Anal. Chem. 1999, 71, 5242-5247
Extraction and Isolation of Linear Alkylbenzenesulfonate and Its Sulfophenylcarboxylic Acid Metabolites from Fish Samples Johannes Tolls,* Manuela Haller, and Dick T. H. M. Sijm†
Environmental Toxicology & Chemistry, RITOXsResearch Institute of Toxicology, Utrecht University, P.O. Box 80.058, NL-3508 TB Utrecht, The Netherlands
Linear alkylbenzenesulfonate (LAS) is the most widely used synthetic surfactant. In fish, assessment of the environmental risk and investigation of the biotransformation behavior of LAS require compound-specific methods for extraction and isolation of LAS and its biotransformation products, sulfophenylcarboxylic acids (SPC). Matrix solid-phase dispersion (MSPD) extraction with subsequent ion-pair liquid-liquid (IP-LL) partitioning of the extract was a time-efficient sample preparation method for analysis of LAS. The recovery of parent LAS from spiked fish exceeded 70%, and the limit of quantitation ranged around 0.2 mg‚kg-1 corresponding to 0.6 µmol‚ kg-1. In a simultaneous determination of LAS and SPC in fish, the analytes were MSPD extracted in different fractions. The target compounds were separated from the sample matrix by protein precipitation and subsequent isolation of (a) SPC by graphitized carbon black solidphase extraction of the supernatant and (b) parent LAS by IP-LL partitioning of the pellet obtained after protein precipitation. The recoveries of the model compounds C12-2-LAS and C4-3-SPC were 84 ( 6 and 65 ( 11%, respectively. The use of C3-3-SPC as an internal standard corrected for the loss of the biotransformation product during sample workup. The suitability of both methods was demonstrated by analyzing fish containing LAS and SPC incurred during aqueous exposure. Linear alkylbenzenesulfonate (LAS) is the most widely used synthetic surfactant with an annual production rate of 1.8 million tons.1 It is a mixture of n-(p-sulfophenyl)alkanes, and the individual constituents are abbreviated as Cn-m-LAS, with n specifying the number of C-atoms in the alkyl chain (10-13) and m identifying the C-atom at which the p-sulfophenyl moiety is attached to the alkyl chain. In technically produced LAS, m can assume values between 2 and 7. Being the “workhorse” surfactant in laundry detergents, LAS is discharged with household wastewater although efficiently removed by wastewater treatment.2-8 The * Corresponding author: (tel) **31/30/2532578; (fax) **31/30/2532837; (e-mail)
[email protected]. † Present address: RIVM-CSR, P.O. Box 1, 3720 BA Bilthoven, The Netherlands. (1) Berth, P.; Jeschke, P. Tenside Surf. Deterg. 1989, 26, 75-79.
5242 Analytical Chemistry, Vol. 71, No. 22, November 15, 1999
