Biotransformation of N-Ethyl Perfluorooctanesulfonamide by

The mobile phase system used consisted of water (A) and methanol (B), both of which ... in the negative ion ES mode using multiple reaction monitoring...
0 downloads 0 Views 64KB Size
Download PDF
Environ. Sci. Technol. 2004, 38, 758-762

Biotransformation of N-Ethyl Perfluorooctanesulfonamide by Rainbow Trout (Onchorhynchus mykiss) Liver Microsomes G R E G G T . T O M Y , * ,†,‡ SHERYL A. TITTLEMIER,§ V I N C E P . P A L A C E , †,| WES R. BUDAKOWSKI,† ERIC BRAEKEVELT,† LYNDON BRINKWORTH,† AND KEN FRIESEN⊥ Department of Fisheries and Oceans, Winnipeg, Manitoba R3T 2N6, Canada, Departments of Chemistry and Zoology, University of Manitoba, Winnipeg, Manitoba R3T 2N6, Canada, Food Research Division, Health Canada, Ottawa, Ontario K1A 0L2, Canada, and Department of Chemistry, University of Winnipeg, Winnipeg, Manitoba R3B 2E9, Canada

Rainbow trout (Onchorhynchus mykiss) liver microsomes were incubated with N-ethyl perfluorooctanesulfonamide [N-EtPFOSA, C8F17SO2NH(C2H5)], to examine the possibility of in vitro biotransformation to perfluorooctane sulfonate (PFOS, C8F17SO3-) and perfluorooctanoate (PFOA, C7F15COO-). Incubations were performed by exposing trout liver microsomes to N-EtPFOSA at 8 °C in the dark. Reaction mixtures were analyzed after incubation periods of 0, 2, 4, 8, 16, and 30 h for N-EtPFOSA, PFOS, PFOA, and perfluorooctanesulfonamide (PFOSA, C8F17SO2NH2), a suspected intermediate. Amounts of PFOS and PFOSA were found to increase with incubation time, but only background levels of PFOA were detected. Three possible reaction pathways are proposed for the conversion of N-EtPFOSA to PFOS: (i) direct conversion of N-EtPFOSA to PFOS by deethylamination accompanied by conversion of the sulfone group to sulfonate, (ii) deethylation of N-EtPFOSA to PFOSA, followed by deamination to form PFOS, and (iii) direct hydrolysis of N-EtPFOSA. These findings represent the first report indicating a possible biotransformation of a perfluorosulfonamide to PFOS in fish and may help to explain the detection of PFOS, which is relatively involatile, and thus not likely to undergo atmospheric transport, in biota from remote regions.

Introduction Sulfonyl-based fluorinated organic compounds (FOCs) are a diverse group of chemicals used in a variety of specialized consumer and industrial products (1). Many of these compounds are used in the textile, paper, and packaging * Corresponding author phone: (204) 983-5167; fax: (204) 9842403; e-mail: [email protected]. † Department of Fisheries and Oceans. ‡ Department of Chemistry, University of Manitoba. § Health Canada. | Department of Zoology, University of Manitoba. ⊥ University of Winnipeg. 758

