Uptake and Elimination Routes of Inorganic Mercury and

Inorganic Mercury and. Methylmercury in Daphnia magna. MARTIN T. K. TSUI AND. WEN-XIONG WANG*. Department of Biology, The Hong Kong University of ...
0 downloads 0 Views 162KB Size
Download PDF
Environ. Sci. Technol. 2004, 38, 808-816

Uptake and Elimination Routes of Inorganic Mercury and Methylmercury in Daphnia magna MARTIN T. K. TSUI AND WEN-XIONG WANG* Department of Biology, The Hong Kong University of Science and Technology (HKUST), Clear Water Bay, Kowloon, Hong Kong, People’s Republic of China

Mercury (Hg) is an important environmental pollutant due to its highly toxic nature and widespread occurrence in aquatic systems. The biokinetics of Hg in zooplankton have been largely ignored in previous studies. This study examines the assimilation, dissolved uptake, and efflux of inorganic mercury [Hg(II)] and methylmercury (MeHg) in a freshwater cladoceran, Daphnia magna, and models the exposure pathways of Hg(II) and MeHg in the daphnids. The assimilation efficiencies (AEs) of both Hg species decreased significantly with increasing algal carbon concentrations. The dissolved uptake of Hg(II) and MeHg was proportional to the ambient concentration (ranging from environmentally realistic to high concentration over a 3-4 orders of magnitude variation), whereas MeHg had a slightly higher uptake rate constant (0.46 L g-1 h-1) than Hg(II) (0.35 L g-1 h-1). Surprisingly, the efflux rate constants of Hg(II) and MeHg were rather comparable (0.0410.063 day-1). The release of both Hg(II) and MeHg via different routes (excretion, egestion, molting, and neonate production) was further examined at different food concentrations. It was found that regeneration into the dissolved phase was important for D. magna to eliminate both Hg species, but maternal transfer of Hg(II) (11-15%) and MeHg (32-41%) to neonates represented another important pathway for the elimination of Hg(II) and MeHg from the mothers. Modeling results suggest that food is an important source for MeHg exposure (47-98%), but water exposure represents 31-96% of Hg(II) accumulation in D. magna, depending on the variation of Hg bioconcentration factor in ingested food. Furthermore, MeHg predominates the bioaccumulation of Hg in D. magna even though MeHg constitutes only a small percentage of the total Hg in the water. The results strongly indicate that maternal transfer of Hg(II) and MeHg in freshwater zooplankton should be considered in many toxicity testings and risk assessment in aquatic food chains.

Introduction Mercury (Hg) has been recognized as a very toxic metal in the aquatic environment and poses a serious threat to human health through fish consumption due to the fact that it can biomagnify along the food chains. Biomagnification of Hg in aquatic food chains is mainly due to its organic form (i.e., methylmercury or MeHg) because it can be efficiently taken * Corresponding author phone: (852) 2358 7346; fax: (852) 2358 1559; e-mail: [email protected]. 808

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 3, 2004

up by aquatic organisms and transferred to a higher trophic level (1, 2). In contrast, inorganic Hg [Hg(II)] is biodiminished through the food chain because of its poorer assimilation and more efficient excretion. Field measurements found that almost all Hg (>90%) in fish muscles was in methylated forms (3); therefore, the estimated bioconcentration factors (BCF) of MeHg in fish were extremely high (close to or even greater than 106) given that the percentage of total Hg as MeHg in the ambient water was only 0.05-2% (4, 5). The BCFs of inorganic and total Hg were 1-2 orders of magnitude less than those of MeHg (2). Most field studies examined the trophic transfer of Hg by collecting various abiotic (water and sediment) and biotic (phytoplankton, zooplankton, and fish) compartments and then analyzing the respective Hg concentrations (1, 2, 6). However, these studies did not provide information regarding the uptake and removal kinetics of the Hg compounds, which are important parameters in interpreting and predicting the food-chain transfer of Hg. It remains essentially unknown whether the accumulated Hg in zooplankton originates from direct uptake from water, ingested phytoplankton, or parental transfer. Empirical evidence exists for the dominance of food for MeHg exposure in fish (7), but no information is available with regard to zooplankton species. Such information is, however, critical for the risk analysis of Hg and for a better and more realistic design of water toxicity testing, especially for Daphnia, which have been widely used as test organisms for toxicity testing. The incorporation of Hg at the base of the food chain is crucial for the subsequent trophic transfer to fish (8). Most previous studies, however, focused on the toxicity and bioaccumulation of Hg by fish (5), with little attention being paid to freshwater zooplankton, which play an important role in ecosystem dynamics. Daphnia are common freshwater zooplankton species in many temperate lakes and also important links between phytoplankton and fish. They are relatively large in size and can carry higher Hg body burdens than other taxa (2), and therefore they can be a good predictor of the Hg contents at higher trophic levels (e.g., fish). Chen et al. (6) observed that the Hg content in macrozooplankton (>202 µm) as opposed to the Hg content in small plankton (45-202 µm) was a strong positive predictor of Hg in fish. Although the importance of macrozooplankton such as Daphnia for Hg cycling in freshwater systems has been appreciated, there have been only limited studies on the uptake and depuration of Hg by the freshwater macrozooplankton (9, 10). Recently, MeHg has been detected in the eggs of freshwater fish exposed to trace Hg concentrations in the environment (11-13). The concentration of total Hg in the fish eggs was found to be proportional to that in the maternal carcasses (11, 13). Whether other aquatic organisms, such as Daphnia, can transfer Hg(II) and MeHg to the eggs during oogenesis is unknown at present. Previously, Yu and Wang (14) showed that reproduction was an important pathway for Daphnia magna to remove selenium (Se) and possibly zinc (Zn) from the body. These trace elements are essential, and their behavior in the zooplankton can constitute a marked contrast to the nonessential and toxic elements such as Hg. In this study, we examined the assimilation, uptake, efflux, and elimination routes of Hg(II) and MeHg by D. magna. We then employed a kinetic model (15, 16) to assess the relative importance of food and water as the exposure pathways and the relative importance of both Hg species as sources of Hg accumulation in D. magna. We further quantified the 10.1021/es034638x CCC: $27.50

 2004 American Chemical Society Published on Web 12/20/2003

elimination of both species of Hg by the animals at different food concentrations, with a focus on the potential maternal transfer of Hg to the offspring.

