Influence of Humic Acid on Bioavailability and Toxicity of Benzo [k

Nov 8, 2008 - Industries, 32 Xiaoyun Road, Beijing 100027, China, and. Department of Chemistry and Biochemistry, MSN 3E2,. George Mason University ...
0 downloads 0 Views 193KB Size
Download PDF
Environ. Sci. Technol. 2008, 42, 9431–9436

Influence of Humic Acid on Bioavailability and Toxicity of Benzo[k]fluoranthene to Japanese Medaka SHAN CHEN,† RUNHUI KE,‡ J I N M I A O Z H A , † Z I J I A N W A N G , †,* A N D SHAHAMAT U. KHAN§ State Key Laboratory of Environmental Aquatic Chemistry, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, P.O. Box 2871, Beijing 100085, China, National Food Quality Supervision and Inspection Center, China National Research Institute of Food and Fermentation Industries, 32 Xiaoyun Road, Beijing 100027, China, and Department of Chemistry and Biochemistry, MSN 3E2, George Mason University, 4400 University Drive, Fairfax, Virginia 22030-4444

Received May 26, 2008. Accepted August 27, 2008.. Revised manuscript received August 25, 2008

Japanese medakas (Oryzias latipes) and triolein-embedded cellulose acetate membranes (TECAMs) were exposed simultaneously to benzo[k]fluoranthene (BkF) in the static exposure system containing different concentrations of humic acid (HA). The concentration-response relationships of induced hepatic 7-ethoxysorufin-o-deethylase (EROD) activity were established in regard to the nominal water concentration of BkF and the free concentration estimated using TECAM, as well as the body residues, respectively. In general, bioaccumulation of BkF and EROD activity in medaka were reduced with an increase of HA concentration in the exposure medium. The concentration-response relationships varied with HA concentration when expressed in nominal concentration. However, these relationships overlapped completely and partially when expressed in body BkF residue and in free BkF concentration estimated by TECAM, respectively. HA treatments were slightly beyond the 0.95 confidence band of HAfree control thereby indicating the participation of BkF-HA complex to the bioavailability and toxicity. On the basis of the bioavailability model, it was estimated that approximately 17-22% and 13-18% of BkF-HA complex contributed to the bioaccumulation and/or to the induced toxic effect, correspondingly.

Introduction The actual process of contaminants uptake into a cell varies because of the enormous diversity of organisms. However, there is one common factor among all organisms: the presence of a cellular membrane that separates the cell interior from the external environment (1). Thus it appears * Corresponding author phone: +86-010-62849140; fax: +86-01062849140; e-mail: [email protected]. † Chinese Academy of Sciences. ‡ China National Research Institute of Food and Fermentation Industries. § George Mason University. 10.1021/es8014502 CCC: $40.75

Published on Web 11/08/2008

 2008 American Chemical Society

that only the freely dissolved form of an organic contaminant can cross membranes by passive diffusion and become toxicology active after reaching its target site. Therefore, interactions with abiotic and biotic components of the environment can significantly influence the bioavailability, accumulation, and toxicity of contaminants (2-4). The magnitude of biavailability is controlled by the abiotic factors including the type and concentration of dissolved organic carbon (DOC) as well as the characteristics of the contaminants (5). To avoid uncertainties in toxicity test interpretation of a contaminant which can ultimately lead to misdirected management decisions, it is indispensable to quantify the bioavailable or free concentration of the chemical that is accessible to the target site. On the other hand the concentration at the target site and the strength of the interactions determine the toxic effect (toxicodynamics) (6). The target dose or biologically effective dose not only plays a prominent role in understanding the impact of pollutants on living organisms but also in deducing descriptive and predictive models in ecotoxicology. As target site concentrations are difficult to obtain directly, critical body residues are often used as a surrogate (7). However, in spite of the biological variability, regional restriction, and ethical disputes, direct measurement of xenobiotic body residues in organisms themselves are complicated by the presence of other bodily substances that may interfere with analyses (8). Therefore, methods of predicting the bioconcentration of hydrophobic organic chemicals by fish from water have been employed based on physicochemical partitioning between water and lipid tissues and diffusion through a series of aqueous and lipid barriers (9). Since uptake of highly hydrophobic organics in fish was mediated by passive diffusion process and the effect of the aqueous boundary layer increases with hydrophobicity, Sijm et al. established a model for hydrophobic chemicals in fish assuming that the uptake of hydrophobic chemicals is controlled by their resistance through the aqueous diffusion layer. Although the model calculations were in good agreement with experimental values, the study could not settle the high dependence of bioconcentration kinetics on variations such as species, size, and age of the organism (10). Recently several biomimetic passive sampling methods have been developed to meet the particular need for techniques to mimic the bioconcentration of aqueous organisms and measure the freely dissolved concentration of contaminants. Several devices, notably the solid phase microextraction (SPME) fibers and semipermeable membrane devices (SPMDs), appear to be very promising and effective (11, 12). These devices offer great advantages over traditional methods for estimating accumulation potential and baseline toxicity of contaminants. However, without any information on either the (intrinsic) toxicity of the compounds involved or the levels attainable in organisms, it is difficult to visualize bioavailability or the interactions (synergism, antagonism) between pollutants in complex mixtures. For this purpose, bioassay data are particularly required to determine the accordance between toxicological end points and freely dissolved concentrations. Some significant studies have been performed to link bioavailability and lethal risk. For instance, Yang et al. used the SPME methodology to predict the actual ecotoxicologic effects of pyrethroids, thereby indicating that the effect of DOC on permethrin bioavailability was toxicologically relevant (13). Escher and Hermens suggested that LC50 for the effect of hexachlorobenzene on the zebra fish was reduced by a factor of 20 based on different concentration measurements (3). VOL. 42, NO. 24, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