fraction remaining in the wastewater eventually reaches surface waters,7-13 therefore exposing aquatic organisms to LAS. The bioaccumulation potential of LAS required evaluation in the course of an environmental risk assessment of this surfactant.14,15 In the past, LAS bioaccumulation research has relied on quantifying total radioactivity without distinguishing between the parent surfactant and its biotransformation products.16 Therefore, these data do not reflect the bioaccumulation potential of the parent surfactant.16 Moreover, quantitative information on the biotransformation of LAS is not available. However, it has been proposed17 that LAS (Figure 1a) is transformed via ω-oxidation and subsequent β-oxidation steps to sulfophenylalkanoic acids (SPC; Figure 1b). Since biotransformation results in a reduction of the concentration of LAS in fish,16-18 it is an important process contributing to the overall LAS bioaccumulation and therefore an issue deserving scientific attention. (2) Brunner, P. H.; Capri, S.; Marcomini, A.; Giger, W. Water Res. 1988, 22, 1465-1472. (3) Gledhill, W. E.; Huddleston, R. L.; Kravetz, L.; Nielsen, A. M.; Sedlak, R. I.; Vashon, R. D. Tenside Surf. Deterg. 1989, 26, 276-281. (4) Prats, D.; Ruiz, F.; Vazquez, B.; Zarzo, D.; Berna, J. L.; Moreno, A. Environ. Toxicol. Chem. 1993, 12, 1599-1608. (5) Feijtel, T. C. J.; Matthijs, E.; Rottiers, A.; Rijs, G. B. J.; Kiewiet, A.; De Nijs, A. Chemosphere 1995, 30, 1053-1066. (6) Moreno, A.; Ferrer, J.; Ruiz Bevia, F.; Prats, D.; Vazquez, B.; Zarzo, D. Water Res. 1994, 28, 2183-2189. (7) Rapaport, R. A.; Eckhoff, W. S. Environ. Toxicol. Chem. 1990, 9, 12451257. (8) Scho ¨berl, P. 4th World Surfactant Congress, Barcelona, 3-7 June, 1996; pp 87-98. (9) Tabor, C. F.; Barber, L. B., II Environ. Sci. Technol. 1996, 30, 160-171. (10) Kikuchi, M.; Tokai, A.; Yoshida, T. Water Res. 1986, 20, 643-650. (11) Gonzalez-Mazo, E.; Honing, M.; Barcelo, D.; Gomez-Parra, A. Environ. Sci. Technol. 1997, 31, 504-510. (12) Di Corcia, A.; Marchetti, M.; Samperi, R.; Marcomini, A. Anal. Chem. 1991, 63, 1179-1182. (13) Scho ¨ster, M. Ph.D. Thesis, Universita¨t Du ¨ sseldorf, Ju ¨ lich, Germany, 1993. (14) Feijtel, T. C. J.; Kloepper-Sams, P.; Den Haan, K.; Van Egmond, R.; Comber, M.; Heusel, R.; Wierich, P.; Ten Berge, W.; A., G.; De Wolf, W.; Niessen, H. Chemosphere 1997, 34, 2337-2350. (15) Kloepper-Sams, P. J.; Cowan, C. E.; Larson, R. J.; Versteeg, D. J. 4th World Surfactant Congress, Barcelona, 3-7 June 1996; pp 213-219. (16) Tolls, J.; Kloepper-Sams, P.; Sijm, D. T. H. M. Chemosphere 1994, 29, 693717. (17) Newsome, C. S.; Howes, D.; Marshall, S. J.; Van Egmond, R. A. Tenside Surf. Deterg. 1995, 32, 498-503. (18) Comotto, R. M.; Kimerle, R. A.; Swisher, R. D. In Aquatic Toxicology; Marking, L. L., Kimerle, R. A., Eds.; American Society of Testing Materials: Philadelphia, PA, 1979; Vol. ASTM STP 667, pp 232-250. 10.1021/ac990235x CCC: $18.00
© 1999 American Chemical Society Published on Web 10/19/1999
Figure 1. Chemical structures of 2-n-(p-sulfophenyl)dodecane (C12-2-LAS, a) and its biotransformation product 3-n-(p-sulfophenyl)butyric acid (C4-3-SPC, b).