Materials and Methods Radioisotopes and Organisms. A radiotracer technique was used throughout the present study because it is a highly sensitive method and the biokinetics of Hg can be followed noninvasively over time. 203HgCl2 [Hg(II)] (in 0.1 N HCl, t1/2 ) 46.6 days, specific activity ) 26.1-52.3 GBq g-1) was purchased from Risø National Laboratory, Roskilde, Denmark. CH3203HgCl (MeHg) (with the same specificity activity as 203HgCl2) was synthesized from 203HgCl2 using an established method (17) and was dissolved in 5 mM Na2CO3. D. magna has been cultured in our laboratory for several years. This clone was originally obtained from the Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan, China. D. magna was cultured in glass-fiber (GF/C) filtered pond water at a temperature of 23.5 °C with a 14:10 h light/dark cycle. The water was collected from an uncontaminated freshwater pond within the campus of the university. The cladocerans were fed with a mixture of Chlamydomonas reinhardtii and Chlorella vulgaris every day at a food concentration of 1-2 × 105 cells mL-1, and the water was renewed every 2-3 days. The algae were maintained in artificial WC medium for freshwater phytoplankton culture (18) in ultrapure water at 23.5 °C, with a 14:10 h light/dark cycle and under a light intensity of 100 µmol of photons m-2 s-1. Assimilation of Hg(II) and MeHg. In this experiment, the algae were radiolabeled and pulse-fed to D. magna at different food levels. Subsequently, the retention of the radioisotopes was followed nondestructively over time in order to quantify the assimilation of Hg(II) and MeHg from the ingested food. Two freshwater green algae (Chla. reinhardtii and Scenedesmus obliquus) were used as labeling algae, because they are common in freshwater environments and good foods for the cladocerans. We did not examine the assimilation of Hg from Chlo. vulgaris, which was used as food for the culture of the animals. To radiolabel the algae, the algae were first grown at an initial concentration of 2.5 × 105 cells mL-1 in a modified WC medium, which was the same as the original WC medium but without the addition of Cu, Zn, and EDTA, to minimize the possible chelation with the radiotracers and metal toxicity due to the absence of EDTA (19). After 3-4 days of growth, the culture was in the exponential phase and the radioisotopes were spiked into the solution [740 kBq L-1, corresponding to 28 µg L-1 for Hg(II), and 185 kBq L-1 or 7 µg L-1 for MeHg]. Different amounts of radioisotopes were spiked to result in a comparable amount of Hg ingested by the animals after the radioactive feeding among the different food concentration treatments. After 2-3 more days of growth, the cells were considered to be uniformly labeled. Any biotransformation between Hg(II) and MeHg in the algae was assumed to be negligible. Afterward, the radiolabeled cells were collected by filtering onto 1-µm polycarbonate membranes and were resuspended into 0.22-µm-filtered pond water. This procedure was repeated two more times to remove the weakly bound radioisotopes on the algal surfaces, and the cell density of the algal solution was counted under the microscope. The algae retained on the 1-µm polycarbonate membrane were also radioassayed to determine the Hg concentration in the phytoplankton. Adult D. magna with similar body sizes were collected from stock cultures, and their gut contents were evacuated in 0.22-µm-filtered pond water for 2-3 h without food addition. Radiolabeled algae [concentrations in algae: about 100 µg of Hg g-1 of dry wt for Hg(II) and 25 µg of Hg g-1 of dry wt for MeHg] were added separately into the feeding beakers containing 10-15 animals and 100-150 mL of 0.22-