9

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

industries, where they are applied to products and raw materials to form stain-resistant coatings (2-4). Some FOCs are also used as fire retardants and insecticides (1). Although FOCs have been manufactured for over 50 years, there are no published reports on their total production volumes: total production estimates for the sulfonyl-based FOCs were 3 × 106 kg in the year 2000 (4-6). The perfluorinated acids are a group of FOCs that has attracted much interest. Of these, perfluorooctanesulfonate (PFOS, C8F17SO3-) and perfluorooctanoate (PFOA, C7F15COO-) have recently received the most attention. Much of the concern surrounds the ubiquitous presence of both compounds in the environment. PFOS and PFOA have been detected in human serum (7), freshwater and marine biota (5, 8, 9), and surface water (10, 11). Their high chemical and biological stability appears to preclude any degradation or metabolism and contributes to the bioaccumulation and persistence of PFOS and PFOA. The detection of PFOS and PFOA in remote regions such as the Canadian Arctic (8) seems to contradict what is known about their physical chemical properties. PFOS, for example, is relatively soluble in water (critical micelle concentration 300-700 mg/L), is completely ionized in water, and has a low vapor pressure [3.3 × 10-4 Pa (20 °C)] (5, 12), whereas other persistent organic contaminants found in the Arctic tend to have very low water solubilities and higher vapor pressures. Because PFOS appears to be an unlikely candidate for long-range atmospheric transport, there has been speculation on the existence of precursors that are more environmentally mobile, and which subsequently degrade to PFOS (5, 13, 14). The perfluorosulfonamide group of FOCs is used in some pesticide formulations, as surfactants, and as intermediates in the synthesis of other FOCs (1, 2, 15). For example, N-ethyl perfluorooctanesulfonamide [N-EtPFOSA, C8F17SO2NH(C2H5), I] commonly known as Sulfluramid, is

an insecticide used to control cockroaches, termites, and ants (1, 16). It has recently been detected in air samples from Toronto, Ontario (14), and in biota from a marine food web from the Canadian Arctic (17) and is suspected of being a precursor of PFOS (14). The toxicokinetics of N-EtPFOSA has been examined in terrestrial mammals (16, 18). In rats and dogs, N-EtPFOSA underwent deethylation to form I. Both the parent and the metabolite compound were found to be potent uncouplers of oxidative phosphorylation (1). Unfortunately, no attempts were made in these studies to follow the formation of any anionic metabolites. Biotransformation of N-EtPFOSA to PFOS and PFOA in fish will confirm that there are neutral precursors of these compounds. It will also support the hypothesis that the presence of PFOS and PFOA in remote and less industrialized areas may be due to long-range transport of volatile precursors, by describing a means for the transformation of precursor to anionic products. To examine whether N-EtPFOSA is a precursor of PFOS and PFOA in fish, an in vitro incubation experiment was performed using liver microsomes from rainbow trout (Onchorhynchus mykiss). The starting compound and products of biotransformation were monitored by high-perfor10.1021/es034550j CCC: $27.50

 2004 American Chemical Society Published on Web 12/03/2003

mance liquid chromatography electrospray tandem mass spectrometry (HPLC-ES-MS/MS).