It appears likely that the shortage of valid data on LAS bioaccumulation and biotransformation is partly due to the lack of compound-specific methods for the determination of LAS and SPC in biological tissues. In contrast, a variety of methods have been employed to determine LAS in other environmental matrixes. HPLC with fluorometric detection (HPLC-FL),10,19-21 as well as gas chromatography of LAS derivatives with either electron capture or mass-selective detection.9,22-26 The more time-efficient HPLC-FL determination method is preferred for selective as well as sensitive determination of LAS for the analysis of large numbers of samples throughout our investigation of LAS bioconcentration. The most frequently employed method for enrichment of LAS from water samples is solid-phase extraction (SPE) with octadecyl (C18)-modified silica, either in the form of extraction cartridges4,10,11,19,20 or as extraction disks.27,28 C2- and C8-modified silica26,29 have also been used. The use of XAD-4,13 as well as a polymer-based anion-exchange resin,30 for extraction of LAS has been described as well. Simultaneous extraction of LAS and SPC can be achieved using graphitized carbon black (GCB) extraction cartridges.31 This stationary phase behaves as an anion-exchanger site upon acidification32 and retains even very hydrophilic aromatic acids.32,33 Extraction and Isolation of Analytes from Solid Samples. Environmental solids such as sediments, suspended particles obtained by filtration, or lyophilized sewage sludge resemble fish tissue with respect to matrix complexity. Hence, the approaches (19) Marcomini, A.; Capri, S.; Giger, W. J. Chromatogr. 1987, 403, 243-252. (20) Matthijs, E.; De Henau, H. Tenside Surf. Deterg. 1987, 24, 193-199. (21) Linder, D. E.; Allen, M. C. J. Am. Oil Chem. Soc. 1982, 69, 152-155. (22) McEvoy, J.; Giger, W. Environ. Sci. Technol. 1986, 20, 376-383. (23) Savric, I.; Reemtsma, T.; Jekel, M. Umwelt und Chemie; Umwelttagung 1996, GDCh, Ulm, Germany, 7-10 October 1996; Vol. 17. (24) Field, J. A.; Miller, D. J.; Field, T. M.; Hawthorne, S. B.; Giger, W. Anal. Chem. 1992, 64, 3161-3167. (25) Osburn, Q. W. J. Am. Oil Chem. Soc. 1986, 63, 257-263. (26) Trehy, M. L.; Gledhill, W. E.; Orth, R. G. Anal. Chem. 1990, 62, 25812586. (27) Yamini, Y.; Ashraf-Khorassani, M. J. High Resolut. Chromatogr. 1994, 17, 634-638. (28) Yamini, Y.; Ashraf-Khorassani, M. Fresenius J. Anal. Chem. 1994, 348, 251252. (29) Castles, M. A.; Moore, B. L.; Ward, S. R. Anal. Chem. 1989, 61, 25342540. (30) Yokoyama, Y.; Kondo, M.; Sato, H. J. Chromatogr. 1993, 643, 169-172. (31) Di Corcia, A.; Samperi, R.; Marcomini, A. Environ. Sci. Technol. 1994, 28, 850-858. (32) Di Corcia, A.; Marchese, S.; Samperi, R. J. Chromatogr., A 1993, 642, 163174. (33) Altenbach, B.; Giger, W. Anal. Chem. 1995, 67, 2325-2333.
taken in the analysis of these samples can serve as a starting point in the development of a method for extraction and isolation from fish samples. Solids are most frequently liquid extracted with polar solvents such as MeOH20,34 or with an apolar solvent in the presence of an ion-pair reagent as phase-transfer catalyst.24 Alternatively, two references report the use of supercritical fluid extraction.24,35 Isolation of LAS from matrix constituents is achieved by anion-exchange chromatography,20 aluminum oxide column chromatography,13 or silica TLC.22 Apparently, ion-pair liquid-liquid (IP-LL) extraction of sewage sludge samples yields extracts that are sufficiently clean for