µm-filtered pond water (i.e., 10 mL of water per animal), which resulted in different algal concentrations (5 × 103-105 cells mL-1 for Chla. reinhardtii and 104 to 5 × 105 cells mL-1 for S. obliquus). Different algal cell concentrations were prepared to result in comparable algal carbon concentrations between the two algal diets. All treatments consisted of three replicates. D. magna were exposed in the dark for 30 min [approximate body burdens immediately after the feeding: 0.4 µg of Hg g-1 of dry wt for Hg(II) and 0.1 µg of Hg g-1 of dry wt for MeHg], after which the animals (by 300-µm mesh) and any egested feces (by 40-µm mesh) were collected and rinsed gently before the radioactivity measurements. The 30 min of radioactive feeding time was chosen because this duration was comparable to or shorter than the gut passage time of the animals (20), to minimize the defecation of radioactive feces. These two fractions (i.e., animals and feces, the latter represented a small fraction of Hg ingested by the animals) represented the total radioactivity ingested by D. magna during the 30-min feeding period. After radioactivity counting, D. magna were immediately returned to the 0.22-µm-filtered pond water containing the same conditions as in the pulse-feeding period, except with nonradioactive food. The depuration lasted for 30 h, during which time the radioactivity retained in the animals was measured periodically. After each radioactivity measurement, the water and food were completely renewed. The Hg assimilation efficiency (AE) was calculated as the percentage of ingested Hg retained in D. magna after 12 h of depuration (14, 15). The physiological turnover rate constant (k) was calculated as the slope of the linear regression between the natural log of the percentage of Hg retained in the animals and the time of depuration between 12 and 30 h, assuming that D. magna completed the Hg digestion within 12 h (14) and that any loss of Hg afterward was due to the physiological turnover (21). Because the first AE experiment testing the influence of food concentration on Hg assimilation employed a relatively high Hg concentration (due to the availability of radioactive Hg with a low specific activity), we subsequently specifically tested the influences of Hg concentration in ingested algae on its assimilation by Daphnia, using 203Hg with a relatively high specific activity (52.3 GBq g-1). The algae Chla. reinhardtii were radiolabeled with 203Hg(II) and Me 203Hg at different concentrations [0.28-28 µg L-1 for Hg(II) and 0.0343.4 µg L-1 nM for MeHg], as described above. The AEs were subsequently quantified for Daphnia at a food concentration of 2 × 104 cells mL-1. Dissolved Uptake of Hg(II) and MeHg. Adults of D. magna were exposed to different concentrations of Hg(II) and MeHg in the dissolved phase for a total of 8 h, and the uptake of radioisotopes was followed nondestructively over time. The Hg(II) and MeHg concentrations chosen ranged from 1.0 to 2835 ng of Hg L-1 and from 0.5 to 709 ng of Hg L-1, respectively. These concentrations were achieved by adding different amounts of radioactivity of 203Hg isotopes and were calculated from the specific activity of the radioisotopes spiked into the solution. No stable Hg was added into the experimental waters. The low concentration range used in this study represented the environmentally realistic concentration of both Hg species in natural waters. These extremely low concentrations were also similar to the concentrations used by a recent study on Hg transfer from phytoplankton to zooplankton in a mesocosm system (22), but were considerably lower than the concentrations used by many previous studies. The highest concentrations used in this experiment were at least 30 times lower than the acute toxicity of Hg(II) and MeHg (Tsui and Wang, unpublished data). Background total dissolved Hg concentrations in the pond water as well as in the experimental water, as quantified using trace metal clean technique, were 0.08-0.18 ng L-1. VOL. 38, NO. 3, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

809

The metals were spiked in 0.22-µm-filtered pond water and equilibrated for 3-4 h. Adults of D. magna with similar body sizes were collected and acclimated for 2-3 h to evacuate their gut contents. The animals (16-20 individuals) were added to each feeding beaker containing 150 mL of test solution without food additions to avoid the scavenging of metals by food particles from the water (23) and causing artifacts from the true uptake from the dissolved phase. We did not quantify whether the filtration rate of the animals was affected without the presence of food. There were three replicates for each concentration treatment. The experiment lasted for a total of 8 h, during which time the test solution was renewed every 4 h to minimize the decrease of Hg in the solution. The radioactivity accumulated in D. magna was nondestructively measured every 2 h by first placing the animals in 1 L of 0.22-µm-filtered pond water for 1-2 min to remove the radioisotopes in the carapace fluid (24) and then rinsing the animals gently. D. magna were immediately returned to the test solution after the radioactivity measurements. The radioactivity of the test solution was also monitored by taking 5-mL samples at 0, 2, 4, 6, and 8 h of exposure. After the exposure, half of the exposed D. magna were collected and dried at 80 °C overnight for dry-weight measurements. Another half of the D. magna were fractionated to determine the relative distribution of Hg(II) and MeHg in the soft tissues and exoskeleton, using the method described in Yu and Wang (14, 24). The dryweight concentration factor (DCF) was calculated as the ratio of the radioactivity in D. magna (ccpm per gram of dry weight) to the radioactivity in the test solution (ccpm per milliliter, calculated as the mean of the initial and final concentrations). The influx rates were calculated as the slope of the linear regression (using model II) between the DCF and the time of exposure (2-8 h) multiplied by the initial Hg concentration in the test solution. Efflux of Hg(II) and MeHg. Chla. reinhardtii was radiolabeled with Hg(II) and MeHg using an approach similar to the one described in the assimilation experiment. The radiolabeled algae were fed to 150 adults of D. magna in 500 mL of GF/F-filtered pond water at a cell density of 5 × 104 cells mL-1. Each day, the radiolabeled algae were added every 3 h for a total of 6 h. Afterward, the animals were removed, rinsed gently, and placed in GF/F-filtered pond water with unlabeled Chla. reinhardtii at a cell density of 5 × 104 cells mL-1. This feeding regime was repeated for a total of 3 days. We fed the animals with only radioactive food for 6 h each day primarily for the purpose of minimizing the potential uptake of Hg from the aqueous phase due to Hg desorption from the radioactive food. The nonradioactive feeding period allowed the radioisotopes to be well incorporated by the animals while the unassimilated metals were purged out of the guts of animals. After 3-day feeding, 10 individuals of D. magna were collected and depurated in 80 mL of GF/Ffiltered pond water at three different food levels (104, 5 × 104, and 2 × 105 cells mL-1). There were three replicates (10 individuals each) for each food concentration treatment. During the depuration, the radioactivity retained in D. magna was measured every 12 h within the first 2 days and every day afterward for a total of 7 days. The water was renewed every 12 h, and new food was provided every 6-12 h. The water, molts, feces, and neonates produced were collected and radioassayed in order to construct a daily release budget of Hg(II) and MeHg by D. magna during the 7 days of depuration. D. magna on days 0 and 7 (at three food levels) of the depuration period were fractionated to determine the Hg distribution in the exoskeleton and the soft tissues using the same method as described above. The efflux rate constants (ke) were calculated from the slope of the linear regression between the natural log of the percentage of Hg retained in D. magna and the time of depuration 810

