day bioaccumulation tests using - Wiley Online Library

Environmental Toxicology and Chemistry, Vol. 30, No. 6, pp. 1366–1375, 2011 # 2011 SETAC Printed in the USA DOI: 10.1002/etc.520

NOVEL CONTROL AND STEADY-STATE CORRECTION METHOD FOR STANDARD 28-DAY BIOACCUMULATION TESTS USING NEREIS VIRENS ERIN R. BENNETT,*yz JEFFERY A. STEEVENS,§ GUILHERME R. LOTUFO,§ GORD PATERSON,z and KEN G. DROUILLARDz yBioengineering Group, Salem, Massachusetts, USA zGreat Lakes Institute for Environmental Research, University of Windsor, Windsor, Ontario, Canada §U.S. Army, Engineer Research and Development Center, Vicksburg, Mississippi (Submitted 14 September 2010; Returned for Revision 3 November 2010; Accepted 3 January 2011) Abstract— Evaluation of dredged material for aquatic placement requires assessment of bioaccumulation potentials for benthic

organisms using standardized laboratory bioaccumulation tests. Critical to the interpretation of these data is the assessment of steady state for bioaccumulated residues needed to generate biota sediment accumulation factors (BSAFs) and to address control correction of day 0 contaminant residues measured in bioassay organisms. This study applied a novel performance reference compound approach with a pulse-chase experimental design to investigate elimination of a series of isotopically labeled polychlorinated biphenyl (13C-PCBs) in the polychaete worm Nereis virens while simultaneously evaluating native PCB bioaccumulation from field-collected sediments. Results demonstrated that all 13C-PCBs, with the exception of 13C-PCB209 (> 80%), were eliminated by more than 90% after 28 d. The three sediment types yielded similar 13C-PCB whole-body elimination rate constants (ktot) producing the following predictive equation: log ktot ¼ 0.09 log KOW 0.45. The rapid loss of 13C-PCBs from worms over the bioassay period indicated that control correction, by subtracting day 0 residues, would result in underestimates of bioavailable sediment residues. Significant uptake of native PCBs was observed only in the most contaminated sediment and proceeded according to kinetic model predictions with steady-state BSAFs ranging from 1 to 3 and peaking for congeners of log KOW between 6.2 and 6.5. The performance reference compound approach can provide novel information about chemical toxicokinetics and also serve as a quality check for the physiological performance of the bioassay organism during standardized bioaccumulation testing. Environ. Toxicol. Chem. 2011;30:1366–1375. # 2011 SETAC Keywords—Biota sediment accumulation factor

Nereis virens

Polychlorinated biphenyl

Pulse-chase

Sediment

potential to alter chemical toxicokinetics in bioassay organisms because of altered feeding or filtration activities, or physiological changes related to inducement of stress, or toxicity [14–16]. Time to steady state is controlled by the whole-body chemical elimination constant (ktot), and for hydrophobic organic contaminants this parameter has been shown to be inversely related to the n-octanol–water partition coefficient (log KOW) of the chemical [17–20]. There is some concern that highly hydrophobic compounds (log KOW > 5) may not achieve steady state in sediment bioassay test organisms during 28-d exposures [21]. For highly hydrophobic organic compounds, model-based correction is required to estimate steady-state concentrations. Similarly, for these chemicals, there is a need to evaluate whether the ktot value remains constant when bioassay specimens are exposed to different sediment types. Another issue commonly encountered in the implementation of sediment bioaccumulation tests involves the source of animals used for bioassays. Ideally, bioassay organisms are derived from laboratory cultures, to reduce genetic variation among individuals and ensure that background contamination of organisms prior to implementing the test procedures is minimized. However, some commonly utilized test species are not widely available from commercial sources and require that animals be field collected. Although collection sites are chosen to reflect uncontaminated conditions, for ubiquitous pollutants such as polychlorinated biphenyls (PCBs), residual contamination of field-collected test specimens will always be present [1]. Therefore, sediment bioaccumulation bioassays may require appropriate control correction procedures in order to correct for day 0 background contamination. As in the case for time to steady state, control correction methods also require an estimate of the chemical-specific ktot value. This is because day 0 contaminant

INTRODUCTION

Evaluation of contaminant bioavailability and bioaccumulation from sediments is often based on standardized laboratorybased bioaccumulation test methods [1–3]. Tissue residues from bioaccumulation tests are analyzed for contaminants and used to compare chemical bioavailability and bioaccumulation against tissue screening values [4,5] and used in food web models [6]. The main assumption of such evaluations is that the test organism achieves steady state with the sediments by the end of the bioassay period. In addition, appropriate modeling techniques can be used to extrapolate time-dependent bioaccumulated residues measured in the organism at the end of the test in order to predict steady-state chemical concentrations [3,7–9]. The steady-state condition occurs when chemical flux into the organism is balanced with chemical flux out of the animal, resulting in a time-independent stable body burden. The time required to achieve steady state is a characteristic of a variety of test conditions, including the bioassay species being utilized [9,10], environmental conditions under which the test is conducted (i.e., temperature, light, feeding rate) [11,12], and the chemical under evaluation [7,13]. Although standardized bioaccumulation assays control for test species and most environmental conditions, the same protocols are applied to fieldcollected sediments that vary in sediment characteristics (e.g., grain size and organic carbon content) and pollutant mixtures. These sediment-specific characteristics have the * To whom correspondence may be addressed ([email protected]). Published online 4 March 2011 in Wiley Online Library (wileyonlinelibrary.com). 1366

Toxicokinetics of PCB congeners in Nereis virens

residues are depurated from bioassay organisms to the bioassay environment over the duration of the experiment. The fraction of initial day 0 chemical remaining in animal tissues decreases with time and may approach a value of zero by the end of the bioassay for low-KOW chemicals. However, the initial day 0 burden will be partially retained for more hydrophobic compounds and will therefore contribute to the total body burden of chemical in the animal at the end of the test. The derivation of steady-state adjustment and the control correction factors are directly related to the whole-body elimination rate constant for a chemical [22]. Therefore, there is a need to provide accurate estimates of this critical parameter when implementing standardized bioaccumulation bioassays for priority pollutants. In this study, PCB elimination rates were quantified in the commonly utilized polychaete Nereis virens, in a standard 28-d sediment bioaccumulation assay used for the evaluation of chemical bioaccumulation potential of field-collected sediments [1]. The study used a novel pulsechase design that permitted simultaneous determination of whole-body elimination of isotopically (13C) labeled PCB congeners from bioassay organisms while at the same time evaluating the uptake and bioaccumulation of in-place PCBs derived from three field-collected sediments. The main objective of the present study was to determine congener-specific PCB depuration rates for a series of 13C-PCBs and to establish a predictive relationship to estimate ktot from chemical log KOW. A second objective was to evaluate whether the predictive relationship described above remained constant across different sediment types. Finally, the data sets were used to reevaluate recommended steady-state adjustment factors and control correction factors issued for Nereis virens during the implementation of standardized 28-d sediment bioaccumulation tests for regulatory purposes. MATERIALS AND METHODS