analysis.24 In contrast, there is only one reported attempt to extract LAS from fish tissue.10 Since LAS is not detected and the recovery rate is not specified, the performance of that method cannot be evaluated. Matrix Solid-Phase Dispersion Extraction. Matrix solidphase dispersion extraction (MSPD) can be applied to a variety of matrixes.36-40 Analytes with a broad range of polarity ranging from polychlorinated biphenyls40 and pesticides39 to sulfonamide and β-lactam antibiotics36,41 can be recovered with sufficient extraction yields. The method allows for a comparatively selective extraction of the analytes of interest, thereby simplifying the subsequent isolation steps.36 Thus, MSPD extraction is a promising approach for extraction of the hydrophobic LAS anions and polar SPC. Quantitation of LAS bioaccumulation and biotransformation requires the determination of the parent compounds and biotransformation products. In the present research, aqueous samples are extracted utilizing C18-SPE for analysis of the parent LAS and GCBSPE when SPC and LAS are extracted simultaneously. We report the use of fractionated MSPD extraction of LAS and SPC from fish samples. Two isolation methods are presented: one for determination of LAS exclusively and one for simultaneous determination of LAS and SPC. The applicability of the methods to determine incurred levels of LAS and SPC in fish is demonstrated. MATERIALS AND METHODS Chemicals and Fish. Solvents (CH3CN, CH3OH, ethyl acetate (EtOAc), CH2Cl2) were of HPLC gradient grade and supplied by Mallinckrodt-Baker (Deventer, NL) or Merck (Amsterdam, NL). Triethylamine (N(Et)3; Merck) was of spectroscopical purity. Technical hexane (Mallinckrodt-Baker) was redistilled in glass prior to use. HCl (Merck) was of analytical grade. Octadecylsilica (ODS) SPE columns and ODS were obtained from MallinckrodtBaker. SPE columns filled with graphitized carbon black were supplied by Supelco (Sigma-Aldrich-Fluka, Zwijndrecht, NL). Water for HPLC was purified with an ELGASTAT (ELGA, Buchs, CH) purification system. Distilled water was glass distilled in a (34) Holt, M. S.; Matthijs, E.; Waters, J. Water Res. 1989, 23, 749-759. (35) Hawthorne, S. B.; Miller, D. J.; Walker, D. D.; Whittington, D. E.; Moore, B. L. J. Chromatogr. 1991, 541, 185-194. (36) Barker, S. A.; Long, A. R.; Short, C. R. J. Chromatogr., A 1989, 475, 353361. (37) Long, A. R.; Hsieh, L. C.; Malbrough, M. S.; Short, C. S.; Barker, S. A. J. Assoc. Off. Anal. Chem. 1990, 73, 379-384. (38) Lott, H. M.; Barker, S. A. J. AOAC Int. 1993, 76, 663-668. (39) Long, A. R.; Soliman, M. M.; Barker, S. A. J. Assoc. Off. Anal. Chem. 1991, 74, 493-496. (40) Ling, Y.-C.; Huang, I.-P. Chromatographia 1995, 40, 259-266. (41) Long, A. R.; Hsieh, L. C.; Malbrough, M. S.; Short, C. S.; Barker, S. A. J. Agric. Food Chem. 1990, 38, 423-426.
Analytical Chemistry, Vol. 71, No. 22, November 15, 1999
5243
Bu¨chi-Fontavapor (Bu¨chi, Flawil, CH). NaOH, KH2PO4, Na2HPO4 (all from Merck, Darmstadt), and tetrabutylammonium hydroxide (TBA-OH; Fluka, Sigma-Aldrich-Fluka) and tetrabutylammonium hydrogen sulfate (TBA-HSO4; Janssen Chimica, Tilburg, NL) were of analytical grade. Spectroscopical grade trifluoroacetic acid (TFA) and n-(p-sulfophenyl)octane (>97%) were purchased from Merck, Darmstadt, and Aldrich (Sigma-Aldrich-Fluka), respectively. Individual n-(p-sulfophenyl)alkanes were synthesized in our laboratory according to methods reported in the literature42,43 and were at least 97% pure. n-(p-sulfophenyl)alkane