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 3, 2004

between 3 and 7 days, whereas the y-intercept represented the percentage of Hg retained in this exchanging compartment in D. magna. The biological retention half-lives (t1/2) of Hg in D. magna were calculated as t1/2 ) 0.693/ke. Modeling the Exposure and the Trophic Transfer Factor. Under steady-state conditions, the Hg concentration in D. magna (Css, µg g-1) can be calculated by the following equation (16):

Css ) (ku × Cw + AE × IR × Cf)/ke

(1)

where ku ) uptake rate constant from the dissolved phase (L g-1 day-1), Cw ) Hg concentration in the dissolved phase (µg L-1), AE ) assimilation efficiency (%), IR ) ingestion rate (g g-1 day-1), Cf ) Hg concentration in the ingested food particles (µg g-1), and ke ) Hg efflux rate constant (day-1). In this equation, the growth rate constant (g) was ignored because it was assumed to be negligible when compared to the efflux rate constant (ke). The fraction of Hg accumulation from the aqueous phase (f) can be predicted by the following equation, assuming that Cf can be estimated from the Cw and the BCF of Hg in phytoplankton:

f ) ku/[(AE × IR × BCF) + ku]

(2)

The fraction of Hg accumulation from the dietary exposure can be calculated as 1 - f. To predict the fraction of total Hg accumulation due to Hg(II) or MeHg, we first calculated the concentration factors (CF) of both Hg species using eq 3 (16, 25):

CF ) ku/ke + (AE × IR × BCF)/ke

(3)

The fraction of total Hg in D. magna due to accumulation of Hg(II) (RHg(II)) and MeHg (RMeHg) can then be calculated by the following equations (16, 25):

RHg(II) ) CFHg(II)/[CFHg(II) + (CFMeHg × CMeHg/CHg(II))] (4) RMeHg ) 1 - RHg(II)

(5)

where CFHg(II) and CFMeHg are the concentration factors of Hg(II) and MeHg in D. magna, respectively. The trophic transfer factor (TTF) is the ratio of Hg concentration in zooplankton to that in phytoplankton prey and can be determined as below (16, 26):

TTF ) (AE × IR)/ke

(6)

Radioactivity Measurements and Statistical Analysis. The radioactivity of Hg(II) and MeHg was determined by a Wallac 1480 NaI(T1) gamma counter (Turku, Finland). All analyses were related to appropriate standards and were calibrated for spillover and radioisotope decay. γ-Emission was measured at 279 keV. Counting times were adjusted to yield propagated counting errors generally 0.05, one-way ANOVA). This result was similar to previous studies that have demonstrated that the physiological turnover rate constants for trace metals were relatively independent of the food concentrations in marine copepods (21, 31) and D. magna (14). For MeHg, the k values were much lower for D. magna feeding on Chla. reinhardtii than on S. obliquus, approaching almost zero at low food concentrations. This indicated that MeHg was well incorporated in the soft tissues and had almost not turned over during this period (between 12 and 30 h of depuration) when the animals fed on Chla. reinhardtii. The difference in the k values between these two algal diets remained unexplained. Nevertheless, the k values of MeHg were generally much smaller than those of other trace metals, whereas Hg(II) had k values comparable to or slightly higher than those of other trace metals (i.e., Cd, Cr, Se, and Zn) studied in D. magna (14). Dissolved Uptake of Hg(II) and MeHg. D. magna exhibited a linear uptake of Hg(II) and MeHg at a wide range (3-4 orders of magnitude) of ambient concentrations over 8 h of exposure (data not shown). It is obvious that different ambient concentrations of Hg(II) and MeHg affected the dryweight concentration factors (DCF) after 8 h of exposure (Figure 4). At the very low concentration range, the DCF of Hg(II) was higher but became lower than that of MeHg at the higher concentration range. The decreasing DCF with increasing concentrations of Hg(II) and MeHg indicated some possibilities of regulation, but it should be noted that the 812

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 3, 2004

FIGURE 5. Influx rates of Hg(II) and MeHg into D. magna at different dissolved Hg concentrations. Error bars are standard deviations of three replicates. quantified DCF may also include metal sorption onto the daphnid’s body. The influx rates for Hg(II) and MeHg into D. magna at different dissolved concentrations are shown in Figure 5. A total of three different experiments employing different dissolved Hg concentrations were conducted. The power coefficients describing the relationship between the influx rate and the ambient concentration were close to unity for both Hg(II) (1.11) and MeHg (0.937). The uptake rate constants were 0.350 L g-1 h-1 for Hg(II) and 0.460 L g-1 h-1 for MeHg. After 8 h of exposure, we found that both Hg(II) and MeHg were mainly distributed in the soft tissues rather than in the exoskeleton. The percent distribution of both Hg(II) and MeHg in D. magna following 8 h of dissolved uptake was not significantly affected by different exposure concentrations and ranged from 80 to 100% for both Hg compounds in the soft tissues. In a previous study, the dissolved uptake of MeHg by D. magna was shown to be proportional to the ambient concentrations from 0 to 4 ng Hg L-1 (10), whereas the MeHg content in D. magna reached steady state after 48 h. In our study employing 8 h of exposure, MeHg and possibly Hg(II) were not saturated in the animals. In addition, we renewed the water at a 4-h interval to minimize the decrease of Hg(II) and MeHg in the exposure media. For the decreasing DCFs with increasing exposed concentrations, a possibility was that all of the binding sites available for Hg on or in the animals were completely saturated at the high concentrations of dissolved Hg used, thus leaving an excess of Hg dissolved in solution. This may explain the decreasing DCF with increasing Hg concentrations after 8 h of exposure (Figure 4). In addition, the DCF values obtained in our experiment (103-104) were considerably lower than the BCFs estimated from the field-collected zooplankton samples (105-106) (1, 2), primarily because no steady state of Hg uptake was reached within our short experimental period (8 h). Mason et al. (27) proposed that the passive diffusion of lipophilic Hg species through the lipid bilayer is the dominant pathway for Hg bioaccumulation by the microorganisms. New experimental evidence, however, demonstrated that the uptake of Hg(II) was through facilitated transport, rather than by simple passive diffusion into microorganisms (32). Another recent study indicated that several freshwater algal species (both prokaryotic and eukaryotic) could actively take up MeHg from the water, but several mechanisms (e.g., passive and active transport) may simultaneously be involved in MeHg uptake in microorganisms (33). One possible reason