Experiment

A pulse-chase assay design was implemented to monitor the depuration of 13C-labeled PCB compounds in bioassay organisms and to measure the uptake of sediment-associated PCBs simultaneously. The test species used for the study was the polychaete worm Nereis virens, recommended for use for dredged materials testing in the Ocean and Inland Testing manuals and used routinely in the North Atlantic Region as part of dredged material environmental evaluations [1,2]. This species of polychaete is an active burrower (8–10 cm deep) and a deposit feeder in sediments. The test specimens used for the present study were acquired from Aquatic Research Organisms and originally collected from the Damariscotta River (Boothbay Harbor, ME, USA). Prior to implementing the study, animals were separated into nondosed control and treatment groups. For nondosed control animals, a 14-d clearance period was completed to allow animals to depurate any assimilated PCBs that were due to prior food- or sediment-borne exposures. During this 14-d clearance period, nondosed controls were held in 42 24 15 cm polyethylene chambers (85 g total biomass/ tank) containing 5 cm depth of New York Bight sand (40 820.210 N, 73 852.190 W) and reconstituted sea water (Instant Ocean; Aquarium Systems) and fed a daily diet of flake food (Aquatic Ecosystems). The New York Bight sand has been designated as a regional reference site for New York Harbor assessments. Treatment animals were housed under conditions identical to those outlined above but fed a flake diet spiked with

Environ. Toxicol. Chem. 30, 2011

1367

seven different 13C-PCB congeners (PCB 28, 52, 101, 138, 153, 180, and 209) obtained from Wellington Laboratories. The nominal food concentration was 450 ng/g total 13C-PCBs. At the end of the 14-d period, three nondosed control and three treatment group worms were removed to determine concentrations of both environmental PCB and 13C-labeled PCBs in the worms. Sediment bioassays

After the clearance/13C-PCB loading periods, animals were added to a series of exposure chambers containing one of three sediment treatments. The first exposure group consisted of the same New York Bight control sediment (here referred to as NY) used for the clearance/13C-PCB loading periods. The second exposure treatment was sediment collected from Sequim Bay, Washington, USA (here referred to as SB), an area of relatively low contamination [23]. The third exposure group consisted of contaminated sediments collected from the New York Harbor section of the Hudson River (40 846.690 N, 74 800.260 W, denoted as HR sediments). The NY sediment had a grain size distribution of 97.8% sand and 2.2% fines. Grain size distribution of the HR sediment consisted of 93.4% fines and 6.6% sand. Additional characteristics for each sediment treatment have been provided by Steevens et al. [24]. The exposure chambers used for bioassays consisted of 8-L glass containers containing 5 cm of sediment and reconstituted sea water. Approximately 10 g wet weight of worms (two or three individuals) was placed into a given chamber, and multiple chambers were used for each sediment treatment and experimental group. In total, 16 worms were placed in the HR experimental group, with 14 and 13 worms placed in the NY and SB experimental chambers, respectively. The exposures were conducted under static renewal conditions with 60% of the overlying water replaced daily and trickle-flow aeration provided for the duration of the study. Temperature was maintained at 20 8C 1 8C in the exposure chambers using water recirculating REMCOR heating/cooling units (REMCOR Products). Photoperiod was maintained at 500 to 1,000 foot-candles on a 16:8 h light:dark cycle with timer-controlled lights. Water quality parameters including pH, salinity, and dissolved oxygen were monitored each weekday in at least one chamber per treatment to ensure that rearing conditions maintained the minimum acceptable criteria outlined in standard methods. The quality of the overlying water was monitored using a model ABMTC handheld refractometer (Aquafauna BioMarine) for salinity, a model 315i meter (WTW) for pH, and a model Oxi 330 meter (WTW) for dissolved oxygen. Experimental worms (i.e., preloaded with 13C-PCBs) were added to exposure chambers containing each of the three sediment treatments. Nondosed control worms were added only to chambers containing contaminated HR sediments. Three replicate worms were destructively sampled from the nondosed control and experimental groups and from each sediment treatment (NY, SB, HR) on days 0, 7, 14, 21, and 28. At each time point, replicate animals from the nondosed control and treatment groups were taken from three different exposure chambers to avoid pseudoreplication artifacts. On removal, animals were rinsed with distilled water and anesthetized in an isotonic MgCl2 solution, and undigested sediment was removed from the gut by gently pressing a clean metal spatula along the length of individual animals. Organisms were then thoroughly rinsed with distilled water, blotted dry, weighed, and subsequently frozen at 80 8C. Once frozen, animals were homogenized by pulverization in a mortar and pestle over liquid nitrogen until

1368


E.R. Bennett et al.

ground to a fine powder according to Boese et al. [25]. Pulverized tissue from each time point replicate was then weighed into separate scintillation vials and stored at 20 8C until PCB analysis. Chemical analysis

Full details on the extraction and analysis methods have been provided by Lazar et al. [26]. Briefly, worm homogenates ( 5 g) were ground with Na2SO4 and spiked with three 13CPCB recovery standards (IUPAC 77, 126, and 169; Wellington Laboratories) and 1,3,5-tribromobenzene (TBB; AccuStandard). The homogenates were wet packed in glass columns containing 50 ml of 1:1 hexane:dichloromethane, allowed to stand for 1 h, followed by elution with another 250 ml of extraction solvent. Extracts were concentrated to 25 ml, and a 2-ml portion of the sample was removed for gravimetric lipid determination, with the remaining extract cleaned up by gel permeation chromatography followed by florisil chromatography [18]. Extracts were concentrated to 1 ml, capped in two gas chromatograph (GC) vials, and stored at 4 8C until instrumental analysis. Sediment samples were soxhlet extracted for 24 h using 1:1 acetone:hexane, and sample extracts were concentrated to 2 ml and subject to florisil as described above. Organic carbon content was determined by subjecting dried sediment to 450 8C for 24 h in a muffle furnace. Native PCB analyses were completed using a Hewlett-Packard 5890 GC equipped with a 5972 mass selective detector (MSD) and a HP-7673 autosampler. Coplanar PCB and 13C-PCB analyses were analyzed on Hewlett-Packard 6890 GC equipped with a Waters GCT-premier-time-of-flight (TOF) detector and a HP-7683B autosampler. Detection limits for PCB congeners, coplanar PCBs, and 13 C-PCBs ranged from 0.28 to 1.7 ng/g wet weight, from 0.046 to 0.073 ng/g wet weight, and from 0.002 to 0.306 ng/g wet weight, respectively. Method blanks and an in-house reference homogenate (Detroit River carp) were coextracted for every batch of five samples. Recoveries for the TBB internal standard averaged 99.2 0.9% and samples were not corrected for recovery. The PCB concentrations in blanks were near or below detection limits, and sample correction was not necessary. The SPCB concentrations for the in-house reference homogenate averaged 4,120 125 ng/g wet weight (mean SE) and were in compliance with the Great Lakes Institute for Environmental Research analytical laboratory’s quality control charts (mean 2 SD). Data analysis and bioaccumulation model