Experimental Section Standards and Reagents. PFOA, the tetraethylammonium salt of PFOS, perfluorobutane sulfonate (PFBS), tetrabutylammonium (TBA) hydrogen sulfate, sodium hydroxide, sodium carbonate, nicotinamide adenine dinucleotide phosphate (NADPH), and Tris-HCl buffer were obtained from Sigma-Aldrich Canada Ltd. (Oakville, ON, Canada). NEtPFOSA (85% purity) was purchased from Interchim (Montlucon, France). PFOSA was a generous gift from Dr. Natalia Vais (Health Canada, Ottawa, ON, Canada). Optima grade methanol and water and reagent grade methyl tertbutyl ether (MTBE) were purchased from Caledon Laboratories Ltd. (Georgetown, ON, Canada). Liver Microsome Isolation and In Vitro Incubation. Microsomes were prepared from the liver of a single rainbow trout (Onchorhynchus mykiss). Liver tissue and concentrations of microsomal protein were determined as described by Shields and Eales (19). The potential for microsomal transformation of N-EtPFOSA was examined by adding 10 µL of its solution of N-EtPFOSA in HPLC grade methanol (1 ng/µL) to a polypropylene microcentrifuge tube containing the following: 890 µL of 0.05 M Tris-HCl buffer (pH ) 7.0) with 0.1 M NaCl, 100 µL of buffer containing 1 mg/mL of NADPH, and 100 µL of microsomes (total protein ) 0.2 ( 0.01 mg). Thirty microcentrifuge tubes containing microsomes reaction mixtures were prepared in this manner. The tubes were incubated in the dark at 8 ( 1 °C to exclude the possibility of phototransformation. Experiments were conducted at 8 °C to reflect temperatures most likely encountered by wild rainbow trout. Since reaction rates vary with temperature, the temperature chosen should therefore produce results that reflect relevant transformation rates. At selected times [0, 2, 4, 8, 16, and 30 h] reactions were terminated (five replications for each of the six reaction times) by adding 0.5 mL of HPLC grade methanol (Sigma Chemical Co., St. Louis, MO) to each tube and vortexing it for 1 min at high speed. The methanol-terminated reaction suspension in each tube was extracted as described below. Incubation Blanks. Two types of experimental blanks were used in the incubation. One consisted of N-EtPFOSA and heat-treated microsomes (i.e., microsomes heated at 100 °C for 5 min in a water bath to render them inactive) plus all of the reaction components. This blank would correct for any abiotic transformation of N-EtPFOSA. The other blank contained microsomes and all of the reaction components without N-EtPFOSA. Each of the two blanks were collected at each reaction time, quenched, and analyzed as described below. Extraction Procedure. The extraction of FOCs from collected aliquots was done in a manner similar to that described by Hansen et al. with small modifications (20). Prior to extractions, samples and blanks were spiked with 500 pg (5 µL of a 100 pg/µL solution) of the recovery standard PFBS. PFOS and PFOA concentrations were recoverycorrected on the basis of the recovery of PFBS. PFBS was used as a surrogate standard because it has the same functional groups as PFOS and was not present at detectable levels in solvent blanks. While perfluorononanoic acid (PFNA) or a similar compound would be better than PFBS as a surrogate for PFOA, the presence of this and other perfluorinated acids in the blanks led us to use PFBS as the surrogate. Further, PFOS recoveries were also checked by spiking a known amount of PFOS into Optima grade water and extracting according to Hansen et al. (20). No suitable recovery standard for the perfluorosulfonamides exists, and the recoveries of N-EtPFOSA and PFOSA from the incubation

samples were corrected on the basis of the recovery of N-EtPFOSA and PFOSA that were spiked into Optima grade water and taken through all phases of the extraction procedure. To monitor the potential for contamination to occur during extraction, five method blanks (hereafter referred to as extraction blanks) consisting of Optima grade water were extracted along with the samples. Separation and Quantitation. The liver microsome extracts were chromatographed on a Supelcosil C8 analytical column (5.0 cm × 2.1 mm i.d., 5 µm particle size; Supelco, Oakville, ON, Canada). The analytical and C8 guard columns (Phenomenex) were installed on an Agilent 1100 Series HPLC system (Agilent Technologies, Palo Alto, CA) equipped with a vacuum degasser, binary pump, and autosampler. The mobile phase system used consisted of water (A) and methanol (B), both of which contained 2 mM ammonium acetate. The flow rate was 300 µL/min, and the injection volume was 3 µL. The gradient employed started at 20% B, increasing to 95% B in 9.5 min, and was held for 2 min. Thereafter the mobile phase composition was returned to starting conditions in 5 min. The column was allowed to equilibrate for 5 min between runs. A blank methanol solution was run twice after each sample or standard using the same elution profile. Due to the known tendency of fluorinated compounds to contaminate systems through carryover, after every 10-15 samples, the HPLC-ES-MS/MS system was rinsed with methanol containing 75 mM ammonium acetate for several hours. Analyses were performed with a Sciex API 2000 triple quadrupole mass spectrometer (MDS Sciex, Ontario, Canada) in the negative ion ES mode using multiple reaction monitoring (MRM). The optimized parameters were as follows: ionspray voltage, -1200 V; curtain gas flow, 15.00 arbitrary units (au); sheath gas flow, 30.00 au; turbo gas flow, 35.00 au; temperature 525 °C; focusing potential, -360 V; collision-assisted dissociation gas flow, 8 au. The reactions monitored and the corresponding ion kinetic energy (KE) for the collisions were as follows: PFBS, 299f80 (KE ) -51.00 eV), 299f99 (KE ) -37 eV); PFOS, 499f80 (KE ) -80 eV), 499f99 (KE ) -63 eV); PFOA, 413f369 (KE ) -9 eV), 413f169 (KE ) -26 eV); PFOSA, 498f78 (KE ) -45 eV); N-EtPFOSA, 526f169 (KE ) -34 eV). Standards were run with every 10-15 samples. A fivepoint calibration curve spanning concentrations from 10 to 300 pg/µL was used to quantify target analytes. Concentrations of PFOS and PFOA were corrected on the basis of the recovery of the surrogate standard PFBS that was spiked into every sample prior to extraction. N-EtPFOSA and PFOSA concentrations were recovery-corrected on the basis of the mean recovery efficiencies determined from spiked Optima grade water