TABLE 2. Efflux Rate Constants (ke), Biological Retention Half-Life (t1/2), and Percentage Retained in the Slow-Exchanging Compartment of Hg(II) and MeHg in D. magna in the 7-Day Efflux Experiment ke (day-1) food concn (cells mL-1) 1× (L) 5 × 104 (M) 2 × 105 (H) 104

t1/2 (days)

% in the slow-exchanging compartment

Hg(II)

MeHg

Hg(II)

MeHg

Hg(II)

MeHg

0.049 ( 0.011 0.045 ( 0.003 0.061 ( 0.007

0.050 ( 0.009 0.041 ( 0.013 0.063 ( 0.012

14.7 ( 3.0 15.6 ( 0.9 11.5 ( 1.4

14.1 ( 2.6 18.2 ( 6.5 11.3 ( 2.2

67.9 ( 1.0 62.5 ( 2.5 62.5 ( 2.7

92.6 ( 7.6 94.7 ( 7.7 98.3 ( 4.6

FIGURE 6. Retention of Hg(II) and MeHg by D. magna at three food levels following 3 days of dietary uptake of Hg-laden algae. Error bars are the standard deviations of three replicates.

FIGURE 7. Percent contribution (on average) of the four routes in the release of Hg(II) and MeHg from D. magna over the 7 day depuration period. Error bars are the standard deviations of three replicates.

for the differences among these studies is the variety of organisms tested. For example, Mason et al. (27) and Moye et al. (33) tested the eukaryotic algae (both freshwater and marine species), whereas Goldling et al. (32) employed prokaryotic bacteria. In our study, we found that the dissolved uptake of Hg(II) and MeHg by D. magna was proportional to the ambient concentrations, possibly through passive uptake, but the mechanism needs further study to elucidate. Efflux of Hg(II) and MeHg. After 3 days of dietary uptake of Hg(II) and MeHg by D. magna, the zooplankton constantly depurated MeHg during the 7 days of depuration but rapidly released Hg(II) within the first 2-3 days of depuration, followed by gradual loss afterward (Figure 6). The percentage of MeHg (57-66%) retained after 7 days of depuration was higher than that of Hg(II) (37-46%) at different food concentrations. As shown in Table 2, the efflux rate constants (ke) and the biological retention half-lives (t1/2) were comparable for both Hg(II) and MeHg at different food levels (p > 0.05, one-way ANOVA). Nevertheless, the percentage retained in the slow-exchanging compartment was much higher for MeHg (93-98%) than for Hg(II) (62-68%), but again, these parameters were not affected by different food levels (p > 0.05, one-way ANOVA). After 3 days of dietary uptake and 7 days of depuration at three food levels, there was no significant difference in the distribution of Hg(II) and MeHg in D. magna soft tissues and exoskeleton at different time and food levels, where both Hg species were mostly distributed in the soft tissues (i.e., 90-100%). We also monitored the elimination routes of the ingested Hg(II) and MeHg during the 7 days of depuration by measuring the radioactivity in the water, molts, feces, and neonates produced by the animals. Overall, regeneration into the dissolved phase was a dominant pathway for D. magna to depurate Hg(II) (78-80% of the overall loss) and MeHg (53-59% of the overall loss) throughout the depuration period, whereas molting (2.1-3.7% of the overall loss) and fecal egestion (3.0-5.7% of the overall loss) were only minor

