Environ. Sci. Technol. 2007, 41, 1783-1789
Variation in the Bioaccumulation of a Sediment-Sorbed Hydrophobic Compound by Benthic Macroinvertebrates: Patterns and Mechanisms PAUL N. GASKELL,* AMY C. BROOKS, AND LORRAINE MALTBY Department of Animal and Plant Sciences, The University of Sheffield, Western Bank, Sheffield, S10 2TN, United Kingdom
Aquatic ecological risk assessment is primarily focused on aqueous exposure, but many hydrophobic contaminants bind to particulate material and accumulate in sediments. The risk posed by such contaminants is partially dependent on the importance of dietary exposure. Here, we describe the bioaccumulation of a highly hydrophobic compound (dioctadecyl-dimethyl ammonium chloride (DODMAC)) to four freshwater macroinvertebrates (i.e., Asellus aquaticus, Chironomus riparius, Gammarus pulex, Lumbriculus variegatus) and investigate the mechanistic basis for observed interspecific variation in bioaccumulation. Although more than 99.99% of DODMAC was sediment-bound, it was bioavailable to all four species via dietary exposure. Interspecific variation in bioaccumulation was apparent despite the lack of selective feeding and biotransformation potential and after normalization for body size and lipid content. Chironomus riparius had the highest lipid-normalized DODMAC concentration and L. variegatus had the lowest. Study species differed in factors affecting uptake (i.e., feeding rate) and absorption efficiency (i.e., gut passage time and gut surfactancy). Feeding rate did not explain interspecific variation in bioaccumulation, but bioaccumulation was enhanced by either high surfactancy and short gut passage time (e.g., G. pulex) or low surfactancy and long gut passage time (e.g., C. riparius). Risk assessment of hydrophobic contaminants should consider dietary exposure and the potential food chain effects of interspecific variation in bioaccumulation.
Introduction Many environmental contaminants are poorly soluble in water and bind strongly to particulate material, resulting in sediment concentrations that are higher than aqueous concentrations (1). Although benthic macroinvertebates are exposed to sediment-associated contaminants, both in the particulate and dissolved (porewater) phases (2), most attention has focused on aqueous exposure leading to the assumption that hydrophobic chemicals that are tightly bound to particulates pose little risk to aquatic organisms (3). Sediment-bound contaminants could pose a risk if dietary exposure is an important route of uptake. Moreover, the magnitude of this risk may vary between species, sediment* Corresponding author phone: +44 114 222 0018; fax: +44 114 222 0002; e-mail:
[email protected]. 10.1021/es061934b CCC: $37.00 Published on Web 01/30/2007
2007 American Chemical Society
bound contaminants being more bioavailable to some species than others (4-6). Interspecific differences in the bioaccumulation of sediment-bound contaminants may be due to differences in exposure (e.g., habitat, behavior (7)), uptake (e.g., feeding rate (8), gut passage time (9), desorption efficiency (10), absorption efficiency (11)), metabolism, and excretion (12). Differences in exposure may arise because of an uneven distribution of contaminants within sediments, with concentrations varying across particle sizes and types (1). In addition, some benthic organisms selectively feed on particular types of particles and some actively avoid contaminated sediments (8, 13). Both selective feeding and avoidance behavior have the potential to affect contaminant exposure and hence bioaccumulation. Once ingested, the efficiency with which aquatic organisms absorb contaminants ranges from zero to 98% and varies between contaminants, species, and measurement techniques (11), selective feeding partially explaining intertechnique variability (13). An important factor in determining the efficiency with which contaminants are absorbed is desorption efficiency. Sediment-bound chemicals must be desorbed from particles before they can be absorbed. This solubilization can occur either in the external medium (aqueous exposure) or in the animal’s digestive tract (dietary exposure). Incubations of gut fluids with contaminated sediments have demonstrated considerable interspecific variation in the ability of marine benthic macroinvertebrates to solubilize metals and organics (14, 15). Moreover, several studies have demonstrated a correlation between the solubilization of particle-bound contaminants and gut surfactancy (16, 17). Although monomolecular dispersions of surfactants reduce gut fluid surface tension, it is the formation of micelles that confers the greatest increase in contaminant solubilization (10, 16, 18). Absorption of dietary particlebound contaminants is also influenced by the length of time contaminated sediment remains in the gut. Positive correlations between absorption and gut passage time have been reported for organics (9) and metals (19). In addition to feeding behavior and gut physiology, variation in organism lipid content has the potential to influence bioaccumulation of organic compounds. Hydrophobic contaminants can accumulate in lipids and normalization for lipid content may reduce variability in reported bioaccumulation values (20). Here, we investigate the bioaccumulation of a highly hydrophobic compound to four freshwater benthic macroinvertebrates. We then investigate the mechanistic basis for interspecific variation in bioaccumulation, with a particular focus on gut physiology. The compound used was dioctadecyl-dimethyl ammonium chloride (DODMAC), a cationic surfactant and a major component of dihardened tallow dimethyl ammonium chloride (DHTDMAC), which has been used as a fabric softener (21). DODMAC was used as a model compound in this study because it is highly hydrophobic and adsorbs strongly to mineral and organic surfaces. Octanol/water partition coefficients (Kow) are widely used to describe the hydrophobicity/liphophilicity of organic compounds. However, “surface-active” compounds, such as DODMAC, interact with the interface between nonpolar (i.e., octanol) and polar (i.e., water) media thus precluding the generation of meaningful Kow values (22). It is possible to produce micellar dispersions of DODMAC in aqueous media under physical shear or temperatures > 47.5 °C (21) and, consequently, aqueous toxicity or solubility values greatly in excess of the true solubility limit are sometimes incorrectly reported (21, 23). However, it is safe to assert that in natural VOL. 41, NO. 5, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1783
