Anal. Chem. 2005, 77, 864-870
Development of a Solid-Phase Extraction-HPLC/ Single Quadrupole MS Method for Quantification of Perfluorochemicals in Whole Blood Anna Ka 1 rrman,*,† Bert van Bavel,† Ulf Ja 1 rnberg,‡ Lennart Hardell,† and Gunilla Lindstro 1 m†
Man-Technology-Environment Research Centre, O ¨ rebro University, SE-701 82 O ¨ rebro, Sweden, and Institute of Applied Environmental Research, Stockholm University, SE-106 91 Stockholm, Sweden
A method for the determination of perfluorooctanesulfonate (PFOS) and perfluorooctanoic acid (PFOA) simultaneously with 10 closely related perfluorochemicals (PFCs) in human whole blood was developed and validated. PFOS and PFOA are used in various applications, for example, as surfactants and plastic additives, and are subject to environmental and health research due to their persistence. The main part of the data on PFCs in human blood is from serum samples, analyzed mainly by ion pair extraction followed by high-performance liquid chromatography (HPLC) and negative electrospray (ESI) tandem mass spectrometry (MS/MS). The analytical method developed here is suitable for human whole blood and involves solid-phase extraction (SPE) and HPLC negative electrospray single quadrupole mass spectrometry (HPLC/ ES-MS). A whole blood aliquot was treated with formic acid and extracted on a octadecyl (C18) SPE column. The PFCs were isolated with methanol, and quantification was performed using single quadrupole mass spectrometry and perfluoroheptanoic acid as internal standard. Validation was performed in the range 0.3-194 ng/mL with recovery between 64 and 112% and limit of detection in the 0.1-0.5 ng/mL range for 11 of the 12 PFCs studied. We applied this method to 20 whole blood samples collected in 1997-2000 from the Swedish population in the ages 24-72. Eleven of the 12 PFCs were detected, and they were quantitatively and qualitatively confirmed using triple quadrupole LC/MS/MS analysis. PFOS, perfluorooctanesulfonamide, perfluorohexanesulfonate, PFOA and perfluorononanoic acid were quantified in all samples. In addition, perfluorohexanoic acid, perfluorodecanoic acid, perfluorodecanesulfonate, perfluoroundecanoic acid, perfluorododecanoic acid, and perfluorotetradecanoic acid were detected in some samples. This study shows that SPE and single quadrupole MS can be applied for extraction and quantification of PFCs in human whole blood, resulting in selectivity and low detection limits. Perfluorochemicals (PFCs) are man-made chemicals used as surfactants, polymer and plastic additives, and a variety of other applications. Their amphiphilic character and thermal, biological, and chemical stability make them useful for many purposes. The 864 Analytical Chemistry, Vol. 77, No. 3, February 1, 2005
concern of this group of chemicals began after the discovery of perfluorooctanesulfonate (PFOS) and perfluorooctanoic acid (PFOA) being widespread in the environment far from production locations at levels comparable to certain PCB congeners.1-4 The 3M Co., the primary worldwide manufacturer of perfluorooctanesulfonyl fluoride (PFOSF)-derived chemicals (including PFOS), declared a phase-out of PFOSF products in May 2000.5 This announcement was closely followed by the U.S. EPA production and import