Anal. Chem. 2001, 73, 2171-2177
Determination of Clenbuterol in Bovine Liver by Combining Matrix Solid-Phase Dispersion and Molecularly Imprinted Solid-Phase Extraction Followed by Liquid Chromatography/Electrospray Ion Trap Multiple-Stage Mass Spectrometry C. Crescenzi,*,† S. Bayoudh,‡ P. A. G. Cormack,‡ T. Klein,† and K. Ensing†
Department of Analytical Chemistry and Toxicology, University Center for Pharmacy, Groningen, The Netherlands, and Department of Pure and Applied Chemistry, University of Strathclyde, Glasgow, U.K.
Matrix solid-phase dispersion (MSPD) is a new sample pretreatment for solid samples. This technique greatly simplifies sample pretreatment but, nonetheless, the extracts often still require an extra cleanup step that is both laborious and time-consuming. The potential of combining MSPD with molecularly imprinted solid-phase extraction (MISPE) was investigated in this study. Liver samples were ground in a mortar with C18 sorbent and the homogenized mixture packed into an SPE cartridge and placed on top of a MISPE cartridge. Subsequently, clenbuterol was eluted from the MSPD cartridge onto the MISPE cartridge using acetonitrile containing 1% acetic acid. The ability of the molecularly imprinted polymer to selectively adsorb analyte in acetonitrile was exploited for re-extracting clenbuterol directly from this acetonitrile extract via the double cartridge tandem system. The analyte was eluted from the MISPE cartridge using acidified methanol. A clear eluate was obtained, which was subsequently evaporated, redissolved, and analyzed by HPLC electrochemical detection (ECD) or ion trap mass spectrometry (LC/IT-MS). The MISPE cartridge used in this study was imprinted using bromoclenbuterol, a structural analogue of clenbuterol, as the template. These MISPE cartridges showed excellent stability. The complete extraction procedure was rapid, and recoveries exceeded 90% for the target analyte. The method detection limit for the LC/IT-MS procedure was < 0.1 µg/kg. This method, therefore, satisfies the stringent requirements of European Union regulation EEC 2377/90.
cerns over the abuse2,3 of these compounds in farm animals as growth promoters, which have arisen as a result of documented poisoning cases, have led to strict law enforcement against their illegal sale and use. Clenbuterol is the only member of the β2agonist sympathomimetic class of drugs approved for therapeutic use within Europe on animals as a bronchodilator, and the FDA has added it to their list of drugs that are prohibited from extralabel use in animals.4 To ensure that meat products suitable for human consumption are virtually free from residues of β-agonists, the maximum residue limit (MRL) is fixed at 0.5 µg/kg.5 Sensitive and specific methods are, therefore required to monitor these drugs below the parts per billion level. Reports have indicated that the persistence of clenbuterol in plasma and urine is low, but it persists in the liver at much higher levels than in other edible tissues and is detectable in the liver for up to two weeks after the withdrawal of the drug from an animal’s diet.6 The liver is, therefore, the tissue of choice for detecting illegal use of clenbuterol. Conventional sample treatment procedures for animal tissue usually involve extraction with solvents7 or enzyme digestion, sonication, and centrifugation or filtration followed by time-comsuming cleanup steps in order to obtain a homogeneous liquid phase containing the analyte of interest and correspondingly low abundances of endogenous interferences. Most of these methods include at least one purification step using large amounts of organic solvent. An alternative to liquid extraction is supercritical fluid extraction (SFE). JimenezCarmona et al. reported large fluctuations in the recovery at the parts per million level when using ion pair SFE.8 Better results were obtained by other researchers9,10 using a post-SFE cleanup
β2-Agonists are widely used in human and veterinary medicine as therapeutic drugs for the treatment of pulmonary disease, especially asthma; however, when administered in amounts above the level recommended as the therapeutic dose, they act as repartitioning agents and function as growth promoters.1 Con-
(1) Ricks, C. A.; Dalrympale, R. H.; Baker, P. K.; Ingle, D. L. J. Anim. Sci. 1984, 59, 1247-1255. (2) Kuiper, H. A.; Noordam, M. Y.; van Dooren-Flipsen, M. M. H.; Schilt, R.; Roos, A. H. J. Anim. Sci. 1998, 76, 195-207. (3) Mitchell, G. A.; Dunnavan, G. J. Anim. Sci. 1998, 76, 208-211. (4) Code of Federal Regulations, Part 530.41, Title 21, Volume 6, 2000. (5) Council regulation EEC No. 2377/90, European Union, Brussels, 1990. (6) Meyer, H. H. D.; Rinke, L. M. J. Anim. Sci. 1991, 69, 4538-4544. (7) Blanchflower, W. J.; Hewitt, S. A.; Cannavan, A.; Elliott, C. T.; Kennedy, D. G. Biol. Mass Spectrom. 1993, 22, 326-330. (8) Jimenez-Carmona, M. M.; Tena, M. T.; Luque de Castro, M. D. J. Chromatogr. 1995, 711, 269-276. (9) O’Keeffe, M. J.; O’Keeffe, M.; Glennon, J. D.; Lightfield, A. R.; Maxwell, R. J. Analyst 1998, 123, 2711-2714.