ecosystems, DODMAC is present almost entirely bound to particulate matter and its most reliably determined aqueous solubility is less than 1 pg/L (21). The four species investigated were Gammarus pulex (Crustacea, Amphipoda), Asellus aquaticus (Crustacea, Isopoda), Chironomus riparius (Insecta, Diptera), and Lumbriculus variegatus (Annelida, Oligochaeta), and a simple clay sediment was used to minimize the effect of selective feeding on DODMAC bioaccumulation. None of the four species is capable of biotransforming DODMAC (23), hence, postexposure elimination could only proceed via loss of the parent compound. Further, comparisons of sediment ingesting and noningesting individuals within a single species indicate that bioaccumulation of DODMAC by two of the four benthic invertebrates (L. variegatus and A. aquaticus) is dependent on dietary exposure (23).
Materials and Methods Model Compound. The 14C DODMAC (>99% purity) used in this study was synthesized at Unilever Research Port Sunlight (Unilever Research, Bebbington, United Kingdom) and had a specific activity of 0.51 MBq/mg. Test Species. Lumbriculus variegatus and Chironomus riparius were cultured in the Department of Animal and Plant Sciences within The University of Sheffield using methods adapted from U.S. EPA (24) and Credland (25), respectively. Culture vessels were rectangular perspex tubs (30-cm long, 15-cm wide, and 10-cm deep) containing 1.5 L of aerated artificial pond water (APW) (26) over 2-cm depth of nonbleached shredded paper towel substrate (L. variegatus) or play pit sand (C. riparius). Lumbriculus cultures contained 300-500 individual worms, and chironomid cultures contained approximately 200 larvae. Cultures were maintained at 20 °C and a photoperiod of 16 h light to 8 h dark. Animals were fed either powdered catfish food (L. variegatus) or tropical fish flake (C. riparius) twice a week. Sufficient food (c. 150 and 180 mg, respectively) was added to support healthy individuals without clouding the overlying water. Asellus aquaticus (10-15 mm in length) and adult male Gammarus pulex (10-15 mm in length) were collected from field populations (national grid reference (NGR) SK 315 881 or NGR SK 323 888, and NGR SK 497 744, respectively) and then were maintained in 20-L aquaria of aerated APW at 20 °C ((2 °C). Animals were fed alder (Alnus glutinosa) leaf material; aquaria holding 300-400 individuals received 1015 leaves per week. Two days prior to use in studies, the feeding of both crustacean species was suspended by removing any leaf material remaining in the aquaria. Crustaceans were retained in laboratory aquaria for a maximum of two weeks prior to use in experimental procedures. Bioaccumulation. One hundred and sixty 200-mL glass test vessels, each containing 10 g dry weight of kaolin clay (mean (SD) particle size 8.08 (0.75) µm), were prepared. Confounding influences of toxicity on bioaccumulation rate were avoided by selecting a sediment loading approximately 200× lower than a toxicity threshold reported for a relevant organism. The lowest observed effect concentration of DODMAC on chironomid emergence is reported as 876 µg per g dry weight sediment (27), and it was this value that was used to establish the exposure concentration. Accordingly, 80 test vessels were dosed with a nominal concentration of 4.3 µg DODMAC per g dry weight of sediment. Two milliliters of a 21.8 mg/L stock solution of DODMAC dissolved in HPLC grade ethanol was added to each test vessel and was mixed thoroughly. The ethanol was then allowed to evaporate over 3 days, during which time the sediments were stirred thoroughly once a day to facilitate evaporation. Constant weight due to full evaporation of ethanol was observed within 60 h using this system. Sediment in each test vessel was 1784
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 5, 2007
moistened with 2 mL of APW and was gently tamped down to produce a cohesive pellet. To minimize sediment disturbance, a plastic disc was placed on the sediment surface prior to the addition of 100 mL of APW. Twenty dosed and 20 nondosed vessels were used for each of the four species. Each vessel contained either five G. pulex, A. aquaticus, or C. riparius larvae (>5-mm length, third/fourth instar) or 20 L. variegatus. All animals had been starved for 24 h at 20 °C before addition to test vessels. The organisms were exposed to sediment for a 48-h period to minimize the opportunity for desorption of compound from sediment particles into pore water. Exposure for 48 h enabled an examination of whether DODMAC bioaccumulation occurred but was not intended to result in “steady-state” tissue loadings. Following exposure, organisms were removed to gut purge vessels for 24 h. Gut purge vessels consisted of 200-mL glass vessels each with a 1.5-cm depth of sand and 2.5-cm depth of APW. After the 24-h gut purge, each organism was observed and the absence of clay within the gut was confirmed visually. The animal tissue collected from each gut purge vessel was rinsed, blotted, and then weighed prior to addition to scintillation vials for analysis. Exposure and gut purge were performed at 20 ( 2 °C. Pore waters were separated by centrifugation of the test vessels at 3000 rpm for 1 h. One-milliliter samples of overlying water, pore water, and 1-g sediment samples were taken at the end of the 48-h exposure period from both nondosed and dosed vessels. In all cases, the mean detected radioactivity in nondosed vessels was assigned to represent background radioactivity. Dosed artificial sediment analyzed at Unilever Research (Bebbington, United Kingdom) contained only parent 14C DODMAC (Stephen Harding, personal communication). Therefore, in the absence of intermediate breakdown molecules, dose vessel sample radioactivity values minus mean calculated background values were attributed to 14C DODMAC. The organisms recovered