Results and Discussion Recoveries of PFBS and of analytes in fortified samples processed and analyzed using the described method were good. PFOS spiked into Optima grade water had an average (( standard deviation) of 93 ( 13% (n ) 5). The average recovery of PFBS that was spiked into the samples and used to correct for PFOS and PFOA was 72 ( 13%. The average recoveries of N-EtPFOSA (n ) 7) and PFOSA (n ) 7) that were spiked into Optima grade water were 40 and 20%, respectively. The low recoveries of the neutral sulfonamides were somewhat expected as the analytical method was designed specifically for PFOS and PFOA, which are ionic. Although the recoveries of the sulfonamides were low, especially for PFOSA, the reproducibility of the extraction was very good: the coefficient of variation was 13% in both cases. Even with its limitations, we chose to use the method and recovery-correct for both N-EtPFOSA and PFOSA using VOL. 38, NO. 3, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

759

TABLE 1. Mass Balance Results (Arithmetic Means ( Standard Deviation) of the Five Replicate in Vitro Incubations Experiments Performed on N-EtPFOSAa material present (pmol)

t (h)

N-EtPFOSA

PFOSA

PFOS

total

0 2 4 8 16 30

14.1 ( 9.7 11.3 ( 4.0 8.9 ( 1.8 9.3 ( 2.1 7.9 ( 2.3 7.1 ( 1.6

0.46 ( 0.05 0.92 ( 0.14 1.34 ( 0.04 1.46 ( 0.01 1.8 ( 1.1 4.3 ( 2.3

4.8 ( 2.0 5.1 ( 2.8 5.1 ( 1.8 5.9 ( 2.2 5.8 ( 1.9 7.8 ( 4.5

19.43 17.38 15.33 16.71 15.51 19.26

a Outliers were removed using the Q-test method (26) prior to statistical treatment of the data.

the average recoveries from spiked water because it allowed us to perform the analyses of all the targeted compounds using a single analytical detection technique. Ideally, spiking microsomes would be the preferred way to determine extraction efficiencies. The main difference between spiking microsomes and water would be differential behavior related to the presence of proteinaceous material. This is especially important since the anionic compounds appear to bind to proteins. Unfortunately, the amount of microsomes available was limited. However, PFBS was added to all the samples prior to extraction and should mimic the extraction behavior of PFOS and PFOA in the presence of biological material. The strong basic conditions employed are also likely to disrupt the binding of sulfonamides with proteins, so extraction from water should adequately reflect their behavior in the presence of biological material. Incubation and instrument blanks (solvent injections of methanol) contained both PFOS and PFOA, with the latter typically being greater. The levels of PFOS in the incubation blanks (whether heat-treated microsomes or blanks without N-EtPFOSA) were approximately equal to the method detection limits (MDL, 5 pg on column, S/N ) 10). For solvent blanks, background levels of PFOS were lower than MDLs. In both cases, the PFOS levels were significantly lower than the extracted samples. The levels of PFOS and PFOA in the extraction blanks were similar to that of the instrument or solvent indicating that contamination occurring during extraction was minimized. The presence of PFOS in the HPLC system is thought to arise largely from trace amounts being present in the mobile phase (which was corroborated by the fact that non-Optima grade HPLC water and methanol displayed larger background signals). PFOA was likely present due to leaching from the Teflon tubing and parts of the HPLC system (20). For the extraction blanks, PFOSA was not observed in instrument or incubation blanks, while N-EtPFOSA was only present in incubation blanks to which it had been added (heat-treated microsomes). The number of moles of PFOS in the incubation vials containing microsomes and N-EtPFOSA increased over time, suggesting a biotransformation of N-EtPFOSA to PFOS (Table 1). Although PFOA was detected in the samples, its concentration did not increase with time. This suggests that the source of PFOA was some background contamination, possibly a prior contamination of the lake trout liver used in the experiment rather than biotransformation of NEtPFOSA. There was no indication of abiotic transformation of N-EtPFOSA, as its concentrations in the heat-treated microsomes did not change during the course of the experiment. The N-EtPFOSA standard used in this study was characterized using full-scan positive chemical ionization, electron impact ionization, and electron capture negative ionization GC-MS (not shown), and its purity was estimated 760