regulations of PFOSF-derived chemicals.6 PFOSF-derived chemicals were used for fabric, leather, and apparel treatment, for protection of food packaging and paper products, and as performance chemicals, e.g., fire extinguishing foam and insecticides.7 The manufacturing of PFOA was discontinued by 3M Co. as well, but several other companies worldwide maintain the production. PFOA is an essential processing aid in the fluoropolymer industry.8 Cookware, carpets, and textiles are treated with fluoropolymers to provide stain, grease, and water repellency. Together with inertness and great heat stability, these properties have resulted in extensive industrial use of fluoropolymers. Subsequent studies have pointed out the environmental persistence and bioaccumulation of this group of chemicals9 together with frequent occurrence in nonoccupationally exposed humans from several countries.10-13 PFOS is the compound in this class of chemicals that has been found most frequent and at the highest concentration in human blood. Levels up to 82 ng/ * Corresponding author. Tel.: +46 19 30 14 01. Fax: +46 19 30 31 69. E-mail:
[email protected]. † O ¨ rebro University. ‡ Stockholm University. (1) Giesy, J. P.; Kannan, K. Environ. Sci. Technol. 2001, 35, 1339-1342. (2) Kannan, K.; Franson, C. J.; Bowerman, W. W.; Hansen, K. J.; Jones, P. D.; Giesy, J. P. Environ. Sci. Technol. 2001, 35, 3065-3070. (3) 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, 1593-1598. (4) Ka¨rrman, A. Report MTM 02-03-PP; O ¨ rebro University, Sweden, 2002. (5) EPA Public docket AR 226-0588, 2000. (6) EPA Fed. Regist. 2000, 65, 62319-62333. (7) EPA Public docket AR 226-0550, 1999. (8) EPA Public docket AR 226 OPPT fact sheet, 2003. (9) Martin, J. W.; Mabury, S. A.; Solomon, K. R.; Muir, D. C. Environ. Toxicol. Chem. 2003, 22, 196-204. (10) Masunaga, S.; Kannan, K.; Doi, R.; Nakanishi, J.; Giesy, J. P. Organohalogen Compd. 2002, 59, 319-322. (11) Kannan, K.; Corsolini, S.; Falandysz, J.; Fillman, G.; Kumar, K. S.; Loganathan, B. G.; Mohd, M. A.; Olivero, J.; van Wouwe, N.; Yang, J. H.; Aldous, K. M. Environ. Sci. Technol. 2004, 38, 4489-4495. 10.1021/ac049023c CCC: $30.25
© 2005 American Chemical Society Published on Web 12/30/2004
Table 1. PFCs Included in the Selected Ion Monitoring Method Including Quantification Ion and Fragmentation Voltage Settings
a
compound
abbrev
molecular formula
quantification ion (m/z)
fragmentation voltage (V)
perfluorobutanesulfonate perfluorohexanoic acid perfluorohexanesulfonate perfluorooctanoic acid perfluorooctanesulfonate perfluorononanoic acid perfluorodecanoic acid perfluorodecanesulfonate perfluorooctanesulfonamide perfluoroundecanoic acid perfluorododecanoic acid perfluorotetradecanoic acid perfluoroheptanoic acida 7H-perfluoroheptanoic acidb 1H,1H,2H,2H-perfluorooctanesulfonic acida
PFBuS PFHxA PFHxS PFOA PFOS PFNA PFDA PFDS PFOSA PFUnDA PFDoDA PFTDA PFHpA 7H-PFHpA THPFOS
C4F9SO3C5F11CO2H C6F13SO3C7F15CO2H C8F17SO3C8F17CO2H C9F19CO2H C10F21SO3C8F17SO2NH2 C10F21CO2H C11F23CO2H C13F27CO2H C6F13CO2H HC6F12CO2H C8F13H4SO3H
299 269 399 369 499 419 469 599 498 519 569 669 319 281 427
95 65 130 75 75 150 83 150 113 85 90 93 70 75 110
Internal standard, i.e., added before extraction. b Recovery standard, i.e., added before injection.