* To whom correspondence should be addressed. Arrhenius Laboratory of Natural Sciences, Department of Analytical Chemistry, Stockholm University, S-106 91 Stockholm, Sweden. Fax: +46 8 156391. E-Mail: carlo.crescenzi@ anchem.su.se. † University Center for Pharmacy. ‡ University of Strathclyde. 10.1021/ac0014360 CCC: $20.00 Published on Web 04/11/2001
© 2001 American Chemical Society
Analytical Chemistry, Vol. 73, No. 10, May 15, 2001 2171
and enzyme-immunoassay determination. Matrix solid-phase dispersion (MSPD) is a one-step technique that simultaneously allows sample homogenization, analyte extraction, and cleanup.11,12 The tissue sample is blended and homogenized with an appropriate solid-phase material (usually derivatized silica) and then packed in standard SPE-reservoirs. This technique has been successfully applied to the determination of β2-agonists in biological tissue13 and in the determination of clenbuterol in liver.14-16 All of these methods, however, still require an extensive cleanup step, with the extent of sample pretreatment being dependent upon the analytical goal as well as the selectivity and sensitivity of the detection system. Typical cleanup steps include liquid-liquid extraction15,17,18 (LLE) and solid-phase extraction14 (SPE) with reversed-phase and ion exchanger or diphasic dialysis.19 Immunoaffinify chromatography (IAC) has also proven to be a particularly valuable tool in the isolation and purification of samples for residue analysis.20,21 This technique exploits antibodyantigen interactions to selectively isolate analytes from complex mixtures for qualitative and quantitative determination by standard chromatographic techniques. It offers simplicity in terms of sample manipulation in comparison to other cleanup methods. An evaluation of the binding capacity of several commercial and in-house affinity columns by Crooks et al. demonstrated that there were large differences in the recoveries of analytes from spiked biological fluid samples.22 The authors suggested that an insufficient concentration of the antibody or antibodies of poor avidity was the limiting feature. IAC was used to purify liver extract for quantitative determination by LC/UV23 and LC/MS.24 Molecularly imprinted polymers (MIPs) are attractive materials that enable the selective extraction of small molecules from complex mixtures.25 Molecularly imprinted polymers have been employed in immunoassays,26 as catalysts,27,28 in sensors,29 and (10) O’Keeffe, M. J.; O’Keeffe, M.; Glennon, J. D. Analyst 1999, 124, 13551360. (11) Barker, S. A.; Long, A. R.; Short, C. R. J. Chromatogr. 1989, 475, 353-361. (12) Barker, S. A. J. Chromatogr. A 2000, 885, 115-127. (13) Le Boulaire, S.; Baudet, J.-C.; Andre, F. J. Agric. Food Chem. 1997, 45, 2134-2142. (14) Horne, E.; O’Keeffe, M.; Desbrow, C.; Howells, A. Analyst, 1998, 123 (12), 2517-2520. (15) Courtheyn, D.; Desaever, C.; Verhe, R. J. Chromatogr. 1991, 564, 537549. (16) Boyd, D.; Shearan. P.; Hopkins, J. P.; O’Keeffe, M.; Smyth, M. R. Anal. Chim. Acta 1993, 275, 221-226. (17) De Wasch, K.; De Brabander, H.; and Courtheyn, D. Analyst 1998, 123, 2701-2705. (18) Lin, L. A.; Tomlinson, J. A.; Duane Satzger, R. J. Chromatogr. A 1997, 762, 275-280. (19) Gonza´lez, P.; Fente, C. A.; Franco, C.; Va´zquez, B.; Quinto, E.; Cepeda, A. J. Chromatogr. B 1997, 693, 321-326. (20) van Haagsma, L. A.; van de Water, C. In Analysis of Antibiotic/Drug Residues in Food Products of Animal Origin; Agarwal, V. K., Ed.; Plenum Press: New York, 1992; pp 81-97. (21) Wong, R. B.; Pont, J. L.; Johnson, D. H.; Zulalian, J.; Chin, T.; Karu, A. B. In Immunoanalysis of Agrochemicals; Nelson, J. O., Karu, A. B., Wong, R. B., Eds.; American Chemical Society: Washington, D.C., 1995. (22) Crooks, S. R. H.; Elliott, C. T.; Thompson, C. S.; McCaughey, W. J. J. Chromatogr. B 1997, 690, 161-172. (23) Lawrence, J. F.; Menard, C. J. Chromatogr. B 1997, 696, 291-297. (24) Lau, B. P.