from each individual gut purge vessel comprised a single tissue sample. Tissues were solubilized for 3 days using 2 mL Soluene prior to addition of 10 mL Hionic Fluor scintillator fluid (Packard Bioscience B.V, Groningen, The Netherlands) (28, 29). The samples were left in darkness for a further 24 h to allow for attenuation of chemiluminescence before counting. Samples were counted using a Tricarb 3000 Liquid Scintillation Counter (Packard Bioscience B.V, Groningen, The Netherlands) with a single 14C counts window (0-156 KeV emission spectrum) and disintegrations per minute (dpm) recorded. The external standards method was used to correct for quench and each vial was counted for 2 min (28, 29). The activities for each sample were expressed as µg DODMAC/mL by dividing them by 29748 (dpm produced by 1 µg DODMAC) and subtracting the background value. Tissue and sediment sample concentrations were expressed as µg DODMAC/g fresh and dry weight, respectively. Methodological comparisons revealed that scintillation counting returned values that were 100-103% of the values determined via combustion analysis of aliquots from a single-spiked sediment batch (23). DODMAC tissue concentrations were normalized for lipid content using average lipid concentrations of 0.7, 1.1, 1.2, and 1.3% fresh mass for A. aquaticus, C. riparius, G. pulex, and L. variegatus, respectively (30-33). Gut Passage Time. Forty individuals of each species were allowed to ingest artificial sediment (70% sand:20% kaolin: 10% powdered cellulose) by placing them in individual 8-cm diameter test vessels, each containing a 2-cm depth of APW overlying a 2-cm depth of sediment. After approximately 2 h, organisms were removed to individual gut purge vessels and were left for 15-30 min. Upon re-examination, the time elapsed (minutes) and the distance (mm) between the trailing (most anterior) edge of the sediment pellet and the pharynx
were recorded to the nearest millimeter. The total length of the alimentary canal was also recorded for each organism. The time required for one complete gut passage was estimated by dividing the total length of the alimentary canal by the calculated pellet velocity (velocity ) distance moved by trailing edge of pellet/time). Depuration. Chironomus riparius, G. pulex, and A. aquaticus were exposed to DODMAC-spiked artificial sediment (4.3 µg/g) for 48 h, after which they were transferred to gut purge vessels. Nonmoulting individuals were removed from gut purge vessels throughout the following 10 days and were analyzed for DODMAC, as described above. The correlation between the logarithm of tissue concentration against depuration time was determined, and if statistically significant, the first-order rate constant for depuration was determined from the slope of this relationship, estimated by least-squares linear regression (34). Surfactancy. Guts were extracted from G. pulex, A. aquaticus, and C. riparius by removing 1-2 mm of the posterior end of the animal and then by pulling away the head capsule with the gut attached. The gut was separated from the head and, if necessary, the hepatopancreas, and was blotted dry before being placed in an Eppendorf tube. Great care was taken to exclude additional tissue or fluid from other internal structures during the separation of animal guts. Each tube contained 12, 30, or 20 guts from G. pulex, A. aquaticus, or C. riparius, respectively, and was frozen until required. Gut homogenate was prepared by defrosting guts, macerating them with mounted needles, and centrifuging them in the Eppendorf tubes for 10 min at 13 000 rpm (Micro Centaur, MSE (UK) Limited, Kent, United Kingdom). Since guts were removed from animals (and any attendant midgut glands detatched) before fluid extraction, dilution of gut homogenate by disrupted cell contents should not vary systematically between species. Gut homogenate surfactancy was measured using a modification of the method described by Ahrens et al. (17). One microliter of the resulting supernatant was dispensed as a single droplet onto a Parafilmcoated microscope slide. The droplet was photographed immediately with a digital camera (Nikon Coolpix 4300, Nikon Corporation, Japan) and the angle of the droplet to the surface of the slide (i.e., the contact angle) was measured using image analysis software (Image J: public domain software rsb.info.nih.gov/ij/). Both the left- and right-hand side of the droplet were measured and the average was taken. The total number of droplets measured for G. pulex, A. aquaticus, or C. riparius were 11, 8, and 9, respectively. The contact angles of 20 1-µL droplets of artificial pond water droplets were measured as a control. Micelle Formation. It is possible to test for the presence of surfactant micelles via serial dilution. At surfactant concentrations high enough to support micelle formation, surface tension is independent of dilution. Conversely, when dilution becomes too great, micelle formation is not supported (critical micelle dilution) and surface tension becomes proportional to dilution (18). Therefore, in addition to overall surfactancy, gut homogenates were investigated for the presence of micelles. Ten additional Eppendorf tubes of gut homogenates were obtained for each of the three species as described above. A commercially available surfactant, sodium dodecyl sulfate (SDS), was used to validate micelle-detection methodology. Thirty millimolar and 20 mM stock solutions of SDS in distilled water were prepared in 25-mL volumetric flasks. Each stock solution was then serially diluted with distilled water four times, each time by a factor of 10. This gave a total of 10 solutions ranging from 0.002 mM to 30 mM SDS. Gut homogenate was serially diluted with distilled water using a graduated 5-µL Hamilton syringe. Between 18 and 25 dilutions were performed per sample, the number being
determined by the volume of gut homogenate in each sample. For each dilution of either SDS or gut homogenate, a 1-µL droplet was sampled and the contact angle was measured as described previously. Contact angle was then plotted against dilution and the form of each plot was examined visually for evidence of micelles via a sudden change in contact angle with dilution.