mL in serum from nonoccupational exposed humans has been reported, and corresponding levels reported for occupational exposed humans are 60-10060 ng/mL.10,11,13-17 The distribution and exposure routes for PFCs are still under investigation, and therefore, information of human exposure to a larger number of compounds is valuable. The toxicology effects on humans are not well understood. PFCs have shown to act as a peroxisome proliferater, affect membranes, and affect the reproductive system. Other effects such as metabolic wasting, increased liver weight, and lowered serum cholesterol have also been observed.18-22 Although studies on rodents and primates have shown the mentioned effects, studies on exposed humans have shown no substantial effects.16,17,23 In the first studies on fluorine in human blood, nonspecific analytical methods such as oxyhydrogen combustion and fluoride ion-selective electrode measurements were applied.24 The first report on organofluorine compounds in human blood was published in the 1960s,25 using nuclear magnetic resonance. In the (12) Olsen, G. W.; Church, T. R.; Miller, J. P.; Burris, J. M.; Hansen, K. J.; Lundberg, J. K.; Armitage, J. B.; Herron, R. M.; Medhdizadehkashi, Z.; Nobiletti, J. B.; O’Neill, E. M.; Mandel, J. H.; Zobel, L. R. Environ. Health Perspect. 2003, 111, 1892-1901. (13) WWFs European policy office, 2004. (14) Hansen, K. J.; Clemen, L. A.; Ellefson, M. E.; Johnson, H. O. Environ. Sci. Technol. 2001, 35, 766-770. (15) EPA Public docket AR226-0978, 2000. (16) Olsen, G. W.; Burris, J. M.; Mandel, J. H.; Zobel, L. R. J. Environ. Monit. 1999, 41, 799-806. (17) Olsen, G. W.; Burris, J. M.; Burlew, M. M.; Mandel, J. H. J. Environ. Monit. 2003, 45, 260-270. (18) Hu, W.; Jones, P. D.; de Coen, W.; King, L.; Fraker, P.; Newsted, J.; Giesy, J. P. Comp, Biochem, Physiol., Part C 2003, 135, 77-88. (19) Sohlenius, A.-K.; Messing-Eriksson, A.; Ho ¨gstro ¨m, C.; Kimland, M.; DePierre, J. W. Pharmacol. Toxicol. 1993, 72, 90-93. (20) Seacat, A. M.; Thomford, P. J.; Hansen, K. J.; Olsen, G. W.; Case, M. T.; Butenhoff, J. L. Toxicol. Sci. 2002, 68, 249-264. (21) Austin, M. E.; Kasturi, B. S.; Barber, M.; Kannan, K.; MohanKumar, P.; MohanKumar, S. M. J. Environ. Health Perspect. 2003, 111, 1485-1489. (22) OECD, ENV/JM/RD(2002)17/FINAL, 2002. (23) Olsen, G. W.; Burris, J. M.; Burlew, M. M.; Mandel, J. H. Drug Chem. Toxicol. 2000, 23, 603-620. (24) Hudlicky´, M.; Pavlath, A. E. Chemistry of organic fluorine compounds II, a critical review, 1st ed.; American Chemical Society: Washington, DC, 1995. (25) Taves, D. R. Nature 1968, 217, 1050-1051.
1980s, PFOA and total organofluorine in plasma and blood were analyzed by gas chromatography with flame ionization detection, electron capture detector, or coupled to mass spectrometry.26-28 The commercialization of interfaced high-performance liquid chromatography-mass spectrometry (HPLC-MS) facilitated selective and sensitive analysis of PFC acids in a more convenient way than before. The most commonly used technique for the analysis of PFCs in human blood has been ion pair extraction followed by HPLC and negative electrospray (ESI) tandem mass spectrometry (MS/MS). Tetrabutylammonium ion as the counterion in the ion pair extraction of PFOS and PFOA has been used together with GC analysis,26 LC-fluorescence,29 LC-ESI-MS/MS14 and LC-NCI-MS/MS.30 Recently, automated solid-phase extraction (SPE) has been used instead of ion pair extraction for the extraction of human serum.31 Previously reported studies on human exposure have mainly been focused on serum. Serum is easier to handle in SPE compared to whole blood since the red blood cells have been removed. Further, studies have shown that PFOS and PFOA bind to plasma proteins.32,33 To the best of our knowledge, studies on the distribution of long alkyl chain PFCs and sulfonamides in whole blood have not yet been conducted and therefore information about concentrations in whole blood are useful. In the present paper, we present a method for the analysis of 12 PFCs (listed in Table 1) in human whole blood samples. The method involves sample treatment with formic acid and sonication (26) Ylinen, M.; Hanhija¨rvi, H.; Peura, P.; Ra¨mo ¨, O. Arch. Environmental Contam. Toxicol. 1985, 14, 713-717. (27) Yamamoto, G.; Yoshitake, K.; Sato, T.; Kimura, T.; Ando, T. Anal. Biochem. 1989, 28, 371-376. (28) Belisle, J.; Hagen, D. F. Anal. Biochem. 1980, 19, 369-376. (29) Ohya, T.; Kudo, N.; Suzuki, E.; Kawashima, Y. J. Chromatogr., B 1998, 720, 1-7. (30) Sottani, C.; Minoia, C. Rapid Commun. Mass Spectrom. 2002, 16, 650654. (31) Kuklenyik, Z.; Reich, J. A.; Tully, L. L.; Needham, L. L.; Calafat, A. M. Environ. Sci. Technol. 2004, 38, 3698-3704. (32) Jones, P. D.; Hu, W.; de Coen, W.; Newsted, J. L.; Giesy, J. P. Environ. Toxicol. Chem. 2003, 22, 2639-2649. (33) Han, X.; Snow, T. A.; Kemper, R. A.; Jepson, G. W. Chem. Res. Toxicol. 2003, 16, 775-781.