9431

Notwithstanding the foregoing it appears that there are limitations in using the lethality observed at higher concentrations in laboratory simulated experiments for expressing the effect of contaminant to organisms under the field conditions. Exposure to the contaminants present at trace and ultratrace levels in the soil, sediment, or natural water bodies can rarely elicit lethal risk to micro- or macro-fauna, especially vertebrates. Although few studies have been reported dealing with the quantitative inhibition of toxicity by DOC and the end point of in vivo toxic effect, there has been some concern shown in this regard by several researchers. For in vitro bioassay, serum content in the culture mediumwasreportedtoresultinlargeshiftingofdose-response curves and increasing EC50 values, while the effective data based on free concentrations measured by SPME represented the intrinsic potency of chemicals (14). In an in vivo study for exposure of fingerling rainbow trout to spiked sediments, Oikari et al. probed the influence of aging on the portion of the PAHs bioavailable and the corresponding declined extent of liver CYP1A induction (15). Also, Koganti et al. evaluated systemic bioavailability of PAH from ingested soils in mice using metabolite levels in urine and chemical:DNA adduct levels in lungs as biomarkers (16). However, quantitative descriptive models concerning bioavailability processes are lacking in these cases. To promote mechanistic understanding and predictive model development, new tools are preferred over conventional empirical approaches (1). The present work was undertaken to provide a method for measuring aqueous free concentration of contaminants using the TECAM method, and an attempt was also made to establish more reliable concentration-response relationship in aquatic organisms for quantitative estimation of toxic effect as well as risk assessment. The experimental methods were designed to predict the effect of humic acid (HA) on bioavailability/toxicity of benzo[k]fluoranthene (BkF) by comparing the measurements of biological response based on body residues to those based on TECAM extracts. Ke et al. have revealed that only the freely dissolved molecules of organic compounds pass through the membrane and are sequestered in the triolein (17). The TECAM method was used to determine the freely dissolved fraction of organic pollutants from HA-bound BkF. Bioassay was used to complement the traditional chemical analysis approach as a measure of bioavailability. The induction of CYP1A1 associated 7-ethoxysorufin-o-deethylase (EROD) activity can act as an appropriate biomarker of persistent aryl hydrocarbon receptor (AhR) agonist; thus, hepatic EROD activity was used as bioassay end point (18). In our study BkF was selected because of its relatively high tendency of bioaccumulation and potential immunotoxicity (19). Thus, a quantitative framework regarding TECAM-based method integrated with in vivo bioassay was applied for relating the chemical and biological phenomena. It was expected that the negation of the actual toxic response in living organisms could be observed in response to the inhibitory role of HA on bioavailability of hydrophobic organic chemicals. Moreover, the TECAM method was used to determine quantitatively the bioavailable portion of HA bound chemical by comparing freely dissolved fraction in TECAM and the fraction that was bioaccumulated or toxicologically effective.