Results At the end of the 48-h exposure period, sediment DODMAC concentrations were between 70% and 86% of nominal and more than 99.99% of the DODMAC recovered was associated with particulate material. There was no significant difference in particulate or pore water concentrations across the four test species (Kruskal-Wallis test: H3 < 4.6, p > 0.05). Despite similar exposure regimes, there was statistically significant interspecific variation in DODMAC tissue concentrations (H3 ) 44, p < 0.001). Chironomus riparius achieved greater tissue concentrations than any other species tested and Gammarus pulex tissue concentrations exceeded those of both Asellus aquaticus and Lumbriculus variegatus, which did not differ significantly (Figure 1). Interspecific differences in bioaccumulation persisted when tissue concentrations were normalized for lipid content (H3 ) 40, p < 0.001). Chironomus riparius larvae had the highest median lipid-normalized tissue concentrations and L. variegatus had the lowest (Figure 1). There was no significant difference in the DODMAC tissue concentration of the two crustaceans, G. pulex and A. aquaticus, after lipid normalization. There was no evidence of elimination of DODMAC over the 10-day depuration study. As there was no significant correlation between the logarithm of DODMAC tissue concentrations and time for any of the three species investigated (r < 0.78, p > 0.07), first-order rate constants for depuration were not calculated. The time taken to make one complete passage of the alimentary canal differed significantly across the four species (H3 ) 46, p < 0.001), with median values of 6.6 h for C. riparius, 2.7 h for both A. aquaticus and L. variegatus, and 1.6 h for G. pulex. There was no statistically significant difference in the gut passage time of L. variegatus or A. aquaticus, but all other pairwise comparisons were significantly different (Nonparametric Tukey-type multiple comparison test: Q > 2.639, p < 0.05). The contact angle of undiluted gut homogenates was used as a measure of maximum surfactancy: the smaller the angle, the greater the surfactancy. Contact angle measurements were made for three of the four test species, C. riparius, A. aquaticus, and G. pulex, and were compared to that of artificial pond water. No data were obtained for L. varigatus because of their small size. There was statistically significant variation in contact angle (F3,47 ) 268, p < 0.001) with APW having the largest contact angle (mean ) 91o). Contact angle decreased sequentially in Chironomus riparius (mean ) 67 o), Asellus aquaticus (mean ) 59 o), and G. pulex (mean ) 51o). All pairwise comparisons were statistically significant (Tukey test: q > 3.78, p < 0.05), indicating that gut surfactancy decreases in the order G. pulex > A. aquaticus > C. riparius. The presence of micelles was investigated by performing serial dilutions of gut homogenate or SDS. Plots of dilution against contact angle were concave, with little or no increase in contact angle with initial dilution, followed by a more pronounced increase in contact angle with increasing dilution (Figure 2). A breakpoint in dilution plots after logarithmic transformation of the dilution axis was used as evidence of the presence of micelles (18). Whereas there was clear evidence of a breakpoint in the SDS curve, with contact angle being independent of dilution down to 2 mM, after which it was proportional to dilution, this was not the case for the VOL. 41, NO. 5, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1785
FIGURE 1. Median DODMAC tissue concentrations for four benthic macroinvertbrates exposed to DODMAC-spiked clay. Tissue concentrations are presented normalized for fresh body mass (dark bars) and lipid content (light bars). Groups sharing the same letter do not differ significantly (Nonparametric Tukey-type multiple comparison test Q > 2.64, p < 0.05).