9

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

FIGURE 1. Change in amount of N-EtPFOSA, PFOSA, and PFOS over the course of the reaction. Plotted are the arithmetic means ( standard error. to be 85%. The most abundant impurities observed appeared to be isomers of N-EtPFOSA and are likely present due to branched chain impurities in synthetic starting materials (5). It can be expected, therefore, that these branched-chain isomers of N-EtPFOSA are also being biotransformed to nonlinear PFOS. It is unlikely that the branched chain isomers are being transformed to linear PFOS or to PFBS due to the recalcitrance of the perfluoroalkyl chain. Isomerically separated standards for PFOS (and PFOA) are unavailable commercially, and at present chromatographic techniques cannot completely resolve the isomers (21). Reported concentrations of PFOS in the environment therefore are likely that of branched- and straight-chain PFOS isomers. Even if the starting material used in this study contains branched-chain isomers and are biotransformed to branched PFOS, this still contributes to the burden of PFOS in the environment. The biotransformation of N-EtPFOSA is dependent on the presence of NADPH (22) and is, therefore, likely to have been mediated by the phase I, or NADPH-cytochrome P-450, system. The active site of these enzymes contains a heme unit that is reduced by an electron transfer from NADPH mediated by cytochrome reductase. After binding molecular oxygen, the enzyme complex incorporates one atom of oxygen into the substrates structure, with the second atom being reduced to water (23). Biotransformation Pathways. Three reaction pathways are possible for the conversion of N-EtPFOSA to PFOS: (i) direct deethylamination of N-EtPFOSA accompanied by conversion of the sulfone to sulfonate, (ii) deethylation of PFOSA followed by deamination to form PFOS, and (iii) direct hydrolysis of N-EtPFOSA to PFOS.

(a) Biotransformation of Starting Material, N-EtPFOSA. Figure 1 shows that the biotransformation of N-EtPFOSA occurs rapidly at first and slows down after 4 h. The presence of PFOSA and PFOS at t ) 0 h is thought to arise from the physical delay in quenching the reaction. If the initial phase of the biotransformation is very fast and there is an unavoidable time delay in stopping the reaction (usually a few minutes), then this might result in the formation of the products.

t ) 30 h remained unchanged, strongly suggesting that no other metabolic products (i.e., PFOA) other than PFOSA and PFOS were produced. In addition, the results of the mass balance also suggest that even though the recoveries of N-EtPFOSA and PFOSA were low, correcting for their loss was appropriate and not a result of an analytical artifact. Implications of Findings. N-EtPFOSA is not registered for use as a pesticide in Canada. However, it is still being used in many parts of the southeastern United States to control termites and cockroaches. The detection of NEtPFOSA in air samples in Canada (14) suggests that this compound is fairly persistent in the atmosphere and may in fact be contributing to the burden of PFOS observed in the Canadian Arctic (8).