Materials and Methods Chemicals and Membrane. BkF was purchased from AccuStandard (New Haven, CT) as a 1 mg/mL methanol solution. The standard solution was prepared by diluting the original stock solution with acetone to 100 mg/L. Humic acid sodium salt was purchased from Aldrich (Steinheim, Germany). A stock solution of 300 mg C/L humic acid was prepared as described previously (17). The organic carbon content of the humic acid stock solution was measured with a total organic 9432

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 24, 2008

carbon analyzer (phoenix-8000, Tekmar-Dohrmann, Mason, OH). The ultrapure water (DOC < 0.1 mg C/L) used in this study was purified by a Milli-Q Gradient system (Millipore, Bedford, MA). “Artificial freshwater” (AFW) was prepared in ultrapure water and continuously aerated before use (20). Solvent and other chemicals used in the study were of GC or analytical reagent grade. The procedure for preparing small pieces of TECAM was the same as described previously (21). Animal. The Japanese medaka (Oryzias latipes) d-rR strain was cultured and maintained for more than five generations in our laboratory before testing. Female medakas selected for bioassay were approximately 8 months post-hatch and outside of spawning period. The weights of the fish were 0.5 ( 0.1 g, and the body lengths were 3.0 ( 0.2 cm. The medakas were starved for 24 h prior to the experiment to clean the gut and exclude feces that may affect the quantity of BkF bound to HA. One medaka was introduced into 1 L of the test solution or solvent control solution (0.1 mL/L acetone) in each conical flask. The medakas were kept without feeding during the exposure under a constant light-dark photoperiod of 16:8 h, and a temperature of 25 ( 1 °C was maintained. The flasks were kept sealed with Teflon-lined caps to avoid volatilization loss of BkF, and the test solutions were renewed every 12 h. All medaka exposure experiments were carried out in six replicates, and three medakas were taken as a bioassay sample. Exposure. Fish was exposed to the sample solutions containing HA, which were produced through diluting the humic acid stock solution with AFW to reach content of 0, 5, and 10 mg C/L (AFW serving as the HA-free control), respectively. The solution was then adjusted to a pH of 7.5 ( 0.3 with 0.1 M HCl and 0.1 M NaOH. Solution was transferred to flasks and spiked with BkF to give a nominal concentration range of 0 (solvent control)-8000 ng/L. The concentration of carrier did not exceed 0.1 mL/L. The flasks were kept sealed and shaken gently on the shaker for 1 d in the dark to establish equilibrium for BkF between HA and aqueous phase (11). Meanwhile, since diverse chemical stressors could elicit similar physiological responses, HA control groups containing different levels of HA without BkF were also explored over the same exposure time frame in triplicate to investigate the disturbance which might introduced by HA. Medakas were introduced and exposed for 72 h according to the preliminary experiment (see Supporting Information), and other exposure conditions were similar as described above. It was observed that the loss of BkF after 12 h of exposure for one medaka in 1 L of aqueous solution was about 7.4%. At the end of the experiment the fish were sacrificed, and livers were carefully removed for EROD assays. First the uptake profiles of BkF in TECAM were produced in HA-free solutions (10-8000 ng/L). Subsequent experiments were carried out in the presence of HA. One piece of TECAM was immersed in each flask with BkF solutions with or without HA. The flasks in triplicate were sealed and shaken for 1 h (100 rpm, 25 °C). During this time period it was observed that TECAM accumulated less than 10% BkF of total amount in exposure solution without HA, thereby suggesting that both the depletion of TECAM and that of medaka were negligible (17). The same protocol was followed for determining the free concentrations of BkF in test solutions after medakas were exposed. Analytical Procedure. Details about sample processing, bioassay (22, 23), and chemical analysis can be found in Supporting Information. Method blanks and procedural blanks were routinely analyzed. Surrogate standard phenanthrene-d10 was added to one matrix spike sample for each batch of samples to monitor matrix effects. The average recovery for BkF was 68 ( 7% in medaka and 83 ( 5% in TECAM, respectively. Measured BkF concentrations were

corrected for the recovery of BkF matrix spikes. No BkF was detected from the blank samples. Statistics. All fits and statistics were determined using Origin-7.5 (Microcal Software, Northampton, MA). Significance was set at p < 0.05. For EROD activity, data are presented as means ( SD. One-way ANOVA followed by Turkey’s test was used to determine differences between HA control groups. All concentration-response models were regressed using the nonlinear least-squares fitter based on the Levenberg-Marquardt (LM) algorithm.