-Y.; Lewis, D.; Lawrence, J. F. J. Mass Spectrom. 1997, 32, 655661. (25) Stevenson, D. Trends Anal. Chem. 1999, 18, 154-158. (26) Andersson, L. I. Anal. Chem. 1996, 68, 111-117. (27) Strikovsky, A. G.; Kasper, D.; Gru ¨ n, M.; Green, B. S.; Hradil, J.; Wulff, G. J. Am. Chem. Soc. 2000, 122 (26), 6295-6296.
2172
Analytical Chemistry, Vol. 73, No. 10, May 15, 2001
as stationary phases for liquid chromatography.30 Molecularly imprinted solid-phase extraction (MISPE) cartridges have been used to extract pesticides31 and drugs32 from different biological matrixes. When compared to antibodies, MIPs offer several advantages in that they are stable at relatively high temperatures, in organic solvents, and over a wide pH range. Moreover, MIPs often can be developed within a matter of weeks, in contrast to the long development times (3-12 months) normally required for antibody production in animals. Quantitative analysis of clenbuterol is often performed using liquid chromatography (LC) or gas chromatography (GC) after chemical derivatization. Both of these methods suffer from low sensitivity and specificity in the absence of mass spectrometric detection (MS) and are considered to be unsuitable for measuring low parts per billion levels of clenbuterol in biological samples. GC/MS with electron impact (EI)33-35 and chemical ionization36 (CI) has been used for clenbuterol determination after derivatization to its trimethylsilyl (TMS) ether derivative. Because of excessive EI fragmentation of the TMS derivative of clenbuterol, only nonspecific low-mass fragments were used, thus limiting the usefulness of the technique. GC/MS based on negative-ion chemical ionization and the pentafluoroacetyl derivative of clenbuterol offer high sensitivity (low picogram levels), but the derivatization procedure may introduce large variability or sample loss and lengthen the analytical procedure. To fulfill EU criteria, a confirmatory method based on a combination of two derivatization methods, two ionization techniques, and tandem mass spectrometry was suggested.37 It is usually most convenient to work with the analyte in its native form, and for polar and basic analytes, liquid chromatography is the method of choice. Unfortunately, quantification of clenbuterol by UV detection does not deliver satisfactory sensitivity, although LC/UV detection can be improved to some extent by postcolumn derivatization.15 Electrochemical detection (ECD) is a very sensitive and selective technique, and it was successfully applied to the analysis of clenbuterol in urine38 and in plasma.39 In the recent decades, efficient interfaces between LC and mass spectrometers have been developed. Clenbuterol was analyzed in urine by coupled column (LC/LC) and thermospray (TSP) tandem mass spectrometry (MS/MS),40 in human plasma at low nanograms per milliliter (parts per billion) by atmospheric pressure chemical ionization (APCI)41,42 and APCI coupled with (28) Sellergren, B.; Karmalkar, R. N.; Shea, K. J. J. Org. Chem. 