mixtures consisting of primarily one single homologue were a gift from PETRESA (Madrid, Spain). Dr. Sarrazin (University of Marseilles, France) kindly provided us with 3-(p-sulfophenyl)propanoic acid (C3-3SPC), and 3-(p-sulfophenyl)butanoic acid (C4-3-SPC) was given to us by Dr. Cavalli (Condea, Milano, Italy) and by Dr. Kanz (EAWAG, Du¨bendorf, Switzerland). The [14C]LAS was uniformly ring labeled and supplied by NATEC (Hamburg, DE). Its composition resembled that of commercial mixtures. Fathead minnows (Pimephales promelas) reared in the hatchery of Utrecht University were used throughout this study. The fish weighed between 0.4 and 1.0 g and were killed by immersion in liquid nitrogen or by cervical dislocation. Chemical Measurements. Extracts or aliquots of extracts containing [14C]LAS were diluted with scintillation fluid (Emulsifier Safe, Packard Research Instruments, Groningen, NL), and the radioactivity was determined by liquid scintillation counting (Packard TRI 2380, Packard Research Instruments). The analyses of “cold” LAS and SPC were performed by HPLC. The gradients were delivered by a Gynkotek M480 pump (Separations, Alblasserdam, NL). A Basic Marathon (Spark Holland, Separations, Alblasserdam, NL) fitted with a 20-µL injection loop was employed for sample introduction. Separation took place on a 100 mm × 3 mm Chromsphere (5 µm) ODS column (Chrompack, Bergen op Zoom, NL). The fluorescence signal of a Jasco FL 920 fluorescence detector (Separations) was collected using Chromcard (Fisons, Interscience, NL) data collection software on a personal computer. When analyzing for parent LAS only, we employed gradient program 1 (Table 1), which allowed for separation of the 2-isomer from the inner isomers of the same homologue.20 The inner isomers, abbreviated as Cn-inner, were quantitated as the sum of the chromatographically unresolved 3-, 4-, 5-, 6-, and 7-alkyl isomers of the homologues with n carbon atoms. For simultaneous determination of the LAS constituents and their sulfophenylcarboxylic acid biotransformation products, a reversed-phase ion-pair chromatographic method was used33 (i.e., gradient program 2 in Table 1). Extraction of Water Samples. When the analysis was for LAS exclusively, water samples (∼50 mL) were extracted with ODS SPE columns (pretreated with 10 mL of MeOH and water) as described by ref 20. LAS was eluted from the column with 3 mL of CH3OH. After evaporation of the solvent, the sample was redissolved in 300 µL of CH3OH and transferred to HPLC vials. Prior to simultaneous extraction of LAS and SPC, GCB columns31 were pretreated by rinsing the columns with 15 mL of CH2Cl2CH3OH (1:1), 5 mL of CH3OH, and 25 mL of 0.01 M HCl in H2O. Water samples (∼75 mL) were then allowed to drain through the (42) Doe, P. H.; El-Amary, M.; Wade, W. H.; Schecter, R. S. J. Am. Oil Chem. Soc. 1977, 54, 570-577. (43) Gray, F. W.; Gerecht, J. F.; Krems, I. J. J. Org. Chem. 1955, 20, 511-524.
5244 Analytical Chemistry, Vol. 71, No. 22, November 15, 1999
Table 1. Analytical HPLC for Constituents of LAS and Their Biodegradation Products p-(Sulfophenyl)alkanoic Acidsa program 1 analytes flow Tcolumn λex/λem eluent
program 2
LAS only 600 µL‚min-1 ambient 225 nm/295 nm time A B
C
0 6 16 22 25
12 12 3 1 1
48 48 72 90 90
40 40 25 9 9
LAS and SPC 500 µL‚min-1 40 °C 232 nm/292 nm time A D 0 1 45 50
5 5 95 95
95 95 5 5
a T column specifies the temperature of the column. Eluents: A, CH3CN; B, H2O; C, 0.5 M NaClO4; D, 5 mM TBA-HSO4, 10 mM PO4, pH 6.