Results and Disscussion Accumulation of BkF. Although some researchers had reported that high levels of complex matrixes enhanced the diffusive mass transfer through the unstirred boundary layer adjacent to SPME (2), our previous study has shown that the uptake of TECAM is HA-independent in relatively low HA content (15 mg C/L) (17). Thereby the TECAM method can be used to determine the freely dissolved concentration of analyte in water samples in the present study. The uptake of the chemical by the TECAM in linear phase of accumulation can be simply described by the following equation: CTECAM ) Cwkut

(1)

where CTECAM is the concentration of analyte in the TECAM (ng/g), Cw is the concentration of analyte in water (ng/L), ku is the rate coefficient for uptake (L/(h g)), and t is uptake time. When HA is present, a nonlinear model according to Freidig et al. (12) could be used to describe the relationship between CTECAM and HA concentration as FTECAM )

Cfree ku′ 1 ) ) Cfree + Cbound 1 + KHACHA ku

(2)

where FTECAM is the fraction of freely dissolved analyte, KHA is a dimensionless partitioning coefficient of analyte between HA and water, CHA is the concentration of HA in water, and ku′ is defined as apparent rate coefficient for uptake determined in the presence of HA. To describe the extent of bioavailability of bound BkF, the bioavailability model is depicted as follows: Fmedaka )

Cbioavailable 1 + RKHACHA ) Cfree + Cbound 1 + KHACHA

(3)

where Fmedaka was used to represent the fraction of bioavailable BkF and R is the portion of bioavailable fraction of the bound BkF. In the study of bioaccumulation of BkF by medaka, Fmedaka can be calculated by the following equation: Fmedaka )

BAF′ BAF

(4)

with bioaccumulation factor (BAF) in AFW samples being the ratio of the concentration of contaminant in the medakas (ng/g wet weight) and nominal concentrations (ng/L) and BAF′ being the BAF expected in HA containing sample solutions. The accumulation profiles of TECAM and medaka are shown in the Supporting Information. Results showed that accumulation of BkF in medaka and TECAM were both reduced by the presence of HA. Data on the decrease of apparent rate coefficient for uptake of BkF in TECAM and bioaccumulation in medaka was fitted to eqs.2 and 4, respectively. Then fitted values of FTECAM were substituted to estimate the values of KHA, which was then substituted into eq 3 together with Fmedaka to calculate R. Concentration-Response Curves with Different Data Sets. It has been proposed that freshwater organisms react on exposure to humic substances in a similar manner as those exposed to xenobiotics (24), and some HA were observed to be able to elicit significant AhR-mediated effects

FIGURE 1. Hepatic EROD activity in relation to BkF concentrations in nominal (A), makada (B), and TECAM (C). The concentration-response curves were measured in the presence of 0 (square), 5 (triangle), or 10 (circle) mg C/L HA in sample solutions. Regression lines were fitted through the data by eq 4. The 95% confidence intervals of regression analyses without HA are indicated as the upper and lower lines in the graphs (dotted line). (25). Thus it became of interest to explore whether introduction of HA caused a bias in bioassay determinations of KHA. One-way ANOVA on the EROD result of HA controls indicated that in the present test set HA had no significant effect on the hepatic EROD activity (p ) 0.284). In AFW samples a strong response pattern could be observed. EROD levels were elevated nonlinearly with an increase in the concentrations of BkF. (Figure 1). The generally similar patterns of exposure-dependent increase in EROD activity were also shown in HA-containing exposures. The response curve patterns were similar to those reported in other published literature. In the study when medaka was treated with dioxin congeners, EROD activity declined after reaching a maximum, which was explained by the cytotoxicity occurring at the higher exposure level (26). Basu et al. (27) observed that EROD activity of BkF-exposed trout reached a plateau maximum at higher concentrations. However, since in our study BkF did not cause depressed EROD activity at the highest concentrations, both exponential and logistic models were used to fit the induction data of HA-free sample solutions. This resulted in similar shifting with increasing HA content. Thus a logistic curve fitting procedure was chosen in this study represented by equation as follows (28): y(x) ) Y2 +

(Y1 - Y2)

(5)

1 + (x/x0)p

where y(x) is EROD activity at the BkF concentration x, Y1, and Y2 are basal and maximal EROD activity, respectively, and x0 is exposure concentration at which 50% effect is VOL. 42, NO. 24, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