Whole-body elimination rate constants (ktot) were determined for each 13C-PCB congener from each sediment treatment using linear regression analysis to calculate the slope of the relationship between natural log-transformed 13C-PCB concentrations versus time. Linear regressions were then used to determine the relationships between log ktot and chemical log KOW for each sediment type. Differences between 13C-PCB congener ktot values among the sediment types were evaluated by analysis of covariance (ANCOVA). Elimination rate constants were used for the control correction procedure and to determine the proximity of worms to the steady-state condition at the end of the bioassay. Assuming firstorder elimination kinetics, the fraction of day 0 residues remaining in animals after a given exposure duration or at the end of the bioassay was estimated as follows. % Remaining ¼ eðktot tÞ 100%

(1)

The steady-state correction factor is given by the following equation. SS ¼

1 ð1ektot t Þ

(2)

These terms provide simple evaluation tools with which to assess the need for control correction or steady-state correction. A bioaccumulation model was used to apply both control and steady-state corrections to determine the steady-state concentration of worms. In the model, tissue concentrations of PCBs (ng/g wet wt) are partitioned into two fractions: the fraction of day 0 PCB residues remaining in polychaetes (COrg(c)) and bioaccumulated PCBs from in-place contaminants and/or pore waters (COrg(s)). The general model describing change in chemical concentrations in worms with time is given by Equation 3. dCOrg ¼ k1 CW þ kA CSed ktot COrgðsÞ ktot COrgðcÞ dt

(3)

where k1 and kA are uptake coefficients from water (ml/g/d) and sediment (g/g/d), ktot is the whole-body elimination rate constant (/d), and CW and CSed are chemical concentrations in water (or pore water; ml/g) and sediment (ng/g dry wt), respectively. Integration of Equation 1 yields the following. COrgðtÞ ¼

ðk1 CW þ kA CSed Þ 1ektot t þ COrgðcÞ ektot t ktot (4)

In practice, k1 and kA are not independently measured. However, the steady-state tissue concentration in the animal (COrg(SS)) is related to these terms according to Equation 5. COrgðSSÞ ¼

ðk1 CW þ kA CSed Þ ktot

(5)

Thus, combining Equations 4 and 5 and solving for COrg(SS) yields the followiong. COrgðtÞ COrgðcÞ ektot t COrgðSSÞ ¼ (6) 1ektot t The terms COrg(t), COrg(c), and ktot in Equation 6 are all experimentally measured and contribute to error propagation of COrg(SS). Monte Carlo simulations were performed to estimate errors associated with congener-specific COrg(SS) estimates. Standard deviations of COrg(28) and COrg(c) were obtained from replicate measurements and assumed to follow log-normal distributions. A relative standard deviation for ktot was derived from the mean residuals established from the predictive relationship between log ktot and log KOW and applied to all congeners. Monte Carlo simulations were run to estimate COrg(SS) over 1,000 iterations, and the standard deviation of the simulated distribution was determined. The biota sediment accumulation factor (BSAF) is operationally defined as follows. BSAF ¼

COrgðSSÞ XOC Xlipid CSed

(7)

where XOC and Xlipid are the mass fractions of organic carbon in sediments and lipids in animal tissues, respectively. Standard deviations associated with BSAF estimates were established using a Monte Carlo procedure similar to that described for COrg(SS) but also considered error propagation related to measurements of XOC and Xlipid.



1369

RESULTS

No mortalities were observed over the course of the clearance, preloading, or sediment bioassay components of the study. Background levels of environmental PCB contamination were detected in control and treatment worms following the initial clearance/13C-loading period. Mean SPCBs in control and treatment worms were 16.1 1.30 ng/g after the 14 d clearance/preloading period (day 0 of sediment bioassay) and were not significantly different between the control and treatment groups ( p ¼ 0.501). For the treatment groups, all seven 13C-PCB congener were assimilated from spiked food during the 14-d preloading period. The mean S13C-PCB concentration in treatment worms on day 0 of the sediment bioaccumulation assay was 182 18.6 ng/g. Some variability was evident in the extent of 13C-PCB assimilation by worms. For example, a coefficient of variation (CV) of 70.9% was determined for 13C-PCB28 in treatments at the end of the 14-d preloading period. However, for the remaining 13C-PCB congeners, CVs ranged from 14.1 to 37.1%. Labeled PCB congeners were not detected in any of the control organisms. Elimination of

13

C-PCBs

Significant elimination of all seven 13C-PCB congeners was achieved by the worms during the 28-d bioassay period. The 13 C-PCB28 concentrations in worms fell below detection limits by 21 d. In contrast, concentrations of 13C-PCB180 measured in worms by day 28 averaged 4.4, 10.6, and 8.9% of the day 0 concentrations for worms from the HR, NY, and SB sediment bioassays, respectively. Whole-body elimination rate constants determined by linear regression analysis ranged from 0.06 to 0.11/d across the 13C-labelled congeners and sediment types. Example elimination profiles for 13C-PCBs 52 and 180 are provided in Figure 1. The magnitude of loss of individual congeners was independent of sediment type as determined by ANCOVAs performed on congener-specific linear regressions of the relationships between 13C-PCB concentrations in worms and time across the sediment treatments. In all cases, there were no significant differences between the observed congener-specific ktot values from treatment worms and sediment types ( p ¼ 0.3). Whole-body elimination rate constants (ktot) for the 13 C-PCB congeners were negatively correlated with congener log KOW (Fig. 2). Combining the data across congeners and sediment types indicated a highly significant (p < 0.001) relationship between log ktot and log KOW, yielding the following predictive equation. Log ktot ¼ 0:09 0:02 log KOW 0:45 0:14; R2 ¼ 0:49; n ¼ 21

Fig. 1. Elimination of 13C-PCB 52 (A) and 13C-PCB 180 (B) from Nereis virens depurated in Hudson River (HR; n ¼ 16; circles), New York Bight (NY; n ¼ 14; squares), and Sequim Bay (SB; n ¼ 13; lozenges) sediments. Solid, dashed, and dotted lines represent the best fit linear regressions for the HR, NY, and SB sediment treatments, respectively.

dosed organisms demonstrated uptake of environmental PCBs. At the end of the 28-d bioassay, control and treatment worms exposed to HR sediments had SPCB concentrations (unlabelled congeners) of 123.4 16.5 and 122.2 35.0 ng/g (wet wt), respectively. In both cases, bioaccumulated residues at the end of the 28-d bioassay period were significantly ( p < 0.001) higher than day 0 residues. There were no significant differences ( p ¼ 0.958) between bioaccumulated environmental PCB residues in treatment versus nondosed control worms after the 28-d HR sediment bioassay.