FIGURE 2. Contact angles for SDS (a) and gut homogenates of Asellus aquaticus (b), Chironomus riparius (c), and Gammarus pulex (d), all diluted with distilled water. Error bars are (1 SD. three species investigated; in all cases, contact angle was proportional to dilution between 100 and 7% gut homogenate (Figure 3).
Discussion Dioctadecyl-dimethyl ammonium chloride is highly hydrophobic and more than 99.99% of the DODMAC recovered from test vessels was associated with sediment particles. Despite this, all four invertebrate species tested accumulated DODMAC, emphasizing the importance of dietary absorption as a route of uptake of extremely hydrophobic organic contaminants. Animals were exposed to DODMAC by spiking a simple clay sediment, thereby reducing the potential for 1786
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 5, 2007
selective feeding. However, interspecific variation in DODMAC bioaccumulation was apparent, with Chironomus riparius achieving the highest tissue concentrations. Hydrophobic compounds such as DODMAC may accumulate in tissue lipids and therefore interspecific differences in bioaccumulation may be a consequence of concurrent differences in lipid content. Lipid content does differ between the four test species (30-33), but interspecific differences in DODMAC bioaccumulation persisted even after normalizing for lipid content although the difference between the two crustaceans, G. pulex and A. aquaticus, was no longer statistically significant.
FIGURE 3. Contact angles with logarithmic dilution scales for SDS (a) and gut homogenates of Asellus aquaticus (b), Chironomus riparius (c), and Gammarus pulex (d), all diluted with distilled water. Error bars are (1 SD. Bioaccumulation rate of dietary contaminants (dCt/dt, µg/ (g tissue × time)) is a function of consumption rate (F, g/(g tissue × time)), contaminant concentration in the diet (CP, µg/g), contaminant absorption efficiency (EP), contaminant elimination rate (E, µg/(g tissue × time)), and growth dilution (GD, µg/(g tissue × time)) (eq 1) (35).
dCt ) dt
∑ (F × CP × EP) - E - GD
(1)
Because of the short exposure period (48 h) and minimal opportunity for selective feeding (simple clay sediment), GD is assumed to be zero and CP is assumed to be constant across all species. This leaves interspecific differences in consumption, absorption efficiency, or elimination rate as possible drivers for the observed interspecific variation in DODMAC bioaccumulation. The four species tested are known to exhibit different feeding strategies. Lumbriculus variegatus and Chironomus riparius live within the sediment and feed on fine sedimentary particles. However, whereas L. variegatus is a conveyor belt feeder that extensively reworks the sediment (36), C. riparius is a deposit-feeding tube dweller (37) and has a lower ingestion rate than L. variegatus (14 µg dry sediment/mg dry weight/h (38) compared to g470 µg dry sediment/mg dry weight/h (39)). The epibenthic crustaceans, Gammarus pulex and Asellus aquaticus, are selective feeders, preferring fungally conditioned coarse particulate organic material (40), and have lower feeding rates than either L. variegatus or C. riparius ( C. riparius > G. pulex > A. aquaticus. However, the observed bioaccumulation pattern was C. riparius > G. pulex > L. variegatus ) A. aquaticus.
Elimination of absorbed compounds may be achieved via metabolic breakdown of the parent compound to smaller, more easily excreted, molecules (43). Whereas there is some evidence in the literature to suggest that the four species investigated differ in their ability to metabolize organic contaminants (12, 44), there is no evidence that the four species used in the present study could metabolize DODMAC (23). Similarly, we found no evidence of depuration of DODMAC over a 10-day period. Therefore, differences in elimination rate cannot account for the observed interspecific variation in DODMAC accumulation. Absorption efficiency of hydrophobic contaminants is determined, in part, by desorption efficiency. For instance, desorption efficiency is the key determinant of bioaccumulation in polychaetes as approximately 100% of the chemical desorbed is subsequently absorbed into the tissues (45, 46). Desorption efficiency is affected by gut chemistry and the length of time material is retained in the gut (i.e., gut passage time) (17). Gut passage time of the four study species varied in the order C. riparius > L. variegatus ) A. aquaticus > G. pulex. Previous studies on marine invertebrates have suggested that longer gut passage times result in increased bioaccumulation of sediment-bound chemicals (9, 17). Therefore, it would be expected that species ranking based on bioaccumulation would be the same as that based on gut passage time. Although