9433

achieved. Optimized parameters of standard curves (regressions without HA) were initialized for sequential fittings in the presence of HA. The only variable parameter P was used to describe the extent of shifting, of which the physical mean was the same as Fmedaka in eq 3. EROD activity was fitted to the model as the function of the nominal water concentration and body residue as well as free concentration of BkF, respectively. The concentration-response curves of nominal concentrations underwent a large shift to the right as the HA content increased (Figure 1A). When the body residue concentration was applied to these results, the curves became independent of the HA content instead (Figure 1B). The HA-independent fitting curve for bioassay data versus internal concentrations of BkF indicated that the speciation difference of a chemical in water did not affect the concentration-response relationship. Differences between the nominal and the actual free concentration in in vivo systems were pronounced for BkF, as expected based on theoretical considerations. However there was an unexpected nonoverlapping between the HA containing curves and the standard curve (significance ) 0.05) shown in Figure 1C. In summary, the presence of HA to a large extent depressed the EROD response induced by BkF, thereby indicating that the inhibitory role of HA on bioavailability of BkF can be expressed by negating toxic effect in medakas. Furthermore, the HA-containing curves by the TECAM method shift to the left, which mean that HA reduced BkF toxicity somehow less than predicted according to adsorption theory. Bearing in mind that HA-bound BkF might have a measurable contribution to bioavailability in acute toxicity experiments, these results must be interpreted with caution. Considering the potential effect of HA on diffusive mass transfer at the gill in medaka, it is hard to distinguish the contribution between the enhanced diffusive BkF flux through the UBL and the utility of the HA bound form. However, one feature of diffusion enhancement is that the ratio of uptake in samplers for matrix to that for pure water decreased with the exposure time (2). In other words, the uptake kinetic modes were different with or without matrix. On the contrary in our study, the bioconcentration kinetics of BkF over 12 h in medaka with different HA concentrations were in the same manner (see Figure S3 in Supporting Information), which made a clear distinction from the UBL concern. What’s more, although bound pollutant releasing occurred inevitably over the time scale in contact with the gills and gut membranes, the process of binding and releasing should be a homeostasis considering that the equilibrium between HA and BkF was not disturbed with a negligible depletion setting. Having excluded the probable interference, it was speculated that the deviation was primarily attributed to the uptake of HA-bound BkF through gastrointestinal route in medaka. The movements of a released contaminant to the membrane of an organism and contaminants still bound to the solid phase are integral fate-and-transport processes that can control an organism’s overall exposure. Nonetheless if contaminant release from the solid phase occurs internally (as in the gut lumen), fate-and-transport processes prior to uptake across a biological membrane may not be the merely factors which should be focused on (1). McCarthy and Jimenez also suggested that the sorption of benzo[a]pyrene to Aldrich humic acid was almost totally reversible (29). From this point of view it appears that a rapid desorption of bound BkF ingested by medaka would occur. Organisms in this study were not fed, but ingestion is often suggested as an extra route of uptake that does not apply to artificial samplers. Uptake of very hydrophobic compounds may also take place via routes other than simple diffusion (16). In such cases ingestion of food or sediment particles may become the predominant routes for uptake (30). 9434

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 24, 2008

TABLE 1. Summary of Results of the Regression Analysis of TECAMs and Medakas HA (mg C/L)

KHAa

FTECAM

Fmedaka

Pb

rc (%)

rd (%)

5

296356 (14931) 332139 (15018)

0.40 (0.02) 0.23 (0.02)

0.53 (0.03) 0.36 (0.03)

0.51 (0.04) 0.33 (0.02)

21.68 (6.91) 16.98 (5.04)

18.29 (7.34) 13.12 (4.13)

10

a The partition coefficient between commercial humic acid (HA) and water estimated by eq 2. b Retrieved parameters of the concentration-response relationship based on different data. c Fraction of bound BkF which was bioavailable to medaka calculated by eq 3 using Fmedaka. d Fraction of bound BkF which was bioavailable to medaka calculated by eq 3 using Pnominal. Errors are indicated in parenthesis.

