Environ. Sci. Technol. 2004, 38, 1003-1008
Cadmium Bioavailability and Accumulation in the Presence of Humic Acid to the Zebra Mussel, Dreissena polymorpha JUDITH VOETS,* LIEVEN BERVOETS, AND RONNY BLUST Ecophysiology, Biochemistry and Toxicology Group, Department of Biology, University of Antwerp, Groenenborgerlaan 171, B-2020 Antwerp, Belgium
Metal speciation in aquatic systems is mainly determined by the type and concentration of ligands present in solution. A very important group of complexing agents is dissolved organic matter (DOM), e.g., humic and fulvic acids. According to the free-ion activity model, only the free metal ion is available to biota. Nevertheless, DOM has been reported to decrease or increase metal uptake, leading to uncertainty concerning the bioavailability of metal-DOM complexes. In this work the effect of Aldrich humic acid on cadmium accumulation by the zebra mussel, Dreissena polymorpha, was studied under laboratory conditions. Mussels, collected in a drinking water reservoir, were exposed to varying environmentally relevant concentrations of cadmium in the presence and absence of humic acid. Cadmium concentrations in the mussel tissues were analyzed, and measurements with a cadmium-ion-selective electrode were made to determine the free cadmium ion activity in the exposure waters. The uptake of humic acid by the zebra mussels was measured by the decrease of the total organic carbon (TOC) concentration in the water over time. The free cadmium ion activity in the water decreased from 51.6% to 19.9% of the total cadmium concentration in the presence of humic acid. This decrease by a factor of 2.6 resulted in a decrease in the cadmium uptake rate in the soft tissue of zebra mussels from 12.9 to 7.9 nmol/g dry wt/day, which corresponds to a decrease by a factor of 1.6. This implies that cadmium uptake rates were higher than predicted by the free-ion activity model and indicates that cadmium-humic acid complexes are partly available to zebra mussels.
Introduction The toxicity of metals in natural waters is greatly dependent on their chemical speciation, and therefore, generally the total metal concentration is a poor predictor of its biological impact (1-3). According to the free-ion activity model (FIAM) for metal-organism interactions in a system at equilibrium, the metal-ion activity reflects the chemical reactivity of the metal, which determines the interaction of the metal with cellular surface sites and hence its bioavailability (4). The free-ion activity of a metal is determined by the total dissolved metal concentration and by the physicochemical charac* Corresponding author phone: 32 321 80349; fax: 32 321 80497; e-mail:
[email protected]; website: www.ecotox.be. 10.1021/es034742e CCC: $27.50 Published on Web 01/08/2004
2004 American Chemical Society
teristics of the water, such as water hardness, pH, and complexation capacity (5-8). The type and concentration of the inorganic and organic ligands present in the solution have an important influence on the free metal-ion activity (9-11). In natural aquatic systems, part of the metals is present as metal complex, and therefore, in most cases their bioavailability and toxicity to biota is lowered (12-14). A very important group of complexing agents in natural aquatic ecosystems, which is assumed to decrease the bioavailability and toxicity of metals, is dissolved organic matter (DOM) (15-17). The most important components of DOM are humic and fulvic acids, which are often present at concentrations many orders of magnitude higher than that of the trace metals (15, 18, 19). Humic and fulvic acids are complex aromatic macromolecules with a wide variety of functional groups, mainly carboxylic and phenolic groups and to a lesser extent amino and sulfydryl groups (10). Cadmium binds to all these functional groups, although the binding strength between cadmium and the nitrogen- and sulfur-containing functional groups is higher, and therefore, all these functional groups play an important role in the complexation characteristics of DOM for cadmium (20, 21). Varying and even controversial effects of dissolved humic material on cadmium bioavailability and toxicity are reported in the literature. The bioavailability of cadmium in the presence of DOM depends on the organism under investigation and the kind and concentration of DOM. For some species the bioaccumulation or toxicity of cadmium was found to be in accordance with the FIAM (14, 22-25). On the other hand, in some studies concerning bivalve molluscs, the accumulation of cadmium increased in the presence of DOM (26-28). George and Coombs found that humic materials increased both the rate of uptake and the final body concentration of Cd in the marine bivalve, Mytilus edulis (26). Hung found a decreased cadmium accumulation in the presence of humic acid in the marine bivalve, Crassostrea virginica, after an exposure of 40 days (29). Roditi et al. found that absorption of dissolved cadmium, silver, and mercury by Dreissena polymorpha increased 32-, 8.7-, and 3.6-fold, respectively, in the presence of high-molecular-weight dissolved organic carbon (DOC) compared to low-molecular weight DOC (30). Most of these studies, however, followed the uptake for only 2-8 h, and long-term accumulation studies are scarce (26, 29). In most cases, no information is provided concerning the free cadmium-ion activity. For freshwater mollusks, very little is known about the effect of humic acid on metal bioavailability (30). The aim of this study was to investigate the effect of high but environmentally realistic concentrations of humic acid on the relative long-term accumulation of Cd in the freshwater mussel D. polymorpha under controlled laboratory conditions. We assessed whether the long-term uptake of Cd is in agreement with the free-ion activity model and if Cd accumulation is related to the Cd2+-ion activity in the water.