routes (Figure 7). Clearly, maternal transfer (i.e., from females to neonates) of Hg(II) (11-15% of the overall loss) and MeHg (32-41% of the overall loss) was also important for D. magna to eliminate Hg from their bodies. On a daily basis, the efflux of MeHg was predominantly controlled by both regeneration into the dissolved phase and reproduction, but the efflux of Hg(II) was mainly dominated by regeneration into the dissolved phase (Figure 8). At the low food level (i.e., 104 cells mL-1), the contribution of reproduction to MeHg loss decreased from 4 to 7 days of depuration, because the energy allocated to reproduction was greatly reduced (i.e., much smaller brood size). Nevertheless, at the higher food levels, the contribution of reproduction to MeHg loss was ∼50% from 3 to 7 days of depuration, indicating the higher energy allocated to reproduction (Figure 8). Over the 7-day depuration period, 5.2 ( 0.93, 8.0 ( 0.61, and 7.4 ( 1.25% for Hg(II) and 5.2 ( 0.04, 9.5 ( 0.46, and 11.3 ( 2.02% for MeHg were transferred from adults to neonates for 104, 5 × 104, 2 × 105 cells mL-1 food treatments, respectively. The percentage contribution of Hg(II) and MeHg loss through reproduction largely depended on the number of neonate produced by the females, and the higher percentage of efflux by reproduction was attributed to the higher number of neonates produced. Several recent studies found that the eggs of freshwater fish from different Hg-contaminated environments contained measurable maternally derived MeHg (11, 13). In the study of Hammerschmidt et al. (11), the total Hg concentrations in the eggs ranged from 12 to 298 ng g-1 (or 2.3-63 pg in individual egg). They also found that the concentrations of total Hg and MeHg in the eggs were proportional to the Hg content in the fish carcassess, suggesting that the maternal transfer of Hg may not be regulated by the female yellow perch. The average percentage of total Hg transferred to the egg was ∼1.9% (ranged from 0.3-2.3%) of the total Hg content in the parental fish body, which was relatively low when compared to highly lipophilic organic compounds such as VOL. 38, NO. 3, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

813

TABLE 3. Kinetic Parameters Used for Modeling the Exposure of Hg(II) and MeHg in D. magna (ku ) Uptake Rate Constant from Dissolved Phase, AE ) Assimilation Efficiency, BCF ) Bioconcentration Factor of Metals in Phytoplankton, IR ) Ingestion Rate) metal/parameter Hg(II) ku (L g-1 day-1) AE (%) BCF (L kg-1) IR (g g-1 day-1) MeHg ku (L g-1 day-1) AE (%) BCF (L kg-1) IR (g g-1 day-1)

range

104-105 0.1-0.5

105-106 0.1-0.5

mean

refs

8.40 37

this study this study 1, 2 42

0.25 11.04 97 0.25

this study this study 1, 2 42

FIGURE 8. Relative contribution (on a daily basis) of excretion into water, molting, fecal egestion, and neonate production in elimination of Hg(II) and MeHg by D. magna. Error bars are the standard deviations of three replicates. PCB and DDT (ranging from 5.5 to 25.5%) (34). Nevertheless, the toxicological significance of such maternally derived Hg is rather controversial. Matta et al. (35) demonstrated that maternally derived MeHg could have a significant impact on the reproduction and sex differentiation of the offspring fish of Fundulus heteroclitus. Latif et al. (12), however, found no effect of maternally derived MeHg on the hatching success and growth of larvae of walleye, and Hammerschmidt et al. (36) observed that the maternal MeHg had no observable effect on the embryos and larvae of fathead minnows. It therefore appears that the toxic effects may be different for different organisms (e.g., zooplankton vs fish) and the findings of these toxicological studies may not be applicable to D. magna and other zooplankton species. Further studies are required to examine the ecotoxicological significance of this maternally derived MeHg on the offspring of zooplankton. For example, transgenerational toxicity tests can be carried out for MeHg, using approaches similar to those of the pesticides diazinon (37) and tetradifon (38). In this study, the efflux rate constants for both Hg(II) (0.049-0.061 day-1) and MeHg (0.041-0.063 day-1) at different food levels were rather comparable (Table 2), and it appears that the reproduction contributed to the “extra” loss of MeHg. The comparable efflux rate constants of Hg(II) and MeHg were in contrast to the general concept that MeHg was eliminated at a much slower rate than Hg(II) by the aquatic animals. For example, Fowler et al. (39) and Wang and Wong (40) found that the efflux rates of MeHg were smaller than that of Hg(II) in marine fish, although others observed very slow efflux rates of both Hg species in marine mussels (41). The extra loss of MeHg due to the maternal transfer has been ignored in the past efflux studies as reproduction of aquatic organisms may not occur during the “short” experimental period. Modeling Exposure Pathways and Trophic Transfer Factor (TTF). Table 3 summarizes the kinetic parameters used for modeling the exposure pathways, taken from our present kinetic measurements and the literature. For the AE values, we used only the data obtained from the low Hg concentration in phytoplankton. Figure 9 shows the predicted 814

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 3, 2004

FIGURE 9. Model-predicted uptake of Hg(II) and MeHg due to dietary exposure by D. magna at different bioconcentration factors (BCF) in phytoplankton, as a function of ingestion rate (IR) of the animals. percentage of Hg(II) and MeHg uptake from the dietary phase at the likely range of BCFs in phytoplankton and IRs of Daphnia in the environment. The uptake of MeHg by D. magna was predominantly due to the food exposure (8198%) at intermediate to high BCFs in phytoplankton, but at the low BCF, the uptake of MeHg through food exposure was 47-81% at different IR values. On the other hand, the uptake of Hg(II) through dietary exposure was much lower than that of MeHg, with only 4-18% at low BCF and 18-69% at intermediate and high BCFs. These modeling results suggest that MeHg was mainly accumulated by D. magna through dietary exposure, whereas Hg(II) in D. magna mainly came from the aqueous phase under most natural conditions. Several studies also indicated that food exposure was important for the bioaccumulation of MeHg in a freshwater community. For example, the MeHg content in freshwater crustaceans was strongly correlated to the MeHg content in seston in 15 northern Wisconsin lakes (2). Hall et al. (7) provided the first empirical evidence that food was the major pathway for fish to accumulate MeHg. The kinetic model employed in our study can account for diverse ecological and geochemical conditions likely encountered by the macrozooplankton in the field and should be applicable to different freshwater ecosystems. The dominance of Hg uptake due to MeHg by D. magna is shown in Figure 10. The modeling results indicate that MeHg uptake predominated at a low ratio of MeHg to Hg(II). For example, MeHg contributed to 57% of the total Hg uptake by D. magna at a ratio of 0.05, a mean value of the percentage