C. riparius had both the longest gut passage time and the highest bioaccumulation, G. pulex has the shortest gut passage time but the second highest bioaccumulation of DODMAC, implying that gut passage time alone is insufficient to account for interspecific variation in bioaccumulation. The desorption efficiency of a species can also be affected by properties of the gut fluid. Surfactants are potentially able to influence desorption in two ways: (1) by enhancing aqueous solubility via incorporation into surfactant micelles and (2) by reducing the overall work required to separate sorbed material from its site of adhesion (18). We tested gut homogenates both for the presence of micelles (via dilution) VOL. 41, NO. 5, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1787
and their overall surfactancy (via minimum contact angle). The overall surfactancy of the gut fluid differed between species in the order G. pulex > A. aquaticus > C. riparius, however, there was no evidence for the presence of micelles in any of the species tested. Previous studies on marine benthic invertebrates have also found interspecific differences in surfactancy with, for example, gut fluid contact angle measurements ranging from 49 to 69o for echinoderms, 66-74o for mollusks, and 36-65o for annelids (15, 47). Evidence of micelles in gut fluid has been reported for deposit feeding and herbivorous marine invertebrates (15, 48, 49), micelles being more prevalent in gut fluids with low contact angles. Using the criterion of Mayer et al. (48), the transition from lack of micelles to the presence of micelles generally occurs in gut fluids with contact angles between 50 and 54o (15, 48), the exception being the holothuroid, Leptosynapta clark, whose gut fluid has a contact angle of 63o and contains micelles. The results obtained in the present study conform to this general pattern in that although there was evidence of gut surfactancy, contact angles were greater than 50o (i.e., 51-67o) and there was no evidence of micelle formation. Nonmicellar compounds in gut fluid can solubilize sediment-bound organic compounds, but this process is greatly enhanced by the presence of micelles (16). Hence, variation in gut surfactancy may contribute to interspecific differences in bioaccumulation, but it is unlikely to be a major determining factor. One possible explanation for the lack of concordance between either gut passage time and bioaccumulation or surfactancy and bioaccumulation is the fact that gut passage time and surfactancy are not independent. Chironomus riparius has a long gut passage time and low surfactancy, whereas G. pulex has a short gut passage time and high surfactancy. Similarly, polychaetes have very short gut passage times but high gut fluid surfactancies (15, 17, 46). This negative relationship, coupled with the finding that the majority of desorption in the gut fluid can occur within a few minutes (46), suggests that a longer gut passage time would not greatly enhance desorption from the ingested sediment when gut fluids have a high surfactancy (50). Consequently, there seem to be two different strategies that animals can adopt: either a long gut passage time or a high surfactancy. The ability of animals to absorb sediment-bound contaminants will therefore depend on which of these strategies they adopt and on their relative effectiveness. In conclusion, hydrophobic chemicals that sorb strongly to sediment particles are bioavailable to a range of benthic macroinvertebrates via dietary exposure, and this route of uptake should be considered as part of the risk assessment of these chemicals. Moreover, attention should also be paid to the types of organisms exposed as not all species will bioaccumulate these contaminants to the same degree and this will have potential food chain effects. On the basis of the results of this study, a predator feeding on C. riparius will be exposed to 5 times more DODMAC than a predator feeding on L. variegatus (on a weight-for-weight basis).
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18) (19)
Acknowledgments This work was performed while PNG was in receipt of a U.K. Natural Environment Research Council studentship (GT04/ 98/253/FS) and was supported by Unilever Research. The authors thank Stuart Marshall and Stephen Harding for their contribution to this work.
(20)
(21)
Literature Cited (1) Kukkonen, J.; Landrum, P. F. Distribution of organic carbon and organic xenobiotics among different particle-size fractions in sediments. Chemosphere 1996, 32, 1063-1076. (2) Timmerman, K.; Andersen, O. Bioavailability of pyrene to the deposit-feeding polychaete Arenicola marina: importance of 1788
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 5, 2007
(22)
(23)