FIGURE 2. Test of the first-order reaction kinetics for the biotransformation of N-EtPFOSA. An attempt was made to try and understand the kinetics associated with the biotransformation of N-EtPFOSA. Because most processes involved in toxicokinetic models, including metabolism, can be described by first-order reactions, i.e., the rate of reaction is proportional to the concentration of the starting material present (24, 25), the disappearance of N-EtPFOSA was monitored by plotting ln[N-EtPFOSA] vs time to test if first-order reaction kinetics described the transformation reaction (Figure 2). The plot shows that there are two distinct linear regions, the slope of the first being much steeper than the second. More than one process appears to be occurring. Second-order reaction kinetics was also tested by plotting the 1/[N-EtPFOSA] vs time. This plot again showed two distinct segments (not shown), the first segment being linear within the short three-point range. On the basis of these results, it is unclear whether the initial reaction is firstor second-order. (b) Formation of the Intermediate, PFOSA. The conversion of N-EtPFOSA in terrestrial mammals to PFOSA proceeds through a deethylation step (16, 18). The formation of PFOSA as a biotransformation product of N-EtPFOSA in this study is consistent with those findings. The increase in concentration of PFOSA with time is shown graphically in Figure 1. (c) Formation of PFOS. The increase in the number of moles of PFOS with time confirms that PFOS is formed from N-EtPFOSA (Figure 1). As mentioned earlier, PFOS may be formed directly from N-EtPFOSA or through the PFOSA intermediate. The kinetics associated with the formation of PFOS from N-EtPFOSA are not straightforward. Interpretation of the reaction kinetics data is complicated by the fact that the concentration of the intermediate, PFOSA, which is thought to proceed directly to PFOS, is also changing with time. A similar experiment using PFOSA as the starting material may help to elucidate the PFOSA f PFOS kinetics, thereby simplifying the interpretation of the kinetics of the N-EtPFOSA f PFOS reaction pathway. (d) Formation of PFOA. Background levels of PFOA were detected in the incubation samples, but there was no increase in the concentration of PFOA throughout the course of the experiment. This suggests that there was a source of background contamination in the experiment. One possibility is that, prior to microsome preparation, the rainbow trout were exposed to an environmental source of PFOA, thus resulting in contaminated liver microsomes. The lack of a correlation of PFOA concentrations with time also indicates that PFOA was not being formed as a result of transformations of N-EtPFOSA. (e) Mass Balance. A mass balance was conducted to account for the change in the number of moles of the starting material and metabolic products formed during the experiment (Table 1). The total number of moles at t ) 0 h and at

The present study confirms that perfluorosulfonamides are in fact precursors of PFOS in fish and, combined with the results of Martin et al. (14), demonstrates that atmospheric N-EtPFOSA (and PFOSA) can be sources of environmental PFOS. There are other compounds similar in structure to N-EtPFOSA that may also be neutral PFOS precursors in fish. The perfluorosulfonamido alcohols, for example, which were also detected by Martin et al. (14), may also be contributing to loadings of PFOS in remote regions. Standards of these compounds, however, are not readily available and examination of their biotransformation could not be performed in this study. Future work will investigate the toxicokinetics of PFOS and other structurally analogous precursors administered to fish. The connection between PFOS, N-EtPFOSA, and PFOSA will also be expanded to a more environmentally realistic in vivo situation.

Acknowledgments We are indebted to the helpful suggestions of Dr. John Westmore (University of Manitoba) on an earlier version of the manuscript. We also thank Drs. Geoff Eales (University of Manitoba) and Jonathan Martin (University of Toronto) for their helpful comments. This manuscript benefited greatly from the comments of two anonymous reviewers. This work was funded in part by a Department of Fisheries and Oceans (DFO) Subvention Grant awarded to K.F and G.T.T.