FIGURE 10. Predicted uptake of Hg by D. magna due to MeHg at different relative ratios of MeHg to Hg(II) in the water.

that in phytoplankton in freshwater lake experiments (1). Eventually, the trophic transfer process results in the predominance of the methylated form in fish (1, 3). In conclusion, using our experimentally determined kinetic parameters for Hg(II) and MeHg in Daphnia, the kinetic model indicates that dietary exposure was important for MeHg, but water exposure was important for Hg(II) in their bioaccumulation in D. magna. Furthermore, MeHg dominates the accumulation of Hg in D. magna even at a low percentage of total Hg as MeHg. The trophic transfer factor for MeHg is >1, confirming its likelihood of biomagnification in freshwater zooplankton largely caused by its high AEs, despite the fact that its efflux rate constant was comparable to that of Hg(II). The biomagnification of Hg(II) is likely to occur under certain ecological and environmental conditions in this algae-daphnid food chain. We found that the efflux rates of both Hg(II) and MeHg were comparable, due to the increasing importance of the reproductive allocation of MeHg in the efflux system. Our study provides the first strong evidence for the role of maternal transfer for D. magna in the process of elimination of Hg(II) and MeHg from their bodies. Modeling the exposure pathways of Hg bioaccumulation and the maternal transfer to offspring has important implications for a realistic design of models to predict the biokinetics of mercury in freshwater ecosystems.

Acknowledgments

FIGURE 11. Trophic transfer factor (TTF) of Hg(II) and MeHg in D. magna as a function of ingestion rate (IR) of the animals. Only a single AE and ke values were used in the calculation (Table 3). Dashed line indicates a TTF value of 1. of MeHg in total Hg in four freshwater lakes in Wisconsin (1). Consistent with our modeling results, field measurements found that the percentage of MeHg in total Hg in Daphnia ranged from 76 to 87% at a MeHg/Hg(II) ratio of 0.17 in the dissolved phase (2). At this ratio, our model would predict that 82% of total Hg bioaccumulation in Daphnia was due to MeHg uptake. Moreover, Watras et al. (2) found that the percentage of MeHg in total Hg was relatively independent of the Hg(II) concentrations. Our modeling can therefore be used to predict the relative importance of MeHg as Hg source in the bioaccumulation in Daphnia if the percentage of MeHg in total Hg in a particular ecosystem is known. The TTFs of Hg(II) and MeHg were related to the ke, AE, and IR (Figure 11). Only a single ke value was used in the calculation. The TTF was smaller for Hg(II) than for MeHg in D. magna that had ingested Hg-contaminated freshwater green algae (Chla. reinhardtii and S. obliquus) because the AEs of Hg(II) were much smaller and the ke values were similar for both Hg(II) and MeHg. The TTF for MeHg was generally >1 under most conditions, indicating the likelihood of MeHg biomagnification from phytoplankton to zooplankton (1, 2, 43). However, the TTF for Hg(II) was unexpectedly higher than one >1 when the IR was >0.14 g g-1 day-1, showing the potential of Hg(II) to biomagnify in this algae-daphnid food chain under certain physiological and environmental conditions (e.g., at a high ingestion rate). The AEs of MeHg were much higher than those of Hg(II) for D. magna at different food carbon concentrations, but the efflux rates were rather comparable between these two Hg species. At the same food level or ingestion rate, the TTF for this freshwater algae-daphnid food chain would be much higher for MeHg. Thus, the importance of MeHg in upper trophic levels becomes greater as the TTF for MeHg was much higher than that for Hg(II). In field studies, the ratio of MeHg to total Hg in zooplankton was 4 times greater than

We thank the three anonymous reviewers for their very constructive and helpful comments on this work. We gratefully acknowledge the general technical assistances by Robert Dei, Rui Guan, and Wai-Kit Wong. This work is supported by a Competitive Earmarked Research Grant from the Hong Kong Research Grants Council (HKUST6097/02M) to W.-X.W.