(8)

Bioaccumulation of sediment-associated PCBs

Treatment worms added to NY and SB sediments did not demonstrate any significant bioaccumulation of sediment-associated PCBs during the bioassay test. The SPCBs averaged 15.2 1.0 ng/g (wet wt) in worms at the beginning of the bioassay and averaged 16.0 0.6 and 9.0 2.1 ng/g in worms after 28 d of exposure to NY and SB sediments, respectively. For the HR sediments, bioassays were completed using both the control and the 13C-PCB-dosed experimental worms. In the case of this more contaminated sediment, treatment and non-

Fig. 2. Elimination rate constants (ktot) of 13C-PCB congeners from Nereis virens depurated in Hudson River (HR; circles), New York Bight (NY; squares), and Sequim Bay (SB; lozenges) sediments. Solid line refers to the linear regression fit to the combined elimination data. Dashed lines indicate 99% confidence limits for pooled regression.

1370


E.R. Bennett et al.

The log ktot versus log KOW relationship derived for 13CPCBs described in Equation 8 was used to estimate ktot values for individual environmental PCBs using congener-specific data on chemical log KOW [24,27]. These data were combined with Equations 1 and 2 to estimate the fraction of initial day 0 chemical residues remaining and the steady-state correction factor for each of the environmental PCBs (Table 1). For all congeners, most of the initial day 0 residues would have been expected to be lost from worms during the 28-d bioassay test. For trichlorobiphenyls, > 95% of initial day 0 residues were predicted to be lost. For decachlorobiphenyl, 84% of initial day 0 residues would be lost over the bioassay period. Similarly, most PCB congeners were predicted to approach steady state with the sediments over the 28-d study duration. Steady-state correction factors remained low and ranged from 1.04 (PCB 17) to 1.19 (PCB 209). Times to 90% steady state (t90) were estimated to range from 19 to 35 d for the different environmental PCB congeners (Table 1). The rapid loss of initial PCB contamination by worms and their approach to steady state with in-place sediment contaminants by this species indicate that neither control correction or the application of steady-state adjustment factors would have substantially changed bioassay conclusions and the magnitude of BSAF estimates compared with raw data. Table 1 summarizes the model-adjusted (control and steady-state adjustment factors) steady-state concentrations of individual environmental PCBs as well as mean 28-d measured congener-specific residues. For the HR sediments, the model-adjusted sum PCB concentrations were 112 ng/g relative to the measured residue value of 103.9 ng/g. The difference between the raw measured

and model-corrected value was within the analytical error range of samples measured after 28 d. The bioaccumulation model outlined in Equation 6, solved for COrg(t), was further used to predict bioaccumulation curves for PCBs 49 and 180 in N. virens exposed to the HR sediment (Fig. 3). The model was parameterized using Equation 8 to estimate ktot for sediment-associated PCBs and assuming that COrg(SS) corresponds to the last sampling time point of the bioassay. The latter assumption was considered valid because the bioassay duration approached or exceeded the estimated t90 times predicted for the congeners of study (Table 1). As a validation exercise, the model was also applied to a longer-term sediment bioassay study completed with N. virens and two types of sediments collected from New York Harbor, including the Hudson River materials sampled for the present study [28]. Model simulations were successful in predicting PCB congener bioaccumulation trends observed for worms across the range of experimental data, including the independent data set. The bioaccumulation model–predicted results generally provided good fit to the range of experimental data, including the independent data set. For the 28-d HR sediment bioassay completed in the present study, steady-state concentrations predicted by the bioaccumulation model were within 11% of concentrations measured in the worms by the conclusion of the bioassay. Model-predicted results for the long-term 60-d HR sediment bioassay of Kennedy et al. [28] were within 2.5% of the concentrations measured in worms at the conclusion of the bioassay. For the Arthur Kill sediment bioassays of Kennedy et al. [28], the bioaccumulation model overpredicted PCB 49 concentrations relative to those measured in worms after 60 d of

Table 1. Polychlorinated biphenyl bioaccumulation (PCB) and toxicokinetic data for Nereis virens exposed to Hudson River sediments during the 28-d bioassay PCB Sediment concn. Mean day 0 concn. congener Log KOWa (ng/g dry wt) (ng/g wet wt)b 17 31/28 33 52 49 44 74 70 95 101 99 87 110 118 105 151 149 153/132 138 187 180 170 194 209 77 a

5.25 5.67 5.60 5.84 5.85 5.75 6.20 6.20 6.13 6.38 6.39 6.29 6.48 6.74 6.65 6.64 6.67 6.92 6.83 7.17 7.36 7.27 7.80 8.18 6.86

33.0 172.6 18.7 67.6 43.5 43.6 22.1 47.0 31.1 44.2 21.4 15.3 43.9 36.8 12.9 7.4 29.7 44.3 38.9 13.5 21.6 10.3 6.0 11.0 4.0

0.06 0.02 0.30 0.05 ND 0.55 0.12 0.13 0.02 0.12 0.03 ND 0.08 0.02 0.68 0.13 0.75 0.17 0.39 0.08 0.11 0.03 0.33 0.08 0.34 0.07 0.22 0.05 0.18 0.06 1.18 0.35 3.05 0.59 2.12 0.45 1.09 0.25 1.20 0.26 0.48 0.14 0.21 0.03 0.30 0.07 0.02 0.01

Mean day 28 concn. (ng/g wet wt)b 2.23 0.23 10.10 5.03 0.44 0.04 7.57 2.11 5.76 1.30 5.02 2.38 1.97 0.87 1.79 1.22 9.11 1.15 8.32 1.62 3.54 0.73 1.15 0.38 5.84 1.29 3.82 1.08 1.91 0.24 1.29 0.28 6.27 1.28 9.58 1.50 8.46 1.47 2.81 0.46 3.69 0.66 1.63 0.34 0.54 0.11 0.47 0.05 0.50 0.11

ktot (/d)c t90 (d)d 0.12 0.11 0.11 0.11 0.11 0.11 0.10 0.10 0.10 0.09 0.09 0.10 0.09 0.09 0.09 0.09 0.09 0.09 0.09 0.08 0.08 0.08 0.07 0.07 0.09

19.2 21.0 20.7 21.7 21.7 21.3 23.4 23.4 23.0 24.2 24.3 23.8 24.7 26.1 25.6 25.6 25.7 27.1 26.6 28.5 29.6 29.1 32.4 35.1 26.7

Percentage SS adjustment remaininge factorf 3.50 4.61 4.41 5.12 5.16 4.85 6.33 6.33 6.08 7.00 7.03 6.66 7.38 8.45 8.07 8.03 8.16 9.25 8.85 10.41 11.35 10.90 13.70 15.90 8.98

1.04 1.05 1.05 1.05 1.05 1.05 1.07 1.07 1.06 1.08 1.08 1.07 1.08 1.09 1.09 1.09 1.09 1.10 1.10 1.12 1.13 1.12 1.16 1.19 1.10