column without applying a vacuum. Subsequently, the column was air-dried for 1 min and the residual water was removed with 1 mL of MeOH. SPC and LAS were eluted with 8:2:1 (v:v) in CH2Cl2-CH3OH-NEt3. The extract was neutralized with TFA, evaporated to dryness, and redissolved in 200 µL of MeOH. Prior to HPLC analysis, 200 µL of eluent D (Table 1) was added to the sample. MSPD Extraction of Fish Samples. A single, whole dead fish was transferred to a mortar, followed by the addition of the standard spike. The fish was then ground thoroughly in the mortar with a pestle. The entire mass was homogenized after addition of 1 mL of MeOH. Four grams of ODS was added to each 1 g of fish. The mixture was ground into a homogeneous paste. The fins were a tough tissue that often remained intact during grinding; however, this had no significant effect on the recovery of the analytes. The paste was allowed to air-dry until a powdery homogenate was obtained. A column for extraction of the ODS matrix was constructed by removing the plunger from a 20-mL injection syringe (Becton & Dickinson), filling quartz wool (Soxhlet extracted for 18 h with MeOH) into the bottom of the syringe, topping the quartz wool plug with a filter paper disk (Whatman No. 2), and washing it thoroughly with EtOAc and CH3OH. After drying, the column was filled with the ODS matrix powder. In fractionated elution experiments, the elution pattern and recoveries of LAS and SPC were determined by eluting the column sequentially with 40 mL of hexane/g of fish, 5 mL of EtOAc, and 20 mL/g of fish each of EtOAc-CH3OH (1:1, v:v), MeOH, and MeOH-H2O, yielding the fractions 1-5, respectively (Figure 2). Isolation of Analytes. I. LAS Exclusively. The fraction containing LAS was blown to dryness, redissolved in 1 mL of MeOH using ultrasonication, and transferred to a shake flask. Subsequently, the extract was partitioned between 25 mL of CH2Cl2-MeOH (3:1, v:v) and an aqueous phase containing 25 mM TBA-OH and 0.7 M NaOH. The partition step was repeated with 15 mL of CH2Cl2-MeOH (3:1, v:v), and the organic phases were combined and taken to dryness. After the sample was redissolved in 300 µL of MeOH by ultrasonication, it was ready for injection into the HPLC. II. LAS and SPC in One Sample. The MSPD extraction fractions 3-5 were blown to dryness, redissolved in 2 × 200 µL
of CH3OH (by ultrasonication), and transferred to a 10-mL centrifugation tube. After addition of 0.5 mL of 0.1 M H3PO4 buffer, the samples were kept at 4 °C overnight and centrifuged for 60 min at 1000g. In this manner, the proteins were precipitated36 and separated from the supernatant. The supernatant of each fraction was diluted with 10 mL of distilled water and extracted using GCB SPE cartridges as described for the water samples. A significant portion of LAS was associated with the pellet of fraction 3. LAS was recovered by resuspending the pellet in the eluate of the GCB extraction of fraction 3, transferring the suspension to a shake flask, and subsequently extracting by IP-LL partitioning. All fractions were analyzed separately after evaporation and redissolution in 200 µL of MeOH by ultrasonication. Prior to analysis, 200 µL of eluent D was added to all fractions. Evaluation of Method Performance. HPLC chromatograms were utilized to evaluate the efficiency of the cleanup steps. At the same time, this allowed for recovery quantification. The limit of quantification (LOQ) was defined as the concentration required to increase the signal to 10 times the noise of the baseline. The applicability of the methods was evaluated by analyzing fathead minnows that had taken up LAS in bioconcentration and biotransformation experiments in which the fish were exposed to LAS in aquariums receiving a constant input of LAS.44 For extraction and isolation of LAS exclusively, fish were exposed to a mixture of pure homologue LAS materials with n ranging from 10 to 13 (internal standard, C8-1-LAS). The method for simultaneous analysis of LAS and SPC was evaluated in fish exposed to C12-2LAS and the parent compound and the analysis of its presumably stable biotransformation intermediate C4-3-SPC. C12-2-LAS and C43-SPC were selected because these compounds were available as authentic standards. C12-1-LAS and C3-3-SPC were selected