Literature Cited (1) Watras, C. J.; Bloom, N. S. Limnol. Oceanogr. 1992, 37, 13131318. (2) Watras, C. J.; Back, R. C.; Halvorsen, S.; Hudson, R. J. M.; Morrison, K. A.; Wente, S. P. Sci. Total Environ. 1998, 219, 183208. (3) Bloom, N. S. Can. J. Fish. Aquat. Sci. 1992, 49, 1010-1017. (4) Wiener, J. G.; Spry, D. J. In Environmental Contaminants in Wildlife: Interpreting Tissue Concentrations; Beyer, W. N., Heinz, G. H., Redmon-Norwood, A. W., Eds.; Lewis Publishers: Boca Raton, FL, 2002; pp 42-71. (5) Boudou, A.; Ribeyre, F. In Metal Ions in Biological Systems. Vol. 34, Mercury and Its Effects on Environment and Biology; Sigel, A., Sigel, H., Eds.; Dekker: New York, 1997; pp 289-319. (6) Chen, C. Y.; Stemberger, R. S.; Klaue, B.; Blum, J. D.; Pickhardt, P. C.; Folt, C. L. Limnol. Oceanogr. 2000, 45, 1525-1536. (7) Hall, B. D.; Bodaly, R. A.; Fudge, R. J. P.; Rudd, J. W. M.; Rosenberg, D. M. Water, Air Soil Pollut. 1997, 100, 13-24. (8) Mason, R. P.; Reinfelder, J. R.; Morel, F. M. M. Water, Air Soil Pollut. 1995, 80, 915-921. (9) Boudou, A.; Ribeyre, F. Bull. Environ. Contam. Toxicol. 1981, 27, 624-629. (10) Monson, B. A.; Brezonik, P. L. Environ. Toxicol. Chem. 1999, 18, 560-566. (11) Hammerschmidt, C. R.; Wiener, J. G.; Frazier, B.; Rada, R. G. Environ. Sci. Technol. 1999, 33, 999-1003. (12) Latif, M. A.; Bodaly, R. A.; Johnston, T. A.; Fudge, R. J. P. Environ. Pollut. 2001, 111, 139-148. (13) Johnston, T. A.; Bodaly, R. A.; Latif, M. A.; Fudge, R. J. P.; Strange, N. E. Aquat. Toxicol. 2001, 52, 73-85. (14) Yu, R. Q.; Wang, W. X. Limnol. Oceanogr. 2002, 47, 495-504. (15) Wang, W. X.; Fisher, N. S. Environ. Toxicol. Chem. 1999, 18, 2034-2045. (16) Wang, W. X.; Fisher, N. S. Sci. Total Environ. 1999, 237/238, 459-472. (17) Rouleau, C.; Block, M. Appl. Organometall. Chem. 1997, 11, 751-753. (18) Guillard, R. R. L. In Culture of Marine Invertebrate Animals; Smith, W. L., Chanley, M. H., Eds.; Plenum Publishing: London, U.K., 1975; pp 29-60. VOL. 38, NO. 3, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

815

(19) Wang, W. X.; Fisher, N. S. Limnol. Oceanogr. 1996, 41, 197207. (20) Peters, R. H. In A Manual on Methods for the Assessment of Secondary Productivity in Fresh Waters; Downing, J. A., Rigler, R. H., Eds.; Blackwell Scientific: Oxford, U.K., 1984; pp 336412. (21) Wang, W. X.; Reinfelder, J. R.; Lee, B. G.; Fisher, N. S. Limnol. Oceanogr. 1996, 41, 70-81. (22) Pickhardt, P. C.; Folt, C. L.; Chen, C. Y.; Klaue, B.; Blum, J. D. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 4419-4423. (23) Taylor, G.; Baird, D. J. Environ. Toxicol. Chem. 1998, 17, 412419. (24) Yu, R. Q.; Wang, W. X. Environ. Toxicol. Chem. 2002, 21, 23482355. (25) Wang, W. X.; Stupakoff, I.; Gagnon, C.; Fisher, N. S. Environ. Sci. Technol. 1998, 32, 2564-2571. (26) Reinfelder, J. R.; Fisher, N. S.; Luoma, S. N.; Nichols, J. W.; Wang, W. X. Sci. Total Environ. 1998, 219, 117-135. (27) Mason, R. P.; Reinfelder, J. R.; Morel, F. M. M. Environ. Sci. Technol. 1996, 30, 1835-1845. (28) Mason, R. P. In Coastal and Estuarine Risk Assessment; Newman, M. C., Roberts, M. H., Jr., Hale, R. C., Eds.; Lewis Publishers: Boca Raton, FL, 2002; pp 127-149. (29) Urabe, J. Freshwater Biol. 1991, 26, 57-68. (30) Urabe, J.; Watanabe, Y. Funct. Ecol. 1991, 5, 635-641. (31) Xu, Y.; Wang, W. X. Mar. Ecol. Prog. Ser. 2001, 218, 227238.

816

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 3, 2004

(32) Goldling, G. R.; Kelly, C. A.; Sparling, R.; Loewen, P. C.; Rudd, J. W. M.; Barkay, T. Limnol. Oceanogr. 2002, 47, 967-975. (33) Moye, H. A.; Miles, C. J.; Phlips, E. J.; Sargent, B.; Merritt, K. K. Environ. Sci. Technol. 2002, 36, 3550-3555. (34) Niimi, A. J. Can. J. Fish. Aquat. Sci. 1983, 40, 306-312. (35) Matta, M. B.; Linse, J.; Cairncross, C.; Francendese, L.; Kocan, R. M. Environ. Toxicol. Chem. 2001, 20, 327-335. (36) Hammerschmidt, C. R.; Sandheinrich, M. B.; Wiener, J. G.; Rada, R. G. Environ. Sci. Technol. 2002, 26, 877-883. (37) Sanchez, M.; Ferrando, M. D.; Sancho, E.; Andreu, E. Comp. Biochem. Physiol. C 1999, 124, 247-252. (38) Villarroel, M. J.; Ferrando, M. D.; Sancho, E.; Andreu, E. Aquat. Toxicol. 2000, 49, 39-47. (39) Fowler, S. W.; Heyraud, M.; LaRosa, J. Mar. Biol. 1978, 46, 267276. (40) Wang, W. X.; Wong, R. S. K. Mar. Ecol. Prog. Ser. 2003, 261, 257-268. (41) Riisgard, H. U.; Kioboe, T.; Mohlenberg, F.; Dabaek, I.; Madsen, P. P. Mar. Biol. 1985, 86, 55-62. (42) Porter, K. G.; Gerritsen, J.; Orcutt, J. J. D. Limnol. Oceanogr. 1982, 27, 935-949. (43) Back, R. C.; Watras, C. J. Water, Air Soil Pollut. 1995, 80, 931938.

Received for review June 20, 2003. Revised manuscript received November 6, 2003. Accepted November 14, 2003. ES034638X