Log KOW values from Hawker and Connell [27]. Day 0 and day 28 concentrations indicate those measured in worms at beginning and conclusion of the HR sediment bioassay. PCB congener whole-body elimination constant (ktot) values predicted from regression log ktot ¼ 0.09 log KOW 0.45. d Times to achieve 90% of steady-state concentration (t90) calculated from ln(10)/ktot e Percentage of initial PCB burden predicted to be remaining in animal after the 28-d bioassay period using Equation (1). f Steady-state (SS) adjustment factor calculated as per Equation (2). b c

Control corrected SS concn. (ng/g wet wt) 2.31 0.24 10.59 5.28 0.46 0.04 7.98 2.32 6.07 1.36 5.28 2.50 2.10 0.93 1.91 1.31 9.70 1.22 8.95 1.75 3.81 0.79 1.23 0.40 6.31 1.39 4.17 1.17 2.08 0.26 1.40 0.31 6.83 1.39 10.56 1.65 9.28 1.62 3.14 0.52 4.16 0.74 1.83 0.39 0.63 0.13 0.56 0.05 0.55 0.11



1371

Fig. 3. Uptake of PCB congeners 49 and 180 by Nereis virens from Hudson River sediments from the present study (A,B), Hudson River sediments from a longterm bioaccumulation study [28] (C,D), and Arthur Kill sediments from a long-term bioaccumulation study [28] (E,F). Solid, dashed, and dotted lines in each panel represent the bioaccumulation model fit derived using the general whole-body elimination constant (ktot) versus log KOW relationships for 13C-PCBs derived from the Hudson River, New York Bight, and Sequim Bay exposures, respectively. Model simulations were completed assuming steady state; sediment concentrations were achieved in worms after 28 d (the present study) or 60 d [28].

exposure. However, PCB 49 concentrations measured in N. virens at the termination of the Arthur Kill bioassay had declined relative to preceding sampling point. For PCB 180, the model-predicted steady-state concentrations were between 1.13 and 1.15 ng/g relative to an average concentration of 1.16 ng/g measured in worms after 60 d of exposure to the Arthur Kill material. Biota-sediment accumulation factors were calculated for bioassay worms exposed to the contaminated HR sediments

and are summarized in Figure 4. The BSAF values were calculated using steady-state adjusted polychaete PCB concentrations. Error bars were estimated by error propagation using Monte Carlo statistics as described above under Data analysis and bioaccumulation model. The BSAF curve demonstrated a typical curvilinear relationship, with linear increases in BSAF predicted for congeners having log KOW between 5.16 and 7 and decreased BSAF’s predicted for chemicals having log KOW values > 7.

1372


Fig. 4. Biota-sediment accumulation factors (BSAF) of PCB congeners in Nereis virens following 28 d of sediment bioaccumulation bioassay tests in Hudson River sediments. Error bars are standard deviation of the ratio estimated by using Monte Carlo analysis. Horizontal dashed line represents 1:1 ratio for BSAF values.

DISCUSSION

The present study utilized a pulse-chase experimental design coupled with performance reference compounds to probe pollutant bioaccumulation kinetics during a standard sediment bioassay procedure. Specifically, this design provided the ability to quantify the uptake of sediment-associated persistent organic pollutants while concomitantly investigating elimination kinetics of labeled reference compounds during the implementation of the bioassay. This technique provides in situ determination of the whole-body elimination rate constants (ktot) required to apply sediment- and chemical-specific control and steady-state correction factors. Control correction of data is necessary for bioassay studies [1–3], especially for investigation of ubiquitously distributed pollutants such as PCBs, for which truly clean animals may not be readily available [1–3]. The need for control correction becomes more important when bioaccumulated residues at the end of the recommended 28-d exposure periods are used for regulatory decision making and are intended to reflect natural conditions. Under such conditions, control correction completed by the subtraction of day 0 concentrations can lead to overly conservative bioaccumulation estimates. The rapid rate of 13C-PCB elimination observed in this study demonstrates the potential to optimize pre-exposure conditions prior to commencing standardized sediment bioaccumulation assays using N. virens such that control correction factors may not be required. From the relationship determined between ktot and log KOW from the elimination of 13C-PCBs by N. virens, elimination of > 80% of background-assimilated PCBs in worms can be completed with pre-exposure clearance periods extending up to 28 d. Significantly, these results demonstrate that such a level of clearance could be achieved for even the most hydrophobic PCB congeners using a 28-d clearance period prior to implementing a bioassay test. Such clearance periods would require clean sediments with animals fed an uncontaminated diet. Regardless, the determination of kinetic parameters such as whole-body elimination rate constants using in situ reference compounds such as 13C-PCBs and test species including N. virens provides an inherent benefit for bioassay procedures. With the performance reference compound and pulsechase approach, control correction with Equation 6 can be

E.R. Bennett et al.

performed without the requirement for a prebioassay clearance period. The only constraint to this correction procedure is that the depuration of the initial chemical burden by the bioassay organism does not substantially alter the mass balance of contaminant in the experimental system. Calculating the potential of animals to achieve the steadystate condition is also critical to determine whether exposure periods are sufficient to characterize the maximum bioaccumulation potential of sediment-associated contaminants in the bioassay organism [1–3]. The results of the present study demonstrated that N. virens would approach 90% of steady state with sediment concentrations for most PCB congeners during the 28-d bioassay period. Other studies investigating PCB bioaccumulation by N. virens have reported times to steady state ranging between 70 and 120 d for sediment bioassay studies [21,29]. However, these studies used worms with a lipid content different from that of the worms used in the present study and animals that had been reared at different temperatures. Environmental and biological factors including temperature and lipid content play critical roles in pollutant toxicokinetics and bioaccumulation potentials during sediment bioassays [30,31]. Specifically, the mass of lipid in an animal is significant for determining the magnitude of steady-state bioaccumulation that can be achieved [32], and temperature has been demonstrated to affect significantly toxicokinetic parameters such as whole-body elimination rates in poikilothermic species [20,33,34]. Polychlorinated biphenyls elimination rate constants have also been demonstrated to differ between Neries spp. populations [35,36]. The differences between the results of the present study and the studies of Pruell et al. [21] and McLeese et al. [29] likely reflect a combination of such experimental conditions and organism body condition. A variety of studies tested the bioavailability and toxicity of sediment-associated PCBs to both marine and freshwater invertebrate species. Using contaminated Hudson River estuary sediments, Meador et al. [37] concluded that a period of 10 d was sufficient for the marine polychaete Armandia brevis to reach steady state with sediment-associated PCB congeners. The marine bivalve Macoma nusata has been demonstrated to reach steady state with sediment-associated PCB concentrations within a time period of approximately 40 d [25,38,39]. For freshwater mayfly nymphs (Hexagenia limbata), Drouillard et al. [40] demonstrated that this species can achieve 95% of the steady-state concentration for Detroit River sediment PCB concentrations within a 32-d exposure period. Similarly to the present study, mayflies were predicted to achieve steady state for PCB congeners ranging in log KOW from 5.4 to 7.4. Landrum et al. [41] demonstrated that freshwater amphipods (Diporeia spp.) were approaching the steady-state condition for tetra- and hexachlorobiphenyl PCB congeners after 23 d of exposure to spiked Lake Michigan sediments. The 28-d time frame predicted for N. virens to approach steady state with sediment-associated PCB concentrations in the present study is consistent with the range observed for many other common benthic invertebrate bioassay species. However, in addition to factors such as temperature and lipid content, other factors such as organism feeding behavior, sediment to water desorption kinetics, particle size distribution, organic matter type, and presence of geosorbants such as black carbon can play important roles in bioaccumulation kinetics and bioaccumulation potential [42–44]. The effects of these different factors on the physiological performance of the test species can be potentially tracked using the performance reference compound and pulse-chase experimental design applied in the present study.