Analytical Chemistry, Vol. 77, No. 3, February 1, 2005
865
followed by SPE with a common C18 sorbent. Evaporation of the extracted sample is unnecessary, which decreases recovery losses. This is the first SPE method reported for extraction of PFCs from human whole blood. Various groups have reported levels of PFCs in human whole blood using ion pair extraction and MS/MS.10,11 The applicability of single quadrupole MS for the quantification of PFCs in human whole blood is demonstrated, which facilitates more laboratories to execute PFC analysis. EXPERIMENTAL SECTION Solvents, Chemicals, and Standards. Ammonium acetate (>99%, pa for HPLC) was purchased from Fluka (Steinheim, Germany), formic acid (98-100%), trichloroacetic acid (pa), and acetonitrile (HPLC-grade) were from Scharlau (Barcelona, Spain), and methanol (HPLC) was from Labscan (Dublin, Ireland). The water used was laboratory-produced ultrapure water. All standard stock solutions were prepared by dissolving or diluting 50 mg (corrected for salt content) of solid or liquid standard in methanol and further dilution to 100 mL with water in polypropylene (PP) volumetric flasks. Dilution to working standard concentrations was made with methanol and PP flasks. Perfluoroheptanoic acid (PFHpA) and 1H,1H,2H,2H-PFOS (THPFOS) were used as internal standards, i.e., added before extraction, and were stored in water at concentrations that made 7.5 µL spiked to 0.75 mL of sample the suitable internal standard concentration (100 ng/mL). 7H-perfluoroheptanoic acid (7H-PFHpA) was used to control the recovery of the internal standard, i.e., added before injection, and was stored in water at a concentration making 5 µL spiked to the final extract (∼500 µL) a suitable concentration (20 ng/mL). All standard solutions were stored at 4 °C. Perfluorobutanesulfonate (PFBuS) tetrabutylammonium salt (g98%), PFOS potassium salt (g98%), perfluorodecanoic acid (PFDA; >97%), and perfluorohexanoic acid (PFHxA; g97%) were purchased from Fluka. PFHpA (99%), perfluorononanoic acid (PFNA; 97%), PFOA (96%), perfluorodecanesulfonate (PFDS) ammonium salt (25 wt % in 2-butoxyethanol (37%) in water), perfluoroundecanoic acid (PFUnDA; 95%), perfluorododecanoic acid (PFDoDA; 95%), and perfluorotetradecanoic acid (PFTDA; 97%) were purchased from Aldrich (Steinheim, Germany and Milwaukee, WI). Perfluorooctanesulfonamide (PFOSA; 97%) and 7HPFHpA (98%) were purchased from ABCR (Karlsruhe, Germany). THPFOS and perfluorohexanesulfonate (PFHxS; 98%) were purchased from Interchim (Montlucon, France). Whole blood samples. A quality control (QC) sample was prepared from a whole blood sample received from the blood center at the University hospital in O ¨ rebro. Various, known amounts of PFCs (2.6-39 ng/mL) were added except for PFOS, which was already present in the blood at a suitable concentration (8 ng/mL). After homogenization, small portions were stored in methanol-rinsed 15-mL PP containers at -20 °C until analysis. Twenty whole blood samples, drawn from the Swedish population registry, were collected in 1997-2000, consisting of 7 females (age 46-72) and 13 males (age 24-46). The samples were initially treated with heparin but stored in glass containers without heparin at -20 °C. Sample Cleanup and Extraction. Samples, QC sample, and blanks were prepared using the same procedure. The extraction cartridge (300 mg/3 mL Bond Elut C18 sorbent material with a particle size of 120 µm) was purchased from Varian (Harbor City, 866
Analytical Chemistry, Vol. 77, No. 3, February 1, 2005