FIGURE 2. Predicted versus observed EROD activities. Predicted values were obtained from regressed model in Figure 1 by the TECAM method (white circle) and body residue (black circle), respectively. The solid lines are the curves predicted by fitting the data to linear regression through the origin. The 1:1 line is indicated in the graph in the broad solid line. The results of the regressed perimeter P (Table 1) support the data observed in Figure 1. The bioavailable fraction of the total chemical contains not only the whole freely dissolved part but also some portion of the bound fraction. To test the validity of the prediction for EROD response by free fraction of BkF, a standard concentration-response model for HAfree control regressed based on BkF concentration in TECAMs was used to calculate the EROD activity from BkF concentration in TECAM of HA-containing solutions (Figure 2). The underestimation of EROD activities predicted by TECAM measurements again indicated that BkF toxicity cannot be explained by the free concentration alone. These results together with more accumulation in medaka strengthen the fact that BkF bound on HA were not completely unavailable to mekada either for uptake or for causing short-term toxicity. Bioavailability of HA-Bound BkF. Bioavailable fraction of the bound BkF, that is, the value of R, was calculated according to eq 3. The data shown in Table 1 also suggest that for every fraction of chemical bound to HA, approximately 17-22% is available for accumulation and 13-18% is available for toxic effect at the present set of HA concentrations. The bound BkF affect bioaccumulation and toxicity in medaka nearly to the same extent at the same HA level. However, it was observed that the concentration of HA in the water affect R to some extent. The values of R at 10 mg C/L HA level are lower than the corresponding contribution of bound BkF at 5 mg C/L HA. This difference is suspected due to more difficult utility of bound BkF by medaka at high HA level. The humic molecules tend to aggregate or coil at higher concentration, and the rearrangement in macromolecular structure probably hinders the access of the fast desorbing BkF to medaka (31, 32). However, it should be noted that the accumulation still depends primarily on the amount of BkF present in freely dissolved form.

Supporting Information Available Details of the analytical procedure, time course of EROD induction, uptake kinetics of BkF in TECAM and medaka, as well as accumulation of BkF by TECAM and medaka. This material is available free of charge via the Internet at http:// pubs.acs.org.

Literature Cited

FIGURE 3. Retrieval curves (dashed line) versus r corrected BkF concentrations fitted through the data with 5 mg C/L (black circle) or 10 mg C/L (white circle). Initial regression lines are indicated as solid lines (overlapping). The 95% confidence intervals of revised regression analyses are indicated as the upper and lower lines in the graphs (dotted line). For the validation of the model veracity, correction was applied to retrieve the concentration-response curves taking R into consideration (Figure 3). Free dissolved BkF reported in the form of concentration in TECAMs were compensated with the bioavailable HA-bound fraction. When the response from the bioassay tests was plotted as a function of calculated concentrations of bioavailable BkF to derive an effective concentration-response curve for the test organisms, retrieval curves of both HA levels shifted back into the confidence band. Therefore, it is important to consider the speciation of a chemical, that is, the chemical form in which the compound occurs to satisfactorily predict its transport and effect. At this juncture it cannot be concluded that there was no introduction of additional errors that have not been accounted for in the deviation. For example, the amount of test compound added to the test system can be partly “lost” due to degradation, accumulation, and adsorption to the flask wall. Nevertheless, TECAM measured concentrations as well as internal concentrations more directly reflect the intrinsic activity of BkF on medaka. In the field, however, fish can absorb PAHs from water, via body surface or gills, from contaminated sediment and food particles. Other factors that cannot be neglected during extrapolations are differences in bioavailability and speciation of chemicals between laboratory tests and real ecosystems. Additionally, reduced bioavailability of the model compounds can also be impacted by the variation in organisms and properties of DOC. DOC generally has much smaller Mw and higher polarity than soil HAs. Other investigators have shown that commercial humic acids tend to have more nonpolar moieties than natural humic substances, exhibiting significantly stronger binding ability (33). Therefore, the limitations of using Aldrich HA as the model natural/dissolved organic carbon species should be noted. Although all of these aspects may affect the ultimate bioaccumulation factor and the corresponding toxic effect of various chemicals in certain biota, the uptake of the majority of compounds is still regulated by free fraction. There is no adequate quantitative data on the nonbioavailable portion of contaminants complexed with HA and the balance of desorption between HA and BkF; more research has to be conducted to perfect the application of passive membrane sampling devices for predicting the exposure effect of biota and developing sound ecological risk assessments.

Acknowledgments The authors thank the following financial sources for making this project possible: National Basic Research Program of China (2007CB407301), Natural Science Foundation of China (20737003), and Chinese Academy of Sciences (KZCX1-YW06). We also thank Jian Li and Wei Li for valuable assistance in the laboratory.