sediment versus water uptake routes. Mar. Ecol.: Prog. Ser. 2003, 246, 163-172. Lu, X.; Reible, D.; Fleeger, J. Bioavailability and assimilation of sediment-associated benzo[a]pyrene by Ilyodrilus templetoni (Oligochaeta). Environ. Toxicol. Chem. 2004, 23, 57-64. Trowbridge, A.; Swackhamer, D. Preferential biomagnification of aryl hydrocarbon hydroxylase-inducing polychlorinated biphenyl congeners in the Lake Michigan, USA, lower food web. Environ. Toxicol. Chem. 2002, 21, 334-341. Fisk, A.; Hoekstra, P.; Gagnon, J.; Duffe, J.; Norstrom, R.; Hobson, K.; Kwan, M.; Muir, D. Influence of habitat, trophic ecology and lipids on, and spatial trends of, organochlorine contaminants in Arctic marine invertebrates. Mar. Ecol.: Prog. Ser. 2003, 262, 201-214. Schuler, L.; Wheeler, M.; Bailer, A.; Lyder, M. Toxicokinetics of sediment-sorbed benzo[a]pyrene and hexachlorobiphenyl using the freshwater invertebrates Hyalella azteca, Chironomus tentans, and Lumbriculus variegatus. Environ. Toxicol. Chem. 2003, 22, 439-449. Leppa¨nen, M.; Kukkonen, J. Toxicokinetics of sediment-associated polybrominated diphenylethers (flame retardants) in benthic invertebrates (Lumbriculus variegatus, Oligochaeta). Environ. Toxicol. Chem. 2004, 23, 166-172. Millward, R. N.; Fleeger, J. W.; Reible, D. D.; Keteles, K. A.; Cunningham, B. P.; Zhang, L. Pyrene bioaccumulation, effects of pyrene exposure on particle-size selection, and fecal pyrene content in the oligochaete Limnodrilus hoffmeisteri (Tubificidae, Oligochaeta). Environ. Toxicol. Chem. 2001, 20, 1359-1366. Wang, W.-X.; Chow, A. T. S. Benzo[a]pyrene absorption and exposure pathways in the green mussel Perna viridis. Environ. Toxicol. Chem. 2002, 21, 451-458. Smoot, J. C.; Mayer, L. M.; Bock, M. J.; Wood, P. C.; Findlay, R. H. Structures and concentrations of surfactants in gut fluid of the marine polychaete Arenicola marina. Mar. Ecol.: Prog. Ser. 2003, 258, 161-169. Wang, W.-X.; Fisher, N. S. Assimilation efficiencies of chemical contaminants in aquatic invertebrates: a synthesis. Environ. Toxicol. Chem. 1999, 18, 2034-2045. Verrengia Guerrero, N. R. V.; Taylor, M. G.; Davies, N. A.; Lawrence, M. A. M.; Edwards, P. A.; Simkiss, K.; Wider, E. A. Evidence of differences in the biotransformation of organic contaminants in three species of freshwater invertebrates. Environ. Pollut. 2002, 117, 523-530. Kukkonen, J.; Landrum, P. F. Measuring assimilation efficiencies for sediment-bound PAH and PCB congeners by benthic organisms. Aquat. Toxicol. 1995, 32, 75-92. Mayer, L. M.; Chen, Z.; Findlay, R. H.; Fang, J.; Sampson, S.; Self, R. F. L.; Jumars, P. A.; Quetel, C.; Donard, O. F. X. Bioavailability of sedimentary contamiants subject to deposit-feeder digestion. Environ. Sci. Technol. 1996, 30, 2641-2545. Mayer, L. M.; Weston, D. P.; Bock, M. J. Benzo[a]pyrene and zinc solubilization by digestive fluids of benthic invertebrates - a cross-phyletic study. Environ. Toxicol. Chem. 2001, 20, 18901900. Voparil, I. M.; Mayer, L. M. Dissolution of sedimentary polycyclic aromatic hydrocarbons into the lugworm’s (Arenicola marina) digestive fluids. Environ. Sci. Technol. 2000, 34, 1221-1228. Ahrens, M. J.; Hertz, J.; Lamoureux, E. M.; Lopez, G. R.; McElroy, A. E.; Brownawell, B. J. The effect of body size on digestive chemistry and absorption efficiencies of food and sedimentbound organic contaminants in Neris succinea (Polychaeta). J. Exp. Mar. Biol. Ecol. 2001, 263, 185-209. Shinoda, K. Solvent Properties of Surfactant Solutions; Edward Arnold: London, 1967. Selck, H.; Decho, A. W.; Forbes, V. E. Effects of chronic metal exposure and sediment organic matter on digestive absorbtion efficiency of cadmium by the deposit-feeding polychate Capitella species I. Environ. Toxicol. Chem. 1999, 18, 1289-1297. Di Toro, D.; Zarba, C.; Hansen, D.; Berry, W.; Swartz, R.; Cowan, C.; Pavlou, S.; Allen, H.; Thomas, N.; Paquin, R. Technical basis for establishing sediment quality criteria for nonionic organic chemicals using equilibrium partitioning. Environ. Toxicol. Chem. 1991, 10, 1541-1583. ECETOC. DHTDMAC:- Aquatic and Terrestrial Hazard Assessment; CAS No. 61789-80-8; European Centre for Ecotoxicology and Toxicology of Chemicals: Brussels, Belgium, 1993. Bureau, E. C. European Union Risk Assessment Report: Dimethyldioctadecylammonium chloride; Institute for Health and Consumer Protection: Dortmund, Germany, 2002. Gaskell, P. Ph.D. Thesis. The Importance of Dietary Exposure in the Determination of Bioavailability of Sediment-Sorbed
(24) (25) (26) (27) (28) (29) (30)
(31) (32)
(33)
(34) (35) (36) (37) (38)
(39)