CA). Additionally, other sorbent materials from Varian were also evaluated (C8, C2, cyclohexyl, phenyl, cyanopropyl, different C18 materials (porosity 60 and 500 Å, particle size 40 and 120 µm and one non end capped), and styrene-divinylbenzene-based ENV and PPL). The frozen blood sample was allowed to thaw at room temperature, and 0.75 mL was then transferred to a 15-mL PP centrifuge tube, prewashed with methanol. An aliquot of 7.5 µL of internal standard solution (100 pg/µL PFHpA and THPFOS) was added using a 10-µL Hamilton syringe. After thorough mixing, 3 mL of 50 vol % formic acid/water was added and the solution was sonicated for 15 min and centrifugated at 10000g for 30 min. An aliquot of 2.5 mL, corresponding to 0.5 mL of blood, was extracted on a SPE column, preconditioned with 2 mL of methanol and 2 mL of water. After washing the column with 2 mL of 40 vol % methanol in water, the column was put under vacuum suction until visual dryness. Elution was performed with 0.5 mL of methanol. All solutions including the sample solution were allowed to run through the column material without applying vacuum. The approximate flow time for 2.5 mL of sample solution was 5 min. Filtration of the final extract was performed through a 0.2-µm PP filter into a PP vial, prior to which 5 µL of recovery standard solution (20 pg/µL) had been added using a 10-µL Hamilton syringe. Separation and Detection. A total of 10 µL was injected into an HP 1100 LC/MSD system (Waldbronn, Germany) equipped with a binary pump, polyetherether ketone (PEEK) tubings, an automatic degasser, and a thermostated column compartment that was kept at 40 °C. Separation was achieved on a Discovery HS C18 (50 × 2.1 mm, 3 µm) column (Supelco, Bellefonte PA) and a 20 mm guard column of the same material. A water and a methanol mobile phase, each containing 2 mM ammonium acetate was delivered with a flow rate of 0.3 mL/min. The gradient started at 35% methanol followed by a 20-min ramp to 90% methanol, a 10-min hold followed by a 10-min washing sequence with 100% methanol, and then reverting to initial conditions allowing 7-min stabilization time. The outside of the needle was washed with methanol between each injection. Detection was performed with an HP 1100 mass spectrometric detector (MSD) with an atmospheric electrospray interface operating in the negative ion mode with the following settings: nitrogen nebulizer gas temperature 350 °C, nebulizer gas pressure 20 psi, nitrogen drying gas flow 13 mL/min, and capillary voltage 3500 V. Selected ion monitoring (SIM) was used measuring the ions at the fragmentation voltage given in Table 1. Quantification. The software Chemstation version A.08.03 was used for acquisition and analysis. For each sequence run, one blank sample (extracted water) and one QC sample were analyzed. After 5-10 samples one midlevel standard injection was done in order to confirm system stability. Methanol injections were carried out after standard injections and after the last sample injection. Quantification was performed using nonextracted standards in methanol and by relating the area of the analyte to the area of an internal standard. PFHpA, THPFOS, and 7HPFHpA were evaluated as internal standards but quantification was performed using PFHpA. Detection and Quantification Limit. The instrumental detection limit was defined as the concentration needed to produce a signal-to-noise ratio of 3:1. The method quantification limit for 0.5
Table 2. Recoveries (arithmetic mean and RSD) of 12 PFCs Determined by Triplicate Human Whole Blood Samples with Added Standards at 3 Different Concentration Levels level 1
PFBuS PFHxA PFHxS PFOA PFOS PFNA PFDA PFDS PFOSA PFUnDA PFDoDA PFTDA a
level 2
level 3
initial concn (ng/mL)
spiked concn (ng/mL)
recovery n ) 3 (%)
RSD (%)
spiked concn (ng/mL)
recovery n ) 3 (%)
RSD (%)
