markich, S.J., Twining, J.R., 2007. Uptake and loss of dissolved 109 Cd and 75 Se in estuarine macroinvertebrates. Chemosphere 67, 1020–1210. Alvarez-Guerra, M., Viguri, J.R., Casado-Martinez, M.C., DelValls, T.A., 2007. Sediment quality assessment and dredged material management in Spain: part II, analysis of action levels for dredged material management and application to the bay of Cadiz. Integr. Environ. Assess. Manag. 3 (4), 539–551. Amiard, J.C., Amiard-Triquet, C., Barka, S., Pellerin, J., Rainbow, P.S., 2006. Metallothioneins in aquatic invertebrates: their role in metal detoxification and their use as biomarkers. Aquat. Toxicol. 76, 160–202. ASTM Standard E1611, 2007. Standard Guide for Conducting Sediment Toxicity Tests With Polychaetous Annelids. ASTM International, West Conshohocken, PA, www.astm.org. Bat, L., 1998. Influence of sediment on heavy metal uptake by the polychaete Arenicola marina. Turkish J. Zool. 22, 341–350. Bat, L., Raffaelli, D., 1998. Sediment toxicity testing: a bioassay approach using the amphipod Corophium volutator and the polychaete Arenicola marina. J. Exp. Mar. Biol. Ecol. 226 (2), 217–239. Bernds, D., Wübben, D., Zuke, G.P., 1998. Bioaccumulation of trace metals in polychaetes from the German Wadden Sea: evaluation and verification of toxicokinetic models. Chemosphere 37 (13), 2573–2587. Blackmore, G., Wang, W.-X., 2002. Uptake and efflux of Cd and Zn by the green mussel Perna viridis after metal pre-exposure. Environ. Sci. Technol. 36, 989–995. Cammen, L.M., 1980. Ingestion rate: an empirical model for aquatic deposit feeders and detritivores. Oecologia 44, 303–310. Chen, Z., Mayer, L.M., 1999. Sedimentary metal bioavailability determined by the digestive constraints of marine deposit feeders: gut retention time and dissolved amino acids. Mar. Ecol. Prog. Ser. 176, 139–151.

17

Geffard, A., Smith, B.D., Amiard-Triquet, C., Jeantet, A.Y., Rainbow, P.S., 2005. Kinetics of trace metal accumulation and excretion in the polychaete Nereis diversicolor. Mar. Biol. 147, 1291–1304. Goto, D., Wallace, W.G., 2007. Interaction of Cd and Zn during uptake and loss in the polychaete Capitella capitata: whole body and subcellular perspectives. J. Exp. Mar. Biol. Ecol. 352, 65–77. Jannasch, H.W., Honeyman, B.D., Balistrieri, L.S., Murray, J.W., 1988. Kinetics of trace element uptake by marine particles. Geochim. Cosmochim. Acta 52, 567–577. Luoma, S.N., Rainbow, P.S., 2005. Why is metal bioaccumulation so variable? Biodynamics as a unifying concept. Environ. Sci. Technol. 37 (7), 1921–1931. Luoma, S.N., Rainbow, P.S., 2008. Metal contamination in aquatic environments. In: Science and Lateral Management. Cambridge University Press, Cambridge, UK. Mayer, L.M., Chen, Z., Findlay, R.H., Fang, J., Sampson, S., Self, R.F.L., Jumars, P.A., Quetel, C., Donald, O.F.X., 1996. Bioavailability of sedimentary contaminants subject to deposit-feeder digestion. Environ. Sci. Technol. 30, 2641–2645. Miramand, P., Germain, P., Camus, H., 1982. Uptake of americium and plutonium from contaminated sediments by the three benthic species: Arenicola marina, Corophium volutator and Scrobicularia plana. Mar. Ecol. Prog. Ser. 7, 59–65. Packer, D.M., Ireland, M.P., Wootton, R.J., 1980. Cadmium, copper, lead, zinc and mangenese in the polychaete Arenicola marina from sediments around the coast of Wales. Environ. Pollut. 22, 309–321. Rainbow, P.S., Ng, T.Y.-Y., Shi, D., Wang, W.-X., 2004. Acute dietary pre-exposure and trace metal bioavailability to the barnacle Balanus amphitrite. J. Exp. Mar. Biol. Ecol. 311, 315–337. Rainbow, P.S., Smith, B.D., Luoma, S.N. Differences in the bioaccumulation kinetics of trace metals among populations of the polychaete Nereis diversicolor from metal-contaminated estuaries. Mar. Ecol. Prog. Ser. in press. Retraubun, A.S.W., Dawson, M., Evans, S.M., 1996. Spatial and temporal factors affecting sediment turnover by the lugworm Arenicola marina (L.). J. Exp. Mar. Biol. Ecol. 201, 23–35. Riisgård, H.U., Banta, G.T., 1998. Irrigation and deposit feeding by the lugworm Arenicola marina, characteristics and secondary effects on the environment. A review of current knowledge. Vie Milieu 48 (4), 243–257. Selck, H., Forbes, V.E., Forbes, T.L., 1998. Toxicity and toxicokinetics of cadmium in Capitella sp. I: relative importance of water and sediment as routes of cadmium uptake. Mar. Ecol. Prog. Ser. 164, 167–178. Simpson, S.L., King, C.K., 2005. Exposure-pathway models explain causality in wholesediment toxicity tests. Environ. Sci. Technol. 39, 837–843. Taghon, G.L., 1988. The benefits and costs of deposit feeding in the polychaete Abarenicola pacifica. Limnol. Oceanogr. 33, 1166–1175. Thain, J., Bifield, S., 2001. Biological effects of contaminants: sediment bioassay using the polychaete Arenicola marina. ICES Techniques in Marine Environ. Sci. 29, 19. Turner, A., Millward, G.E., 2002. Suspended particles: their role in estuarine bioceochemical cycles. Estuar. Coast. Shelf Sci. 55, 857–883. (WO/2003/007701). Aquaculture of marine worms. Available online at www.wipo.int. Wang, W.X., Stupakoff, I., Fisher, N.S., 1999. Bioavailability of dissolved and sedimentbound metals to a marine deposit-feeding polychaete. Mar. Ecol. Prog. Ser. 178, 291–293. Wang, W.X., 2002. Interactions of trace metals and different marine food chains. Mar. Ecol. Prog. Ser. 243, 295–309. Wang, W.X., Fisher, N.S., 1999. Assimilation efficiencies of chemical contaminants in aquatic invertebrates: a synthesis. Environ. Toxicol. Chem. 18 (9), 2034–2045. Wang, W.-X., Rainbow, P.S., 2005. Influence of pre-exposure on trace metal uptake in marine invertebrate. Ecotoxicol. Environ. Saf. 61, 145–159. Wang, W.X., Rainbow, P.S., 2006. Subcellular partitioning and the prediction of cadmium toxicity to aquatic organisms. Environ. Chem. 3, 395–399. Zhou, Q., Rainbow, P.S., Smith, B.D., 2003. Tolerance and accumulation of the trace metals zinc, copper and cadmium in three populations of the polychaete Nereis diversicolor. J. Mar. Biol. Assoc. U K 83, 65–72.