Literature Cited (1) Key, B. D.; Howell, R. D.; Criddle, C. S. Environ. Sci. Technol. 1997, 31, 2445. (2) U.S. EPA. Fluorochemical Use, Distribution and Release Overview; Rep. AR226-0550; U.S. EPA: Washington, D.C., 1999. (3) U.S. EPA. The Science of Organic Fluorochemistry; Rep. AR-2260547; U.S. EPA: Washington, D.C., 1999. (4) Giesy, J. P.; Kannan, K. Environ. Sci. Technol. 2002, 36, 147A. (5) Renner, R. Environ. Sci. Technol. 2001, 35, 154A. (6) Kannan, K.; Newsted, J.; Halbrook, R. S.; Giesy, J. P. Environ. Sci. Technol. 2002, 36, 2566 (7) Olsen, G. W.; Burris, J. M.; Mandel, J. H.; Zobel, L. R. J. Occup. Environ. Med. 1999, 41, 799. (8) Giesy, J. P.; Kannan, K. Environ. Sci. Technol. 2001, 35, 1339. (9) Kannan, K.; Koistinen, J.; Beckmen, K.; Evans, T.; Gorzelany, J. F.; Hansen, K. J.; Jones, P. D.; Helle, E.; Nyman, M.; Giesy, J. P. Environ. Sci. Technol. 2001, 35, 1593 (10) Moody, C. A.; Kwan, W. C.; Martin, J. W.; Muir, D. C. G.; Mabury, S. A. Anal. Chem. 2001, 73, 2200. (11) Moody, C. A.; Martin, J. W.; Kwan, W. C.; Muir, D. C. G.; Mabury, S. A. Environ. Sci. Technol. 2002, 36, 545. (12) U.S. EPA. Perfluorooctane Sulfonic Acid and Its Salts; Rep. AR2260978; U.S. EPA: Washington, D.C., 2000. (13) U.S. EPA. PFOSsTransport between Environment Compartments (Fugacity); Rep. AR226-0060; U.S. EPA: Washington, D.C., 2000. VOL. 38, NO. 3, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

761

(14) Martin, J. W.; Muir, D. C. G.; Moody, C. A.; Ellis, D. A.; Kwan, W. C.; Solomon, K. R.; Mabury, S. A. Anal. Chem. 2002, 74, 584. (15) Kissa, E. Fluorinated Surfactants and Repellents, 2nd ed.; Marcel Dekker Inc.: New York, 2001. (16) Grossman, M. R.; Mispagel, M. E.; Bowen, J. M. J. Agric. Food Chem. 1992, 40, 2505. (17) Tomy, G. T.; Budakowski, W.; Halldorson, T.; Helm, P.; Stern, G.; Tittlemier, S. Organohalogen Compd. 2003, 62, 323. (18) Arrendale, R. F.; Stewart, J. T.; Manning, R.; Vitayavirasuk, B. J. Agric. Food Chem. 1989, 37, 1130 (19) Shields, C. A.; Eales, J. G. Gen. Comp. Endocrinol. 1986, 63, 334. (20) Hansen, K. J.; Clemen, L. A.; Ellefson, M. E.; Johnson, H. O. Environ. Sci. Technol. 2001, 35, 766. (21) Hansen, K. J.; Johnson, H. O.; Eldridge, J. S.; Butenhoff, J. L.; Dick, L. A. Environ. Sci. Technol. 2002, 36, 1681.

762

9

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

(22) Munkittrick, K. R.; van den Heuvel, M. R.; Metner, D. A.; Lockhart, W. L.; Stegeman, J. J. Environ. Toxicol. Chem. 1993, 12, 1273. (23) Landis, W. G.; Yu, M.-H. Introduction to Environmental Toxicology; Lewis Publishers: Boca Raton, FL, 1995. (24) Barron, M. G.; Stehly, G. R.; Hayton, W. L. Aquat. Toxicol. 1990, 17, 187. (25) Ballantyne, B.; Marrs, T.; Turner, P. General and Applied Toxicology, College Edition; Stockton Press: New York, 1993. (26) Dean, R. B.; Dixon, W. J. Anal. Chem. 1951, 23, 636.

Received for review June 3, 2003. Revised manuscript received October 15, 2003. Accepted October 23, 2003. ES034550J