Biota sediment accumulation factors determined for PCBs in N. virens from the present study ranged from 0.2 to 3.0. These results agree well with other polychaete sediment bioassays in which BSAFs ranged from approximately 0.1 to 5.7 for Aroclor mixtures and individual PCB congeners [21,29,37,45,46]. The BSAF results of the present study also demonstrated a hydrophobicity pattern consistent with the patterns determined by Pruell et al. [21] and Meador et al. [37]. Increases in BSAF with chemical hydrophobicity occurred for congeners having log KOW between 6.4 and 7.0. For highly hydrophobic congeners with log KOW > 7.0, BSAFs were observed to decrease with increasing chemical hydrophobicity. Lake et al. [42] observed a similar range for field-collected species, in which BSAFs for PCB congeners of log KOW > 7.0 were lower than those for less hydrophobic ones. The relationship between PCB BSAFs and log KOW observed in the present study confirms prior conclusions that highly chlorinated and hydrophobic compounds are not as effectively bioaccumulated during sediment exposures as are more water-soluble compounds [21,38,42,47] The results of the present study demonstrate that N. virens can reach steady state with sediment-associated PCB congeners within the recommended 28-d bioassay exposure period. Kennedy et al. [28] demonstrated that this species reached 80% of the steady-state condition with PCB concentrations in sediments collected from multiple locations within New York Harbor within the same time period. The bioaccumulation model from the present study was able to predict PCB congener bioaccumulation profiles accurately for worms from the experiments of Kennedy et al. [28]. Significantly, these simulations were completed using the relationship between ktot and log KOW determined from the elimination of the 13C-PCBs by the worms used in the present study. The elimination rates determined for PCBs by Kennedy et al. [28] were somewhat slower than those calculated in the present study. However, Kennedy et al. used nonlinear regression methods to deduce elimination rates by fitting the bioaccumulation curve to a one-compartment model similar to that described in Equation 3. The direct measurement of elimination rate constants as determined in the present study is considered a more accurate method of determining ktot. The use of performance reference compounds such as isotopically labeled or environmentally rare PCB congeners has been implemented in experiments designed for the calibration of species toxicokinetics [48,49]. An advantage of this technique is that it provides the ability to calibrate in situ toxicokinetics under experimental exposure conditions operating within the bioassay. In such a manner, the influence of bioassay conditions on organism physiology such as food availability and quality, temperature, and toxicant–organism interactions can be evaluated with the performance reference compounds and pulse-chase experimental design [48]. The use of performance reference compounds at the dosing concentrations used in the present study appeared to have little impact on the physiological performance of the bioassay organism. No differences in the uptake of HR sediment-associated PCBs had been observed between worms that had been dosed with 13C-PCBs relative to nondosed control worms. Combined with the lack of mortality observed in this study, these results indicate that it is unlikely that the bioassay sediments used or the 13C-PCB dose treatment altered animal toxicokinetics during the bioassay. Whole-body elimination rate constants represent the combined sum of pollutant elimination via metabolic transformation, waste egestion, losses across respiratory surfaces, and growth dilution [50,51]. Thus, in addition to quantifying critical toxicokinetic parameters such as ktot, toxicokinetic tracers such as

Environ. Toxicol. Chem. 30, 2011 13

1373

C-PCBs can be considered metabolic reference compounds that provide indices of physiological performance in situ when used in conjunction with routine bioaccumulation experiments. Although the pulse-chase design does entail greater costs and time investment for dosing of animals, the routine implementation of such methods could be modified to minimize costs and enhance regulatory test protocols. For example, rather than sampling at multiple time points during exposure, the loss of the reference compounds by study completion relative to day 0 can provide a simple calibration check to ensure that in situ toxicokinetics during the bioassay period did not deviate significantly from control exposures. The validity of such a calibration increases with the number of reference compounds utilized, especially when reference compounds are available across a wide range of chemical hydrophobicities that could be used to track metabolic functioning in the test species [17,18,48,49,52]. From a regulatory perspective, this technique provides a valuable quality assurance procedure for interpreting sediment bioassays. Exposure of invertebrate species to highly contaminated sediments has been demonstrated to impact feeding and respiratory activities negatively [53]. For depositfeeding species such as N. virens, such changes in animal bioenergetics could result in low bioaccumulation rates, leading to underestimates of the true bioavailability of sediment-associated contaminants. Thus, the utilization of performance reference compounds in a pulse-chase design immediately flags deviations in the toxicokinetic performance of the test species and performance attributes of a given bioassay test, which greatly aids in the risk assessment of contaminated sediments and dredged materials. The approach also has some limitations when extended to predict BSAFs outside of the empirically derived bioassay value. The pulse-chase design chiefly measures the physiological performance of the test organism but does not track sediment–toxicant associations and desorption kinetics that contribute to differences in bioavailability between sediment types. Geosorbants such as black carbon strongly influence contaminant bioavailability and the magnitude of the BSAF [44,47,54]. If geosorbed contaminants remain essentially unavailable to the bioassay organism, then differences in geosorbant content and desorption kinetics between individual sediment components will be expected to influence the magnitude of the BSAF but should not strongly affect the time required to reach steady state. Thus, the performance reference compound approach will still provide value in establishing appropriate steady-state and control correction factors to facilitate an appropriate empirical BSAF measurement. The method will not, however, provide a mechanistic determination of why contaminant bioavailability differs between sediment types unless additional characterization of sediment components and the degree of contaminant sequestration to these individual sediment components is performed [47]. Alternatively, if a portion of the geosorbant-associated contaminant is bioavailable and this fraction has a lower uptake rate coefficient relative to an organic matter–associated chemical, then hysteresis in contaminant uptake and elimination may occur. Under the latter condition, a multiphasic approach to steady state would necessitate implementing a more detailed, temporal sampling strategy and an appropriately designed bioaccumulation model to reflect steady-state correction factors more accurately. However, these artifacts are expected to be small when the bioavailability of geosorbant-associated contaminant remains low, and the life span of the organism is short relative to geosorbant– water desorption kinetics.