Organic compounds to Benthic Macroinvertebrates. The University of Sheffield, Sheffield, U.K., 2002. U.S. EPA. Standard operating procedure: culture of Lumbriculus variegatus; Environmental Research Laboratory, U.S. Environmental Protection Agency: Duluth, MN, 1993. Credland, P. A new method for establishing a permanent laboratory culture of Chironomus riparius Meigen (Diptera: Chironomidae). Freshwater Biol. 1973, 3, 45-51. Health and Safety Executive Methods for the determination of ecotoxicity. Approved Code of Practice 8; Her Majesty’s Stationary Office: London, 1982. Pittinger, C.; Woltering, D.; Masters, J. Bioavailability of sediment-sorbed and aqueous surfactants to Chironomus riparius midge. Environ. Toxicol. Chem. 1989, 8, 1023-1034. Bransome, E. D. E. The Current Status of Liquid Scintillation Counting; Grune and Stratton: New York, 1970. Neame, K. D.; Homewood, C. A. Introduction to Liquid Scintillation Counting; Butterworth: London, 1974. Montan ˜ e´s, J. F. C.; van Hattum, B.; Deneer, J. Bioconcentration of chlorpyrifos by the freshwater isopod Asellus aquaticus (L.) in outdoor experimental ditches. Environ. Pollut. 1995, 88, 137146. Meier, G. M.; Meter, E. I.; Meyns, S. Lipid content of stream macroinvertebrates Arch. Hydrobiol. 2000, 147, 447-463. Leslie, H. A.; Oosthoek, A. J. P.; Busser, F. J. M.; Kraak, M. H. S.; Hermans, J. L. M. Biomimetic solid-phase microextraction to predict body residues and toxicity of two chemicals that act by narcosis. Environ. Toxicol. Chem. 2002, 21, 229-234. Kukkonen, J. V. K.; Landrum, P. F.; Mitra, S.; Gossiaux, D. C.; Gunnarsson, J.; Weston, D. The role of desorption for describing the bioavailability of select polycyclic aromatic hydrocarbons and polychlorinated biphenyl congeners for seven laboratoryspiked sediments. Environ. Toxicol. Chem. 2004, 23, 1842-1851. Fundamentals of Aquatic Toxicology; Rand, G., Ed.; Taylor & Francis: Washington, DC, 1995. Lee, H., II. In Sediment Toxicity Assessment; Burton, G. A., Jr, Ed.; Lewis Publishers: Boca Raton, FL, 1992; pp 267-293. Brinkhurst, R. O.; Jamieson, B. G. M. Aquatic Oligochaeta of the World; Oliver and Boyd: Edinburgh, U.K., 1971. Oliver, D. R. Life history of the Chironomidae. Annu. Rev. Entomol. 1971, 16, 211-230. Bervoets, L.; De Bruyn, L.; Van Ginneken, L.; Blust, R. Accumulation of 137Cs by larvae of the midge Chironomus riparius from sediment: effect of potassium. Environ. Toxicol. Chem. 2003, 22, 1589-1596. Kukkonen, J.; Landrum, P. F. Effects of sediment-bound polydimethylsiloxane on the bioavailability and distribution of
(40)
(41)
(42)
(43)
(44)
(45)
(46)
(47)
(48)
(49) (50)
benzo[a]pyrene in lake sediment to Lumbriculus variegatus. Environ. Toxicol. Chem. 1995, 14, 523-531. Grac¸ a, M. A. S.; Maltby, L.; Calow, P. Comparative ecology of Gammarus pulex (L.) and Asellus aquaticus (L.) II: fungal preferences. Hydrobiologia 1994, 281, 163-170. Adcock, J. Engergetics of a population of Asellus aquaticus (Crustacea, Isopoda): respiration and energy budgets. Freshwater Biol. 1982, 12, 257-269. McLoughlin, N.; Yin, D.; Maltby, L.; Wood, R.; Yu, H. Evaluation of sensitivity and specificity of two crustacean biochemical biomarkers. Environ. Toxicol. Chem. 2000, 19, 2085-2092. van Leeuwen, C.; Hermens, J. Risk Assessment of Chemicals: An Introduction; Kluwer Academic Publishers: Dordrecht, The Netherlands, 1995. Van Hattum, B.; Montanes, J. F. C. Toxicokinetics and bioconcentration of polycyclic aromatic hydrocarbons in freshwater isopods. Environ. Sci. Technol. 1999, 33, 2409-2417. Weston, D. P.; Mayer, L. M. Comparison of in vitro digestive fluid extraction and traditional in vivo approaches as measures of polycyclic aromatic hydrocarbon bioavailability from sediments. Environ. Toxicol. Chem. 1998, 17, 830-840. Ahrens, M. J.; Hertz, J.; Lamoureux, E. M.; Lopez, G. R.; McElroy, A. E.; Brownawell, B. J. The role of digestive surfactants in determining bioavailability of sediment-bound hydrophobic organic contaminants to 2 deposit-feeding polychaetes. Mar. Ecol.: Prog. Ser. 2001, 212, 145-157. Rust, A.; Burgess, R.; McElroy, A.; Cantwell, M.; Brownawell, B. Influence of soot carbon on the bioaccumulation pf sedimentbound polycyclic aromatic hydrocarbons by marine benthic invertebrates: an interspecific comparison. Environ. Toxicol. Chem. 2004, 22, 2594-2603. Mayer, L. M.; Schick, L. L.; Self, R. F. L.; Jumars, P. A.; Findlay, R. H.; Chen, Z.; Sampson, S. Digestive environments of benthic macroinvertebrate guts: Enzymes, surfactants, and dissolved organic matter. J. Mar. Res. 1997, 55, 785-812. Tugwell, S.; Branch, G. Effects of herbivore gut surfactants on kelp polyphenol defenses. Ecology 1992, 73, 205-215. Penry, D.; Weston, D. Digestive determinants of benzo[a]pyrene and phenanthrene bioaccumulation by a deposit-feeding polychaete. Environ. Toxicol. Chem. 1998, 17, 2254-2265.
Received for review August 11, 2006. Revised manuscript received November 30, 2006. Accepted December 21, 2006. ES061934B
VOL. 41, NO. 5, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1789