(1) Ehlers, L. J.; Luthy, R. G. Contaminant Bioavailability in Soil and Sediment. Environ. Sci. Technol. 2003, 37 (15), 265A–304A. (2) Oomen, A. G.; Mayer, P.; Tolls, J. Nonequilibrium solid-phase microextraction for determination of the freely dissolved concentration of hydrophohic organic compounds: Matrix effects and limitations. Anal. Chem. 2000, 72 (13), 2802–2808. (3) Escher, B. I.; Hermens, J. L. M. Internal exposure: Linking bioavailability to effects. Environ. Sci. Technol. 2004, 38 (23), 455A–462A. (4) Hermens, J. L. M.; Ankley, G. T.; Sumpter, J. P. Ecotoxicology-A multidisciplinary, problem-driven science. Environ. Sci. Technol. 2004, 38 (23), 446A–447A. (5) Akkanen, J.; Tuikka, A.; Kukkonen, J. V. K. Comparative sorption and desorption of benzo[a]pyrene and 3,4,3′,4′-tetrachlorobiphenyl in natural lake water containing dissolved organic matter. Environ. Sci. Technol. 2005, 39 (19), 7529–7534. (6) Escher, B. I.; Hermens, J. L. M. Modes of action in ecotoxicology: their role in body burdens, species sensitivity, QSARs, and mixture effects. Environ. Sci. Technol. 2002, 36 (20), 4201–4217. (7) Verhaar, H. J. M.; Busser, F. J. M.; Hermens, J. L. M. Surrogate parameter for the baseline toxicity content of contaminated water: simulating the bioconcentration of mixtures of pollutants and counting molecules. Environ. Sci. Technol. 1995, 29 (3), 726–734. (8) Leslie, H. A.; ter Laak, T. L.; Busser, F. J. M.; Kraak, M. H. S.; Hermens, J. L. M. Bioconcentration of organic chemicals: Is a solid-phase microextraction fiber a good surrogate for biota. Environ. Sci. Technol. 2002, 36 (24), 5399–5404. (9) Kwon, J. H.; Katz, L. E.; Liljestrand, H. M. Use of a parallel artificial membrane system to evaluate passive absorption and elimination in small fish. Environ. Toxicol. Chem. 2006, 25 (12), 3083– 3092. (10) Sijm, D. T. H. M.; van der Linde, A. Size-dependent bioconcentration kinetics of hydrophobic organic chemicals in fish based on diffusive mass transfer and allometric relationships. Environ. Sci. Technol. 1995, 29 (11), 2769–2777. (11) Akkanen, J.; Kukkonen, J. V. K. Measuring the bioavailability of two hydrophobic organic compounds in the presence of dissolved organic matter. Environ. Toxicol. Chem. 2003, 22 (3), 518–524. (12) Freidig, A. P.; Garicano, E. A.; Busser, F. J. M.; Hermens, J. L. M. Estimating impact of humic acid on bioavailability and bioaccumulation of hydrophobic chemicals in guppies using kinetic solid-phase extraction. Environ. Toxicol. Chem. 1998, 17 (6), 998–1004. (13) Yang, W.; Spurlock, F.; Liu, W.; Gan, J. Effects of dissolved organic matter on permethrin bioavailability to Daphnia species. J. Agric. Food Chem. 2006, 54 (11), 3967–3972. (14) Heringa, M. B.; Schreurs, R. H.; Busser, F.; van der Saag, P. T.; van der Burg, B.; Hermens, J. L. M. Toward more useful in vitro toxicity data with measured free concentrations. Environ. Sci. Technol. 2004, 38 (23), 6263–6270. (15) Oikari, A.; Fragoso, N.; Leppanen, H.; Chan, T.; Hodson, P. V. Bioavailability to juvenile rainbow trout (Oncorynchus mykiss) of retene and other mixed-function oxygenase-active compounds from sediments. Environ. Toxicol. Chem. 2002, 21 (1), 121–128. (16) Koganti, A.; Spina, D. A.; Rozett, K.; Ma, B. L.; Weyand, E. H.; Taylor, B. B.; Mauro, D. M. Studies on the applicability of biomarkers in estimating the systemic eioavailability of polynuclear aromatic hydrocarbons from manufactured gas plant tar-contaminated soils. Environ. Sci. Technol. 1998, 32 (20), 3104–3112. (17) Ke, R; Wang, Z; Huang, S; Khan, S. U. Accurate quantification of freely dissolved organochlorine pesticides in water in the presence of dissolved organic matter using triolein-embedded cellulose acetate membrane. Anal. Biochem. Chem. 2007, 387 (8), 2871–2879. (18) White, K. L.; Kawabata, T. T.; Ladics, G. S. Mechanisms of polycyclic aromatic hydrocarbon immunotoxicity. In ImmunoVOL. 42, NO. 24, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