2000, 65 (13), 4009-4027. (29) Kriz, D.; Ramstro¨m, O.; Mosbach, K. Anal. Chem. 1997, 69, 345A-349A. (30) Remcho, V. T.; Tan, Z. J. Anal. Chem. 1999, 71 (7), 248A-255A. (31) Muldoon, M. T.; Stanker, L. H. Anal. Chem. 1997, 69, 803-808. (32) Mullett, W. M.; Lai, E. P. C. Anal. Chem. 1998, 70, 3636-3641. (33) Hooijerink, H.; Schilt, R.; van Bennekom, E. O.; Huf, F. A.; J. Chromatogr. B 1994, 660, 303. (34) Van Ginkel, L. A.; Stephany, R. W.; Van Rossum, H. J. J. AOAC Int. 1992, 75 (3), 554-560. (35) Dumasia, M. C.; Houghton, E. J. Chromatogr. 1991, 564, 503-513. (36) Leyssens, L.; Driessen, C.; Jacobs, A.; Czech, J.; Raus, J. J. Chromatogr. 1991, 564, 515-527. (37) Batjoens, P.; Courtheyn, D.; De Brabander, H. F.; Vercammen, J.; De Wasch, K.; Logghe, M. J. Chromatogr. A 1996, 750, 133-139. (38) Koole, A.; Bosman, J.; Franke, J. P.; de Zeeuw, R. A. J. Chromatogr. B 1999, 726, 149-156. (39) Diquet, B.; Doare, L.; Simon, P. J. Chromatogr. 1984, 336, 415-421. (40) Hogendoorn, E. A.; van Zoonen, P.; Polettini, A.; Marrubini Bouland, G.; Montagna, M. Anal. Chem. 1998, 70, 1362-1368. (41) Doerge, D. R.; Bajic, S.; Lowe, S. Rapid Commun. Mass Spectrom. 1993, 7, 462-464.
ion trap (IT) multiple mass spectrometry.43 Good results were also obtained from electrospray (ES) and tandem mass spectrometry detection after a reversed-phase chromatographic separation.24,44 The objective of the present study was to evaluate the potential of combining MSPD with molecularly imprinted solid-phase extraction (MISPE) to develop a rapid and sensitive method to identify clenbuterol in liver samples. This simple and versatile technique could be used as a screening method, with unequivocal identification being obtained by LC/ES-MS. EXPERIMENTAL SECTION Reagents and Chemicals. Clenbuterol hydrochloride {Benzenemethanol,4-amino-3,5-dichloro-R-[[(1,1-dimethylethyl)amino]methyl]-(9CI)}, trifluoroacetic acid (TFA), methacrylic acid, ethylene glycol dimethylacrylate, and azobis(isobutyronitrile) were purchased from Sigma-Aldrich Corporation (St. Louis, MO). Clenbuterol-d6 was supplied by RIVM (Community Reference Laboratory/Laboratory for Analytical Residue Research, National Institute for Public Health and the Environment; Bilthoven, Netherlands). Bromoclenbuterol {Benzenemethanol, 4-amino-3bromo-5-chloro-R-[[(1,1-dimethylethyl)amino]methyl]-(9CI)} was supplied by Boehringer Ingelheim. Acetonitrile for ECD was purchased from Fisher Scientific (Acros Organics; Geel, Belgium). Acetonitrile and methanol HPLC grade were purchased from Labscan (Dublin, Ireland). Water was purified using a Maxima ultrapure water system (Elga, Salm & Kipp BV; Breukelen, Netherlands). Glacial acetic acid, ethanol, and ether, all of analytical quality, were obtained from Merck (Darmstadt, 92 Germany). 1-Octanesulfonic acid, sodium salt monohydrate 98% came from Acros Organics (Geel, Belgium). LiChrolut strong cation exchanger (SCX) was purchased from Merck. Polymer Preparation. The bromoclenbuterol-imprinted polymer (MIP) and the nonimprinted control polymer (NIP) were prepared via methods similar to those described elsewhere.45 Briefly, EGDMA (3.26 g), MAA (296 mg), bromoclenbuterol (200 mg), AIBN (74 mg), and acetonitrile (4 g), were mixed in a glass tube, and the solution sparged with nitrogen for 5 min. The photochemically initiated polymerizations (UV radiation) were performed overnight at 5 °C in a thermostatically controlled water/ ethylene