1374


In conclusion, the performance reference compound approach in conjunction with a pulse-chase experimental design can provide novel information about the physiological performance of test organisms during standardized sediment bioaccumulation bioassays. The design also provides additional information used to evaluate the need for quantification of toxicokinetic parameters necessary to apply control and steady-state correction factors. Future studies should be performed to apply the method in a bioassay species to determine variations in ktot across a broader range of sediment types than explored in the present study. For example, the use of highly toxic sediments in conjunction with performance reference compounds would demonstrate whether or not toxicity experienced by the bioassay organism influences its ability to achieve steady state within the time frame of the bioassay study. Finally, using the performance reference compound and pulse-chase design in conjunction with sediments that vary in geosorbant content or characteristics could provide useful information to determine whether geosorbants not only affect the magnitude of contaminant BSAF but also alter contaminant toxicokinetics and approach to steady state. Acknowledgement—The present study was supported in part by the U.S. Army Corps of Engineers (USACE) New York District (Monte Greges) and USACE Engineer Research and Development Center Dredging Operations and Environmental Research (Todd Bridges, Program Manager). The authors acknowledge Jerre Sims for laboratory support and David Qiu for analytical support. Permission was granted by the Chief of Engineers to present these research results. REFERENCES 1. U.S. Environmental Protection Agency/Army Corps of Engineers. 1991. Evaluation of Dredged Material Proposed for Ocean Disposal Ocean Testing Manual. EPA/503/8-91/001. Washington, DC. 2. U.S. Environmental Protection Agency/Army Corps of Engineers. 1998. Evaluation of Material Proposed for Discharge to Waters of the U.S.— Testing Manual. Inland Testing Manual. EPA/823/B-98/004. Washington, DC. 3. American Society for Testing and Materials. 1998. Standard guide for determination of the bioaccumulation of sediment-associated contaminants by benthic invertebrates. E1688-00a. West Conshohocken, PA. 4. Steevens JA, Reiss M, Pawlisz AV. 2005. A methodology for deriving tissue residue benchmarks for fish exposed to 2,3,7,8-tetrachlorodienzop-dioxin. Integr Environ Assess Manag 1:142–151. 5. McElroy AE, Barron MG, Becvar N, Kane Driscoll SB, Meador JP, Parkerton TF, Preuss TG, Steevens JA. 2010. A review of the tissue residue approach for organic and organometallic compounds in aquatic organisms. Integr Environ Assess Manag 7:50–74. 6. Sappington KG, Bridges TS, Bradbury SP, Erickson RJ, Hendriks AJ, Lanno RP, Meador JP, Mount DR, Salazar MH, Spry DJ. 2011. Application of the tissue residue approach in ecological risk assessment. Integr Environ Assess Manag 7:116–140. 7. Landrum PF, Lee H, Lydy MJ. 1992. Toxicokinetics in aquatic systems—Model comparisons and use in hazard assessment. Environ Toxicol Chem 11:1709–1725. 8. Ingersoll CG, Ankley GT, Benoit DA, Brunson EL, Burton GA, Dwyer FJ, Hoke RA, Landrum PF, Norberking TJ, Winger PV. 1995. Toxicity and bioaccumulation of sediment associated contaminants using freshwater invertebrates—A review of methods and applications. Environ Toxicol Chem 14:1885–1894. 9. Bridges TS, Wright RB, Bray BR, Gibson AB, Dillon TB. 1996. Chronic toxicity of Great Lakes sediments to Daphnia magna: Elutriate effects on survival, reproduction and population growth. Ecotoxicology 5:83– 102. 10. Harkey GA, Landrum PF, Klaine SJ. 1994. Comparison of whole sediment elutriate and pore water exposures for use in assessing sediment associated organic contaminants in bioassays. Environ Toxicol Chem 13:1315–1329. 11. Bruner KA, Fisher SW. 1993. The effects of temperature, pH, and sediment on the fate and toxicity of 1-naphthol to the midge larvae Chironomus riparus. J Environ Sci Health A28(6):1341–1360.

E.R. Bennett et al. 12. Airas S, Leppanen M, Kukkonen JVK. 2008. Effects of temperature and oxygen concentration in sediment toxicity testing. Ecotoxicol Environ Saf 70:475–482. 13. Lyytikaïnen M, Pehkonen S, Akkanen J, Leppa¨nen M, Kukkonen JVK. 2007. Bioaccumulation and biotransformation of polycyclic aromatic hydrocarbons during sediment tests with oligochaetes (Lumbriculus variegatus). Environ Toxicol Chem 26:2660–2666. 14. Driscoll SK, Harkey GA, Landrum PF. 1997. Accumulation and toxicity of fluoranthene in sediment bioassays with freshwater amphipods. Environ Toxicol Chem 16:742–753. 15. Driscoll SK, Landrum PF. 1997. A comparison of equilibrium partitioning and critical body residue approaches for predicting toxicity of sediment-associated fluoranthene to freshwater amphipods. Environ Toxicol Chem 16:2179–2186. 16. Kukkonen JVK, Landrum PF. 1994. Toxicokinetics and toxicity of sediment associated pyrene to Lumbriculus variegatus (Oligochaeta). Environ Toxicol Chem 13:1457–1468. 17. Gewurtz SB, Drouillard KG, Lazar R, Haffner GD. 2002. Quantitative monitoring of PAHs using the Barnes mussel (Elliptio complanata). Arch Environ Contam Toxicol 43:497–504. 18. O’Rourke S, Drouillard KG, Haffner GD. 2004. Determination of laboratory and field elimination rates of polychlorinated biphenyls (PCBs) in the freshwater mussel, Elliptio complanata. Arch Environ Contam Toxicol 47:74–83. 19. Drouillard KG, Chan S, O’Rourke S, Haffner GD, Letcher RJ. 2007. Elimination of 10 polybrominated diphenyl ether (PBDE) congeners and selected polychlorinated biphenyls (PCBs) from the freshwater mussel, Elliptio complanata. Chemosphere 69:362–370. 20. Paterson G, Drouillard KG, Haffner GD. 2007. PCB elimination by yellow perch (Perca flavescens) during an annual temperature cycle. Environ Sci Technol 41:824–829. 21. Pruell RJ, Rubinstein NI, Taplin BK, Livolsi JA, Bowen RD. 1993. Accumulation of polychlorinated organic contaminants from sediment by 3 benthic marine species. Arch Environ Contam Toxicol 24:290–297. 22. Mackay D. 1982. Correlation of bioconcentration factors. Environ Sci Technol 16:274–278. 23. Ramsdall KA, Strand JA, Cullinan VI. 1989. Amphipod bioassay of selected sediments from Sequim Bay, Washington. Proceedings, Oceans 89 Conference, Seattle, WA, USA, September 18–21, pp 443–448. 24. Steevens J, Kennedy A, Farrar D, McNemar C, Reiss M, Kropp R, Doi J, Bridges T. 2008. Dredged Material Analysis Tools: Performance of Acute and Chronic Sediment Toxicity Methods. U.S. Army Engineer Research and Development Center Report ERDC/EL TR- 08-16. Vicksburg, MS. 25. Boese BL, Winsor M, Lee H, Echols S, Pelletier J, Randall R. 1995. PCB congeners and hexachlorobenzene biota sediment accumulation factors for Macoma nasuta exposed to sediments with different total organic carbon contents. Environ Toxicol Chem 14:303–310. 26. Lazar R, Edwards RC, Metcalfe CD, Gobas FAPC, Haffner GD. 1992. A simple, novel method for the quantitative analysis of coplanar (non-ortho substituted) polychlorinated biphenyls in environmental samples. Chemosphere 25:493–504. 27. Hawker DW, Connell DW. 1988. Octanol–water partition coefficients of polychlorinated biphenyl congeners. Environ Sci Technol 22:382–387. 28. Kennedy AJ, Lotufo GA, Steevens JA, Bridges TS. 2010. Determining steady state tissue residues for invertebrates in contaminated sediments. U.S. Army Corps of Engineers Report ER/EL TR- 10-2. Vicksburg, MS. 29. McLeese DW, Metcalfe CD, Pezzack DS. 1980. Uptake of PCBs from sediment by Nereis virens and Crangon septemspinosa. Arch Environ Contam Toxicol 9:507–518. 30. Landrum PF. 1988. Toxicokinetics of organic xenobiotics in the amphipod, Pontoporeia hoyi: Role of physiological and environmental variables. Aquat Toxicol 12:245–271. 31. Landrum PF, Tigue EA, Kane Driscoll S, Gossiaux DC, Van Hoof PL, Gedeon ML, Adler M. 2001. Bioaccumulation of PCB congeners by Diporeia spp.: Kinetics and factors affecting bioavailability. J Gt Lakes Res 27:117–133. 32. Mackay D, Paterson S. 1981. Calculating fugacity. Environ Sci Technol 15:1006–1014. 33. Goerke H. 1984. Temperature-dependent elimination of 2,4,6,20 ,40 pentachloro[U-14C]biphenyl in Nereis virens (Polychaeta). Arch Environ Contam Toxicol 13:347–355. 34. Buckman AH, Brown SB, Hoekstra PK, Solomon KR, Fisk AT. 2004. Toxicokinetics of three polychlorinated biphenyl technical mixtures in rainbow trout (Oncorhynchus mykiss). Environ Toxicol Chem 23:1725– 1736.