as internal standards due to their close structural similarity to the analytes. The recovery of the procedure for simultaneous extraction of LAS and SPC was calculated by summing up the recoveries in the individual fractions. RESULTS AND DISCUSSION Recovery of LAS and SPC from Aqueous Samples. At spiking levels ranging between 40 and 200 µg‚L-1, the recovery of the ODS extraction of LAS was 85% and higher (all experiments in triplicate) for the individual LAS constituents tested. The recovery of the GCB extraction of SPC and LAS exceeded 90% (n ) 3) for the SPC as well as the LAS constituents tested. The method LOQ was 5 µg‚L-1 when the analysis was for LAS only and 7 and 30 µg‚L-1 for LAS and SPC, respectively. The respective values in molar units were 15, 21, and 77 nM. MSPD Extraction of Radiolabeled LAS from Spiked Fish. The suitability of MSPD for extraction of radiolabeled LAS from spiked fish was assessed by fractionated elution from MSPD columns. The hexane, the EtOAc, and the MeOH fractions contained 0.4 ((0.2%), 1.1 ((1.1%), and 4.2% ((5.5%), respectively, of the radioactivity applied initially (10 µg‚g-1). In the EtOAcMeOH (1:1, v:v) the recovery was 94.1% ((5.3%) and therefore almost quantitatively. Neither hexane, nor toluene (results not shown), nor EtOAc eluted significant portions of LAS from the MSPD column, indicating that protic solvents were required to elute LAS from the column. (44) Tolls, J.; de Graaf, I.; Thijssen, M. A. T. C.; Haller, M.; Sijm, D. T. H. M. Environ. Sci. Technol. 1997, 31, 3426-3431.
MSPD Extraction of SPC. The fractionated elution of C4-3SPC and C3-3-SPC from the MSPD column revealed that fractions 1 and 2 did not contain SPC. In contrast, significant portions (>5% of the total) of the SPC were eluted in fractions 3-5 (Figure 3). The extraction yields were not quantitative (54 and 65% for C3-3SPC and C4-3-SPC). The portions of SPC recovered in the individual fractions decreased with increasing solvent strength. An increase of elution volumes and/or solvent strength is therefore not likely to increase the recovery. The variability in the size of the portions of SPC recovered in the different fractions was high relative to that observed for LAS, although the extraction yields of the two model SPC covaried strongly (r > 0.99, data not shown). This suggested that C3-3-SPC could be used as an internal standard to correct for the analytical losses of C4-3-SPC during sample preparation, thereby permitting quantitative determination of C4-3-SPC. Isolation of Analytes I. LAS Exclusively. The EtOAcMeOH extracts containing LAS were colored. In addition, repeated injection (n > 10) of MSPD extracts (concentrated to 1 mL) led to decreased chromatographic performance, indicating that the samples required further cleanup. Proteins have been reported to be the predominant constituents in tissue extracts obtained by eluating MSPD columns with polar solvents.45 Given that proteins are denatured in an alkaline medium and that IP-LL extraction recovered LAS quantitatively from 1 M NaOH (data not shown), it was expected that proteins were retained in the aqueous phase while LAS was extracted as a TBA-LAS ion pair into the organic phase. This was confirmed by the recovery data of C10-2-, C11-2-, C12-2-, C13-2-, C9-1-, C12-1-, and C13-1-LAS from fish (spiked with 0.6 and 8 µg‚g-1), which ranged between 90 and 130% for all compounds (Figure 4) and could thus be regarded as quantitative. Moreover, recoveries were independent of the amount spiked in a range of concentrations that was expected in the bioconcentration experiments. HPLC chromatograms of the fish extracts were free of interferences in the region of analytical interest after treatment of the MSPD extracts by IP-LL partitioning. The method allowed for concentration of the fish sample fractions to 300 µL, thereby decreasing the LOQ of individual constituents of LAS to an injection amount of 10-15 ng. For fish samples weighing 1 g, this translated to a LOQ of ∼0.2 mg‚kg-1. The proximity of the lower spike level to the LOQ explained why the standard deviation (n ) 3) was higher than at the higher spike level. In a second recovery experiment (data not shown), the recovery of LAS constituents relative to that of C8-1-LAS had a small standard deviation (