Material and Methods Test Organism. Zebra mussels (D. polymorpha) were collected in December 2001 at a drinking water reservoir of the Antwerpse Waterwerken (AWW) in Duffel (Belgium), selected by length (19-25 mm), and cleaned. The mussels were acclimated in the laboratory for 28 days at 15.0 ( 0.5 °C in acryl tanks filled with artificial freshwater and were subjected to a 16/8 light/dark period. Water was filtered using a closed filter system (type EHEIM 2213) and continuously aerated. The artificial water was prepared by dissolving the analyticalgrade reagents (Union Chemique Belge; UCB) CaCl2‚2H2O VOL. 38, NO. 4, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1003
(294 mg/L), MgSO4‚7H2O (123 mg/L), NaHCO3 (64.8 mg/L), and KCl (5.8 mg/L) in deionized water. During acclimation the mussels were held at a density of six individuals per liter and fed yeast cells (20 000 cells/L; Lansy PZ, INVE) every 2 days. The ammonium, nitrite, and nitrate concentration in the water was tested twice a week, and the water was renewed when the concentration reached 2, 0.25, and 20 mg/L, respectively. Ion-Selective Electrode Potentiometry. To determine the free cadmium-ion activity in the Cd exposure solutions, the effect of humic acid (HA) on the free cadmium-ion activity was measured at different humic acid and cadmium concentrations. The humic acid solution used in this study was prepared with commercially available purified solid material from Aldrich (Germany). Aldrich humic acid and aqueous humic acid have a similar 13C NMR spectrum, displaying broad resonances in the aliphatic (0-105 ppm) and aromatic regions (105-165 ppm), which points to comparable functional groups. Furthermore, the cadmium binding properties of Aldrich humic acid are similar to that of aqueous humic acids (10, 11). The humic acid stock solution was prepared by dissolving 1 g of Aldrich humic acid in 1 L of deionized water (Millipore). This suspension was adjusted to pH 10 with KOH, stirred for 24 h, and centrifuged at 5000g for 1 h in a Sorval RC-5 centrifuge (Du Pont). The supernatant solution was filtered through a 0.45 µm membrane filter (Millipore) to remove particulates (11). The obtained humic acid stock solution of 1 g of HA/L was measured with a total organic carbon analyzer (Shimadzu TOC-5000) and contained 498.6 ( 3.7 mg of C/L (n)3). The cadmium standard solutions were prepared with Cd(NO3)2 and NaNO3 and adjusted to pH 6.0 with HNO3. NaNO3 (8.4 mM) was used to adjust the ionic strength to that of the artifical water. In the first experiment the effect of different humic acid concentrations on the free cadmium-ion activity was determined. The sample solutions were prepared with reagent-grade CdCl2 (Sigma Chemicals) in artificial water with a salt composition as described earlier. A series of sample solutions were prepared with cadmium concentrations varying from 0.01 to 500 µM and humic acid concentrations varying from 10 to 75 mg of HA/L. Cadmium-ion activities were measured with a cadmium(II)-ion-selective electrode (Cd(II)-ISE, model no. 9448SC, Orion Co.) and a doublejunction reference electrode (model no. 900200) coupled to an ion meter (Metrohm 654). The electrode responded linearly in the range 0.3-500 µM Cd with a Nernstian slope of -27.3 mv. All the Cd2+ ion activity measurements were performed in a climate room at 15.0 ( 0.5 °C. Humic acid also binds calcium, and this may influence the competition between cadmium and calcium at the biological uptake sites and as a result influence Cd uptake. Therefore, additional measurements with a calcium-ionselective electrode (Ca(II)-ISE model 93-20) were performed to assess the effect of humic acid on the free calcium-ion activity in the exposure water. The solutions were prepared in artificial water with 53.3 and 56.4 mg of HA/L and varying concentrations of cadmium. Standard solutions were prepared with reagent-grade CaCl2, and the ionic strength was adjusted to 8.4 mM. Cadmium Accumulation Experiment. The experiments were run in a climate room at 15.0 ( 0.5 °C with a 16/8 light/dark period. Zebra mussels were exposed in 80 L of water for 31 days. Each aquarium contained 150 individuals at the start of the experiment. Five experimental aquaria and one control experiment were run: (1) 0.3 µM Cd, (2) 0.3 µM Cd + 56.44 mg of HA/L, (3) 0.1 µM Cd, (4) 0.1 µM Cd + 53.3 mg of HA/L, (5) 0.033 µM Cd, (6) control group without Cd and HA. A humic acid concentration of 50-60 mg of HA/L is a high but environmentally relevant level. Natural aqueous 1004
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 4, 2004