glycol bath. The tubes were then transferred to a water bath set at 60 °C for 24 h. The tubes were then crushed, and the recovered polymers were ground and wet-sieved with acetone to a particle size in the range of 25-38 µm. For the polymerization, all of the reagents and solvents were purified via standard literature methods. Sample and Cartridge Preparation. Samples (0.5 g) of beef liver were ground in a porcelain mortars with 2 g of C18 sorbent (Alltech; Breda, Netherlands), as shown in Figure 1a. Standard solutions used for spiking were prepared in ethanol and stored at 4 °C. The analyte was added to the samples during the homogenization. The homogenized mixture was packed into a SPE polypropylene tube (6.5 × 1.3 cm i.d.) and compressed to a volume of approximately 4.5 mL. MIP and NIP (60 mg) and 100 mg of (42) Doerge, D. R.; Churchwell, M. I.; Halder, C. L.; Rowe, L.; Bajic, S. Anal. Chem. 1996, 68, 1918-1923. (43) van Hout, M. W. J.; Hofland, C. M.; Niederla¨nder, H. A. G.; de Jong, G. J. Rapid Commun. Mass Spectrom. 2000, 14, 2103-2111. (44) Cai, J.; Henion, J. J. Chromatogr. B 1997, 691, 357-370. (45) Crescenzi, V.; Masci, G.; Fonsi, M.; Casati, G. Polym. Prepr. Am. Chem. Soc., Div. Polym. Chem. 1998, 39, 699-700.
Figure 1. Sample preparation. (a) MSPD cartridge preparation; (b) tandem extraction by MISPE.
SCX were packed in smaller diameter SPE polypropylene tubes (6.5 × 0.8 cm i.d.). Polypropylene frits were located above and below both homogenized mixture and sorbent bed. Plastic adapters for the in-series connection, as well as cartridges and frits, were supplied by Alltech. Blank liver samples were extracted and analyzed prior to the recovery experiments to ensure the absence of clenbuterol in the samples. MISPE cartridges were regenerated by washing them sequentially with 10 ml of acetone, 10 mL of water, 10 mL of 1 M NaOH, and 20 mL of water followed by conditioning using 1 mL of acetonitrile. Before their first use, MISPE cartridges were washed by simulating two complete extraction cycles from distilled water and regeneration. Extraction Procedure. The two-cartridge tandem system MSPD/MISPE (Figure 1b) was connected to a Vac elut SPS 24 (Varian; Walnut Creek, CA) vacuum extraction unit. Clenbuterol was eluted from the MSPD cartridge using 20 mL of acetonitrile containing 1% acetic acid (method A). The eluent was forced to pass through the two-cartridge tandem system by regulating the vacuum to obtain a flow rate of 1 mL/min. Method B was a modification of an existing method.14 In this case, the MSPD cartridge was washed with 8 mL of a mixture of hexane/diethyl ether (60:40, v/v) before it was connected to the MISPE cartridge. After passage of the extract through the MISPE cartridge, the MISPE cartridge was disconnected and washed with 5 mL of the same eluent. The vacuum was applied for one minute after the last drop of liquid had passed through the column. The analyte was eluted from the MISPE cartridge using 8 mL of a mixture of methanol containing 10% acetic acid. Eluate was collected in roundbottomed vials (10 × 1.3 cm i.d). and evaporated at 50 °C under Analytical Chemistry, Vol. 73, No. 10, May 15, 2001
2173
Table 1. Percentage Recovery of 1 µg of Clenbuterol from Acetonitrile and Acetonitrile/Acetic Acid 1%a
sample vol (mL), acetonitrile breakthrough + 10 mL of washing phase I elution fraction (6 mL) II elution fraction (6 mL)
MIPb
MIPc
MIPc
MIPb
NIPc
SCXc
10 5.6 ( 2.6 110 ( 3.1