9435

(19)

(20) (21)

(22)

(23) (24) (25) (26)

9436

toxicology and Immunopharmacology, 2nd ed; Dean, J. H., Luster, M. I., Munson, A. E., Kimber, I., Eds.; Raven Press: New York, 1994. Willett, K. L.; Gardinali, P. R.; Sericano, J. L.; Wade, T. L.; Safe, S. H. Characterization of the H4IIE rat hepatoma cell bioassay for evaluation of environmental samples containing polynuclear aromatic hydrocarbons (PAHs). Arch. Environ. Contam. Toxicol. 1997, 32 (4), 442–448. Haitzer, M.; Akkanen, J.; Steinberg, C.; Kukkonen, J. V. K. No enhancement in bioconcentration of organic contaminants by low levels of DOM. Chemosphere 2001, 44 (2), 165–171. Xu, Y.; Wang, Z.; Ke, R.; Khan, S. U. Accumulation of organochlorine pesticides from water using triolein embedded cellulose acetate membranes. Environ. Sci. Technol. 2005, 39 (4), 1152– 1157. Ma, T.; Wan, X.; Huang, Q.; Wang, Z.; Liu, J. Biomarker responses and reproductive toxicity of the effluent from a Chinese large sewage treatment plant in Japanese medaka (Oryzias latipes). Chemosphere 2005, 59 (2), 281–288. Braford, M. M. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 1976, 72 (1), 248–254. Pflugmacher, S.; Tidwell, L. F.; Steinberg, C. E. W. Dissolved humic substances can directly affect freshwater organisms. Acta Hydrochim. Hydrobiol. 2001, 29 (1), 34–40. Janosˇek, J.; Bittner, M.; Hilscherova, K.; Blaha, L.; Giesy, J. P.; Holoubek, I. AhR-mediated and antiestrogenic activity of humic substances. Chemosphere 2007, 67 (6), 1096–1101. Chen, C. M.; Cooper, K. R. Developmental toxicity and EROD induction in the Japanese medaka (Oryzias latipes) treated with

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 24, 2008

(27)

(28)

(29)

(30)

(31) (32) (33)

dioxin congeners. Bull. Environ. Contam. Toxicol. 1999, 63 (4), 423–429. Basu, N.; Billiard, S.; Fragoso, N.; Omoike, A.; Tabash, S.; Brown, S.; Hodson, P. Ethoxyresorufin-O-deethylase induction in trout exposed to mixtures of polycyclic aromatic hydrocarbons. Environ. Toxicol. Chem. 2001, 20 (6), 1244–1251. Kennedy, S. W.; Lorenzen, A.; James, C. A.; Collins, B. T. Ethoxyresorufin-O-deethylase and porphyrin analysis in chicken embryo hepatocyte cultures with a fluorescence multiwell plate reader. Anal. Biochem. 1993, 211 (1), 102–112. McCarthy, J. F.; Jimenez, B. D. Interactions between polycyclic aromatic hydrocarbons and dissolved humic material: Binding and dissociation. Environ. Toxicol. Chem. 1985, 19 (11), 1072– 1076. Van Loon, W. M. G. M.; Verwoerd, M. E.; Wijnker, F. G.; van Leeuwen, C. J.; van Duyn, P.; van de Guchte, C.; Hermens, J. L. M. Estimating total body residues and baseline toxicity of complex organic mixtures in effluents and surface waters. Environ. Toxicol. Chem. 1997, 16 (7), 1358–1365. Ghosh, K; Schnitzer, M. Macromolecular structures of humic substances. Soil Sci. 1980, 129 (5), 266–276. Engebretson, R. R.; von Wandruszka, R. Microorganization in dissolved humic acids. Environ. Sci. Technol. 1994, 28 (11), 1934– 1941. Chin, Y. P.; Aiken, G. R.; Danielsen, K. M. Binding of pyrene to aquatic and commercial humic substances: the role of molecular weight and aromaticity. Environ. Sci. Technol. 1997, 31 (6), 1630–1635.

ES8014502