Toxicokinetics of PCB congeners in Nereis virens 35. Goerke H, Weber K. 1990. Population dependent elimination of various polychlorinated biphenyls in Neries diversicolor (Polychaeta). Mar Environ Res 29:205–226. 36. Goerke H, Weber K. 2001. Species-specific elimination of polychlorinated biphenyls in estuarine animals and its impact on residue patterns. Mar Environ Res 51:131–149. 37. Meador JP, Adams NG, Casillas E, Bolton JL. 1997. Comparative bioaccumulation of chlorinated hydrocarbons from sediment by two infaunal invertebrates. Arch Environ Contam Toxicol 33:388–400. 38. Ferraro SP, Lee H, Ozretich RJ, Specht DT. 1990. Predicting bioaccumulation potential: a test of a fugacity based model. Arch Environ Contam Toxicol 19:386–394. 39. Boese BL, Lee H, Echols S. 1997. Evaluation of a first-order model for the prediction of the bioaccumulation of PCBs and DDT from sediment into the marine deposit-feeding clam Macoma nasuta. Environ Toxicol Chem 16:1545–1553. 40. Drouillard KG, Ciborowski JJH, Lazar R, Haffner GD. 1996. Estimation of the uptake of organochlorines by the mayfly Hexagenia limbata (Ephemeroptera: Ephmeridae). J Gt Lakes Res 22:26–35. 41. Landrum PF, Dupuis WS, Kukkonen J. 1994. Toxicokinetics and toxicity of sediment-associated pyrene and phenanthrene in Diporeia spp.: Examination of equilibrium partitioning theory and residue-based effects for assessing hazard. Environ Toxicol Chem 13:1769–1780. 42. Lake JL, Rubinstein NI, Lee H, Lake CA, Heltshe J, Pavignano S. 1991. Equilibrium partitioning and bioaccumulation of sediment-associated contaminants by infaunal organisms. Environ Toxicol Chem 9:1095– 1106. 43. Landrum PF, Lee H, Lydy MJ. 1992. Toxicokinetics in aquatic systems: Model comparisons and use in hazard assessment. Environ Toxicol Chem 11:1709–1725. 44. Moermond CTA, Zwolsman JJG, Koelmans AA. 2005. Black carbon and ecological factors affect in situ biota sediment accumulation factors for hydrophobic organic compounds in floodplain lakes. Environ Sci Technol 39:3101–3109. 45. Pruell RJ, Taplin BK, McGovern DG, McKinney R, Norton SB. 2000. Organic contaminant distributions in sediments, polychaetes (Nereis


46. 47. 48.

49.

50. 51. 52.

53.

54.

1375

virens) and American lobster (Homarus americanus) from a laboratory food chain experiment. Mar Environ Res 49:19–36. Ruus A, Schaanning M, Øxnevad S, Hylland K. 2005. Experimental results on bioaccumulation of metals and organic contaminants from marine sediments. Aquat Toxicol 72:273–292. Koelmans AA, Jonker MTO, Cornelissen G, Bucheli TD, Van Noort PCM, Gustafsson O. 2005. Black carbon, the reverse of its dark side. Chemosphere 63:365–377. Raeside AA, O’Rourke SM, Drouillard KG. 2009. Determination of in situ polychlorinated biphenyl elimination rate coefficients in the freshwater mussel biomonitor Elliptio complanata deployed in the Huron–Erie corridor, southeast Michigan, USA, and southwest Ontario, Canada. Environ Toxicol Chem 28:434–445. Drouillard KG, Norstrom RJ. 2001. Quantifying maternal and dietary sources of 2,2,4,40 ,5,50 -hexachlorobiphenyl deposited in eggs of ring doves (Streptopelia risoria). Environ Toxicol Chem 20:561– 567. Connolly JP, Pedersen CJ. 1988. A thermodynamic-based evaluation of organic chemical accumulation in aquatic organisms. Environ Sci Technol 22:99–103. Clark KE, Gobas FAPC, Mackay D. 1990. Model of organic chemical uptake and clearance by fish from food and water. Environ Sci Technol 24:1203–1213. Paterson G, Huestis SY, Whittle DM, Drouillard KG, Haffner GD. 2005. In situ measurement of tissue turnover and energy conversion efficiencies in lake trout (Salvelinus namaycush) using a novel toxicokinetic approach. Can J Fish Aquat Sci 62:464–471. Toro B, Navarro JM, Palma-Fleming H. 2003. Relationship between bioenergetics responses and pollutants in the giant mussel, Choromytilus chorus (Mollusca: Mytilidae). Aquat Toxicol 63:257– 269. Cornelissen G, Breedveld GD, Naes K, Oen AMP, Ruus A. 2006. Bioaccumulation of native polycyclic aromatic hydrocarbons from sediment by a polychaete and a gastropod: Freely dissolved concentrations and activated carbon amendment. Environ Toxicol Chem 25:2349–2355.