concentrations of dissolved humic material (DHM) range from 1 to 70 mg of DHM/L (10, 18, 19). At these humic acid concentrations, the free cadmium-ion activity is strongly decreased and Cd-HA complexes contributed significantly to the Cd speciation. Therefore, these conditions give the opportunity to observe even small deviations from the FIAM. The Cd speciation data used for the setup of the Cd accumulation experiment are based on measurements of the free cadmium-ion activity at different humic acid concentrations with the Cd(II)-ISE. The free cadmium-ion concentrations in the accumulation experiment were below the regular detection limit of the Cd(II)-ISE, and therefore, measurements of the cadmium-ion activity in the presence of humic acid were performed at higher concentrations. After log transformation, the relation between [Cd2+] and [CdT] was linear in the presence of a fixed concentration of humic acid. The linear equation was used to calculate [Cd2+] at lower concentrations. The pH of the exposure solutions was adjusted daily to the desired test pH (7.6) using analytical-grade HCl or NaOH and a pH glass electrode (Ingold 104573002). The resulting pH ranged from 7.6 to 7.8. During the experiments, the animals were fed every 4 days with yeast cells (20 000 cells/ L). Previous experiments proved that this feeding regime is sufficient to maintain the animals in good condition. The water was continuously aerated. Every 2 days the ammonium, nitrite, and nitrate concentration in the water was tested, and the water was renewed weekly or when the concentration reached 2, 0.25, and 20 mg/L, respectively (after 6, 14, and 21 days). Water samples for metal analyses were taken every 2 days. At the start of the experiment and after 1, 3, 7, 10, 14, 18, 21, 24, and 31 days, 10 individuals from each aquarium were sampled and dissected. Sample Preparation and Metal Analyses. Wet tissues were rinsed with ultrapure water (Milli-Q), transferred to preweighed polypropylene vials, and dried for 24 h at 60 °C. Subsequently, the dry weight was determined and the biological material digested in a 5:1 mixture of HNO3 and H2O2 using a microwave digestion procedure as described by Blust et al. (31). Water samples taken during the Cd accumulation experiment and water samples of the Cd solutions, prepared for the measurements of the Cd2+ activity, were acidified to 1% with HNO3. Cd levels in all samples were determined by graphite furnace atomic absorption spectrophotometry (GF-AAS, Varian SpectrAA-800). Metal concentrations are expressed as µmol of Cd per gram dry weight (µmol of Cd/g dry wt). Analytical accuracy was determined using process blanks and certified reference material of the Community Bureau of Reference standard for trace elements in mussel tissue (CRM 278). Recoveries were within 10% of the certified values. Consumption of Humic Acid by Zebra Mussels. To determine whether humic acid is available as a food source to zebra mussels, the change in humic acid concentration, measured as total organic carbon (TOC), was determined as a function of time in the presence of zebra mussels. The mussels were acclimated in the laboratory and starved for 4 days. In this period the water was renewed daily until no traces of faeces were visible in the aquarium. To correct for the potential loss of humic acid from the water, due to adsorption of humic acid to the mussel shells or the plastic containers, and/or loss as CO2, three exposure conditions were run in triplicate. All the exposure conditions contained 600 mL of HA solution containing artificial freshwater with humic acid (18.2 mg of TOC/L) and with Cd (0.3 µmol). The first exposure condition contained HA solution with six zebra mussels. The second and third exposure conditions were controls and contained, respectively, HA solution with shells of six zebra mussels and only HA solution, without mussels or shells. The TOC concentration in the water was measured
TABLE 2. Total Cd Concentrations (n ) 7) and Calculated Cd2+ Ion Activities in the Exposure Watersa
TABLE 1. Summary of the Regression Parameters for the Linear Relations between log(Cd2+) and log(CdT) at Humic Acid Concentrations of 10 to 75 mg/La
group
y ) ax ( b
CdT
Cd2+
Cd2+
Cd (µm) HA (mg/L) mean ( SD (µm) % of CdT mean + SD (µm)
mg/L HA
a ( SE
b ( SE
r2
n
10 20 50 75
1.018 + 0.011 1.045 ( 0.009 1.131 ( 0.010 1.172 ( 0.017
-0.341 + 0.007 -0.382 ( 0.005 -0.529 ( 0.012 -0.675 ( 0.019
0.999 0.999 0.999 0.999
4 4 4 4
a The regression line y ) ax + b is a fitting of the Cd2+-ion activity at the different CdT. The Cd2+ ion activity and the CdT are obtained from measurements with the Cd-ISE and the GF-AAS.
FIGURE 1. Free Cd-ion activity as a function of humic acid concentration at 0.072 and 0.218 µM CdT in water. The data points are calculated with the linear regressions presented in Table 1. The nonlinear regressions are used to calculate the free Cd ion activity in the exposure water at 53.3 and 56.44 mg of HA/L. every 2-8 h with a TOC-analyzer (Shimadzu TOC 5000) for 45 h. Statistics. Analyses of variance (ANOVA, with post-hoc Duncan’s multiple range test), Friedman tests, unpaired t-tests, and linear and nonlinear regressions were used to analyze the data, as appropriate. If necessary, data were tested for homogeneity of variance by the log-anova test and for normality by the Kolmogorov-Smirnov test for goodness of fit. All tests were performed with STATISTICA 5.0 (StatSoft, Inc.). Data were considered statistically different when p-values were
